V

*
/
 \

'

\
*
This book reviews the long-standing debate over the relative merits of a
high-protein versus a low-protein diet. When protein (or “animal sub¬
stance”) was first discovered in vegetable foods, it was hailed as the only
true nutritional principle. Justus Liebig, the leading German chemist of the
mid-nineteenth century, believed that it provided the sole source of energy
for muscular contraction. In contrast, health reformers argued that high
intakes were overstimulating, leading to dissipation and decline. U.S. gov¬
ernment publications in the 1890s recommended that working men receive
1x5 grams of protein per day, but work at Yale indicated that men main¬
tained their strength on half that intake.
In the 1950s kwashiorkor, a disease of infants in many Third World
countries, was judged to be the result of simple protein deficiency. The
United Nations declared a world protein shortage. But the causes of kwa¬
shiorkor were reassessed, and projects to produce novel protein sources
were eventually abandoned. Today there is again concern that overcon¬
sumption of protein in affluent societies may damage health. This book puts
the protein controversy into a historical perspective that sheds light not only
on the subject itself, but on the scientific process as well.
'

\
Protein and Energy
 \

«
 Protein and Energy
A Study of Changing Ideas
in Nutrition

KENNETH J. CARPENTER
University of California, Berkeley

national institutes of health

N!H LIBRARY

AUG 5 1 1994
eUlLOlMG 10
BETHESD/1, MD 20892

Kim
***** saw Cambridge
In® UNIVERSITY PRESS
Published by the Press Syndicate of the University of Cambridge
The Pitt Building, Trumpington Street, Cambridge CB2 irp
40 West 20th Street, New York, ny 10011-4211, USA
10 Stamford Road, Oakleigh, Melbourne 3166, Australia

© Cambridge University Press 1994

First published 1994

Printed in the United States of America

a
Library of Congress Cataloging-in-Publication Data
Carpenter, Kenneth J. (Kenneth John), 1923-
Protein and energy : a study of changing ideas in nutrition /
Kenneth J. Carpenter
p. cm.
Includes bibliographical references and index.
isbn 0-521-45209-0 (he)
1. Proteins in human nutrition. 2. Proteins in human nutrition -
Government policy - History. 3. Energy metabolism. I. Title.
QP551.C37 1994
613.2'8 — dc20 93-3S.130
CIP

A catalog record for this book is available from the British Library.

isbn 0-521-45209-0 hardback

*
 Contents

List of Tables and Figures page vii

Preface xi

1 Nutritional Science before the Chemical Revolution,

1614-1773 1
2 Nutrition in the Light of the New Chemistry,
1773-1839 17
3 Protein Discovered and Enthroned, 1838—1845 40
4 Things Fall Apart, 1846—1875 55
5 Vegetarian Philosophies and Voit’s Standards,
1875-1893 77
6 Chittenden versus the U.S. Establishment, 1883—1912 100
7 Vitamins and Amino Acids, 1910—1950 119
8 Protein Deficiency as a Third World Problem,
1933-1957 !42-
9 International Efforts to Produce High-Protein
Supplements, 1955—1975 161
10 Reappraisals of the Third World Problem, 1955—1990 180
11 Adult Needs for Amino Acids: A New Controversy,
1950—1992 2.04
12 In Retrospect 2.19

Appendix A: Chemical Structures of Amino Acids 236

Appendix B: The Measurement of Protein Quality 23 8
Appendix C: Calculations of Amino Acid Balance
Using an Isotope Label 2.45
References 2.47
Index 2.71

v
 Tables and Figures

Tables

2.1 Nitrogen content of foods and estimates of their

nutritional equivalence page 34
2.2 Changes in carbon and nitrogen contents of plants
grown in sand, expressed relative to 100 g original
seed dry matter 36
2.3 Results of Boussingault’s first nitrogen balance
experiments with a cow and a horse 37
2.4 Chemical cycles in the plant and animal kingdoms 38
4.I Mean results from Smith’s 1863 survey of the diets of
low-income indoor workers and their families
compared with results from later surveys of relatively
low income families 63
5-1 Results of short-term nitrogen balance experiments in
a human subject 91
5-2- Nitrogen balance in a subject eating different
quantities of white bread 96
6.1 Food eaten by Mendel in one day of Chittenden’s
study and its calculated nitrogen content 109
10.1 Calculations leading to estimates for safe protein
allowances in successive FAO/WHO reports 189
10.2 Comparison of suggested requirement patterns with
concentration of four critical amino acids in the
protein of low-cost diets 194
B.i Protein efficiency ratios of four materials 240
B.2 Summary of results obtained with rats by Mitchell’s
nitrogen balance procedure 244

vu
 Tables and Figures

Figures

1.1 The title page of what appears to have been the first
book on nutrition in English 4
2.1 Lavoisier’s list of elementary bodies from his Elements
of Chemistry (1790) I9
2.2 Lavoisier experimenting on human respiration with
Madame Lavoisier taking notes 2.0
3.1 “Liebig in His Laboratory at Giessen” as portrayed on
an advertising card issued with boxes of Liebig’s Meat *
Extract 47
4.1 “Liebig Is Summoned to Meet King Max” as
portrayed on an advertising card issued with boxes of
Liebig’s Meat Extract 58
4.2 Prisoners sentenced to hard labor at work on the
treadmill at Brixton prison, London 64
4.3 Edward Smith demonstrating the apparatus that he
used to measure his excretion of carbon dioxide when
working on the Brixton treadmill 65
4.4 Cutaway sketch of Pettenkoffer’s apparatus for
measuring respiratory exchange in human subjects
over a period of several days 71
5.1 Caricature exemplifying the common view that
vegetarians necessarily became weakly 85
5.2 The “protein man” (1987), a familiar figure in ^
London’s central shopping district for many years 88
5.3 Short-term nitrogen balance results of Rubner, as
calculated by Bowie and plotted in relation to Voit’s
standard of 105 g digestible protein for an “average
worker of 70-75 kg” 92
5.4 Photograph of a 54-year-old Dane who lived on a
low-protein diet and cycled 363 miles in 37 hours
over boggy roads in a timed trial 98
6.1 Photograph used by Chittenden to illustrate the good
physical condition of his subjects after 5 months on a
low-protein diet in
7.1 Comparison of two female rats, each 140 days old, on
diets with different protein sources 130
7.2 Reproduction of the final summary of the rat’s
requirements for amino acids, as determined by Rose
and his colleagues in 1948 133

viii
Tables and Figures

Photographs of African children with kwashiorkor

and marasmus
Brock and Autret’s 195z diagram relating the protein
concentration in foods to human needs at different
ages for average-quality protein
Material from a 1971 U.S. Department of Commerce
booklet entitled “FPC — Fish Protein Concentrate: It’s
for People” (1971)
Mean results from Osborne and Mendel’s 1919 paper
comparing 4-week protein efficiency ratio values for
casein and lactalbumin when fed at different levels in
the diet
 Preface

1 his book has its roots in my desire to understand, first, how the
idea arose, and became endorsed by the United Nations Organization in
1965, that a growing nutritional problem requiring immediate attention
was a worldwide shortage of protein and, second, what caused the idea to
be suddenly discarded, amid some passionate debate and name calling.
Although it is of considerable importance that such an about-face be under¬
stood, and appropriate lessons drawn from it, the nutrition community
seems largely to have ignored the subject as an embarrassment.
However, this controversy from the 1950s and 1960s was really the third
of a series of disputes on whether human diets commonly contained either
too little or too much protein — the first coming in the mid-nineteenth century
and the second at the beginning of this century. I have therefore tried to
set out the development of ideas about protein from their recognizable
beginnings so that the whole sequence will appear in a single volume.
Of course, in the seventeenth and eighteenth centuries the terms “protein”
and “energy” did not exist, but investigators were struggling toward the
idea that while one part of our food was used to replace worn-out tissues,
or provide the material for growth, another part provided “fuel for the
flame of life” and was the source of animal heat. At first, after protein was
discovered, it was thought also to be the source of mechanical energy.
Although this notion was disproved, investigators still thought that a high
intake of dietary protein was responsible for nervous energy in the sense of
vigor, or “get up and go.” One school of thought held that the relative
prosperity of people in Western countries was a result of their vigor, which
was dependent in turn on their eating substantial quantities of meat. Others
believed that this practice was overexciting and stimulated uncontrollable
sexual energy, which had debilitating and demoralizing effects, so that a
low-protein, .vegetarian diet was to be preferred.
Particularly in the earlier years, the key advances came from only a small
number of people, and I have tried to say something about their interests

xi
 Preface

and backgrounds at times when there was no established career pattern for
someone wanting to work in this field. New ideas in science do not just
happen: they are products of individual minds, influenced by specific back¬
grounds. And it is this personal side of the history of science that increases
its interest, at least for some of us.
The historical chapters have all been written, to the best of my ability,
in terms of the problems and questions as seen by investigators at the time,
and without an immediate reinterpretation of their observations from a
modern point of view. Such an approach seems the only way for us to try
to put ourselves in the shoes of our predecessors. Readers with an interest
only in modern nutritional concepts could begin at Chapter 6, but I hope
that they will, in time, work their way back to the origins of the subject
and come to respect and appreciate the contributions made in the earlier
periods.
Although this book is intended to be a serious study of the development
of ideas about the human diet against the background of general scientific
concepts in each period, I have tried to put the issues in terms that will not
deter the general reader. At some points it has seemed necessary to explain
in detail, for the sake of those interested, how authors did their calculations
to arrive at their estimates of protein needs, but the reader Can skip these
without losing the thread of the arguments. Because “nutrition” is something
that involves us all in everyday decisions, and because its scientific study
has not, in general, been very mathematical or dependent on abstruse con¬
cepts, it is an area in which nonspecialists can study the gradual development
of scientific methods of investigation and their application in practice.
Because of the many relevant papers published since 1850, I can only
offer a survey of this period. I apologize for omitting the work of some
significant contributors. It could be said that a study should be limited to
the intensive and more comprehensive scrutiny of a single period of the
past. Rightly or wrongly, I have had the different aim of trying to connect
the past with the present, and still to use direct quotations from leading
individual scientists within each period.
There are many other aspects of dietary protein and related issues that
this book does not cover. For example, it is not concerned with the social
causes of poverty-related malnutrition or with the design of nutrition policies
and their implementation in the Third World. Nor is it concerned with
debates on the long-term effects of malnutrition in children or, in another
direction, with the biochemical mechanisms that control protein synthesis
in the body. All of these are matters of great interest and importance, but
others have written about them more knowledgeably than I could. One
chapter does discuss scientists’ attempts to produce new protein foods as a
direct response to the conclusion that there would (or could) be a world
protein crisis. Otherwise, the book is concerned solely with the evolution

Xll
 Preface

of scientific ideas about the need for protein and how our actual require¬
ments could be determined.
As far as I know, this is the first book to review the development of these
ideas over such a long time span. However, any author in such a field is
indebted to those who “went before” as specialists in particular aspects of
the subject. I feel particularly indebted to Professor J. R. Partington, among
the dead, for his four-volume History of Chemistry and to Professor F. L.
Holmes, among the living, particularly for Claude Bernard and Animal
Chemistry and his introduction to Liebig’s Animal Chemistry, as well as
for his encouragement.
I am also grateful to my colleagues Marc Schelstraete and George Wolf
for assistance with some convoluted German writing from the preceding
century, to Doris Calloway and Patricia Swan for their comments on por¬
tions of the manuscript and to Frank Smith of Cambridge University Press
and his anonymous reviewers, as well as to the production editor, Mary
Racine, for their suggestions. Finally, the secretaries in the Department of
Nutritional Sciences have put a handwritten manuscript, and many revi¬
sions, onto a word processor with skill and good humor, and this has been
much appreciated.
»
1 Nutritional Science before the
Chemical Revolution,
1614-1773

The object of this book is to trace the origins of and changing ideas
about the role of protein in our diet and the quantities needed for optimal
health. The word “protein” was not coined until 1838, and the chemistry
of materials falling into this class was only beginning to be understood in
the preceding 40 years. However, scholars were interested much earlier than
this in the basic question of whether animals (and humans) had the power
to turn any kind of digestible food into the material of their own tissues,
or only certain fractions of it, having basically similar properties. There
were also differences of opinion as to whether growth in the young, and
tissue replacement in adults, was the sole, or even the major, function of
nutrition.
In classical Greek medicine in particular, “diet” was an important con¬
sideration in the maintenance of health, although at that time the word
meant the whole manner of one’s life and environment, as well as one’s
food.1 2 In the second century A.D., Galen had written: “Our bodies are
dissipated by the transpiration that takes place through the pores in our
skin that are invisible to us; therefore we need food, in a quantity propor¬
tionate to the quantity transpired.”1 Foods were classified primarily ac¬
cording to what were thought to be their immediate effects on a person’s
mood. One food would be sedative, another aphrodisiac, and so on.3 Dif¬
ferent foods were recommended for people with different temperaments
(hot, cold, dry, moist) in order to bring them more into the norm or “ideal”
balance.4
Recommendations originally made in the classical period were still being
proposed in the sixteenth century. Persons of the choleric type were advised

1. Tremolieres (1975); Fidanza (1979).

2. Cited by Temkin (i960), p. 86.
3. Cosman (1983); Mauron (1986).
4. Eliot (1541).

I
 Protein and Energy

to eat coarse meat (such as beef rather than chicken), which was only slowly
digested, and to have frequent meals; otherwise, the intense heat of their
digestive system would scorch and damage the empty stomach, with the
fumes rising to cause a headache. By contrast, naturally melancholic persons
should eat moist and easily digested, boiled meats and should drink milk.5
These ideas seem strange and incoherent to the modern reader, but they
were in fact built up from a complex framework of assumptions.6

The New Beginning

There is little connection between traditional belief and the science of nu¬
trition that was stirring in the seventeenth century on the basis of new
developments in physical science and physiology. The first influential pub¬
lication in this period came from the Italian scientist Santorio (usually known
by his latinized name, Sanctorius). He was a professor in Padua, the seat
of a famous university but best known to English-speaking people from
Shakespeare’s references to it in the Merchant of Venice and the Taming of
the Shrew, which were written about the time that Santorio was studying
there. Santorio repeatedly weighed his food and drink, on the one hand,
and his excreta (urine and feces), on the other, and also measured changes
in his own weight. He reported in 1614 that, on average, his daily intake
amounted to about 8 lb. and his excreta only to about 3 lb. Since there was
no significant change in his weight, there was an unexplained daily disap¬
pearance of some 5 lb. of material. At that time respiration was thought to
serve only as a means of cooling the heart. And since it seemed that this
amount of food and drink was necessary, Santorio concluded, for want of
any other explanation, that the daily “disappearance” must be due to the
breakdown of this amount of body tissue, which was then excreted largely
through the skin in the form of “insensible perspiration.” The losses were
made good by the nourishment ingested.7 This was only a more quantitative
restatement of Galen’s view in the 2nd century A.D., but Santorio’s work
made a great impression.
The second significant advance was published in 1628 by William Harvey,
an Englishman who had obtained his medical degree at Padua. This was
the discovery of the circulation of the blood from the left side of the heart
throughout the body, via arteries, capillaries, and veins and back to the
right side of the heart, with a second circulation to the lungs. Harvey’s
demonstration of a constant, rapid perfusion of the body’s tissues by arterial
\

5. Boorde (1567).
6. O’Hara May (1977).
7. Santorio (1614); Bylebyl (1977), pp. 377-8.

2
 Nutritional Science from 1614 to 1773

blood provided a mechanism to explain the extensive movements of old

material away from the tissues and the incorporation of new.8
The French philosopher Rene Descartes, in his Discourse on Method, first
published in 1637 and still regarded as one of the seminal works in the
development of scientific thinking, made use of Harvey’s finding. He believed
that the living heart was extremely hot and argued that it was the repeated
exposure (perhaps 100 to 200 times per day) of absorbed nutrients circu¬
lating in the blood to this “hot spot” that gradually “distilled” them into
the “stuff” of animal tissues. The process might take longer with some foods
than others, but all would eventually be converted.9 This was not a totally
original concept, as has been shown elsewhere.10
We see a somewhat different view in The Natural History of Nutrition,
Life and Voluntary Motion, published in 1659 and considered the first book
on human physiology to have been written in English (Figure 1.1). The
author, Walter Charleton, born in 1620, was exposed to the new ideas of
Harvey, Descartes and others as a student at Oxford, where he earned an
M.D. degree. He was a fluent writer, in particular trying to explain recent
scientific discoveries and to reconcile them with Christianity, but his career
had its ups and downs. He sided with the royalists and remained in Oxford
when the English civil war began in 1642 and, while still in his early twenties,
became one of the king’s personal physicians. After the restoration of
Charles II, he became a founding member of the Royal Society and president
of the Society of Physicians.”
In his book on nutrition, Charleton wrote that animal heat continually
agitates the minute particles of the body and “dissolves, dispenses or con¬
sumes” them, so that there must be a continuous assimilation of equivalent
particles from our food. That is the first function of nutrition. The second
is to provide fuel (or oil) for the vital flame burning within us, which can
be extinguished either by suffocation or by want of sustenance.” This seems
consistent with the modern view that, as adults, we need food partly for
“maintenance” due to constant wear and tear, or turnover, of our tissues
and partly for combustion as a source of energy and secondarily of heat.
Charleton went on to say that he did not believe that the different parts
of the body needed correspondingly different materials for their restitution.
He believed that a principle similar to that found in the white of an egg
would be suitable for all, and he used the simile of a common rain supporting
the growth of trees of different species.13 (At that time, before the existence

8. Debus (1978), pp. 66-73.

9. Descartes (1649), p. 79.
10. Mendelsohn (1964); Hall (1969).
11. Rolleston.(f94o); Gelbart (1971); Fleitmann (1986).
12. Charleton (1659), p. 10.
13. Ibid., p. 9.

3
Protein and Energy

natural,
HISTORY
{ NUTRITION,

LIFE, and
VOLUNTARY MOTION.
Containing
f An the HEW DISCOVERIES
of A N A TOM IS T *5, and mod probable
► Opinions of P H T S I C I A N%

Cancernwg she
6EC0N0ME OF HUMAN NATURE 5
Methodically delivered in
EXERCITATIONS PHTSICO- ANATOMIC At
Jly Wait. Charlton : M.D.

1# 11 ■ 1 ■■■ 1 i ■■ ———

1 LONDON,
Printed for UfttryMerringmatti and are to btf &ld at
y his (hop at the Anchor m the lower walk iirihe
| new Exchange, l 6 } 9.

Figure i.i. The title page of what appears to have been the first book on
nutrition in English. (The author’s name is more commonly spelled “Char-
leton” but here the “e” is omitted.) (Bancroft Library, University of Cal¬
ifornia, Berkeley)

4
 Nutritional Science from 1614 to 1773

of carbon dioxide in the air had been discovered, it was thought that rain¬
water actually provided the “substance” of trees.)
He further thought that, after the food had been worked on in the acid
conditions of the stomach, the usable portions were divided up according
to their function — the fuel going into the blood for its distribution, and the
“restorative principle,” or succus nutritius, being circulated by another
route, which he thought must be the nerve fibers.14 Fifteen years later, John
Mayow expanded this concept in relation to rickets, acknowledging that
the idea originally came from the Cambridge professor Francis Glisson. His
argument, paraphrased, was as follows: In this disease the child’s head
appears unusually large but the legs weak and poorly developed. The size
of the head shows that there can be nothing wrong with the blood, which
is the same wherever it circulates. The only other fibers going to the legs
are the nerves. There must therefore be some blockage in the spinal tract
which hinders distribution of the nutritious nervous juice from the brain to
the lower areas.15

Mechanical Models

Charleton did not refer to the qualities of individual foods, nor did he
indicate which were good sources of the two classes of nutrients that he
discussed. But it was obviously more difficult to understand how plant foods
could provide the substances needed for growth or “maintenance” than
how animal products such as meat could be reconverted to animal (or
human) tissue. The next 70 years were dominated by the successes of New¬
ton and others in explaining phenomena in terms of physics, and mechanics
in particular. Not unexpectedly, enthusiasm developed for trying to explain
the function of living things as well by a mechanical physiology entirely free
of chemistry.16
It had been suggested as early as 1663 that “the human body itself seems
to be but an engine,”17 and Hermann Boerhaave, the Dutch physician and
leading medical teacher of his time, began his textbook for medical students
with 99 pages devoted to physics and the admonition that students must
familiarize themselves with Newton’s Principia.18 Andrew Pitcairn, one of
the first lecturers at the Edinburgh Medical School, founded in 1695, con¬
cluded that the “animal economy” was no more than a hydraulic system;
disease resulted from obstruction of the circulation, and animal heat resulted
from friction among the particles in the bloodstream tumbling against the

14. Ibid., pp. 156—6z.

15. Mayow (1674), pp. 306-9; Davis (1973), pp. 87-8.
16. Brown (1968), pp. 1x2-235; Hall (1969), vol. 1, pp. 250-63, 367-408.
17. Boyle (1684).
18. Boerhaave (1719), pp. 1-99.

5
 Protein and Energy

vessel walls and one another. This also had the effect of mechanically abrad
ing (i.e., grinding down) absorbed nutrients to the appropriate size for
deposit in the tissues, whenever there were holes waiting to be filled. 9 The
holes had arisen in the first place because of wear on particles in the wall,
which then fell out. In the rough and tumble of the bloodstream, these were
then finally broken down into pieces small enough to pass out through the
pores of the skin as insensible perspiration.
British workers confirmed Santorio’s observation of weight loss that could
be explained at the time only as being due to insensible perspiration, though
not in quite such quantity. Keill reported that with an average intake of 74
oz. food and water, 5 oz. were lost as feces, 3 8 as urine and 31 by per¬
spiration.10
The final conversion of absorbed aliment to usable animal material was
thought to result from the combination of the mechanical beat of the heart
and the pressure of the air: “In the heart, briskly rubbed together they
subtilize and rarefy each other, the more volatile parts grinding and breaking
in pieces the grosser and less subtilized; also dulling the edges of the sharper
parts.”11 People doing hard physical work - whose hearts were beating
more strongly and whose breathing was deeper and more rapid than that
of more sedentary people - broke down absorbed food particles more
quickly. Delicate food was therefore unsuitable for this class of people, on
the assumption that it contained smaller particles, which would quickly be
broken down too far and perspired before they could be used to fill in gaps
in the tissues. This provided an argument for coarse food being natural and
preferable for laborers. By contrast, the richer, and physically l|ss active,
classes did not have the digestive vigor to utilize such materials, and actually
needed more refined and dainty refreshment.11
The same line of argument was used to explain why inactive people needed
meat and laborers did not:
There is a similarity and homogeneity between the muscular flesh of tender, sweet
animals and that of the human body; the integral particles of their solids and the
component globules of their juices are ready formed... to build up the flesh and
furnish out the juices of the latter... with less labour or struggle than those of
vegetables in general; as a mason will sooner and more strongly build the walls of
a house, who has plain rectangular stones at hand, than one who has rough stones
only... which must be first figured or prepared for a solid, durable building
.15

Enthusiasm for the mechanical approach to medicine was on the wane

after 1740, for despite the all-embracing claims for its worth, it had failed

19. Pitcairn (1718), pp. 11, 20, 36-7.

20. Keill (1728), p. 332.

21. Crawford (1724), p. 65.

22. Ibid., pp. 174, 333; Arbuthnot (1731), p. 35.

23. Cheyne (1740), p. xiv.

6
 Nutritional Science from 1614 to 1773

to produce more effective treatments and was subjected to ridicule. John

Arbuthnot wrote in The Memoirs of Scriblerus:

It is well known to anatomists that the brain is a congerie of... canals of great
length variously intorted and wound up together.... Simple ideas are produced by
the motion of the spirits in one simple canal; when two canals merge they make
what we call a proposition, and when two empty into a third they make a syllogism.
... That some people think so perversely proceeds from the bad configuration of
these canals.14

A direct attack on the mechanical system came from the French scholar
Frangois Quesnay:

The idea, which many people hold, that digestion consists of mechanical pulveri¬
zation or grinding is a fiction devoid of any probability. It bears no relation to the
character of the digestive organ or of the foods which are digested there. The stomach
is a soft and supple pouch which can only be gently agitated.... Also, foods are not
typically friable, but doughy or cartilaginous, and usually also soaking in drink in
the stomach.15

Another of the basic assumptions of the system had been that animal heat
could be explained by friction among particles circulating in the blood. This
was challenged in the Proceedings o f the Royal Society for 1745 by a writer
who said that, just as in the polishing of marble, greasing or watering was
used to prevent friction; so would the fluids of the blood prevent heat from
being produced by its circulation/6 In his opinion, heat came from the
fermentation of food and from animal tissues coming into contact with
air/7

Digestion by Fermentation

In fact, the idea that something similar to fermentation occurred during the
digestion of food, or its “concoction” as it was called, was not a new one.
These were changes that were easily observed outside the body when foods
were kept moist and warm and lost most of their characteristic differences.
The Belgian scholar Jean-Baptiste Van Helmont (1577-1644) had argued
that the Ancients’ explanation of “vital heat” causing digestion was inad¬
equate. Cold-blooded creatures like fish were as efficient as hot-blooded
ones at digestion. There must be some chemical processes at work, and
these he attributed to specific “ferments.” In the case of the stomach an
acid ferment was involved, but one with properties not possessed by simple

24. Aitken (1892), pp. 352.-3-

25. Quesnay ji<747)> Vol. 3, p. 14.
26. Mortimer (1745), pp. 473~4-
27. Ibid., 477.

7
 Protein and Energy

acids such as vinegar or oil of vitriol (sulfuric acid). The direction of fer¬
mentation was also different from that observed to occur spontaneously
outside the body/8 Thus meat left in a warm place undergoes, putrefaction,
giving off a foul odor, but digestion of meat in the stomach follows a
different course.
Thomas Willis (1621-75), a leading medical writer of his time and an
Oxford professor before becoming a successful London physician, strongly
supported the fermentation concept and considered that it operated through¬
out the body tissues/9 He referred to the use of old, sour dough as a starter
for making a new batch of bread and suggested, by analogy, that old,
“perfected chyle” lodged in the folds of the stomach wall might have the
same action in speeding the fermentation of food as it arrived in the
stomach/0
In some ways it made very little difference whether the digestion of food
was considered a mechanical principle or one of fermentation. In either case,
the particles present were thought of as being jumbled about and rearranged
to give products with different properties. However, later writers did not
subscribe to Van Helmont’s belief that the fermentation in the stomach was
of a particular kind; they assumed that the changes in the digestive system
were the same as those seen in the breakdown of materials outside the body.
This in turn led to some novel ideas about the principles of good nutrition.

Acid-Alkali Balance

The spontaneous fermentation of grains, flours and starchy or sugary foods

such as potatoes or grapes resulted in their becoming acidic. Vinegar pro¬
duction was a well-known example of this. Animal products, however,
underwent quite different changes and became alkaline. Robert Boyle, an
aristocrat and one of the leading members of the Royal Society from its
foundation in 1662, was a pioneer in the attempt to dissociate chemistry
from alchemy and to establish it on a rational, scientific basis, and he took
an interest in the subject. A material, or vapor, was alkaline, according to
Boyle, if it turned a paper soaked in syrup of violets from blue to green;
the use of litmus paper, going from red to blue with alkali, was a later
development. In 1684 he wrote that putrefied blood and strongly heated
hartshorn (deer antlers) each gave off a volatile alkali, which, to him, was
identical with that of smelling salts (ammonium carbonate). However, he
added that society ladies who visited his laboratory told him that, while the
\
28. Helmont (1662), pp. 115, 208; Pagel (1956), pp. 524-6; Mendelsohn (1964), pp. 41-3.
29. Davis (1973), pp. 82—6.
30. Willis (1684), p. 12.

8
 Nutritional Science from 1614 to 1773

smell of the salts was pleasant and refreshing, the smell of the others was
disgusting.31
Francois de le Boe (1614—72), who latinized his name to Sylvius, was an
influential professor in Holland who argued for the chemical nature of the
digestive process. He believed that good health depended on acid and al¬
kaline materials in the body being in balance and thus neutralizing each
other. An excess of either caused a condition of acridity, which irritated the
tissues and was the basic cause of most diseases.31 From this it seemed to
follow that, if people were to remain healthy, they needed a balance of the
two types of food that went respectively alkaline and acid when decomposed
in the body. This seemed to explain the value of such traditional dietary
combinations as bread, which was acescent (acid-producing), and meat,
which was alkalescent. A diet tipped slightly to the alkalescent side could
be balanced by the addition of vinegar. Scurvy was considered to result
from an alkaline acridity, and ships’ surgeons were provided with sulfuric
acid as a corrective when naturally “sharp” (i.e., acidic) fruits like lemons
were not available.33
It soon became clear to people pursuing the matter that not all plant
materials were acescent. Many types of green leafy materials gave off a
volatile alkali either when left to go putrid or when distilled over a fire when
still fresh. To the surprise of some, even the leaves of the plants valued as
antiscorbutics (e.g., watercress and scurvy grass) proved to be alkalescent.34
A committee of the Royal Academy of Science in France was set up for the
systematic study of the chemical characteristics of different plants.35 One
finding was that mushrooms, either distilled or fermented, gave off so much
volatile alkali that “if one did not know better, one would think them to
be of animal origin.”36
Herman Boerhaave, who was, as mentioned before, the most influential
teacher of medical theory at the beginning of the eighteenth century, did
not agree with all of Sylvius’s ideas about acids and alkalies in the digestive
processes.37 However, he pointed out that, in view of animals’ ability to
live entirely on acescent vegetable foods, the predominantly acid chyle that
they produced must gradually be changed to animal tissues with their al¬
kalescent qualities. Therefore, the process of animalization of plant foods,
as he called it, was one of making them more potentially alkaline. He thought

31. Boyle (1684), pp. 116, 121-2.

32. Boas (1956), pp. 14-15-

33. Boerhaave (1715), pp. 313—21; Arbuthnot (1731), pp. 176—7, 181; Carpenter (1986a),

pp. 56-7.
34. Homberg (1702), p. 51; Boerhaave (1715), PP- 19-2-0.
35. Dodart (1731); Holmes (1971), P- x36; Stroup (1990), pp. 89-100.

36. Lemery (1721), PP- 3°-

37. Jevons (1962).

9
 Protein and Energy

that the heat of the body fostered these chemical changes. Again, he argued
that laboring peasants with their more vigorous physique were more capable
of this than the physically inactive. Bread and vegetables were therefore
suitable for peasants, while the rest of society needed inherently alkalescent
foods, like meat and eggs.38

The Discovery of Wheat Gluten

The next important finding was made by an Italian scholar who had also
combined the study of chemistry and medicine. Jacopo Beccari (1682^1766)
spent his whole working life in Bologna at the university and academy. The
latter was a novel institution supplied with scientific equipment that the
university departments of that period did not have.39
In the present context Beccari is remembered for a paper published in
1745 (though, in fact, it was an extended minute of a lecture given in 1728).40
He argued that “our bodies must, presumably, be composed of the same
substances which serve as our nourishment,”41 yet the obvious properties
of wheat flour, like those of other plant foods, appeared quite different and
almost opposite to the basically gluey nature of muscle, blood and egg white.
His new finding was that, in fact, wheat flour did contain* some of this
“animal glue.” It was normally hidden by the larger quantities of other
constituents. He had separated it by first sieving coarsely ground wheat so
as to remove the branny particles. Then he added a little water to the flour
and kneaded a ball of dough (as in bread making, but without yeast), then
continued to knead it under water until all the white floury particles in it
(i.e., the starch) had dispersed. The residue was a soft, elastic, glutinous
mass insoluble in water. When heated strongly, or kept moist and warm
for several days so that it began to putrefy, it gave off a “volatile alkali”
with a urine-like smell. This was exactly what happened with animal tissues,
but it was quite different from the behavior of whole wheat, or of the wheat
starch, which went acidic with a wine-like fermentation if left warm and
wet.
Beccari believed, therefore, that he had demonstrated that this vegetable
food contained a proportion of what could be called “animal substance”
and that this provided an explanation for the reputation of bread as being
highly nutritious. He was puzzled, however, at being unable to find similar

38. Boerhaave (1719).

39. Heilbron (1991), pp. 62—3.
40. Beach (1961).
41. Beccari (1745). An English translation has been published by Bailey (1941), but translations
of this sentence differ greatly. The original Latin is “Nam si corpus tantum spectemus,
immortalemque ac divinum animum excipiamus, quid aliud sumus, nisi id ipsum, unde
alimur?”

IO
 Nutritional Science from 1614 to 1773

glutinous material after washing out the floury particles from either barley
or beans.
Two historians of science have criticized the priority accorded to Beccari
in recent years as the “discoverer of vegetable protein.”4* They point out
that many workers had already (i.e., before 1728) shown that some vegetable
materials gave off volatile alkali when distilled or allowed to ferment. In¬
deed, some species had been described as “animal plants” on this basis, and
this was also the crucial test used by Beccari to identify gluten as an “animal”
substance.
We have already referred to these findings, as well as to Boerhaave’s
conclusion that animal bodies must have the power to turn acescent foods
into alkalescent tissues, since some animals and birds can live almost
entirely on grain. We do not, unfortunately, have Beccari’s full lecture in
his own words, but it seems likely from his introductory point, namely,
that “our bodies must, presumably, be composed of the same substances
which serve as our nourishment,” that he doubted the need for such
drastic changes. The expansion of this point would be: Within this food
(i.e., bread), traditionally classed as acescent, there is a portion that is
already “animal” in nature; therefore, this fraction could be the part used
to repair, or replace, body tissue while the remaining, starchy part is
either got rid of or, possibly, “form[s] oil for the lamp,” to use Charleton’s
phrase.
Another interesting aspect of Beccari’s work was that he had shown the
presence of gluten, with its physical, elastic properties similar to those of
coagulated animal tissues, by the very mildest of procedures - that is, wash¬
ing in cold water. There was a concern at this time to develop new and less
destructive methods of analysis using solvents,43 because it was feared that
the older methods of analysis by distillation or fermentation might have
yielded products that were not present in the starting materials.44 Thus
volatile alkali was produced by distilling materials that were not initially
alkaline. Certainly the repeated references to Beccari’s work in the second
half of the eighteenth century point to the importance attached to it at that
time.4S

A Physiologist’s Perspective

Albrecht von Haller, still regarded as one of the most brilliant physiologists
of the mid-eighteenth century, sets out interesting views of digestion and

42. Jevons (1963); Goodman (1971).

43. Venel (1755O, P- 3I9-
44. Holmes (1971), pp. 138-9-
45. Van-Bochaute (1786), p. 114; Fourcroy (1789b), pp. 2-52-~3-

II
 Protein and Energy

nutrition in his general textbook of physiology. He followed earlier writers

in thinking of the living body as in a state of continual turnover: “Like the
timbers of a ship, it is repaired every day during life, till at length not two
jots of the old or first material remain - we may venture to say that in three
years the change is universal.”46 “Even bones have a slow dissolution and
a perpetual renovation.... The whole liying body is in a perpetual state of
fluxion, consumption and renovation.”47
He says that the blood contains “dissimilar parts of various natures.”48
This simple statement can easily be overlooked, but it is important because
so many others had thought of it as a single “thing,” with regards to both
its manufacture and its conversion into tissue. He goes on to say that “the
red parts of the blood (i.e. the globules) seem to be chiefly used to produce
heat.... The hardening serum is more especially designed for the secretions
and nutrition of the parts.”49 These are made up of chalky particles cemented
together by glue or jelly.50 Elsewhere Haller suggests that the various uses
of fat are: as a lubricant for muscles moving against each other, “giving a
comely form and cushioning against knocks,” and having a primary role
in the formation of bile.51
Haller believed that the action of the stomach was mainly mechanical,
with food being rapidly moistened and gradually “opened1 up” to allow
extracted juices to diffuse out and move into the blood.5Z Only the flesh
of animals contains “the gelatinous lymph, ready prepared for the recruit
both of our solids and fluids..., and from our canine teeth and the shortness
of our cecum it appears that animal food is a necessary patf of our nour¬
ishment.” However, vegetable foods do have “a small portion of j«lly drawn
from their farinaceous [starchy] parts which, after many repeated circula¬
tions, is converted into the nature of our indigenous juices.” Vegetables
were also thought to serve a kind of “diluting” function: “They are necessary
to avoid over-repletion with blood... [which is said] to breed a fierce or
savage temper... only to be avoided by a change of diet in which a vegetable
acidity abounds. Hence it is that we are furnished with but few canine teeth,
and our appetite is strongly for acidulous vegetables.”53
Yet Haller recognizes that cattle and other species live on nothing but
vegetable materials, and he argues that these animals have the power to

46. Haller (1754), Vol. 1, p. 10.

47. Ibid., pp. 195-6.
48. Ibid., p. 136.
49. Ibid., p. 160.
50. Ibid., p. 6.
51. Ibid., p. zi.
52. Ibid., Vol. z, p. 157.
53. Ibid., pp. izz-3.

IZ
 Nutritional Science from 1614 to 1773

convert the plant materials that become viscid (sticky, viscous) when mixed
with water into a “coagulable lymph.”54
Other physicians at this time held the opinion that any viscous solution,
whether from cooked starch or gums, was automatically nutritious and
could be turned into blood, another sticky fluid. A ship’s surgeon facing an
outbreak of scurvy at sea had said that “the blood was so thin as to be
unfit for circulation.” From this it seemed to him that foods of a glutinous
nature, such as salt fish and bread, were the most suitable for long voyages.
And in 1777 James Lind was to write that an ounce of powdered salep
(dried orchid root) and another of “portable soup” (dried cakes of boiled-
down cattle offal, with seasoning) would make two quarts of a thick jelly
and serve as complete sustenance for a man for a day.55 The modern nu¬
tritionist would, of course, say that these 2 oz. of dry matter could provide
only 10% of a man’s daily needs, but we will see that exaggerated ideas of
the value of soup were to persist for at least another century.

Potatoes as a Substitute for Grains?

The next personality to contribute to our subject, a generation later, is the

French food scientist Antoine Parmentier. Born in 1737, he trained as a
pharmacist and eventually became chief pharmacist to the French army.
The records show that he had a long-term concern for the welfare of ordinary
people and the effects of periodic, local famines that afflicted the French
countryside during this period.56 While serving with the army in Germany
during the Seven Years’ War, he noticed the importance of potatoes in the
peasant diets in the areas where he was stationed. This crop was still almost
unknown in France, although it had been brought to Europe from the New
World at least 200 years earlier. Back in France, he demonstrated that
potatoes could be grown successfully in sandy soils that did not support
grain production. However, people were generally skeptical as to whether
potatoes were safe and “meant to be eaten” since they were not mentioned
in the Bible. Parmentier sought official support for his “Grow more pota¬
toes” campaign, and was successful, in part, because of a seemingly irrel¬
evant gesture by the king and Marie Antoinette, who accepted from
Parmentier some flowers from a potato plant and wore them as corsages
for a day. This was regarded as symbolic of royal support for the campaign.57
In his writings Parmentier was concerned to demonstrate that the potato

54. Ibid., p. 199.

55. Carpenter (1986a), pp. 49, 67.
56. Berman (1974).
57. Grande Encyclopedic (1899), Vol. 25, pp. 1177-8.

13
 Protein and Energy

could serve as a substitute for wheat flour, and he submitted the winning
essay to a French provincial academy of science that had offered a prize for
the best advice on substitute foods that could be used when wheat was
scarce.58 There was no problem over the quantitative yield of dry matter
from the new crop — it was at least equal to, and usually greater than, that
from wheat. However, with regard to quality, Parmentier had to take into
account Beccari’s demonstration of the gluten, with its animal character, in
wheat. He also referred in his writing to a thesis by Johanne Kesselmeyer,
published in Strasburg in 1759. This reported a series of experiments with
wheat gluten and the author’s conclusion that its properties made it virtually
identical with the material of blood serum, which explained why wheat was
so nourishing.59 Parmentier was unable to extract a similar, glutinous frac¬
tion from potatoes, though he, too, did many experiments with wheat gluten,
demonstrating that it was the factor that allowed fermented dough to rise
during baking and then not to collapse.60 In addition, and this is an inter¬
esting general point, he accepted that the fact that both insects and birds
seemed particularly attracted to wheat must be evidence of its general nu¬
tritional value.61
But if wheat consisted essentially of a large proportion of starch and a
small proportion of gluten, and only the latter was truly nutritive, what
was the nutritive value of potato, in which only starch had been identified?
It seemed to follow that it had to be zero. Parmentier considered the problem
and concluded that, since people obviously could thrive on a mainly potato
diet, Kesselmeyer and Beccari must be mistaken in their conclusions; there
was, in any case, too little of the gluten for it to be the majg nutritive
principle of wheat. Given the facts, as he knew them, he could only conclude
that the true nutritive material in both foods was the starch.6i
In so many puzzling questions that nutritionists have had to consider in
the past, the apparently simple choice of “either this or that” has turned
out to be spurious. As a French scholar has written in this century: “Logic
is an unreliable instrument for the discovery of truth, for its use implies
knowledge of all the essential components of an argument - in most cases
an unjustified assumption.”63
In the present example, vital additional information was published in the
very same year that Parmentier’s book appeared - 1773. The author was
H. M. Rouelle, born in 1718 and the younger, and by then the survivor,

58. Balland (1902.), pp. 1—20.

59. Kesselmeyer (1759).
60. Parmentier (1773), p. xx.
61. Ibid., p. 244.
62. Ibid., pp. 237—9.
63. Dubos (1950), p. 131.

14
 Nutritional Science from 1614 to 1773

of two brothers who were both famous French chemists of this period.64
He described in two papers how he had discovered soluble “animal sub¬
stance or, literally, “animal—vegetable glutinous material” in juices
squeezed out of fresh green plants as different as sorrel, hemlock and
chervil. In his second paper he gave a full account of this procedure using
hemlock taken just at the point of flowering.66 The plant material was
carefully crushed in a mortar and then pressed. The juice was filtered through
a cloth and then heated to a point at which he could just bear to immerse
his fingers in it for a few minutes. (He had no thermometer.) A green
precipitate formed, and this was retained on a filter cloth, transferred to a
glazed pot and stirred with 8 to 9 pints of added water. After 24 hours,
the water was poured off and the process repeated twice more. The re¬
maining precipitate was transferred to another filter cloth and blotted to
take up as much water as possible. It was then subjected to further extrac¬
tions, this time with the same quantity of hot alcohol in a double boiler,
until all the green coloring material had been removed. The alcohol was
“spirit of wine,” containing perhaps 80% by weight of ethanol, in view of
the crude distilling apparatus available at the time.
The residue was a dirty whitish gray, and darkened further on being
dried. When it was placed in a retort and heated, the distillate contained
an “alkaline volatile spirit,” that is, ammonia, and finally a “concrete volatile
alkali (crystals of ammonium carbonate?). The sequence of observations
and the appearance of the residue were exactly the same as that obtained
with either wheat gluten or the caseous curd of milk and was proof for
Rouelle that he had obtained another glutinous or vegeto-animal sub¬
stance.67

Summing Up

In 1773 chemistry was about to be revolutionized by some sensational

developments. This is therefore a good place to summarize the knowledge
gained and the changes in ideas that occurred over the 120 years from the
1650s to the 1770s.
Constant throughout the period was the assumption that respiration
served the purposes of cooling the heart and, through air pressure, aiding
physically in the digestion, or concoction, of foods for their conversion into
blood. Whether this conversion was achieved purely mechanically or
whether it was the result of specific types of fermentation was a subject of

64. Rappaport (i960); Partington (1962), Vol. 2, p. 76.

65. Rouelle (1773a), pp. 264—5.
66. Rouelle (1773b).
67. Ibid., pp. 61—6.

15
 Protein and Energy

debate, with the evidence for neither view strong enough to discredit the
other completely.68 In either case it was agreed that the main function of
nutrition was to replace rapidly turning over tissues and that the degraded
tissues broke down further into particles small enough to diffuse through
the skin in the form of insensible perspiration.
Toward the end of the period, it was suggested that only the fraction of
vegetable foods that had alkalescent properties was truly nutritive - that
is, capable of replacing worn-out animal tissue. At first, such a fraction was
found only in grains of wheat, but at the very end of the period, similar
fractions were found in green plants as well. However, it was thought
unlikely that these alkalescent fractions could be the only nutritive portions
since they made up such a small percentage of the whole. Was everything
else useless?
At the beginning of the period, Walter Charleton had referred to a portion
of ingested food being needed as oil or fuel for the “vital flame burning
within us.” However, with the growth of the mechanical philosophy, animal
heat was assumed to come from friction among particles in the bloodstream.
Similarly, for those who considered fermentation to be the main mechanism
of digestion, the violent reactions in progress were thought to be the source
of animal heat. The idea of something analogous to a flame requiring a
specific fuel dropped from sight. It seemed to make more sense that the
body was capable of turning every kind of digestible food into animal tissue
by one means or another.
The developments described in the next chapter were to throw entirely
new light on all these matters in less than a generation. *

68. Holmes (1975), p. 13s-

2 Nutrition in the Light of the New
Chemistry, 1773—1839

The period from 1770 to 1790 saw the development of a “new chem¬
istry” that changed the ways in which the utilization of foods could be
studied. The story has been told many times, at length, elsewhere.
It began with the studies of combustion in Britain over the preceding
hundred years. Why did the flame of a candle go out after a period of
burning in an enclosed space? Why did a mouse die quite quickly when put
into the space in which a candle had been burning, whereas a green plant
seemed to thrive in it? “Was there only one kind of air, or was ordinary
air a mixture?” Such questions were puzzled over in the Royal Society and
elsewhere. They were also thought to have medical importance, since ill
health was often associated in people’s minds with bad air and lack of
ventilation in slums and prisons, or below deck in ships of war where sailors
developed scurvy. There was also a philosophical problem: Why was it that,
despite the general principle of attraction between individual particles, as
expressed in the law of gravity, the particles in gases seemed to repel one
another? And yet a material like chalk seemed to contain a highly com¬
pressed gas that expelled itself into a hugely greater volume when the chalk
was heated and converted to lime.1

Antoine Lavoisier

The person who did the most to construct a rational scheme of chemistry
that has survived to this day was the French scientist Antoine-Laurent La¬
voisier. To quote a specialist: “The chemical Revolution was the work of
many hands... [it] was prepared on French soil with materials quite con¬
spicuously of British origin.”* It was completed under the storm clouds of
a political revolution.

1. A short introduction to these findings and their relation to Lavoisier’s work has been given
by McKie (1965).
z. Guerlac (1961), p. xiv.

17
 Protein and Energy

Lavoisier was a complex person, and his work as well as his personal
characteristics have been analyzed in great detail, as befits someone who
made such important contributions to science.3 He came from a family
whose members had in successive generations risen in status from postilion
to postmaster to lawyer. He himself was brought up in some luxury and
given every opportunity for study. He was expected to be a lawyer and to
hold official positions. After studying hard and qualifying for the practice
of law, he was already wealthy enough from his inheritance to be able to
work, as an amateur, in a geological field survey with a powerful patron.
He interested himself mainly in mineralogy.
In 1771, at the age of 27, he married a girl only 14 years old, but intelligent
and well educated, as well as an heiress. Through this new family connection
he became a member of the national tax-gathering syndicate (la Ferme-
Generale), which collected taxes in return for prearranged payments to the
state. The syndicate also had a monopoly on the sale of tobacco, and La¬
voisier was assigned to the supervision of that part of its activities. Apart
from his work for the syndicate, his substantial income allowed him to give
the remainder of his time to his own scientific interests.
In 1772 he became a member of an official committee of the Royal
Academy of Sciences, examining the claims in a paper regarding the complete
disappearance of diamonds when they were heated strongly in a furnace.
It was regarded as extremely mysterious for a gemstone to behave in this
way. Was it a case of combustion? At that time, combustion was viewed
as a material losing something, a “matter of fire,” that could be seen leaving
the material in the flame and in some way combining with the ay:, until the
air could take no more.
Lavoisier began to look critically at this view of combustion and to make
crucial observations for himself, doing quantitative weighing experiments
using the best possible balances. His first experiments confirmed that both
sulfur and phosphorus gained weight when they burned. In later work he
showed that a given weight of “inflammable gas (hydrogen),” obtained from
the reaction of metals with acids, yielded a greater weight of water when
it was burned in air. This dealt a decisive blow to the idea that water was
an element and made it easier to believe that “air” was not a single element
either.
Finally, in 1789 Lavoisier assembled the new scheme in his Traite ele-
mentaire de cbimie, which was immediately translated and published in
Edinburgh as Elements of Chemistry. Part of one table from the English
translation is reproduced here (Figure 2.1). What he called “simple sub-
V

3. Detailed studies of Lavoisier’s chemical work have been made by Guerlac (1961) and
Holmes (1985). Recommended biographies, including parallel studies of his British con¬
temporaries, are those of Aykroyd (1935) and Davis (1966).

18
 Nutrition in the Light of the New Chemistry

TABLE OF SIMPLE SUBSTANCES.

Simple fubftancet belonging to all the kingdoms ot na¬
ture, which may be considered as the elements of bo¬
dies.
New Names. Corre/pondent old Names.
Light Light.
Heat.
Caloric Principle or element of heat.
Fire. Igneous fluid.
Matter of fire and of heat.
Dephlogiflicated air.
Oxygen Empyreal air.
Vital air, or
Safe of vital air.
A rote Phlogifticated air or gas.
Mephitis, or its bafe.
Hydrogen Inflammable air or gas,
or the bafe of inflammable air.
Oxydable and Acidifiable Ample Subftances not Metallic.

Sulphur
Phofphorus
Charcoal
Muriatic radical • y
Fluoric radical . - C Still unknown.
Boracic radical - - j
Oxydable and Acidifiable Ample Metallic Bodies.

Antimony Mercury.
Arfenic Molybdena.
Bifmuth Nickel.
Cobalt Platina.
Copper Silver.
Gold Tin.
Iron Tungftein.
Lead Zinc.
Manganefe

Figure 2.1. Lavoisier’s list of elementary bodies from his Elements of

Chemistry (1790).

stances, that have not yet been decomposed,” we would call “elements.”
The major elements of living bodies - carbon, hydrogen, oxygen and ni¬
trogen - are all there. Lavoisier gave “hydrogen” its name because it could
be converted to (or was a source of) water when burned. (“Hydro” is the
Greek root for “water” and “gen” gives the idea of source, as in “genesis”
or “generate.”) He coined the term “oxygen” (or “acid former”) because
of his experience with the combustion products of carbon, sulfur and phos¬
phorus. In each case the oxides dissolved in water to form acids. He con-

19
 Protein and Energy

vJaxHHJtep' d<MJ oo/i /a/>oraioire

<> .
Cl\y>enic •c,i jus* /a /‘cofnretfion Oe /'jCam/ne -rift, -repo<J

Figure z.z. Lavoisier experimenting on human respiration with Madame

Lavoisier taking notes. (Wellcome Institute, London)

sidered using the word “nitrogen” (or “nitigen”) but eventually chose (and
the French have retained) “azote,” meaning the gas that does not support
animal life.4 (“Nitrogen” was the name adopted in English since, when
sparked with excess oxygen, this element gives a gas that dissolves in water
to yield nitric acid, well known at that time.)
Lavoisier also conducted physiological experiments with a younger col¬
league, Armand Seguin. They measured the uptake of oxygen and produc¬
tion of “fixed air,” or “carbonic acid” (as he called it), during the respiration
of both humans and guinea pigs, and concluded that the ultimate source of
the carbon expired was the food eaten. However, they believed that atmo¬
spheric nitrogen was neither consumed nor excreted through the lungs.
Lavoisier also found that the rate of combustion greatly increased when the
subject was doing physical work (Figure z.z). From experiments in a cal¬
orimeter with guinea pigs, he concluded that animal heat could be explained
as coming from the oxidation of carbon and hydrogen in our tissues (or
ultimately from our food).5
In their report of these experiments, Lavoisier and Seguin pointed out

4. Lavoisier (1790), pp. 48-53.

5. Seguin and Lavoisier (1789).

ZO
 Nutrition in the Light of the New Chemistry

that respiration now had to be considered in terms of using up the body’s

tissues. It was ironic, therefore, that the rich person who did no physical
work needed less food than the laborer, although it was the latter whose
ability to purchase food was so limited. They ended by expressing the hope
that the reforms then in progress would correct this imbalance.6 Lavoisier
was obviously taking the opportunity to show that, despite his own wealth
and membership in the tax syndicate, his sympathies were with the new
regime and the common people. But it is also true that his new findings had
far-reaching implications for the understanding of nutritional needs, and
these will be considered in due course.
One wonders what Lavoisier would have done next. But in November
1793, although he had been heavily engaged in work for the revolutionary
government, in the organization of gunpowder supplies and an improved
system of weights and measures, and was also a commissioner of the Na¬
tional Treasury, he was arrested. He was imprisoned with other former
members of the Ferme-Generale, who included his father-in-law. They were
ordered to complete an audit of their accounts up to 1791, when the syn¬
dicate had been suppressed.7 The government hoped to confiscate large,
fraudulent profits, but apparently there had been no excess profits. A cynic
has said that the syndicate’s legal profits were large enough to make fraud
unnecessary. Its accusers then relied instead on charges that the tobacco it
sold had been adulterated with water in order to increase its weight. La¬
voisier tried to explain that this was a normal part of the “curing” process,
but to no effect. It seems likely that the syndicate had been tempted to add
more water than was really necessary, despite Lavoisier’s protests at the
time. And, of course, if the man in the street had had difficulty over the
years in keeping the tobacco in his pipe lit, there was an obvious target for
his abuse - the greed of the tax collectors - and now there was an oppor¬
tunity for revenge.
In May 1794 the syndicate members were brought to a hurried trial, and
all 28 were guillotined the same afternoon. Lavoisier is believed to have
asked for a short reprieve in order to carry out one last experiment on
transpiration, but the presiding judge is said to have replied: “The Republic
has no need of savants [scientists or scholars].” Later generations have
expressed surprise that leading scientists among Lavoisier’s contemporaries
did not do more to try to save him. During the Reign of Terror at this time
it was obviously dangerous to appear to be in sympathy with someone
accused of being a traitor, but it has also been the general conclusion that
Lavoisier had made enemies and that others were jealous of his wealth and
scientific success.

6. Ibid., pp. 578—9.

7. Aykroyd (1935), pp. 155-92-

21
 Protein and Energy

Identification of the Volatile Alkali

We return to the subject of the alkalescent “animal substances” in food.

Lavoisier did no work on them himself, but in 1778 he wrote a report for
the Academy of Sciences on three memoirs submitted by Parmentier on
methods of measuring the quality of wheat and flour, and how to make
best use of samples that were beginning to deteriorate. We discussed in
Chapter 1 Parmentier’s view that starch was really the nutritive factor in
wheat, and this was repeated in the memoirs. Lavoisier wrote: “It has to
be accepted that other grains such as barley and oats, which do not^contain
gluten, are also nutritious. Starch must therefore be nutritious, but Par¬
mentier seems to us to go too far in singling out starch as the only nutritive
element in wheat. It seems obvious that gluten, with its chemical properties,
is more ‘animalized’ and therefore eminently nutritive.”8
The first advance came in 1785 when Claude Berthollet told the Academy
of Sciences in Paris that he had discovered the composition of the volatile
alkali associated with animal substances.9 Translating his paper into more
modern chemical terms, we may describe three of the results as follows:

1. Reacting “volatile alkali” (ammonia gas) with chlorine resulted in the

production of nitrogen gas.
2. Reacting ammonia gas with hot metallic oxides resulted in the production
of hydrogen.
3. Electric sparking in ammonia gas gave a greatly increased volume of gas.
This was then burned in oxygen with the production of water (proving the
presence of hydrogen) and leaving an unreactive gas (assumed to be nitro¬
gen).

Berthollet estimated that the compound contained one-sixth of its weight

as hydrogen. (This is close to the modern value of 1 part in 5.6.)
Lavoisier himself did further work on the composition of isolated plant
fractions and concluded that sugars, gums and starches consisted of com¬
binations of carbon, hydrogen and oxygen only.10 By contrast, animal tis¬
sues, in the light of Berthollet’s finding, obviously contained nitrogen as
well.
Berthollet argued that ammonia was not present as such in healthy tissue
and that it was formed by a combination of nitrogen and hydrogen during
the process of either distillation or fermentation.11 He was impressed by
repeated observations that treating animal substances with nitric acid re¬
sulted in the immediate release of nitrogen gas. (Presumably his “nitric acid”

8. Lavoisier (1778).
9. Berthollet (1785a, b).
xo. Lavoisier (1790), pp. 117, 142—3.
11. Berthollet (1786), p. 273.

22
 Nutrition in the Light of the New Chemistry

was impure and included a proportion of nitrous acid that would really
have been responsible for this reaction.) There was no such reaction when
he added nitric acid to ammonium salts.

How Many Animal Substances?

Antoine Fourcroy, another French chemist interested in applying his science

to medicine, read a paper to the Royal Academy of Medicine in August
1788.11 He supported and extended Berthollet’s findings of the presence of
nitrogen in isolated animal substances. He said that there was a perfect
correlation between the quantity of nitrogen produced by the action of nitric
acid and the quantity of ammonia that distilled over when the products
were heated.13
Fourcroy also said that there were three classes of animal matter. The
first was gelatin, the jelly that one could extract (by boiling) from skin,
tendons, membranes, and other such tissues. The second was albumin, which
was soluble in water but was precipitated by heat, acids or alcohols; it was
found in the white of egg, blood serum and milk casein, and it contained
a higher proportion of nitrogen than gelatin did. The third was fibrin, which
formed clots in blood and which “existed so abundantly in muscles.” It
contained the highest level of nitrogen.14 He was surprised to find little more
nitrogen in the flesh of carnivorous animals than in that of fruit eaters. Fish
were also similar to land animals in their nitrogen content.
In a second paper, published in 1789, Fourcroy reported his further anal¬
yses of plants. He confirmed Rouelle’s work (described at the end of Chapter
1), though he concluded that the “animal substance” prepared from the
juice of leaves was not a form of gluten but had all the properties of animal
albumin, including coagulation upon heating and redissolution by alkalis.
He found similar material in the roots of plants and in wheat. He made the
last discovery by following Beccari’s original procedure for separating the
gluten from wheat flour by washing the dough in cold water. He then
strained the washings to separate out the suspended starch particles. When
the clear filtrate was heated, an albuminous precipitate appeared. Fourcroy
concluded that wheat contained two “animal substances” — albumin, which
was analogous to that found in blood serum, and gluten, which was anal¬
ogous to fibrin.15
Fourcroy’s final finding was that acescent plants, and acidic fruits in
particular, contained absolutely no albumin. Instead, they consistently

12. Smeaton (1962), pp. 19—37.

13. Fourcroy.(¥7893), p. 43.
14. Ibid., pp. 41—2.
15. Fourcroy (1789b), pp. 255-9.

2-3
 Protein and Energy

yielded a gelatin-like material. He suggested that this vegetable jelly was

the product of albumin reacting with acid. Then, since all acids were believed
to contain oxygen, he suggested that gelatin differed from albumin only in
containing a higher proportion of oxygen.16 Fourcroy was to be vigorously
attacked some iz years later for some of these claims. In particular, it was
said that his “vegetable jelly” was well known to be only an aromatic resin,
soluble in alcohol and with none of the special properties of a glutinous
substance.17 Certainly Fourcroy made no mention of the “jelly” yielding
either nitrogen or ammonia, and one wonders, with hindsight, if he had
isolated pectin. A more detailed study of vegetable glue confirmed that it
was not gluten-like or an “animal substance.”18
Like so many of the leading chemists of this period, Jacob Berzelius of
Sweden took an interest in “animal chemistry.” From a study of the com¬
position of blood he concluded that fibrin, albumin and the coloring matter
(i.e., hemoglobin) resembled one another so intimately that they could be
considered modifications of a single substance. He called them “albumi¬
nous” and concluded that secretions such as the casein of milk were essen¬
tially “modified albumin.”19 This was not a novel conclusion by 1813, but
carried weight because of Berzelius’s reputation for careful, accurate work.
In passing, it has been said that “he left school with the usual report on
pupils of genius: that there was no hope for him.”zo We will return to his
fundamental contributions in the next chapter.

A Nutritional Reassessment

Already in 1786 Van-Bochaute had developed the argument that there was
one basic “animal substance” and that from Beccari’s and then Rouelle’s
work it could be concluded that it was manufactured only by the vegetable
kingdom/1 Tactfully, he suggested that Rouelle must have seen the impli¬
cations of his finding for animal physiology, but had been too modest to
make a show of them.
He made fun of the physiologists who tried to persuade their readers that
digested foods, whether they consisted of starch, sugars or fat could, almost
in a moment, be metamorphosed into “animal substance.”

Thanks to modern chemistry which can throw new light on these obscure sciences.
... we can now understand the process of digestion; we see that it really consists of

16. Ibid., pp. 260-1.

17. Proust (1802), pp. 109—10.
18. Bouillon-Lagrange (1805), p. 36.
19. Berzelius (1813), pp. 71, 115-6.
20. Partington (1964), Vol. 4, p. 142.
21. Van-Bochaute (1786).

24
 Nutrition in the Light of the New Chemistry

a solubilisation of the food materials accompanied by slight decompositions or

fermentations, which allow the animal substance, already formed by the vegetable
economy, to separate itself from the other components... and to pass into the blood
circulation by resorption from the surrounding veins.11

Van-Bochaute went on to say that this absorbed “animal substance” had

different uses, forming the casein of milk for the nutrition (i.e., building
new tissues) of the young and for the nutrition (i.e., restoration) of tissues
throughout the adult body. The other components absorbed from the gut
were not wasted. The “buttery part,” which was a vegetable oil, furnished
the fat laid down by animals; other portions were used, for example, as
antiseptics that corrected the “corruption leading to scurvy.” All the evi¬
dence had led him to believe that animal matter was manufactured solely
by plants. They were particularly suited for it because of the slowness of
their chemical processes and their organization, which was designed to
utilize the primary elements under the influence of heat, light, water and
the minerals of the earth.
Van-Bochaute’s paper, published in February 1786, was probably written
before Van-Bochaute had read Berthollet’s discovery of “animal substance”
containing nitrogen. But that discovery, together with Seguin and Lavoisier’s
1789 paper in which they said that atmospheric nitrogen took no part in
respiration, would only have strengthened his argument. In essence, of
course, it was that the building of new tissues and repair of existing tissues
was achieved only by absorbed “animal substance” and that the other
digested components of foods - sugars, starch, fat and so on - were used,
but only for other purposes.

The Concept of Animalization

Van-Bochaute’s paper was an exception to the main line of thought after

the recognition that animal substance(s) contained nitrogen in greater abun¬
dance than plants. Others seemed to have in mind two assumptions. The
first was still that the only function of nutrients was to replace worn-out
tissues, or to produce new ones, which meant that only nutrients in the
form of “animal substance” were of value. The second (already mocked by
Van-Bochaute) was that materials of lower nitrogen content could be turned
into “animal substance” by being topped up with nitrogen; that is, the
character of organic materials was determined entirely by their relative
contents of carbon, hydrogen, nitrogen and oxygen. This process was de¬
scribed as “animalization,” and the only debate concerned whether the
nitrogen came from the air or from organic nitrogen left over when the
carbon and hydrogen of some food molecules were oxidized so that the

zz. Ibid., p. 113.

25
 Protein and Energy

residual “organic” nitrogen could be transferred to other materials to raise

their nitrogen content.13 The oxidized material might be decomposing,
worn-out tissue rather than newly arrived nutrients, so that the process
would not really be as wasteful as it seemed.
Halle, writing in 1791, developed an argument on the basis of his belief
that all the nutritious components of foods (sugar, gum, starch jelly, al¬
bumin, casein, fibrin, etc.) when treated with nitric acid gave a residue of
oxalic acid. They therefore contained the same “oxide hydro-carboneux”
(carbon—hydrogen-oxygen compound) combined with either extra carbon
or extra nitrogen. He went on to suggest that, when ingested foods contained
this “base substance” with additional carbon, the extra carbon was removed
in the digestive tract or in the lungs and skin, where it came into contact
with air, and oxidized to form carbonic acid. Nitrogen then took its place.14
Medical school teachers in the “physiology” tradition continued to write
about nutrition with little regard to the new views of chemists that elements
could not be interconverted to one another. A standard history of medicine
tells us that “among clinical teachers of the 18th century there is no name
more highly esteemed than that of William Cullen (1712-90).”15 Cullen
did, in his final work, make the point that Halle’s gelatinous matter from
vegetables was interconvertible with starch and sugar, rather than being
albuminous in nature, and also that there was only a little difference between
starch and the gums extractable from salep. However, he said that, despite
Beccari’s finding, he was “still of the opinion that the chief part of the
aliment of vegetables is afforded by acid, sugar and oil, which are com¬
pounded by the powers of the animal economy.”16
Benjamin Thompson (1753-1814) was another scientist who i^hored the
“new chemistry” in his writing on nutrition. He was an American, but one
who took the royalist side during the War of Independence and became a
colonel in the British army. In 1784 he moved to Bavaria, where he rose to
head the army and receive a title as Count Rumford. He was at home in
both London and Paris, active in both the Royal Society and the Institut de
France, and even married Lavoisier’s widow, though they subsequently sep¬
arated. He had a genius for technological innovation to increase the effi¬
ciency, first of an army, and then the operations of ordinary life. He was
particularly interested in the conservation of heat and the design of stoves
and, for example, is credited with the invention of a drip-type coffeemaker.17
He was also concerned with developing cheaper, but still nutritious food
for the poor and believed particularly in the value of soups. He accepted

23. Fourcroy (1792), Vol. 2, p. 486; Holmes (1963), pp. 61-5; Florkin (1972), pp. 117-23.
24. Halle (1791), pp. 162-3.
25. Garrison (1929), pp. 357-8.
26. Cullen (1789), Vol. 1, pp. 169-73.
27. Brown (1976).
 Nutrition in the Light of the New Chemistry

the idea discussed in Chapter i that all thick soups were equally nutritious.
Therefore, since barley flour would thicken three to four times as much
soup as the same weight of wheat flour, he believed that it had three to four
times the nutritional value, and he put this idea into practice in the provision
of soup for the poor in Munich. In an essay written in 1798 he rationalized
this principle by suggesting that the discovery of water being a “compound”
would lead to further nutritional findings. He said that “our knowledge in
regard to the science of nutrition is still very imperfect,” but he thought it
likely that, if provided in a thickened form, water was capable of decom¬
position by the body, thus itself serving as a food.z8
In a book published in Paris in 1806, it was in fact suggested that nitrogen
was formed from the partial removal of oxygen from water, and also that
carbon, in turn, was formed from nitrogen and hydrogen in one proportion,
and ammonia from another proportion of the two.*9 Today, of course, we
have abundant evidence that carbon, nitrogen and hydrogen are all elements,
whereas ammonia is a compound made from nitrogen and hydrogen. But
the proofs for some materials being “elements” were not that strong at the
beginning of the nineteenth century, so it is not really surprising that La¬
voisier’s “new chemistry” was not universally accepted.

The Work of Francois Magendie

Magendie is an important person in our history — famous for carrying

out what now seem extremely simple experiments. Of course, it is with
hindsight that the experiments seem obvious and simple. But why had no
one previously done them? Was there something special, or different,
about Magendie’s background? Born in 1783 and brought to Paris at the
age of 8, with his father heavily involved in revolutionary politics and
his mother dead a year later, he was 10 years old before he learned to
read or write. Then he asked to enter elementary school, and at 16 became
apprentice to a surgeon friend of his father, for whom he carried out
anatomical dissections. At 20 he became a formal medical student. After
graduation he practiced surgery but then changed to physiology, becoming
a respected teacher and in time the accepted leader of experimental phys¬
iology in France.30
His first paper that concerns us, “The Nutritive Properties of Substances
That Do Not Contain Nitrogen,” was read to the Academy of Sciences in
1816.31 In view of its historic interest, I will set out Magendie’s argument

28. Rumford (1800), pp. 194-9, 290-3.

29. Thouvenel (1806), Vol. 3, pp. 26—8.
30. Grmek (1974).
31. Magendie (1816).

27
 Protein and Energy

at some length. He starts by saying: “Nutrition has often been the subject
of conjectures and ingenious hypotheses by physiologists - but our actual
knowledge is so insufficient that their only use is to try to satisfy our imag¬
ination. If we could arrive at some more exact facts, they would be important
for physiology and could well have applications in medicine.”
Magendie then reviewed the differing ideas as to the source of the nitrogen
in animal tissues. Those who believed that it need not come from food had
referred to reports of Arabs who ate nothing but gum when crossing the
African desert and of colonial natives who ate nothing but sugar. He was
skeptical of these claims and pointed out that people who asserted that
herbivores had a nitrogen-free diet were simply wrong: all plant leaves had
been found to contain nitrogen.
His own plan was to use dogs — since they, like ourselves, could eat both
plant and animal foods — and to feed them on a single food that was accepted
as nutritious but free of nitrogen. He gave his first dog an unlimited supply
of pure sugar and distilled water. It maintained a reasonably good appetite
for about 2. weeks, but was already losing weight and then developed an
ulcer on the cornea of each eye. After a month it died. Postmortem ex¬
amination of the carcass showed a total absence of fat and only one-sixth
the normal muscle mass. 4‘
Magendie thought it important to verify that these events were the in¬
evitable consequence of living on sugar alone. So he repeated the trial with
two more dogs and obtained essentially the same results. He then decided
to find out whether the results were caused by a particular property of sugar,
and he tested other materials that were generally considered to be flourishing
even though they contained no nitrogen. He performed separate feeding
experiments with olive oil, gum and then butter. All of the dogs died in less
than 40 days with the same characteristics at autopsy, but there was no
ulceration in the two dogs receiving olive oil.
Magendie wrote that these facts seemed to throw doubt on the belief that
oils, fats, gum and above all sugar were preeminently nutritive, even though
they were well absorbed and supported life in dogs for a longer period than
if the animals were starved completely. The work also seemed to make it
probable, if not certain, that the majority of the nitrogen present in animal
tissues came from the food eaten.31
He suggested that the results also indicated a promising treatment for
people suffering from kidney stones. Most of these stones contain high
proportions of uric acid and phosphate. The patients tend to be heavy people
who are big eaters of meat, fish and dairy products, so that they have a
high nitrogen intake and therefore a high excretion rate. Would it not be
reasonable to have them reduce their intake of these foods so as to reduce

32. Ibid., pp. 66-75.

28
 Nutrition in the Light of the New Chemistry

the concentration of uric acid in their urine? Magendie said that his expe¬
riences so far had supported the idea, though the change of diet seemed to
take a month to have an effect.33
One’s first reaction to this account is, perhaps, revulsion at the suffering
of the dogs. And in his own time Magendie was heavily criticized for his
physiological experiments on animals. But we have to remember two things.
First, the results with the first animals were so contrary to general expec¬
tation that they had to be repeated if they were to be believed. Second,
Magendie hoped that his nutritional work would improve human health.
Ill health requiring surgery involved acute pain and suffering — there were
no anesthetics at the time. Therefore, it must have seemed reasonable to
advance knowledge at the cost of some animal suffering.
We must now evaluate the work, and the conclusions, in terms of what
was known at the time. Certainly, the results showed clearly that dogs could
not live on any one of the test materials alone. The property that Magendie
recognized them to have in common was their lack of nitrogen. To rule out
the possibility that sugar had some uniquely bad quality, he repeated the
test with three other highly regarded foods containing little or no nitrogen.
He again found that the dead animals had less than normal nitrogenous
material in their urine and feces. But, of course, even if a healthy animal
were capable of making nitrogenous compounds from atmospheric nitrogen,
this does not mean that a sick animal too weak even to eat would have the
same ability. One critic argued that the results showed only that “an animal
cannot be supported by a highly concentrated aliment.”34 Presumably he
meant that a healthy diet had to include some fiber or ballast, and it is true
that Magendie’s diets were all lacking in it.
Two modern scholars, making a passing reference to Magendie’s work,
wrote: “He reported in 1816 that dogs fed a diet completely lacking in
nitrogen shortly died, and that the addition to such diets of nitrogen-
containing protein materials prevented this early mortality.”35 In fact, they
were using their imagination in this description. The really weak point in
the work was that Magendie had no positive controls, that is, with an added
nitrogenous material such as gluten or albumin. He fell into the same trap
as those he had criticized, by saying: “Everyone knows that dogs can live
very well on bread alone.” Was this a fact any more rigorous than the belief
that Arabs could live on nothing but gum? Magendie probably asked himself
the same question, or perhaps others put it to him. In any case, the second
edition of his textbook of physiology, published in 1831, contains these
short statements:

33. Ibid., p. jj.

34. Paris (18z6), pp. 96-9.
35. Greenstein and Winitz (1961), Vol. 1, p. 247.

2-9
 Protein and Energy

Since the publication of these facts in the former edition of this work, I have been
enabled to establish some other very important facts.... A dog, eating at discretion
pure wheaten bread, and drinking common water, does not live above fifty days. It
expires at that period, with all the known marks of final decay recorded above. A
dog, eating exclusively of soldier’s biscuit, lives very well, and his health is not in
any degree impaired. Some dogs, fed exclusively upon cheese, and others upon hard
eggs, lived a long time, but they were weak, meager: they lost their hair, and their
appearance announced an imperfect nutrition....
[A section on experiments with other species including rabbits and rodents is
omitted here.]
The most general, and the most important consequence to be deduced frpm these
facts, which it would be worth while to follow up and investigate anew, is that
diversity and multiplicity of aliments is an important rule of hygiene; which is,
moreover, indicated to us by our instincts, and by the variation induced by the
seasons over all nature, particularly in the species of alimentary substances.56

So here were upsetting results: Dogs were receiving nitrogenous foods

and still not thriving. Obviously there were problems to be investigated.
Magendie had shown the possibility of doing long-term nutritional exper¬
iments with animals, and he continues to be credited with having provided
experimental proof that animals need nitrogenous materials in their diet in
order to thrive.

The Controversy over Gelatin

The next nutritional experiments to be reported were concerned with a

question of considerable practical importance. Gelatin is an “animal sub¬
stance,” rich in nitrogen and obtainable by cooking bones in steam. It
dissolves in boiling water and gives an attractive, “sticky” physical character
to clear soup, or bouillon. Although flavorless itself, it can be seasoned with
condiments and vegetables. In France at the end of the eighteenth century,
where methods had been developed to extract gelatin from bones, it was a
much cheaper source of nitrogen than was meat. It was important, therefore,
to know whether its nutritional value relative to meat was in proportion to
its nitrogen content. If it was, philanthropists were anxious to use it to
provide economical but still nourishing soup for the poor.37 The conclusion
from the first trials using bouillon made from gelatin in a Paris hospital was
that the patients seemed to do neither better nor worse than when they had
received an ordinary bouillon made from boiling meat.
In 1814 there was a further report on the use of gelatin prepared from
bones after their inorganic material had been extracted with hydrochloric
acid. Hospital patients received a mixed bouillon, three-quarters from gel-

36. Magendie (1831), pp. 478-9.

37. Magendie (1841), pp. 237-42; Holmes (1974), PP- 6-7.

30
 Nutrition in the Light of the New Chemistry

atin and one-quarter from meat. The change was well accepted, but the
patients received the extra meat, spared from bouillon preparation, as a
separate roast. The authors of the report considered it to be a generally
accepted fact that the nutritive value of meat derived from its content of
gelatin.3* We have already seen references to starchy materials being re¬
garded in this way because of their capacity to thicken soup and so give it
a consistency closer to that of blood.
Hospitals in Paris persevered with the use of gelatin-bouillon as an econ¬
omy measure, but in 1831, at the Hotel-Dieu, a group of physicians, sur¬
geons and pharmacists (that included Magendie) submitted a formal
statement to their administration urging that the use of gelatin be discon¬
tinued. Their reasons were that substituting part of the regular bouillon
with gelatin made it go putrid more quickly, that gelatin had a disagreeable
taste, which reduced the patient’s appetite, and that its nutritional value
was unproven.39 These complaints were made even though the gelatin was
prepared each day in the hospital kitchen by extraction with steam from
fresh bones.
In 1833 two French scientists published the results of a series of feeding
trials with puppies.40 They gave some of the animals bread alone, and other
bread supplemented with either gelatin or meat bouillon. The puppies did
better with a supplement of gelatin bouillon than with bread alone, but still
failed to grow consistently and became feeble after some weeks. In one dog
that died after a period of receiving the mixture, the tissues were wasted
and extremely pale. The puppies that received pure meat bouillon with their
bread or even a small proportion mixed into the gelatin bouillon did perfectly
well. The final conclusions were, therefore, that gelatin had some positive
value but was not equivalent to meat in nutritional value and had to be
used in combination with other materials, as was always the case in practical
human diets.
The French Academy of Sciences had, in 1831, set up a commission to
investigate the subject after receiving a letter from a chemist who reported
that he and a group of medical students had tried to live on bread and
gelatin with little success.41 Magendie was made a member of the commis¬
sion and began a long series of feeding experiments with dogs. He did not
report the results of the work until after 10 years of effort, and even then
said that much more still needed to be done: “As so often in research,
unexpected results had contradicted every reasonable expectation. ”4Z
There was no doubt that, after a few days of being offered nothing but

38. Leroux et al. (1814).

39. Petit et al. (1831).
40. Edwards and Balzac (1833).
41. Holmes (1974). P- 9-
42. Magendie (1841), p. 239.

31
 Protein and Energy

gelatin, dogs preferred to starve. Adding fat, or flavoring the gelatin with
salts and ham, would revive their interest for only a few days. However,
the dogs were equally averse to eating egg white, either raw or lightly cooked,
as their sole food. In separate trials with raw beef tendons or washed blood
fibrin, dogs continued to eat for several weeks, but eventually lost a great
deal of weight or died.
The only foods on which dogs continued to remain vigorous were raw
bones (but not cooked bones), bouillon made from meat, meat (but not the
residual fibers after extraction of lean meat in water for 24 hours) and crude
wheat gluten containing some residual starch and fat. 4
Magendie suggested that chemists investigate what the extractable parts
of meat were that combined with gelatin, albumin or fibrin to make it a
complete food; it could perhaps be iron or other salts, fatty material or
lactic acid. There was clearly more work to be done. Nevertheless, it seemed
clear to Magendie that gelatin was not nutritious, and the reason was that
it was not an “organic element” present, as such, in raw bone but a product
of the reaction of bone material with water and heat.43

The Work of Boussingault

Yet another Frenchman to make an original contribution to our subject in

the 1830s was Jean Baptiste Boussingault (1802-87). As with Magendie, it
is interesting to consider whether Boussingault’s originality can be linked
to his unusual background and experience. He also grew up in Paris in
turbulent times. His teachers regarded him as a blockhead, arid he left school
at the age of 14, thus losing the chance of going on to a university. But his
parents, who ran a modest tobacconist’s shop, bought him a four-volume
textbook of chemistry, which had caught his interest, and he avidly attended
free public lectures on science over the next two years. At 16 he was admitted
to the School of Mines near Lyons. The school had been set up to produce
knowledgeable mine managers, and the two years of training combined
chemistry and mineralogy with the practice of mining. In his second year
Boussingault was selected as “student-demonstrator” with permission to
work in the laboratory at all times. Surprisingly, one of his interests lay in
the production of glucose by the fermentation of different materials, and
when still only 18 he submitted a paper on platinum, which was published
in the principal French chemical journal.44
He was then appointed manager of a mine in Alsace, but in 1822 he
accepted the offer of an academic position at the National School of Mines
in Bogota, capital of the newly independent Colombia, which was still at

43. Ibid., pp. 153-81.

44. McCosh (1984), pp. 3-12.
 Nutrition in the Light of the New Chemistry

war with Spain. He spent the next io years in South America. There was
little teaching to be done, but he devoted a great deal of time to exploring
dangerous country with a portable laboratory, making geological reports
and improving methods for extracting metals from ores. In addition, he
arranged for a supply of iodine compounds to be used in the treatment of
endemic goiter. Some 50 papers from this period were published in France
detailing his analyses of minerals, volcanic ash and foods, as well as his
meteorological observations at high altitudes.45
In 1835 he married a woman whom he had first met in Alsace twelve
years earlier. Her father owned a farm there, and from that time on Bous-
singault applied his scientific mind to agricultural problems. He began a
long series of quantitative experiments on the farm, with both plants and
livestock. But he also made a home in Paris, where in 1837 he obtained a
university teaching position and began to collaborate with J. B. Dumas on
the chemical composition of organic materials.46 He then spent half of each
year in Paris and the other six months at the farm.

The Nutritional Equivalence of Plant Foods

Boussingault’s first paper based on studies made at the farm was published
in 1836.47 He set forth the results of analyzing a range of animal feedstuffs
for their nitrogen content. The values for those that are also used as human
foods are presented in Table 2.1.
The author then developed an argument that we must follow step by
step:
1. All the vegetable substances used as food have been found to contain
nitrogen.
2. Magendie has shown that foods that do not contain nitrogen do not support
life for more than a very limited period.
3. It is accepted, therefore, that the nutritional quality of a vegetable food
resides principally in the gluten and vegetable albumin that it contains.
4. From this it follows that the nutritional value of a vegetable substance is
proportional to its content of “animal substances.”
5. It is true that some nitrogenous substances obtained from vegetable prod¬
ucts are not nutritive, and are even violent poisons, but such substances
are not present in appreciable quantities in plants eaten by humans or
animals.
6. I am therefore expressing as “nutritionally equivalent” the relative weights
of different foods that contribute the same amount of nitrogen, using
average hay as the standard for animal feeds and wheat flour for human

45. Ibid., pp.17-57; Cowgill (1964).

46. McCosh (1984), pp. 58-71.
47. Boussingault (1836).

33
 Protein and Energy

Table 2.1. Nitrogen content of foods and estimates of their nutritional

equivalence

Total
Nitrogen nitrogenous
content nutrients0 Nutritional
Foodstuff (%) (%) equivalents*

Wheat flour 2.27 14.2 (100)

Rice 1.28 8.0 177
Corn flour 1.64 10.2 138
Potato flour 1.80 11.2 126
Barley meal 1.90 11.9 119
Dried carrots 2.39 14.9 95
Dry peas 3.40 21.2 67
Lentils 4.00 25.0 57
Haricot beans 4.08 25.5 56

“Estimated as nitrogen content x 6.25, i.e., assuming that they contain 16% nitrogen
and are the only major nitrogen-containing materials present. This calculation was
suggested by Boussingault in 1843.
‘These are the quantities that contribute the same total nitrogen as 100 parts of
wheat flour; i.e., the lower the number, the higher the nutritional value of the food.
Source-. Boussingault (1836); Boussingault (1838a).

foods. (In other words, if a food has twice the nitrogen content of wheat,
only 50 grams [g] of it will be needed to replace 100 g of ^heat in our
daily diet.)

In the subsequent discussion, however, he went on to say that these were

“theoretical equivalents,” and it was important to check how far they agreed
with the practical experience of stock feeders.
He qualified his conclusions slightly in a second paper, published in 1838,
saying that he was doubtful whether humans could live on bread alone!
This appeared to be because of the large quantity he thought would be
required and the human capacity for food being so much smaller than that
of herbivores. Some peoples were said to subsist on potatoes and others on
rice. However, he knew that the potato-eating peasants in Alsace also con¬
sumed a good proportion of milk curds. And an authority on Indian life
had written that the quantities of rice eaten there were very much greater
than what a European could consume. In addition, the rice was eaten with
a supplement, typically fish or legumes. When Boussingault himself had
been in charge of a detachment of men (presumably in South America), he
had found 1 lb. of haricot beans to be equivalent in value to about 3 lb. of
rice, and that fitted with their nutritional equivalence, based on their relative

34
 Nutrition in the Light of the New Chemistry

nitrogen contents. Finally, he added values for manioc (i.e., cassava or

tapioca) and for plantains (starchy bananas), assigning them only one-
seventh the nitrogen content of wheat flour.48 He was almost certainly the
first person to draw attention to the nutrition problem associated with these
Third World staples, which was to become a major concern to nutritionists
in the 1950s.
What are we to think of Boussingault’s idea that the nitrogen content of
a food gave a measure of its nutritional value? Did his conclusions follow
necessarily from the facts that he used in his argument? We have to say
that they did not. Magendie’s extended results (including those published
in 1831) no longer supported the idea that a food could be judged by its
nitrogen content, so that 60 g of beans (or of any other food containing
4% nitrogen) would have the full nutritional value of 100 g wheat (con¬
taining 2.3% nitrogen). Boussingault had no justification for narrowing
down his criteria for a good food, as compared with the conclusions of
Vauquelin and Fourcroy, who had written 30 years earlier that “beans are
so nourishing, and capable of replacing all other food because they contain
starch, an animal matter, phosphate, lime, magnesia, potash and iron. They
yield at once the aliments and the materials proper to form and color the
blood and to nourish the bones.”49 However, Boussingaut’s hypothesis was
stimulating, even if it served in the short term to get things out of perspective.
In his book Rural Economy, published in 1843-4 and quickly translated
for British and American readers, Boussingault added some further impor¬
tant qualifications. First, he said: “I am far from regarding nitrogenous
materials alone as sufficient for the nutrition of animals; but it is a fact that
where nitrogenous materials are present at high levels in vegetables they are
generally accompanied by the other organic and inorganic substances which
are also needed for nutrition.”50 Then, at a later point, he wrote:

... in balancing diets for nitrogen we are considering only the flesh contained in the
foods, and although it is unquestionably the principle of highest value, and the one
most likely to be deficient, it is still not everything. Starch, sugar, gum and oil are
indispensable as auxiliaries in the alimentation of cattle.... Potato and meadow hay
brought to the same state of dryness contain the same proportions, i.e. 1.3 -1.5%
nitrogen, or about 8.5% of albumen and gluten, i.e. of flesh. But in the potato
almost the whole of the remainder consists of starch, while in hay by far the largest
proportion is woody fiber, which we presume to be inert. And this explains the
higher value of the same weight of dry potato as an article of sustenance. To give
our theoretical equivalents all the precision that is really desirable, it would be
necessary to ascertain the quantity of organic matter that escaped digestion.

48. Boussingault (1838a).

49. Vauquelin and Fourcroy (1806), p. 2.Z0.
50. Boussingault (1845), p. 387.
51. Ibid., pp. 390—1.

35
 Protein and Energy

Table 2.2. Changes in carbon and nitrogen contents of plants grown in

sand, expressed relative to 100 g original seed dry matter

Clover Wheat

C N C N

Initial content (g) 50.8 7.2 46.6 3.5

Changes with germination -11.4 0.0 -6.9 + 0.1
Changes with growth
2 months + 32.6 + 0.7 + 25.8 * -0.3
3 months + 80.5 + 2.6 + 42.0 + 0.2

Source: Boussingault (1838b).

The Fixation of Nitrogen by Plants

Boussingault was particularly interested in the source of the nitrogen in

plants. In his memoirs, written late in life, he said that it was in 1832,
toward the end of his time in South America, that he saw the value of
Peruvian guano, “formed almost exclusively of ammonium shits,” in making
sterile soils fertile and that this caught his interest.5* (Guano came from the
breakdown of bird droppings where birds living on the abundant fish off
the coast made their nests.)
His next farm studies addressed this question, beginning with elegant
trials of the growth of seedlings in sand that had previously be^n heated in
an oven until all the organic matter present had been burned off.53 First,
he compared clover and wheat seeds, and their change in composition as a
result of germination, and then the composition of the products, some
harvested after 2, and the remainder after 3, months of growth. The results
are summarized in Table 2.2. Boussingault concluded that clover had the
capacity to make use of atmospheric nitrogen and incorporate it into organic
compounds, whereas wheat did not. In further trials of the same kind, peas
also appeared capable of “fixing” nitrogen, whereas oats did not. Wheat
and oats, presumably, normally obtained their nitrogen from nitrogenous
compounds present in the soil. These results provided an explanation for
the practical experience of farmers that cereal crops depleted the subsequent
fertility of soils, whereas crops of clover or of legumes, such as beans or
peas, increased it. Boussingault later confirmed, in a 5-year trial on his farm,
that cereal crops yielded more nitrogen after a legume crop.54

52. Aulie (1970), p. 438.

53. Boussingault (1838b).
54. Aulie (1970), p. 448.

36
 /

Nutrition in the Light of the New Chemistry

Table 2.3. Results of Boussingault’s first nitrogen balance experiments

with a cow and a horse (expressed per 24 hours)

Cow Horse

N N N N
Materials (%) (g) Materials (%) (g)

Inputs Hay, 6.32 kg 2.4 152 Hay, 6.46 kg 1.5 97

Potatoes, 1.2 50 Oats, 1.93 kg 2.2 42
4.17 kg dry wt
Total in 202 139

Outputs Milk 46 —
Urine 37 38
Dung 92 78
Total out 175 116

Apparent
balance + 27 + 23

Source: Boussingault (1839a,b).

The First Nitrogen Balance Trials with Animals

Boussingault’s next experiments were designed to investigate whether her¬

bivorous animals obtained all the nitrogen they needed from their food. As
discussed earlier, short-term respiration experiments had not given a con¬
sistent answer as to whether respiration involved the absorption of nitrogen,
and the herbivores were thought most likely to be in need of such an ability.
In 1839 Boussingault reported work on both a cow5S and a horse.s
He explained in his papers that he chose levels of feed for each animal
that kept it at constant weight. Under such conditions he assumed that an
animal would require as much of each element every day as it lost - either
by excretion or by the secretion of milk. If the output of nitrogen were more
than the amount obtained from the food, it would have to be assumed that
additional nitrogen was being obtained from the air; otherwise not.
The animals were kept in stalls and fed for at least a month on the same
rations before the 3-day period during which their outputs were collected
and analyzed. The cow yielded just over 8 liters of milk per day. The results
for nitrogen are summarized in Table 2.3 We see that for each animal the
food eaten apparently contained more nitrogen than left the animal each
day either as a product (milk) or in its excreta. Boussingault concluded that,
9

55. Boussingault (1839a).

56. Boussingault (1839b).

37
 Protein and Energy

Table 2.4. Chemical cycles in the plant and animal kingdoms

Action of Action of
Materials plants animals

Neutral nitrogenous compounds,

fats, sugars, starches, gum Produce Consume
Oxygen Release Absorb.
Carbonic acid, water Take up Produce
Heat Absorb Produce
%

Source: Dumas and Cahours (1842), p. 976.

because of the direction of the balance, it was extremely improbable that

these animals assimilated nitrogen from the atmosphere since their food
more than met their requirements. In fact, he believed that the results con¬
firmed other reports that animals exhaled nitrogen incessantly. He also
presented corresponding results for carbon, hydrogen and mineral balance
and expressed his deep conviction that such elementary analysis could lead
to rapid increases in our understanding of life itself.57 This >was the first of
many thousands of feeding experiments with either animals or human sub¬
jects in which the balance of nitrogen and other elements was to be studied.
With hindsight, we can judge that the large positive balances in animals
not gaining weight were an artifact. Boussingault’s analyses for the feeds
and the milk were in good agreement with modern valufes, but he kept
samples of the excreta in a hot oven to evaporate them to dr^iess before
beginning their analysis, and this must have led to a loss of some of their
nitrogen in volatile ammoniacal compounds. In the following period, an¬
other method was developed in which fluids could be analyzed directly for
their nitrogen content. In a rather similar experiment carried out 50 years
later on a horse fed hay and oats, more than 60% of the nitrogen fed was
recovered in the urine, and the animal was calculated to be in almost exact
nitrogen balance.s8

The Grand Cycles of Nature

Boussingault’s quantitative work with both plants and animals confirmed

the general concept that had been developing among French scientists since
the 1780s. This was that the plant kingdom alone was capable of building
up complex organic materials from sirqple starting compounds and pure

57. Boussingault (1839a), pp. 118, 115.

58. Atwater and Langworthy (1898), p. 357.

38
 Nutrition in the Light of the New Chemistry

elements. The animal kingdom provided the other side of the cycle, breaking
down complex materials, largely by way of oxidation, into carbonic acid,
water and urea or other relatively simple nitrogenous compounds.
In 1842 Dumas and Cahours put this concept into the form of a table
(see Table 2.4). It was realized that plants could use simple ammonium salts
as a source of nitrogen. Why the legumes, but not cereals, could apparently
also utilize atmospheric elemental nitrogen was unknown.

Summing Up

The advances in chemistry in this period made it possible to study nutrition

in a quantitative way. Lavoisier has been called “the father of nutrition
science.” His first contribution was his recognition of the distinction between
compounds that could change their character and “simple substances” (or
elements), carbon, hydrogen, nitrogen, oxygen and others, which could not
either appear or disappear. Therefore, once accurate methods of analysis
had been developed, the metabolism of foods could be followed by means
of balance studies of the elements carbon and nitrogen. Lavoisier’s second
basic contribution, in this context, was his recognition that combustion and
respiration involved basically similar processes of oxidation, which could
explain the phenomenon of “animal heat.”
Magendie’s and Boussingault’s work indicated that animals were unable
to use atmospheric nitrogen for the synthesis of the organic nitrogen com¬
pounds that, apart from water, made up the bulk of their tissues. Animals,
and humans, were therefore ultimately dependent on plants for these. This
led Boussingault to suggest that the nutritional value of different plant foods
would prove to be proportional to their nitrogen content.
It required developments in atomic theory, to be described in the following
chapter, for a further understanding of the nature of even the simple nitrog¬
enous compounds.

«•

39
3 Protein Discovered and Enthroned,
1838—1845

The preceding chapter described the immediate implications of

the advances in chemistry at the end of the eighteenth century, as well as
some of the nutritional experiments that followed from them, dealing with
the nitrogen balance of the animal kingdom. The work to be described in
the present chapter depended, in turn, on the further chemical advances
made at the beginning of the nineteenth century. The basis for these was
John Dalton’s New System of Chemical Philosophy, which contained the
idea that the smallest units of particular compounds consisted typically of
one or two atoms of each of the elements present.
Dalton was a largely self-taught Quaker schoolmaster in the north of
England. His evidence for this “new system” was, first that every compound
seemed to have exactly fixed proportions, by weight, of each element that
it contained. Second, when there was more than one compoundAxmtaining
just two elements, then for a given weight of one element, the weights of
the other element in the two compounds were in a simple proportion, such
as 2:1 or 3:2. To use modern terminology, carbon monoxide has 133 g
oxygen combined with 100 g carbon, while carbon dioxide has exactly
twice as much oxygen, that is, 266 g per 100 g carbon. This relationship
was called the law of multiple proportions.1
Dalton started with the assumption that, unless there was evidence to the
contrary, the smallest unit of simple binary (i.e., two-element) compounds
consisted of one atom of each element. Thus water consisted of one hydrogen
and one oxygen atom. From the observations of others that gases combined
in simple volume ratios and Avogadro’s hypothesis that, in the gaseous
state, each molecule occupied the same volume, modified formulas were
developed. Since 2 volumes of hydrogen and only 1 of oxygen gave 2
volumes of water vapor, it seemed that the oxygen “unit” must split in two
or, in other words, that the molecule consisted of two oxygen atoms (given

1. Dalton (1808); Meyer (1891), pp. 177-81.

40
 Protein Discovered and Enthroned

the symbol O); expressed in the modern convention, it is Or Again 2

volumes of ammonia were formed from i volume of nitrogen (N) and 3 of
hydrogen (H), so that nitrogen gas also apparently existed as a double
molecule (NJ. With hindsight it would seem to follow that free hydrogen
was also a doublet, to allow for a formula NH3. However, practice was
inconsistent and some chemists still thought of water as H—O rather than
HzO. This meant that, if the relative atomic weight of hydrogen was taken
as 1, that of oxygen was 8, since the ratio of weights of oxygen to hydrogen
in water is 8:1 and other atomic weights were also one-half of their modern
values, with allowance for errors in the earlier determinations. Thus carbon
(C) was 6 and nitrogen was thought at first to have either one-half (or even
one-third) of the modern value of 14.z

The “Dark Forest” of Organic Chemistry

There was doubt, at the beginning, whether the law of fixed proportions,
which applied to minerals, could be extended to the multitude of naturally
occurring organic materials, that is, those containing carbon, hydrogen and
oxygen or in some cases nitrogen sulfur (S), and phosphorus (P) in addition.
But it began to be discovered that the simpler, crystallizable materials,
particularly organic acids and sugars, gave proportional compositions that
corresponded to relatively small whole numbers of atoms.3 The Swedish
chemist Berzelius played a leading role in the work on the composition of
organic materials both with his own analyses and by collecting new pub
lished information in chemistry in a long series of yearbooks from 1822 to
1848 4
One crystallizable organic compound was urea, which had been obtained
from urine by Rouelle as early as 1773 and had been found to give off half
its weight as “volatile alkali” on being heated.5 With the discovery of ni¬
trogen, it was realized that this was the major nitrogen-containing excretion
product in humans and other mammals. In 1819 Prout published an ex¬
traordinarily accurate analysis of it and concluded that it had the molecular
formula CzH4NzOz. With modern atomic weights, this becomes CH4NzO.
In 1828 Friedrich Wohler found that treating lead cyanate with ammonium
hydroxide gave a white precipitate that also consisted of urea. This was a
surprising result, because both cyanic acid and ammonia could in turn be
synthesized from their elements, and urea was a product of nitrogen me-

z. Ladenburg (1900), pp. 47~66; Ihde (I964)> PP- 95-I23> Mo-154-

3. Berzelius (1814).
4. Partington (=1964), Vol. 4, pp. 147-%-
5. Rouelle (i773c)-
6. Prout (1819), p. 376-

41
 Protein and Energy

tabolism in the living body.7 For the first time, therefore, a piece of “vital”
chemistry, the synthesis of urea, could be reproduced in the laboratory.
The general view in this period was that the production of an organic
molecule, whether in a plant or an animal, involved the use or even incor¬
poration of a life force and that it would therefore be impossible to reproduce
in a laboratory. This concept, which differed in detail from one school to
another, had the general name of “vitalism.”8 Urea was, of course, only a
final breakdown product of living tissues and Wohler’s work certainly did
not constitute a refutation of vitalism, but it was still a surprise to scientists
that he had been able to produce a chemical that until then had been known
only as a product of the kidney.
We must introduce two more developments in organic chemistry before
we can get to their nutritional implications. The first was the concept of
“isomerism.” This was based on the finding that two compounds with quite
different properties could have the same elementary composition. Thus silver
cyanate and silver fulminate both contained one atom each of silver, carbon,
oxygen and nitrogen and were said to be “isomers.” Again, Lavoisier had
attributed the acidity of acetic acid to its high proportion of oxygen, but
cellulose was found to have almost identical proportions of carbon, hydro¬
gen and oxygen though with no trace of acidity. The only explanation for
such differences was that the atoms were combined in different ways in the
different isomers and that knowing the overall proportions of elements was
not sufficient to predict the properties of compounds.9
The second discovery was that in organic chemistry some atomic group¬
ings, or “radicals,” seemed to persist and to behave almost' as if they were
elements in inorganic chemistry. One was the gas cyanogen (formed from
carbon and nitrogen), which formed an acid and salts and reacted with
organic compounds in a manner similar to chlorine.10 Another was the
benzoyl radical. Distilling bitter almonds with water yielded an oil that on
exposure to air formed crystals of benzoic acid, which had already been
obtained from resins of the benzoin laurel tree. The oil formed more com¬
pounds with chlorine or bromine and also a cyanide derivative. But benzoic
acid was re-formed from all these on reaction with water. It appeared that
the “benzoyl” group (with the formula C7HsO) was a stable radical, though
if the compounds were ignited in oxygen they were fully oxidized to carbon
dioxide and water. This work, by Wohler and Liebig, was hailed by Berzelius
as beginning a new day in vegetable chemistry and “opening a way into a
dark forest.”11

7- Wohler (i8z8); Ihde (1964), pp. 164-5.

8. McKie (1944); Toulmin and Goodfield (1962^, pp. 322-6; Teich (1965), 44-5.
9. Partington (1964), Vol. 4, pp. 256-7.
10. Ibid., pp. 253—4.

11. Vickery and Osborne (1928), p. 394; Holmes (1973), p. 333.

42
 Protein Discovered and Enthroned

The Protein Radical

This preliminary review brings us up to the year 1838, where the preceding
chapter broke off, and sets the scene for a famous paper published by Gerrit
Mulder, a Dutch physician who had taught himself chemical analysis.11 He
reported analyses for egg albumin, serum albumin, fibrin and wheat gluten
and concluded that they contained an identical common radical. The mol¬
ecules of the individual products differed only because they were combined
with different numbers of sulfur atoms and, in some cases, phosphorus
atoms. He said that these extraneous atoms could be removed from the
naturally occurring nitrogenous materials by treatment with potassium hy¬
droxide. The radical could be precipitated by adding acetic acid. He assigned
it a formula of C4OH6lNIO0I2 on the basis of the reaction product obtained
from treating egg albumin with strong, cold sulfuric acid. He assumed that
one molecule of the acid had combined with one molecule of the radical,
later given the symbol Pr. However, in both lead and silver salts of albumin,
the proportion of the metals was much lower, so that one atom was ap¬
parently combined with 10 molecules.13 When purified serum or egg albumin
or fibrin was analyzed, each was also found to contain sulfur and phosphorus
at such low levels that again it appeared that the formulas were as follows:

Serum fibrin and egg albumin: Prlo-SP

Serum albumin: PrIO-SJP

Mulder sent a summary of his findings to Berzelius, who gave an enthu¬

siastic reply and suggested a name for the new radical:

The name protein that I am proposing for the organic oxide of fibrin and albumin
I chose to derive from the Greek word irpcoTeob (proteios), because it appears to be
the fundamental or primary substance of animal nutrition which plants prepare for
herbivores, and who in turn supply it to the carnivores. If one were to choose a
name based on the Greek word for fiber it would be less appropriate since the
organic oxide is also the base of albumin [in addition to fibrin] and probably of the
coloring matter [hemoglobin] as well as of still other materials.

The ancient Greeks had a god, Proteus, who had the power to change
himself into different shapes. Mulder followed Berzelius’s suggestions for
the name of his new radical, and expressed his conclusions as follows: .

It seems certain that animals obtain their basic materials from the plant kingdom.
It is possible that plant albumins have proportions of sulfur and phosphorus that
differ from those in animal albumins, but the ternary [four-element] organic body
is “protein” in each case.... It is not yet known whether starch and other substances

xz. Brouwer (1952), p. 4; Glas (1975), pp. 291-9.

13. Mulder (1839), p. 138; Snelders (1982), pp. 200-1.
14. Vickery (1950), p. 389.

43
 Protein and Energy

which have been found to be of nutritive value, can also be converted into protein
in the animal body.IS
(
>. 1
Joining the Bandwagon

The idea was well received by other chemists. In 1841, in the Annalen der
Chemie und Pbarmacie, of which he was the editor, Justus Liebig compared
the analyses of nitrogenous materials extracted from plants with analyses
reported from Mulder’s laboratory for serum albumin, casein and fibrin.
He concluded that they all contained a ratio of exactly 8 equivalents of
carbon to 1 of nitrogen.16 (Because of the continuing controversy over
atomic weights, Liebig had reverted at this time to “equivalent weights”
and was using 6 for carbon and 14 for nitrogen.) Therefore, with modern
atomic weights of 12 and 14, respectively, this corresponded to a ratio of
4 atoms of carbon to 1 of nitrogen, which was in agreement with Mulder.
He also said that when “the casein of sweet almonds or legume grains” was
suspended in an alkaline solution (potassium hydroxide) with gentle heat
and then acidified, there was an abundant production of hydrogen sulfide
and a precipitate of protein.17
Liebig recorded in his paper some general thoughts: “The similarity of
the vegetable nitrogenous compounds to the essential parts' of the blood is
not confined to their chemical composition; their chemical reactions are
also the same.” Later he added that the inorganic elements adhering to
them, magnesium, calcium, iron, sulfur and others, were also the same. “A
carnivore eats itself, from a chemical point of view, because its nourishment
is identical with the component parts of its body; but a herbiv&e does the
same, because its foods are also identical with its flesh or its blood.”18
Elsewhere he wrote: “It may be laid down as a law, founded on experience,
that vegetables produce, in their organism, compounds of protein; and that
out of these compounds of protein the various tissues and parts of the animal
body are developed by the vital force, with the aid of the oxygen of the
atmosphere and of the elements of water.”19
In 1842 Dumas and Cahours, working in Paris, reported that they had
followed Mulder’s procedure for the preparation of protein from both casein
and serum albumin. In each case the products had a composition that
corresponded best (after conversion to modern atomic weights) to the for¬
mula C48H74NI2Ois, which again has a ratio of exactly 4 atoms of carbon

15. Mulder (1839), p. 140.

16. Liebig (1841), pp. 145-9.
17. Ibid., p. 152.
18. Ibid., pp. 151, 154.

19. Liebig (1842a), p. 102.

20. Dumas and Cahours (1842), pp. 985-6.

44
 Protein Discovered and Enthroned

to i of nitrogen. However, the French workers differed from Liebig in finding

that the nitrogenous material, “legumin,” that could be extracted from peas
and beans, and that Liebig called “vegetable casein,” contained only 3.25
atoms of carbon per atom of nitrogen. They added that, although legumin
therefore had a chemical composition closer to that of gelatin, in view of
the proven nutritional value of legumes (in contrast to the apparent worth¬
lessness of gelatin), it must really be a compound of albumin with some
other product in fixed proportions/1
Dumas and Cahours also discussed the metabolism of protein in the
body/1 An adult man consumed each day 100 to 125 g “neutral nitrogenous
material,” containing 16 to 21 g nitrogen. Almost the same quantity of
nitrogen was recovered in the urine each day in the form of urea, which
could be regarded as an oxidation product of albumin or casein. Ignoring
intermediate steps, the overall reaction could be represented (after con¬
verting to modern atomic weights) as
C „H N O +100 “O” -* 6 CH.NiO (urea) + 42 COz (carbonic acid)
+ 25 FLO (water)

From an average daily intake of protein, 50 g carbon and 6 g hydrogen

were not recovered in urea, but were combusted to carbonic acid and water,
respectively. Since 1 g carbon had been found to give 7.3 kilocalories (kcal)
on combustion and hydrogen gave 35 kcal/g, the total daily heat from the
combustion amounted to 575 kcal. However, it was known that a man
produced 2,500 to 3,000 kcal of heat each day and that this sufficed to
maintain his body temperature. With only 575 kcal, he would die of cold.
Therefore, he needed to burn another 240 g fat per day, or its caloric
equivalent (2,200 kcal) in starch or sugars. This explained a man s nutri
tional needs for non-nitrogenous organic materials in addition to the nitro¬
genous ones. (1 kcal is the amount of heat needed to raise the temperature
of 1 kilogram [kg] water by i°C.)
In the same year (1842), Liebig set out a series of 20 chemical equations
to explain the production of different materials in the body/3 Thus he
attempted to demonstrate that 2 molecules of protein could be broken down
to yield 1 of fat and 6 of urea. He also wrote that 1 molecule of gelatin
appeared to be made up of 2 molecules of protein, 3 of ammonia, 1 of
water and 1 of oxygen and drew the conclusion that by means of these
formulae we can trace the production of the different compounds from the
constituents of blood.”14
Berzelius was alarmed at these “speculations to describe the chemical

21. Ibid., pp. 992—4-

22. Ibid., pp. 998-9-
23. Liebig (1842a), p. 147.
24. Ibid., pp. izi-2.

45
 Protein and Energy

phenomena which occur in living bodies.... readers will be misled into

regarding probabilities as truths, and it will be the more difficult later on
to eradicate errors.”15 But despite this criticism of speculative chemical
physiology, it was generally accepted by 1842 that the different nitrogenous
compounds in animal and plant tissues for which there had been no unifying
name were all based on a common large unit, or radical, called “protein.”16
\

The Position of Justus Liebig

In this period Liebig was such an influential and controversial figure in the
fields of both animal and plant nutrition that it seems useful to review how
this came about. He was born in 1803 in Darmstadt (some 20 miles south
of Frankfurt, Germany). He was not very successful at school, where the
principal told him that he was “the plague of his teachers and the sorrow
of his parents.” However, he became interested in chemistry while helping
his father, who was a wholesaler, and to a small extent a manufacturer, of
dyestuffs. At 16, he managed to gain enrollment at the University of Bonn,
where there was a professor of chemistry. The chemistry taught there was
very theoretical, but after graduation at 19 Liebig obtained a grant from
the Grand Duke of his area to study chemistry abroad. The choices open
to him were Berzelius’s modest laboratory in Sweden and the University of
Paris with a variety of laboratories in its neighborhood. Liebig chose the
latter and greatly enjoyed the French tradition of lecture demonstrations.
He was lucky enough to catch the attention of Gay-Lussac and to become
his personal assistant in the latter’s research laboratory Tor nearly two
years.17 Gay-Lussac was experienced in the analysis of organic Compounds
and interested in their fine structure as a result of observations of isomerism.
He was more concerned with the discovery of general laws than with the
proliferation of individual observations.18
At the early age of 21 Liebig was appointed assistant professor in the
small University of Giessen, some 30 miles north of Frankfurt. Here he set
up a laboratory, at first in an unused barracks, where his students could
work on individual research projects under his supervision (Figure 3.1). This
was the first facility of its kind anywhere, and Liebig soon attracted students
from many countries. Here he generated a previously unheard of rate of
experimentation, principally on the preparation and characterization of or¬
ganic compounds. Liebig was also extraordinarily active in teaching and
writing as well as in editing his own Annalen der ChemieA9

25. Berzelius (1842), pp. 535—6.

26. Fruton (1972), pp. 97—8.
27. Shenstone (1901), pp. n-19; Partington (1964), Vol. 4, pp. 294-5; Munday (1990).

28. Crosland (1978), pp. 87-136.

29. Browne (1942); Partington (1964), Vol. 4, pp. 296-7.

46
 Protein Discovered and Enthroned

Figure 3.1. “Liebig In His Laboratory at Giessen” as portrayed on an

advertising card issued with boxes of Liebig’s Meat Extract. (Dr. P. A.
Munday)

In 1837 Liebig was invited by a group of his old pupils to England, and
gave a lecture to the British Association for the Advancement of Science.
In his talk he stressed the practical applications of organic chemistry, par¬
ticularly in connection with the agricultural advances in Britain and the
work in progress there, in both animal and plant physiology. The association
expressed great interest in his ideas and asked him to prepare a report.30
This has frequently been taken as the stimulus that completely changed
Liebig’s interests and caused him to write Organic Chemistry in Its Relation
to Agriculture and Physiology, which appeared in 1840. However, a modern
scholar has argued that Liebig had always been interested in the practical
applications of chemistry and that because of his fascination with the met¬
amorphosis of materials, combined with the new information linking the
components of plants and animals, it was a natural progression to consider
the synthetic powers of plants.31
Some, like Berzelius, thought that plants had only a limited ability to
produce living material out of purely inorganic substances and that it was
therefore the soil’s complex organic materials - that is, the humus fraction

30. Rossiter (1975), pp. 2.7-8; Brock and Stark (1990), pp. 134-5.
31. Munday (1991).

47
 Protein and Energy

from decayed plant residues - that was important for new crop growth.
Liebig did not see any fundamental distinction between inorganic and or¬
ganic chemistry, nor did he feel that a plant’s power of synthesis was limited
in this way. He believed that what plants needed from the soil was water,
minerals and ammonia - the last being supplied by rainwater, in which it
naturally dissolved. Fertilizers would increase crop growth if they supplied
minerals in short supply in soils. These ideas were set out in his 1840 volume.
The direct, authoritative tone of the book helped to make it a “best-seller.”
It was translated into many languages and was particularly influential in
the United States.31
Liebig’s ideas led him to take a financial interest in the British prqduction
of “patent fertilizers” carrying his name. They were apparently marketed
without prior testing because of his certainty that they would be effective.
In fact, they were a fiasco, because they were too insoluble to have any
worthwhile effect in the short term.33 The subject of fertilizers is not directly
related to this book, but it shows the range of Liebig’s activities and illus¬
trates his willingness both to plunge fearlessly into a new field and to set
out dogmatic opinions that were not always correct.

Protein: The Only True Nutrient

After completing his book on plant nutrition in 1840, Liebig turned his
energies to the subject of animal (and human) nutrition, although he had
never carried out a feeding trial or any kind of metabolism experiment with
animals, and only one rough balance trial on a group of soldiers. He was
spurred on, it seems, by his rivalry with Dumas and Boussing^ult in Paris
and his fear that they would publish and gain priority for ideas that he was
already introducing in his lectures. The first edition of Animal Chemistry
was published in 184Z and was immediately translated into English. It has
already been the subject of a thorough study.34 I will consider it only in
relation to its direct implications for the science of nutrition. The main
points of Liebig’s argument can be compressed into the following statements
(using my words rather than his):
1. The bodies of stall-fed domestic animals may contain fat, but it is not
an essential component since animals in their natural, wild state do not
accumulate fat. In nature it is found only as a relatively small component
of an animal’s brain and nerves.35
2. The dry matter of the lean functional tissues contains approximately

32. Liebig (1840).

33. Rossiter (1975), p. 445 Unlay (1991), pp. 1^7-8.
34. Holmes (1964),
35. Liebig (1842a), pp. 80, 85-6.

48
 Protein Discovered and Enthroned

17% nitrogen. This is also the nitrogen content of the two principal com¬
ponents of the blood plasma (i.e., the blood without its globules, which are
assumed to be oxygen carriers and to play no part in nutrition). These
components are albumin and fibrin, which differ in some properties and
therefore have a different arrangement of atoms. However, they have the
same chemical composition, and it has been demonstrated that they can be
converted to one another. The conversion of the constituents of blood into
muscle fiber (and vice versa) involves, again, only a change of form. Muscle
fiber, and in fact all organized tissues, must of course be formed from the
blood that bathes them.36
3. The only true nutrients are therefore those that can be converted to
blood. In carnivores the process of nutrition is easily understood. The ni¬
trogenous tissues of the meat consumed are solubilized by the digestive
juices, and if necessary by a type of fermentation that gives the rearrange¬
ments needed for their conversion to soluble albumins. These are then ab¬
sorbed directly and become part of the blood.37
4. The food of grain- and grass-eating animals is apparently very different,
but it too contains vegetable fibrin, albumin and casein, with a composition
identical to the animal components, and they become integrated into the
animals’ tissues exactly as in the carnivores. The other nitrogenous con¬
stituents of plant foods are rejected by the animals.38
5. The functions of the non-nitrogenous components that exist in larger
quantities in these animals’ foods will be discussed later, but it is certain
that they cannot be “animalized” by reaction with atmospheric nitrogen.39
6. The turnover of proteinaceous foods by adult animals, as shown by
the continued excretion of urea even when none of these materials is con¬
sumed, is explained by muscles consuming themselves when they exert their
muscular force. This force is released by the molecule breaking into two
fragments. One, rich in nitrogen, is broken down to form urea, together
with carbonic acid and water. The urea is excreted in the urine, and the
amount of mechanical force exerted by a man or animal in a particular
period is proportional to the quantity of urea excreted. It cannot be supposed
that the nitrogen of food can pass into the urine as urea without having
previously become part of an organized tissue.40
7. The second type of fragment formed during muscular exertion is car¬
bon-rich; it is also released into the blood and picked up by the liver,
converted to choleic acid and secreted into the bile. When the bile, in turn,

36. Ibid., pp. 39-41, Z2.0.

37. Ibid., pp. 43-4.
38. Ibid., pp. 44-7.
39. Ibid., pp. 71-2.
40. Ibid., pp. 231-3.

49
 Protein and Energy

is secreted into the gut, the choleic acid is largely reabsorbed and oxidized
to carbonic acid and water.41
8. The breakdown of muscle that occurs during the day is compensated
for by re-formation of tissues during sleep, and the “force” or vitality of
the muscles is regained. For an active adult, 7 hours of sleep are required.
An old man, who is necessarily less active, requires only 3r/4 hours. In each
case “waste is in equilibrium with supply.” However, in the infant who
sleeps 20 hours and is awake for only 4, there is an excess of supply, and
this explains the child’s ability to gain weight and grow.4*
9. Since only those substances that are capable of conversion to blood
can properly be called nutritious, or considered to be food, the ^protein
elements of foods are the only true nutrients, that is, the only ones capable
of forming or replacing active tissue. The gelatin of the skin and connective
tissue is not itself a form of albumin. It can be formed from it, but the
reverse process does not take place. Gelatin in the diet can thus spare the
small proportion of dietary protein otherwise needed for the production of
skin and connective tissue, but it cannot meet the larger need for replacement
of muscle fiber.43
10. The non-nitrogenous components of the diet also have a role, which
is to support the process of respiration. They fall into two main classes.
The first includes starch, gum and the sugars. Their chemiqal composition
can, in each case, be represented as so much carbon and so much water.
Also, on digestion or fermentation, starch can be converted to sugar. The
second class of materials comprises the fats, and these can also be formed
in the body from sugar when the latter is being ingested at a faster rate
than is needed to react with oxygen, that is, when there is a relative shortage
of oxygen.44
11. The oxygen of the air serves two vital functions in animals, but it
can also be damaging. Its first function is to take part in the digestive
reactions in the stomach, and it is carried there by saliva, whose mucus has
the special property of forming bubbles that trap air. The surplus nitrogen
in these bubbles is reexcreted through the skin. The second function of
oxygen is to take part in the reaction by which muscle fibrin decomposes,
as described earlier, with the release of mechanical force. The danger is that
too much oxygen may enter the circulation and cause excessive oxidation
and loss of protein tissue. This can be prevented only by non-nitrogenous
foods reacting preferentially with the excess oxygen. The secondary effect
of both types of oxidation is the production of necessary animal heat.45

41. Ibid., p. 58.

42. Ibid., pp. 2.36-7.
43. Ibid., pp. 38, 122-4.
44. Ibid., pp. 68-71, 78-81, 85-6, 90.
45. Ibid., pp. 2, 108, 209, 232.

50
 Protein Discovered and Enthroned

12. The special role of body fat in protecting the essential, nitrogenous
body tissues is clearly shown by the course of events during starvation when
foods are no longer present to protect the system from the oxygen of the
air. First, there is gradual oxidation of body fat. When this is exhausted,
the muscles become shrunken as the albuminous structure is gradually ox¬
idized. Finally, the brain and nerves are attacked and death ensues.46
13. The rate at which oxygen enters the body, and thus the rate at which
sugars and fats have to be used up to protect the tissues, depends on the
rate at which it enters the lungs. At higher atmospheric temperatures the
air is thinner and, where the climate dilutes the air with moisture, there is
in each case less oxygen in a single respiration. This explains why an En¬
glishman transferred to the hot, moist climate of Jamaica experiences a loss
of appetite; the carbon and hydrogen of his food are oxidized at a lower
rate. Fortunately, the foods most abundant in the tropics are watery fruits.
Conversely, someone living in the cold of the Arctic and inspiring more
concentrated air has a much greater appetite, and nature has arranged that
the foods there are concentrated and fatty. It is also part of nature’s ar¬
rangement that the higher rate of metabolism produces proportionately
more heat, which is of course required to maintain body temperature in
that climate.47
14. Someone who feels cold is induced to engage in physical activity. This
stimulates respiration, part of which is needed for the breakdown of muscle
fibers, but it admits more oxygen, which also results in more combustion
of the non-nitrogenous protectors and, thus, in more heat production. As
Lavoisier showed, physical activity results in greatly increased carbonic acid
production.48
15. For the carnivore who eats only lean meat, the oxygen can react only
with organized tissues. Thus lions and tigers housed in cages in a zoo are
in incessant motion so as to furnish the “material necessary for respiration.”
Similarly, a savage living only on meat is forced to make the most laborious
exertions.49
Obviously, in a book of more than 300 pages Liebig said a great deal
more, but this summary presents the principles of human and animal nu¬
trition as he saw them. Modern readers probably see the scheme as so
“upside-down” as to be painful. Some of its implications are seen in Lyon
Playfair’s paper delivered to the Royal Agricultural Society in Britain, im¬
mediately after the publication of Animal Chemistry. Playfair had worked,
as a student, in Liebig’s laboratory and had translated his earlier volume

46. Ibid., p. 25.

47. Ibid., pp. 15-17, 21-3.

48. Ibid., pp. 20, 85—6.

49. Ibid., p. 75.

51
 Protein and Energy

into English. Now he said that Liebig’s contribution had important practical
applications for animal breeders. Previously, animals had been selected for
well-developed lung capacity. However, now that it was understood that
allowing a larger quantity of oxygen into the body only resulted in a greater
oxidation and loss of sugars and fat, it was obviously desirable to select
narrow-chested animals for breeding, because these would take in less ox¬
ygen with each breath, and thus would burn less carbon and hydrogen, and
so leave more available for fattening.50 The conclusion was logical enough,
given Liebig’s premise, but what an extraordinary one it was, with its im¬
plication that we would all be better off half-choked, so that less oxygen
reached our lungs. *

The Reception of Animal Chemistry

The book greatly impressed many of the general reviewers in medical jour¬
nals. They commonly questioned particular statements, but the work as a
whole was thought to represent a step forward into a new era.
Liebig had derided physiologists as being in no position even to comment
until they had also made chemists of themselves. One of them wrote a whole
book in reply: The Inter-relationships between Physiology and Chemistry
as Illuminated by a Critique of Liebig’s “Animal Chemistry.*’sx The author
contrasted the typically meat-rich diet of the sedentary, well-off businessman
with the cheaper, high-carbohydrate diet of the day laborer who earned his
smaller income by the sweat of his brow. How could this be reconciled with
Liebig’s dogma of albuminoids being the only source of muscular energy
and of physical activity being the only way in which they coulc^be broken
down and eliminated? An alternative hypothesis, which fitted the facts bet¬
ter, was that the physically active person on a high-carbohydrate diet was
able to keep going because, as muscle tissue broke down, the nitrogen-rich
fragments recombined with carbohydrates to provide replacement muscle
material, so that there was no longer any clear differentiation of function
between albuminoids and respiratory foods.51 It had already been shown
that, as seemed inevitable, very large amounts of urea could be excreted by
someone on a diet rich in albuminoids without any increase in muscular
activity.53
J. R. Mayer, now remembered mainly as one of the principal independent
discoverers of the conservation of energy, also criticized Liebig for his in¬
consistencies and contradictions concerning the relationships between mus-

50. Playfair (1843), pp. 257-9.

51. Kohlrausch (1844).
52. Ibid., pp. 51—2.
53. Lehmann (1842), pp. 273-4.

52.
 Protein Discovered and Enthroned

cular work and heat production. The oxidation of carbon compounds must
be the source of chemical energy for muscular contractions, and this ap¬
peared partly in the form of mechanical energy and partly as heat. Using a
rough value for the mechanical equivalent of heat, as well as estimates of
the amount of mechanical work done by laborers and the extra food they
needed to eat as compared with men at rest, he calculated the mechanical
efficiency of the muscular “engine” as about 33%.54 However, Mayer was
still considered something of a crank at this time, and his fundamental
contribution was not to be appreciated for another io years.

The French Challenge to Liebig

Dumas and Boussingault agreed with Liebig as to the primary importance

of the nitrogen-containing components of food, but were angry with him
for having claimed priority for discovering the identity of plant and animal
materials — the albumins, caseins and so on. But, ironically, the claim in
Animal Chemistry that they chose to challenge was one of the few points
in Liebig’s scheme that a modern reader would consider to be correct.
The French workers’ summary of the relative roles of the plant and animal
kingdom in nature (Table 2.4) included the generalization that plants were
capable of “reduction” reactions — that is, those adding hydrogen or yielding
free oxygen - but that animals were capable only of oxidation, which used
oxygen and resulted in the breaking down of organic compounds. Now the
conversion of starch or sugar to fat came into the category of “reduction,”
as Liebig himself had said.55 Using one of his reaction schemes (converted
to modern atomic weights), it could be expressed as
(QTL.CXJ -* CfIHlo0 + CO, + 7 “O”
Starch unit “Fat”

Dumas and Boussingault believed that Liebig had ignored a general law
that they had discovered, and set out to prove him wrong. Some of their
own colleagues in the Academy of Sciences warned them against being so
dogmatic about the limited powers of synthesis by animals. Magendie
stressed that “nutrition remains one of the most obscure questions in sci¬
ence,” and another member wrote that it would seem to follow from their
law that snakes could secrete venom only when they had obtained it in their
food.s6
Boussingault and others in France were stimulated to do very careful
balance trials with successive carcass analyses, but finally had to admit, in

54. Mayer (1845), pp. 52, 57-8.

55. Liebig (1843a), pp. 86-7.
56. Holmes (1974), pp. 48-117.

53
 Protein and Energy

1845, that fat could be synthesized from starch or sugars. As another critic
was to put it some zo years later, “This decision in his [Liebig’s] favour
contributed greatly to the extension of his reputation.”57 The French sci¬
entists had lost the debate that had been waged on ground of their own
choosing. Dumas continued to have a distinguished career but withdrew
from the field of nutrition, and Boussingault concentrated his subsequent
work on crop husbandry and the problems of nitrogen fixation in the soil.5

Summing Up

By this period it had begun to be shown that even organic corrypounds

obeyed the “law of fixed proportions,” and atomic proportions had been
worked out for urea, benzoic acid and other compounds. Then in 1838
Gerrit Mulder claimed to have discovered the existence of a protein radical
with the formula C40H6JN10Oiz and that this combined with sulfur and
phosphorus to form albumin, fibrin and so on. Justus Leibig adopted this
concept and went further, concluding that the protein radical, a chemical
product of the plant kingdom, was the only true nutrient. It was therefore
the essential ingredient for both body building and physical activity. This
was an elegant concept and apparently was firmly based on recent devel¬
opments in both organic chemistry as well as a multitude of observations
from animal experiments. The broad conclusions were to survive in people’s
minds for many generations, but the scientific bases for them were to dis¬
integrate quite quickly, as we shall see in the next chapter.

57. Voit (1870), p. 371.

58. Kapoor (1971); McCosh (1984), pp. Z48-59. 4

54
4 Things Fall Apart, 1846—1875

There was little criticism from 1842 to 1846 of Liebig’s scheme

of Animal Chemistry outside a few German university circles. It appeared
to be a giant intellectual synthesis. And coupled with Mulder’s discovery
of “protein” as the common repeating unit in the molecules of the albu¬
minoids, as well as related nitrogenous compounds, it seemed to introduce
an attractive harmony into the principles of nutrition and to have explained
its basic principles. However, this scheme was not to last.

The Disappearance of “Protein”

As a result of further work in his laboratory, Liebig published a note in his

Annalen early in 1846 saying that he and his colleagues had been unable
to prepare sulfur-free protein from any of a variety of starting materials;
he ended by asking Mulder to give precise details of how he had accom¬
plished this and what tests he had conducted to check that his products
were truly free of sulfur.1
A detailed paper by Laskowski, a Polish visiting scientist in Liebig’s lab¬
oratory who had done the bulk of the work, appeared later that year. He
had started with egg albumin and treated it with potassium hydroxide in
solutions of different strengths and at different temperatures. This dissolved
the albumin. Following Mulder’s instructions he then acidified the solution,
which usually resulted in a precipitation of an albumin-like material. How¬
ever, these precipitates, when tested by drying and heating with solid po¬
tassium hydroxide, and then re-acidifying and heating again, all gave off
hydrogen sulfide (the gas with the characteristic smell of bad eggs); and
paper, moistened with lead acetate solution and held over the mixes, turned
through the formation of lead sulfide. If the original alkaline albumin so¬
lution was heated at its boiling point for a period before acidification, the

1. Liebig (1846a).

55
 Protein and Energy

albumin did lose much more sulfur as hydrogen sulfide, but then no pre¬
cipitate was formed - the albumin having presumably undergone a greater
degree of decomposition/
Laskowski went on to describe other experiments with casein, and also
presented analyses of egg albumin blood fibrin and other materials that did
not agree with Mulder’s models of these being a combination of io to 15
molecules of protein with 1 to 2 atoms each of sulfur and phosphorus.2 3 His
final conclusion was that there was no basis left for believing in the existence
of Mulder’s fundamental substance, “protein.”4
Mulder was enraged by Liebig’s condescending attitude and wrote urging
him to apologize publicly in the Annalen.5 When he received no reply, he
hurriedly published a booklet denouncing Liebig both for his science and
his character, defending the existence of “protein” and proposing a modified
method for its preparation.6 Mulder’s reply attracts some sympathy from
the reader. Liebig apparently considered himself to be in a superior position
that allowed him to sit in moral judgment over others. He had already called
Dumas a plagiarist who had been wasting his time in worthless speculations
and had described Laurent and Gerhardt, two scientists still honored for
their contributions to organic chemistry, as conceited, self-complacent cocks
strutting about on a dung hill.
Mulder added that his original paper, written in Dutch, gave full exper¬
imental details and that it was only an abbreviated translation, appearing
in a German journal, that omitted them. He also pointed out that Liebig
had himself referred to “protein” having been obtained in his laboratory
and, in a rebuttal to Dumas, had claimed a formula of C4gH74NIiOIS for
it, based on its supposed decomposition to choleic acid and jjric acid.7
However, Mulder’s revised claims still could not be confirmed by other
chemists. His ideas were therefore discarded, and English-peaking authors
went back to the term “albuminoids” when referring to the group of ni¬
trogenous materials as a whole.

Liebig’s Revised Views

The first edition of Animal Chemistry was, of course, prepared when Liebig
was a believer in Mulder’s “protein radical.” In the 1846 edition, after his
rejection of the concept, he omitted all mention of it and simply restated:
“It was established as a universal fact that the sulfurized and nitrogenized

2. Laskowski (1846), pp. 152—8.

3. Ibid., pp. 133-9.
4. Ibid., p. 165.

5. Snelders (1982), p. 212.

6. Mulder (1846).
7. Scherer (1841), p. 44; Liebig (1842b), p. 355.

56
 Things Fall Apart

constituents of vegetable foods have a composition identical with that of

the constituents of the blood.”8 He seemed not to consider the growing
knowledge of digestion as worthy of mention. Presumably he was still of
his former opinion that “in the digestive process the food only undergoes
a change in its state of cohesion without any other change of properties.”9
We will return to the subject in Chapter 7.
In 1847 Liebig took another sharp turn. In his Researches on the Chem¬
istry of Food, he began by lamenting how few securely based facts there
really were in animal chemistry. “This was because the subject had been in
the hands of adventurers.... The protein theory had no foundation and was
never regarded by those intimately acquainted with its chemical groundwork
as an expression of the knowledge of a given period.”10 This was going too
far. He had chosen, consciously or unconsciously, to forget his own enthu¬
siastic use of the concept in his earlier writings. And, in fact, in the very
next section he said that it was under the influence of Mulder’s authority
that he had come to believe that fibrin, casein and albumin all had exactly
the same chemical composition and to ignore results, from his own and
many other laboratories, that indicated the opposite.11
So he now withdrew what he had said only the preceding year to be
“established as a universal fact” and wrote that it was, after all, not necessary
for all the components of blood to have been already present in food. Even
if food contained three different compounds — one containing nitrogen,
another sulfur, and a third additional carbon, they could all react together
finally to form blood if the appropriate attractive forces preponderated.11
As has been pointed out already, this nullified all his former arguments that
carbohydrates and fats were purely respiratory and had no role in the
production of tissues.13 This may have come home to Liebig also, because
the promised Part II of his 1846 edition of Animal Chemistry never, in fact,
appeared.
For the next 20 years Liebig kept a relatively low profile on the subject
of nutrition, except for his proprietary “extract of meat” (Figure 4.1) and
“malt food for babies,” which we will discuss later in the chapter. He also
gave up supervising an experimental laboratory with his move to Munich
in 18 5 2. However, he never publicly withdrew his view that non-nitrogenous
foods were purely “respiratory” and that only albuminoids were truly nu¬
tritional.

8. Liebig (1846b), p. 135.

9. Liebig (1842a), p. 115.
10. Liebig (1847), pp. 1, 2, 19.
11. Ibid., p. 21.
12. Ibid., pp. 19-21.
13. Holmes (1964), p. lxxxix.

57
 Protein and Energy

^as dem leben Liebig's

F Bcrufung Uebsq'sdurchKonia Max.
11 mM

Ellgp

LIEBIG’S FLEISCH-EXTRACT.

Figure 4.1. “Liebig Is Summoned to Meet King Max” as portrayed on an

advertising card issued with boxes of Liebig’s Meat Extract. (Dr. P. A.
Munday)

Critical Comments

Liebig’s writings were widely circulated and rapidly translated into other
languages. His critics were not as widely read, but they could be biting,
particularly in the light of the changes in his ideas between successive pub¬
lications and his withdrawal of many of the hypothetical chemical equations
found in the first edition.
The following are typical comments from an English review:
A bold speculator will never be in want of earnest admirers and enthusiastic followers
... but his statements appear to us to require impartial examination at the hands
of men who have no hypotheses to support. To take one instance, he refers to the
consumption by people living in the cold of the far North of much blubber and
other kinds of highly concentrated fatty food. Liebig says this is absolutely necessary
to them to prevent the combustion of their bodies by the condensed state of the
oxygen in the cold air which they breathe. But the reindeer living in the same area
defy the Arctic oxygen while living on dry, farinaceous moss and other plant foods.
There is an entire want of proof that man’s food is necessarily regulated by the
variable proportion of oxygen contained in a given bulk of the atmosphere.14

14. Anonymous (1846).

58
 Things Fall Apart

In the subsequent decade, interest in metabolism and nutritional questions

was developing in the United States. In 1856 a 21-year-old medical student,
John C. Draper, published the short thesis that he had written for his medical
degree at the New York University Medical College. Its title was “Is Mus¬
cular Motion the Cause of the Production of Urea?” He said that he had
kept his diet constant, that on a day of rest he excreted 26.4 g urea and
that on two active days (one of them including a 13-mile walk) the excretion
was not significantly different.15 His father, John W. Draper, was president
of the college and had developed a new method of quantitative analysis for
urea a few years earlier.
The paper was certainly referred to by European workers. However,
neither of the Drapers extended his work, and another paper published in
the United States in the preceding year had given contradictory results. W.
A. Hammond, who was to be a controversial surgeon-general for the north¬
ern armies at the beginning of the Civil War, was interested in Liebig’s belief
that the first product from the breakdown of protein was uric acid and that
urea was produced from it in its turn. He reported that on a day of rest he
excreted 1.6 g uric acid and 31.6 g urea. On another day of strenuous
exercise he found only one-third as much uric acid in his urine, but nearly
twice as much urea.16 He too reported no further experiments. The United
States was not yet an important source of scientific or medical advances,
and since the two papers appearing from this country largely neutralized
each other, it is not surprising that little notice was taken of them.

The British Approach to Nutritional Questions

We have already referred to the work of one English scientist active in

physiological chemistry in the first half of the nineteenth century. This was
William Prout, who lived from 1785 to 1850 and received his training in
medicine at Edinburgh University. He was then attracted to organic chem¬
istry and to the atomic theory. In particular, he became convinced that there
must be a special significance to the fact that so many atomic weights (e.g.,
for carbon, oxygen and nitrogen) were whole-number multiples of the
atomic weight of hydrogen. He also demonstrated that the acid secreted
into the stomach was hydrochloric acid and is credited with having been
the first to divide the principal organic constituents of foods into the three
categories of fats, carbohydrates and albuminoids. He was a religious man
and argued that God would not have included members of all three groups
in milk if they did not each have an essential nutritive function.17

15. Draper (1856).

16. Hammond (1855).
17. Brock (1975).

59
 Protein and Energy

Prout’s approach was very much that of the basic scientist-philosopher,

but the typical approach to nutritional problems in Britain at this period
was more applied. First there was a continuing interest in England in ap¬
plying scientific knowledge to the improvement of farm animal production,
through both selective breeding and efficient husbandry. Lawes and Gilbert,
who had already publicly disagreed with Liebig over the value of nitrogenous
fertilizers, carried out detailed animal feeding studies at the Rothamsted
Experiment Station in the 1850s. From their work on the growth of sheep
and pigs on different diets, they concluded that the importance attached to
the nitrogenous fraction of feedstuffs by both Boussingault and Liebig had
been greatly exaggerated. They commented further that they had beep “un¬
able to discern a sufficient basis of facts” for the evaluation of human diets
in terms of their nitrogen content. It appeared to them that people doing
hard physical work instinctively increased their intake of fat and other non-
nitrogenous materials.18 In addition, they pointed out that their detailed
studies of the composition of carcasses of animals killed in the “fat” con¬
dition desired by the market showed that they actually had a lower ratio
of nitrogen to carbon than was present in bread. Fat meat could not therefore
be considered to be primarily a source of albuminoids. They suggested that
the proven value of meat as a nutritional supplement to bread must therefore
have some other, still unrecognized basis.19
A second practical problem in Britain was the feeding of people who were
the responsibility of the state. Members of Parliament, and others interested
in social problems, were concerned that prisoners and paupers in institutions
should receive food adequate to support health but not as “luxurious” as
the food that even the poorest independent working family could afford to
provide for itself.10 It was generally agreed that standards were required for
both quantity and quality.
The general practice was for prisoners to receive only bread and gruel
for the first 14 days of imprisonment. Gruel at that time meant oatmeal
boiled with water, or sometimes with a proportion of skimmed milk.11 For
the next 6 weeks, prisoners received meat and potatoes twice a week, and
after that four times a week, with soup on the remaining days. Some critics
said that it seemed irrational that those who had committed the most serious
crimes, and therefore had the longest sentences, were fed the best. But prison
doctors and administrators countered that continuing the more restricted
diet would result in wasting and ill health, so that men would no longer be
able to carry out the hard labor that was part of their sentence.11

18. Lawes (1853), pp. 539—40.

19. Lawes and Gilbert (1859), pp. 574-8.
20. Drummond and Wilbraham (1957), pp. 363-72; Tomlinson (1978); Johnston (1985).
21. Mayhew and Binny (1862), pp. 347—8.
22. Ibid., pp. 349—51.

60
 Things Fall Apart

Scurvy as a Symptom of Protein Deficiency

The supply of bread, potatoes and other vegetables varied slightly according
to what was available at particular times. It had been concluded by 1843
that the numerous outbreaks of scurvy in English prisons in the previous
20 years had been associated with the omission of potatoes from the daily
diet even when the total intake of albuminoids was normal/3 A textbook
of nutrition published in that year criticized Liebig’s assertion in the first
edition of his Animal Chemistry that albuminoids were the only true nu¬
trients; it seemed clear also from the experience of victualing in the navy
that nutritional requirements were more complex/4
In 1845 and 1846 northern Europe was hit by a devastating blight that
destroyed the greater part of the potato crop. In Ireland, where the potato
had become the major staple crop, the result was a terrible famine. In
England and Scotland, where grains were still the major food crops and
potatoes secondary, the significance of the antiscorbutic value of potatoes
had not been appreciated, and there were widely distributed outbreaks of
scurvy as a result of the blight/5 A major outbreak at a prison in Scotland
was investigated by Dr. Christison of Edinburgh University, remembered
now for his work showing the danger of poisoning from lead pipes where
the water supply was “soft.”16
Christison had been convinced by Liebig’s thesis that nitrogenous com¬
pounds were the only true nutrients. Yet the suspect prison diet supplied
135 g per day of albuminoids, an ample amount by either contemporary
or modern standards. However, all but 23 g were in the form of gluten
from bread. He therefore concluded that there were limitations to the extent
that gluten could replace (or be converted to) albumin and fibrin. This was
a small modification of the Liebig “paradigm.” He explained that the suc¬
cesses reported from the use of “succulent vegetables” to treat scurvy were
due to their content of “vegetable albumen.”17
The argument was quickly refuted. As a physician from Glasgow pointed
out, 4 oz. of lemon juice per day, which was known to prevent scurvy,
contained only 2.5 g dry matter, and very little even of this small quantity
was nitrogenous/8
In 1853 Lyon Playfair, who has already been referred to as a British
protege of Liebig, gave a lecture entitled “The Lood of Man under
Different Conditions of Age and Employment.” He reported extensive

23. Baly (1843); Carpenter (1986a), pp. 99-102.

24. Pereira (1843), p. 47.
25. Carpenter (1986a), pp. 101—3.
26. Anonymous (1887b).
27. Christison (1847), pp. 888-90.
28. Anderson (1847), pp. 177-8.

61
 Protein and Energy

calculations of the carbon and nitrogen contents of the diets furnished

to paupers, prisoners and servicemen, and compared them with the diets
of agricultural laborers. The average daily diet for paupers contained 14
g nitrogen and 230 g carbon, equivalent to approximately 88 g protein
and 2,300 kcal per day in modern terms. Playfair said that no positive
answer could yet be given to the question of how much “flesh-forming
matter” (albuminoid) was required to support an adult man under normal
conditions. It was determined by the amount of labor performed. He
thought it appropriate therefore that prisoners sentenced to hard labor
received a proportionally greater increase in “nitrogenous materials” than
in fats or carbohydrates/9 4

The Work of Edward Smith

Another physician interested in nutritional questions during this period was

Edward Smith.30 In 1863 he had made the first survey, for the government,
of the food eaten by low-income families.31 Some of the results are sum¬
marized in Table 4.1 together with the calculated nutrient values of the
diets, made more recently by a group at London University, and comparisons
with the results of later surveys.31
Bread formed an extraordinarily large part of the ordinary diet in the
1860s. Even so, it had greater variety than the short-term prison diet, and
Smith had argued that even short-term prisoners should be given a more
substantial diet. Deterioration was admittedly not so marked in the short
term, but the state should endeavor to return a prisoner to society in a
strong rather than a weak condition, so that he could better compete for a
job of honest labor and be less tempted to return to crime.33 *
Smith was also concerned that the “hard labor” that formed part of long
sentences was unduly arduous.34 The standard form of hard labor in London
prisons consisted of continually climbing up the paddles of a large treadmill,
rather like the side paddle wheels of a Victorian steamship. A row of them
were linked together on the same axle and connected to a sail on the roof
of the prison to provide resistance (Figure 4.2).35 With the weight of the
prisoner on a paddle, it fell, and he had to step up to the next one. He
climbed in this way for 15 minutes, then rested for the same period while

29. Playfair (1853).

30. Chapman (1967); Barker, Oddy and Yudkin (1970); Carpenter (1991).
31. Smith (1864), p. 217.
32. Barker, Oddy, and Yudkin (1970), 35-49; U.S. Department of Agriculture (1983); U.S.
Department of Agriculture (1985).
33. Smith (1857a).
34. Smith (1857b).
35. Mayhew and Binny (1862), pp. 303-7; Chapman (1967), pp. 5-9.

62
 Things Fall Apart

Table 4.1. Mean results from Smith’s 1863 survey of the diets of low-
income indoor workers and their families compared with results from
later surveys of relatively low income families

U.K. u.s.
1863 1933 1965 1977

Basic Bread 145.6 65.6 40.0 25.6

Victorian Potatoes 38.4 54.4 57.6 24.0
foods, per > Sugar 7.9 22.8 16.9 18.2
adult per Fats 4.7 10.2 8.3 11.5
week, oz. Meats 12.3 23.0 23.0 68.8
Milk 16.0 36.0 80.0 90.0

Estimated Protein (g) 55 63 58 101

nutrient Fat (g) 53 72 81 138
intakes, per > Carbohydrates (g) 370 350 290 311
adult per Total energy (kcal) 2,190 2,320 2,040 2,880
day Calcium (mg) 360 370 820 1,030
> 10.8
Iron (mg) 12.5“ 8.0fc 15.7

“During this period, significant additional iron was obtained from iron cooking pots
and pans.
fcDuring this period, there was some changeover from iron to aluminum cooking
vessels.

another prisoner took his place. The standard day’s labor consisted of 15
stints and corresponded to a climb of approximately 1.4 miles. Each prisoner
did this 3 days per week, and the same food ration was provided each day
regardless of the work done.
Smith set out to measure how arduous the work was by doing it himself
while connected to an apparatus that he had developed for measuring the
output of carbon dioxide from his lungs as the increase in weight of po¬
tassium hydroxide onto which it was absorbed, after the air had first been
dried (Figure 4.3). He found that, on average, in the combined “15-minute
work + 15-minute recovery period” he expired 19.6 g carbon per hour
more than when resting. Since his z4-hour expiration when leading a restful
life was altogether 223 g, the labor of a standard workday, that is, of 7 Vi
hours “work + recovery,” at this rate would have increased his estimated
24-hour output by 66%.36
Smith organized the collection of urine from a group of four prisoners
for a 3-week period and analyzed it for the quantity of urea produced during

36. Smith (1859), pp. 692, 709-11.

63
 Protein and Energy

Figure 4.2. Prisoners sentenced to hard labor at work on the treadmill at

Brixton prison, London (British Register, August 1, 1823). (Watkinson
Library, Trinity College, Hartford, Connecticut)

each day plus the following night. The overall average was 15.5 f nitrogen
excreted in the form of urea. There was variation from day to day, but the
average production on workdays was only 3% higher than on rest days.37
This small, and not statistically significant, difference was in contrast to the
very large increase in the carbon dioxide production of a man engaged in
labor. Smith concluded that urea excretion was determined chiefly by the
nitrogen content of the food and that the production of carbon dioxide was
“the best measure we have of the vital functions attending muscular exer¬
tion.”38

An Elegant Experiment in the Swiss Alps

Edward Smith did so much work, with thousands of analyses stretching

over a period of 5 years, but he published it in such a way that some of

37. Smith (1862), p. 827.

38. Ibid., pp. 804-9, 831.

64
 Things Fall Apart

Figure 4.3. Edward Smith demonstrating the apparatus that he used to

measure his excretion of carbon dioxide when working on the Brixton
treadmill. {Phil. Trans. Roy. Soc. London, 1859)

the crucial points in his findings, which we have just described, were
buried. However, two scientists in Switzerland saw the implications of
Smith’s work and, tying it in with a recent development in physics, were
able in 1865, with no more than 4 days of experimental work and no
subjects other than themselves, to write a paper that became one of the
classics in the history of nutritional science.39 Slightly the older of the
two, at 35, was Adolf Fick, who had received his medical education in
Germany and was now a professor of physiology at the University of
Zurich. He had a long-standing interest in studying biological phenomena
in terms of physics. His colleague, Johannes Wislicenus, was a 30-year-
old associate professor in chemistry.40
Their experiment was designed to ask just one question: Does the amount
of body substance broken down, and converted in part to urea, always
provide enough energy to account for the energy expended in muscular
exertion? It was Liebig’s generalization that this was the only source of
muscular energy, so that one well-documented exception should be enough
to disprove it.
Fick and Wislicenus decided that, by ascending a mountain, they could

39. Fick and Wislicenus (1866); Coleman (1977), PP- i34~5-

40. Rothschuh (1971); Costa (i976)-

65
 Protein and Energy

establish a definite minimum amount of work, that is, by lifting their body
weight through a definite height. There was a convenient mountain, the
Faulhorn, on which they could ascend some 2,000 meters (m) by a steep
path and rest overnight in a hotel at the top. Also, if they refrained from
eating any albuminoids from noon on the day before their ascent until the
following night, all the urea would have to come from their tissues rather
than from any surplus food intake. They therefore prepared small cakes
made from starch paste fried in plenty of fat and took along sugar dissolved
in tea. There would also be some sugar in the beer and wine that they would
drink en route. They would analyze their urine production for urea on the
day of the ascent and reanalyze the samples for total nitrogen on returning
to their laboratory.
They set out soon after 5 a.m. and reached the hotel at 1:20 p.m. They
were enshrouded in a cold mist the whole time and did not become over¬
heated, so that they did not consider losses by perspiration to have been a
significant factor. They rested until 7 p.m., collecting “after-work urine,”
and then had a dinner consisting mainly of meat. On the next morning they
collected their overnight urine and returned immediately to their laboratory
to complete the analyses.
Their object was to compare the work done with the energy that could
have been liberated from the breakdown of the muscle albuminoids. There
was no doubt as to the external work achieved. Fick, with his hat, clothes
and stick, weighed 66 kg and Wislicenus 76 kg. The endpoint of the
climb was 1,956 m higher than the starting point. By simple multipli¬
cation, it could be shown that the external work done by the two men
against the force of gravity was therefore 129,000 and T49,ooo kg-m,
respectively. *
Then they used the concept of the conservation of energy, and its inter-
convertibility from one form to another, following the lead of J. R. Mayer
as described in Chapter 3, but with more precise data. The work of James
Joule and others had shown by then that the effort, or “energy,” needed to
raise a weight of 423 kg against the force of gravity by a height of 1 m was
the same as that which, if all the effort were dissipated as friction, would
give the heat required to raise 1 kg of water by i°C (i.e., 1 kcal). Using this
factor for the mechanical equivalent of heat,” one can say that the external
work of Fick and Wislicenus was equivalent to 305 and 352 kcal, respec¬
tively.
Returning to the estimation of the energy that could have been obtained
from the albuminoids, they found, as expected, that the value for the total
nitrogen content of the urine was always slightly higher than that for urea
nitrogen. To avoid controversy on this point, they used the higher values.
Expressed as grams nitrogen per hour, the results can be summarized as
follows:

66
 Things Fall Apart

Collection period Fick Wislicenus

The previous night 0.63 0.61

The morning of ascent (8 hours) 0.41 0.39
Rest after the ascent (6 hours) 0.40 0.40
The following night 0.45 0.51

The authors assumed that nitrogen from muscle which decomposed as a

result of work could have appeared in the urine both during the work and
in the following rest period. There was some variation in published analyses
for the nitrogen content of muscle albuminoids, but no one had reported
less than 15%. This conversion value would therefore give the highest pos¬
sible weight of albuminoids metabolized for a given excretion of nitrogen.
For the combined ascent and rest periods, the corresponding total weights
of albuminoids metabolized became 37.2 and 37.0 g for Fick and Wislicenus,
respectively. They then had to estimate the amount of energy that could
have been released during the partial combustion of this quantity of material.
Given the heats of combustion of carbon and hydrogen, they believed that
the highest possible value would be 6.73 kcal/g albuminoid. On this basis
the maximum energy obtainable would be 250 and 249 kcal, respectively.
These values were less than the external work accomplished by the two
men, and no account had been taken of the internal work, such as the
pumping of the heart and filling and emptying of the lungs, which the authors
estimated to have been not less than another 30,000 kg-m per head. They
therefore concluded that they had disproved Liebig’s assertion. It was im¬
possible for muscle substance to have provided the energy needed. Therefore,
fats and carbohydrates must have provided the staple fuels for the muscles,
since their metabolism was greatly increased during mechanical effort while
that of the albuminoids was virtually unchanged. They suggested that the
steam engine provided a useful analogy, since the iron machinery did not
consume itself, but a separate fuel that was oxidized and provided the energy
for the work indirectly through steam as an intermediate.
Later in the same year an Englishman, Edward Frankland, who was also
Fick’s brother-in-law, published an important supplement to the Swiss
work.41 His contribution was to measure directly the heat released by the
complete combustion of a large number of foods and also of urea. He got
them to burn by mixing them with potassium chlorate, and he did control
runs with chlorate alone or mixed with substances whose heats of com¬
bustion were already known. He calculated the heat produced by immersing
the combustion chamber in water and measuring the rise in temperature.
In this context Frankland’s most important values were 5.10 kcal/g for
dried beef muscle, from which the fat-soluble materials had been extracted
9

41. Frankland (1866a, b).

67
 Protein and Energy

with ether, and 2.206 kcal/g for urea. Since 1 g albuminoids gave 0.33 g
urea in metabolism, the remaining energy contained in the urea from 1 g
albuminoids was 0.33 x 2.206 = 0.73 kcal. By difference, therefore, the
energy released from metabolism of 1 g material was 5.10 — 0.73 = 4.37
kcal/g. This was considerably less than the indirect estimate of 6.73 kcal
used by Fick and Wislicenus. Applying the revised figure, he found that the
calories obtainable from the albuminoids that the two men metabolized
were reduced to 163 and 162 kcal, equivalent to little more than half the
external work performed.
Frankland also pointed out that all studies, both with mechanical systems
and with isolated animal muscles, indicated that chemical energy wa$ never
realized entirely as mechanical work; it was accompanied by at least an
equal amount of heat energy and usually more. In what he referred to as
“Smith’s highly important experiments on himself,” the extra energy re¬
leased from the body during treadmill work, as estimated from the carbon
dioxide produced, was three to four times greater than the external work
performed. The muscle, viewed as a machine, appeared to be operating at
no more than 25 to 33% efficiency. This was a repetition of conclusions
that Helmholtz had also drawn from Smith’s data in a lecture given 5 years
earlier.41
The actual calculations are not given in either paper, bpt we can re¬
produce them. As stated already, Smith expired an extra 19.6 g carbon
per hour when he was working the treadmill “15 minutes on, 15 minutes
off.” Fat, carbohydrate and protein all yield approximately 10 kcal per
gram carbon when oxidized, so that 19.6 g carbon is equivalent to 196
kcal. This in turn, using Joule’s equivalent of 423 kg-m/kcal, comes to
82,900 kg-m. In an hour (i.e., with two work periods) Smith would have
climbed 287 m. Since he weighed 89 kg, the external work done was
25,600 kg-m. This corresponds to 30% of the total energy estimated to
have been liberated.
One group in Britain for whom these results were of immediate concern
were army surgeons. They were responsible for ensuring that soldiers
received rations that would allow them to engage in strenuous activity in
wartime without becoming unduly fatigued. Professor Parkes of the Army
School of Medicine was concerned to replicate the Swiss type of experiment
and to check whether a 6-hour “after-work” period was enough to collect
the protein breakdown products resulting from a period of labor. He
organized a long series of trials with soldiers on both high-nitrogen and
low-nitrogen diets, where days of long marches with full equipment were
followed by rest periods of varying duration. He was finally convinced

42. Helmholtz (1861), pp. 355—7.

68
 Things Fall Apart

that “force necessary for great muscular work can be obtained by the
muscles from fat and starch.”43

Could the Liebig School Admit It Had

Been Wrong?

In the 1850s Liebig’s group expanded to include a physiologist, Theodor

Bischoff. He was an admirer of Liebig and eager to extend and support his
views of chemistry in relation to physiology. Later, he recruited an assistant,
Carl Voit (in some cataloging systems, Karl Voit), who had been another
admirer of Liebig as a medical student. They carried out very careful balance
experiments with dogs, the results of which persuaded them that with their
improved analytical methods they could obtain reliable nitrogen balance
data and that little or no nitrogen was being lost through the lungs or skin.44
Bischoff and Voit then carried out trials with dogs fed meat, starch and
fat in various proportions. They found, as had several other groups by then,
that urea production was greater in dogs on high-nitrogen diets. Others had
concluded that, although Liebig was obviously right in thinking that some
dietary albuminoids went to muscle and other tissues to replace “broken-
down” material, the excess, because of its rapid conversion to urea, must
be directly metabolized in the blood or liver, with urea as one product, and
that the remainder of the molecule was utilized for the same purposes as
the starch and fat in the original diet.45 But the Munich workers were
unwilling to accept an idea so clearly contrary to Liebig’s overall scheme.
As an alternative, they produced a complicated argument to the effect that
increasing the level of albuminoids in the diet resulted in the body having
to increase the volume of blood plasma and of organs to hold them, and
the heart having to do a proportionate amount of extra work. They ended
by saying that “it is established for all time that the nitrogen-containing
substances are the sources of physical power.”46
With hindsight, we might rephrase that as: “Don’t confuse me with facts;
my mind is made up.” Another German scientist wrote: “This scheme would
resemble a miserably built machine which used all its resources to overcome
its own internal resistance... and could be compared to the Danae, con¬
demned by the Greek gods to spend an eternity trying to bail water out of
a well with a sieve.” He went on to say that “to the unprejudiced it would
be immediately clear, from the observations of Bischoff and Voit, that the

43. Parkes (1871), p. 59.

44. Holmes (1987), pp. 251-7.
45. Lusk (1922), pp. 57-64-
46. Bischoff and Voit (i860).

69
 Protein and Energy

breakdown of albuminoids has nothing to do with the muscle work of the

organism... the only plausible explanation is that it occurs in a special
organ, where it is activated by an enzyme [ferment], and everything indicates
that this is the liver.”47

Voit’s “Work” Experiments

In Voit’s next experiments he used a large exercise wheel, on the same

principle as wheels employed in pet-mouse cages but large enough for 70-
lb. dogs. On their running days in the wheel, Voit’s dogs covered 16 km in
6 periods of 10 minutes each.48 A typical set of results for 2 days iij which
a dog received water, but no food, was that on its running day it excreted
17.1 g urea and on its rest day 16.7 g. When it was being given 1,500 g
lean meat per day, it excreted approximately 117 g urea on running days
and no g on rest days.
The work involved in running and turning the wheel was calculated by
Voit to be approximately 155,000 kg-m, equivalent to 366 kcal. This energy
could be produced from the metabolism of approximately 85 g albuminoids,
which would then yield 34 g urea. This was much greater than the total
output of urea on fasting days, whether the dog was running or not, and
vastly more than the 7-g increment on running days when the dog was
eating 1,500 g lean meat.49
What Voit had expected to find was increased urea production as a result
of work in sufficient quantity to account for the energy needed for the
muscular work having come entirely from protein breakdown. This was
not what he found. But he again thought of an argument thal: circumvented
the obvious conclusion. He assumed that the dog on the days m which it
was not fed was metabolizing its “flesh” and that this had the same com¬
position as the “flesh” (i.e., meat) that he was feeding to the dog in the
other periods. Then, assuming that the urinary nitrogen content reflected
the nitrogen content of the flesh metabolized, and knowing the nitrogen
content of the meat he fed to the dog, as well as an estimate of its caloric
value, he could calculate the estimated total calories released from the dog’s
metabolism of its flesh. He reckoned that the dog that had been fasting, on
a running day metabolized only 23 g flesh dry matter, which would yield
113 kcal. Since this was so much less than the external work calculated to
have been equivalent to 366 kcal, he concluded that an animal’s energy
came not just from ordinary chemical reactions, but from intermolecular
attractions and rearrangements that were not yet understood.50

47. Traube (1861), pp. 405-6.

48. Voit (i860), p. 154.
49. Ibid., pp. 156-72.

50. Ibid., pp. 190-1.

70
 Things Fall Apart

Figure 4.4. Cutaway sketch of Pettenkoffer’s apparatus for measuring res¬

piratory exchange in human subjects over a period of several days. (Bulletin
21, U.S. Department of Agriculture Office of Experiment Stations, 1895)

One would think that 17.1 g urea must have arisen from at least 43 g
albuminoids, so that his “23 g flesh dry matter” is unintelligible. Also, since
only nitrogen metabolism was measured, we have no way of knowing how
much body fat may also have been metabolized. The modern reader would
expect this to have been considerable. In contrast, the mechanical work was
probably much less than Voit had calculated in terms of the dog moving
in a series of kangaroo hops rather than running steadily up only a slight
incline. This appears to have been another instance of avoiding an obvious
conclusion at any cost.
Voit went on to enlist the help of Professor Pettenkoffer in designing and
operating an apparatus containing a chamber in which a human subject
could live for several days (Figure 4.4). The equipment included a means
of measuring the oxygen absorbed in the chamber and the carbon dioxide
and water vapor expired, in addition to the usual measurements of urinary
nitrogen compounds.51
From our point of view, the most interesting comparisons were again for
a human subject who, as in the preceding experiments with dogs, rested on
some days and did physical work on others. The work consisted of giving
a heavy crank some 7,500 turns by hand over a period of 9 hours. Unfor¬
tunately, the work done could not be quantified into meter-kilograms, but
the subject said that he felt as tired from performing the task as if he had
been on a long march.

51. Pettenkoffer and Voit (1868), pp. 472—8.

71
 Protein and Energy

The experiments were repeated with the 70-kg subject receiving either
nothing but water and a little extract of meat or else a mixed diet that
included meat, eggs, bread, milk, fat, sugar and beer.5i The mean results
can be summarized as follows, with all the values expressed as grams per
24 hours:

C out as C02
N out in urine
in breath

N Rest Work- C Rest Work-

in days days in days days

Hunger periods 1.5 12.4 12.3 3 203 * 333

On mixed diet 19.5 19.5 19.3 307 283 355

The development of the equipment was a considerable technical achieve¬

ment, and the attention to detail was such that even a book taken into the
chamber by a subject was weighed in and out to check whether it might
have absorbed moisture. However, in relation to the principles of nutrition,
it only confirmed that physical work was accompanied immediately by
greatly increased oxidation of carbon, but resulted in little or no change in
nitrogen metabolism.
The authors wrote that these results did not necessarily piean that the
physical energy came from fat or sugar oxidation. Such a conclusion
would require proving that not all of the energy released by oxidation
appeared as heat — that is, that some was being diverted to work. They
added that they were planning to develop equipment which would allow
them to measure heat production directly while a subject was doipg different
kinds of work. But even if it turned out that oxidation of these otRer carbon
compounds was providing the energy, the albuminoid muscle fibers must
still be involved, since it was they that attracted the oxygen out of the blood
and condensed it, and it was they that changed physically during muscle
contraction.
The authors’ hypothesis, though they realized it could only be tentative,
was that the muscle albuminoids gradually, in periods of rest, built up a
“tension” comparable to that in a wound-up spring, which could later be
released as mechanical work by an act of free will, and that during this
release oxygen was transferred from the albuminoids to fat, which oxidized
with the release of heat. The truth of this mysterious matter would have
to wait for the outcome of further experiments.”53 This hypothesis did not,
of course, explain the great increase in oxygen uptake during exercise.

52. Ibid., pp. 478-502.

53. Ibid., pp. 570-3.

72
 Things Fall Apart

Liebig’s Last Interventions

Three years later, in 1870, Liebig returned to writing on the source of

muscular energy, a subject that he had not touched for the previous 2.0
years, and two of his papers appeared in English translation in the following
year. First, he acknowledged that he had himself been responsible for an
error, namely, his assumption that urea production would always be pro¬
portional to work done. This had been disproved by Bischoff and Voit,
though he claimed that he deserved a portion of the credit, since he had
developed the improved method of analysis for urea that had given their
work the required precision.54
Nevertheless, he still believed that the energy needed for muscular work
came from albuminoids alone — not necessarily from their immediate com¬
bustion, but from internal molecular stresses gradually built up over a period
of as long as 3 days. The spring of a wound-up clock could serve as an
analogy. It was quite possible that the oxidative reactions occurring in the
body could release more energy than those occurring in Frankland’s com¬
bustion apparatus. It also seemed unreasonable that the daily loss of urea,
even in someone on a nitrogen-free diet and at complete rest, should be so
great if albuminoids were needed only to replace worn-out tissues and not
as an energy source.55
In a third paper, after writing that he believed the formula of blood
albumin to be Cil6Hl69Ni7Sz068, he attacked his old disciple and colleague,
Carl Voit. The latter’s main offense had been to believe it possible that,
under some circumstances, a portion of a high intake of albuminoids could
be converted to stored fat. Liebig wrote, “In his [Volt’s] hands facts are like
wax, to which the wished-for form is given by kneading it,” and he added
that he was unhappy at the way in which, nowadays, people were drawing
physiological conclusions before the necessary data had been sufficiently
established.56 This from Liebig of all people: what extraordinary impudence
and a classic case of the pot calling the kettle black!
It was, understandably, too much for Voit, and he tried to set the record
straight in a paper covering 99 pages. In it he pointed out where Liebig had
adopted his (Voit’s) views without acknowledgment, as in the concept of
resting muscle fibers gaining tension like a wound-up spring. Then in his
general conclusion he wrote: “He [Liebig] has brought about great progress
through his ideas about the processes within the animal body but he has
forgotten, to the sorrow of those who know and value his service to science

54. Liebig (1870a), p. 162.

55. Ibid., pp. 182-4; Liebig (1870b), pp. 77~8-
56. Liebig (1870c), p. 283.

73
 Protein and Energy

better than do his flatterers, that these are all mere ideas and possibilities,
whose validity has to be tested by actual animal studies.”57

Liebig’s Proprietary Products >

We will return to the “muscle question” after a brief detour. In the period
just under review, and for the remainder of the nineteenth century, the
general public was to associate Liebig with two nutritional products that
carried his name, but have not yet been described. Both were causes of
contention between the famous chemist and other scientists and medical
men. t
One of these was Liebig’s Infant Food, or Malt Soup (to translate the
German name literally), which was advertised as a complete substitute for
breast milk.58 At the Medical Academy in Paris, speakers said that Liebig’s
great reputation as a chemist had created public interest in the product but
that young infants did not thrive on it and the public must be warned about
the danger of using it.59 A Munich physician also reported adverse results.60
The other proprietary product was Liebig’s Extract of Meat. Liebig wrote
that it was prepared from finely chopped lean meat mixed with its own
weight of cold water and slowly heated. After boiling for a minute or two,
the liquid was decanted and strained. It could be used immediately for soup
or boiled down to the consistency of honey, when it would keep indefinitely.
(The commercial product was sold in the concentrated form.) Liebig con¬
cluded, “On the average, the soluble matters of 4 kg of flesh, after the
coagulation of albumin and coloring matter, do not exceed 100 g and of
this a very considerable proportion consists of organic salts, fhe phosphates
being particularly abundant, while the remainder is formed of no less than
five organic compounds.” Creatine was the organic compound to which he
devoted the most time. He found its formula to be C8N3H904 (which, with
modern atomic weights, becomes C4N3H9OJ, and showed that it readily
decomposed to creatinine, which is excreted in the urine.61
Liebig put forward two arguments in support of the value of this product.
The first was that, although plant albuminoids such as wheat gluten were
fully equivalent to the albuminoids of muscle fibers, both our own muscles
and the meat of animals contained, in addition, soluble or “juice” factors
such as creatine, which were not found in plants. Our bodies therefore had
to do extra work to convert plant material into these substances unless we
either ate meat or supplemented plant foods with “extract of meat.” These

57. Voit (1870), pp. 317, 399; Holmes (1964), p. cviii.

58. Liebig (1867).
59. Guibourt and Depaul (1867). V
60. Poppel (1869).
61. Liebig (1847), pp. 39-60, 131-2.

74
 Things Fall Apart

substances were true nutrients because they were identical with compounds
produced in the body from albuminoids.61
He offered a second argument in reply to critics who complained that
the public was being misled by commercial advertising of the product, which
seemed to imply that i oz. of concentrated extract was the equivalent of z
lb. of meat and that invalids in particular would be strengthened by it when
they were unable to eat solid food. For example, Edward Smith had said
at a public meeting in England in 187Z that the product had been sold in
large quantities since Liebig had lent his name to it, but it had little nutri¬
tional value; the flavor of meat disguised the real poverty of the substance.
However, with hot water it made a stimulating drink that was comparable
to tea or coffee.63
These views were summarized in the London Times, which published a
reply from Liebig a few weeks later. In his unique style, he characterized
Smith’s views as comical and “without even a faint notion of the science
of chemistry,” but then agreed with Smith that “extract of meat is not
nutriment in the ordinary sense-Like tea, it possesses a far higher im¬
portance by certain medical properties of a peculiar kind.... taken in proper
proportions it strengthens the internal resistance of the body to the most
various external injurious influences.”64 In the following year he also wrote
that meat extract increased the working capacity of the body through its
effect on the nerves; he added that people did not judge foods by their
carbon and nitrogen contents, but paid more for meat (than for liver or
cheese) because it contained certain other substances.65
The claims for extract of meat having special qualities have never been
substantiated.

Summing Up

By the end of the 30-year period covered in this chapter most of the beliefs
held at the beginning had disintegrated. On the purely chemical side, dif¬
ferences were found in the carbon-nitrogen ratio of different albuminoids,
and the existence of Mulder’s “protein radical” could not be confirmed.
This meant that animal proteins had, at least to some extent, to be man¬
ufactured by animals themselves, which was contrary to the earlier doctrine.
Experiments with both humans and animals also showed that physical
work was not necessarily accompanied by a significant breakdown of tissue
protein and increased excretion of urea, though it was always accompanied

6z. Liebig (1870b), pp. 108, 114.

63. Smith (1874), pp. 87-9; Finlay (1992.). PP- 4n-I3-
64. Liebig (187Z).

65. Liebig (1873).

75
 Protein and Energy

by increased oxidation of carbon, which must have come from the body’s
stores of fat and/or carbohydrate. By 1870 the matter had been settled for
most of those interested, though the German school continued to try to find
arguments that circumvented the obvious conclusion. But by the end of the
century reviewers were unanimous in concluding that protein was not the
main, or obligatory, source of energy.66
Now that Mulder’s concept is well in the past, we can use the word
“protein” again, in the modern sense of any type of “albuminoid.” The
development of knowledge as to how muscles actually worked was slow
and extremely complex, and is the subject of a comprehensive work.67 In
some aspects, Liebig’s picture was correct. The protein fibers do store energy,
and creatine does serve as an intermediate energy store involved in the
process of replenishment, but the processes are fully reversible, and it is the
oxidation of carbohydrates and fatty acids that we rely on for continued
physical work.
One last point of terminology: It is now becoming standard practice in
scientific publications to replace the kilocalorie as a measure of energy
expenditure with the joule (J). One calorie is usually taken as equivalent to
4-I9 J- F°r example, what used to be described as a daily energy expenditure
of 2,000 kcal/day now becomes 8.38 megajoules (MJ). In trying to connect
this relationship to James Joule’s own conversion factor, one njust remember
that the force of gravity at the earth’s surface is approximately equivalent
to 10 m/sec1, and the present unit is the work required to move a mass of
1 kg a distance of 1 m against a force equivalent to 1 m/secL

66. Paton (1895); Armsby (1906), pp. 194-109.

67. Needham (1971).
5 Vegetarian Philosophies and Voit’s
Standards, 1875—1893

In the preceding chapters we followed the gradual development in

Europe of scientific knowledge about the chemical composition of foods
and our requirements for different types of nutrients. The writers considered
up to this point judged the suitability of different foods — meats, bread,
fruit and so on - entirely in terms of their digestibility and their capacity
to meet the needs of tissue growth or replacement.

The Romantics’ Abhorrence of Killing

From the beginning of the nineteenth century, another school of thought

developed in England and then in the United States. Its central belief was
that it was morally wrong to kill animals in order to eat them. This was
not an entirely novel idea. Many early Greek philosophers had had the same
belief, as had followers of several Eastern religions. Some earlier Europeans,
such as Leonardo da Vinci, had also been vegetarians. However, there had
been no continuing tradition of vegetarianism in the West.1 The Catholic
church forbade the eating of meat on fast days, but by implication considered
it acceptable on other days. And the Old Testament contained references
to Jehovah’s pleasure at the sacrifice of animals on his altars.
The reappearance of the vegetarian ideal in the 1790s and early 1800s
seems to have been related to the Romantic movement, which began as a
reaction against the formalism and arid rationality of the eighteenth-century
Age of Enlightenment in France. Jean-Jacques Rousseau had attacked the
artificiality of French society and argued that a simple kindness of heart
was more important than sophisticated knowledge.
One of the first Englishmen to write on vegetarianism in this period quoted
extensively from Rousseau - for example, “The only way in which man
can be happy is to be mild, benevolent and humane, in contrast to the

1. Barkas (1975), PP- 69-73; Dombrowski (1984)-

77
 Protein and Energy

present bloodshed and cruelty.” The author, Joseph Ritson, argued that
humans had an animal body, so that the killing of other animals was halfway
cannibalism. Moreover, it did not appear from the anatomy of our teeth
and digestive tract that we were intended to be meat eaters. The animal
closest to humans in the arrangement of its teeth, the orangutan, was a
vegetarian in its natural environment. Ritson cited examples of peoples with
a reputation for strength and good health who lived without meat. Among
these were Irish laborers, whose diet consisted virtually entirely of potatoes
and buttermilk. Barbarous killing to obtain meat was therefore not essential
and done only “to gratify luxury”.1
Nine years later, John Newton published The Return to Nature, ip which
he argued that “man, of all races, is the most diseased.” This was because
he had “quit the nutriment on which alone Nature had destined him to
enjoy a state of perfect health.” Newton personally knew no less than 2.5
vegetarians and all were in excellent health. “In the happy state of innocence
[i.e., in the Garden of Eden], man was commanded to eat, not fish or meat,
but the fruits of the Earth of every sort.”2 3 Certainly, if the Creator desired
humanity to follow a moral order, and if He regarded killing other animals
as immoral, it would seem to follow that He would also have designed our
own system so that optimal health could be maintained without our having
to kill.
The poet Shelley, after only one year as an undergraduate at Oxford, was
expelled for writing a pamphlet in defense of atheism. He became a friend
of John Newton and was converted to vegetarianism. When still only 21,
he published a pamphlet on the subject, claiming, like his predecessors, that
“the present depravity of the physical and moral nature of man originated
in his unnatural habits of life.” He added that he believed that fire and
cooking had been used to disguise man’s horror of eating bits of bleeding
carcass, and “from this moment his vitals were devoured by the vultures of
disease.” Like Newton he assumed that “man, and the animals whom he
has depraved by his dominion, are alone diseased.” Except in children, no
instinct remained for what was natural.4 In arresting terms, he wrote:
The whole of human science is comprised in one question: how can the advantages
of intellect and civilization be reconciled with the liberty and pure pleasures of
natural life; how can we take the benefits and reject the evils of the system which
is now interwoven with the fiber of our being?5

These ideas were, of course, entirely at odds with both the orthodox
science of the time and the views of “the man in the street.” As we saw in

2. Ritson (1802), pp. 40, 44, 78.

3. Newton (1811), pp. 63, 65, 109.
4. Shelley (1813), pp. 9-11.
5. Ibid., p. 12.

78
 Vegetarian Philosophies and Voit’s Standards

Chapter 3, the ideal food was considered to be one with the composition
of animal tissues, and plant foods were judged by how nearly they ap¬
proached that ideal in their nitrogen content. The popular feeling in England
during this period of the Napoleonic Wars, and one that contributed to
morale in the armed services, was that the relatively high meat diet in Britain,
as compared with the diet in France, made “one Englishman equivalent to
at least two Frenchmen.” The complications of French cooking were re¬
garded contemptuously as being merely ways of “making a little meat go
a long way.”6
The first preacher to urge vegetarianism on members of his congregation
appears to have been the Reverend William Cowherd, founder of the in¬
dependent Bible Christian Church, in Lancashire in 1807.7 Cowherd and
his congregation were said to have become unpopular in the area because
of their criticisms of their neighbors’ habits. One of Cowherd’s assistants,
William Metcalfe, emigrated to Philadelphia with some of the flock in 1817
and made further converts there.
The new environment was probably more favorable for these immigrants.
There was a mood of optimism in the United States at this time and a feeling
among ordinary people that there was no limit to their opportunity for a
good life if they worked hard and abstained from every kind of personal
excess. Many writers urged, in particular, that ill health and epidemics not
be accepted as trials sent by God to be endured as a means of acquiring
merit in His eyes. Health was a duty and sickness a sin, reflecting gluttony
and immoderate living.8 One writer has said that during this period “per¬
fectionism swept across denominational barriers and penetrated even secular
thought.... The progress of the country suddenly seemed to depend upon
the regeneration of the individual and the contagion of example.”9

Sylvester Graham

In 1846 Metcalfe wrote that he had one hundred people who had been
abstaining from animal foods and intoxicating drinks. The convert who
was eventually to be the most famous was Sylvester Graham. Since it has
been suggested that Graham’s early experiences explain some of his later
teachings, I will summarize these briefly. He was bom in 1794 as the young¬
est of a large family. His father, a retired Connecticut clergyman, died two
years later, and soon afterward his mother lost her mind. He was cared for
by a series of relatives, his education interrupted by bouts of ill health.

6. Drummond and Wilbraham (1939), p. 2.54.

7. Forward (1898), p. 7; Carson (1957), pp. 15—17.
8. Fowler (1851), p. 49; Whorton (1982), pp. 13—37.
9. Thomas (1965), pp. 658—9.

79
 Protein and Energy

Finally, at the age of 30 he became a Presbyterian preacher and, after private

tutoring, a regular minister in New Jersey. He began to purchase books on
physiology, anatomy and nutrition. In 1830 the Philadelphia Temperance
Society appointed him as a lecturer, and he began to speak not only on the
ill effects of alcohol, but also on the importance of chastity, the prevention
of cholera and diet reform, including the elimination of meat.
Graham was a very successful lecturer and spoke in New York, Boston
and Philadelphia for fees as high as $300 per night, at a time when a nurse
might earn $3 per week. However, protesters attempted on several occasions
to stop his lectures. Some were butchers and others people protesting his
discussion of sexual subjects. In 1837 he arranged for the publication of
the Graham Journal of Health and Longevity, and in several cities Graham
boardinghouses were set up where his recommendations could be followed.
In 1839 he published a large work, The Science of Human Life, and then
went into semiretirement. He died in 1851 at the age of 57. His aggressive
and impassioned style made him a successful public speaker, but he was
apparently aloof and crotchety toward his admirers.10
Graham is more complex and interesting than one might expect the orig¬
inator of “Graham crackers” to be. First, he was not a straightforward
puritan, suspicious of any kind of pleasure. For example, he wrote:
1

Exercise is important, but it must be enjoyed. Religious prejudice against dancing

is altogether ill-founded_better that people should come together to sing and
dance, in the health and exhilaration of their spirits than ... that they should endure
a miserable existence in moping melancholy.... Children require much exercise in
the open air. Action is as instinctively natural to children as breathing, and it is
unnatural and improper to restrain them from it... for any considerably time. Girls
should be allowed as much freedom of action, in childhood, as boys.11

Graham’s philosophy began with a consideration of human nature -

which has occupied the minds of so many thinkers.... In the human body, matter
and vitality, and mind and moral feelings are mysteriously associated.... By this
wonderful union of intellectual and moral powers with organized matter, man alone,
of all terrestrial beings, is brought into a two-fold relation with his Creator-The
animal nature of man may be considered as the basis of human existence. Its passions
and desires... constitute the primary and principal elements of activity to his mental
powers, and tend continually to cause his rationality to concur with his animal
indulgence.... Man, unlike animals, has both the voluntary and intellectual powers
and the natural propensity to violate the constitutional laws of his nature, and thus
deprive, deteriorate and destroy himself.11

10. Naylor (1942); Nissenbaum (1980), pp. 9-18; Hammond (1987), pp. 43—53-
ix. Graham (1839), Vol. 2, pp. 119, 655-6.
12. Ibid., Vol. 1, pp. 424-5, 428.

80
 Vegetarian Philosophies and Voit’s Standards

He added that an addiction could “establish in the body an appetite whose

despotic and often irresistible influence... compels the understanding and
will to comply with its demands.”13 The obvious addiction that would have
sprung to the reader’s mind at that time was alcoholism, but Graham went
on to develop a broader argument. History had repeatedly shown that
savage” peoples could work their way up to a life of civilization and luxury,
only to become debilitated and effete and to fall back again into poverty
and servitude.

Disease Due to “Overstimulation”

Graham’s ideas as to the kind of life that would keep people vigorous and
free of disease were in line with some of the orthodox medical ideas of the
period. Doctors believed that the body was healthiest and had the greatest
resistance to disease when in balance between the two extremes of weakness,
or debility, on the one hand, and “plethora” (i.e., ruddiness with excess of
blood and overheating), on the other. The purpose of their practice was
therefore to bring their patients back to the ideal mean position, in part by
prescribing the foods they could eat. Meat, eggs and spices were all
“heating,” as were coffee, tea and alcohol, and were therefore denied to
plethoric patients; they were to receive fruit and vegetables, regarded as
“cooling.” Pregnant women, for example, were regarded as susceptible to
plethora, and one aristocratic lady expecting her first child in 1810 wrote
to a friend that, because of her physician’s instructions, “I am now living
exactly like a horse on grass food and water.”14
Benjamin Rush, the doyen of the Philadelphia Medical School at the
beginning of the century, had taught that the direct cause of disease was
“morbid excitement,” that is, excessive stimulation or irritation, either phys¬
ical or mental, operating on a body already debilitated, either indirectly by
previous overstimulation or directly by cold, hunger, grief, and so on. The
main route of exterior physical stimulation was the gastrointestinal tract,
and overeating, especially of spiced foods, was therefore a major source of
overstimulation of the body.15
In 1826 a translation of the French physiologist Broussais’s similar views
on overstimulation as the cause of debility and disease had been published
in Philadelphia. He wrote: “Strong and black meats, full of extractive matter
... high seasoned and fermented liquors have the double bad effect of sup¬
plying an abundant and substantive chyle, and of exercising [i.e., straining]
the assimilative power of the stomach.... While the man is yet young he

13. Ibid., pp. 437-8.

14. Lewis (1986), pp. 130—1.

15. Nissenbaum (1980), pp. 54—7.

8l
 Protein and Energy

resists for a length of time such excesses. But there is a limit to everything.
[Finally] the fundamental organs, after having resisted acute congestions,
sink under chronic irritations.”16
Graham followed these precepts in teaching that people should eat as
little food as they found to be consistent with maintaining their energy and
that they should cut out all meat and stimulants, including alcohol, tea,
coffee, pepper and mustard. He also implied, at one point, that they should
eat foods that were easily digested.17 Ffowever, William Beaumont, in his
famous work with a patient who had a wound that provided direct access
to his stomach, had shown that meat was “digested” (i.e., dissolved) there
more quickly than were vegetable foods.18 »
Graham and other vegetarian writers accepted Beaumont’s finding, but
claimed that the stimulation and “heat” felt by people after having a meat
meal was evidence that their rapid digestion required a large expenditure
of vital energy.19 (The modern scientist would agree that the digestion and
absorption of high-protein foods such as meat result in an increase in met¬
abolic rate. This is attributed principally to the complex, energy-using chem¬
ical reactions involved in handling the absorbed digestion products.)
With regard to vegetable food, Graham argued that the slower digestion
was gentle and unstimulating, and allowed people to work for a longer
period, without the need for a further meal. He accepted tl^at a portion of
vegetable foods remained indigestible, but said that the structure of the gut
showed that it was designed to deal with a certain amount of “innutritious”
matter (i.e., fiber, in modern terminology) and that this was “necessary to
sustain the functional powers of the alimentary organs.” He used this point
in criticizing the conclusions drawn from Magendie’s experiments, in which
dogs died when fed pure sugar or white bread. These did not prove the
need for a high intake of nitrogen, since no diet lacking non-nutritious
matter could be expected to keep an animal healthy. He was, in any case,
critical of attempts to characterize diets by their proportions of carbon and
nitrogen, because living beings had vital powers that did not obey the law
of dead chemistry.10
Graham argued that wheat bread was a particularly good food. He did
not seem to regard the leavening with yeast and subsequent baking as
“unnatural.” He did, however, disapprove of commercially prepared bread
because of the chemicals added to it.11 (There is evidence that this was a
real problem in that period.) He also argued against the fractionation of

16. Broussais (1826), pp. 32.5-6.

17. Graham (1839), Vol. 1, p. 540.

18. Beaumont (1833), p. 46.

19. Graham (1839), Vol. 2, pp. 108-12; Whorton (1982), pp. 81-2.

20. Graham (1839), Vol. 1, pp. 542-4; Vol. 2, p. 116.

21. Ibid. Vol. 2, pp. 423-5.

82
 Vegetarian Philosophies and Voit’s Standards

ground wheat to separate the white flour from the branny fraction. White
bread was too concentrated and constipating. Bran was non-nutritious, but
it was “not a mere mechanical irritant”; it corrected chronic diarrhea as
well as constipation, “probably from the soothing action of the mucilage.”zz
Whole-meal bread seemed to have a symbolic importance for Graham.
It should be baked by the mother of a family — not by servants — and her
love for her family would ensure that the process received the care it de¬
served/3 The bread, made with clean wheat, freshly ground, would be the
centerpiece for the family coming together at their dinner table. This was
a romantic picture painted by someone whose own mother had been unable
to carry out this role.

Diet and Sex

To the modern reader it seems strange that health reformers in the 1830s
should have considered even modest stimulation a danger/4 Perhaps the
greatest danger in their eyes was that such stimulation or irritation would
be transmitted through nervous connections to the reproductive system.
Earlier writers had referred to the exhausting effects of loss of semen - each
emission resulting in a shortening of life by so many days, or “the loss of
1 ounce of this rectified spirit being more weakening than the loss of 40
ounces of blood.”zs
However, this was not Graham’s main concern; his strictures applied to
either sex. In his lectures on chastity he said that “the convulsive paroxysms
attending venereal indulgence... cause the most powerful agitation to the
whole system that it is ever subject to_The powerfully excited and con¬
vulsed heart drives the blood, in fearful congestion to the principal viscera
- producing depression ... rupture, inflammation and sometimes disorgan¬
ization.”^ Apparently, orgasm, or even “the actual exercise of the genital
organs,” was not required for this dreadful sequence of events. “Day dreams
and amorous reveries” or even “an aching sensibility” could themselves lead
eventually to debility, disease and death/7
What shocked people at the time about Graham’s views was that inter¬
course between man and wife was made to seem as undesirable as adultery,
since he was arguing on hygienic grounds rather than moral ones. To counter
this point, Graham could only say that sex in marriage was less damaging

22. Ibid., pp. 428—30.

23. Ibid., pp. 454-67.

24. Whorton (1982), p. 78.

25. Sokolow (1983), p. 79.

26. Quoted by Whorton (1982), p. 93.

27. Quoted by 'Nissenbaum (1980), p. 113.

83
 Protein and Energy

than adultery because, in general, it was less exciting; but even vigorous
young couples should indulge no more than once a month.
Why Graham, and others, in this period should have professed such views
about sex has been discussed elsewhere.19 It is more relevant, in the present
context, to consider what evidence he had for associating meat eating with
lust. Benjamin Rush, writing on diseases of the mind in 1812, referred to
cases of excessive sexual appetite and their treatment. He believed that one
remote cause was “excessive eating, more especially of highly seasoned
animal food,” and that treatment should include “a diet, consisting simply
of vegetables, without any of the condiments that are taken with them.”
He added that Dr. Stark, who had reported the results of his own dietary
experiments, “found his venereal desires almost extinguished by living upon
bread and water.”30 This was perhaps not surprising, since he died as a
result of this self-experimentation, probably of scurvy.31 Graham himself
did not cite Rush or offer specific evidence for his belief.
Graham’s lectures were well attended and his books sold in large numbers
but his followers formed only a tiny portion of the U.S. population. The
more common view was that vegetarianism was inconsistent with physical
vigor (Figure 5.1). Graham’s type of teaching was subjected to some witty
attacks:

Here is an exact portrait of a modern lecturer on Dietetics... general ignorance with

a smattering of medical knowledge;... all the great objects of amelioration and
amendment are to be accomplished by the substitution of unbolted flour in the place
of pure wheat and solid animal food.... man has no imagination, no heart, no soul
... but is all stomach_Vegetable food is said to preserve a delicacy of feeling
seldom enjoyed by those who live principally on meat.... Green peas ihd cabbage
are become meta-physical.31

John Harvey Kellogg

Other proponents of vegetarianism continued to lecture after Graham’s

death in 1851, and one religious group adopted his teaching virtually in
toto. These were the Seventh Day Adventists, so called because they believed
that Saturday, the seventh day of the week, should be set aside for worship
and rest, rather than Sunday, and that the Advent, or Second Coming of
Christ, would be in the near future. They started as three small congregations

2.8. Ibid., pp. 118-19.

29. Sokolow (1983), pp. 91-5.

30. Rush (1812), pp. 350-3.

31. Carpenter (1986a), p. 201.

32. Anonymous (1837), pp. 337, 342, 350-1.

84
 Vegetarian Philosophies and Voit’s Standards

Figure 5.1. Caricature exemplifying the common view that vegetarians

necessarily became weakly. Members of the Vegetarian Society are por¬
trayed as having to be carried to their annual banquet on stretchers. {Le
Charivari, Paris, 1853)

in New England in 1845. By 1855 they had moved their headquarters to

Battle Creek, Michigan.33
So that church members could maintain healthy living in preparation for
the Advent, prohibitions were gradually imposed — on alcohol, tobacco,
tea, coffee, pepper and other condiments and finally meat in the 1860s. In
1866 the Western Health Reform Institute was established at Battle Creek
for the purpose of teaching and bringing invalid Adventists back to fitness
through “healthful living” in a religious atmosphere, rather than through
the use of drugs. At first those in charge had no medical qualifications, but
in 1876, when the institute’s affairs were languishing, John Harvey Kellogg,
24 years old, was appointed as physician in chief. Raised as an Adventist,
he had been converted to Grahamism as a boy, and in 1872 the organization
had sent him to New York to obtain an orthodox medical education. On
his return he received permission to create a much larger medical sanitarium.

33. Land (1986), pp. 39-51.

85
 Protein and Energy

Although this facility attracted patients from all over the country, the church
organization was unhappy because Kellogg appeared to be putting health
reform activities ahead of religion, placing too much emphasis on bringing
in celebrities such as Henry Ford and John Rockefeller, and providing en¬
tertainment with a resident string orchestra in the dining room.34
Kellogg lived until 1943 and wrote many books, which continued to
appear in new editions in the 1920s and 1930s and made him wealthy.
However, his main ideas were set in place within a few years of his taking
charge of the sanitarium. On the damaging effects of sex his views were at
least as strong as those of Graham. He was particularly concerned as to
“the almost universal vice of self-abuse among the young”; “mucl^ of the
nervousness, hysteria and general worthlessness of the girls of the rising
generation originates in this cause alone.”35 “Meat of all kinds... is stim¬
ulating and should not be freely used by anyone whose nervous system is
already over-excited and irritable.”36
He believed that use of the generative organs caused a nervous shock to
the whole system and that unnecessarily frequent intercourse in marriage
could lead to insanity. Although he himself married, he said that he wanted
to show that continence was still possible; he and his wife had separate
rooms and adopted children but had none of their own.37
1

Autointoxication

In his writing on diet Kellogg referred not only to the dangers of general
“overstimulation” from meat eating, but also to the poisoning and the
individual diseases associated with the practice. In particular he had col¬
lected data suggesting that there was a lower incidence of cancer among
vegetarians.38
Unlike Graham, he believed that the characteristic effects of foods could
be explained in terms of their chemical components. The bad effects of meat
he attributed to its high protein content.39 In the main, these effects came
indirectly from the action of bacteria, whose significance had become ap¬
parent only after Graham’s time. It was now realized that these organisms
were responsible for the phenomenon discussed in earlier chapters - namely,
that, in moist storage, foods high in nitrogen putrefied and went alkaline,
while high-carbohydrate foods fermented, with the production of acid. The
chemical products of the acid fermentation were generally harmless, but the

34. Carson (1957), pp. 88-103; Whorton (1982), pp. 202-3; Land (1986), pp. 69-70, 133—

8.
35. Kellogg (1893), pp. 144, 148.

36. Ibid., p. 221.

37. Sokolow (1983), p. 162; Carson (1957), pp. no-n; Money (1985), p. 84.

38. Kellogg (1877), p. 55; Kellogg (1921), p. 491; Kellogg (1923), pp. 145-6, 253-5.

39. Kellogg (1922), pp. 35-7.

86
 Vegetarian Philosophies and Voit’s Standards

bacterial putrefaction of proteins gave some products that not only smelled
repulsive but were poisonous. In fact, given the favorable light in which
nitrogen-containing materials were considered in earlier contexts, the as¬
sociations that Kellogg made come as something of a shock:
The offensive odor arising from the burning of protein is due to the poisonous
compounds which are formed from nitrogen and sulphur. Compounds of these
substances, formed either in the chemical laboratory or in the mysterious laboratory
of the plant, are among the most poisonous known to man. Nitrogen is an essential
constituent of the high explosives used in warfare. It enters into the venoms of snakes
and the virulent poisons produced by bacteria.40

It was believed by many physicians in the 1880s that this putrefaction

could occur in the human colon, or large intestine, especially when someone
had eaten more protein than could be completely digested in the small
intestine. The result would be autointoxication from the toxic products.41
It was further supposed that bacteria of the putrefactive type had reached
the colon originally from meat that was already infected. Kellogg was con¬
vinced of the importance of this concept and added that constipation — the
disease of civilization resulting from overrefined foods — gave the bacteria
more time to do their deadly work. To ensure that food residues did not
stagnate in the colon, one should keep to a diet that promoted three bowel
movements per day, was free of meat and was low in total protein content.41
Obviously, cereals, fruit and vegetables were desirable types of food. Kellogg
and his wife developed recipes for precooked, ready-to-eat wheat and oat
granules that they called “granola.” His brother William then developed
flaking procedures on an industrial scale. Wheat flakes were officially in¬
troduced at the Seventh Day Adventists’ annual meeting in Battle Creek in
1895, ar|d by 1906 a method for producing corn flakes was also perfected.43
Today, few of the millions who eat the Kellogg Company’s breakfast
cereals know of the beliefs of their inventors, and if they did would probably
consider them unbalanced. However, to jump forward in time for a moment,
it is interesting that the Seventh Day Adventists are a thriving church and
that modern studies of the health records of members following the tradi¬
tional prohibitions have shown that they do indeed have a lower risk of
developing cancer than their neighbors who have no such prohibitions.44
It is also of interest that, as visitors to the West End shopping area of
London know, from at least 1973 to 1990 an anonymous man has consis¬
tently walked the streets carrying a banner proclaiming, “Less lust from less

40. Kellogg (1921), p. 31.

41. Bouchard (1894), pp. 138-53; Hudson (1989), pp. 396-400.

42.. Kellogg (1923), pp. 632-3.

43. Money (1985), pp. 24—6.

44. Berkel and de Waard (1983).

87
 Protein and Energy

Figure 5.2. The “protein man” (1987), a familiar figure in London’s central
shopping district for many years. (Art Director’s Photo Library, London)

protein” and selling a pamphlet now in its 35th edition that gives details
of his ideas (Figure 5.2). But one must wonder whether some people who
have seen his banner have been inclined to change their diet in the opposite
direction to the one he intended.

The Opinions of the Establishment

We left Chapter 4 with scientists’ general acceptance by the 1870s that

dietary protein was not the essential energy source for muscular contraction.

88
 Vegetarian Philosophies and Voit’s Standards

The uses of protein for the growth of new tissue and the replacement of
existing, worn-out tissue were not contested.
Carl Voit, who had attacked Liebig in his “25-year review” in 1870
(as discussed in Chapter 4), continued to organize a long series of studies
on protein nutrition in Munich and became the new authority, though
not an undisputed one. In 1875, in a lecture to people with responsibility
for providing food in different kinds of institutions, he proposed a daily
dietary standard for an average worker. This consisted of 118 g protein,
56 g fat and 500 g carbohydrate, with the proviso that one-half of the
protein be from animal sources. The quantities of fat and carbohydrate
could also be modified as long as they included no more than 500 g
carbohydrate and their combined carbon content remained the same. Later
he was to modify this by saying that their combined energy value should
remain the same.45
Voit emphasized that his standard did not apply to the average man, but
to a physical worker, that is, a necessarily well-muscled person. Inactive
people, like senior citizens in a nursing home or confined prisoners, needed
less. He understood that at first sight this did not seem to be consistent with
his own experiments showing that protein metabolism, as judged by the
excretion of urea, did not increase as a result of physical activity. However,
the protein requirement was proportional to muscle mass, and the ability
to do heavy work required the maintenance of a larger mass. Everyone
accepted that a large draft horse at rest needed more dietary protein than
a pony did. In the same way the physical worker needed to maintain his
protein consumption even on his rest days so that he would not lose strength
before returning to work.46
Voit accepted the criticism that his standard was based largely on what
men ate rather than what they needed to eat, but in 1881 he published an
expanded explanation of his standard and referred to another publication
by a visiting worker in his laboratory, claiming that it provided data con¬
firming the need for 118 g protein (equivalent to 105 g digestible protein)
by the average worker.47 The author was Hamilton C. Bowie, from San
Francisco, who was in Munich in 1878-9. Bowie had begun by citing papers
of scientists who had carried out studies on themselves and found that they
maintained nitrogen balance on diets that contained 100 g protein or less.
But he pointed out that these people did not even have to remain standing
all day. Voit’s “average worker” was a man who could maintain physical
work for 10 hours per day at a trade such as carpentry or bricklaying -
that is, something more strenuous than tailoring, but less active than work-

45. Voit (1876), pp. 21-2; Voit (1889), p. 243.

46. Voit (1876), pp. 23-4.
47. Voit (1881), p. 525.

89
 Protein and Energy

ing as a blacksmith. There was no contradiction, therefore, in scientists,

with their smaller muscle mass, requiring less protein.48
Bowie’s own calculations were based on his reanalysis of data that had
been obtained with a laboratory assistant (P.P.), 43 years old, weighing 74
kg and described as “big and muscular.” His work included cleaning glass¬
ware and keeping the fire going in the laboratory oven, and his normal diet
included cheese, sausage, meat dumplings, bread and beer. In one 24-hour
period his urine was found to contain 18.5 g nitrogen, and in another 14.4
g. Assuming that he also had a typical loss of 2.3 g nitrogen per day in his
feces, the average combined loss per day was 18.75 g- Bowie assumed that
he was in equilibrium (i.e., neither gaining nor losing body mass) at this
time and that his average daily diet must therefore also contain 18.75 g
nitrogen. Taking a factor of 6.45 for converting nitrogen to protein, this
corresponded to an average of approximately 120 g protein per day.49
Max Rubner, another colleague in the Munich laboratory, had published
the first results of the work with this subject, which had been designed to
measure the digestibility of individual foods or simple combinations. For
each study the subject consumed on the first day nothing but about 2 liters
of milk. On the next 2 days (or occasionally 3 days) he ate the test diet,
and finally spent the last day with nothing but another 2 liters of milk. The
“milk days” were used to yield fecal material with a different character.
The analysts could then separate for analysis the fecal material appearing
between the “milky” ones. This material could be assumed to have origi¬
nated from the test diet even though it appeared a day or so later. s° Alto¬
gether 10 such studies were carried out using subject P.P. (who was recoded
“D” in this paper). Rubner had also reported the nitrogen contents of the
urine produced on the test days, although he made no further u$e of these
results.51
Bowie calculated the nitrogen balance in each test period by subtracting
from the nitrogen present in the test diet over the 2 (or 3) days both the
corresponding fecal nitrogen and the urinary nitrogen during the test pe¬
riod. 5i His findings are summarized in Table 5.1. In the original paper he
also calculated balances for some of the i-day periods on milk, but the
results were erratic, and I have omitted them. Figure 5.3 shows graphically
the relation between the amount of “digestible protein” eaten and the ni¬
trogen balance of the subject (i.e., the balance of nitrogen consumed minus
nitrogen excreted).
The data show that, in dietary periods when he ate less than 65 g digestible

48. Bowie (1879), pp. 466-71, 475.

49. Ibid., p. 474.

50. Rubner (1879), pp. 119—20.

51. Ibid., pp. 138-77.

52. Bowie (1879), p. 479.

90
 Vegetarian Philosophies and Voit’s Standards

Table 5.1. Results of short-term nitrogen balance experiments in a

human subject (g/day)

Primary data Calculated data

N Digestible
Food N in Fecal N Urine N balance protein"

White bread, yeast* 7.6 2.0 11.2 -5.6 36.1

Macaroni noodles 10.9 1.9 16.0 -7.0 58.0
Black bread 13.3 4.3 12.5 -3.5 58.3
Dumplings* 11.9 21.3 13.8 -4.2 61.9
White bread* 13.0 2.4 12.6 -2.0 64.9
Macaroni, gluten 22.6 2.5 18.0 + 2.1 129.5
Meat, bread, lard (200 g) 23.5 3.3 18.3 + 1.9 130.3
Milk, bread, butter (240 g) 23.0 2.6 16.2 + 4.2 131.6
Meat, bread, lard (100 g) 23.7 2.9 23.5 -2.7 133.5
Meat, bread, fats (380 g) 23.4 2.1 15.5 + 5.8 137.1

"The “digested crude protein” was calculated as (“N in” minus “fecal N”) X 6.45.
*These test foods were given for 3 days, the remainder for 2 days.
Sources: The primary data are from Rubner (1879) and the calculated data (with
one correction) from Bowie (1879), all rounded off to one decimal place.

protein per day, the subject was in negative balance; at 120 g or above, he
was in positive balance. It is unfortunate that no diet provided digestible
protein in the critical range from 65 to 105 g. However, even though there
is one aberrant negative balance from 133.5 g digestible protein, the results
roughly fit a regression line, going through “zero balance” at an intake level
corresponding roughly to Voit’s standard. The one negative balance in the
high-protein periods was explained by a low calorie intake. At the time, the
results did seem to provide clear support for Voit’s standard of 118 g total
mixed protein, equivalent to about 105 g digestible protein, and for his
conclusion that feeding an average worker less would cause his body to
“degenerate,” that is, to lose protein.53

Circulating and Organ Protein

In encouraging these simple conclusions, Voit was disregarding his own

findings with dogs in the 1860s, even though he had just discussed them
again in his 1881 treatise The Physiology of General Metabolism and Nu¬
trition.54 The first finding was that if he took a dog that had been receiving

53. Voit (i88i),:p. 525-

54. Ibid., pp. 103-18.

91
 Protein and Energy

+6 % 9
The Voit
Standard
+4

Apparent /
+2 y-
/
N balance,
1 /
0 /
g/day

-2 /
/
6
-4
/
-6

y
20
6 /

40
6
60 80
1100 120
Digestible protein eaten per day, g
140

&
Figure 5.3. Short-term nitrogen balance results of Rubner, as calculated
by Bowie and plotted in relation to Voit’s standard of 105 g digestible
protein for an “average worker of 70-75 kg.”

a generous ration of meat for a considerable period and gavejt no food for
a period of 4 days, its daily output of urinary nitrogen would decline day
by day. The average values from six experiments were (day 1) 9.9, (day 2)
3.9, (day 3) 2.6 and (day 4) 1.4 g nitrogen.55 Dogs of the size that Voit was
using would have had at least 700 g nitrogen in their body proteins, so that
the fall in excretion rate could not be explained by a significant reduction
in total body protein. Voit’s explanation was therefore that, in addition to
the “fixed” protein that constituted the working muscles and the other body
organs, there was a smaller quantity of what he first called Vorrathseiweiss
- that is, storage or stock protein — and later circulating protein.56 He further
hypothesized that organ proteins metabolized slowly, less than 1% per day,
and were replaced from the pool of circulating protein. Protein newly ac¬
quired from digested food went first into the circulating pool, and this
material had a much higher rate of decomposition, perhaps as much as 80%
per day, so that it normally provided the majority of urinary nitrogen.
This idea also provided an explanation for Voit’s observations that a dog

55. Voit (1867), p. 47.

56. Voit (1869), p. 330.

92
 Vegetarian Philosophies and Voit’s Standards

could be brought into nitrogen balance, within certain limits, at different

levels of nitrogen intake. As intake increased, the pool size increased day
by day until the daily metabolism also reached a level equivalent to the new
intake, and the animal was once more in balance. Similarly, if intake was
reduced, metabolism at first exceeded intake; that is, there was a period of
negative balance. But this caused the pool of “circulating protein” to shrink,
and after a few more days metabolism decreased to a point where it again
balanced intake.57
Later scholars, considering Voit’s model, have drawn the analogy of a
wash basin with the drain unplugged and water pouring in from a tap. At
one flow rate the basin will remain half full, with the pressure from this
height of water pushing water out at the same rate as it is entering, so that
the system is in equilibrium. If the flow from the tap is increased a little,
the water will at first rise, but the pressure on the exit will increase and
eventually the extra flow out will balance the extra in and a new equilibrium
will result. Obviously there are limits — with a very low input rate none
will accumulate in the basin, and at a very high rate the basin will overflow.
In the light of this model Bowie’s calculations look very different. It could
simply be said that at nitrogen intakes lower than those provided by the
subject’s habitual diet (roughly 105 g digestible protein) the subjects were,
for short (2- to 3-day) periods, still in negative balance because the new
equilibrium had not yet been reached, and conversely for the periods of
higher nitrogen intake. On this view the regression line indicated, by the
point at which it crossed zero balance, what the subject’s habitual intake
was, rather than what his requirement was. If he had habitually consumed,
say, 130 g digestible protein, then placing him for 2 days on 105 g digestible
protein would have put him into negative balance, but would it be reasonable
to say that this amount was therefore inadequate?
One cannot help wondering whether Rubner, who had originally obtained
the data, presumably under the instructions of Voit as his professor, may
have declined to do the calculations of balance, because he had such con¬
siderations in mind, so that the job was then given to Bowie. It is known
that Voit and Rubner sometimes saw things differently and that the younger
man soon moved away to head an independent laboratory at Marburg.58
In any case, Voit never seems to have looked more critically at the results
of human studies of nitrogen balance. As a medical man he had made out
his “prescription,” and he seems for the rest of his life to have made use
of data that appeared to support it and ignored everything that went against
it.
In 1887 a paper was published which demonstrated that gradual adap-

57. Voit (1881), pp. 110-13.

58. Chambers (1952), pp. 4“55 Rothschuh (1975), P- 585-

93
 Protein and Energy

tation to a reduced protein intake could occur in humans exactly as had

been found in dogs. A German physician changed from a high protein diet
to one containing only 42 to 45 g protein per day (i.e., approximately 7 g
nitrogen), with most of it coming from rice, potatoes and beer and a little
from milk and an occasional egg. The values for urinary nitrogen excretion
for the first four days were 9.6, 7.5, 6.8 and 5.5 g, respectively.59 He did
not determine fecal nitrogen content, but we can see that a 2- or 3-day
balance experiment would have been qtnte misleading as to the ability of
a subject to be in nitrogen balance on such a diet. The author pointed out
that in one of Rubner’s earlier experiments, in which he gave a 23-year-
old, 72-kg soldier a diet consisting of potatoes, butter, oil and salt, and
containing 11.5 g nitrogen, the successive urinary nitrogen values for 3 days,
were 12.8, 7.6 and 6.0 g. This indicated adaptation and then nitrogen
retention on a relatively low protein diet.60

Voit and Vegetarianism

Voit had always said that his 118-g standard applied to a mix of animal
and vegetable protein and that a higher quantity would be required if veg¬
etable proteins alone were consumed, since they were generally of lower
digestibility.61 This seemed so different from the claims madehy vegetarians
that he finally took a vegetarian as a subject. He was a 28-year-old pap-
erhanger’s assistant, 162 cm tall, weighing 57 kg and described as “well
muscled.” He had been a vegetarian for 3 years and continued to eat his
self-selected food while living in Voit’s institute for 14 days. During this
period his measured daily intakes were approximately 400 g Graham bread,
125 g pumpernickel, 20 g oil and fruit (apples, dates, figs and oranges). His
diet contained 8.2 to 8.9 g nitrogen, equivalent to 53 to 57 g protein
(nitrogen content x 6.45), and was estimated to provide 2,550 kcal.
Since the subject did not change his diet, Voit was justified in taking the
results for the whole 14-day period. The mean daily intake of nitrogen was
8.5 g and the losses were 8.8 g (3.5 g in feces and 5.3 g in urine), giving a
negative balance of 0.3 g.6i This small value would probably not be regarded
as significant now, given the known errors of the analytical methods. A
modern physiologist might also predict a small diminution of muscle mass,
and therefore a negative balance, because of the inactivity of the subject,
in contrast to his usual working life.
Voit was clearly surprised that the subject had done so well and wondered

59. Hirschfeld (1887), p. 547.

60. Ibid., p. 559; Rubner (1879), pp. 147—8.

61. Voit (1876), pp. 21—2.

62. Voit (1889), pp. 255—65.

94
 Vegetarian Philosophies and Voit’s Standards

if this was because his gut had over a long period adapted its function to
this diet. He tested how well the same diet would be tolerated by a non¬
vegetarian. The second subject is described in a heading of Voit’s paper as
a “worker” (a loaded term to make the contrast with the “vegetarian,” who
was also, in fact, a working man) and then in the text as “the servant in
the Institute, weighing 74 kg, who has been used for most of our balance
studies, and no lover of vegetable foods.” This appears to have been the
same “P.P.” who had been used by Rubner and Bowie 10 years earlier and
who liked to eat sausages and meat dumplings, taking in something like
120 g protein per day.
He received the same quantities of each food that were eaten by the
vegetarian, although he was a bigger man, for a 3-day period. With a daily
intake of 8.25 g nitrogen (equivalent to 53 g total protein or 31 g digestible
protein) his successive daily excretions of urinary nitrogen showed a de¬
clining trend from 11.3 to 9.3 to 8.5 g. Voit took the value of 3.5 g fe¬
cal nitrogen per day and averaged the balance over the 3-day period as
“ — 4.95 g N.” He added that even taking the third day alone the balance
was “ — 3.75 g N,” equivalent to a loss of 24 protein per day from the body.
His conclusion was that “the protein in the vegetarian diet did not in the
least suffice for the heavier and more vigorous worker; it is essential that
he ingests more since he metabolizes 82 g protein per day.”63 Of course, it
would be expected that the energy intake was also inadequate and contrib¬
uted to the negative nitrogen balance.
Analysis of the feces from the two subjects showed no significant differ¬
ences in the digestibility of the vegetarian diet. In each case the fecal excretion
of nitrogen corresponded to just over 40% of the intake. There was therefore
no evidence of a long-term adaptation of the digestive system to such a
diet.64
Voit wrote that the vegetarian metabolized only 39.5 g protein per day.
I have not been able to find his basis for this particular value; it may come
from urinary nitrogen plus a 5-g allowance for protein secreted into the gut
and recovered in the feces. Voit accepted that this was an extremely low
value, even for a man of only 57 kg, since in proportion to body weight it
corresponded to only 51 g protein for someone of 74 kg, the weight of the
second subject. His explanation for this finding was that in some way the
high starch content of the vegetarian diet had reduced the need to metabolize
protein. The vegetarian’s mean intake of carbohydrates was 560 g, with
sugars accounting for slightly more than starch, so that it was really only
high in “total carbohydrates.” Analysis of the diet also indicated that it
contained 20 g fat, of which 15 g were digested. From these values Voit

63. Ibid., pp. 2.71—5.

64. Ibid., p.,276.

95
 Protein and Energy

Table 5.2. Nitrogen balance in a subject eating different quantities of

white bread

Eaten >. '

White N out, Balance,

Day bread Carbohydrate N (protein) urine (feces) “ N'(protein)

1 698 540 10.5 (68) 12.8 (2.0) -4.3 (-27.7)

2 834 645 12.5 (82) 12.4 (2.4) -2.3 (-14.8)
3 1,068 825 16.1 (104) 12.4 (3.0) + 0.7 ( + 4.5)
Total 2,600 2,010 39.1 (254) 37.6 (7.4) -5.9 (-38.0)

Note: Data are expressed as grams per day.

“Only the total nitrogen content of the feces from the whole experiment was actually
analyzed, and the total has been divided into the proportions of bread eaten each
day.

calculated that the diet provided altogether 2,550 kcal. In relation to his
size this was some 20% higher than the calorie value of the average diet of
workers at rest, and the proportion of the calories coming from carbohy¬
drates (88%) was also considerably higher.65
He attempted to substantiate this hypothesis, namely, that carbohydrates
reduced protein breakdown, by selecting some i-day values from Rubner’s
old digestibility experiments. Because Rubner had himself been interested
only in the digestibility of a food that had not been thought to vary with
the quantity eaten, he had sometimes given different amounts on different
days according to how much the subject could consume.66 On 3 ^recessive
days of a trial using white bread, the quantities were increased considerably,
and the results were as shown in Table 5.2. As we can see, the subject was
responding in the first 2 days in the usual way to a decline in protein intake
(i.e., from his habitual 120 g or so), but he was then fed an increased amount
and was presumably taking time to react to that. The small positive balance
on the third day (which is also the first in which he consumed 104 g protein)
may be another transient effect. The carbohydrate content is certainly also
increased, but that is not the only variable. Voit seems here to be grabbing
at straws in concluding that the improved balance on day 3 was due to the
extra starch eaten.
At any rate he sounded confident in his conclusions:

It may be possible for a man to maintain himself in nitrogen balance for some time
with a lower intake of protein and a great deal of starch, but such a diet carries

65. Ibid., pp. 2.64-5, 276.

66. Rubner (1879), pp. 153-4.

96
 Vegetarian Philosophies and Voit’s Standards

disadvantages with it which require one, over the longer term, to return to a diet
richer in protein and poorer in starch. It is an interesting fact for physiologists but,
from a hygienic point of view, one has to aim always in the direction of using less
starch and more digestible vegetables and more protein.... The vegetarian exposes
himself to disadvantages when he eats a large volume of poorly digested vegetables.
It is possible to compose a sufficient and healthy diet just with vegetables for someone
with a healthy digestive system, but it is easier with a mixed diet which has a balance
of meat.67

It is sad to read these unsubstantiated statements from a man who showed

such care and precision in his experimental work. Presumably he felt that,
as an “authority,” he had to sound “authoritative” even when none of the
data that he had collected were relevant to the issue at hand. Hindhede, an
early critic who analyzed the work in detail, wrote:

Inasmuch as Voit and his pupils did not carry out a single experiment with a moderate
worker on less than 118 g protein... for any acceptable period of time... he could
not have had the slightest notion to what extent such a high standard is necessary.
... When confronted by the question as to what ought to be the daily allowance of
protein if a human being is to continue to live in perfect health and vigor, no attention
must be paid to Voit.68

Hindhede himself had studied several subjects in Denmark who had shown
strength and endurance despite living on low-protein, largely vegetarian
diets (Figure 5.4).
Nevertheless, Voit’s standard, implying the necessity of high protein in¬
takes by vigorous men, continued to be accepted by most textbook writers
for the remainder of the century and were also made use of by economists.
An Italian writer, on the basis of data indicating that the average worker
in Naples in the 1890s consumed only 70 g protein per day, concluded: “It
is lack of albumen which renders them so idle, so apathetic, and often so
absolutely degenerate.”69 He also quoted from a report which indicated that
French laborers had been less productive than the Englishmen working with
them in the construction of the Paris-Rouen railway line, until they were
given the same portions of roast beef. A few tens of thousands of well-fed
English carnivores were, he wrote, able to hold in subjection a hundred
million vegetarian Hindus. It was true, he added, that experiments had
shown that for short intervals of time people could remain in equilibrium
on small quantities of protein, but these experiments were too brief to show
the statistical effect of such quantities on the collective life of peoples: “low
stature, a low power of resisting disease and slackness in muscular vigor.”70

67. Voit (1889), p. 288.

68. Hindhede (1913), P- 3°-

69. Nitti (1896), p. 62.

70. Ibid., pp. 31, 49—52..

97
 Protein and Energy

Figure 5.4. Photograph of a 54-year-old Dane who lived on a lo^-protein

diet and cycled 363 miles in 37 hours over boggy roads in a timed trial.
(Hindhede, 1913)

The striking claims of vegetarians were therefore disregarded by the or¬

thodox or brushed aside on the grounds either that the subjects were “not
real workers” or that the results were “short-term phenomena.” But this
situation was not to continue much longer, as will be described in the next
chapter.

Summing Up

Vegetarianism was originally adopted by some Westerners early in the nine¬

teenth century on the religious or philosophical grounds that it was wrong
to kill. U.S. vegetarians later urged that their diet was physically advanta¬
geous because it was less stimulating. Kellogg and others argued that it was
the high protein content of a meat diet that made it undesirable, both because

98
 Vegetarian Philosophies and Voit’s Standards

it was overexciting and because undigested protein in the gut gave toxic
breakdown products.
Meanwhile, on the basis of a study of what healthy people unconstrained
by poverty actually ate, Voit, in Germany, proposed a standard of 118 g
protein per day for an “average worker.” A vegetarian subject consuming
only 5 5 g protein per day was found to be in virtual nitrogen balance, but
Voit rejected such a diet because it relied on too high a starch content, and
his 118 g standard remained generally accepted.

99
6 Chittenden versus the U.S.
Establishment, 1883—1912 i

As we have seen already, U.S. scholars wishing to gain advanced

training in chemistry or nutrition in the nineteenth century typically spent
one or two postgraduate (or postdoctoral) years in a German research
laboratory. Wilbur Atwater, who has been described as “the founder of the
science of nutrition in the United States,” was one of these. Born in 1844,
the son of a New England minister, he obtained a Ph.D. at Yale in 1869
for a study of the composition of corn (maize) and then spent z years at
Leipzig and Berlin. In 1874 he became professor of chemistry at Wesleyan
University in Connecticut, but had to struggle to find funds fot the research
in which he was interested. However, in 1879, with the help of grants from
the Smithsonian Institution and the U.S. Commission of Fish and Fisheries,
he began a systematic study of the composition of North American foods
and their economic value as sources of nutrients.1

Atwater’s Valuations of Foods

In his report to the commission, Atwater included 50 analyses of seafoods

(fish, oysters, lobsters, etc.) for water, albuminoids, fats, minerals and ex¬
tractives.2 He had followed what was being adopted as the “proximate”
system of food analysis - the water (or moisture) value being the loss of
weight of the sample after heating at ioo°C; the “crude fat” or “ether
extract” value being the weight of material extractable with ether from the
dried sample; and the “ash” or “mineral” content being the residual weight
after the sample had been held in a furnace until the organic matter burned
off. The “crude protein” content of the original sample was determined by
the Kjeldahl procedure, that is, heating with concentrated sulfuric acid and
a catalyst to convert the protein nitrogen to ammonium sulfate, then cooling,

1. Maynard (1962), pp. 3-4; Aronson (1982a), pp. 465-6.

2. Atwater (1883), p. 264.

IOO
 Chittenden versus the U.S. Establishment

adding excess alkali, distilling the ammonia and measuring it by titration.

Crude protein was estimated as ammonia nitrogen multiplied by 6.2.5.
Finally, if these various fractions determined for an animal food together
came to less than 100%, the shortfall was called “nitrogen-free extractives”
(NFE) and considered to represent the small amount of carbohydrate pres¬
ent. For vegetable materials an additional analysis was carried out — the
dried, fat-extracted sample was boiled with dilute acid, then with dilute
alkali and finally filtered. The insoluble residue was dried and weighed, then
kept at red heat in a furnace, cooled and reweighed. The material lost on
ashing was referred to as “crude fiber.” (Materials like cellulose fall into
this category.) NFE was calculated as the shortfall from 100% for the total
of all the other analyses including that for crude fiber.
Atwater compared his analyses for fish with German analyses for different
types of meat and dairy products. He used a German method for expressing
the relative economy of different foods.3 Konig had found that in Germany
it cost approximately three-fifths as much to buy 1 lb. of fat as 1 lb. of
protein, and one-fifth as much to buy a pound of carbohydrate. He therefore
prepared an index of the economic value of a particular food as

protein content + V$ x fat + V, x carbohydrate

He then obtained relative values by comparison with average beef, which

he used as a standard and gave a value of 100. We can follow the principle
with three examples:

Protein Fat Extractives Economic Relative

Food (%) (%) sum value
(%)

Beef 21.4 5.2 — 24.5 100

Halibut 18.2 6.3 — 22.0 90
Oysters 5.8 1.1 3.4 7.1 29

By Atwater’s calculations, halibut would have to cost less than 90% of

the price of beef before it became an economic choice, and oysters would
have to be very much cheaper. In a paper read to the American Association
for the Advancement of Science in 1884, he expanded the list of foods to
include beans, grains and potatoes and showed that, in these terms, they
were more economical than nearly all animal or fish products.4
In commenting on his analyses of fish, Atwater had pointed out that true
comparisons between fish and meat also depended on their relative diges¬
tibility. One suspects that it was the U.S. Fish Commission that gave him
funds to carry out such experiments in the winter of 1882-3 at Voit’s

3. Konig (1876).
4. Atwater (1884).

IOI
 Protein and Energy

Institute of Physiology in Munich. He used the flesh of haddock, which

contained 17% crude protein and only 0.1% crude fat. A test dog first
received the fish with a little added fat for 6 days, and then beef steak for
another 6 days. In a second experiment a medical student volunteer lived
for 3 days on the fish, with a little butter, vinegar and Worcestershire sauce,
followed by 3 days on beef, the technique being that described earlier, in
Chapter 5. In all four experimental periods the digestibility of the test protein
was calculated to be 98%, regardless of whether it was man or dog eating
fish or beef.5
Probably no one had doubted that fish was a class of well-digested food,
so that the direct results of Atwater’s work in Munich are not very impor¬
tant. However, Atwater’s admiration for Voit’s techniques and his adoption
of Voit’s ways of thinking were to have considerable consequences for the
direction of nutrition research in the United States.

Definitions and Standards

Atwater’s next project was to collaborate with the Massachusetts Bureau

of Labor Statistics in analyzing the findings from a study of the foods
consumed by different groups of workers in that state. The final report was
obviously written by Atwater, although no author is stated and he is thanked
only in the introduction of the volume of the bureau’s annual reports for
1886.6
He began with a general discussion of the components of foods, and with
regard to protein he wrote that it was in general the most important, as
well as the most costly, of the nutrients. It served to repair'the muscles,
tendons, skin and other organs that were being worn out by constant use,
but that part was probably transformed into fat and into glycogen, a car¬
bohydrate that occurred in the liver.7 He added that the dietary standards
proposed by Professor Voit were more commonly accepted than any others.
They had been estimated in two ways: first, by observing the amounts
actually consumed by people whose life circumstances would permit rea¬
sonably good nourishment and at the same time preclude any considerable
waste of food and, second, by direct experiments.
Atwater gave Voit’s standards as 100 g protein per day for an old man,
118 g for a laborer performing a moderate amount of work and 145 g for
one doing arduous work. The corresponding total energy standards for these
classes were 2,400, 3,000 and 3,300 kcal, respectively.8 He commented that

5. Atwater (1888).

6. Anonymous (1886), p. 239.

7. Ibid., pp. 243-4.

8. Ibid., p. 262.

102
 Chittenden versus the U.S. Establishment

these figures represented only general averages, but that if an ordinary

laboring man, doing an ordinary amount of work, had much less protein
and kept up his muscular exertion, he would be apt, sooner or later, to
suffer.9

The First U.S. Dietary Surveys

Atwater went on to summarize analyses of the foods purchased by different

social groups. Some groups comprised families, and others consisted of
workers living in boardinghouses. The values were then expressed as “per
man per day” on the assumption that women’s needs, on average, were
80% those of men; those of children under the age of 2 were 25%; those
between 2 and 6 were 50%; and those between 6 and 15 were 70%. Children
over the age of 15 were assumed to have adult requirements.10
The averages of results obtained from three boardinghouses were 126 g
protein (52% from animal sources), 192 g fat and 539 g carbohydrate per
man per day. This we can equate to a total of nearly 4,400 kcal. More
controlled studies of food consumption in later years have led us to expect
a maximum need of about 3,200 kcal per man. Of course, more physical
work was required during that period, but even in a boardinghouse where
the occupants were “operatives in a shoe factory, clerks and dressmakers
the daily food purchases were equivalent to 3,900 kcal per man, and in a
household where the husband was a skilled printer (i.e., did not perform
heavy manual labor) the corresponding value was 4,100 kcal.11 One can
imagine that at a boardinghouse the staff also ate from the general stock
and even took home to their families portions that could be regarded as
“surplus,” but this would not apply to a workingman’s family. Atwater was
surprised that the fat values were so much larger than those in the records
accumulated in Europe: 190 to 200 g versus 40 to 70 g for European men
considered to be adequately fed, or 56 to 100 g by Voit’s standards. He
thought that the overall waste of food was probably higher among the
working class in Massachusetts than in Europe, and that this reflected their
being better off, but that nevertheless true consumption of food was also
higher.
Atwater noted that it was a common practice of the poor to purchase
more expensive foods, especially meats, when “food obtainable at only a
fraction of the cost would be equally wholesome and nutritious. Here
he was referring back to tables included in the preliminary section of his

9. Ibid., pp. 263-5.

10. Ibid., p. 267.
11. Ibid., pp. 290, 306-9.
12. Ibid., pp. 324-6.

103
 Protein and Energy

paper showing that an expenditure of 25 cents on different animal foods

bought very different quantities of protein: beef sirloin, 68 g; round of beef,
132 g; salmon, 18 g; smoked herring, 380 g; and skim-milk cheese, 545

§/3
In a series of articles that appeared in the Century Magazine over the
next two years, Atwater elaborated this theme. He referred to a coal laborer
who “made it an article of faith to give his family the best of flour, the
finest sugar and very best quality of meat.... [and] spent $156 per year on
the meat alone and only $72 for rent in a crowded tenement-house where
they slept in rooms without windows or closets.”14 Since food was the
principal item of expense for most people, economies here would allow for
better housing and the possibility of building a reserve of savings. *
Atwater had been hurt by the president of Wesleyan University having
told him that his type of work “was not in consonance with the intellectual
dignity of a university.” He referred to this in his article and replied with
an emotion rarely seen in his writing:

The place of the scholar, as of the saint, was once that of the recluse; now they are
both busy among their fellow-men and doing their best to help them_At the great
universities which are the fountain heads of knowledge, speculative philosophy and
technology, Sanskrit and sanitation, are studied side-by-side with equal intellect and
ardor-This is following the precept and the example of the great Teacher... a
part of whose work on Earth was to feed the hungry and heal the s‘ick.15

It seems that Atwater was not the only one the president had offended. By
the time Atwater’s article appeared he had been “egged” by the students
and described in their newspaper as “odious,” and on the combined rec¬
ommendation of the professors and a committee of alumni, the trustees had
dismissed him.16
Naomi Aronson, a sociologist, has argued that, in fact, Atwater’s writing
went against the interests of the working class. The readers of his articles
were mainly middle-class people with an interest in social problems, and
the impression they would have gained was that the poor could help them¬
selves and manage quite well if only they would eat more sensibly. This
would, in turn, have reduced their willingness to support workers’ demands
for higher wages. The hidden implication was that workers had no “enti¬
tlement to the pleasures of the table, with a variety of fruits and vegetables,
but only to the tougher cuts of meat requiring extra hours of cooking by
the woman of the family.17

13. Ibid., p. 254.

14. Atwater (1887-8), p. 437.

15. Ibid., pp. 445-6.

16. Peterson (1964), pp. 115-18, 228-9.

17. Aronson (1982a), pp. 482-3; Aronson (1982b), p. 52.

104
 Chittenden versus the U.S. Establishment

It may well have been the case that some large employers thought nu¬
tritional science” desirable because it would show that workers could live
more cheaply. Yet I believe that Atwater was sincerely trying to provide
poor families with a wider range of alternatives for spending their money
and that it never occurred to him that his writing might have a negative
effect.

Federal Funding for Nutritional Studies

In 1887 the Hatch Act was passed in the United States, providing for an
agricultural experimental station in each state. In the next year Atwater was
appointed the coordinator of their work and served in Washington for the
next three years. His personal interest was, of course, particularly in nu¬
tritional studies, and special federal funding for these began in 1894 with
Atwater as coordinator. He remained in Connecticut, but stimulated people
in different regions to make dietary surveys, coordinated their work and in
many cases published his comments as an appendix to their findings.
In 1891 Atwater explained his belief that U.S. standards should be higher
than those developed in Europe. First, his subjects apparently ate more,
even after allowing for waste. Second, the European standards were based
on people whose “plane of living” was lower that of people in the United
States, and “to bring a man up to the desirable level of productive capacity,
to enable him to live as a man ought to live, he must be better fed.”19 His
new standards, including “125 g for a man at moderate muscular work,”
were “a compromise between the currently accepted European standards
and the actual dietaries observed in New England.’”0 Atwater added that
in the United States many people were eating well above their needs and
might be damaging their health as a result.11 He admonished the farming
community for producing overfat meat and mainly low-protein plant crops.
He felt they should be concentrating on the production of lean meat and
legumes.11 ,.
Typical of Atwater’s comments on the results of the dietary studies were
those he made on results from Purdue University. For three mechanics’
families the average daily intakes per man unit were calculated to be 105
g protein and 3,570 kcal. Atwater wrote that “the so-called dietary standards
are for the most part based on the observed facts of food consumption...
[and] are to be understood to represent simply tentative estimates... and

18. Maynard (1962.), pp. 4-5-

19. Atwater (1891), p. 14s-

20. Ibid., p. 149-

2.1. Ibid., p. 161.

22. Ibid., pp. 168—71.

105
 Protein and Energy

in no sense to be considered exact or final... but viewed from this point

the dietaries are deficient in protein.”13
In most studies people’s diets were found to be at least in the neighborhood
of the standards, which is not surprising since Atwater had adjusted the
standards in order to make them closer to the intakes found in his first
studies. The great exceptions were the families of black farm laborers in
the southern region of the United States. Here the food purchased consisted
mainly of fat pork, cornmeal, wheat flour and molasses. In addition, some
of the families grew their own greens and owned a cow that yielded milk.
For 18 families the average intake (per “man unit,” as usual) was 6z g
protein and 3,270 kcal, compared with average intakes for 50 families in
the poorest parts of Philadelphia and Chicago of 114 g protein and 3,330
kcal. As a northerner, Atwater probably felt it politic to make little comment;
he therefore limited himself to stating that more extended surveys were
needed and that “while the diet of the negro in the South is a very important
fact of his character and condition, its effect is hardly to be separated from
that of the other conditions of his existence.”14
Atwater’s biographer has pointed out that he maintained his high protein
standards even though he knew that men could remain in nitrogen balance
on 44 g protein per day and also that muscular work did not increase
nitrogen losses provided that energy intake was adequate.15 He was probably
influenced by a German paper that he reprinted, in translation, in the Ex¬
periment Station Record. There Professor Zuntz argued that although
steady, moderate physical work did not increase the need for dietary protein,
intensive work over a short period, which led to a deficiency of oxygen in
the tissues, resulted in increased metabolism of protein.16
Atwater referred to Zuntz’s idea when commenting on the results of
dietary studies of university boat crews. Their daily intakes had averaged
15 5 g protein and 4,085 kcal. He concluded that the uniformly high intake
of protein by athletes in training must indicate a real demand, rather than
individual idiosyncrasy, and that this seemed to confirm “a view coming to
be entertained by many physiologists that metabolism... is regulated not
simply by muscular work but also by the nervous effect required for its
performance.”17 A more skeptical reviewer might attribute the high protein
intakes of athletes, first, to their eating more to meet their higher requirement
for energy, and thus necessarily to their consuming more protein also if they
maintained a normal dietary pattern, and, second, to their being fed at a
special table and encouraged by their coaches to eat meat in the belief that

23. Atwater and Woods (1896).

24. Atwater and Woods (1897), pp. 22, 64-5.

25. Maynard (1962), p. 7.

26. Zuntz (1897), pp. 549-50.

27. Atwater and Bryant (1900), pp. 69-71.

 Chittenden versus the U.S. Establishment

this would develop their muscles. Atwater’s observation that this was hap¬
pening, and his conclusion that it represented demand, would thus encour¬
age coaches to continue the practice in a self-fulfilling circle.
Atwater had other nutritional interests. He and Edward Rosa, the pro¬
fessor of physics at Wesleyan, built and operated the first human calorimeter
in the United States, in which heat output could be measured directly, as
could the balance of gas exchanges. They used this to measure the me¬
chanical efficiency of human work, including mental activity, and the use
of alcohol as an energy source/8 In 1904 Atwater suffered a stroke and
died 3 years later.

Professor Chittenden’s Doubts

Russell Chittenden, the next important contributor to our story, was born
in 1856. Like Atwater, he grew up in Connecticut. His family was not well
off, but he managed to study at Yale University, which was near his home,
by working as a technician in the chemistry laboratory and assisting with
lecture demonstrations. For his bachelor’s degree he carried out a small
research project on the composition of scallops, showing that they contained
free glycine. This was an unexpected finding, and his paper was accepted
for publication in Germany. This, in turn, enabled him to spend a year in
a leading laboratory in that country. He was then employed again in teaching
at Yale and became the country’s first professor of physiological chemistry
in i88z, at the early age of 25.Z9
For the next 20 years Chittenden’s main, ongoing line of research involved
the action of different digestive enzymes on food proteins. He had not been
directly concerned with dietary standards for humans. However, at the turn
of the century, considerable interest had been aroused by the claim of a
wealthy, middle-aged American, Horace Fletcher, that if one chewed every
morsel of food eaten until it was in very fine suspension, the quantity of
food one required would be much reduced. Fletcher offered to serve as a
volunteer subject for research at Yale and spent some months as Chittenden’s
houseguest in 1902. Fletcher’s physical condition was certainly exceptional,
and the amount of food, including protein, that he ate was remarkably
small. Chittenden thought that the main effect of the prolonged chewing
was to reduce appetite rather than to increase digestibility. The observations,
however, made it seem doubtful that dietary requirements were really as
high as was currently believed, and Chittenden decided to investigate the
matter in a systematic way.30

28. Maynard (1962), pp. 6—9.

29. Vickery (i94S)> PP- 59“68-

30. Ibid., pp, 80-1.

107
 Protein and Energy

In November of the same year he gradually reduced his overall intake of

food, particularly that of meat, cutting out breakfast altogether and taking
only a light lunch. Over the next 7 months, he lost 7 kg of his original
weight of 65 kg, but felt in better health, was free of headaches and “rheu¬
matic trouble in the knee joint” and fully maintained his enfergy for both
mental and physical activity. His weight from then on remained stable.31
Chittenden did not refer to any calculations of his protein intake during
this period, but it seems that he must have made them. In the next year,
when he continued the same type of diet, his published values showed a
mean intake of approximately 40 g protein per day. Allowing for his rel¬
atively low weight, this was still equivalent to no more than 48 g for the
“average man” weighing 150 lb. (68 kg), compared with Atwater’s standard
of 100 g for even an “aged man.”

Human Studies

Chittenden was sufficiently impressed by the importance of his finding to

organize quantitative studies starting in the fall of 1903 with three different
groups of subjects. In the first group were Chittenden himself, along with
four colleagues at Yale, all busy, but at work not requiring muscular
strength. The second group consisted of 13 soldiers (corpspien normally
employed in army hospitals). During their 6 months at Yale their daily
routine included z hours in the gymnasium and another hour of drill. The
third group, which started in January 1904 and spent 5 months in the study,
comprised seven leading student athletes at Yale who continued to train
and compete in their sports.
Every subject’s urine was collected daily throughout the trial and analyzed
in duplicate for its total nitrogen content by the Kjeldahl procedure. For
the first 1 to 2 weeks all the subjects remained on their habitual diet. Then
they gradually reduced their protein intake and, to a lesser extent, their
total energy intake. Also, typically for two periods of 4 to 7 days each, they
made a complete record of the weight of each food item that they ate; their
feces were also collected and analyzed for nitrogen. The nitrogen intake was
also calculated after analysis of each of the foods eaten, and energy intakes
were estimated from Atwater’s published tables of values for different foods.
The way in which Chittenden handled the data and drew conclusions can
be illustrated for a single subject, 32-year-old Lafayette Mendel, whose later
work will be discussed in the following chapter. At the beginning of the
study he weighed 76 kg; in the following 3 rA months his weight fell gradually
to 70 kg and then remained steady for the remaining 4 months.31

\
31. Chittenden (1904), pp. 19—2.2.

32. Ibid., pp. 5 2-76.

108
 Chittenden versus the U.S. Establishment

Table 6.i. Food eaten by Mendel in one day of Chittenden’s study and
its calculated nitrogen content

Food Weight of food eaten (g) N (%) N (g)

Bread 58 + 120 + 67.5 1.71 4.20

Coffee (plus milk) 210 + 107 + 210 0.11 0.58
Sugar 21 + 21 + 21 — —
Custard — 76 — 0.83 0.63
Milk — 250 — 0.48 1.20
Potato — — 150 0.37 0.56
Lima beans — — 80 0.90 0.72
Apple dumpling — — 131 0.72 0.94
Candy 27
8.83
Total

In the middle of October, Mendel began to modify his diet, mainly by

reducing and then eliminating meat, but urine collection and analysis started
only on October 26. The average daily urinary nitrogen content was about
iZ g during the first week of collection, equivalent to the metabolism of
about 75 g protein. It continued to fall for another week, then from No¬
vember 10 to the end of the study, on June 23, the mean daily excretion
was 6.53 g, which Chittenden considered equivalent to the metabolism of
only 41 g protein per day.
The following daily average results were obtained during the two balance
periods:

Period I Period II
(Feb. 9-14) (May 18-24)

1,975 2,448
Energy intake (kcal)
7.83 9.19
Ingested N (g)
-7.51 -6.31
Urinary N (g)
-1.48 -1.50
Fecal N (g)
-1.19 + 0.38
Balance (g)

The diet varied considerably from day to day; estimated energy intakes
varied from 1,700 to 2,300 kcal in the first period and from 1,900 to 3,200
in the second. Daily intakes of nitrogen ranged from 6.5 to 10.4 g. A typical
day’s food record in the first period for Mendel’s three daily meals was
calculated to contribute 1,930 kcal and consisted of the foods shown in
Table 6.1. . . ,
Chittenden commented that the negative balance in the first period was
probably due to inadequate energy intake, which resulted in an increase in

109
 Protein and Energy

tissue protein metabolism. He added that having to weigh and record food
intake led unconsciously to a reduction in the quantity and variety of food
eaten. He pointed out that Mendel had lost 1.2 kg in weight during this 6-
day period, and certainly the records for the weeks preceding and following
the balance period show lower mean values of 6.82 and 6.45 g,respectively,
for daily urinary nitrogen.
He concluded that Mendel, who was a big man and on his feet all day
working in the analytical laboratory, clearly needed more than 2,000 kcal/
day to be in energy balance. Therefore, the results from the second balance
period, when his energy intake was greater and he kept his weight constant,
were more representative. He further concluded that with Mendel being in
nitrogen balance when he had a urinary excretion rate of 6.31 g nifrogen
per day, it was safe to conclude that his average daily metabolism, over a
period of 7 months, of 41 g protein (i.e., 6.53 g nitrogen x 6.25) was fully
adequate to maintain his health and strength. Allowing for the loss to the
feces of 18.3% of the ingested nitrogen, the actual total intake needed was
50 g, which corresponded to 0.715 g protein per kilogram body weight.
The reader will notice that no allowance has been made for daily loss of
protein from the body by way of hair and nail growth, sweat or rubbing
off of skin. These miscellaneous losses are thought (as we shall see in Chapter
11) to be no more in normal circumstances than 0.03 g protein per kilogram
body weight. After allowing for partial indigestibility as before, this would
raise Mendel’s estimated requirement slightly to 53 g, or 0.753 g/kg. but
this is still less than one-half of Atwater’s standard.
The records for the other four subjects in the first group indicate that
they all remained healthy and vigorous on estimated protein intakes (ad¬
justed to the standard 68 kg body wt) of 46, 58, 74 and 70 g/daf, respec¬
tively. We do not, of course, know whether they would have remained as
fit on even lower intakes.

Soldiers and Athletes

Eleven soldiers completed the study.33 Two dropped out, apparently finding
it intolerable that they were not allowed to go out on the town in their free
time. The restriction was to prevent unrecorded consumption of food and
drink. Over the study, the subjects’ average weight fell very slightly, from
62.1 to 61.4 kg, and all the men showed an improvement in physical con¬
dition as assessed by their medical officer, and from the results of strength
tests (Figure 6.1). The level of hemoglobin in their blood was also fully
maintained.
The soldiers’ urinary excretion values for the first 2 weeks, when they

33. Ibid., pp. 131—326.

no
 Chittenden versus the U.S. Establishment

Figure 6. i. Photograph used by Chittenden to illustrate the good physical

condition of his subjects after 5 months on a low-protein diet. (Chittenden,
1904)

were still on regular army rations, indicated an average protein consumption

of about 105 g/day. Once the diet pattern was fully changed, the daily intake
fell to about 55 g/day (which corresponded to 61 g for the standard 68-kg
man) and the individuals’ average urinary nitrogen excretion per kilogram
of body weight ranged from 0.106 to 0.150 g. This was similar to the range
observed in the first group.
In the first balance period, the men were all in negative balance, but the
energy intake was only about z,ooo kcal/day. By the second study period,
intakes had increased to 2,500 kcal and there was an average positive
balance of 0.6 g nitrogen. The one man in negative balance was described
in the gymnastic report as being “nervous, irritable and aggressive.” If an
uncooperative subject were to scrounge extra food from the kitchen, this
would of course increase his urinary nitrogen values and make him appear
to be in negative balance in relation to his recorded intake. This is at least

hi
 Protein and Energy

a possible explanation, since this subject did not lose weight over the trial
and gained in strength.
The third group, consisting of student athletes, were in the study for only
5 months (from mid-January to mid-June).34 In the initial week, while still
on their usual diet, their urinary nitrogen values indicated they were taking
in, on average, about 130 g protein per day. By the last 2 months the level
of urinary excretion had almost halved. On average the men lost 2.7 kg in
body weight during the trial, mostly in’the first 2 months. However, they
said that they felt better, their strength tests improved and their athletic
performances were at least as good as previously.
Only one balance study was carried out with this group. From the results,
two of the seven subjects appeared to be in negative balance. Chittenden
again attributed this to these two having reduced their intake as a conse¬
quence of the measuring required. Certainly this effect was seen in the first
balance periods of the other groups, for whom there was more than one
balance study. From the data as a whole, it appeared that 64 g protein
would be adequate for an average 68-kg man engaged in athletics. The
corresponding Atwater standard would again be roughly double this value.
We see, therefore, that the three groups gave very similar results, and
Chittenden felt justified in concluding that people could safely live on much
less protein than the standards set out by Voit and Atwater. Further, they
did not require excessive quantities of carbohydrates in order to achieve
nitrogen balance, as had been argued by Voit in his paper on vegetarianism
that was discussed in the preceding chapter. Chittenden’s second conclusion
was that low-protein diets were actually more healthy than high-protein
diets. He believed that this was because the body had smaller quantities of
uric acid and other protein breakdown products to dispose off which ex¬
plained the general feeling of well-being experienced by the subjects in the
trials.35

The Great Debate

Chittenden’s publication made a considerable stir, even though his conclu¬

sions were not entirely original. As we saw in the preceding chapter, ve¬
getarian groups, including the Seventh Day Adventists, had long claimed
that their low-protein diets kept them in superior health. But orthodox
scientists and physicians could, and did, shrug off such claims on the grounds
that they came from enthusiasts who reported mainly their feelings and were
therefore unscientific. Also, although several scientists had reported exper-

34. Ibid., pp. 327-453.

35. Ibid., pp. 454-63.

112
 Chittenden versus the U.S. Establishment

iments on themselves in which they remained in nitrogen balance on low-

protein diets for periods of i to z weeks, it was easy to consider this a
purely short-term phenomenon.
The work at Yale could not be dismissed or ignored so easily. Professor
Chittenden was himself a member of the scientific establishment and had
presented his results to the National Academy of Sciences. He had also
worked with physically active men, as well as relatively sedentary academics.
In addition, he had continued the studies for many months and accumulated
quantitative data on strength and physical performance, as well as on ni¬
trogen balance.
By 1906 Francis Benedict had taken over Atwater’s program, and in a
paper published in that year he paid tribute to the quality of Chittenden’s
work. It was “scientific research of the highest type, and for the most part
defying criticism.”36 However, he was still unwilling to accept his conclu¬
sions. With regard to the soldiers, he felt that their steady routine with
regular moderate exercise and no opportunity for dissipation would have
led to an improvement in physical condition even if their diet had been
unchanged. No individual was worked to a condition of fatigue, when the
wear on tissues was out of proportion to the work performed. With regard
to the athletes, he wrote that it was difficult to understand why, if the low-
protein diet had really suited them so well and helped their performance,
they had almost all (so he had heard) reverted at the end of the study to a
more traditional diet rich in meat.37
As he and Atwater had written earlier, “A man may maintain bodily
equilibrium on either a higher or lower nitrogen level.” The answer as to
which was most advantageous “must be sought in broader observations...
efficiency, health, strength and welfare.”38 He, Benedict, remained more
impressed by the observation in so many countries that men engaged in
heavy work consumed considerably more protein than those engaged in less
physically demanding activities:

Dietary studies made in England, France, Italy, and Russia show that a moderately
liberal quantity of protein is demanded by communities occupying leading positions
in the world.... It certainly seems more than a remarkable coincidence that peoples
varying so widely in regard to nationality, climate and geographical conditions...
should show such agreement.39 The negro and poor white of the South, the laborers
in Southern Italy, all partake of diets relatively low in protein. That their sociological
condition and commercial enterprise are on a par with their diet, no one doubts.

36. Benedict (1906), p. 424.

37. Ibid., pp. 418—19, 421—2.

38. Atwater and Benedict (1902), p. 130.

39. Benedict (1966), p. 416.

 Protein and Energy

when people accustomed to a low protein diet are fed on a higher protein plane,
as is the case when Southern Italians come to America, their productive power
increases markedly.40

Another writer expanded on this point:

The only people who fail to attain the ordinary [Voit] protein standard are those
who are too poor to afford the cost.... As soon as financial considerations are
surmounted, the so called ‘vegetarian Japanese’ or Hindu raises his protein intake
to reach the ordinary standard of mankind in general.... If Chittenden is right, then
all the world up to this time, with the exception of a few faddists, has been wrong.
It is ‘Chittenden against the world’ and it is inconceivable that all mankind, under
the most diverse conditions, should have fallen into the same mistake.4 *

Criticism from India

The most systematic attack came from D. McCay, a British professor of

physiology in Calcutta. He had been greatly impressed by the differences
in physique among members of the various communities in India. In par¬
ticular, the typical Sikh in the Punjab was muscular and vigorous in tem¬
perament, whereas the Hindu Bengali was usually, in his experience,
relatively flabby, lethargic and incapable of performing a hard day’s work.
He related these differences to the characteristic diets of the two groups.
He found that the Sikh community, with wheat as its staple grain, consumed
generous quantities of meat and milk, so that an adult male could receive
as much as 135 g protein per day. By contrast, the average Bengali’s rice
diet was supplemented with only small quantities of animal protein, and
men typically received only 54 to 6 5 g protein per day. By £hittenden’s
standards this intake should have been healthful, and indeed optimal, but
in fact the Bengalis were not virile and energetic.41
McCay accepted that men could remain in nitrogen balance on low levels
of protein. Bengalis, for example, must do so because they spent a lifetime
on such diets. But whereas these low amounts were adequate for existence,
the question remained as to what quantity was required for maximum
efficiency. McCay reprinted a quotation: “Energy is not to be confused with
muscular strength... [it] is a property of the nervous system-Muscles
give us the power to work; the nervous system gives us the initiative to start
it_If protein be regarded as a nervous food... it is not without reason
that the more energetic races of the world have been meat-eaters.”43
McCay, too, praised Chittenden for his work and for doing so much to

40. Ibid., pp. 427-8.

41. McCay (1912), p. 102.

42. Ibid., pp. 78-9, 90-2, 153-6, 203-4.

43. Ibid., pp. 106-7, x53-

114
 Chittenden versus the U.S. Establishment

awaken interest and lift the subject of dietetics from the playground of
quacks and charlatans.”44 But he went on to subject Chittenden’s results to
detailed examination. He first criticized the use of urinary nitrogen as a
direct measure of protein metabolized. According to further results pub¬
lished by Chittenden in 1911, six men working in his laboratory had changed
to a low-meat diet; both their food intakes and total excreta had been
recorded and analyzed continuously for 18 weeks.45 All gave similar results,
the mean values (expressed as grams nitrogen per day) being as follows:
Intake 12.02
Fecal loss -1.36
Urine loss -9.16
Apparent balance + 1.50

McCay said that these results demonstrated that urinary nitrogen gave
an underestimate of protein metabolized. Since there had been a net ab¬
sorption of 10.66 g nitrogen (i.e., 12.02 — 1.36 g), and the men were not
said to have been growing or gaining weight, this quantity must have been
metabolized - some through the skin, and some lost as hair and so on. The
difference of 1.5 g “lost” nitrogen (equivalent to 9.4 g protein) should
therefore be added to all of Chittenden’s estimates of metabolized protein
in his main experiments, where food and fecal nitrogen values were not
available for the full periods of the trials.46
If we apply this correction to the results that Chittenden gave for Mendel’s
experiment, which were examined in detail earlier, 1.5 g must be added to
the 6.53 g nitrogen found in the urine. This makes 8.03 g nitrogen (equiv¬
alent to 50.2 g digested protein). Allowing for incomplete digestibility as
before, we find that the revised estimate of Mendel’s protein intake when
he was in balance is 61.4 g. This is a significant increase, but of course still
well below Voit’s standards.

Were the Data Reliable?

By the time of McCay’s writing, Chittenden himself had remained on his

modified diet for 7 years and claimed that he had never been fitter or more
mentally active. However, McCay questioned whether the data given for
Chittenden’s two 6-day balance periods — the only times that actual intakes
were determined — were representative or even correct. The calculated daily
intake of 1,600 kcal corresponded to only 28 kcal/kg body wt. “Yet the
average amount hitherto considered absolutely necessary is about 40 kcal/
kg. Either this is far above the true requirements or there is some fallacy in

44. Ibid., pp. 68, 145.

45. Chittenden (1911), pp. 659—60.

46. McCay (191 zf, p. 84.

 Protein and Energy

Chittenden’s data or, if the results from the experiment are to be extended
to mankind in general, the law of conservation of energy in human beings
must be seriously called in question.”47 McCay had a strong point. Energy
requirements can be determined either directly from heat output or indirectly
from respired carbon dioxide, and modern recalculations confirm that some¬
one of Chittenden’s weight and age range would be expected to use ap¬
proximately 40 kcal/kg daily even with very light activity.48
With reference to the results obtained with the soldiers, McCay drew
attention to the considerable (i.e., twofold) differences in fecal nitrogen
content among the subjects when they were supposedly eating virtually the
same diet during the balance periods. Thus the apparent percent digesti¬
bilities of the nitrogen varied from 75% to 90%. McCay suggested that
this was inconsistent with the individuals truly eating the same diet.49 Ben¬
edict had also noticed the variability and considered it to be evidence of
malabsorption among some of the men, their diets being too low in protein
to sustain optimal body function.s° However, this is not a satisfactory ex¬
planation, since it is the higher values for apparent digestibility that seem
out of line.
Examination of the detailed data published by Chittenden indicates an¬
other explanation. Considering the full 6 days of the first balance period,
which included 12 men, and the first 6 days of the second (7-day) trial,
which included 11 men, if every man had produced a fecal sample every
day, there would have been altogether 23 fecal samples on the first days of
the two trials, 23 on the second and so on. If some men did not have a
bowel movement on a particular day, there would be some number fewer
each day, but one would expect the number of “missing samples” to be
randomly distributed. In fact, in the 6 successive days, the missing totals
were 15, 12, 4, 4, 1 and 1. Such a skew distribution could have occurred
by chance less than 1 in 10,000 times. One seems forced to conclude that
many of the men were unwilling to turn in their fecal material for weighing
and analysis in the first portion of each trial. And, indeed, the calculated
protein digestibility value for those subjects who said they had no bowel
movement on 3 or more of the 6 days is 88%, and that for the remainder
of the subjects is 81%. This seems to confirm the idea that not all the fecal
material was collected from the first group.
If this conclusion is correct, it certainly indicates that the control of the
“soldier experiment” was imperfect, and McCay may have been right also
in doubting whether the full amount of the daily urine was always handed

47. Ibid., p. 115.

48. FAO/WHO/UNU (1985), pp. 72-6.

49. McCay (1912), pp. 125-6.

50. Benedict (1906), pp. 420-1.
 Chittenden versus the US. Establishment

over for analysis. He quoted another author who had written that some of
the soldiers had later admitted that they had eaten unauthorized meals
during the experiment.5 Certainly the soldiers’ computed mean energy in¬
take of 2,300 kcal/day does, as McCay points out, seem extremely low in
relation to the directly determined energy expenditures in rest and activities
of various kinds that have since been recorded.Si

Chittenden’s Response

The British Medical Association held two debates during this period on
nutritional requirements, and summaries of the discussions were published.
The first was held in Canada in 1906 and the second in England in 1911.53
Chittenden was the opening speaker on each occasion, and in the second
debate he made some new points. One was that critics had tried to argue,
on the basis of grams protein consumed per kilogram body weight, that the
suckling infant s intake of breast milk, “nature’s food,” corresponded
roughly to the Voit standard for adults. However, as Rubner had recently
set out in great detail, human breast milk (in contrast to the milk of other
species) was quite low in protein, which contributed only some 10% of its
total calories. Further, it was agreed that growth itself required a high
proportion of protein, and Rubner had calculated that, of the nutrients
remaining for maintenance in the growing infant, only 5% of the energy
came from protein.54 This corresponds to a ratio of 37.5 g protein to 3,000
kcal.
Chittenden also argued that it was the relative prosperity of Westerners
that allowed them to consume a high-meat and therefore a high-protein
diet. His critics were trying to reverse the true sequence of cause and effect.
Moreover, it was more likely that the bad effects of the poorest Indian diets
were the consequence of their lack of trace nutrients: “There are many
factors, aside from nitrogen and calories, that play a part in determining
proper nutritive conditions. I have seen dogs, for example, go to pieces on
a vegetable diet, while the substitution of a little meat or milk, with the
same total of available nitrogen, is followed by a satisfactory condition of
health.”s5
Chittenden was referring here to feeding experiments that he had under¬
taken in response to reports from Europe that, although dogs remained in
nitrogen balance for some time on low-protein diets, they eventually weak¬
ened and succumbed. This was a powerful argument for caution in adopting

51. Loeser (191Z), p. 156.

52. McCay (19x2), pp. 145-8.

53. Anonymous (1906); Anonymous (1911).

54. Rubner (1908), pp. 96—7, 108—9; Chittenden (1911), p. 662.

55. Chittenden (1911), p. 661.

 Protein and Energy

Chittenden’s ideas, and he was in turn responding with a bold prediction

that the deficiency was one of hitherto unknown nutrients rather than of
protein. But to follow this line of investigation and its successive revelations
will require a new chapter.
>. '

Summing Up

Wilbur Atwater, who was the leader of the first U.S. dietary surveys in the
1880s and 1890s, set the protein standard for physically active men at 12.5
g/day. This was made slightly higher than Voit’s standard on the ground
that Americans were more vigorous and lived at a higher level of activity
than Europeans. Working people, however, were urged to economize on
food purchases by buying cheaper cuts of meat that were just as nutritious
as more expensive ones, even if less appetizing.
Russell Chittenden, an established physiological chemist, doubted the
validity of this high standard and studied the physical performance and
nitrogen balance of volunteers who consumed only about one-half the stan¬
dard quantity of protein for 6 months, but remained vigorous and apparently
in nitrogen balance. However, others were not persuaded that such a diet
could be safely maintained for longer periods, because it seemed a general
rule in the world that those who could afford to do so ate more and that
the poor who did not were generally lacking in energy and initiative. Chit¬
tenden tried unsuccessfully to argue that the deficiencies of the “poor” diets
were more likely to be those of unidentified trace nutrients than those of
protein, and that it was affluence that caused people to eat a “rich” diet,
and not vice versa, as his critics claimed. „
4

118
 Vitamins and Amino Acids,
1910-1950

One of the serious criticisms of Chittenden’s advocacy of low-

protein diets was that, while dogs on such regimes remained healthy for
many weeks, they eventually became ill, lost their appetite and died. This
had been the experience in two German laboratories. In each the daily diets
contained about 1.5 g protein and 90 to 100 kcal/kg body wt and consisted
of minced meat (or dehydrated meat powder), animal fat and rice. In one
laboratory, where meat powder was used, the dogs’ digestion became less
efficient; they vomited and lost their appetites.1 2 In the other laboratory there
was no change in digestibility but, after 6 weeks or so, the dogs lost their
appetite and finally died.* In both laboratories, autopsies showed fatty de¬
generation of the gut wall and of the liver.
In the next decade a worker in Finland undertook further trials with two
dogs using a simple mix of meat and sugar. The dogs did quite well for
many months and digestibility remained normal. But finally each dog died
after a short acute illness, which was diagnosed as resulting from a severe
infection.3
Chittenden, skeptical as ever, wrote:
We must first be sure of our facts before arguments or conclusions of any kind are
warranted. It is to be remembered that dogs are as sensitive in many ways as man,
and no physiological experiment covering a long period of time can be carried out
with any hope of success unless there is due regard for proper hygienic conditions,
some degree of variety in diet, and reasonable opportunities for fresh air and oc¬
casional exercise. Is it strange that dogs confined in cages barely large enough to
permit their turning around, and fed day after day and month after month with
exactly the same amount of desiccated meat, fat, and rice, should show signs and
symptoms, if nothing worse, of disturbed nutrition?4

1. Munk (1893).
2. Rosenheim (1893).

3. Jagerroos (1902), pp. 389-412.

4. Chittenden .^907), p. 242.

 Protein and Energy

He therefore went on to carry out a large series of experiments of his own.

Chittenden’s technique was to alternate the conditions of the dogs be¬
tween io days of confinement in a cage, where urine and feces could be
collected in order to determine nitrogen balance, and 3 weeks of running
free in an airy room, and even in an outdoor paddock for several hours on
sunny days. The diet was also varied from month to month. For example,
in one trial it started with meat, lard and cracker dust. Then the cracker
dust was replaced by bread, and in the next month half the meat was replaced
by milk. The total nitrogen and energy values of the diet were, of course,
kept constant. Under these conditions the dogs remained in good health
and vigor for many months, though they received only 1.6 to 1.7 g protein
and 70 kcal/kg.5 This contrasted with the conclusion from the German
studies that even 1.9 g protein per kilogram was inadequate, and Chittenden
attributed their failures to the fact that the dogs were kept continuously in
small cages.
However, when Chittenden removed meat and milk entirely from the
diets, and the same nitrogen intakes came from bread, beans and/or peas,
the animals quickly developed gastrointestinal disturbances and lost their
appetite and energy. He wrote that these results could have been due to
some specific differences between animal and vegetable proteins or to “the
possible need of the animal’s body for extraneous principles which only
meat, milk, or other animal products can supply-Other substances with¬
out any appreciable fuel value are quite likely to be of primary importance
in controlling and regulating the various processes of the body, which com¬
bine to maintain the condition of normal nutrition. With a diet restricted
to one or two vegetable products, it is quite conceivable that something
may be lacking which the system demands, though it cannot be'Pleasured
in terms of nitrogen or calories.”6

The Search for Unknown Nutrients

Both possibilities (i.e., the importance of so far unknown nutrients and

nutritional differences among proteins) that Chittenden put forward in 1907
were to become subjects of active research. We will begin with the first type
of investigation, but in abbreviated form since it is peripheral to the primary
subject of this book.
Writing in 1905 about the dangers of low-protein diets, Benedict argued
that the disease beriberi had been eliminated from the Japanese navy by
increasing the protein content of rations to European standards.7 This had

5. Ibid., pp. 2.44-52. \

6. Ibid., pp. 253-6.

7. Benedict (1906), p. 429.

12.0
 Vitamins and Amino Acids

been achieved by Takaki, who replaced a proportion of the predominantly

white rice” diet with wheat bread, peas, beef, fish and miso (fermented
soybean).8
Beriberi was sometimes referred to as the “national disease of Japan”
because of its prevalence there, but it was a problem in other parts of Asia
too. It was also to be experienced by many U.S. and British soldiers held
in Japanese prison camps in World War II. The principal symptoms were
weakening of the heart and degeneration of nerve fibers, particularly in the
legs, causing paralysis. The mortality could be very high.9
By the 1880s European doctors were seeing beriberi in British Malaya
and the Dutch East Indies. And with the great influence of the “germ theory
of disease” at that time the disease was generally thought to be either a
direct infection or the result of poisoning from a toxin produced by a micro¬
organism. Research was stimulated by the observation that a disease in
chickens characterized by damage to peripheral nerves was associated with
their being fed white rice. The disease did not occur if their diet was sup¬
plemented with “rice polishings,” the fragments removed when brown rice
was refined to white rice by abrasive milling. Further inquiries showed a
comparable distribution of beriberi among prisons in both countries; that
is, it was a continuing problem where white rice was the prisoners’ staple
food, but not where they were given brown rice. The hypothesis first ad¬
vanced to explain this finding was that white rice contained a toxin, but
that a benevolent Nature had provided the antidote in the outer layer of
the grain.10
There was little difference in the protein content of the two forms of rice.
Moreover, the “chicken disease” could be produced with a number of
starchy foods, so that it was not due to a specific effect of white rice. It
could also be cured by microscopic quantities of extracts made from either
rice polishings or brewer’s yeast. In one series of tests, even 4 milligrams
(mg), equivalent to Vy.ooo of an ounce, of a still impure extract was a curative
dose.11

The Discovery of Vitamins

In 1912 Casimir Funk, a 28-year-old Polish chemist attempting to isolate

the factor in rice polishings at London’s Lister Institute, put forward the
bold hypothesis that there were at least four organic compounds that must
be present in our diet (though at quite low levels) in order to prevent the

8. Takaki (1885); Anonymous (1887a).

9. Williams (1961), pp. 53—8, 68—80.

10. Ibid., pp. 36-48.

11. Funk (1913)*. p. 179.

*

121
 Protein and Energy

following four diseases: beriberi, scurvy, pellagra and rickets. He further

hypothesized that each was a compound containing an amine (-NHJ group,
and he named these hypothetical vital factors vitamines.x:L For the next 30
to 3 5 years a large proportion of the nutritional research carried out through¬
out the world was devoted to establishing the identity of these new organic
nutrients. It soon became clear that they were not all amines, so the term
was shortened to vitamins. The general history of these complex develop¬
ments has already been well covered,13 and I have myself contributed vol¬
umes on pellagra and scurvy specifically.14
An early finding was that young rats failed to grow normally when given
“purified” diets consisting of protein, fat and carbohydrates together with
a mixture of minerals. However, normal growth rates could be obtained if
these diets were supplemented with butterfat (or cod-liver oil) and skim
milk from which the protein had been removed. It was concluded that
butterfat and cod-liver oil contributed an unknown fat-soluble factor “A,”
and the protein-free milk, a water-soluble factor “B.” After a short time
they were named “vitamin A” and “vitamin B.”IS It was then realized that
the active factor in lemon juice and cabbage that cured an experimental
scurvy-like condition in guinea pigs was neither of these, and it was named
“vitamin C.”
After that, things became more complicated. More fat-soluble factors were
discovered and given the letters D, E and K. It was also realized that vitamin B
was a complex, and individual factors preventing different conditions were
coded B„ Bz and so on. However, some of the discoveries remained uncon¬
firmed, and the system became muddled. Therefore, as soon as the water-
soluble vitamins were isolated and identified, they were generally given
names reflecting their chemistry. However, cobalamin (the cobaltcontaining
vitamin) is still remembered by its original code of BIZ and pyridoxal (con¬
taining a pyridine ring) is often referred to as B6, partly because several slight
variations of pyridoxal have the same type of biological activity.
By 1950, as a result of collaborations between nutritionists working with
laboratory animals and organic chemists, at least 14 vitamins had been iso¬
lated and identified. In addition to the diseases listed by Funk, blindness
among Third World children was often due to a deficiency of the first fat-
soluble factor (vitamin A). In the same period it was also realized, or con¬
firmed, that people could suffer from deficiencies of inorganic elements that
were needed only in very small quantities. For example, deficiency of iodine

12. Funk (1912).

13. McCollum (1957), pp. 201-318; Guggenheim (1981), pp. 111-276.

14. Carpenter (1981); Carpenter (1986a).

15. McCollum (1957), pp. 114, 2.91.

122
 Vitamins and Amino Acids

resulted in goiter and cretinism, and that of zinc in dwarfism and dermatitis.
Iodine and zinc are therefore two of the “trace elements” required in our diet.
The Japanese succeeded in eradicating beriberi from their navy by giving
sailors a greater variety of foods that were richer in protein than white rice
was. However, it is now appreciated that these additional foods were also
richer in thiamine, the antiberiberi vitamin, and that it was this, rather than
the additional protein, which explained the success.
Chittenden’s provisional hypothesis, that his dogs maintained on a diet
of dried peas, cracker meal and lard died because of a deficiency of an
unknown trace nutrient, also proved to be correct. It was found that the
disease condition could be prevented by supplementing the dogs’ diet with
a small quantity of carrots, because, as was later shown, carrots contained
the yellow pigment carotene, which animals can convert to the active form
of vitamin A.16 The observation that animals could do well for a period on
a diet lacking vitamin A was explained by their accumulation of a reserve
of the vitamin in the liver during an earlier period on a rich diet consisting,
for example, of dam’s milk, and the gradual use of these reserves. In general,
the water-soluble vitamins are not stored as well in animals and humans,
though cobalamin (BIZ) is an exception, and people may subsist for several
years on a vegan diet containing none of this vitamin before irreversible
nerve damage occurs. (A “vegan” is a strict vegetarian who takes no milk
or eggs.)
Because of the skills of chemists and the very low levels of vitamins that
are needed, vitamin supplements can be purchased for only a small pro¬
portion of the cost of one’s ordinary food needs. In the wealthier countries
of the world most people can afford, and in fact choose, a varied diet that
meets their vitamin requirements. But in the poorer countries, where many
families can afford only a much more limited diet, young growing children
are more likely to be vitamin-deficient.
The main problem for governments or other organizations that want to
provide aid is the cost not of purchasing vitamins, but of distributing them
to those actually in need, often in remote villages. Where staple foods, such
as white rice, are processed in relatively large amounts, it is practicable to
have the foods fortified with the important vitamin(s) lost during processing.
In the use of white rice, a supplement of only 5 parts per million of thiamine
is sufficient to prevent the occurrence of beriberi. In most developed coun¬
tries, white wheat flour is fortified with three or more vitamins to the levels
found in whole wheat grains.
Although, as just mentioned, the common “affluent” diets contain suf¬
ficient amounts of each vitamin to eliminate any risk of deficiency conditions

16. Underhill and Mendel (1928), pp. 632—3.

 Protein and Energy

developing, some people recommend taking “megavitamin” dosages in or¬

der to induce a state of “supernutrition.” For example, the ordinary daily
recommended intake of vitamin C is 60 mg, but Linus Pauling has written
that he takes some 200 times this quantity, that is, about 12 g/day, in order
to reduce his risk of developing cancer and other diseases. The benefits from
such a practice are still a subject of controversy.17 What is noncontroversial
is that our knowledge of vitamins and trace elements has provided a better
understanding of the effects of food processing and increased the public’s
ability to make healthy food choices.

The Constituents of Proteins *

Once it had been found possible to achieve good growth in young rats with
purified diets supplemented with protein-free sources of vitamins and min¬
erals, the way was open for rapid development in understanding what made
some proteins apparently of higher nutritional value than others.
We saw in Chapter 2 that gelatin, though rich in nitrogen, could not
replace the nitrogenous compounds of meat in the diet of dogs. Then Mulder,
in 1839, hypothesized the existence of a chemical unit, called “protein,”
which, in combination with varied proportions of sulfur and phosphorus,
formed albumin, fibrin and gluten (as explained in Chapter 3)^ But this idea
soon collapsed. Although the proteins of vegetable foods had properties
and carbon—nitrogen ratios that were very similar to those in animal tissues,
they were not identical. Nevertheless, the animal kingdom apparently had
to use vegetable proteins as a basis for forming their tissues.
Chemists like Liebig realized that they were dealing with large molecules
and seemed to visualize that the changes animals made to the albuminoids
they ate involved the removal where necessary of a few atoms or radicals
and the addition of others, but that when an animal or human ate meat,
the proteins could be deposited with virtually no modification. They knew
that, by boiling proteins with either strong sulfuric acid or strong alkali,
some much simpler crystalline products could be obtained. These — named,
for example glycine, leucine and tyrosine — all belonged to the same class,
called amino-bodies at first and then amino acids.18 However, at the time,
this seemed to have no biological significance, since the conditions of break¬
down were so extreme, in contrast to the moderate temperature and neutral
reaction of animal tissues. (The chemical structures of the amino acids are
given in Appendix A.)
During the same period (i.e., the 1830s and 1840s) there was an increase
in knowledge of the digestion of proteins. Extracts made from the dried

17. Carpenter (1986a), pp. zio— 20.

18. Vickery and Schmidt (1931), pp. 187-213.

124
 Vitamins and Amino Acids

mucus of stomach linings changed the solubility and coagulability of dried

albumin in acid suspension in the same ways as had been observed in the
stomachs of animals killed some time after receiving a meal of dried albumin.
The active factor in these extracts was named pepsin. However, chemical
analysis of the digestion products showed no significant change in the car¬
bon-nitrogen ratio. This allowed both Liebig and Dumas to conclude that
there had been no real chemical change in the food protein, “only a change
in its state of cohesion,” so that it could pass more readily through the gut
wall.19
After the discovery in the 1860s that pancreatic juice secreted into the
small intestine exerted a powerful digestive action on proteins, the free
amino acids leucine and tyrosine were recovered in traces in intestinal con¬
tents (i.e., in vivo) and in quite large quantities from the incubation of
proteins with pancreatic juice in glass bottles (in vitro) at body tempera¬
ture/0 At first, German scientists considered that this was probably an
artifact from the putrefactive action of bacteria present in the gut. It seemed
to them that “such a profound decomposition would be a waste of chemical
potential energy, and a reunion of the products of such a profound decom¬
position is highly improbable.”11
Little attention was therefore paid to these findings by students of nutri¬
tion. It was assumed that under natural conditions an animal organism
would make only minimal changes to the proteins it consumed before in¬
corporating them into its tissues. Scientists labored for manv years to ca¬
tegorize the products of peptic (i.e., stomach) digestion and spoke of
different classes of proteoses, peptones and so on, according to the solu¬
bilities under different conditions of the fractions they isolated. Presumably
they assumed implicitly that they might discover a universal “core” around
which all the individual animal proteins could be built up.11 There was no
proof that whole proteins consisted only of pre-formed amino acid units;
the combined yield of amino acids recovered, even from acid hydrolysates
in this period (i.e., before 1890), corresponded to less than 50% of the
original weight of the proteins/3
It had also been observed repeatedly that if a large supplement of a single
amino acid such as glycine were given to a dog, almost all of the added
nitrogen would be recovered in the urine as urea within the next 24 hours/4

19. Holmes (1974), pp. 163-76, 181-6; Matthews (1991), pp. 7-19.

20. Lea (1890), pp. 243—4.

21. Bunge (1902), pp. 167-71; Greenstein and Winitz (1961), Vol. 1, pp. 249-63; Holmes

{1979), pp. 179-80.

22. Fischer and Abderhalden (1903), p. 83; Vickery and Osborne (1928), pp. 406-7; Fruton

(1979), p. 5; Fantini (1983), pp. 9-15.

23. Moore (1898), pp. 401—40; Chibnall (1939), p. 15.

24. Salkowski (1880), pp. 100—7.

I25
 Protein and Energy

This threw further doubt on the nutritional value of free amino acids and
led to an ingenious alternative hypothesis: that the cells of the intestinal
wall could limit the absorption of proteoses and peptones to the quantity
required by the body for growth and repair. It was the quantities remaining
in excess in the gut that were then further broken down by pancreatic juice
to free amino acids. These were rapidly absorbed and metabolized by the
liver to urea, which was excreted."5 Chittenden, for example, wrote in 1895:
“We may well consider the formation of these amino-acids in pancreatic
proteolysis as a means of quickly ridding the body of any excess of ingested
protein food, with the least possible expenditure of energy on the part of
the system. This has always seemed to me the probable purpose of the
profound changes which the pancreatic ferment is capable of inducing.”"6
The scientists who believed that proteoses and peptones were absorbed
by the gut wall were puzzled by the fact that they were unable to find these
compounds in the blood of animals that were digesting a protein meal. They
proposed that these molecules disappeared because they were immediately
recombined by the cells in the gut wall into ordinary protein before being
passed into the blood. In one experiment carried out to test this idea, a
horse was given large meals of the wheat protein gliadin, which had been
found to yield about 36% of glutamic acid on hydrolysis. Blood samples
were taken from the horse before and after the digestion of the gliadin, but
analysis indicated no rise in the glutamic acid content of th$ mixed blood
proteins from their normal value of about 8%. It did not appear, therefore,
that “re-formed” gliadin was transferred to the bloodstream."7 Moreover,
when other workers attempted to study the changes undergone by proteoses
and peptones in contact with isolated pieces of small-intestine wall, they
found no evidence of protein synthesis. Instead they discovered another
enzyme, erepsin, which caused them to break down further to Free amino
acids."8
In several experiments, when proteins were incubated for long periods
with pepsin and trypsin, the digests retained their nutritional value. How¬
ever, it was still possible that some “special” molecule remained. And
it had been repeatedly confirmed that neither mixtures of the amino acids
that were then available nor the mixture of products from boiling proteins
with mineral acids could support sustained growth or nitrogen retention in
dogs or rodents."9 Here, then, was a paradox: The evidence from digestibility
studies tended to show that there was a complete breakdown to amino acids

25. Lea (1890), pp. 2.57-8.

26. Chittenden (1895), p. 113.

27. Abderhalden and Samuely (1905).
28. Conheim (1901), p. 462; Kutscher and Seemann (1902), p. 543; Matthews (1991), pp. 25-

47-
29. Henriques and Hansen (1904), pp. 427-39; Abderhalden and Rona (1905), pp. 202-3.

126
 Vitamins and Amino Acids

before absorption, yet feeding studies indicated that amino acids could not
replace protein.
Results obtained with sheep were exceptional, since adding a supplement
of the amino acid asparagine, or even of ammonium acetate, to a low-
protein diet improved their nitrogen balance. However, it was realized that
the digestive apparatus of ruminating herbivores was different from that of
single-stomach species such as humans, dogs and rats. The food eaten by
these herbivores goes first to the rumen, a large bag in which the food is
fermented for many hours before it passes to the true stomach, into which
hydrochloric acid and pepsin are secreted. It was hypothesized that bacteria
living and multiplying in the rumen were capable, like plants, of synthesizing
protein from simple nitrogen sources and that dead bacteria passing down
the gut might then be digested in the small intestine and utilized as witb
ordinary food protein. This idea was strengthened by Muller’s finding that
bacteria taken from a sheep’s rumen could use asparagine and that the
bacterial protein obtained was utilizable by dogs.30

The Discovery of Tryptophan

In 1872. Heinrich Ritthausen, who had spent many years identifying amino
acids present in hydrolysates of plant proteins, concluded that the quantities
of individual acids present differed from those in hydrolysates of animal
albumins. He suggested that such differences might result in differences in
the nutritional value of individual proteins.31 Several more amino acids were
discovered in protein hydrolysates during the next 30 years, and even the
crude analytical methods available confirmed the differences in composition
between animal and vegetable proteins.3Z
The discovery that some of the color reactions given by free ammo acids
in contact with particular chemical reagents were also given by intact pro¬
teins strengthened the idea that the amino acid “units” were already present
in the protein rather than created by the procedures used to break the protein
down. Because of a color reaction that they gave, most proteins were sus¬
pected to include an unknown amino acid, given the name “tryptophane,
which contained an indole radical.33 It was not found in acid hydrolysates
but finally, in 1902, Hopkins and Cole at Cambridge University isolated it
from a pancreatic digest of casein, after precipitation of its mercury salt.34
(The name has been retained, but without the final “e. ) It was found to
be destroyed when heated with the strong acids used to hydrolyze proteins.

30. Muller (1906); Cathcart (1912.), PP- 4^~2-

31. Ritthausen (1872), p. 236; Chibnall (1939), P-13-

32. Abderhalden (1905), PP- u-3; Vickery and Schmidt (1931), P- J73-
33. Greenstein and Winitz (1961), Vol. 3, PP- *3i7-«-

34. Hopkins and Cole (1902).

 Protein and Energy

Four years later, Willcock and Hopkins carried out an important feeding
experiment using mice. The basic diet had zein as its only protein source.
Zein was an alcohol-soluble protein from maize that did not give the color
reaction for tryptophan. Without supplementary tryptophan, the mice lost
weight rapidly and survived on average only 16 days, but with the addition
of this amino acid their weight remained almost constant and they survived
for 30 days.35 In one of a long series of experiments, Abderhalden found
that the addition of tryptophan to an acid hydrolysate of casein was enough
to make it capable of maintaining nitrogen balance in adult dogs.36 These
results certainly indicated that tryptophan had an essential role in animal
biology but, as the authors pointed out, it was not necessarily involved in
the synthesis of protein, since the animals did not gain weight. It could, for
example, serve as a precursor to a hormone.
Eight years later, after Osborne and Mendel had developed protein-free
vitamin sources that would support the growth of rats (i.e., protein-free
milk and butterfat), they too did a feeding experiment with zein. They found
that, when it was supplemented with both tryptophan and the amino acid
lysine (which it also lacked), young rats doubled, or even tripled, their weight
within 7 weeks. If only one amino acid was added they would not grow,
but there was a difference between the responses. In the absence of tryp¬
tophan, body weight fell rapidly (as in the earlier mouse experiment), but
in the absence of lysine, it remained almost constant for seyeral weeks.37
The authors suggested, first, that lysine was needed only fof the synthesis
of additional protein tissue. Second, it seemed possible that the “main¬
tenance” requirement of adult animals for protein could not really be
explained by “wear and tear” and a consequent need for synthesis of re¬
placement protein. It could be that, in the main, tissue proteins were broken
down when the intake was inadequate, simply to provide tryptophan for
another function that had even higher priority.38
Another worker tested Liebig’s old idea that vegetable proteins were used
less efficiently because the animal consuming them had the additional work
of synthesizing the creatine and other soluble nitrogenous compounds found
only in animal tissues. He found that adding Liebig’s “Extract of Meat” to
the wheat protein gliadin did not improve its capacity to keep dogs in
nitrogen equilibrium.39
The indications, therefore, all pointed to differences in amino acid com¬
position as the explanation for young rats and mice not growing well when
receiving only some vegetable proteins. However, there were also indications

35. Willcock and Hopkins (1906).

36. Abderhalden (1909).
37. Osborne and Mendel (19x4), pp. 346—7.
38. Ibid., pp. 334, 340.
39. Michaud (1909), pp. 445-5°, 45^-7-

128
 Vitamins and Amino Acids

that some amino acids — specifically glycine - could be synthesized. In

contrast, lysine and tryptophan apparently could not.
It seemed clear, as Rubner spelled out in 1911, that there could not be
one value for the protein requirement of a human or animal; it must depend
on the character of the protein.40 But that “character” meant no more than
“amino acid composition” was still no more than a hypothesis. Sorting this
out was to require another 30 years of intensive work. On the chemical
side, it required the development of more sensitive and reliable methods of
amino analysis, as well as the search for additional compounds and their
identification. It also required the development of standardized procedures
for putting different protein sources on a scale of relative values, and we
will begin there.

Measuring Protein Quality

The pioneers in the field were Osborne and Mendel, one of whose experi¬
ments has already been described. Like Atwater and Chittenden, they were
both natives of Connecticut and educated at Yale, though only Mendel went
on to spend a period of further training in Germany. Each then spent the
remainder of his working life at a laboratory in New Haven and in each
case married the daughter of his first chief (S. W. Johnson, director of the
Connecticut Agricultural Experiment Station in the case of Osborne and
Professor Chittenden of Yale’s Department of Physiological Chemistry in
the case of Mendel).
Thomas Osborne (born in 1859) was the older of the two. When the
,
collaboration began, in 1909 he had already spent 30 years studying the
characteristics of proteins isolated from different plant materials. He had
made careful analyses of their contents of individual amino acids to the
extent possible at the time and now wanted to see how far these values
could be used to predict a protein’s nutritional value. Lafayette Mendel
(born in 1872), whose work had been mainly in the physiology of the
digestion of foods, was pleased to collaborate in such studies, and in the
next 20 years these two men were to publish well over 100 papers together.
Their early experiments with growing rats, using diets consisting of pur¬
ified materials, had shown them that a diet could appear to be adequate
for a short period but then fail. They had, as described already, overcome
this problem by including rich sources of vitamins, and they preferred to
evaluate a diet by capacity to support rapid growth (measured as weight
gain) and a healthy appearance over a period of several weeks, rather than
by its capacity to support nitrogen retention over a few days in a balance

40. Rubner (1911), p. 46.

41. Smith (1956); Vickery (i956)-

129
 Protein and Energy

Figure 7.1. Comparison of two female rats, each 140 days old, having
received (a) a diet containing 18% casein as the protein source and (b) the
same diet but with 18% gliadin (from wheat) as the protein source. (Os¬
borne and Mendel, 1911)

cage (Figure 7.1). One of their aims was to establish a standardized pro¬
cedure that would allow workers to assign a reproducible numerical value
to different individual proteins, or combinations of them. Controversies
grew up around this subject. They do not need to be understood in order
to follow the general thread of this book. However, the subject has had
some practical importance, and the main developments, the definition of
terms such as biological value and selected results are summarized in
Appendix B.
Once satisfactory standardized methods had been developed, it became
clear that proteins varied considerably in their capacity to support the
growth of young rats. This remained true even after allowing for differences
in the digestibility of the proteins. In general, plant proteins tended to be
inferior to animal ones, though the earliervpoor reputation of gelatin was
confirmed.

130
 Vitamins and Amino Acids

Complementation between Proteins

In practice, human diets always include proteins from different foods. Work¬
ers were interested, therefore, in determining whether there were special
advantages to be obtained from combinations of protein sources, particu¬
larly those used in traditional diets. The first experiments were carried out
with mixtures of white flour and an animal food, with the white flour
providing two-thirds of the total nitrogen, as well as with the two foods
separately. Dried cow’s milk fed as the sole source of protein gave a bio¬
logical value of 84, and the flour fed similarly gave a value of 54. The
arithmetic average for a 1:2 mixture of these materials is 64 (i.e., V3 x 84
+ y3 x 54). However, the actual biological value of the mix was 14%
greater than this. With whole eggs or meat, the “complementary” effect
with white flour was of the same order.4* Such an effect would be expected
if the proteins in the mix had different essential amino acids in short supply.
Obviously it was now hoped that it would be possible to relate the biological
value of food proteins to their amino acid composition and thus to design
new, and perhaps improved, complementary mixtures. But first it was nec¬
essary to know which of the amino acids were nutritionally essential, or
“indispensable,” an alternative adjective used in some papers.

Essential Amino Acids

The first important contributor to this part of the story is W. C. Rose, who
did his graduate work under Lafayette Mendel at Yale and, after two rel¬
atively junior university appointments elsewhere, became professor of phys¬
iological chemistry (later changed to biochemistry) at the University of
Illinois, Urbana, in 1922, at the age of 35-43 Rose and Mitchell (whose work
on protein quality is summarized in Appendix B), both regarded by their
graduate students as admirable men, apparently had no personal contact
during their working lives, although each had his laboratory on the Urbana
campus. They finally appeared on the same platform when the University
of Illinois organized a symposium in their joint honor, and they were both
in their mid-seventies.44
Rose began a 20-year project to prepare a diet containing only free amino
acids (rather than protein) that would support growth in rats and then to
determine which were, or were not, dietary essentials. After a great deal of
work in collaboration with chemists preparing sufficient quantities of 19
known amino acids, he began the feeding experiments in 1930. Although

4x. Mitchell and Carman (1926); Block and Mitchell (1946), PP- 272-3-

43. Roe (1981).

44. Edman, Forbes and Johnson (1968); Roe (1981).
 Protein and Energy

he included them all in his first trials, his rats failed to grow, as had those
of his predecessors using fewer amino acids.45
With hindsight, it seems strange that he failed to include in his initial
mixture the amino acid methionine, which had been discovered and iden¬
tified as a component of proteins in 192.3.46 Its discovery had actually come
from efforts to isolate an unknown growth factor for bacteria. But even
though the work had been done in a department of bacteriology, the results
had been published in the same publication — the Journal of Biological
Chemistry — that was reporting nearly all the U.S. work on the nutritional
requirements of rats. Finally in 1932, another laboratory indicated that
methionine gave as good a growth response as cystine when it was added
to a rat diet known to respond to the latter amino acid.47 It was assumed
that methionine, since it too contained sulfur, must be converted by rats to
cystine, which had been classified as “essential” since the earliest work on
amino acid supplementation. In later work, Rose fortunately included me¬
thionine, though regarding it even in 1936 only as an alternative to cystine.48
The next development was that he and his colleagues fractionated protein
hydrolysates and found two that gave a growth response when fed together
with their “amino acid diet.” Further fractionation showed that the active
factor in one fraction was isoleucine (a well-known amino acid and already
in the basal mix but at a level well below that needed for growth). The
other active factor was isolated and found, after further wprk, to be an
amino acid not previously considered to have any significance for nutrition.
It was to be named threonine.49 Rose had achieved his first aim and could
also work out exactly which molecules were or were not necessary for rat
growth. Figure 7.2 is taken from Rose et al.’s final paper.s° All but glycine
can exist in either a “left-handed” (l) or “right-handed” (d)' form. This is
explained in any textbook of organic chemistry. Naturally occurring com¬
pounds (i.e., those synthesized in living tissues) are all in the left-handed
form. When amino acids are produced by ordinary chemical synthesis, they
are mixtures of equal parts of the two forms, the so-called racemic product.
In some cases - for example, methionine - the right-handed d form can be
converted in the animal body to the usable L form, and the cheaper dl-
methionine mixture is often used to balance commercial poultry diets. How¬
ever, with lysine, the D form is nutritionally useless, and it is the natural L
form, produced by fermentation, which is sold commercially as a supple¬
ment.

45. Rose (1931).

46. Mueller (192.3).
47. Jackson and Block (193Z).
48. Rose (1936).
49. McCoy, Meyer and Rose (1935).
50. Rose, Oesterling and Womack (1948).

132
 Vitamins and Amino Acids

Classification of Amino Acids with Respect to Their Growth Effects in the Rat

Essential Non-essential

Lysine Glycine
Tryptophan Alanine
Histidine Serine
Phenylalanine Cystine*
Leucine Tyrosinef
Isoleucine Aspartic acid
Threonine Glutamic “ {
Methionine ProlineJ
Valine Hydroxyproline
Arginine § Citrulline

* Cystine can replace about one-sixth of the methionine requirement, but has no
growth effect in the absence of methionine.
t Tyrosine can replace about one-half of the phenylalanine requirement, but has
no growth effect in the absence of phenylalanine.
J Glutamic acid and proline can serve individually as rather ineffective substi¬
tutes for arginine in the diet. This property is not shared by hydroxyproline.
§ Arginine can be synthesized by the rat, but not at a sufficiently rapid rate to
meet the demands of maximum growth. Its classification, therefore, as essential or
non-essential is purely a matter of definition.

Figure 7.2. Reproduction of the final summary of the rat’s requirements

for amino acids, qs determined by Rose and his colleagues in 1948.

Amino Acid Analyses

It seemed clear that nutritionists now had to think not of “protein nutrition”
but of “amino acid nutrition,” since it was the amino acids alone which
tissues received from digested food and which they needed, and that the
whole array of 12 essential amino acids was needed for an organism to
remain healthy. . .
In the nineteenth century there were great advances in chemical analysis,
but virtually all the methods depended finally on weighing a product -
usually an insoluble compound formed by reaction of the compound of
interest with some other chemical. Analysis of amino acids was particularly
difficult. Hydrolysis of proteins yielded a mix of up to 20 amino acids, many
having very similar chemical and physical properties. In 1910 Osborne
lamented the slow progress being made in this field and the difficulty of
accounting for much more than one-half of the material present in a hy¬
drolysate Even a mix of known amino acids gave only 66% recovery with
the separation methods available.51 Thirty-five years later, the advances were

51. Osborne and Jones (1910).

133
 Protein and Energy

reviewed in a monograph listing more than 700 references. Some amino

acids were still being isolated and weighed, but for others specific color
reactions were being used, with the depth of color obtained being taken as
a measure of the amount of a particular amino acid present.5*
In the 1940s a new approach was being developed, stemming from the
work of microbiologists interested in developing purified media for growing
lactic bacteria. These bacteria, living normally in milk that is uniformly rich
in nutrients, have lost most of the powers of synthesis characteristic of other
bacteria and have nearly all the nutrient requirements of the higher animals.
Lactic bacteria were first used to assay materials for the content of the B
vitamin riboflavin; the medium employed was designed to support rapid
growth only when the vitamin was added. The rate of multiplication or
acid production was then found to be proportional to the dose of vitamin
added at suboptimal levels.53 By 1943 it was realized that lactobacilli could
similarly be used to assay an individual amino acid by using a medium
containing all the other growth factors required, including a mix of all the
amino acids except the one being tested.54 The quantities required, as com¬
pared with those used in either traditional analytical methods or rat ex¬
periments, were tiny. Good responses could be measured in a small test tube
after 48 hours of incubation. The expense was also small, and a large number
of assays could be run at any one time.
By 1946 Block and Mitchell, in a now-classic review, related the amino
acid composition of 23 protein-containing foods to their biological value
determined with rats.s 5 As a starting point, they hypothesized that the amino
acid composition of whole hens’ eggs, designed by nature for the synthesis
of animal tissue, was ideal, and the level of each essential amino acid in
another food’s protein was expressed as a percentage of the corresponding
value for egg protein. It was judged that the amino acid present at the lowest
level (in proportion to egg) would be the first-limiting amino acid for the
synthesis of growing animal tissue. Thus if a particular protein had only
60% of the lysine content of egg protein and this was the lowest percentage
value, then 10 g of this protein would result in the synthesis of 6 g of mixed
animal protein, but then there would be no lysine left and synthesis would
stop. The excesses of the other amino acids would then be metabolized to
provide urea plus carbon skeletons or carbon dioxide and energy.
Originally such a protein was described as having a “40% deficit,” but
it later became the usual practice to refer instead to it having a “chemical
score” of 60% or just 60. A plot of the original 23 sets of results showed

51. Block and Bolling (1945).

53. Snell (1979).
54. Mitchell and Block (1946).
55. Block and Mitchell (1946). pp. 160—7.

134
 Vitamins and Amino Acids

a clear tendency for the materials with a higher chemical score to have a
higher biological value. However, there was not an exact correspondence.
To use the statistician’s term, there was a “correlation of 0.86” (compared
with 1.0 if all the values were to fit exactly onto a straight line when plotted
in the same way). This means that we can expect, if we know that the
chemical score of a sample is 50, that the corresponding biological value
will be within 7 units of 70 - that is, to be in the range from 63 to 77 -
and that 1 sample in zo might be twice as far from the best estimate of 70.
The plot also showed that even a sample with a zero chemical score might
still be expected to have a positive biological value.
The last point is explained by the apparent capacity of rats to maintain
nitrogen balance with a protein completely lacking lysine, this being the
limiting amino acid for growth for many of the samples. Other factors
adding variability to the plot are inherent inaccuracies in the analyses, as
well as variability in rat responses. Finally, the amino acid pattern of egg
proteins intended for the growth of chicks may not contain exactly the
minimal requirement of each amino acid for optimal utilization by rats.
Despite these shortcomings, it was possible to confirm for a good number
of foods that the amino acid calculated to be first-limiting was in fact
the only supplement that gave the food an improved biological value when
it was tested with rats. It was also calculated for several mixtures of two
foods that showed a complementary effect, as discussed earlier in the
chapter, that this could have been predicted from their amino acid analyses.
To return to the example of 1 part of milk protein and z parts of white
flour protein, the first-limiting amino acid in white flour was calculated
to be lysine, giving it a score of only z8, and that in milk to be methionine
plus cystine, with a score of 68. The score for the mixture was 53, though
by simple proportion it would have been only 41.56 In popular writing,
the protein in a vegetable food is sometimes described as “incomplete.
This term is misleading, since all the amino acids are present, but not in
ideal proportions.

Studies with Humans

Obviously, workers were eager to know how far the results obtained with
rats were relevant to human nutrition. To determine the complete list o
amino acids essential for humans by direct experiment required much larger
quantities of L-amino acids than were needed for the work with rats, but
Rose was able to publish the results in 1949 in preliminary form. Eight
of the 10 amino acids found to be essential for maximal growth in young

56. Ibid., pp. 2.yz-3.

57. Rose (i949)-

135
 Protein and Energy

rats were also found to be required by adult men. The exceptions were
histidine and arginine. Further work on the quantitative requirements of
adults for amino acids was carried out after 1950 and will be considered
in Chapter 11.
It had been relatively easy to carry out nitrogen balance trials of the type
pioneered by Thomas using protein foods as the sources of nitrogen. Un¬
fortunately, the results in the 1920s and 1930s continued to be contradic¬
tory. This was thought to be due in part to adaptation periods being too
short and to total energy intakes not being controlled.58 Mitchell, using his
long experience in nitrogen balance work, collaborated on a carefully con¬
trolled study published in 1945.59 Nine women students served as subjects,
energy intakes were controlled to maintain body weight and each protein
food, or mix, was fed at five different levels. It was estimated that the proteins
of cow’s milk had a biological value of 74, and the proteins of white flour
a value of 41. Both of these are lower than the values of 84 and 54 obtained
with rats for the same materials; however, the relative proportions are
similar. The mixture of proteins in a typical U.S. diet had an intermediate
biological value of 65 and 90% digestibility. The mean daily requirement
of the mix for nitrogen balance was estimated to be 50 g crude protein for
someone weighing 70 kg, after making an allowance for what the authors
described as “adult growth,” including losses of nitrogen from skin, hair
and sweat. Similar results were obtained in the next study, with a diet in
which one-third of the protein came from meat and the remainder from
mixed plant sources. When the meat was replaced by an equivalent level of
protein from white bread, 23% more total protein was required, on average,
to achieve balance.
The results from the more carefully designed human studies seemed to
indicate that measures of protein quality that had been obtainedVith rats
would prove applicable to humans as well.60 Yet the quantities of protein
needed to produce nitrogen balance in human adults, even with relatively
poor quality protein foods, were lower than the protein intakes in the
practical dietaries that investigators had studied. There was therefore no
obvious practical problem of protein deficiency, and a common saying
among nutritionists at the end of the period was: “Look after the calories
and the protein will look after itself.”
This did not apply to subjects who had suffered severe traumas such as
bone fractures, major surgery, certain types of infections or extensive burns.
They went into a state of shock, which resulted in greatly increased losses

58. Leitch and Duckworth (1937).

59. Bricker, Mitchell and Kinsman (1945), pp. 278-9.
60. Ibid., pp. 273-8; Hegsted, Tsongas, Abbott and Stare (1946), pp. 272-9.

136
 Vitamins and Amino Acids

of urinary nitrogen.61 It was believed that paying special attention to their

subsequent intake of protein aided greatly in their recovery.
There was concern, during the Great Depression in Western countries in
the 1930s, that low-income or unemployed workers and their families were
malnourished because they could not afford a balanced diet. However, the
conclusion from the best-known study from this period was that the com-
mon deficiencies were of vitamin A and calcium rather than of protein.
After World War II, when food was still scarce in Germany, studies in two
orphanages showed that 9- to 10-year-old children thrived and made up
for deficiencies in both height and weight when they received as much bread
as they wished to eat, together with vitamin and mineral supplements, but
less than 9 g of animal protein per day. In fact, they did just as well as
other children receiving three times as much animal protein.

The Dynamic State of Body Protein

Meanwhile, the use of new techniques based on developments in physics

and chemistry during the 192.0s and 1930s were beginning to have a rev¬
olutionary effect on biologists’ ideas about the behavior of the constituents
of the body. These new techniques involved the use of isotopes. It had come
to be realized that ordinary nitrogen (to take an example) is actually a
mixture of atoms having an atomic weight of 14 and a small proportion
(0.38%) having a weight of 15. The ISN atoms, as they are called, are
isotopes of nitrogen; they have an additional neutron in their nucleus but
the same pattern of electrons in orbit around the nucleus. It is the latter
that gives the isotope the same chemical and biological properties as ordinary
I4N.64
Similarly, ordinary hydrogen contains a small proportion, less than 1
atom in 10,000, of “heavy hydrogen,” or “deuterium,” with an additional
neutron in its nucleus and an atomic weight of 2 instead of 1. Once physicists
and chemists had discovered ways of accurately measuring the ratios of the
different isotopes present in a particular sample, and then of preparing
materials with high proportions of the same isotope, biological experiments
could begin. Fortunately it appeared that living organisms treated these
isotopes exactly as they did the more abundant forms of hydrogen and

H^Th^scientist whose name is particularly associated with the early work

61. Cuthbertson (1931); Cuthbertson (1941); Lund and Levenson (1948).

62. Orr (1937)) P- 41-
63. Widdowson and McCance (1954b PP- 5Z~9-
64. Clarke (1948), PP- 3-J5-

137
 Protein and Energy

in this field is Rudolf Schoenheimer. He came to Columbia University in

New York in 1933 as a Jewish refugee from Nazi Germany. Columbia was
the leading center for the separation of isotopes at that time* and he was
able to collaborate there with David Rittenberg and others experienced with
these isotopes. Tragically, Schoenheimer was to commit suicide in 1941,
when still only 43 years old, but by then their work had totally changed
people’s perception of the body’s metabolism.65
We will discuss one experiment in enough detail to give an idea of the
approach, though not of the work and technical expertise required.66 A
quantity of the amino acid leucine was prepared with a high proportion of
the nitrogen atoms labeled withISN and the hydrogen atoms in the side
chain labeled with deuterium. This was then added to the diet of four mature
rats for a period of 3 days at a level equivalent to that present in the protein
of the diet (15% casein). The rats were then killed and their tissues analyzed.
From the deuterium content of the leucine isolated from the rat carcass
protein, it was concluded that 6.7% of it had come from the leucine con¬
sumed in the previous 72 hours; this amounted to 32% of the leucine
consumed.
This was in striking contrast to the ideas of protein metabolism held
by nutritionists. They had found that the endogenous loss of urinary
nitrogen (i.e., the loss on a nitrogen-free diet with adequate energy intake)
was approximately 58 mg/day for rats of the size (325 g body weight)
used by Schoenheimer.67 The body protein of such a rat contains ap¬
proximately 8,300 mg nitrogen. The daily urinary loss therefore corre¬
sponds to 0.7% of the body’s store of protein nitrogen. It had therefore
been thought that this corresponded to the daily “wear and tear” of body
protein and that for an animal to remain in nitrogen balance |he same
quantity of new protein would have to be synthesized each day. In the
3 days of Schoenheimer’s trial this would correspond to 2.1% of body
protein. And some workers had believed that the breakdown (and need
for replacement) would actually be less if the animals were in a better
nutritional state.
However, Schoenheimer’s results indicated that more than three times
this amount of leucine was incorporated into body tissues. He postulated
that the peptide chains of body proteins opened and reclosed, allowing
individual amino acids to escape and be replaced by others. It was a difficult
concept to accept. One suggestion was that the newly arrived leucine might
be lodged in a kind of temporary storage material rather than in true “work¬
ing” body components. Schoenheimer and his colleagues countered this with

65. Ratner (1979); Guggenheim (1991).

66. Schoenheimer, Ratner and Rittenberg (1939).
67. Smuts (1935), pp. 419-21.

138
 Vitamins and Amino Acids

a demonstration of the rapid labeling of a specific antibody protein in the

blood at a time when its concentration in the blood was actually declining.68
The same research group also demonstrated that the rapid appearance of
labeling in body proteins was not peculiar to leucine. Moreover, when the
ISN label was followed, it seemed clear that different amino acids were
actively exchanging amino groups. Thus when leucine was fed, only one-
third of the I5N found in the rats’ body proteins was still attached to leucine;
the remainder was attached to other amino acids, most notably glutamic
acid. Only the lysine remained entirely free of ISN. All these findings were
to be confirmed and extended by other workers.
These were disturbing results. Work on the metabolism of body fat had
similarly startled biochemists as to the Dynamic State of Body Constituents,
the title of Schoenheimer’s final monograph. As he wrote, If the starting
materials are available, all chemical reactions which the animal is capable
of are carried out continually,” and at another point, “The synthesis of
amino acids, like that of fatty acids... proceeds even where there is no
obvious need for it.”69
Schoenheimer did not envisage that whole proteins were being dismantled
but, rather, that proteins while still in position in the tissues opened up at
particular points to release an individual amino acid, which was rapidly
replaced by another molecule of the same compound. To use an analogy,
it was as if one visited Venice and saw builders swarming over the face of
every building, chipping out stones, throwing them into boats drifting by
on the canals and replacing them with exactly comparable stones from the
next boat that came along. Then to complete the analogy one would have
to imagine there being one particular part of the city where any building
stones that drifted by were broken into rubble and sent to an external dump,
so that the city required a continual supply of new building stones arriving
from outside, in exactly the same condition as those being systematically
destroyed elsewhere in the system.
The picture indeed boggles the mind, and we will return to it in Chapter
ii One recalls Claude Bernard’s criticism of balance studies made some
So years earlier, when he said that one could not hope to understand what
went on inside a house just by analyzing what went in the doors and what
came out of the chimney.70 And the new developments were certainly a
blow to the self-respect of nutritionists, who had felt that because they had
discovered nutrient requirements and interrelationships (such as methionine
being able to replace cystine, but not vice versa) they were at the cutting
edge of biochemistry.

68. Schoenheimer (1942), P- 36-

69. Ibid., pp. 31* 39-
70. Holmes (1964), p. xcviii.

139
 Protein and Energy

Mitchell was one of the few nutritionists even to discuss the significance
of the findings of Schoenheimer and his successors.71 He accepted the im¬
portance of the work, but argued that there must still be some unknown
regulating system at work that allowed organisms to come into nitrogen
balance on widely different intakes. Although it could no longer be assumed
that all of the extra nitrogen that appeared in the urine of someone on a
high-protein diet came directly from digested amino acids, the reality of
determined nutritional requirements was unchanged and a net nitrogen
balance was still a measure of dietary adequacy of protein for adults.

Summing Up *

In the 40 years covered by this chapter many things were explained, but a
new mystery developed. The earlier failure of dogs to thrive indefinitely on
some low-protein diets was explained by the diets being deficient in the
newly discovered trace nutrients - vitamins. The failure of early amino acid
mixtures to replace proteins in nutrition was also explained. In the case of
acid hydrolysates of good-quality protein, it was due to tryptophan having
been destroyed, and in work with mixtures of purified amino acids it was
due to their lacking methionine and threonine. It now seemed clear, at least
from animal experiments, that the true nutritional requirements were for
amino acids rather than protein and that, when vitamins were also supplied,
the level of protein did not need to be high provided that it contained a
reasonably good balance of the essential amino acids. Many individual
isolated proteins (such as gelatin and zein) were of low biological value,
and those of whole vegetable foods also tended to be somewhat poorer
nutritionally than those of whole animal foods, but the combinations present
in ordinary mixed diets were at least of adequate quality in all the examples
studied. The newer studies also indicated that nitrogen balance was obtained
at levels of protein intake below even those recommended by Chittenden,
and there seemed no reason to be concerned about protein being deficient
in human diets.
Mendel put it eloquently in 1923: “The recognition of the relative im¬
portance of the nitrogenous foods for nutrition... brought about a glori¬
fication of the albuminous substance... which has persisted in its extreme
form almost until the present time.” Now, he added, “the pendulum of
enthusiasm about the proteins has swung from one extreme to the other.”71
The new mystery was the unexpectedly dynamic state of body constitu¬
ents, as found in the work with isotopes. Specifically with regard to proteins,

71. Mitchell (1942), pp. 257-62.

72. Mendel (1923), p. 14.
 Vitamins and Amino Acids

‘amino acids were apparently constantly leaving and reentering tissue pro¬
teins. The fact that amino acids leaving protein were then reusable made it
difficult to continue explaining the adult need for amino acids by the old
concept of “wear and tear.”
8 Protein Deficiency as a Third World
Problem, 1933—1957

The Archives of the Diseases of Childhood for 1933 included a

short article that described a serious disease, and apparently a form of
malnutrition, in the Gold Coast (then a British colony, but now Ghana) in
West Africa. The author said that it appeared to be distinct from any of
the usual deficiency diseases.1 * The only response to this article was a note
in the next volume of the journal from H. S. Stannus, the recognized British
authority on deficiency diseases in the tropics, saying that the author was
in error and that, from its description, the disease was clearly the infantile
form of pellagra/ Nevertheless, 50 years later the Archives reprinted the
article in full, under the editorial comment that it was perhaps the most
important article the journal had ever published, and much of the work to
be covered in this and the following chapters can be seen to have stemmed
from it.
*

Kwashiorkor

The author of the article was Cicely Williams, an Englishwoman born in

1893 who had grown up in Jamaica and received her medical training in
Oxford and London. Jobs were scarce at that time for young women doctors,
and she applied to the Colonial Medical Service, hoping to be posted to
Jamaica.3 After a z-year delay she was appointed in 19Z9 to a position in
West Africa — still a dangerous place at that time for Westerners without
resistance to yellow fever and other endemic diseases. Her special interest
was in infants and young children, and she was disturbed by the high death
rate among them. One disease, in particular, the subject of the article already
referred to, seemed to occur almost entirely in children 1 to 4 years old

1. Williams (1933).
z. Stannus (1934), pp. 733—4.
3. Dally (1968); Darby (1973); Craddock (1983).

14Z
 Protein Deficiency as a Third World Problem

who had been weaned early (by local standards) and then fed entirely on
gruel made from white corn (maize) after some degree of fermentation.
Those who were not seriously ill generally recovered if given cod-liver oil
and canned condensed milk as a supplement. Although goats were kept
locally, they were not milked. It was not customary for anyone to drink
animal milk.
In her introduction Williams said that the disease was characterized by
“edema [swelling from accumulated fluid], chiefly of the hands and feet,
followed by wasting, diarrhea, irritability, sores (chiefly of the mucous mem¬
branes), and desquamation [peeling off] of areas of the skin in a manner
and distribution which is constant and unique.... It appears to be due to
some dietetic deficiency and to be uniformly fatal unless treated early.”
She then described, as a typical case, that of an 18-month-old child. First
there was a slight edema; then patches of skin became “dark, thickened and
crumpled” in a sort of “crazy pavement” pattern. As the older patches
matured they stripped off, leaving a pink, raw surface. The child was now
in a state of misery, had persistent diarrhea and would die in a few more
days if untreated. Later photographs of children suffering from the condition
and from marasmus (“simple” starvation) are reproduced here (Figure 8.1).
Hurried postmortem examinations carried out under difficult conditions
showed nothing abnormal except fatty livers.4 Finally, Williams apologized
for a lack of references to published literature, explaining that she had no
access to a library.
In 1935 she published a paper in the Lancet based on her experience with
60 further cases.5 Here she used for the first time the local name for the
disease — kwashiorkor — which a nurse from the local community told her
meant “the sickness that the older child gets when the next child is born.”6
Stannus again issued a quick rejoinder, criticizing Williams for failing to
accept his earlier conclusion that the disease was infantile pellagra and also
for choosing to introduce a native name for the condition.7

Was It Pellagra?

What was the significance of this dispute? First, if the condition were really
a known disease, then the proven forms of treatment should be adequate.
Second, since intensive research was in progress in the United States on the
identification of the pellagra-preventive factor using the dog as a model
animal, there would be no reason for a separate study of the factor needed

4. Williams (1973), 337; Craddock (1983), pp. 63-4.

5. Williams (1935).
6. Dally (1968), p. 59.
7. Stannus (1935).
9

143
 (d)

Figure 8.1. (a) An African child with kwashiorkor; (b) another with mar¬
asmus (“simple” starvation) for contrast; (c) edema of the face of a child
with kwashiorkor; (d) skin lesions typical of kwashiorkor. (Dr. R. G.
Whitehead)

144
 Protein Deficiency as a Third World Problem

to prevent the development of kwashiorkor. Cicely Williams’s connection

of the disease with a corn diet must have strengthened the idea that it was
a form of pellagra, since pellagra’s association with corn had been seen in
so many other parts of the world.8
A similar controversy had arisen in the case of another deficiency disease,
scurvy. Barlow’s disease in infants showed some differences from adult
scurvy, and many physicians had therefore refused to consider it essentially
the same disease despite its occurrence with the same type of foods as were
associated with adult scurvy. But in the 1920s, with better histological
studies of the changed tissues, the condition was accepted as infantile
scurvy.9 Although Stannus’s dogmatic and even contemptuous reaction may
be partly explained by the alleged identification of kwashiorkor as a new
disease having come from a doctor inexperienced in tropical work, and from
a woman, the identification of the disease with infantile pellagra was made
by others in Africa as well as himself.
More constructively, another medical officer on the Gold Coast published
a note in the Transactions of the Royal Society of Tropical Medicine and
Hygiene in 1935, saying that he had seen a case exactly like those described
by Dr. Williams and that, although its origin was still obscure, he entirely
agreed that it was “impossible to confuse it with any of the well-known
and often described diseases.”10 This journal circulated widely in the British
tropical colonies and a response came, within the same year, from a medical
officer in Kenya, saying that an identical condition had attracted attention
there in the past two ye;ars, though pallor of the skin and hair changes
seemed to be more constant features than had been seen by Williams. The
majority view was that “some form of food deficiency is the root of the
trouble.”11 There was no reference to the possibility of the disease being a
form of pellagra.
We can now see from literature reviews that cases with the characteristics
of kwashiorkor had been reported in at least 11 other papers by this time
from different parts of the world and in different languages, but without
their being linked.11 At this point Cicely Williams found herself unexpectedly
transferred to Malaya, where she was to remain through World War II and
to endure great hardship in Japanese prison camps.13 But the disease that
she had been so interested in was, from then on, to receive intensive study,
particularly in East and South Africa.
Hugo Trowell, an English physician working in Kenya in the early 1930s,

8. Harris (1919), pp. 33-132.

9. Carpenter (1986a), pp. 166-72.
10. Sharp (1935).
11. Carman (1935).
12. Trowell, Davies and Dean (1954), pp. 2-8.
13. Craddock (1983), pp. 105—23.

145
 Protein and Energy

wrote, much later, that medical missionaries were already familiar with the
disease and had provisionally thought of it as “protein deficiency” because
of the edema. However, they and Trowell himself had accepted Stannus s
apparently authoritative statement that it was infantile pellagra. In 1935,
when Trowell was transferred to Uganda, he saw many more cases, which
were described there as “congenital syphilis.”14
Trowell published a review of the work done up to 1940 under the title
Infantile Pellagra. He cited papers indicating that the same disease occurred
in Mexico and Costa Rica as was seen in Africa (including the Belgian
Congo, as it was then). One had called it pellagra-beriberi, attributing the
skin changes to pellagra and the edema to beriberi (recognized as being
caused by thiamine deficiency). However, he had injected thiamine^and not
obtained any reduction of edema.15 He therefore suggested that protein
deficiency caused the edema (as it did under wartime conditions in adults),
but that the other changes were a form of pellagra, which had been shown
in the United States to respond to the vitamin niacin (nicotinic acid).
Trowell injected seven sick children with niacin. Of these, two died and
two more were removed from the hospital by their parents after they had
shown no rapid improvement, but the three remaining all showed significant
improvement after 5 days. All the children also received in this period “as
much milk as they could take, an egg each day, and a little bread in addition
to the usual hospital diet of steamed plantains [starchy bananas].” Trowell’s
tentative conclusion was that “the major deficiency is one of acute pellagra
associated probably with nutritional edema_treatment should be by ni¬
cotinic acid injections, blood transfusions, marmite [yeast extract] and a
diet rich in protein.”16
Williams replied from Malaya to Trowell’s article in a rathenconciliatory
way, agreeing that the East African cases were essentially the same as those
she had seen and going on to say: “What’s in a name? It matters very little
as long as we understand each other.” Nevertheless, the African children
did not display the same type of skin changes as those that she had, by then,
had the opportunity to see in New Orleans, where there had been a few
cases of pellagra in children; their skin changes were all on the areas exposed
to sun, as in adults. Williams was willing to accept that individuals could
be suffering from complex deficiencies but concluded that, until much more
was known of the biochemistry of these things, “it is unwise and confusing
to the issue to label a variety of symptoms with a name that implies an as
yet unproved identity.”17

14. Trowell, Davies and Dean (1954), p. 18.

15. Trowell (1940), pp. 389-98.
16. Ibid., pp. 398-403.
17. Williams (1940).

146
 Protein Deficiency as a Third World Problem

Trowell and Muwazi in 1945 used the new term “malignant malnutri¬
tion” for the disease.18 They explained that they did not like the term
“kwashiorkor” because they understood that it actually meant “red boy”
in the Ga language of the Gold Coast, and although a change in hair color
had been noticed it was probably specific to Africans and not a necessary
condition for diagnosis of the disease.19 However, it was agreed later that
the word really did mean “the deposed child.”10
The authors had seen the condition in 48 children from 7 months to 3
years of age. Almost all the children in the region of Kampala, the Ugandan
capital, had contracted malaria by their third month, and hookworm disease
was also common after the second year. “Malignant malnutrition” was seen
mostly in the local Ganda tribes, whose babies were not weaned until well
into their second year, and then restricted to three meals per day of cooked
plantains, sweet potatoes and tea/1 The absence of corn in the diet of these
children may also have influenced the authors’ decision to give up describing
the disease as a form of pellagra.
Another striking finding was the greater number of cases seen among
adults than among children. Most of the 144 adult sufferers were from the
Belgian mandated territory of Ruanda-Urundu and had walked 500 to 800
miles to obtain work in the plantations around Kampala. They had been
short of food on their journey and, after arrival, lived typically on sweet
potatoes together with very small amounts of meat and vegetables. It was
estimated that they had daily intakes of approximately 2,000 kcal, 25 g
protein, 1 g fat, 250 mg calcium and 8 mg iron. They had very low body
weight, edema, “crazy pavement” dermatosis and diarrhea, but not always
pallor of the skin and hair. All autopsies showed fatty livers. Both the
children and adults had abnormally low levels of albumin in the blood
serum, but increased globulin levels. The adults responded, but only very
slowly, to a standard hospital diet of plantains, meat, beans and potatoes,
supplemented by liver, milk, thiamine and nicotinic acid/1 That the adult
patients did not have the appearance of pellagrins may have been another
reason for Trowell’s change of name for the infantile disease.

Failures with Vitamins

In the same year, 1945, another paper appeared, this time from Johannes¬
burg, reporting more cases, still described as “infantile pellagra” but char¬
acterized by edema. The authors carried out liver biopsies and reported that

18. Trowell and Muwazi (1945), p. 2.41.

19. Ibid., p. 142..
zo. Trowell, Davies and Dean (1954), p. n.
zi. Trowell and Muwazi (1945), p. Z31.

zz. Ibid., pp. Z30—9.

147
 Protein and Energy

fatty infiltration of the liver was a constant and early feature of the condition.
They treated seven patients with combinations of nicotinic acid, riboflavin
and thiamine, but all died. Another six patients were treated with io g/day
of “Ventriculin,” a proprietary, dried extract of hog’s stomach, and within
a week they all showed a spectacular recovery.13 Further trials with the
same batch of Ventriculin were also highly successful, and Trowell, too,
reported successes.14 Unfortunately, later trials with stomach extracts were
much less successful, and even a later .batch of Ventriculin appeared rela¬
tively ineffective.15 No explanation has been found for these apparently
contradictory results.16
In 1946 a medical officer in Southern Rhodesia (now Zimbabwe) reported
further cases. He considered that the cause could not be “dietetic” since the
disease did not occur during the season of greatest food scarcity and was
seen in children whose diet differed in no way from those who remained
healthy. His suggestion was that the central problem was liver damage from
some toxic factor in certain batches of corn and cassava.17
From this period on, papers on the subject began to appear in increasing
numbers.18 In the next two years, it was reported from South America that
the disease could occur in white children as well as people of African stock.
There was general agreement as to the value of milk and to the opinion
that this was due to its protein fraction. It was also suggested that damage
to the pancreas might precede liver damage and that treatment of the disease
should be directed to restoring pancreatic function, since this was judged
to be the organ most damaged.19
In 1949 Trowell again reviewed the subject. He still believed that it was
a complex syndrome: “The strands of the weft are... total calories, protein,
vitamin B complex; the warp... many different strands of an infective na¬
ture, helminths, pneumonia and so forth.” Even apparently trivial infections
had to be treated rigorously. “All sources of animal protein should be high.
Milk should be at least one pint.” Dietary protein should always be suspected
as the bottleneck to recovery if edema was severe and serum albumin levels
very low, and the latter was the most constant feature of the condition. “To
correct this we give powdered milk, like a medicinal powder several times
a day.... all vitamins are forbidden, so are unnatural foods like dehydrated
liver, desiccated stomach, proteolysed meats, casein hydrolysates. Every¬
thing must be directed to overcoming the false economy which would de-

23. Gillman and Gillman (1945).

24. Trowell (1946).
25. Gillman and Gillman (1951), pp. 301-2.
26. Trowell, Davies and Dean (1954), p. 182.
27. Gelfland (1946).
28. Trowell, Davies and Dean (1954), pp. 1-46.
29. Davies (1948); Waterlow (1948a); Waterlow (1948b).

148
 Protein Deficiency as a Third World Problem

prive wards of essential diets but would flood them with expensive vitamins
and quack foods.”30

Investigation by the United Nations

After World War II, the new United Nations Organization (UNO) set up
subsidiary agencies, including the World Health Organization (WHO), with
its headquarters in Geneva, and the Food and Agriculture Organization
(FAO), in Rome. In October 1949 representatives of these two agencies met
to coordinate their plans for nutritional work. First on the agenda was
endemic goiter and the practical problems of supplying iodine preparations
in ways that would prevent its appearance in deficient areas. Second on the
list was kwashiorkor. It was believed that cirrhosis (fibrous degeneration
of the liver) found in many adults in Africa and Central America might be
the result of their having suffered from the disease as children. It was im¬
portant for WHO to extend its studies to areas where the disease did not
occur but in which the diet was apparently similar to those where it did.
There were no suggestions as to what nutritional deficiencies might be
responsible for the condition, but it was urged that more animal milk should
be made available for weanlings, or “where that was difficult, the production
and use of foods and/or preparations which can act as a partial substitute
for milk should be vigorously encouraged.”31
Dr. J. F. Brock, professor of medicine at Cape Town, and Dr. M. Autret
(on the staff of FAO) toured Africa in 1950 to investigate the occurrence
of kwashiorkor and the different methods used for its treatment.31 They
reported that essentially the same disease was a widespread and serious one
among infants in many areas, though they did not make any estimate of its
actual prevalence. Growth retardation, edema and low serum protein levels
were the most constant factors. Dyspigmentation, dermatosis and diarrhea
were all variable. Livers were uniformly found to be fatty, but not always
enlarged. Mortality was very high unless the condition was treated. It seemed
inappropriate to call the condition “malignant malnutrition,” since it had
been treated successfully with skimmed milk.
Brock and Autret concluded that both the epidemiology (i.e., freedom
from the disease in areas where meat, fish and/or milk were relatively abun¬
dant) and the successes with skimmed milk and some good-quality vegetable
protein sources pointed to protein deficiency as the cause of the disease.
However, it was still possible that the missing factors were particular amino
acids or a vitamin such as vitamin BIZ that occurred in association with

30. Trowell (1949).

31. Joint FAO/WHO Expert Committee on Nutrition (1950), pp. 15-16.
3Z. Brock and Autret (1952.), pp. 32.-4.

149
 Protein and Energy

these proteins. Methionine was a possible limiting factor since a deficiency

of this amino acid led to fatty deposition in the livers of experimental
animals.
Despite the proven value of skim milk, it was impractical to expect milk
production to increase rapidly in Africa, and imported material would be
too expensive for general use, though it was felt that skim-milk powder
should be made available to medical centers for use in intensive treatment.
Substitutes for milk were required, and these would have to be mostly
vegetable products: cereals in place of cassava, which was particularly low
in protein, and legumes (peas and beans) acceptable to African people.
Studies were required of the ability of infants to digest different protein
sources and of the effects of traditional processing that involved fermen¬
tation or germination of cereal foods.

Effects of Protein Deficiency in Adults

Brock and Autret also made a separate point. Two adult diseases were
unusually common in the areas studied - cirrhosis and primary carcinoma
(cancer) of the liver. It was thought that parasitic infestations could be
responsible in some areas. However, in the Johannesburg area these diseases
were common even though parasitism was not a major problem. Since
kwashiorkor, characterized by liver damage, now seemed to be caused by
a dietary deficiency of protein, it seemed likely that adult cirrhosis was
also the result of prolonged protein deficiency, though the two diseases
were not necessarily directly related to each other. Kwashiorkor therefore
pointed to a much wider problem: “It would not be too far-fetched to
attribute to protein deficiency, at least in part, the backwarcjpess of the
African people_Protein deficiency, though alleviated after the post-
weaning phase of life, appears at least in a part of the African population
to be permanent throughout life... causing a great volume of premature
morbidity and mortality.”
According to Brock and Autret, it was difficult to estimate the nutrient
content of African diets because quantitative data were not available. “Ac¬
cordingly, a unit of ioo calories has been adopted for comparing different
foods.” (Actually, the authors should have written “kcal” instead of “cal¬
ories” because they were referring to the energy required to raise the tem¬
perature of i kg water by i°C.) They compared the concentration of protein
in different foods with the estimated requirement for protein (in the same
units) at different ages. The maximum protein requirement was given as
4.5 g protein per 100 kcal for children z to 3 years old.33 Since the time of
this report, it has become more common to convert such values to a simple

33. Ibid., pp. 56-8.

150
 Protein Deficiency as a Third World Problem

Sklm-mllk powder
Meat

I I I
Beans and peas

Whole milk

Ground-nuts

Wheat flour
Millet; sorghum
Maize
Yams
Rice (half-milled)
White rice ; taros
Plantains ; sweet potatoes
Fresh cassava
Cassava flour
0 5 10 IS 20 25
AGE (YEARS)

Figure 8.2. Brock and Autret’s 1952 diagram relating the protein concen¬
tration in foods to human needs at different ages for average-quality pro¬
tein, all per 100, kcal with the protein units changed from grams to
kilocalories to give PCals%.

percentage by assuming that dietary protein has a usable energy equivalent

of 4 kcal/g. Then 4.5 g is equivalent to 18 kcal, and the requirement for 2-
to 3-year-olds becomes “18 PCals%,” meaning that 18% of the usable
energy in the diet must come from protein. The corresponding value for
20-year-olds at 3 g protein per 100 kcal was “12 PCals%.” The authors’
diagram is reproduced here in these units (Figure 8.2). It was clear that
plantains, sweet potatoes and cassava, with only 0.5 to 1 g protein per 100
kcal (2—4 PCals%), were grossly deficient in protein. Even the cereal grains
(wheat, rice, corn) would be judged inadequate for the youngsters. Where
plantains were the staple food, the diet as a whole could reach the standards
for protein only if it included considerable quantities of foods rich in protein,
such as fish, meat, milk and beans.
The authors also estimated the supplementary food requirements of a 12-
month-old African child receiving only 86% of its calorie requirements from
breast milk with 2.2 g protein per 100 kcal (i.e., 8.8 PCals%). Their cal¬
culations were as follows:
 Protein and Energy

kcal Protein (g)

Daily requirement of the child 730 20

Supplied by breast milk 630 14
Deficit 100 6

For ioo kcal of supplementary food to supply the full protein deficit it
would have to have at least 6 g protein per ioo kcal (or 24 Pcals%), that
is, the level in beans or equal parts of skim-milk powder and cereal.
Using the same values, one would find that, even if the 12-month-old
child were to receive enough breast milk to satisfy his or her full energy
requirements, this would provide only 16.2 g protein, which is short of the
requirement by 3.8 g. It seems strange that human milk should itself appear
“unbalanced” even when consumed at a level that meets calculated energy
demands in full, but the authors were forced to choose standards for protein
of “average quality,” and milk protein is clearly “above average.”
The joint FAO/WHO Expert Committee met for its second session in
1951. Brock and Autret’s draft report was accepted. It was agreed that milk
production in Africa could not be increased significantly in the short term
and that the use of imported dry skim milk could be only a temporary
measure. Policymakers must stress the provision of locally produced foods.
In some areas it should be possible for fish production tp be increased
rapidly. More information was needed on the kinds of pulse and vegetable
combinations that were effective in preventing kwashiorkor, and obtaining
such information should receive priority in the allocation of research funds.
The portion of the report that dealt with liver disease in adults and its likely
association with protein deficiency was not, it appears, discussed in any
detail. However, the committee noted, “Prevention of the syndrome calls
for a general improvement in the diet of the whole population.”34
Autret and Moise Behar (a research worker in Guatemala and consultant
to WHO) then made a regional study in Central America. There, too, a
disease equivalent to kwashiorkor in its essentials was widespread, regard¬
less of race. Within the area it was called sindrome policarencial infantil
(the multiple deficiency condition of infants). Helminthics (drugs used to
treat worm infestations) and purges were commonly thought to precipitate
the disease, because the infant’s resistance to the side effects of the drugs
was so low.35 One difference between the situation in Central America and
that in Africa was that in the former region 40% of the cases were seen in
children over 4 years old. In general, giving vitamins had had a bad effect.
Milk, and especially skim milk, had proved the most effective for treatment.

34. Joint FAO/WHO Expert Committee on Nutrition (1951), pp. 19, 22.-8.
35. Autret and Behar (1954).

152
 Protein Deficiency as a Third World Problem

Dietary studies had indicated that intakes of calories as well as of protein

were inadequate and that both should be corrected. The following records
for three 3-year-olds in the initial stage of the syndrome illustrate the point
(RDAs indicate recommended daily allowances of the U.S. National Re¬
search Council):36

Actual intakes (% of allowance)

RDAs Child 1 Child 2 Child 3

kcal (1,200) 44 61 37
Protein, g (40) 45 33 38

The Value of Vegetable Proteins

In 1951 the British research group at Kampala was joined by R. F. A. Dean,

who had been a member of the Medical Research Council’s team that studied
diets in German orphanages after World War II. He had obtained very
promising results in the orphanages, even among infants, with mixtures of
cooked soybeans and malted cereals as the main sources of both energy and
protein.37 In Uganda he began with milk powders to provide a standard of
comparison, but found that severely ill children continued to have diarrhea
for long periods when given a large quantity (170 g) of dry skim milk daily.
He thought that this was caused by the large amount of lactose being
consumed. Further tests used a more concentrated source of milk protein,
and better results were obtained when “Casilan,” containing 80% milk
protein, replaced part of the milk powder. When this mixture was given to
severely sick patients, the death rate dropped from approximately 20% to
10%. Diarrhea cleared up more quickly if the milk powders were mixed
with either cooked plantains or mashed sweet bananas, and these mixtures
were also more palatable. The usual aim of all these treatments was to
provide 50 to 75 g protein and 1,000 kcal per day, which had a balance
of 20 to 30 PCals%.38
Dean also carried out trials with soybeans. The dry beans were soaked
in water for 2 days, rubbed to remove the husks, then minced, packed into
jars with water and boiled in water for 10 hours, a procedure designed to
remove their bitter flavor. The resulting “mush” was mixed with the other
ingredients of the test diet in a blender to the consistency of a thick cream.
The first mixes had included a proportion of corn flour. However, such a

36. Ibid., p. 45.

37. Dean (1949); Dean (1953a), PP- 63-110, 115-28.
38. Dean (1952); Dean (1953b), PP- 769-70-

153
 Protein and Energy

large volume of water was needed to bring the mix to a suitable consistency
that the children could not take in enough nutrients. Sick children had a
greatly reduced appetite.
A mix that proved more successful consisted of 300 g cooked soy, 750
g sweet banana, 75 g sucrose, 8 g calcium lactate and 3.5 g of a vitamin
mix. The vitamins included 15 micrograms (|xg) cobalamin (vitamin BIZ).
With this mixture, children recovered from kwashiorkor, though not quite
as rapidly as with the mix based on skim-milk powder and Casilan. The
oil in the soy appeared to be better digested than milk fat. Soy protein had
been found in rat feeding experiments to require supplementation with
methionine in order to support optimal growth, but Dean felt that slower-
growing infants would not have such a high requirement. Methionine, to¬
gether with cystine, the second sulfur-containing amino acid, was needed
in relatively large amounts by young rats because of their growth of fur,
which has a high cystine content.
The practical problem was that soybeans, though grown in large quan¬
tities in some countries bordering the East African coast, were not a common
crop in Uganda. Sunflowers were grown, and although the seeds required
more processing to produce an edible flour, it seemed possible to simplify
the procedure. The final problem of acceptance of a new food might have
been exaggerated, because there was no problem with the acceptance of a
can of condensed milk by people who had the money to pay for it.39 This
work was described and discussed in more detail in the book Kwashiorkor
that appeared in 1954.40 It gives a good summary of knowledge up to 1952.
While this work was still in progress, the FAO/WHO Expert Committee
held its third session at the end of 1952, this time in West Africa, and
“concentrated its attention on protein deficiency and its effect on child
health.”41 The condition was reported in an increasing number of*countries.
In India, there was a condition called “nutritional dystrophy” or “malnu-
tritional oedema syndrome” in many 1- to 3-year-olds living on rice gruel
and little or no milk.4Z In some the condition was complicated by signs of
vitamin A deficiency, which might have been due, at least in part, to poor
absorption of fat. The children typically had diarrhea and heavy roundworm
infestation, and also showed extreme apathy.
The committee’s report listed local names for diseases resembling kwa¬
shiorkor - for example, “infantile edema” (Guatemala), “farinaceous dys¬
trophy” (Uruguay), “red baby” (French Cameroons) and “sugar baby”
(Jamaica) - and put them all under the heading “Names for Protein Mal-

39. Ibid., pp. 771-81.

40. Trowell, Davies and Dean (1954), pp. 195—202.
41. Joint FAO/WHO Expert Committee on Nutrition (1953), P- 4-
42. Achar (1950); Gopalan and Patwardhan (1951).

154
 Protein Deficiency as a Third World Problem

nutrition in Children.” The introduction to the report stated that it was

concerned with “disease in children associated with deficiency of protein in
the diet” and “the prevention of protein malnutrition through appropriate
developments in food production... a problem of fundamental importance
throughout the world.” There was also reference, as before, to the likelihood
that low-protein diets would reduce the ability of men and women to engage
in sustained work.
On the assumption that it was protein deficiency specifically that was
causing the problem, more definitive recommendations could be made as
to desirable changes in food patterns. Cassava (also called manioc or tap¬
ioca), which had originally been brought to Africa as a famine (i.e., reserve)
crop, was becoming accepted as a staple, replacing traditional grain crops.
This was undesirable since it (like plantains) had only 2 to 3% of its calories
in the form of protein, while grains had 8 to 12%. The report stated that
breast feeding should be prolonged where high-protein weaning foods such
as cow’s milk were not available, and that parents should think of their
weanlings as needing “special” foods rather than having to rely on what
they could manage to consume from the adult diet.
The committee warned against technical experts coming in from other
cultures and making recommendations before they had learned existing
customs and beliefs; they should be helpful in their attitude rather than
authoritarian and superior. More research was required on the quantitative
study of diets associated with kwashiorkor, on the nutritive value of the
foods available in problem areas and on the means of detecting malnutrition
in its early stages. Kwashiorkor was distinguished from marasmus, which
was regarded as the result of simple starvation, or calorie deficiency. This
led to a gross loss of both muscular tissue and subcutaneous fat so that the
victims had a shrunken look, in complete contrast to the swollen, plump¬
looking kwashiorkor victims. Hair changes and edema were rarely seen and
the sufferers maintained their appetite. If marasmus was the result primarily
of calorie deficiency, then kwashiorkor must presumably have a different
cause, that is, a deficiency of protein. It was recognized, however, that there
were intermediate conditions with some signs of one disease and some of
the other - for these the term “marasmic kwashiorkor” was sometimes
used.
In 1953 Waterlow (from Jamaica) and Vergara (from Chile) carried out
a study in Brazil, though the report was not published until 19 5 6.43 The
authors .saw numerous cases of the disease known in Brazil as distrofia
pluricarencial and judged it to be essentially kwashiorkor. Skin and hair
changes were not as common as those reported in Africa. The condition
was described in the title of the report and elsewhere in the text as “protein

43. Waterlow and Vergara (1956).

155
 Protein and Energy

malnutrition,” though it was stated that “the intake of calories is also

frequently inadequate.” In certain regions of the country one-third of the
children appeared undernourished. Most of these children were from city
slums and in their second year. It was considered of interest that the disease
occurred in white children as well as in those of other races. Serum protein
levels were very low, but fatty liver was a variable feature and therefore
was not the essential lesion. Cirrhosis and primary carcinoma of the liver
were rare in Brazil. There was no confirmation, therefore, of the idea, based
on the African experience, that these diseases in adults might be long-term
sequelae of kwashiorkor and of continuing protein malnutrition.
In 1953 a conference was held in Jamaica, with z6 participants, from
every continent.44 The title of the conference was Protein Malnutrition,
though several members doubted whether the disease(s) under discussion
could be described in this way. Scrimshaw said that “in INCAP [the Institute
of Nutrition for Central America and Panama] we are far from convinced
that the biochemical changes encountered are specific to protein malnutri¬
tion; rather, they seem to be non-specific.” Another participant said, “The
hypothesis of protein malnutrition in kwashiorkor, fertile though it looks,
has not yet been proved in terms of either qualitative or quantitative de¬
pletion of protein.” However, there was general agreement that “what we
see clinically are cases in which the dominant theme is protein malnutrition,
and that there are local variations in signs and symptoms brought about by
an overplay or a variation in the theme, caused by deficiencies of one or
more of half a dozen nutrients.”45
Gopalan, from India, said that in his experience the diet of children in
poor communities was low in calories. Dean, too, said that children in
Uganda who showed signs of kwashiorkor had quite unexpectedly low
caloric intakes. Even the “sugar babies” in Jamaica, who had been thought
to have a high calorie intake, were reported by Katerina Rhodes to lose
their appetite some z to 4 weeks before developing kwashiorkor.46
Waterlow raised the paradoxical point that marasmus, in which both
calories and protein were equally deficient, seemed easier to cure than kwa¬
shiorkor, in which there was thought to be a better supply of calories (usually
in the form of carbohydrate). Yet it was known that normally a low calorie
intake led to more rapid depletion of protein. There was reason to think
that the kwashiorkor state needed to be triggered by some stimulus, such
as an infection.47
The idea that adult cirrhosis of the liver followed from kwashiorkor was

44. Waterlow (1955).

45. Ibid., pp. 4, 82, 158.
46. Ibid., pp. 43, 179, 250.
47. Ibid., pp. 8, 16, 48.

156
 Protein Deficiency as a Third World Problem

agreed to be no longer tenable since, in addition to its rarity in Brazil, a

high incidence of cirrhosis had been found in parts of Africa where there
was no kwashiorkor and meat was relatively abundant.48
The participants in the conference agreed that the digestive enzymes of
the sick children were grossly deficient, mainly as a result of the abnormal
state of the pancreas, and this made them unable to tolerate fat. Skim milk
had proved about the best treatment everywhere. However, there was agree¬
ment, as at an earlier conference, that locally available plant protein sources
would have to be developed. Work in both India and Mexico had confirmed
that giving victims vegetable mixes based on beans could result in their cure
(and therefore could presumably be used also for prevention), though re¬
covery was usually slower than with mixes containing milk protein. One
report to the conference described the acceptance by schoolchildren of bread
made with a proportion of fish flour, but others commented that 6-month-
old infants might react with diarrhea to a food tolerated by older children.49
In late 1954 the Committee on Nutrition met again in Geneva to review
the main nutrition problems. “Protein malnutrition” was first on the agenda,
with a review of measures being taken, or under consideration, for supple¬
mentary feeding programs. During 1953 enough dried skim milk to provide
about 9 g protein per head daily had been distributed through UNICEF
(United Nations International Children’s Emergency Fund) to 6,000 children
in the Belgian colonies in Central Africa, and it was judged that the condition
of the children had improved. It was hoped that after these supplies ceased
“the protein needs of the people will be met by the extension of fish
farming.”s°
In 1955 another meeting was held in the United States, with the title
Human Protein Requirements and Their Fulfilment in Practice.51 Partici¬
pants were puzzled by reports of kwashiorkor appearing occasionally in
breast-fed infants. Recent analyses had indicated that milk secreted by
women in conditions of poverty still had a normal concentration of protein.
Scrimshaw (from INCAP) also referred to cases developing suddenly in
children although there had been no change in their diet. A stress factor
such as infection was thought to be responsible.51 Senecal (French West
Africa) and Hansen (Cape Town) raised a new point - that kwashiorkor
patients were as depleted of potassium as they were of protein and that
giving potassium salts was important in their recovery.53

48. Ibid., p. 109.

49. Ibid., pp. 207, 220, 253-7.
50. Joint FAO/WHO Expert Committee on Nutrition (1955), pp. 7~8; Holemans, Lambrecht
and Martin (1956), p. 5.
51. Waterlow and Stephen (1957).
52. Ibid., pp. 85-7.
53. Ibid., pp. 91-2.

157
 Protein and Energy

With regard to protein requirements, it was agreed that “the allowances

recommended by previous committees, and based mainly on intakes in
Western countries, are at all ages too high, and to reduce them is not only
realistic but scientifically sound.” Allowances of ioo g protein per day for
adults were thought unrealistic, when even nursing mothers Were lactating
successfully with 70 g or less, and there was no evidence of adolescents
being at risk from protein deficiency during their “growth spurt” period.54
Information on the requirements of both infants and adults for individual
amino acids was reviewed. It was agreed that where the protein in a diet
had a poor amino acid balance, more of it would be needed, but it was not
yet possible to calculate Third World diets in terms of each separate amino
acid; the data were inadequate.55 *
With regard to protein supplies, Autret said that FAO’s first principle was
to develop milk production wherever practicable: “The next approach is
how to use for feeding children various protein-rich foods now used for
animal feeding only. Of one million tons of fishmeal now being produced
in the world each year, 30 to 40 per cent...could be used for feeding
children. In Africa... 500,000 tons of peanut meal are available, which
might well be turned into an edible protein food; there are also large quan¬
tities of soy flour.” Gopalan commented: “The general premise seems to be
that protein malnutrition in under-developed countries is directly and solely
the result of the primary dietary inadequacy of protein_A greater con¬
tribution to the prevention of protein malnutrition in certain regions of the
world may lie, not in supplying supplements, but just in eradication of
malaria, the improvement of poor home economic conditions, etc.... factors
... responsible for the precipitation of protein malnutrition.”
There was debate at the meeting as to how much testing of different
products would be needed before they could be officially recommended.
While some urged the danger of using untried materials, and the possibility
of long-term adverse effects, others said that children were continuing to
die for lack of high-protein supplements and, where similar products had
been fed for years to young chicks and pigs, it seemed unreasonable to hold
them back from infants.56
In 1956 Professor Brock’s group in Cape Town reported that they had
given 33 infants with kwashiorkor a diet containing sugar, minerals, vita¬
mins and mixtures of amino acids (but no protein as such). In 27 of the
children, cure was definitely initiated. In some cases vitamins were withheld
at first, and there was still a positive response.57 The authors concluded it

54. Ibid., pp. 114—16, 133.

55. Ibid., pp. 105-7,
56. Ibid., pp. 138-9, 160-7.
57. Hansen, Howe and Brock (1956).

158
 Protein Deficiency as a Third World Problem

was no longer necessary to suppose that an associated factor (such as an

unknown vitamin) explained the value of milk and other high-protein foods
in the treatment of kwashiorkor.
In 1957 the Committee on Nutrition met for the fifth time, though its
composition had changed considerably over the years. With regard to pro¬
tein malnutrition, the members reported that the first stage of their work
had involved the collection of information about the extent of the condition
and its importance as a public health problem. The second had been the
technical analysis of the problem by experts. The third stage, now in prog¬
ress, was the development of preventive measures, including the effective
use of protein-rich foods other than milk. In late 1955 WHO had set up
the Protein Advisory Group “to advise on the safety and suitability of
proposed new protein-rich food preparations,” and the Rockefeller Foun¬
dation had given $250,000 toward the work. UNICEF had also given
$100,000 toward the procurement of materials and general support of the
program.58 The progress of work of this kind will be the subject of the next
chapter.

Summing Up

In 1933 a fatal disease of young West African children was described for
the first time in the English-language literature. Called “kwashiorkor,” it
was characterized by edematous swelling of the legs, peeling of the skin, an
enlarged liver, great misery and loss of appetite. The sufferers had usually
subsisted mainly on corn gruel since weaning. It was later realized that the
same condition had already been described in other poor countries under
a variety of names in French- and Spanish-language publications and was
widely distributed.
In Africa the condition was first considered to be the result of deficiencies
of B vitamins, but this was not confirmed. And since the condition was
typically associated with the low-protein staple foods plantains and cassava,
and could be treated successfully with reconstituted milk powders, it came
to be classified as “protein malnutrition,” though a further stress such as
infection might be needed to precipitate the acute condition. It was realized
that those affected had low intakes of calories as well as of protein. However,
the clinical picture was quite different from that of marasmus, the condition
of babies suffering from starvation, which was considered primarily a calorie
deficiency. Children with kwashiorkor were also slower to respond to treat¬
ment.
The specialist organizations set up by the United Nations after World
War II considered that correcting the problem should be given the highest

58. Joint FAO/WHO Expert Committee on Nutrition (1958).

159
 Protein and Energy

priority. It was feared that low-protein diets in early childhood and contin¬
ued into adulthood in the Third World resulted in impaired development
that was never made up. There was particular need for alternative protein
sources where it was not practical to expect a local dairy industry to increase
milk supplies greatly.
Brock was to say, with some justification, in i960 that “in human nu¬
tritional studies and in international public health this has been a protein
decade,”59 and another member of the FAO/WHO committee said, “We
have moved from the era of vitamin research to protein research.”60 Autret
wrote in 1962 that it was the assumption of the Nutrition Division in FAO
that “deficiency of protein in the diet is the most serious and widespread
problem in the world.”61 *

59. Brock (1961), p. 1.

60. Sebrell et al. (1961), p. 541.
61. Autret (1962).

160
9 International Efforts to Produce
High-Protein Supplements,
I955~J975

As we have seen, international committees of nutritional experts con¬

cluded in the period 1950-5 that substitutes for milk were urgently needed
to solve the world’s most critical nutrition problem - a shortage of protein.
They set out to stimulate and organize the production of new kinds of high-
protein materials that would help to fill the “protein gap.”
At their fifth session, in 1957, the FAO/WHO experts set criteria for these
materials. They had to be capable of being produced locally, to be affordable
by all sections of the population and to have a long storage life even under
tropical conditions. They also had to be acceptable, nontoxic and easily
included in ordinary diets. Finally, their protein must effectively supplement
existing diets.
Six protein sources were studied: fish, soybean, peanut, sesame, cotton¬
seed and coconut. With regard to the last four, attention was focused on
the low-fat press cakes produced by the vegetable-oil industry.
The initial step would be to produce a batch of each type of high-protein
product under carefully defined conditions and then to study its biological
value and safety with animals before testing its value for children and adults.
The next step would be to study the possibilities of manufacturing the
product in the area concerned. The program had already reached this point
with two of the products in question — fish flour and saridele (a soy extract).
The experts concluded that, although such foods could not permanently
solve the problem of meeting the protein requirements of vulnerable groups,
because in underdeveloped countries transportation, processing and distri¬
bution tended to make these products relatively costly, they could be ex¬
pected to make an important contribution. And although the ultimate aim
was to teach mothers to feed their children adequately by giving them
sufficient foods of the right kind, this could be effective only if the necessary
foods were available.1

1. Joint FAO/WHO Expert Committee on Nutrition (1958), pp. 2.2-3.

 Protein and Energy

The U.S. National Research Council had recently established the Com¬
mittee on Protein Malnutrition, which could submit applications for re¬
search money and, in turn, distribute it to particular projects, in
collaboration with WHO’s Protein Advisory Group (PAG).1 In 1961 the
PAG formally became a joint responsibility of FAO and UNICEF as well
as of WHO. The original members of the group were clinical nutritionists.
To this group were added food scientists, technologists associated with the
development and manufacture of protein foods and marketing experts.
Despite this, the FAO/WHO Expert Committee on Nutrition, meeting in
1966, said that “the marketing phase of the [protein-rich food] program,
including consumer studies and promotion campaigns, has not been given
the attention it requires.... Industrial production of protein concentrates
... has been slow and uncertain_Plans should be developed in co¬
operation with existing distribution organizations.”2 3
In the late 1960s the problem of the “world protein gap” came to the
attention of the UN headquarters staff in New York. The Advisory Com¬
mittee on the Application of Science and Technology to Development re¬
cruited a panel of experts to advise them as to what should be done about
the problem. Its report was accepted and published in 1968 as International
Action to Avert the Impending Protein Crisis.4 The report stated: “The
protein gap in the nutrition of the population of our planet is becoming a
most important scientific, technological and public health problem and a
national and international policy issue.” The panel recognized that diffi¬
culties had arisen with many of the schemes designed to attack the problem,
some of them stemming from neglect of human motivation and behavior,
others from a lack of basic training among the personnel assigned to them.5
The panel’s list of specific proposals included the developnj^nt of fish
protein concentrate and oilseeds, as in the 1957 proposals listed earlier, and
they added the following: the development of genetically improved plants;
greatly intensified research on “single-cell protein sources” (i.e., yeasts and
bacteria); and the use of synthetic amino acids to fortify and improve the
nutritive value of dietary proteins. They also recommended the development
of research and training centers and schemes.6
The estimated cost of implementing these proposals was $75 million, but
they were well received by the General Assembly and the UN Secretary
General, and another booklet elaborating the points in more detail was
published in 1971.7 It was agreed that the work of the PAG should be

2. Darby (1975), pp. xv-xvii.

3. Joint FAO/WHO Expert Committee on Nutrition (1967), pp. 60-1.
4. Advisory Committee on the Application of Science and Technology to Development (1968).
5. Ibid., pp. 40, 59.
6. Ibid., pp. 29—31.
7• P<tnel of Experts on the Protein Problem Confronting Developing Countries (1971).

162
 Efforts to Produce High-Protein Supplements

broadened; the World Bank became an additional sponsor, and sociologists

and economists were added to the group. In 1974 the name was changed
to the Protein—Calorie Advisory Group to the United Nations System.8 This
reflected a changing view of the nature of kwashiorkor, which will be dis¬
cussed in the following chapter, and in 1977 the work of the Group was
terminated.9
Many projects to produce high-protein materials that were started in
different parts of the world within the period 1956—77 were independent
of the PAG, but its meetings and quarterly bulletins did provide a forum
for discussing the problems and progress in this field.

Fish Protein Concentrate

The first project that FAO and UNICEF were working on at the time the
PAG was established was a “fish flour” plant in Chile. Later such material
came to be called “fish protein concentrate” (FPC). The attraction of this
project is easy to understand. The Chilean government was concerned about
the nutritional state of young children from the poorest segment of the
country. It was proving difficult and expensive to expand the dairy industry
significantly, and importing dry skim milk was an economic drain. Yet Chile
had a long coastline, and considerable quantities of fish were already being
caught for conversion to a dry powder (i.e., fish meal) that had proved, in
many parts of the world, to be a valuable source of supplementary protein
for chickens and pigs and could be produced more economically than milk
powders. Surely, with some small modifications, an upgraded flour could
be prepared for young children, who should have a higher priority than
animals for such a source of good-quality protein.10
The great bulk of the world’s fish meal in the 1950s was made from
clupeiform species such as herrings and anchovies. They are generally small,
less than 12 inches long when mature. They also live in “schools,” com¬
prising as many as 1 million or more fish, which keep tightly together.
Though often moving hundreds of miles over a season, they characteristi¬
cally stay close to shore. When a school moves into their area, fishermen
can catch enormous quantities relatively cheaply. Some are sold whole, while
others have their heads and guts removed and are either smoked or canned.
But when very large catches are landed, not all of the fish can be processed
before they deteriorate. Fish meal plants were therefore developed and used
to process large portions of the catch - particularly in Scandinavia, in South
Africa and on the coasts of Chile and Peru.

8. Anonymous (1974).
9. Schatan (1977).
10. Pariser, Wallerstein, Corkery and Brown (1978), PP- i44~7-

163
 Protein and Energy

Clupeiform fish have a relatively high content of fat, which is distributed

throughout their tissues. If the whole fish were dried, the product would
contain 30% fat or more. This would be unacceptable for animal feeding.
The high fat content would give a “fishy” flavor to the carcasses or to hens’
eggs. Also, the fat (or oil) would quickly oxidize when in co.ntact with air,
making the meal indigestible and possibly toxic. At the same time the heat
evolved during the oxidation could be sufficient to set a whole store of fish
meal on fire. \

The processor of clupeiform fish therefore removes most of the oil by

first steaming the fish to soften and break up the tissues, and then pressing
out the liquor from them. The hot liquor is then centrifuged, and the oil
can be sold separately. The aqueous layer can be returned to the “press
cake” (i.e., the solids), and the mix dried in a current of hot air and ground
to a powder. The resulting meal may have some 7% fat, which is acceptable
for animal feeding if relatively low levels of the meal are included in the
animals’ diets.
The fish flour first tested in Chile to determine whether it could be intro¬
duced in foods such as bread without consumers noticing or objecting had
come from South Africa. There, ordinary fish meal made from clupeiform
fish had been extracted with hot ethanol (ethyl alcohol) and the residual
solvent removed by heating in a vacuum oven. This reduced the fat content
to below 1%. It still had a slightly fishy flavor, but this wis undetectable
when added as 4% of the dry ingredients of brown bread. Tests with rats
showed no loss in the value of the protein and no evidence of toxicity,
though the process removed a large proportion of the water-soluble B vi¬
tamins from the meal. More than 800 tons of fish flour were incorporated
into brown bread in South Africa at government expense in the period from
1956 to 1959, but the project was stopped because it was realized that
those at risk from protein deficiency (i.e., preschool children of the poorest
nonwhites) did not consume a significant amount of bread.11
With the success of the small tasting trials of the South African product
in Chile, in which enriched bread was given to 140 children each school
day for 6 weeks, the UN agencies agreed to assist in setting up a fish flour
plant there.11 The goal was to produce an essentially fat-free flour with no
fishy flavor. The plant was set up to process hake, a lean white fish in low
demand. The fish were first dried in the usual way and then extracted, first
with hexane and then ethanol. The first trial runs in 1958 were unsuccessful,
but by mid-1961 1 ton of FPC had been produced. When further difficulties
were encountered with the hexane extraction, the Chilean government de¬
cided, in view of the continuing problems in developing an economical

11. National Nutrition Research Institute (1959),'pp. 59-61, 144-7.

12. Roels (1972), pp. 5-6.

164
 Efforts to Produce High-Protein Supplements

source of FPC, to drop the program and concentrate instead on increasing

the supply of milk powders for children.13
At this time there were plans to produce FPC in a number of countries.14
We will follow that of the Bureau of Commercial Fisheries in the United
States, whose history has already been analyzed in an important book.IS In
1961 John F. Kennedy, just elected as president of the United States, was
anxious to demonstrate the country’s technological abilities — mainly in the
space race, where the USSR had so far been leading - but also in other
ways. The Bureau put forward the idea of developing FPC as a new form
of “food from the sea,” which would help the U.S. fishing industry and also
provide technical expertise for the Third World to adopt in turn (Figure
9-1)-
Research was begun in the Bureau’s laboratory, adjoining the campus of
the University of Maryland. In 1963 an experimental model plant was built
that would process meal with no fishy flavor from a fish that again had
little fat. It was thought prudent to begin with hake, which would be easier
to process than the more abundant fatty fish. The procedure chosen consisted
of extracting the wet, chopped-up fish with hot isopropanol (isopropyl
alcohol), a chemical generally similar to ethyl alcohol but a better solvent
of fat. This was designed to remove fat and water at the same time, with
the solvent being recovered fairly cheaply by distillation. The principles of
this procedure had already been worked out in a Canadian government
laboratory.16
Under political pressure to produce rapid results, and before the problems
of economic, continuous extraction had been worked out, the Bureau agreed
to ask for funds to build a commercial plant as a pilot for private industry.
In April 1971 the plant, set up on the West Coast, was ready for operation
and 1,500 tons of hake were processed in the next few months. However,
the hake resource, which was thought to have been abundant, failed. If the
plant was to work with a reasonable output, it would therefore have to be
modified, at considerable expense, to operate with fatty fish. But the prices
of fuel and of all species of fish were rising rapidly, and Congress no longer
seemed willing to believe that FPC would be economically attractive. No
private company was interested in taking over the plant, and it was sold
for scrap. The detailed histories of FPC projects in other countries were
different, but the final outcome was the same.
The Bureau also had problems meeting the strict specifications of the
Food and Drug Administration (FDA) for a human food product. The first

13. Pariser, Wallerstein, Corkery and Brown (1978), pp. 148-51.

14. Rods (1972.), pp- 3-11; Pariser, Wallerstein, Corkery and Brown (1978), PP- H7-43,
161-213.
15. Pariser, Wallerstein, Corkery and Brown (1978), pp. 19-113-
16. Idler (1969)..
Protein and Energy

More people need more protein

FPC
FISH PROTEIN CONCENTRATE

ITS FOR PEOPLE

The amount of fish now caught
is sufficient to provide nearly
an ounce of fish daily to
everyone in the world; however, because of uneven distribution,
many people living inland get little or none. It is believed that the
fish catch could be increased threefold and thus make an impor¬
tant contribution to the protein shortage, but the problem jj to
preserve the fish in an economical manner so that it can be shipped,
stored, and put into other foods cheaply and safely.
Fish protein concentrate, known as FPC, can provide an answer.
It is made by removing the water and oil from ground fish, leaving
just the protein and the minerals in a concentrated form. FPC is a
light, neutral powder containing about 80 percent high quality
protein. It may be added to many foods without changing their
appearance or flavor, yet greatly increasing and improving their
protein value.

Figure 9.1. Material from a 1971 U.S. Department of Commerce booklet

entitled “FPC - Fish Protein Concentrate: It’s for People” (1971).

166
 Efforts to Produce High-Protein Supplements

such criterion was that FPC be free of disease-causing bacteria. This meant
that the processing plant had to be built to satisfy the hygiene rules laid
down in the United States for any human food product. Since farm animals
appear to be less sensitive to bacterial food poisoning, the same precautions
are not required for the production of animal feedstuffs and therefore of
fish meal. The second was the requirement that foods be free of “filth.” The
public would expect the FDA to ensure that, say, ground beef for a ham¬
burger was not contaminated with cow dung or with intestinal contents.
Yet when thousands of small clupeiform fish were being processed, as in
the South African procedure, the whole fish was used, including the guts
and gut contents. The cooking killed the bacteria present — and fresh sea
fish, in any case, did not carry bacteria that cause disease in humans — but
there remained the aesthetic concept of filth, and the dairy industry objected
that FPC represented unfair competition for milk powders, which had to
meet the most stringent hygienic standards. Finally, it was ruled that FPC
could be sold only in i-lb. packages, so that purchasers would be buying
a clearly labeled product. This eliminated the use of FPC as one ingredient
in proprietary mixed foods or formulas, where consumers might not realize
what they were getting.17
Another problem arose from the bone content of dried fish. The major
mineral in bone was calcium phosphate, and this was generally regarded as
a nutritional asset. However, fish bones also contained a certain amount of
fluoride. This was again desirable in the human diet, up to a point, because
it strengthened tooth enamel and increased resistance to decay. But at higher
levels it could cause mottling of the teeth and, at extremely high levels,
rigidity of the spine. The FDA required that the level of fluoride in FPC be
kept below 80 parts per million. This could be achieved only by removing
a proportion of the bones, either before processing or by sieving the partly
ground product. Either approach added to the cost of production.18
The Bureau had assumed that it would be politically impossible to export
to Third World countries, even as free gifts to the needy, material that had
been considered by another U.S. government agency to be unacceptable
human food. It was also believed that the product would have to be virtually
tasteless in order to be acceptable and that this would mean almost complete
extraction of its fat. Some people argued that this was unnecessary, and a
fishy flavor would, in some cultures anyway, be welcomed in an otherwise
bland diet. In any event, there has been no established use, or adoption, of
either type of material.
The projects did not fail because of the nutritional value of the protein

17. Pariser, Wallerstein, Corkery and Brown (1978), p. 87.

18. Ibid., pp. 83-5.
9
✓

167
 Protein and Energy

in FPCs. The few feeding trials with infants all gave good results, indicating
that the quality of the protein was close to that of milk.19 We will return
later to Pariser’s analysis of why these projects failed.

Oilseed Flours

The protein sources, other than fish, referred to in the FAO/WHO group’s
1957 program of protein foods to be studied as substitutes for milk were
all oilseed flours. Seeds are commonly classified as “oilseeds” if they contain
at least 15% of fat (or oil) on a dry weight basis. Vegetable “oils” are, in
chemical terms, fats, but the word “oil” is ordinarily used if they are liquid
at room temperature. The protein of oilseeds typically provides about 20%
of their calorie value. However, removal of the oil can leave materials
containing 50% protein or more, so that they are indeed protein concen¬
trates.
The pressing of seeds to obtain cooking oils has a long tradition. Olive
oil, in particular, was a staple of the Greco-Roman world. In the tropics,
peanuts and palm nuts, along with coconuts, have been important sources
of cooking oil. Vegetable oils have also been in greater demand in the past
60 years with the development of chemical processing methods, including
hydrogenation, that allow them to be used as substitutes for animal products
such as butter and lard (shortening). Even with high-pressilre equipment,
the residue is left with 5% oil. Solvent extraction is more efficient, and most
oilseed meals made in the technically advanced countries are the residues
from extraction with hexane. This solvent is relatively inert, except to com¬
bustion, and there is no evidence that it has any harmful effects. In poor
communities, where small operators have little capital, inefficient pressing
methods may still be used, leaving residues that rapidly go rancid and can
be used only as fertilizer or fuel.
By the 1950s the industrial processing of oilseeds was substantial, with
the meals being used as economical protein supplements for farm animals.
Even young chicks, which were thought of as delicate, made good use of
them, and since they had even higher requirements for essential amino acids
than infants, it seemed that the materials could be used as human foods
also. There actually was already experience in the production of food-grade
flours from some oilseeds, though they were used only at low levels, largely
to improve the physical texture of foods such as bread rather than as nu¬
tritional supplements.
In general, the protein fraction of oilseeds is about 85% digestible, as
compared with 90 to 95% for most animal foods. Their amino acid balance
is also somewhat inferior, but they can still be valuable supplements in
V
19. Donoso, Maccioni and Monckeberg (1967).

168
 Efforts to Produce High-Protein Supplements

human diets, with 3 parts of their protein (as a rough average) having the
value of 2 parts of milk protein/0 At the same time, oilseeds have certain
advantages over raw materials such as fish: They are less perishable, so that
processing can be spread out over the year, and they are grown all over the
Third World in one form or another in great abundance.
Most oilseed flours contain unwelcome constituents, but ways have been
developed to remove or inactivate most of them. One approach is to dissolve
most of the protein by stirring the flour in an alkaline solution and then
precipitating the protein from the extract by acidifying it. This kind of
“protein isolate” has been used as a component of sophisticated processed
foods in the United States, but the relatively high costs have made it irrel¬
evant to helping poor people on a large scale. We will consider the work
done with just three oilseeds, which have rather different characteristics.

Peanuts

The peanut, or groundnut, Arachis hypogaea, is a native of South America

but has been distributed throughout the tropical and subtropical world. It
was estimated that in 1955 some 13 million tons of the nuts were harvested.
It was also by far the major oilseed crop on the African continent, with
much of it exported to Europe/1
The main advantage of peanuts is that they are very palatable. Peanut
butter is made by shelling, roasting, skinning and then blending the nuts
into a paste. Peanuts have also traditionally been used in China to prepare
“milks,” the nuts being ground to a paste, stirred with water and filtered/1
The Indian government’s Central Food Technology Research Institute set
up a demonstration plant for producing such milk, but the idea was ap¬
parently not taken up commercially/3 The same institute prepared batches
of “Multi-Purpose Food” (MPF), a powder containing some 40% protein,
mostly from peanut flour tested for freedom from aflatoxins (discussed later)
and fortified with vitamins and minerals. A commercial food company also
prepared batches, but only when government aid schemes or relief orga¬
nizations purchased it; there was no regular commercial demand. In addi¬
tion, the institute developed weaning food mixes, Bal-Ahar and Bal-
Amul” with 22% protein, of which peanut flour was an important con¬
stituent. These sold well to people with middle incomes and were also bought
by organizations for emergency relief schemes, but production was not taken
up commercially/4

20. Orr and Adair (1967).

21. Ibid., p. 13; Rosen (i958)-
22. Rosen (1958), PP- 432--3-
23. Dean (1958), pp. 2.22-3.
24. Parpia (1969); Orr (1977), PP- 8“9-

169
 Protein and Energy

In Nigeria, two commercial companies attempted, with government en¬

couragement, to manufacture high-protein weaning foods with peanut flour
as the base. The first of these, “Amama,” was developed by the Glaxo
Company as a cheaper alternative to its ordinary baby foods based on dried
milk. In 1961, just as marketing was beginning, it was discovered that peanut
meals could be seriously contaminated with carcinogens. Further research
revealed that these “aflatoxins” came from fungi growing on improperly
dried nuts after harvesting and that their presence could be measured with
chemical tests.15 In the meantime, however, the Glaxo Company, unwilling
to be responsible for marketing a possibly dangerous product, closed its
operation. In 1963 the Nigerian government undertook to purchase a second
product, “Arlac,” for free distribution through institutions and clinics.
UNICEF also offered aid, on the understanding that the free distribution
would be only a preliminary to the development of a regular commercial
market for the product. The market did not materialize; UNICEF withdrew
its support in 1968 and production ceased.16

Soybean Flour

The soybean (or soya bean) is native to China and has been an important
food there for many centuries. From China it spread throughout East Asia
and was brought to the United States from Japan in the 1850s. After a very
slow start, it had become a major U.S. crop by the 1950s with an annual
harvest of 14 million tons, which was a little more than one-half of the
world harvest, and production has continued to increase. It is the world’s
major oilseed crop and well adapted to mechanized agriculture.17
The mature bean is almost spherical, small (less than 1 cm across), usually
yellow and very hard. It contains about 20% oil and 40% protein. Most
of the crop is processed by solvent extraction to yield oil that can be modified
for cooking uses. The residual flour, which was originally of lesser economic
value, has proved to be an excellent source of protein to supplement corn
in pig and poultry rations. In particular, its protein is richer in lysine (the
first-limiting amino acid in most grains) than is the protein of most other
oilseeds. The protein in soy is rather low in methionine but, fortunately,
synthetic DL-methionine is commercially available at an economic price and
can be blended into compound animal feeds when necessary.18
One problem is that the beans contain at least two antinutrients - a
trypsin inhibitor and a hemagglutinin, which depress the growth of exper¬

ts- Orr (1972), pp. 14, 16.

26. Ibid., p. 12.
27. Smith and Circle (1972).
28. Cravens and Sipos (1958).
 Efforts to Produce High-Protein Supplements

imental animals. Fortunately, they are inactivated by heat. It is therefore

standard practice for soy flour that is to be used for animal feeding to be
heat-treated after the solvent-extraction step. The product is referred to
commercially as “toasted soy flour” because it is somewhat browner than
untoasted flour, though the heating is carried out under moist conditions/9
The highly successful use of soy as a supplementary protein at an economic
price for pigs and poultry led to the recommendation, particularly by U.S.
experts, that it be adopted as a supplementary protein source for infant
feeding. Two of its proponents said, “Soybean protein outranks all of the
other proposed supplements in the worldwide nutrition programs.”30 How¬
ever, if peasant families in Africa, for example, with malnourished children
are encouraged to grow soybeans for themselves, there is a problem. The
mature seed, when cooked like other beans, by boiling in water, remains
firm and somewhat bitter. And since Africans probably do not know any
other way of preparing them, the crop will, almost certainly, be abandoned.
In Asia, where soy has been a traditional food, there are three main
approaches to preparing palatable foods from them. One approach is to
keep the seeds moist and in darkness for 4 days. They then sprout to a
length of about 3 inches, and the sprouts can be eaten either raw or cooked.31
They are similar to the mung bean sprouts more commonly sold in Western
countries, but are not, of course, suitable as ingredients of weaning foods.
The second traditional treatment of soy, particularly in Indonesia, is to
make tempeh by fermentation with a particular type of fungus, Rhizopus.
The beans are soaked until the hulls can be easily removed by hand; they
are then boiled for 30 minutes and spread out to dry. At this stage they are
inoculated with the fungus, using a little of a previous batch of material,
then patted into balls that can each be wrapped in a banana leaf. During
a further 24 hours at room temperature, the fungal mycelium binds the
beans together into a “cake” with a pleasant odor. It is then generally fried
in slices and eaten within a day.3" Some attempts were made to market a
dried form of tempeh in Africa, but the product was not popular.”
The third type of processing begins with the preparation of a “milk.”
The beans are soaked in water for 3 hours or more and then ground, with
the water, to a slurry, heated and held near the boiling point for about 20
minutes and finally filtered through cloth. The liquid, really a suspension,
has to be drunk within the same day if no special precautions are taken to
preserve it.34 This type of material can be fortified with vitamins and min-

29. Liener (1980), pp. 40-9, 91-4-

30. Smith and Circle (1972), P- I9-
31. Ibid., p. 380.
32. Ibid., p. 404; Dean (1958), P- zl8-
33. Orr and Adair (1967), p. 67.
34. Smith and Circle (i97z)> PP- 356~9-

171
 Protein and Energy

erals. Soy milk has been sold commercially in Hong Kong ever since 1945
as “Vitasoy,” being promoted as a thirst-quenching drink in competition
with colas.35 In this context, it is presumably consumed by older people
with money to spend rather than weanlings at risk for malnutrition.
A government project in Indonesia undertook to spray.-dry soy milk,
which was sold as “Saridele,” starting in 1957. The UN agencies gave
assistance on the understanding that 30% of the output would be distributed
free to charitable institutions. However, virtually all of the output was
apparently sold on the retail market. Despite its fairly high price, it was
popular among the fairly well off as a basis for making an after-dinner
drink. After 10 years of operation the equipment needed replacing. The UN
agencies were unwilling to provide further capital because the project did
not appear to be achieving its original aim, and production ceased.36
In Asia soy milk is traditionally used mainly in the first step of tofu
production. In this process calcium sulfate is added to the milk and a ge¬
latinous curd formed. This is pressed gently in a series of small filter boxes,
and each “cake” is then turned out. It still contains about 88% water, and
only 6% protein, so that it is not a protein concentrate and has only a short
“life” if not refrigerated. It has a bland flavor and can be either eaten as it
is or deep-fried. About 65 to 75% of the protein originally present in the
soybeans is recovered in tofu preparation.37 Tofu is said to be an important
protein source in a good proportion of East Asian households. However,
it has not been developed as a supplementary protein source for malnour¬
ished infants in other areas.
Another product, “Pronutro,” was specifically designed as a nutritional
supplement. It is based on full-fat soy flour that has been heat-treated and
given a special, additional treatment to remove the bitter flavor ^f soy. This
was “developed, produced and marketed by a private food firm in South
Africa without any governmental or international backing, as a high-protein
foodstuff from vegetable protein sources.”38 Dr. H. de Muelenaere, a nu¬
tritionist associated with the development, has said that, when production
began in 1962, the soybeans were imported, but the company indicated its
interest in buying locally grown crops. By 1969 farmers had responded to
the existence of this market by growing enough to meet the company’s needs
in full. The other major ingredient was locally grown corn. It was decided
that the product should be “instant,” that is, precooked so that it would
appeal to overworked mothers of lower-income groups. It was also decided
that it should be made attractive and first advertised to white groups. If

35. Orr (1972), p. 15.

36. Ibid., p. 13.
37. Orr and Adair (1967), pp. 67—8; Smith and Circle (1972), pp. 376—8.
38. De Muelenaere (1969), p. 266.
 Efforts to Produce High-Protein Supplements

advertising and sales were directed solely, or mainly, to the black, Bantu
people, they would suspect that it was a second-class food, lacking in pres¬
tige, even though clinical tests had demonstrated its value for malnourished
black infants.39
The product proved attractive and economical to a wide range of people.
After i year, it was being produced at a rate of 75 tons per month, and
production was thought to have reached nearly 600 tons per month by
1967, but to have fallen back to 300 tons by 1976. By 1969 it was estimated
that about one-half of that produced was going to Bantu customers. There
are anecdotes to the effect that white customers found it so economical that,
on the evidence of the advertising as to its high nutritional value, they began
buying it for their pets; this in turn led some Bantu people to reject it on
the grounds that they felt demeaned by the idea of sharing a food with dogs.
The price is now said to have risen to a point where it is not within the
reach of low-income groups, for whose children it was originally designed.
“Fortifex,” another product based on soy flour, was marketed in Brazil
in 1963. It was unpopular because of the “beany” flavor of the soy, and
production ceased in 1966.40 Many products based on soy have been dis¬
cussed by reviewers, but none, apart from Pronutro, appears to have been
commercially successful for any considerable period.41

Cottonseed Flour

Cotton plants are grown on a large scale because of the hairs (or lint)
attached to the outside of the seeds. These can be separated and spun to
form cotton thread and then woven into fabric. But for every ton of lint
produced there remains, as a by-product, about 1.7 tons of seed. The world¬
wide yield of cottonseed is in the range of 15 to 20 million tons per year,
and it contains about 18% oil and 20% crude protein. The United States
has the biggest crop, but many Third World countries with malnutrition
problems also grow sizable amounts.44
The seed, after the cotton lint has been removed, consists of about 45%
by weight of hull (i.e., outer cortex) and 55% of kernel (or “meats”). The
fibrous hull is of some value for ruminant animals, but is merely a source
of undesirable bulk for single-stomach animals, especially if the seed is to
be used in a weaning food for infants. With careful decortication there is
no more than 5% fiber left in the final product. The oil is the most valuable
part of the seed and, for technical reasons, is usually separated by screw-

39. Ibid., pp. 269-73; Orr (1972), p. 24.

40. Orr (1972), pp- i3—14-
41. Ibid., p. 66.
4Z. Altschul, Lyman and Thurber (1958), pp. 469-72.

173
 Protein and Energy

pressing. The residual oil content of the meal can be below 6%, and this
was made a requirement for cottonseed flours acceptable by PAG standards.
The crude protein level in such meal is approximately 50%.43
Cottonseed contains a pigment, gossypol, which has adverse effects on
animals. However, its molecule has reactive chemical groups (both aldehyde
and diphenyl groups). Gossypol is stable when in the intact seed, where it
is stored in pigment glands, but during screw-pressing the glands burst, and
at the elevated temperature in the press the gossypol reacts with the protein
and probably with other components in the seed. This “bound gossypol”
becomes indigestible and apparently harmless. Cottonseed varieties have
been bred without pigment glands, but these were not commercially grown
when high-protein products based on cottonseed flour were being developed
in the 1950s and 1960s. Amino acid analysis showed a somewhat lower
level of lysine in cottonseed meal than in soy, but a slightly higher methionine
level.44
In Central America during that period, cottonseed was the major oilseed,
and not fully utilized. The Institute for Nutrition in Central America and
Panama (INCAP) therefore set out to develop mixtures based on cottonseed
flour as the major protein source, with a high nutritive value. INCAP’s
mixture 9B consisted of 28 parts of ground corn, 28 of sorghum grain, 38
of cottonseed flour, 3 of dried yeast, 1 of calcium carbonate plus synthetic
vitamin A. This gave good results in preliminary laboratory tests with chicks
and rats. Then it was tested, in a cooked form, with young children re¬
covering in the hospital from malnutrition, and it performed nearly as well
as mixtures based on milk powder. Finally, it was tested for acceptability
among needy families in rural areas as a substitute for the cereals they were
accustomed to cooking into an atole, or thin gruel, with added^ flavorings,
for their young children.45
Given the satisfactory results of these tests, INCAP decided to encourage
commercial production of this mixture (and others of the same general type),
under license and supervision from INCAP, using the trade name “Inca-
parina” and with the price fixed as low as possible. It was retailed as 75-g
portions in plastic bags, each enough for three glasses of atole, and also
sold in bulk to charitable institutions. Commercial production of Inca-
parina began in Guatemala in 1961 and was still in production in 1980 at
the rate of 300 tons per month. In other Central American countries, pro¬
duction was started but ceased after a few years. There were various reasons
for this, including competition from material donated by international agen-

43. Ibid., pp. 484-505; Protein Advisory Group (1970).

44. Altschul, Lyman and Thurber (1958), pp. 489—92, 500.
45. Scrimshaw et al. (1961); INCAP (1962).

174
 Efforts to Produce High-Protein Supplements

cies.46 Incaparina was produced in Colombia for 12 years, from 1963 to

i975j by the Quaker Oats Company. In 1969 it appeared to be doing well
and was sold at a price (based on equivalent quantities of protein) roughly
one-eighth that of milk powder, but higher ingredient prices and increasing
competition from other weaning foods finally made its production uneco¬
nomical.47
There has been vigorous debate over whether Incaparina should be re¬
garded as a success or a failure. One paper pointed out that the Guatemalan
production in relation to a population of about 5.6 million corresponded
to only 1.2 lb. per person per year. There was also evidence that the poor
bought less than the better off. From survey data on the use of Incaparina,
it was estimated that 87% of the children got none, 5% consumed between
one-fifth and one-tenth of the recommended level and 6% consumed a little
more than one-third.48 The director of INCAP wrote in 1976 that recent
increases in the prices of ingredients had resulted in the price of Incaparina
being placed “a little beyond the purchasing power of the people who needed
it most — the underprivileged.”49
Scrimshaw, the director of INCAP when Incaparina was being developed,
has defended the record:

There was never any claim or expectation that a commercially produced and mar¬
keted vegetable mixture could “solve” the problem of protein-calorie malnutrition
in Guatemala or any other country. This could only be realized through an increase
in income among the groups in need, or by effective programs of subsidized distri¬
bution. There was, however, a need for a weaning food that could be recommended
for malnourished mothers and children in a country where milk was too costly and
in short supply... it would be the height of arrogance to deny developing countries
the benefits of a nutritionally comparable food at the lowest practical cost. The
objective, then, was to provide a beverage with the nutritional equivalent of milk
... for the benefit of a sector of the population with modest purchasing power,
leaving to other programs the problem of reaching that part of the population unable
to purchase weaning foods.so

Single-Cell Proteins

Single-cell proteins (SCPs) include protein concentrates prepared from any

type of single-celled micro-organism. There are several readily available

46. Scrimshaw (1980); Shaw (1969), pp. 32.1-2; Orr (1972-), PP- 7~n; Orr (*977), P■ 4-
47. Dimino (1969).
48. Wise (1980).
49. Icaza (1976), P- 37-
50. Scrimshaw (1980), p. 1.

175
 Protein and Energy

books on this interesting group of materials.51 One attraction of growing

micro-organisms on a large scale as a food supplement is that they can
reproduce so rapidly. One proponent of their use has compared the daily
production of 0.5 kg milk protein by a 500-kg cow to the 1,250 kg protein
that could (in theory) be produced by 24 hours’ fermentation of the same
weight of yeast.5*
Yeasts had been grown on carbohydrate wastes for many years, but
French work in the 1950s demonstrated that some strains of yeast could
also use hydrocarbons as their energy source, and “protein from petroleum”
became a catchy concept when there was a general fear of a world shortage
of protein from conventional sources.53 Production commenced, at least on
a pilot-plant scale, of a variety of organisms, including bacteria *and fungi
as well as yeasts, and in general the products gave good results in animal
feeding tests, with both rats and pigs. However, there were considerable
difficulties in producing materials that satisfied concerns about their safety
as human food ingredients.
The first problem was that, because micro-organisms reproduce so rapidly
a larger portion of their nitrogen is present in nucleic acids (i.e., their genetic
material) than is the case with conventional plant or animal foods. Two
constituents of the genetic material are purines, adenine and guanine, and
after the digestion and absorption of foods, these components are largely
oxidized to uric acid. In most mammals an enzyme then ^converts the uric
acid to another more soluble compound, allantoin, which is freely excreted
by the kidneys. However, the primates, including humans, do not have this
enzyme, and high concentrations of uric acid can crystallize out, as stones
both in the urinary system and in joints, causing the pains of gout.
There is considerable variability among individuals as to t^e quantity of
nucleic acids that they can tolerate, and safety limits have to be set for the
most sensitive. The general recommendation has been that SCPs should
contribute no more than 2 g nucleic acids to a day’s diet. Since the true
protein content of yeast, for example, may be no more than four times the
nucleic acid content, such a product cannot therefore be safely consumed
at a level contributing more than 8 g protein per day.54 Other yeasts have
a more favorable ratio, but some bacterial products have a ratio of little
more than 3:i.55 It is technically possible to reduce the nucleic acid content
of products by further processing, but of course this adds significantly to
the costs of production.

51. E.g., Mateles and Tannenbaum (1968); Tannenbaum and Wang (1975); Moo-Young and
Gregory (1986).
52. Champagnat, Vernet, Laine and Filosa (1963), p. 14.
53. Ibid., p. 13.
54. Moo-Young and Gregory (1986), pp. 109—11.
55. Mauron (1980-1), pp. 73-4.

176
 Efforts to Produce High-Protein Supplements

The second problem with a number of SCP products has been that, al¬
though they have passed exacting toxicological tests with a variety of animal
species, about 15 % of the human subjects receiving yeasts developed eczema
and other allergic reactions, and bacteria have caused vomiting and diar¬
rhea.56
A third concern has been that production facilities might be contaminated
by less desirable micro-organisms without their being detected. Precautions
undertaken to minimize this risk have included the sterilization of very large
air (or oxygen) intakes, which again has raised costs significantly.57
Because of these considerations, new yeast and bacterial products have
so far been used only as protein supplements for animal feeding, though
one fungal product has been marketed as a food.58

Lysine-Enriched Grain

The protein of grains was known to be first-limiting in lysine, as judged by

feeding experiments with rats and by comparison with whole-egg protein.
Plant breeders at Purdue University therefore set out to produce corn (maize)
of improved lysine content. In 1964 they published a preliminary description
of “opaque-2 corn” with a lysine content of approximately 4 g per 100 g
crude protein, as compared with 2.5 to 3 g per 100 g in ordinary corn, and
nearly a doubling of the tryptophan level.59 In 1966 a large conference was
held to present the first nutritional evaluations of the new corn. The intro¬
ductory speaker suggested that it would become a historic meeting and that
“within the next five years millions of undernourished people in Latin Amer¬
ica should find their diets improved markedly due to the availability of high-
lysine corn.”60 Feeding experiments with young animals confirmed its value,
but farmers were not interested in growing the new variety, since the yields
were relatively low and its soft kernels made it exceptionally subject to
insect attack and unsuitable for the usual milling equipment. A further 20
years of work resulted in “protein quality maize” without these disadvan¬
tages, though some agronomic questions remained, and it is still not in
general use.61
A further approach was to fortify ordinary staple grain foods with syn¬
thetic lysine. In 1956 a scientist working for the DuPont Chemical Company,
which had begun to manufacture lysine, urged that it would be a beneficial
additive for human diets, even in the United States, for the young and the

56. Waslien, Calloway and Margen (1969); Scrimshaw and Dillon (1979).
57. Moo-Young and Gregory (1986), pp. 97-102.
58. Ibid., pp. 19-26.
59. Mertz, Bates and Nelson (1964).
60. Liebenow (1966).
61. Mertz (1992).

177
 Protein and Energy

disadvantaged.61 In a number of trials, young Third World children re¬

covering from protein-calorie malnutrition in medical units retained more
nitrogen when their mainly cereal diets were supplemented with lysine.63
This result encouraged a U.S. government agency to finance large projects
in Guatemala, Tunisia and Thailand to study the value of fortifying corn,
wheat and rice, respectively, with lysine.64 In Thailand the rice was also
supplemented with threonine. In each country different groups received
either no supplementation, vitamin and mineral additions only, or amino
acid(s) plus vitamins and minerals. The areas studied were chosen because
grains formed a particularly large proportion of the local diets. After a great
deal of work, the project leaders summed up their results: “In Guatemala
there was very slight and inconclusive evidence of a decrease in morbidity
and mortality rates; the other studies showed no detectable benefits.”65 It
was concluded that low energy intake was the factor most responsible for
restricting children’s growth in these areas, along with possible marginal
deficiency of zinc and/or other trace minerals. Under such conditions a
response to lysine was not to be expected.66

Summing Up

From 1956 on, the Protein Advisory Group encouraged the development
of many new types of high-protein material. Funding was provided by
individual governments, and in 1970 the United Nations agreed on the need
for a $75 million program.
The first general goal was to upgrade the fish meals and oilseed meals
already used in animal feeding to a quality suitable for use with young
children. But this led to a variety of problems. One was thafUo produce
fish protein concentrate at a low cost required whole fish to be used, and
their intestinal contents, even after cooking, were technically unhygienic.
Another problem was the undesirably high content of fluorine in their bones.
The West African program to market an infant formula based on peanut
flour came to an end in 1961 because of the discovery that peanuts stored
in humid tropical conditions commonly became contaminated with a car¬
cinogenic fungus. Preparations based on “milks” made from soy were de¬
veloped in Asia, but were generally too expensive for adoption as a low-
cost weaning food. A precooked mix, Pronutro, with soy flour as its main
protein source was marketed by a commercial company in South Africa. It

62. Flodin (1956).

63. Austin (1979), pp. 269-81.
64. Ibid., pp. 131—268.
65. Kennedy, El Lozy and Gershoff (1979), pp. 27-8.
66. Vaghefi, Makdani and Mickelsen (1974).

178
 Efforts to Produce High-Protein Supplements

proved palatable and sales reached 600 tons per month, but increases in
the prices of raw materials then put it out of reach of those most in need.
The Institute for Nutrition in Central America and Panama developed
mixes with cottonseed flour as the high-protein component. These were
cooked and used as gruels for young children. They were marketed in several
Central American countries but, again, only a small proportion was con¬
sumed by the poorest groups.
There were many attempts to produce single-cell proteins (yeasts, fungi
or bacteria) as a supplementary food for children, but problems of quality
control and adverse reactions to them among some people prevented pro¬
duction and commercial marketing on a significant scale.
These and other problems with high-lysine grains were great disappoint¬
ments to those involved. Even the costs of simple packaging and transport
could put otherwise economic materials beyond the purchasing power of
villagers. It had been the original policy to have everything produced in the
area of need. Yet to quote someone concerned with the U.S. fish protein
project: “FPC represented a technological application conceived in the in¬
dustrialized world in an environment and time period that were relatively
free from constraints on capital, resources and expertise. This product was
to be applied in the developing world where all of these constraints existed
in abundance.”67
With the altered interpretation of the malnutrition problem that will be
described in the following chapter, most of these developments lost their
funding, so that we do not know what they might, or might not, have
achieved.

67. Pariser, Wallerstein, Corkery and Brown (1978), p. 231.

179
10 Reappraisals of the Third World
Problem, 1955—199°

In chapter 9 we saw some of the practical problems that arose in trying

to provide additional protein sources for people in the poorer countries of
the world. In the same period that these attempts were being made, there
was a progressive reassessment of the human requirements for protein and
other nutrients that was to alter greatly the general view of the “world
protein crisis” and the cause of the major malnutrition problems among
young children in the Third World. This forms the subject of the present
chapter, which picks up where we left off at the end of Chapter 8.
i
v

Bulk and Limiting Calories

From 1955 on, there was an increasing interest in the calorie component
of infant malnutrition, in kwashiorkor as well as marasmus. Dr. Hebe
Welbourn, a medical officer with a long experience with child ^elfare clinics
in Uganda, wrote that the eating behavior of the children she saw had also
been well described by Williams for West African children: “Each one
pinches off a small piece [of maize porridge] from the lump. He gently
shapes it with his fingers, dips it in the soup, and puts it in his mouth. He
delicately licks his fingers and repeats the operation. The whole process is
deliberate and mature; there is no scramble or hurry. But their appetites
are generally poor.”1 Welbourn, returning to her own experience, wrote:

One of the most noticeable features of all cases of kwashiorkor was anorexia....
The daily caloric intake recommended by the League of Nations (1936) for children
aged 2-3 years was 1,000 kcalories. Therefore most of the children, but especially
those of peasant families [for which we recorded a mean intake of 762 kcal,] were
taking [in] insufficient [amounts]. This was, no doubt, partly due to their poor
appetites, and partly due to having only two main meals daily. A child’s stomach
is small, and his needs are relatively large, while African food is very bulky. He

1. Williams (1938), pp. 97-102.

180
 Reappraisals of the Third World Problem

cannot pack in enough of this bulky food in two meals to supply his caloric needs
for the day.1

In her summary, she stated:

It has been shown that the diets of Baganda children are inadequate from the time
that their mothers’ milk alone is no longer sufficient for their growing needs. The
deficiency in diet which is maximal during the weaning period is associated with a
profound disturbance of growth and the appearance of signs of kwashiorkor....
Severe kwashiorkor, as seen in hospital, would appear to be the end result of a
prolonged period of feeding on an unbalanced diet which is mainly of carbohydrate
and is deficient in protein. The protein deficiency is usually aggravated by the oc¬
currence of infections, particularly those of the upper respiratory tract, which are
often followed by prolonged anorexia.3

In 1953 Dr. Katerina Rhodes had presented data at the Macy Founda¬
tion’s conference in Jamaica indicating that the kwashiorkor victims she
had studied there had had a low calorie intake while the disease was de¬
veloping.4 This was contrary to the previous assumption that such “sugar
babies” consumed ample or even excessive calories, and her data were
closely questioned. However, in 1957 she published them in full and con¬
cluded that, although the carbohydrate intake was high in relatk n to pro¬
tein, there was still an absolute calorie deficit. She also referred to similar
data from India, Indonesia and South Africa.5
In 1958 a group working in South Africa described the results of a trial
in which 60 kwashiorkor patients were treated with fresh skim milk sup¬
plemented with 5% of either carbohydrate (dextrimaltose) or protein (ca¬
sein) and, for a third treatment, dried skim milk alone (15 g per 100 g
water). The recovery rate and the rise in serum albumin levels were similar
in all the groups. It was concluded that expensive protein supplements to
skim milk were not necessary, since the addition of carbohydrate as a source
of additional calories was equally effective.6
In the following year, D. B. Jelliffe, who had worked in a number of
Third World countries with kwashiorkor and marasmus patients, wrote an
extensive review, introducing the term “protein-calorie malnutrition” in its
title.7 This quickly came into general use, with “PCM” as the accepted
abbreviation after the FAO/WHO Expert Committee adopted it at its meet¬
ing in 1961. The committee report stated:

1 z. Welbourn (1955), p. 105.

3. Ibid., p. 169.
4. Waterlow (1955), pp. 43-5.
5. Rhodes (1957), PP- 143-50, 166-7.
6. Pretorius and Smit (1958).
7. Jelliffe (1959).

181
 Protein and Energy

In developing countries a high percentage of children from lower-income families

may show retardation of growth and development due to inadequate feeding. The
prevalence of protein-calorie-deficiency diseases is increasing in some countries as
a consequence of the rapid spread of industrialization and urbanization and the
resulting adverse changes in food habits. One common result is the premature
termination of breast-feeding and the use of overdiluted cow’s milk, thin gruels and
cooking water from cereals as substitutes.... As the widespread use of the term
marasmic kwashiorkor suggests, kwashiorkor may be superimposed on any degree
of marasmus. The high prevalence of deficiency-disease syndromes makes it urgent
that attention be directed to practical methods for their prevention. Kwashiorkor
has tended to engage the exclusive attention of many workers. The attention of these
investigators and of those responsible for preventive and corrective programmes
should be directed, without decreasing the interest in kwashiorkor, to all‘aspects of
the problem of protein-calorie-deficiency disease.8

Recommended Protein Levels for Infants

In 1958 the U.S. Food and Nutrition Board issued its regular four-yearly
revision of recommended dietary allowances (RDA). For the first time, there
was a blank space in its tables - the RDA for protein among infants in the
first year of life. An asterisk indicated that breast feeding was recommended
for this period and that it provided no more than z g protein per kilogram
body weight daily. The majority of formulas had a higher level of protein,
which provided approximately 3.5 g/kg, and these, too, had proved to be
successful in practice. Opinion had been divided as to what value should
be listed in the table.9 Reviewers in the following year set out the points of
debate in more detail. They argued that the RDAs were designed for their
own country (the United States), which had an abundance offood. It had
been stated that the intention of the RDAs was to provide “ample allow¬
ances” and that “they may be more generous than would be practical for
feeding large groups under conditions of limited food supply or economic
stringency.”10 Moreover, it could not be assumed that human infants were
able to use cow’s milk protein as efficiently as breast milk protein. “The
protein of unheated cow’s milk was less well digested because of the hard
curd formed in the infant stomach, but there was not the same problem
with heated or dried milk preparations.”11
Waterlow and his colleagues in Jamaica reported, in 1961, the results of
a series of balance studies in children (iz months old on average) who were
being treated for kwashiorkor. Surprisingly in view of the pancreatic damage

8. Joint FAO/WHO Expert Committee on Nutrition (1962), pp. 23—4.

9. Food and Nutrition Board (1958).
10. Gordon and Ganzon (1959), p. 503.
11. Ibid., p. 508.
 Reappraisals of the Third World Problem

associated with the disease, the cow’s milk protein that the children received
was digested with 90% efficiency. The secretion of protein-digesting en¬
zymes into the gut was therefore adequate, though it may have had less
than the usual margin of safety. Also, they found that even 2 g protein per
kilogram body weight per day supported a rate of nitrogen retention three
times that of healthy babies. The “catch-up” phenomenon was well known,
but it seemed reasonable to conclude that, if this was possible on 2 g cow’s
milk protein per kilogram, then the normal rate of growth should easily be
achieved with the proposed standard of 1.5 g/kg for healthy babies.11 Wa-
terlow also found that the rate of recovery of babies depended very much
on their energy intakes. He obtained improved responses by giving 145 kcal/
kg daily in mixtures with 10.5 PCals%. He suggested that, in the past, the
rate of recovery from kwashiorkor had most commonly been limited by the
number of calories provided rather than the amount of protein.13

Supplementary Calories Needed

In 1962 B. M. Nicol, who had become Autret’s deputy at FAO, reviewed

“the utilization of protein-rich foods in the prevention of protein-calorie
deficiency diseases.” He wrote:
Probably the commonest cause of protein-calorie malnutrition is lack of food, which
may result from... an overall shortage of supplies affecting the whole community,
but more often results from the parents’ failure to realize that nutritional require¬
ments of young children are relatively much greater than their own. [Let us take as
an example]... the diet of a poor Indonesian community. It contained rice, a little
fish, sweet potatoes, vegetable oil, fruit, green vegetables and seasoning. This diet
provides adequate energy and protein if enough of it is available for consumption.
The amounts actually fed to weaned infants and pre-school children only provide
two-thirds of their energy and protein needs. Small amounts of dried skimmed milk
or full-fat soya flour correct the protein deficiency but do not supply sufficient calories
to bridge the gap between intake and requirements which must be achieved for
optimal protein utilization. For this purpose the supplement should also provide
concentrated sources of energy such as sugar or vegetable oil. The diets eaten in
Southern Nigeria are very bulky, being composed of starchy roots, very small
amounts of meat and pulses, and reasonable quantities of green vegetables. These
staple foods are freely available but even adults cannot consume sufficient of this
diet to meet their needs. Pre-school children can eat only enough to provide 65%
and 70% of their requirements for protein and calories respectively. In view of its
physical bulk, supplementation of such a diet is undesirable. The replacement of
roots by a good quality cereal such as rice, to provide a more concentrated source
of protein and calories, is necessary.14

12.. Waterlow and Wills (i960), pp. 194, 196.

13. Waterlow (1961), pp. 21-2.
14. Nicol (1962), pp. 152-5.

183
 4

Protein and Energy

It is interesting that this view was expressed by a senior FAO official during
the period when the PAG was accelerating its work on new types of protein
concentrates.
In 1963 Graham and his colleagues, working in Peru, said that use of the
term “protein-calorie malnutrition” for the whole spectrum of cases from
kwashiorkor to marasmus was resulting in unsuitable treatment of the latter
condition, which was much the commoner. Children with marasmus had
a serious caloric deficit and “failure tp appreciate [this] leads to a waste of
expensive protein and to continued failure in treatment.”15 In a second paper
they reported that 2.0 g protein per kilogram body weight daily was suf¬
ficient for hospitalized malnourished children but that daily caloric intakes
as high as 175 kcal/kg were highly beneficial to rapid recovery.16*
In 1965 FAO/WHO published its second report entitled “Protein Re¬
quirements.” The difference between the approach used by the Expert Group
on this occasion and that described in the 1957 report will be considered
later in a comparison of all the reports on the subject. The main point to
be noted here is that the estimates of requirement were much reduced. For
example, the need of a 1-year-old child for protein of the quality of breast
milk was estimated to be 1.1 to 1.2 g/kg per day, as contrasted with 2.0 g/
kg per day specified in the previous report.17
In 1966 Chan and Waterlow reported that 1.25 g cow’s milk protein per
kilogram received daily seemed to give optimum growth in children who
had almost recovered from PCM when they were receiving a total of 120
total kcal/kg daily. The balance of the calories came from sucrose, corn
starch and peanut oil.18 This mixture had a protein concentration of only
4.2 PCals% (i.e., lower than that of breast milk).
McLaren wrote that, though the incidence of marasmus w^s increasing
at a faster rate than that of kwashiorkor and was a more important type
of PCM, it was neglected.19 “Millions of dollars and years of effort that
have gone into developing these [high-protein] foods would have been better
spent on efforts to preserve the practice of breast feeding... being aban¬
doned everywhere.”i0 In 1967 Jelliffe wrote that, because tubers and plan¬
tains were so bulky owing to their water and fiber content, they needed
partial replacement with “compact calories” and that red palm oil, sugar
and avocados, for example, formed a good mashable source of them.zl
It would be wrong to give the impression that the papers just quoted were

15. Graham and Morales (1963), p. 486.

16. Graham, Cordano and Baertl (1963).
17. Joint FAO/WHO Expert Group (1965).
18. Chan and Waterlow (1966).
19. McLaren (1966).
20. McLaren (1967).
21. Jelliffe (1967).

184
 Reappraisals of the Third World Problem

typical of the period 1955—67. They were not. A much larger number of
papers dealt almost exclusively with the evaluation of different kinds of
protein mixtures for children and gave little or no attention to the question
of the caloric adequacy of the whole diets in which these high-protein foods
might play a part. Nevertheless, it is interesting that experienced people
were already trying to draw attention to the energy aspect at a time when
the UN system seemed to be entirely concerned with the “world protein
crisis.” The UN paper “International Action to Avert the Impending Protein
Crisis” was approved in 1968, and governments were asked to report what
actions they were taking on the “protein food problem.”11

Marasmus and Kwashiorkor

The seventh meeting of the FAO/WHO Expert Committee on Nutrition

accepted, with only a few minor reservations, the 1965 report on protein
requirements, but did not discuss the implications of the large reduction in
the estimates. In connection with reports of the increasing incidence of
marasmus as a result of urbanization, they commented that the PAG’s
protein-rich food program was “designed for the prevention of any form
of protein-calorie deficiency.”13 This statement may have been made in
direct response to the comments of Graham and others (quoted earlier),
who were of exactly the opposite opinion.
Some further differences of opinion emerged at a conference organized
by McCance and Widdowson in England in 1967. McCance, who had by
then spent some years at the Medical Research Council’s unit in Uganda,
opened the discussion by saying:

The pioneer workers were right after all, and there really are two quite separate
syndromes... marasmus, which we can describe as an insufficiency of all food,
however good, and kwashiorkor, which is essentially a deficiency of protein in a
diet potentially high in calories.... Both prevent growth, marasmus for obvious
reasons which have been recognized for centuries, and kwashiorkor because protein
deficiency produces anorexia and ultimately destroys all desire to eat.

He also argued that “protein-calorie deficiencies” was a better term than

“protein-calorie malnutrition,” which implied a single syndrome.14 Hansen,
in contrast, speaking from his experience in South Africa, said that, in every
type of PCM,

a deficiency of protein is the overriding problem, which may or may not be com¬
plicated by a calorie deficiency.... From the clinical and public health point of view

22. Kertesz (1968), p. 72.

23. Joint FAO/WHO Expert Committee on Nutrition (1967), pp. 8-10, 60.
24. McCance (1968), pp. 1-2.

185
 Protein and Energy

the similarities [between kwashiorkor and marasmus] are far more telling than the
differences in detail_Both are associated with a protein-deficient diet_Individ¬
ual children can drift from one syndrome to another/5

At the same meeting Gopalan described the studies carried out by workers
in India and their surprise at being unable to find any difference between
the diets of children who developed kwashiorkor and of those who devel¬
oped marasmus. There was no evidence of any of the former having been
“force-fed” starchy foods. The average daily diet of these children contrib¬
uted 1.8 g protein and 85 kcal/kg body wt. Even if the protein mixture,
being mostly vegetable, had a value only 60% that of “reference protein,”
this would still be equivalent to 1.0 g reference protein per kilogram and
so would apparently meet the requirement, whereas the caloric intake was
clearly inadequate. He hypothesized that marasmus represented an extreme
form of adaptation to such a protein—calorie deficiency, whereas kwashior¬
kor represented a dysadaptation — perhaps as a result of some further stress
such as infection. Possibly the different syndromes involved differences in
endocrine response to the stresses on the children/6
McLaren drew attention to PCM being so often complicated by vitamin
A deficiency, which had not received proper attention.17 Other participants
referred to the potassium and magnesium status of kwashiorkor patients as
a cause of concern and to the need for appropriate supplements to minimize
mortality in the most seriously ill.18 These comments emphasized the danger
of considering only one or two nutrients at a time.
From this point on, the relevant papers and discussions centered around
three themes, which we will follow separately for the remainder of this
chapter:
*
1. the idea that kwashiorkor represented a dysadaptation,
z. successive estimates of human requirements for protein and
3. the controversy as to whether it was primarily protein deficiency that caused
protein—calorie malnutrition.

Kwashiorkor as the Result of Dysadaptation

Marasmus, sometimes considered to be simple “wasting,” is not usually

accompanied by a serious fall in the level of serum albumin. Nor does the
liver become fatty. In kwashiorkor both of these changes occur, even though
there is less wasting than in marasmus. It was difficult to understand why,

2.5. Hansen (1968), p. 46.

26. Gopalan (1968), pp. 52-3, 57-8; Jaya Rao, Srikantia and Gopalan (1968).
27. Gopalan (1967); McLaren (1968).
28. Wharton, Howells and McCance (1967); Wharton (1968), pp. 150-1; Garrow, Smith
and Ward (1968), pp. 61-87; Whitehead (1969), p. 41.

186
 Reappraisals of the Third World Problem

in marasmus, muscle protein was catabolized to provide the liver with the
material needed to synthesize the different serum proteins so crucial for
preventing edema, while in kwashiorkor this did not happen, at any rate
to the same extent.
As discussed by Whitehead and Alleyne in 1972, on the “classical” theory
that kwashiorkor developed when carbohydrate intake was high or normal,
the endocrine response (e.g., raised serum insulin levels) would inhibit mus¬
cle protein breakdown, while the virtual starvation of marasmus (with e.g.,
lower levels of insulin) would encourage such breakdown and so add to the
bloodstream a complete mixture of free amino acids, which could be used
by the liver for resynthesis of serum proteins.19 It was accepted that, once
kwashiorkor had developed, one of the effects was a loss of appetite, but
by this time the damage had presumably been done.
Cortisol (hydrocortisone), an important hormone secreted by the adrenal
gland, has an activity generally opposite to that of insulin. In particular, it
increases protein breakdown in peripheral tissues, but stimulates protein
synthesis in the liver. Stress stimulates its secretion into the bloodstream.
In 1973 Whitehead’s group reported results from a detailed study of the
levels of cortisol and other hormones in children developing PCM in Uganda,
as well as a review of the literature. The average serum cortisol levels of
children with severe kwashiorkor were raised, but there were some indi¬
vidual children with normal values. The values showed a close correlation
with the severity of the children’s recent infections. Also, there was a similar
average elevation in cases of marasmus. Values were normal among children
in the early stages of kwashiorkor, when serum albumin values were only
slightly reduced. Analyses for serum insulin levels after fasting overnight
indicated that values were raised in the early stages and then, with fully
developed kwashiorkor, fell to subnormal levels. The decreases again seemed
to coincide with bouts of infection. Finally, a rise in the level of growth
hormone in the serum was found, but only in combination with severe
reduction in serum albumin levels, that is, in the most serious cases. To sum
up, it appeared that at first the only abnormality was a raised insulin level
but that, later, this level fell, and cortisol and growth hormone levels were
raised. The increase in growth hormone concentration was seen in kwa¬
shiorkor but not at all in marasmus and was thought to be particularly
significant.30
It had been the presumption of the group in Uganda that dietary differ¬
ences were primarily responsible for the development of two different dis¬
eases. However, Jaya Rao from India presented another hypothesis based
on the assumption that there were no significant differences in diet between

29. Whitehead and Alleyne (1972.), pp. 73, 76-7.

30. Lunn, Whitehead, Hay and Baker (1973).

187
 4

Protein and Energy

those developing kwashiorkor and those suffering from marasmus. Most

children were able to respond to the stress of a diet extremely low in calories
and low (or marginal) in protein, to adapt and gradually become marasmic.
They did this, primarily, by greatly raising the level of circulating cortisol
- which stimulated the movement of amino acids from muscle to liver.
Children who did not react in this way to produce a sufficient increase in
cortisol suffered the effects of a liver starved of the limiting amino acids
needed to synthesize serum proteins.31 Presumably the failure of the adrenal
cortex to respond was due either to some additional stress, such as infection,
or to an inherent, genetic characteristic of the individual.
This hypothesis seems not to have been generally accepted. Two replies
appeared in the Lancet, making a number of points. First, it had been found
elsewhere that cortisol levels were raised only in cases of PCM where there
was infection and that the rise then seemed independent of whether the
basic condition was one of kwashiorkor or marasmus. Second, the pro¬
portional incidence of the two diseases varied from one area to another,
and this could be related to the prevalent dietary pattern. Finally, the hy¬
pothesis ignored the finding that a consistent, kwashiorkor-like state could
be produced in animals purely by dietary manipulation.31
Because, once a disease is established, there may be complicating effects
of anorexia, Whitehead and co-workers carried out a long-term study of
children in their homes for the first 3 years of their life in rural communities
in both the Gambia (West Africa) and Uganda. It appeared that children
in Uganda, where kwashiorkor was more common than marasmus, were
better off in the first year of life, then suffered a greater shortage of food
in the second. However, they did not appear to react to this, as would be
expected, with lower insulin and higher cortisol levels in their blood. In
fact, it was the children in the Gambia, where marasmus was more common,
who showed this response, perhaps because of the greater incidence of
infections, mostly gastrointestinal. It is interesting that, although the mean
level of protein in the diets of the two groups of children was very similar,
at about 8.4% total calories, there was greater variability in the level from
one child to another in Uganda. And among the individuals whose food
intake was studied, it was those with the lowest protein intake who seemed
most at risk from malnutrition.33 The picture was complicated and provided
some support both for those who argued that kwashiorkor represented a
failure of endocrine adaptation (since cortisol levels were lower in Uganda)
and for the opposing view that it was due to protein deficiency. Most
surprisingly, the data did not fit what seemed to be the noncontroversial

31. Jaya Rao (1974).

32.. Rosen, Buchanan and Hansen (1974); Waterlow (1974).
33. Whitehead, Coward, Lunn and Rutishauser (1977).

188
 Reappraisals of the Third World Problem

Table io.i. Calculations leading to estimates for safe protein allowances

in successive FAO/WHO reports

Mean daily Approximate

requirement protein Safe protein
Allowance
for good score for allowance on
for
protein poorest poorest diet
individual
(g/kg body wt) national diet (g/kg body wt)
Year of variability
report A B (%) A B A B

1957 0.35 1.30 + 50 60 60 0.88 3.25

1965 0.59 0.92 + 20 60 60 1.18 1.84
1973 0.44 0.98 + 30 70 70 0.82 1.82
1985 0.60 1.08 + 25* 85fe 51fc 0.88 2.70
+ 27.5C

Note: Columns labeled A contain values calculated for young adult males; columns
labeled B contain those for 1- to 2-year-old children.
*For young adult males.
fcThe actual scores, based on amino acid analyses, were 100 and 58, respectively,
but an additional 85% digestibility factor was applied to the poorest diets with r
high proportion of vegetable protein.
cFor 1- to 2-year-old children.

view that marasmus could be simply explained by inadequate food intake,

since the Gambian children ate more in their second year in relation to their
size.

What Are Normal Protein Requirements?

In this section, we will review the series of four FAO/WHO reports on this
subject published in 1957, 1965, 1973 and 1985, and refer to other reports
only when they use different means to arrive at their estimates. We will also
concentrate on just two classes of subjects - 1-year-old children and young
adult males - the former because they are the ones at risk of developing
kwashiorkor in the Third World, and the latter because they are the group
that has been most used for direct experimentation. All the reports followed
the same general procedure, that is to say, they first considered the require¬
ment for proteins of excellent quality (i.e., those of milk or eggs) and then
tried to assess how the requirement should be adjusted upward for other
protein sources. The results are summarized in Table 10.1.
Because so much of the work in nutrition in the previous 40 years had
been concerned with vitamins, and so little with protein, the authors of the
first (1957) report had to go back to Sherman’s paper, published in 1920,

189
 Protein and Energy

for their estimate of adults’ daily requirements.34 They assumed a value of

0.35 g/kg body wt for good-quality protein (that of eggs or milk). Their
second assumption was that young suckling infants needed all of the protein
(2 g/kg) they received from breast milk.35 Scrimshaw has referred to the
debates within the committee as to the rate at which the requirement declined
as a child aged, but the final estimate for a 1-year-old child was that, on
average, he or she required 1.3 g/kg.36
Next came the problem of converting these values to safe practical al¬
lowances. As seen in Table 10.1, the decision was made first to increase the
values by 50% to allow for individual variability — that is, for some children
needing more than the average. Then there was the problem of allowing
for the lower quality of protein in practical diets. This was based on’chemical
scoring similar to that developed by Mitchell and others, as discussed in
Chapter 7. From an examination of dietary patterns in different countries
it appeared that the worst combination likely to be encountered had a value
approximately 60% that of ideal protein. From the inverse of this percentage
it followed that 1.67 g of such protein was equivalent to 1 g of the ideal.
Therefore, as seen in the table, a 1-year-old’s safe daily allowance in such
circumstances would be 1.3 x 1.5 x 1.67 = 3.25 g/kg body wt.
The 1965 report followed a different approach. It reviewed the data
available as to how much nitrogen per day was lost by human subjects
when they were on a diet free of protein (and nitrogen), through feces, urine,
skin, sweat and so on. It was argued that the “perfect” dietary protein,
providing the same quantity of nitrogen, would exactly replace these “oblig¬
atory” losses and bring the subject back into balance. Protein needed for
body growth, growth of a fetus or milk secretion would also be obtained
with 100% efficiency from such a perfect dietary protein.37 FJpwever, the
estimates were increased by 10% to make allowance for the effect that
stresses such as minor infections and trauma were believed to have on protein
breakdown and losses from the body.38 The requirements of ideal protein,
before allowance for individual variability, came out to 0.92 g/kg for 1-
year-olds and 0.59 g/kg for adult men. Practical recommendations were
made by multiplying by a small variability factor and a “scoring” factor.
The next study, published in 1973, considered energy as well as protein.
Since the preceding study, there had been a number of tests of its assumption
that egg protein could compensate for obligatory losses with virtually 100%
efficiency. In practice, with young adults, the finding was that it could not.
The recalculated obligatory nitrogen losses (incorporating some new data)

34. Sherman (1920); Irwin and Hegsted (1971a).

35. FAO Committee on Protein Requirements (1957), p. 15.
36. Scrimshaw (1976), p. 138.
37. Joint FAO/WHO Expert Group (1965), pp. 10—16.
38. Ibid., pp. 17-18.

190
 Reappraisals of the Third World Problem

were lower at 54 mg/kg per day, equivalent to 338 mg crude protein, but
463 mg egg protein had been found necessary, on average, to achieve bal¬
ance. In addition, this was very similar to the quantities required to obtain
nitrogen balance with soy flour or any of a variety of mixtures of plant and
animal proteins.39
Two conclusions were drawn. One was that supplying dietary protein
increased the rate of amino acid catabolism, so that obligatory losses on a
nitrogen-free diet were irrelevant to the estimation of protein requirements.
The second conclusion was that, at least for adults, the levels of essential
amino acids in dietary proteins could be considerably lower than those
present in milk and eggs without their value being reduced. There were
fewer new data for young children, but it was judged that, as with adults,
the actual requirements (even for milk protein) were approximately 30%
greater than the amount calculated from obligatory losses, plus the protein
content of growth. However, this change was again largely offset by lower
estimates for obligatory losses.40
The committee considered that, where there were no reliable values for
either the “score” or the biological value of the proteins in particular diets,
it would be reasonable to use a “relative protein value” (i.e., relative to a
value of 100 for milk protein) of 80% for diets in more affluent countries,
where a good proportion of the protein came from animal sources, and of
70% for most of the poorer countries. Where 70% was the relevant value,
the safe daily levels of protein intake became 1.82 g/kg for 1-year-olds and
0.82 g/kg for young adults.41 Since the daily energy requirements of the
children and of moderately active young adults were estimated to be 105
and 4 6 kcal/kg, respectively, the safe protein intakes corresponded in each
case to approximately 7 PCals%, respectively. Examples were also given as
to how a national balance sheet of protein and energy requirements could
be drawn up.
In 1975 another group of scientists was brought together to comment on
the 1973 report. Its first criticism was that allowances should have been
made for the protein in vegetable foods having a lower digestibility than
milk protein; correcting just for differences in “score” based on amino acid
composition was inadequate.4* Also, in practice, habitual energy intakes
were for many people less than the assumed energy requirements. For such
people the level of dietary protein, expressed as PCals%, would have to be
higher than one would calculate from the data given in the report. In ad¬
dition, there was doubt as to whether the “safe” level of intake of egg

39. Joint FAO/WHO Ad Hoc Expert Committee (1973). P- 49-

40. Ibid., p. 52..
41. Ibid., p. 74.
4Z. Joint FAO/WHO Informal Gathering of Experts (1975)-
 Protein and Energy

protein for adults — that is, 40 g (0.44 x 1.30 x 70) for a 70-kg man —
was adequate over a long period. Finally, the group argued that calculation
of the adequacy of national food supplies by adding up the requirements
of each group within a country could be misleading. Many individuals would
choose a dietary pattern with more than the “safe level” of protein. But if
the total supply only balanced the total need, it meant that others were
inevitably consuming less.
In the same year, Scrimshaw, in a published lecture, discussed the history
of committee recommendations on protein requirements. Fie said that com¬
mittees were sometimes responsible for serious errors and that “the dem¬
ocratic approach to scientific truth is a contradiction in terms.” After
reviewing new information on adults’ requirements for egg protein, he con¬
cluded that the 1973 report (to which he had, of course, been a party) “with
its overly confident low estimates of protein requirement has had... a variety
of undesirable policy, educational, and public-health consequences.”43
In 1977 there was another small meeting of FAO/WHO consultants. They
agreed that calculation of safe intakes of protein should include a correction
for digestibility, and in the absence of actual data, a value of 90% could
be assumed for developed countries and 85% for others. This would au¬
tomatically increase the safe intake levels for practical diets by 11% and
18%, respectively. They also attempted to estimate how much protein and
energy a stunted child would need to “catch up” to normal size. For a child
weighing only 7.5 kg at 12 months of age to reach a normal weight of 11.4
kg at 18 months, the estimated average daily needs (per kilogram body
weight) were 1.81 gprotein (of milk quality) and 153 kcal (i.e., 4.7 PCals%).
If, in addition, the child had an infection, which inhibited growth, for 20%
of the time, the estimates increased to 2.41 g protein and 15 6 k<^d (i.e., 6.2
PCals%). Further, if the child, because of individual variability, needed 30%
more protein and 15% less energy than average, that child would apparently
need a protein concentration of 9.5 PCals%, still in terms of milk protein.
Nevertheless, it was thought that, in practice, it was the bulk of the diet
and the lack of appetite, rather than protein, which limited a child’s possible
rate of catch-up growth. Thus a child could be helped by the addition of
oil to a diet of cereals and starchy roots, even though this reduced PCals%,
because it increased palatability.44
The next official FAO/WHO report (published in 1985) was the out¬
come of a meeting in the fall of 1981. Just before that, a large body of
new experimental results had been presented at two conferences sponsored
by the UN University World Hunger Program.45 The report agreed that

43. Scrimshaw (1976), p. 136.

44. Beaton, Calloway and Waterlow (1979).
45. Torun, Young and Rand (1981); Rand, Uauy and Scrimshaw (1984).

192
 Reappraisals of the Third World Problem

more recent experiments indicated that the requirement of young men

for egg protein was higher than the estimate of 0.44 g/kg in the 1973
report. In nine studies using single, high-quality proteins, the mean daily
requirement for short-term (14-day) nitrogen balance in young men with
a habitually higher protein intake was 0.63 g/kg. It was suggested that
the difference might be explained by (a) the earlier work having been
done generally at energy intakes that promoted weight gain rather than
balance, (b) the balance point being extrapolated from points of negative
balance in some studies and, finally, (c) the allowance for nitrogen “losses
other than in feces and urine” having been set at 5 mg/kg per day rather
than the 8 mg used for the newer calculations. In longer (i.e., 1- to 3-
month) studies, the mean protein requirement was apparently met by
0.58 g/kg. Therefore, 0.6 g was accepted as the best estimate of the
average requirement.46
The group then set the safe level of intake of such protein 2.5% higher
(i.e., “ + 2 standard deviations”), assuming that the between-subject vari¬
ability was ±12.5%. The safe level of intake, for all but 2.5/0 of adult
men, thus became 0.75 g milk protein per kilogram body weight per day.
In the case of adults it was now thought unnecessary to correct for the
amino acid score of the diet because in none of a series, even of national
dietaries low in animal protein, was the level of any of the essential amino
acids limiting when calculated as milligrams amino acid per gram protein
and related to the actual amino acid requirements of adults as determined
in nitrogen balance studies (Table 10.2). In general, the results of more
recent research confirmed the findings of Rose for young adult males, as
described in Chapter 11. However, it was agreed that correction was needed
where digestibility of the protein was lower than that of egg or milk pro¬
tein.47
For 1- to 2-year-old children, it was considered prudent to add an extra
50% to the allowance for growth, since it was known that growth rate
fluctuated from day to day, yet there was little or no storage of dietary
proteins in low-growth periods. New research had also indicated that these
young children had considerably higher requirements for essential amino
acids (per kilogram total protein) than did adults, so that additional protein
was recommended where quality was low.48 The safe allowance corre¬
sponded to approximately 10 PCals%. The group confirmed the point made
in 1977 that stunted children needing to catch up to a normal size had
higher requirements.49

46. FAO/WHO/UNU (1985), PP- 79-8i.

47. Ibid., pp. 117-20.
48. Torun, Pineda, Viteri and Arroyave (1981).
49. FAO/WHO/UNU (1985), PP- 118-25, 142-8.

193
 ■o a

Comparison of suggested requirement patterns with concentration of four critical amino acids in the protein of
o o o o o o o o o
o o o o o o o O O C c
« «
U
C Q,
K S
2 a
'uf J5 b
u «
H hJ d d
d. d
"o
On O CO 1-H 0 0 ON 10 ON o H
00 O oc 0 0 00 ON 00 JC • *c
o
u CQ r"H T-H 1 CJ 1)
CO
>,
gQ

C nJ
(U ..
H d d ^^ W ^^ 42 o
ro Tt- ON r\ r-
2 o
< 00 co CO ON ON co CO 2 ^

o 2
O <u
c M -Q
rt a .2
as E 2
a, a g
o ^-v U
4->
Oh < ^
tu <u
U H ON 00 O <N
H ^ -
(75
_
a
a;
G
c
'S <u E ^
OJD
-t-* c 07
o u. .33
u *8
CU o ’3 •-
<u ^ E
60 Ul <u .5
as Tf 00 ON ^ ro O rv| 00 CO O co u —
'3b
H rn rd rD rn rn m ro ro rO ro rn ^3 >N
D Tj
S G
3 S
<j s
« <u
3
rt «
a
a,
o <u rt
c '5 c
o 'S
• —• yj
E _c So 5/5 • —
< *-» u
<U G U
^ (N r\ Tf 00 o co 00 T* (N
s + N N h ro rn ro rn ro co <N <N rn 'S
60 O
M C
E
<L) *3
c
‘3
00 no (\ ON rn O CO CO ON <N ON *- -c
hJ T* TH ro CO co ro
•c a
0 «
O O
«J 33
cw ^
U o .S
3
T3
C/> U5
« W
Source-. FAO/WHO/UNU (1985).

O 03 CO
u o
a, « -s
- c
b <U <U
TO
47
c/5 -a
-s a
ns o o*
g JO c
1 > Ul
.2
’C
4-»
n3 <U CO
C-U u
<J
5 ns 47 K 8 C/5 « OJ
4-»
<17
-C .3
rt o
«
T3 u t:
60 £ T3 07 §3 Ji
low-cost diets

a> 07 C/5
S.3 > W -I 3 T3 C/5 ns
oj (75 47 ns JD 0 _c
Table 10.2.

r» T3 u
13 c/5 a> <u > •~ C C fli 1
‘ ‘ (75
* ^
(J 0 -a g £ -o ns E o
S o IS
"3 OS
JJ Cc"0 -O -a C -3 > 3 w
Uc
2 o 1 <u ns -C
c a> 3.2 T>
2 §>-2^.» § C/5 0JD -C 3
_c
X O o
o <j
x C/5 ui c -a
(S' <D OS 3
^ .2 c « ^ 3 ns 47 C3 QO w rt
u -a •S -T3 a OS Jj « OS S 3 .SP U co _C —
Q< CO < Q 2 H PQ o H U
 Reappraisals of the Third World Problem

Is Protein First-Limiting in Third World Diets?

We can now return to the subject, left off at the end ofThe section titled
“Marasmus and Kwashiorkor,” as to whether kwashiorkor was to be con¬
sidered primarily a protein deficiency condition. This had, as we saw, begun
to be questioned in the period 1955—7, and increasing emphasis was being
placed by some workers on the low energy intakes of children at risk of
developing the disease. The greatly reduced estimates of safe levels of protein
intake in the 1965 FAO/WHO Report, discussed earlier, also made it seem
less likely that kwashiorkor could be considered simple protein deficiency.
In 1959 a study of children’s diets in South India indicated, for 1-year-
olds, an average daily intake of 1.75 g protein and 76.5 kcal per kilogram
body weight from a combination of breast milk and supplements. Compared
with the then-current Indian standards, the protein intake was only 50%
of requirement and the energy 69%, so that protein appeared to be limit¬
ing.50 By the time a similar study was published in 1969, the Indian Council
of Medical Research had modified its standards to bring them in line with
FAO/WHO recommendations.51 As a consequence, although the estimated
intakes were similar (i.e., 1.8 g protein and 74 kcal per kilogram body
weight daily), it was now concluded that protein intake was adequate and
that energy was the factor limiting the growth of these children, who were
well below the standards for both height and weight.51
P. V. Sukhatme, a statistician working at FAO, calculated, by the same
standards, the adequacy of a large number of household diets from data
collected in dietary surveys conducted in villages in the state of Madras
during the 1950s. In earlier papers he had concluded that these gave evidence
of a critical shortage of protein in India.53 He reanalyzed them against the
new standards, which, for a typical household, amounted to a need of 2,800
kcal and 30 g “reference protein” per “nutrition unit,” equivalent to 4.3
PCals%. For the analysis he assumed that animal protein in the diets had
an NPU of 80% and vegetable protein a value of 50%. The average cal¬
culated intakes were 28.5 g “reference protein equivalents” and 2,300 kcal
per “unit,” that is, 95% and 82% of the protein and energy standards,
respectively.
Sukhatme made two further points: first, that the protein standard rep¬
resented a mean requirement with 2 standard deviations added to allow for
individual variability, whereas the energy standard was a mean without any
added variability factor; and second, that nearly all of the households whose

50. Rao, Swaminathan, Swarup and Patwardhan (1959).

51. Nutrition Expert Group (1968).
52. Prahlad Rao, Singh, and Swaminathan (1969).
53. Sukhatme (1966), pp. 228—9.

195
 Protein and Energy

diets were low in protein also had low energy intakes. His final conclusion
was that, in India, “the critical factor is the calorie intake and not protein”
and that signs of protein deficiency arose because dietary protein had to be
used as a source of energy. There was no justification, therefore, for an
increased production of protein in all forms regardless of whether these
forms were high or low in calories.54 In a further paper he extended these
conclusions, saying that most other national surveys also showed that the
average diets had at least 5 PCals% (with protein adjusted to the equivalent
level of milk protein), indicating that any protein deficiency would be largely
the result of inadequate energy intakes. Under these conditions, supple¬
menting diets with protein or amino acids was costly and inefficient. The
only exceptions were those areas where the staples were starchy plantains
or cassava.55
In late 1970 the FAO/WHO Expert Committee met once more, with
Waterlow in the chair. Again, PCM was the subject of a large portion of
the report. The committee concluded:

There has been a tendency to over-emphasize the importance of either protein or

calorie deficiency alone, whereas in fact the two almost always occur together. The
questions of whether the clinical pictures of kwashiorkor and nutritional marasmus
reflect real differences in etiology... have an obvious bearing on diagnosis and on
the planning of preventive measures.... Dietary studies fail to provide conclusive
evidence. In the “sugar babies” in the West Indies,... the presence of ample amounts
of subcutaneous fat suggested that the calorie intake had been relatively high. On
the other hand, a study in India of the dietary habits of patients with ... nutritional
marasmus and kwashiorkor did not show any quantitative or qualitative difference
in the previous diets.... In general, the clinical picture of PCM tends to be that of
kwashiorkor in regions where breast-feeding is continued into the secqpd year of
life, whereas marasmus is the typical result when weaning occurs early_However,
the type of diet available is not always a good indicator of the type of PCM that
develops_We need to obtain more information about the diets of pre-school
children, which might serve as a basis for the planning of preventive programmes.56

This appears to be a carefully drafted, diplomatic statement, likely to be

acceptable to a group with wide differences of opinion.
In 1971 Nicol again wrote about the importance of the bulk factor limiting
the nutrient intake of most Third World children and added that, where
cereals were the basic staple, the new data supported the old phrase “Look
after the calories and the proteins will look after themselves.”57
Data published the next year from Uganda confirmed that children in
their second and third years, with starchy plantains as their staple, and even

54. Sukhatme (1970a).

55. Sukhatme (1970b).
56. Joint FAO/WHO Expert Committee on Nutrition (1971), pp. 51—2.
57. Nicol (1971), p. 85.

196
 Reappraisals of the Third World Problem

when the availability of food was unrestricted, had a mean daily intake
equivalent to only 67 kcal/kg body wt, as compared with the standard of
100 kcal. However, to the authors’ surprise, most of the Ugandan children
showed a normal rate of growth in this period. It was concluded from
energy studies that their rates of basal metabolism were similar to those of
Western children, but that they were much less physically active, and this
was enough to spare at least 20 kcal/kg.58
As Whitehead pointed out in a further paper, by comparison with current
standards this meant that the apparent energy deficit was greater than the
protein deficit, which at 1.5 g/kg and an assumed NPU of 55%, was 71%
of the standard.” One consequence of the lower energy intake was, of
course, that to obtain a standard protein intake the PCals% of the diet had
to be that much higher, in inverse proportion — for example, 7.5 PCals%
at 67 kcal/kg gives the same protein intake as 5 PCals% at 100 kcal/kg.
One could argue that the lower energy intake was, in fact, adequate. How¬
ever, it left little margin of safety for periods when appetite decreased as a
result of infection or other stress. Eating the standard 100 kcal/kg would,
for 1- to 2-year-old Ugandan infants, require eating some 1,000 g food, as
compared with the 250 g required to provide the same nutrients in a typical
Western child’s diet. Since a successful catch-up period after an infection
would require an even higher intake, it seemed essential to make the Ugan¬
dan diet more concentrated, which would be most easily achieved by adding
fat.6° Indeed, the very low fat content of African weaning diets had been
thought by one worker to result in a condition of essential fatty ac 1 defi¬
ciency.61
A further study from India reported the results of giving 300 malnourished
preschool children a dietary supplement that provided cereals, fat and sugar,
but with a low protein content (4 PCals%), that is, 310 kcal and 3 g protein.
The diet of the young children before they received the supplement consisted,
on average, of 700 kcal and 18 g protein. Over a 14-month period these
children grew significantly faster than 100 control children drawn at random
from the same initial pool of subjects. Children receiving the supplement
who had the measles showed little or no decline in growth, whereas those
in the unsupplemented group who had the measles actually lost weight over
the 3 months of the epidemic. The conclusion drawn was that diets based
on cereals could be consumed in sufficient quantities if offered in many
relatively small meals through the day and that there was no need for the
children to receive special protein concentrates; nor did it seem that par-

58. Rutishauser and Whitehead (1972).

59. Whitehead (1973), p- 105.
60. Ibid., pp. 115-17.
61. Naismith (7973).

197
 Protein and Energy

ticularly high protein levels were needed to provide resistance to infection.

The authors commented that people had begun to use the term “protein-
calorie malnutrition,” but so often still forgot the calorie aspect.61
McLaren suggested that “in view of the convincing evidence that had
been building up that the main deficiency was energy not protein, the term
‘energy—protein malnutrition’ should be adopted.”63 To this Scrimshaw and
the PAG made a spirited response. They accepted that there was no “protein
gap” in the sense that the amount of protein available in the world could
theoretically be distributed so that everyone received enough. However, in
the real world the situation was very different, both among families and
within families. Further, the diet on which a child developed kwashiorkor
was often not its normal diet, but consisted of something like rice (or barley)
water given during a long bout of diarrhea. It was accepted that the problem
stemmed basically from poverty and ignorance, but it was unacceptable to
allow malnutrition to continue until, somehow, in a future generation, pov¬
erty was eliminated. The PAG would continue to encourage improved pro¬
duction of protein of all types.64

Accusations and Reconsiderations

The battle was now on, and in 1974 McLaren wrote another, more abrasive
article for the Lancet entitled “The Great Protein Fiasco.” He said that the
PAG was caught in an identity crisis after having “evolved into a major
force in the fight to close the protein gap” and was setting in motion “a
long and disastrous train of events.” Marasmus was a more widespread
disease than kwashiorkor and its incidence was continually increasing as a
result of urbanization and the decline in breast feeding. It was because of
the biochemical changes, which fascinated laboratory-oriented scientists,
that kwashiorkor had received undue attention. Scores of protein-rich food
mixtures had been produced; the majority never reached commercial pro¬
duction, and most of the others sooner or later proved financial failures.
Several attempts to have the matter discussed at a policy-making level had
been thwarted by “the establishment,” and criticisms voiced at one seminar
had been deleted from the final report, the secretariat “keeping the party
line.” The true cost in money and time of the numerous projects, countless
meetings and innumerable publications was what might have been achieved
with the same resources, “whilst children were lost in the unchecked scourge
of malnutrition.”65

62. Gopalan, Swaminathan, Krishna Dumari, Hanumantha Rao and Vijayaraghavan (1973).
63. McLaren (1973).
64. Protein Advisory Group (1973); Scrimshaw (1973).
65. McLaren (1974).

198
 Reappraisals of the Third World Problem

There were some immediate responses. Brock, the South African professor
of medicine who had been the co-author with Autret of the 1952- monograph
Kwashiorkor in Africa, agreed with McLaren’s point about the relative
neglect of marasmus: “We could have described many undernourished and
marasmic African children — they were all around us. But they did not
represent what we had come to study and report on.’ However, he believed
that to question the central role of protein deficiency in kwashiorkor was
pushing the pendulum too far.66 Hansen, too, wrote from South Africa. He
accepted the point that “the protein-rich food programme had perhaps been
over-emphasized.” However, he believed that there were important simi¬
larities between marasmus and kwashiorkor, and that even a low-energy
diet based on cereals was likely to be low in protein, so that McLaren was
“perhaps overstating his case and confusing rather than clarifying issues.
For the preceding 2.0 years, two of the leaders of research groups over
the whole period had been Scrimshaw in Guatemala and then at MIT, and
Waterlow in Jamaica and then at the University of London. Scrimshaw had
already made his position clear on the controversy, but what would be
Waterlow’s view? It came in 1975 and was also quite clear. In a paper with
Payne, published in Nature, he withdrew his former position: “The concept
of a worldwide protein gap, derived from the diagnosis of kwashiorkor as
a protein deficiency state, is no longer tenable: current estimates of children s
protein and energy requirements are considered realistic, and by these cri¬
teria the problem is mainly one of quantity rather than quality of food.
The paper ended: “The evidence we have put forward leads to the conclusion
that the protein gap is a myth, and that what really exists, even for vulnerable
groups, is a food gap and an energy gap. This point of view is now fairly
widely accepted, particularly in this country [Great Britain] where it has
become part of aid policy.”69
The last point is reflected in the report of the Advisory Committee on
Protein, which was set up by a British government department to advise on
how best to provide overseas aid for the relief of malnutrition. Waterlow
was the only member with direct experience in treating malnutrition in the
Third World. The summary of the committee’s recommendations made no
mention of protein. The group advised that “the primary attack on mal¬
nutrition should be through the alleviation of poverty... in the interests of
better nutrition, aid should be directed to projects that will generate income
among the poor, even where such projects do not have any marked effect
on the national income of the country concerned.”70

66. Brock (1974).

67. Hansen (1974).
68. Waterlow and Payne (1975)5 P- XI3-
69. Ibid., 117.
70. ODA Advisory Committee on Protein (1975)5 P- 39-

199
 Protein and Energy

The report of the World Food Conference organized by the United Na¬
tions was also published in 1975. The resolution on “policies and pro¬
grammes to improve nutrition” extends to some 1,400 words, but “protein”
is mentioned only once:

Governments should explore the desirability and the feasibility of meeting nutrient
deficiencies, through fortification of staples or other widely-consumed foods, with
amino-acids, protein concentrates, vitamins and minerals, and with the assistance
of the World Health Organization in co-operation with other organizations con¬
cerned, should establish a world-wide control programme aimed at substantially
reducing deficiencies of vitamin A, iodine, iron/folate, vitamin D, riboflavin, and
thiamine as quickly as possible.71
*
There was no mention of a “world protein gap.”
The next report of the FAO/WHO Expert Committee on Nutrition
avoided the issue as to whether protein or energy was the main limiting
factor in PCM. This was done, it would seem, in part by changing the
composition of the committee. Neither Scrimshaw nor Waterlow or Jelliffe
was present, and Whitehead and McLaren had never been members. Most
of those invited were economists or development planners. According to
their report, the evidence indicated that 100 million children were affected
by moderate to severe PCM and that “the primary cause of PCM can be
overcome only by significant changes in the socio-economic characteristics
of the community.”71 They did not state what the primary cause was, but
they referred to the report of the preceding meeting, where, in fact, there
had been no agreed-upon conclusion. The report then went on, at some
length, but in a general way, to recommend that planning be aimed at
reducing poverty within countries rather than directly trying to develop
“gross national production,” which might leave the poor no bettef off than
before, or even worse off. However, there was a comment that “in earlier
years... undue emphasis on protein and the so-called ‘protein gap’ led to
undesirable approaches in national and international efforts in the field of
food and nutrition.”73
McLaren had already complained, after the 1971 report of the Joint FAO/
WHO Expert Committee, that the series of reports showed prejudices and
gross omissions, and that the choice of invitees was political rather than
based on scientific merit. “When a group was brought together for only a
few days, and with little preparation, a thorough study could not be ex¬
pected, and it was usually left to the secretariat with one or two of their
friends to write the final report.”74 His response to the 1976 report, ex-

71. United Nations (1975), pp. 9—10.

72. Joint FAO/WHO Expert Committee on Nutrition (1976).
73. Ibid., p. 33.
74. McLaren (1972).

ZOO
 Reappraisals of the Third World Problem

pressed in an editorial entitled “Nutrition Planning Daydreams at the United

Nations,” was even stronger. He described it as being full of undefined,
pseudotechnical verbiage concerning the creation of a bureaucratic planning
system that would, in some unstated and naively optimistic way, ‘ increase
the opportunities for employment.” In his opinion, there was a disturbing
unwillingness to grapple with the true causes of world hunger and mal¬
nutrition in the real world of vested interests and corruption.75 For whatever
reason, the Joint FAO/WHO Expert Committee appeared to go out of
existence at this time.
In 1980 a review was published on the role of protein deficiency in
kwashiorkor. The authors went into most detail in their consideration of
data obtained in Jamaica, but commented that data obtained elsewhere
were in agreement with their conclusion. In Jamaica, porridge, primarily
cornmeal, was the main weaning food, with about 70 kcal per 100 g. A
typical dietary of 1 bottle of bush tea, 1 bottle of milk, 2 bottles of porridge
and 1 meal from the family pot provided 57% of the 1090 kcal required
daily between 6 and n months of age.” If the children were to be given
enough porridge to meet their energy standard, the protein intake would
be considerably in excess of the RDA. The conclusion was that “there is
no basis for the view that kwashiorkor is a protein deficiency disease. The
authors stated that the true causes were complex and included infection
(especially measles) and possibly a variety of mineral deficiencies, including
those of potassium, magnesium, vanadium and zinc, as well as essential
fatty acid deficiency.76 They felt that the bush tea itself might have a dam¬
aging effect on a child’s liver.77
One of the signs of kwashiorkor that seemed particularly likely to be a
result of protein deficiency was edema. This had been thought to result from
the low levels of serum albumin usually found in infants with kwashiorkor.
However, a study from the same group in Jamaica showed that this condition
responded to increased energy intake, even when protein was kept down
to the daily maintenance level of 0.6 g/kg body wt. The authors said that
they did not know why children with kwashiorkor developed edema, but
that a low level of protein in the blood did not appear to be the basic cause.
In the 1980s Waterlow, now retired, had some further thoughts on the
subject and wrote that he found it hard to believe that edema was not due
to protein deficiency. Perhaps, after all, the i973 estimates of protein re¬
quirements for children had been set too low, and there was new evidence
indicating that the energy requirements had been set unnecessarily high.

75. McLaren (1978), PP- i296“7-

■76. Landman and Jackson (1980).
77. Bras, Berry and Gyorgy (1957)-
78. Golden, Golden and Jackson (1980).
79. Waterlow (1984); Waterlow (1990).

201
 Protein and Energy

The Complexity of Breast Milk

During this period there was more investigation of the different proteins
present in human milk. It was discovered that as much as one-quarter of
the total protein consisted of immunoproteins that resisted digestion by gut
enzymes and appeared to function as antibacterial agents.80 The fraction
present in the greatest amount consisted of secretory antibodies of the SIgA
type, in a form particularly resistant to enzymic attack. It was thought,
though there was no rigorous proof, that these proteins played an important
role in breast-fed babies generally having fewer episodes of infection than
those weaned early and fed in other ways.81
Another component found at a higher concentration in human milk than
in cow’s milk was lactoferrin, a protein with a high affinity for iron. It was
hypothesized that this compound acted within the gut to reduce the level
of free iron, and thus the proliferation of bacteria that have a high require¬
ment for iron. Other mechanisms helpful to the infant have also been sug¬
gested without real proof as to the important function of lactoferrin in
vivo.8z
Finally, it became clear that a significant proportion of infants could
develop eczema and other allergic reactions to the proteins in cow’s milk.
It was an oversimplification, therefore, to think of different protein sources
entirely in terms of the amino acids they yielded on hydrolysis. The overall
benefits of breast feeding in the Third World and the dangers of mothers
thinking it is “modern” or “Western,” and therefore that substitutes are
preferable, have been argued forcefully by D. B. Jelliffe.83

Summing Up 4

From 1955 on, attention was increasingly focused on the low energy intakes
of Third World infants having to subsist on bulky foods. The term “protein-
calorie malnutrition” replaced “protein malnutrition,” and it was discovered
that children recovered from kwashiorkor when given foods of relatively
low protein content provided that energy intakes were high. There was
controversy as to whether the cause of a malnourished child’s developing
kwashiorkor rather than marasmus was a lower level of dietary protein or
some stress such as infection and/or a failure of hormonal adaptation. In
practical terms, prominence was given to ways of increasing the caloric
density of Third World infant foods.

80. Raiha (1988).

81. Hanson et al. (1988).
82. Sanchez, Calvo and Brock (1992).
83. Jelliffe and Jelliffe (1978).

202
 Reappraisals of the Third World Problem

It appears that the controversy over kwashiorkor will continue, but with
the majority being of the opinion that it is usually the result of the stress
of infection in infants receiving overbulky food and that UN programs to
meet a world protein crisis by producing protein concentrates were based
on a misunderstanding of the situation. Another review of the whole subject
was published in 199Z.84

84. Waterlow (199Z).

Note added to proof. It has now been reported that children with kwashiorkor, but not those
with marasmus, have greatly increased blood levels of the leukotrienes LTCH and LTE4. These
compounds stimulate vascular leakage and edema formation. It is suggested that the stresses
leading to the development of kwashiorkor result in increased production of these endocrine
materials and that they, in turn, contribute to the development of edema [E. Mayatapek, K.
Becker, L. Gana, G. F. Hoffman & M. Leichsenrig. (1993)- Leukotrienes in the pathophysiology
of kwashiorkor. Lancet 342:958—60].

ZO3
 11 Adult Needs for Amino Acids: A
New Controversy, 1950—1992

It was generally accepted by 1950 that human needs were not really
for protein as such but for each of the essential, or “indispensable,” amino
acids, together with nitrogenous compounds that could be used to synthesize
the “nonessential” amino acid units. The latter are those that are not “essen¬
tial” in the diet because we can synthesize them for ourselves. But determining
the adequacy of diets to meet these needs required knowledge of the quantities
of individual amino acids supplied by each type of food, as well as reliable in¬
formation on the requirements of individuals at different stages of life.
In the 1950s methods for analyzing the amino acid composition of foods
were greatly improved. Martin and Synge in England had used paper chro¬
matography to separate the amino acids in protein hydrolysates. Then, upon
exposure of the paper to the reagent ninhydrin, each amino acid was seen
as a colored spot. Moore and Stein in New York developed an automation
of the same principle using a chromatographic column.1 2 All four won Nobel
Prizes for their work in this field.
There was still the question as to how digestible the proteins were, and
also whether individual amino acids in some processed foods might undergo
chemical reactions that would make them no longer “bioavailable” (i.e.,
usable by the person or animal consuming them). There was particular
concern about possible reactions between the amino acid lysine and the
sugars present in a food such as milk powder, and special analytical methods
were developed to check for such binding of lysine.2. Another concern was
that excessive levels of one or more amino acids might block the utilization
of another. Such an imbalance” effect could be demonstrated in experi¬
ments with rats, but it appeared unlikely to be of significance with the
mixtures of foods eaten by humans.3

1. Fruton (1972.), pp. 165-72.

2. Carpenter (1973).
3. Harper (1974), pp. 155-60.

ZO4
 Adult Needs for Amino Acids

Amino Acid Requirements

The work of W. C. Rose and his colleagues in developing rat diets that
contained free amino acids but no actual protein was described in Chapter
7 together with their first work on the requirements to keep young men in
nitrogen balance. Their final conclusions were published as “The Amino
Acid Requirements of Adult Man” in 1957.4 These confirmed that the eight
amino acids that they found to be essential were among those needed by
the growing rat, but that histidine and arginine were apparently not needed
by their subjects.
Similar work was being done in the same period by Ruth Leverton and
her colleagues at the University of Nebraska using young women as subjects,
and by others.5 Later work gave some indications of a requirement for
histidine.6 Almost all of the authors referred in their papers to the individual
variability of their subjects and to the strange phenomenon in which subjects
receiving amino acids in place of dietary protein of the same amino acid
composition apparently needed higher energy intakes in order to come into
nitrogen balance.
Another unexpected but consistent finding was that, even when the es¬
timates for requirement were based on the highest levels of each essential
amino acid that any individual apparently needed to come into nitrogen
balance, the total was still quite low relative to estimates of total protein
needs. The total daily requirement for all the essential amino acids including
histidine came to only about 94 mg/kg body wt. In contrast, the minimum
daily protein intake to maintain nitrogen balance was about 600 mg/kg, so
that the apparent needs for essential amino acids amounted to only 16%
of this value, whereas in the human body, or in whole egg, the essential
amino acids make up about 45% of the total protein content. And it had
seemed likely that the dietary requirements for replacing broken down tis¬
sues in adults would prove to be similar in amino acid composition to that
of the tissues being replaced.
In 1957 FAO proposed a hypothetical reference protein for use in scor¬
ing the protein of human diets.7 The essential amino acid requirements
amounted in total to 31.4 g 100 g protein. Histidine was not included,
but if it had been the total would probably have been increased by 2 g. This
was a compromise between the pattern of infant requirements, which were
assumed to be the pattern of amino acids provided by human milk, and the
pattern based on nitrogen balance experiments with adults. However, when

4. Rose (1957)-
5. Hegsted (1963); Irwin and Hegsted (1971b).
6. Kopple and Swendseid (1975).
7. FAO Committee on Protein Requirements (1957)-

205
 Protein and Energy

the results of a large number of protein-quality trials using nitrogen balance

with adult subjects were compared, it was clear that the reference protein
pattern did not give reliable predictions. For example, a food with a chemical
score (corrected for digestibility) of only 30% showed a performance, based
on nitrogen requirements, that was more than 70% that of egg protein.8
Successive international committees tried to design a single compromise
pattern, but the participants at an international meeting in Rome in 1981
decided to recommend different scoring patterns for different age groups
with the adult pattern being based on the results of nitrogen balance ex¬
periments summarized by Bodwell.9 Dietary protein mixtures that had ap¬
peared by calculation on the older “chemical score” systems to be seriously
deficient in one or more amino acids now appeared to be adequate if con¬
sumed by adults at a level providing the safe level of digestible protein (i.e.,
750 mg/kg body wt). To take a hypothetical extreme, cooked whole wheat
containing 12% protein on a dry-matter basis, if it provided 75% of the
total energy needs (i.e., 2,800 kcal) of a 70-kg man and was the only source
of protein, would still meet all of the essential amino acid requirements,
even though wheat protein had only one-third the lysine context of mixed
animal tissue proteins.

“Health Foods” and Amino Acids

In the 19 70s amino acids began to appear in health food stores as dietary
supplements - ironically, at the same time that scientific work was indicating
lower requirements for individual amino acids. The public in Western coun¬
tries was becoming increasingly nutrition-conscious. After World War II
“health food” stores marketing products that catered to the new^interest in
nutrition proliferated. Previously, “patent” medicines such as “liver pills”
with secret formulas and extensive advertising had been widely purchased
by people with a vague feeling that they were “below par,” but governments
had increasingly restricted both the kinds of medicines that could be sold
without a physician’s prescription and the claims that could be made for
their healing or preventive powers. Almost the only products that could
now be sold as “purveyors of health” were nutrients, with accompanying
claims that the foods in an ordinary diet did not provide them in optimal
amounts.
In the 1950s, the common products were mostly vitamin mixtures or
combinations of vitamins and minerals. Then health food stores increased
their product range to include protein powders (often based on casein),

8. FAO/WHO/UNU (1985), p. 121. \

9. Bodwell (1981), p. 350.

206
 Adult Needs for Amino Acids

which were advertised as a supplement for people wishing to increase their

muscle mass or to overcome a feeling of weakness and acquire more energy.
In the late 1960s a “liquid protein” diet was marketed for people who
wanted to lose fat but not lean tissue. The protein chosen was unfortunately
something like gelatin and was usually partially hydrolyzed to increase its
solubility. As we have seen from the earliest work in the nineteenth century,
gelatin is not a complete protein; it is, in particular, totally lacking in
tryptophan. Several deaths occurred among people trying to live on nothing
but “liquid protein,” and its production was stopped.10
As a consequence, the health food industry became aware of the signif¬
icance of essential amino acids, and they marketed a range of them, ascribing
different virtues to each. The public was encouraged to take one or more
as supplements according to the response they desired. An advertisement
that appeared in Prevention magazine for December 1984 described tyrosine
as an “amazing amino acid that resists aging.” A flyer found in a California
health food shop in 199Z described aspartic acid as “an amino acid for
your health.... in studies with human athletes they showed an increase in
stamina and endurance... it aids in the disposal of harmful ammonia and
thereby helps to avoid damage to the central nervous system.” Aspartic acid
is one of the amino acids that humans can synthesize for themselves from
other nutrients. Glycine was said to assist in prostate health, and lysine was
recommended because it was said to lessen the absorption of arginine and
thus to stop herpes and other viruses. It was also claimed that one could
be more certain of the digestibility of amino acid mixtures than of protein
and that these mixtures were available to the tissues in minutes rather than
hours.
The material bought most was tryptophan, which was promoted as an
aid to people who had a problem sleeping, and apparently it was effective
for a proportion of those trying it. However, in 1989 its use was discovered
to be associated with the development of a serious nervous disease and even
to have been responsible for some deaths. The batches of tryptophan from
one supplier were found to be contaminated with a very small quantity of
two impurities, each quite similar to tryptophan in structure and not pre¬
viously suspected of being highly toxic. As a precaution, tryptophan, from
any source, was withdrawn from sale.11
By 1990 most of the people taking special amino acid mixtures as part
of complex supplements were probably body builders. Such supplements
were advertised in muscle-building magazines as having “anabolic” (tissue¬
forming) activity. Some of them, made mostly from intact proteins, were
advertised as sources of “sustained-release aminos,” a euphemism for

10. Lantigua, Amatruda, Biddle, Forbes and Lockwood (1980).

11. Medsger (1990); Kilbourne (1992).

Z07
 Protein and Energy

“slowly digested.” By contrast, orthodox scientific opinion held that dietary

protein was not the factor normally limiting muscle building, which was
under hormonal control.

Nitrogen Balance Reexamined

Returning to orthodox nutrition, the unexpectedness of the low apparent

requirements for the essential amino acids led investigators to question the
validity of the nitrogen balance procedure as it had ordinarily been carried
out. In many studies energy intake had deliberately been kept high to ensure
that calories were not limiting, and this had resulted in greater nitrogen
retention.11 Adjustment periods had also usually been inadequate. In ad¬
dition, there had been either a rather arbitrary allowance for losses from
sweat, skin and hair growth or no allowance at all.
Adult protein requirements had not been a high-priority area of research
since 1920, because they seemed to be easily met in practice. However, by
i960 NASA, the U.S. space agency, was concerned as to how astronauts
remaining in space for long periods could be fed. A mission to Mars and
back might take 3 years. Plans were developed for recycling their excreta
by redistilling the water present and converting the urea and carbon dioxide
back into food, perhaps by photosynthesis, using simple plant organisms.
It was essential, in such planning, to have reliable values for the minimum
quantities of nutrients that would be needed to constitute a safe, long-term
diet. The agency was therefore willing to fund studies that would reexamine
the nitrogen balance technique.
Such studies were begun by Doris Calloway, Sheldon Margen and others
on the Berkeley campus of the University of California in 1963. A special
penthouse unit was set up where six subjects at a time could live under
constant supervision for up to 4 months. Only a single food was provided
- a premix powder, homogenized with water before each meal and drunk
under the eyes of a dietitian. Each subject had a rubber spatula, which he
used to scrape out any residues in his mug and then licked clean. The mix
included vitamins and minerals, as well as carbohydrate, fat and the protein
source being tested. The total energy value was normally adjusted so as to
maintain body weight. The standard protein source was powdered egg
white.
Physical activity was also controlled, with subjects normally spending 1
hour per day walking on a standard treadmill, but otherwise only sitting
or strolling in the unit. The toilets were tied shut, and each subject had his
own collection vessels for excreta, which were analyzed. Soiled toilet paper
was also analyzed weekly. Skin and sweat losses were measured as follows:

12. Calloway and Spector (1954); Manatt and Garcia (1992).

208
 Adult Needs for Amino Acids

After showering, each subject was sown into a one-piece cotton suit for the
next 3 days. After this, he washed again in a tub with a known volume of
water, and the water and the suit were then analyzed for nitrogen. Further
analyses were made for ammonia lost in the breath, and for nitrogen in
rinsing after toothbrushing and in semen losses. These so-called miscella¬
neous losses totaled approximately 350 mg nitrogen (equivalent to 2.2 g
protein) per day.13
Maintaining these procedures was exacting, even sordid, work, but one
can see its importance. If, for example, in earlier studies the food provided
was estimated to contain 10.0 g nitrogen per day and the collected excreta
gave, on analysis, 9.7 g, this result was assumed to constitute an example
of positive balance. However, if just 3% of the measured food had really
gone as “plate waste” rather than being eaten, and 3% of the excreta had
also been lost in being transferred from vessel to vessel, the losses did not
cancel out. The true intake of nitrogen was 9.7 g, and the true amount
excreted was 10.0 g (i.e., a small negative balance), and if the extra mis¬
cellaneous loss was 0.35 g, the true balance was really —0.65 g nitrogen
per day, equivalent to 4 g protein. If the efficiency of retention of additional
protein at this level of intake had been 50%, it would have required an
additional 8 g protein per day to bring the subject into true balance. (There
is a problem in presenting results of this kind. Authors refer to losses of
milligrams nitrogen per person or per kilogram body weight. However,
requirements are thought of in terms of grams protein per head. For con¬
sistency I will, from now on, convert nitrogen values to the equivalent weight
of protein, that is, “nitrogen content X 6.25,” and convert “per kilogram
body weight” values to “per head” by assuming 70 kg as the weight of the
standard man. Men were used in all the metabolic studies for which results
are quoted in this chapter.)
Despite the refinements in carrying out the balance procedure, there was
no significant reduction in the apparent variability of subjects. The mean
level of endogenous urinary excretion (i.e., after 7 to 10 days on a virtually
nitrogen-free diet) was equivalent to 16.6 g protein, with a coefficient of
variation (c.v.) of ± 18% among individuals.14 A year later, similar values
for altogether 83 young men were reported from a second institute: a mean
of 16.2 g protein, with a c.v. of ± i5%.15 The second set of results came
from the Massachusetts Institute of Technology (MIT), where Nevin Scrim¬
shaw had moved after his work at INCAP in Guatemala, referred to in
Chapters 8 through 10. Of course, the variability of results among individual
subjects was not necessarily due to faults in the procedure. There was no

13. Calloway, Odell and Margen (1971).

14. Calloway and Margen (1971).
15. Scrimshaw, Hussein, Murray, Rand and Young (1972).

209
 Protein and Energy

reason to assume that different individuals would show identical rates of

endogenous protein catabolism.
Another concern was that earlier experiments had indicated that there
were large, continuing retentions of nitrogen when subjects consumed very
large amounts of protein. Again, in a carefully conducted new study, six
subjects who received 225 g protein per day had large retentions for the
first week, as would be expected, but then apparently continued to retain
the equivalent of 10 g/day without any indication of coming back into
balance.16 Could this be a genuine retention, or were some routes of nitrogen
loss still not being measured? The question has remained unanswered.
The first practical application of the refined procedure was to determine
how far the requirement of subjects for good-quality protein exceeded their
endogenous losses. It had been assumed in the 1965 FAO/WHO report on
human protein requirements that the requirement for protein of ideal amino
acid composition and digestibility could be obtained by summing endoge¬
nous losses and, where necessary, adding nitrogen retained for body growth,
milk secretion and so on. In other words, there was an assumption of a
possible 100% efficiency of conversion from food to tissues.17 Although the
welfare of young children was really the major concern of most investigators,
the subjects most readily available for testing the principle were student-
aged males.
The results showed that the requirements considerably exceeded the the¬
oretical estimate. The mean daily endogenous losses (urine -I- fecal + mis¬
cellaneous) from a large number of studies were equivalent to 25 g protein,
and the mean requirement for balance (using eggs, fish or meat) was 42 g
for the standard man.18 In other words, adults used these first-class proteins
with no more than 60% efficiency, whereas they were virtually completely
digestible, and on any “scoring” system based on amino acid analyses their
scores would have been more than 85%. There was again considerable
variability between the calculated requirements of individual subjects, and
it was concluded that 52.5 g (or 0.75 g protein per kilogram) from first-
class protein would be necessary to meet the needs of 97% of the adult
population.19
The next question was, what quantity of less well balanced proteins was
needed? If a protein had a chemical score of 50 (i.e., 50%) compared with
egg protein, did that still mean that twice as much of it would be needed
for nitrogen balance? Both the Berkeley and MIT groups carried out relevant
studies with protein concentrates made from soybeans.

16. Oddoye and Margen (1979).

17. Joint FAO/WHO Expert Group (1965). ^
18. FAO/WHO/UNU (1985), pp. 56, 79—80.
19. Ibid., pp. 81-2..

210
 Adult Needs for Amino Acids

At Berkeley, subjects received varied amounts of either egg white or a

soy protein concentrate, with the total nitrogen content of each day’s diet
made up to the equivalent of 5 6 g protein with the nonessential amino acids
glycine and alanine. It was estimated that the mean minimal amounts of
egg and soy proteins required to produce nitrogen balance were 22.5 and
29.5 g, respectively. When the soy protein was supplemented with extra
methionine, the minimal requirement was 24.5 g. It appeared, therefore,
that the sulfur amino acids were first-limiting in soy for humans, but that
even so, soy could be diluted with nonessential amino acids and still maintain
nitrogen balance.*0
In the studies at MIT the protein sources were fed without nonessential
nitrogenous supplements. There was again considerable variability among
subjects, but when median values were considered, the requirements for
nitrogen balance with egg and soy appeared to be 39 and 51 g protein,
respectively. Supplementation with methionine did not give consistent im¬
provement, but analyses had indicated that the sulfur amino acid content
of the product used in this work was significantly higher than that used at
Berkeley.*1 The results from both centers were consistent with the inter¬
national standard of adult men needing a total of 0.9 g methionine + cystine
per day.
Many other nitrogen balance experiments, too many to list individually,
have been carried out using particular proteins, with and without amino
acid supplements. By and large, they too seem consistent with the inter¬
national standards for amino acid requirements. However, the results ob¬
tained with amino acid mixtures and no protein have been contradictory.
In 1955 Rose’s group had reported results indicating that they had obtained
nitrogen balance (even after adjustment for miscellaneous losses) with two
subjects who received the equivalent of 9 g protein from eight essential
amino acids and a further 29 g from glycine.** They concluded, therefore,
that the equivalent of 38 g protein was adequate for a male subject of
average weight. Their energy intakes were not stated, but the usual level in
Rose’s work, as mentioned previously, was about 5 5 kcal/kg.
Other groups were unable to obtain true nitrogen balance with such a
low level of total nitrogen. In one study the six subjects received the same
level of total nitrogen and of essential amino acids as in the experiment just
described, and 45 to 46 kcal/kg daily.*3 In another study, one treatment
provided two-thirds of the quantities of essential amino acids given in the
two studies just described (but still 30% more than Rose’s estimated re-

20. Zezulka and Calloway (1976).

21. Young, Puig, Queiroz, Scrimshaw and Rand (1984).
22. Rose and Wixom (1955).
23. Kies, Shortridge and Reynolds (1965).

211
 Protein and Energy

quirement for each) and with additional nonessential amino acids to provide
a total equivalent to 44 g protein. The subjects also received approximately
41 kcal/kg. Again, the subjects failed to come into true nitrogen balance/4
The authors suggested that the failure might have been due to the omission
of histidine, which Rose had considered unnecessary. But Rose too had
found that his subjects, who received amino acid diets, needed more than
50 kcal/kg to come into nitrogen balance, and it had been repeatedly con¬
firmed since the original work of Voit that nitrogen retention was improved
by an increased energy intake.

Amino Acid Turnover Studies «

Vernon Young, working at MIT, urged repeatedly that it was unsafe to

accept the 1985 international standards for adult requirements for amino
acids based solely on the results of nitrogen balance experiments/5 He had
three arguments. First, the procedure itself was suspect because the results
indicated that the subjects on a high-protein diet for long periods continued
to retain nitrogen. It seemed more likely that this was an artifact and that
some nitrogen was actually leaving the body in a form that went unmeasured
by the traditional analytical procedures - perhaps as molecular nitrogen in
the breath. If this source of error became obvious at high levels of protein
intake, it presumably occurred to some degree and without detection at
lower intakes. Second, the only attempts to test the full combination of
essential amino acids, all included at what Rose had recommended as “safe
levels for balance,” had in fact not been found to support nitrogen balance
in the test subjects. Third, although “balance can be achieved over a range
of protein intakes ... the rate of protein turnover in cells and organs might
well differ_[Hence] the question that arises is whether a given rate of
protein turnover is better than another for health reasons.”*6
Young and his colleagues therefore began a major research program using
a new approach to measure indirectly the balance of individual amino acids.
Typically the subjects received a powdered diet with energy intakes adjusted
so that body weight remained almost constant, as in the nitrogen balance
trials described earlier. The daily diets contained no protein but amino acid
mixtures contributing nitrogen equivalent to 56 g protein. The amino acid
under study was included at different levels, with the other essential amino
acids at generous levels and the level of the nonessential aspartic acid ad¬
justed to keep the total nitrogen constant. Usually five or six subjects were
used in a study of requirements for a particular amino acid. Each diet was

24. Weller, Calloway and Margen (1971).

25. Young (1986).
26. Ibid., p. 701.

212.
 Adult Needs for Amino Acids

maintained for 7 days, and altogether four or five diets might be tested,
each with a different level of the amino acid under study.
After a period of adjustment to a diet, each subject received for several
hours in the morning a continuous infusion, into an arm vein, of a saline
solution containing the amino acid under study, labeled with the nonra¬
dioactive isotope I3C (i.e., carbon with an atomic weight of 13 rather than
12). In some runs the subjects received their food in successive small por¬
tions; in others they continued their overnight fast/7 The rate of expiration
of carbon dioxide was measured and portions were analyzed further for
their I3C content. At the same time, blood samples were taken and analyzed
for their content of the free amino acid under study, and again portions of
the amino acid were also isolated and analyzed for their content of I3C.
The analyses reached stable plateaus toward the end of 3 hours. The method
of calculating balance values from the data obtained in this way is described
in Appendix C.
The general conclusion drawn from this work was that the adult require¬
ments for the individual essential amino acids leucine, lysine and threonine
appeared to be considerably greater than the standards arrived at as a result
of the earlier nitrogen balance studies/8 Young’s conclusion was that the
pattern of amino acid requirements of adults (and presumably teenagers
also) were not significantly different from the higher standards for 2- to 5-
year-olds (shown in Table 10.2) and were close to the actual composition
of animal tissues. He believed that this was the only “biologically credible”
pattern, in view of the fact that the amino acid pool entering the bloodstream
came from the breakdown of body tissues (mostly muscle). Since the proteins
of cereals are, in general, considerably lower in lysine than is animal protein,
one would expect that people with a marginal intake of protein, most of it
coming from cereals, would be deficient in lysine/9 However, it is clear
from experiments with adult rats that their requirement for lysine in relation
to other amino acids is very much lower than its relative level in rat tissues,
even though rats cannot synthesize it for themselves. These findings have
already been described in Chapter 7, and they indicate that the pattern of
essential amino acid requirements for maintenance can be very different
from the pattern found in animal tissues.
The discrepancies among the results obtained by different procedures are
troubling. It is hard to believe that someone could at the same time be in
overall nitrogen balance and in negative balance for an essential amino acid,
because the amino acid composition of individual proteins is unvarying.
The more likely explanation is that there were sufficient errors in one of

27. Meguid et al. (1986).

28. Young and Pellet (1988), pp. 13-18.
29. Ibid., pp. 26-9; Young, Bier and Pellett (1989), pp. 83-9.

213
 Protein and Energy

the two procedures to distort the conclusions. It has been a pertinent crit¬
icism of the nitrogen balance work on amino acid requirements that the
energy intakes were excessive, as judged by the usual standards. It would
therefore be of great interest to see the results from both types of measure¬
ment (i.e., nitrogen balance and isotope turnover) on the same subjects at
the same time, so that all the conditions, including energy intake, level of
stress, and so on, were identical.
Those who have carried out the newer isotope work have themselves
discussed possible sources of error in their own relatively complex studies.
However, it would seem that most of these would lead to underestimates
of requirements for amino acids so that their correction would only increase
the present discrepancies.30 Another possibility, though it seems remote, is
that humans receive significant quantities of indispensable amino acids from
the colon, where bacteria can use recycled urea as their nitrogen source. It
appears that, in people on a low-protein diet, as much as two-thirds of the
urea produced by the liver is secreted into the colon and recycled in one
form or another.31

Is “Balance” Enough?

One obvious question arising from the MIT studies was whether 7 days
were a sufficient period for the subjects to adapt to intakes kof an essential
amino acid that were considerably lower than those supplied by their pre¬
vious diets. To test this point, a study was carried out at MIT in which the
level of leucine was the variable and each level was ingested for a 3-week
period by a group of four or five subjects. One level was 1.0 g/day, corre¬
sponding to the 1985 international standard, and another was*2.i g/day,
the level the MIT group had previously found to be necessary to achieve
leucine balance. After 1 week of treatment, all the subjects were calculated
to be in negative balance, but appeared to have come very close to balance
at the end of 3 weeks. However, the authors did not conclude from this
that the lower level was satisfactory. Rather, they described the subjects
who had received it as being in a state of “accommodation.”3Z
This term requires explanation. Waterlow, in particular, has stimulated
an interest in the subject of adaptation to changes in the intakes of different
nutrients.33 For example, how is that the rate of urea synthesis adjusts so
rapidly to changes in protein intake? He has also made the point that some
“adaptations” may be at the expense of full functional ability. A malnour-

30. Meguid et al. (1986), p. 778.

31. Millward, Price, Pacy and Halliday (1991), p. 213; Langran, Moran, Murphy and Jackson
(1992.)-

32. Young, Gucalp, Rand, Matthews and Bier (1987), pp. 14—15.
33. Waterlow (1985).

ZI4
 Adult Needs for Amino Acids

ished child may survive, but with a reduced rate of growth, and become a
“small-sized adult” with less than average muscular mass and physical
strength. In this example the “adaptation” has been at the cost of some
degree of physical function. Such a result has now been termed “accom¬
modation” to distinguish it from true, or complete, adaptation with no loss
of function.34
The MIT group’s argument that its 3-week results at a relatively low (i.e.,
1.0-g) leucine level represented accommodation rather than full adaptation,
even though there was “balance,” was as follows:

1. On the 1.0-g diet there was a fall in the rate of protein turnover in the
second and third weeks of the diet, whereas this did not occur at the 2.1-
g level. (This involved further calculations that have not been described
here.)
2. When the subjects were returned to a 5.6-g level of intake, there was a
period of considerable nitrogen retention for those who had previously
received 1.0 g, but much less with those who had received 2.1 g, which
indicated that subjects on the 1.0-g treatment had lost a significant quantity
of body nitrogen, and there was therefore concern as to their “maintenance
of adequate nutriture.”3S

These arguments are analogous to those used by Voit and Atwater 100
years earlier for recommending higher total protein intakes than the mini¬
mum levels needed to obtain nitrogen balance. As was discussed in Chapters
5 and 6, they referred to the greater protein “stores” in the bodies of people
living on high-protein diets, and they believed that this “labile” protein was
a source of nervous energy and drive, as well as of resistance to disease.
The authors of the new studies relate the advantage of higher intakes of
essential amino acids to increased rates of protein synthesis and, presumably,
the potential for more rapid synthesis of immunoproteins, and therefore a
more effective response to infection. But to try to measure what is required
can lead us in a vicious circle.
The level of protein required for subjects not to lose “labile” protein, and
not to show a reduced rate of protein synthesis, is usually the level they
were eating before entering the study. This is seen in a study from another
center where measurements of amino acid turnover were made just in the
first 24-hour period in which the lysine intake was reduced from a high
level, in order to determine its requirement by adults.36 As Hegsted has
written of balance studies for calcium, “They represent primarily a study
of previous intake.”37

34. Scrimshaw and Young (1989).

35. Young, Gucalp, Rand, Matthews and Bier (1987), pp. 15-16.
36. Zello, Pencharz and Ball (1993).
37. Hegsted (1952.), p. 199-

215
 Protein and Energy

With the realization that the amino acids released from protein turnover
are available for reuse, it is difficult to explain adult needs for amino acids
in terms of wear and tear. The nitrogen lost each day in rubbed-off skin,
hair and sweat is much less than our minimal total loss, largely as urea. I
have suggested elsewhere that, by thinking only of the requirement for
protein, we forget our continual need to discard accumulated amino acids.38
This is achieved by deamination and further metabolism of the carbon
fragments. Complete enzyme systems are required to carry out these func¬
tions. And since our food supply has throughout our long evolution con¬
tained a fair amount of protein, these systems have had to be in continual
operation. There would have been no evolutionary advantage in an ability
to reduce the activity of the systems further than is actually found. On this
view, it would be the detoxification mechanisms for amino acids that de¬
termine our minimal requirement rather than any true need of the tissues.
Another worker has argued that “maintenance amino acid needs may be
driven in large part by a continuing loss of protein in the gut.”39 This would
come from digestive enzymes and other secretions that are not completely
digested in the small intestine and are metabolized by bacteria in the large
intestine. Amino acids lost in this way come from proteins relatively rich
in threonine and the sulfur-containing amino acids.
Unfortunately, we still have no clear data for the relative health, longevity
and vigor of peoples whose protein and amino acid intakes differ greatly
but whose intakes of trace nutrients are satisfactory and whose environment,
life-style and intake of fats, fiber and so on are all generally similar. People
living on diets low in protein, or with protein mixtures of doubtful quality,
usually have marginal intakes of one or more trace nutrients also. Even the
results of animal experiments have been equivocal, and we stili lack firm
evidence as to whether higher intakes of protein and amino acids confer a
long-term advantage.40
Fortunately, it appears that even the proposed higher amino acid stan¬
dards for adults are generally met in practice. The most recent national
survey in the United States, using 24-hour recall data, shows a median intake,
even for people living below the poverty line, of 35 g protein per 1,000
kcal consumed (i.e., 14 PCals%).41 This quantity should be ample to meet
the proposed higher amino acid standards in view of the relatively high
proportion of animal protein in these diets.
There has also been a recent and detailed study of food intake in three
villages in different parts of the world where malnutrition problems had

38. Carpenter (1992.), pp. 914-15.

39. Reeds (1990), pp. 493-5.
40. Munro (1964), pp. 410-12; Millward (1985), p. 144; Munro (1985), pp. 163-5.
41. U.S. Department of Agriculture (1985).
 Adult Needs for Amino Acids

been expected. The mean daily protein intakes of the adult women subjects
in Egypt, Kenya and Mexico, respectively, were i.i, i.o and 1.26 g/kg body
wt.41 Adjusting these values for the calculated digestibility of the proteins
and their amino acid score (based on “toddler” standards, as suggested by
Young), the quality-corrected values come to 0.95, 0.73 and 0.93 g/kg,
respectively; these levels should still meet their requirements.

Upper Limits of Desirable Protein Intake

A concern has arisen in the past few years that high protein intakes may
place an undesirable stress on the system. For the first time, the 1989 edition
of the U.S. Recommended Dietary Allowances included a recommendation
that protein intake not exceed twice the RDA of 0.8 g/kg body wt.43 The
recommended upper limit for the standard man therefore becomes 112 g/
day. This highlights the change in views during the twentieth century, since
even 115 g was claimed to be insufficient for a working man in official U.S.
government reports in 1900.
However, there is no more evidence for 112 g as an upper limit than
there was for 115 g being too low. Admittedly, the new recommendation
is expressed in tentative terms, but nevertheless once a number gets into the
textbooks, it will be assumed that there is hard evidence to support it, as
with the high standard of a century earlier. Those who proposed the rec¬
ommendation felt that higher intakes conferred no benefit, and that there
were suggestions from animal experiments that they might accelerate a
decline in kidney function.44 Also, they did not wish to encourage a high
level of beef (or other meat) consumption because of the indirect adverse
effects of the saturated fat that inevitably accompanies the protein. However,
it is a dangerous precedent to assume such possible indirect effects in one
direction or the other. In some contexts, as we have seen, meat may also
be a vital source of cobalamin (vitamin B, J and zinc.

Summing Up

In the 1950s a series of results were published which indicated that adult
requirements for essential amino acids (EAAs) were small, corresponding
in total to no more than about 16% of the overall quantity of protein (or
amino acids) required to maintain nitrogen balance. This is very different
from the 45% of EAAs present in our body proteins. One implication of
these results was that protein quality (other than digestibility) would be of

42. Calloway, Murphy and Beaton (1988), p. 22; Calloway et al. (1992-). P- 55-
43. National Research Council (1989), PP- 7X~3-
44. Committee on Diet and Health (1989), PP- 260-6.
 Protein and Energy

practically no importance in the formulation of adult diets. In practice,

virtually any mix of foods that provided the standard quantity of digestible
protein would be as good as any other.
However, the correctness of this conclusion depended entirely on the
validity of the nitrogen balance procedure. This was simple in principle, but
had shown unexplained prolonged retention of nitrogen in subjects contin¬
uing to receive diets very high in protein.
A group at MIT has attempted to obtain independent measures of the
requirements for individual amino acids by labeling them with stable iso¬
topes and following the balance of the labels. The conclusion drawn from
these experiments has been that the requirements are at least double the
estimates obtained by the nitrogen balance procedure. Further, feeding the
minimal EAA levels required for balance has been found to result in lower
turnover rates, and this was thought to be undesirable. The “desirable”
levels of EAAs were therefore even higher and roughly equivalent (as per¬
centage of total protein need) to those of toddlers.
This conclusion is still controversial because of the complex assumptions
that need to be made when drawing conclusions from studies using isotopes.
However, the results of dietary surveys have indicated that, in general, actual
adult diets do provide the higher levels of EAAs recommended by the MIT
group. A new concern is that people may be consuming an undesirably high
intake of protein, which could, with general aging, accelerate the loss of
kidney function.

218
12 In Retrospect

Looking back at the changing ideas about protein nutrition

described in this book, we see, of course, a great increase in knowledge
during each century, from 1700 to 1800, from there to 1900, and then
again during the present century. However, on looking at the story in more
detail, we see that from decade to decade there was not steady progress,
but often “two steps forward and one step back.”
It is no surprise for historians of science that in any scientific field ideas
which in one generation seemed to be firmly based truths should be over¬
turned in the next. However, it is still a shock for experimental scientists
to find that some conclusions which appeared to follow logically from
carefully collected data could still turn out to be untrue. It seems to us of
prime importance to discover how this could be, so as to be able to reduce,
or help our successors to reduce, the danger of drawing wrong conclusions
again in the future.

The Complexity of Things

One obvious source of error was the assumption of investigators that they
had in their hands all the essential information they needed to propose a
sensible question and answer it. We saw this after Beccari’s extraction of
the gluten fraction from wheat flour and the demonstration that it was
similar to “animal substance” in its properties. The question was asked:
“Which is the nutritive part, the gluten or the starch?” Then because po¬
tatoes, rich in starch but lacking gluten, appeared to be a nourishing staple
food for peasants in some areas, it was concluded, at least by some, that
starch was the essential nutrient. Now we know that it was wrong to
conclude that potatoes lacked “animal substance” just because they did not
have gluten, which happens to be insoluble and gluey in cold water. More¬
over, the question implied as a starting point that there was only one nu-

219
 Protein and Energy

trient, an assumption without any experimental basis. In fact, we now realize

that nutritional needs are quite complex.
There was a similar error in Takaki’s conclusion in the 1880s that the
disease beriberi was the result of a deficiency of protein in the diet. As
described in Chapter 7, he noted that in the Japanese navy, where the disease
was a serious problem, the nitrogen—carbon ratio of the rations issued was
well below the German standards for the same period. He therefore hy¬
pothesized that this was the cause of the problem, and the Japanese au¬
thorities agreed to a trial voyage with foods chosen to provide a higher
nitrogen—carbon ratio. The men remained healthy and Takaki understand¬
ably believed that he had proved the correctness of his hypothesis. We now
know that the foods added to the rations, largely in place of white rice,
changed the level of many other nutrients also and that an increase by a
few parts per million in the thiamine content of the ration had been the
important change. In this instance, the problem had been solved, but the
result was actually misleading and confusing to others who had obtained
promising results by modifying the procedures used for preparing rice for
consumption.
Kwashiorkor was another disease that it was believed at one time could
be classified in a simple way as a “protein deficiency” condition, and im¬
portant allocations of resources were made on that basis. However, further
studies have forced workers to see it as another example of a problem whose
complexities have not yet been fully unraveled.
There have been a number of studies indicating that, even in communities
where there was no obvious malnutrition, the growth of individual children
correlated significantly with their intake of animal protein.1 It is now realized
that animal products contribute more than protein, so that “animal protein”
could, for example, serve only as an indirect indicator of the level of vitamin
BIZ in the diet. Milk is also a good source of calcium, and meat of iron and
zinc. In addition, the minerals from these sources are usually better digested
than are those from unleavened grains and vegetables, where they can be
bound to fiber and/or phytate.

Falsifying Hypotheses

One idea as to how science progresses, associated particularly with the name
of Karl Popper, is that scientists formulate hypotheses, then test and try
hard to refute them by carrying out critical experiments. Logically, any
number of results may seem to support a generalization but still fail to prove
its correctness. Takaki’s test of his theory that beriberi was due to protein
deficiency is a clear example of this. At the same time, a single result contrary

1. Golden (1988), pp. 153-5; Allen et al. (1992).

220
 In Retrospect

to what is predicted by the hypothesis should be sufficient reason to abandon

it. But doubts have been raised as to whether this is what happens in practice.
Certainly scientists are not usually enthusiastic about trying to refute their
own hypotheses, and it is more commonly others who take on that task.
In the protein story, the clearest example of an attempt to refute a hy¬
pothesis with a single experiment is that of Fick and Wislicenus in 1865 (as
described in Chapter 4). They set out to show that the amount of protein
that their bodies broke down and converted to urea during a day’s climb
would be insufficient to provide the energy required to lift their bodies up
the mountain. They expected this to result in a general and immediate
abandonment of Liebig’s teaching that muscular energy was obtained en¬
tirely from the breakdown of a portion of the muscles’ own protein tissue.
The results were clear-cut; the authors believed that they had achieved
their end and that Liebig would immediately withdraw his hypothesis. In
fact, this did not happen. Some scientists were convinced by the work of
Fick and Wislicenus, but Liebig and the Munich group “modified” the
hypothesis in ways that were thought to preserve its validity: “The broken
down protein is not immediately converted to urea, and even when produced
there is a delay in its being secreted into the urine: The flaw in the mountain
experiment was therefore that urine collection was not continued for a
further Z4 hours.”
When further investigations of the response to physical work showed
little evidence of a “second-day” peak in urea secretion, a further modifi¬
cation was suggested: “Muscle protein also breaks down during rest, and
this increases the potential energy in the muscle, in a way comparable to
the winding up of a spring, that gives it a stored, latent power.” Growth
of the idea that muscular energy was linked to the oxidation of fats and
carbohydrates rather than of protein appears to have been gradual and the
cumulative result of many studies, including observations of greatly in¬
creased carbon dioxide production during energy expenditure.
As a student of the process of science has written: “Scientists go to great
lengths to salvage the views to which they are publicly committed. That
way scientific hypotheses get an honest run for their money.”* In Chapter
6 we saw an example of a scientist being proved right to defend and salvage
his views despite an apparent refutation. In 1912 McCay had argued, from
a study of different ethnic groups in India, that those receiving the least
protein were the least vigorous and that, therefore, Chittenden’s recom¬
mendation for reducing the protein content of Western diets could also be
expected to have a weakening effect in the long term. Chittenden responded
by suggesting that the “poor” Indian diets were also lacking in unidentified
trace nutrients and that this would prove to be the true basis of their

2. Hull (1988), p. 13.

221
 Protein and Energy

inadequacy. It seemed a long shot at the time, but subsequent research has
lent support to this claim.
In studies of kwashiorkor, one important trial was designed to refute the
hypothesis that the disease was caused by some kind of vitamin deficiency.
This was the South African work, reported in 1956 (and referred to in
Chapter 8), in which sick children were given a mix containing only sugar,
minerals and amino acids. The children’s positive response was good evi¬
dence that kwashiorkor was not due to a deficiency of a vitamin, even an
unrecognized one present in milk powders.
After good results were obtained by giving patients mixes with 25% or
more of calories coming from milk protein, it became the working as¬
sumption that the disease represented a protein deficiency, and there was
no deliberate attempt to refute this hypothesis. However, variations in the
formulas, designed to raise calorie intakes, had the indirect effect of reducing
the proportion of energy coming from protein. Finally, in 1966 (as described
in Chapter 10), Chan and Waterlow reported excellent results with a high
calorie intake, less than 5% of it coming from protein, and this finding was
clearly incompatible with the simple “protein deficiency” hypothesis.
However, I do not believe that any one observation changed the general
perception about kwashiorkor. In medicine, individual cases can have special
features. The observer therefore develops an “ideal” or “model.” As in other
fields, “the model represents a simplified structuring of reality... [it] gives
prominence to some features of our knowledge, and obscures and distorts
other features.”3 We hold tenaciously to our models, as providing a satisfying
substitute for the overwhelming complexity of the real world. But at some
point the accumulation of discordant evidence is sufficient to force us into the
unwelcome effort of reorganizing our thoughts and finding a new.pattern, or
model, for them. With hindsight, it seems that there was a steady accumula¬
tion of evidence relating to anorexia and low energy intakes in kwashiorkor
from 1955 to 1967. Then suddenly some “last straw” was enough to force a
complete revolution of thought, with different pieces of evidence brought to
center stage. This is comparable, on a smaller scale, to what Kuhn described
as causing a change of paradigms in major scientific revolutions.4

The Overextended Generalization

Sometimes a general principle becomes so firmly established in people’s

minds that they accept it, without question, as a law of nature rather than
as a testable hypothesis. The obvious example of this in the history of protein
is the principle that only the plant kingdom is capable of building up complex

3. Chorley and Haggart (1967), pp. 22-3.

4. Kuhn (1970).

222
 In Retrospect

molecules and that, as a part of the balance of nature, the animal kingdom
can only utilize them and/or break them down by oxidation. Both the
German and French schools claimed the credit for making this generaliza¬
tion, as we saw at the end of Chapter z. Undoubtedly this preconception
delayed the adoption of iron salts as a simple treatment for the iron-
deficiency anemia so common among women in Victorian times. German
authorities stated that “just as there was no possibility of humans being
able to make protein from potassium nitrate and starch, it was equally
improbable that they could use inorganic iron to make hemoglobin.”5
The belief that animals could not synthesize proteins came partly from
the general similarity between proteins found in plants and those in animals.
Thus Liebig wrote of “vegetable casein” and “vegetable albumin.” However,
it was realized in the second half of the nineteenth century that, when
proteins were exposed to digestive juices, their properties changed consid¬
erably. At first, researchers minimized the importance of this by describing
the changes as merely a breakdown of adhesive forces between molecules
so that they become more soluble and diffusible across the gut wall. When
this no longer seemed an adequate explanation, physiological chemists, as
we saw in Chapter 7, spent years trying to categorize the products of gastric
digestion as peptones, proteoses and other molecules according to their
solubility under different conditions. The unstated reasoning behind this
effort was, presumably, that a protein core might be identified that diffused
into the bloodstream and formed the basis of our body machinery. When
actual crystalline amino acids were recovered from proteins being digested
either in vitro or in vivo, they were at first dismissed as artifacts resulting
from the abnormal conditions of the experiment. And when the evidence
for their existence in the lumen of the small intestine became irresistible, it
was argued that this must be nature’s way of disposing of surplus protein
when the intake exceeded the requirements.
It seems that a general principle can lie so strongly at the back of scientists’
minds that it, rather than the observations, is the unconscious starting point
of discussion, so that interpretation is not truly open to discussion. To quote
another writer, “The way to be objective is not to get along without the con¬
cepts and theory, but to get [them]... out into the open, so that nothing is
smuggled in surreptitiously and everyone can see precisely what [they]...
are.”6

The “Great Man” Syndrome

Trowell has referred bitterly to Stannus’s dismissal, from his Harley Street
office in London, of the idea that kwashiorkor could be anything but in-

5. Carpenter (1990), p. 14Z.

6. Clark (1959), P- xo7-

Z2.3
 Protein and Energy

fantile pellagra: “Experts should never be consulted about the status of a

new idea in medicine.... they have climbed on the greater steps of their
own pinnacle of great eminence; they are hostile to a new idea which may
challenge their pre-conceived notions.”7 Such behavior has not been uni¬
versal, and there are many examples of people reminiscing, in later life,
about the help they received from their seniors in their early work. However,
in my study of the history of scurvy, it became clear that, time after time,
men who had become justly famous for some other discovery were heeded
as “authorities” when they gave their opinions that the cause of scurvy was
an infection, lack of potassium or of protein and so on, with very little basis
for these opinions.8
The first obvious example of this phenomenon in relation to pfotein is
Liebig’s dictum that protein provided the sole source of the energy required
for physical work and that the body could get rid of protein only through
the work of muscular contractions. Undoubtedly this statement had great
effects on the following generation of teachers. Indeed, it may have been
indirectly responsible for thousands of deaths among poor housewives in
the southern United States at the beginning of this century who believed
that, in addition to giving the available milk to their growing children, they
should give most of the meat to their husbands — the “bread winners”
engaged in physical work. This left the women with little but cornmeal and
white flour, and prey to pellagra, the disease caused by a deficiency of niacin
and other B vitamins.9
When he published his opinion in 1842, Liebig had already deservedly
gained recognition as the leading organic chemist of his day. But as explained
in Chapter 3, he had carried out no experimental work in animal physiology,
and his elaborate conjectures formed a striking series of hypotheses and
nothing more. They carried weight because of Liebig’s public reputation,
and the energy and ability with which he disseminated them. His critics
were outspoken, but their writing appeared only in specialist publications,
and the bulk of Liebig’s readers did not know of their existence.
Why was Liebig so unwilling to abandon his belief that muscles could
obtain energy only from decomposition (or other changes) in their own
substance? In part we could say that it was pride and a psychological
difficulty in admitting error. Yet he did change his mind on other matters,
even if he usually found someone else to blame for having initially misled
him. Perhaps the real key lies in his concept of “blood” and his insistence
that blood was the only source of nourishment for the tissues. That seems
obvious enough, but the picture that a modern biologist has of the blood-

7. Trowell (1982), p. xxii.

8. Carpenter (1986a), pp. 25 i-z.
9. Carpenter and Lewin (1985), p. 550.

224
 In Retrospect

stream is something like a river carrying an enormous assortment of com¬

pounds. Some of these, like the immunoglobulins, go around and around,
carrying out their function within the bloodstream. In contrast, compounds
like glucose and free amino acids are rapidly picked up by some tissues,
while other organs introduce more, so that turnover rates bear little relation
to the actual concentration of substances in the blood.
All this was unimagined by Liebig when he wrote his Animal Chemistry.
There was no procedure even for detecting the 0.1% (approximately) of
sugar in blood. William Prout himself, who believed that sugar played a
special role in the “animal economy,” and had specialized in the analysis
and identification of sugars, made no mention of it in his table published
in 1845 listing the components of blood.10 It was only in 1847 (after the
last edition of Animal Chemistry) that Claude Bernard began 10 years of
work on the source of the sugar that he found in blood.”
Liebig, it seems, thought of blood as a solution (or suspension) of two
albuminoids that had essentially the same composition as flesh, so that its
conversion, as a whole, seemed only reasonable. Of course, we can think
of all sorts of questions that we would like to have asked him, such as
“Where did absorbed fats and sugars go, if not into the bloodstream?” But
they did not seem to occur to him at the time. Moreover it is understandable
that Liebig (and Voit also) found Frankland’s analogy with the steam engine
difficult to accept. This implied that sugars and fats were burned within the
working machinery of the muscles. Yet the evidence at that time indicated
that the muscle fibers, which contracted, consisted only of albuminoids and
that fats were deposited in separate compartments of the body.
Carl Voit, who succeeded Liebig as the generally accepted “authority”
on protein, spent his life doing physiological experiments with both human
subjects and dogs. With hindsight, his most important contributions appear
to have been his demonstration that increasing total energy intake improved
nitrogen balance, and that when an animal in balance was changed to a
diet of higher protein content, it first retained nitrogen and then gradually
came back into balance but with an increased content of relatively “labile”
nitrogen that would be rapidly lost if protein intake was again reduced.
Voit’s dictum was that an active man of standard weight should consume
115 g protein per day. But as pointed out at the end of Chapter 5? he never
tested whether active men could come into balance and maintain their
strength on moderately lower levels (say, 80 to 100 g). The standard was
derived entirely from the impression that this was the quantity of protein
contained in the food that working men in Voit’s area liked to consume
when they could afford it, combined with the assumption that when people

10. Prout (1845), pp. 437—9> 467-

11. Holmes (1974), pp. 4i3-4i-

225
 4

Protein and Energy

were not limited by poverty they would instinctively select the level of
nutrients they needed. Again, we have an example of a standard that the
public believed to be authoritative and derived from the scientific work of
an expert although, in fact, the standard and the work had little or no
connection.

The Contribution of Women

A considerable literature argues that the male scientist’s choice of research

subjects in biology and the conclusions that he draws tend to be biased
toward what are seen as male characteristics of aggression and competition,
challenge and defense, to the neglect of cooperation and sharing of resources.
In more subtle ways too, male science may have an unconscious bias toward
the need for control and mechanization. The obverse of this concept is that
women can take a different and complementary approach that may throw
an entirely new light on particular topics.
Before 1935 no woman appears in this story. This is to be expected, since
scientific research and academic medicine were almost entirely men’s worlds.
Then Cicely Williams published her first short paper on kwashiorkor. It
had never been suggested that this was a new disease in Ghana, so that
presumably her attention focused on a problem that was present but not
“seen” by her male predecessors in the colonial medical service. As described
in Chapter 8, she was publicly snubbed by the current male “authority” on
tropical diseases and transferred to Malaysia - presumably to get her quietly
out of the way. However, her fate might have been even worse. Her paper
was accepted for publication by the editors of the journal to whom she sent
it, and men elsewhere in Africa responded in print, stating thaPt they had
observed a condition similar to the one she had described.
Twenty years later it had become conventional to categorize kwashiorkor
as straightforward protein deficiency and, particularly in the West Indies,
to assume that the victims were “sugar babies” receiving, if anything, an
excess of calories. At an international meeting on the subject held in Jamaica,
a local physician, Katerina Rhodes, claimed that her experience was differ¬
ent: babies that she had seen developing the disease seemed to have little
appetite. The regular, international participants at such meetings were re¬
luctant to accept this, but she stood her ground.
Another 2 years later, Dr. Hebe Welbourn, who had been working in
child welfare clinics in Uganda, also reported anorexia in her cases, and she
reminded readers that Cicely Williams, too, had spoken of poor appetites.
I am left with the impression that these women spent more time actually
observing the children’s behavior in the company of their mothers and nurses
- a style contrasting with the caricature (if one likes) of a male doctor
hurrying back to the laboratory after gaining a quick clinical impression.

22.6
 In Retrospect

In any case, they clearly made important observations that had not previ¬
ously received attention.

Committees and Consequences

In the nineteenth century, nutritional standards were put forward by indi¬

viduals (e.g.. Smith, Voit and Atwater), as we have seen. In the 1920s and
1930s, with debates over issues such as whether the poor were suffering
from malnutrition serious enough to warrant government intervention, com¬
mittees began to be established in order to obtain agreement as to what the
standards of consumption should be. The period since 1945 has seen the
growth of national and international organizations, which are the major
sources of funding both for research and for actions designed to help the
disadvantaged. It is important that such organizations be seen to be objective
and to rely on conclusions agreed to by a committee of experts as a basis
for distributing funds. One committee function has been to set out dietary
standards of requirements or allowances.
Thanks to rapid air transport, specialists can be assembled from every
continent for such committee meetings. We have seen this system at work
in the material covered in Chapters 8 through n. The committees followed
in most detail were those organized by the UN system to define the world’s
nutrition problems and to suggest ways of tackling them. The time period
for each meeting has to be settled in advance so that participants can book
return flights and plan ahead for the following period. The committee’s job,
while it meets, typically no more than a week or 10 days, is to produce a
report that can be published. This usually means writing with haste in
evening sessions and dividing the task among subgroups.
Obviously, senior UN administrators are pleased with a committee that
gets the report completed on time and comes up with recommendations
that can be carried out within an acceptable budget. Only then does the
expenditure on air fares and per diem honoraria seem to be justified. The
chair in turn has a heavy responsibility in trying to discipline a group of 8
to 10 senior people, all used to having their opinions regarded as authori¬
tative in their own institutes and free to talk there as long as they like. It
is understandable that a chair, in looking over the full list of those qualified,
is likely to recommend as committee members people who will probably
not haggle over details and will accept majority opinions after a reasonable
amount of discussion. The negative side of this is, of course, that the com¬
mittee is willing to emphasize subjects on which members can agree, so as
to give positive answers to sponsors. Conversely, an area of disagreement
is deemphasized (“swept under the rug”) and referred to in words that,
without being actually wrong, still leave a vague impression of overall agree¬
ment.

227
 Protein and Energy

Another characteristic of committee members, particularly when more or

less the same group is brought together at fairly frequent intervals, is that
they develop a team loyalty and feel individually committed to the opinions
expressed in their earlier reports. If these reports are criticized, they naturally
want to defend them in an “us-versus-them” situation. Under these circum¬
stances, to defect from the team view is a big decision. We have seen an
example of this in Waterlow’s dramatic article in 1975, which began with
the statement: “It is a most nourishing and stimulating diet to eat one’s
own words.” After chairing one of the first UN committee meetings at which
kwashiorkor was classified as a condition of protein deficiency, and then
taking the chair at many subsequent meetings with the same theme, he was
converted to an opposing view and wrote (with Payne): “The concept of a
worldwide protein gap ... is no longer tenable.”11 One cannot help admiring
someone who can say publicly, on reviewing new evidence, “I was wrong.”
Scrimshaw has said that “the democratic approach to truth is a contra¬
diction in terms.”13 This is true, and the committee system is not perfect,
but as with democracy, we know nothing better. A key point is that minority
voices should not be suppressed as a consequence.

The Conduct of Controversy

1

The question of the reality of the “world protein problem” Was brought to
a head by Donald McLaren in 1974 with the appearance in the Lancet of
his article titled “The Great Protein Fiasco.”14 It then became impossible
to shrug the matter off as a technical detail relating to a slight difference
in emphasis. In that article McLaren directly associated the Protein Advisory
Group with “a long and disastrous train of events.” This was n©t the type
of language one expects to find in a scientific periodical. McLaren also wrote
that, after he had made some criticism of the current concepts at an inter¬
national meeting, several people said privately that they agreed with him
but did not feel free to speak out, fearing that to do so would cause them
to lose future research grants. This is a serious point and we will return to
it later. But, clearly, McLaren had no such concern. Indeed, a mutual friend
has told me that he was using his normal style of speech and correspondence,
and that he could not understand anyone taking it amiss. The decision of
the editor of the Lancet to accept an article written in that style was pre¬
sumably based on respect for the author’s qualifications and experience.
When an eminent Victorian such as Justus Liebig could run his own
journal, he had no compunction at being offensive to people he disagreed

12. Waterlow and Payne (1975).

13. Scrimshaw (1976), p. 203.
14. McLaren (1974).

2.2.8
 In Retrospect

with, even calling them plagiarists or “cocks crowing on a dunghill,” as we

saw in Chapter 4. The traditional language in British science has continued
to be relatively skeptical and direct, with a bite to it. One paper from London
referred to the results of the MIT studies, discussed in Chapter 11, as
“predictable and irrelevant” to the conclusions being drawn from them.
The same group titled another paper “The Winged Bean: Will the Wonder
Crop Be Another Flop?” Particular phrases carry different connotations in
different cultures. However, it is interesting to find in an editorial review
in the American Journal of Physiology: “American science has become
vitiated by too much politeness.... Conciliatory smoothness is the life blood
of diplomacy; it is the death of science. Diplomacy consists of producing
agreement (or more usually the illusion of agreement) under a patina of
vacuous formulae_The growing point of science is discrepancy.”15
The moral perhaps is that we should be as polite as we can be without
endangering the clarity of the point we are trying to make. The scientific
community should also be careful to include in symposia those with dissident
opinions and to ensure that people speaking up will not risk having funding
for subsequent investigations disapproved by their peers as a result.

The Responsibilities of Applied Scientists

In recent years, publicity has been given to a few scientists who deliberately
falsified data in order to claim the credit for findings that not only were
misleading and wasted the time of others, but were just plain dishonest. We
all feel it as a stain on our profession that this sort of thing occurs, even
rarely. Fortunately, there was no such incident in the present history.
A more common problem is that people become so certain of the truth
of a theory or generalization that they select data that fit it and (perhaps
unconsciously) reject observations that do not. I am sure that Mulder, for
example, truly believed in 1838 that he had discovered that there was a
common protein radical present in different albuminoids, as described in
Chapter 3. One can also well imagine, in view of the inherent variability
in organic analyses at the time, that he regarded those values which agreed
best with his preconception as satisfactory, and others that did not as re¬
quiring further replication. In the long run, this resulted in his being dis¬
credited and in little notice being taken of his later work.
Today there is a more subtle temptation for scientists who have to prepare
competitive research proposals in order to obtain funds for their research.
This is to look at their task as compiling an argument for support, rather
as lawyers on one side of a court case build up the case for their client and
omit any contrary evidence on the ground that this is the opposing lawyer s

15. Yates (1983).

229
 Protein and Energy

responsibility. Papers written for publication have to fit with the preexisting
grant application and to act as supporting evidence for the next one, so that
even though what is published is “the truth, and nothing but the truth,” it
may still not be “the whole truth” about a subject, particularly in its promise
of practical application. This kind of thing is not often written about, per¬
haps because people are afraid of giving offense to their peers whose support
they need. But one nutritionist has written: “No scientist is simply involved
in the single-minded pursuit of truth, he is also engaged in the passionate
pursuit of research grants and professional success.”16 This was a coura¬
geous statement by John Rivers, who was then at midcareer and could not
have known that he had only a few more years to live.
Rivers and his colleagues also criticized a paper in which it wds urged
that increased funding should be allocated for the development of the winged
bean (Psophocarpus tetrogonolobus) as a protein-rich food for the tropics.
They commented that the dietaries into which the winged bean was to be
introduced were not limited in their nutritional value by either protein
quality or quantity, and it seemed likely that there would be an overall
adverse economic effect on a family unit that adopted the winged bean.
They regarded the advocacy of this bean as “merely the latest in a tradition
of technological fixes designed to solve the problem of malnutrition... [and]
the outcome of a continuing process of justifying scientific enthusiasms by
the drawing of facile and tenuous links between research which is intellec¬
tually exciting to the investigator and problems which are of sufficient public
concern to make it politically attractive to devote funds to them.”17
Some similar criticisms are to be found in a review by one of the leaders
of the work done with U.S. federal funding to develop the production of
fish protein concentrates: “Much of the motivation for FPC development
had little or nothing to do with the ostensible and well-publicized human¬
itarian goal..., it often served as a convenient mask for the varied objectives
of the different development groups.”18
It is a truism in nutrition that most attention has to be paid to the limiting
factor, as with the identification of the first-limiting amino acid in a protein.
But this principle was not followed in the large body of projects devoted
to the development of “single-cell proteins.” Early proponents of these
materials predicted the quantity of protein that would be produced, by
multiplying the nitrogen content of yeast and/or bacteria by the usual 6.2.5
factor. This neglected the known fact that a considerable proportion of the
nitrogen in cells capable of such rapid reproduction is incorporated into
nucleic acids rather than into protein. This portion of the nitrogen therefore

16. Rivers (1979).

17. Henry, Donachie and Rivers (1985), p. 337.
18. Pariser, Wallerstein, Corkery and Brown (1978).

230
 In Retrospect

represents not an asset but a liability because of the production of uric acid
when nucleic acids are ingested by humans. It would have been logical first
to have investigated whether these components could be reduced to safe
levels by some economical treatment before attempting scaled-up produc¬
tion, but the enthusiasts apparently did not want to raise an issue that might
have delayed, or stopped altogether, the chance of their developing new
fermenters.

The Pros and Cons of Enthusiasm

Poleman, after concluding that FAO’s estimates of world food shortages

from 1946 to 1974 considerably exaggerated the problem, and that the
concept of 500 million people struggling at the brink of starvation was a
myth, went on to consider how this had happened. After referring to the
possible influence of the special interests of an entrenched bureaucracy and
of authors eager to sell books with a dramatic story, he finally took the
more generous view that it “reflected nothing more than the persistence of
honorable men attempting to dramatize their case through exaggeration.”19
Nowadays, to describe people as enthusiastic is to praise them for the
energy they put into an activity. However, an “enthusiast,” when the word
was coined in the eighteenth century, was also someone whose ideas were
out of perspective or balance. This difference brings us back to the points
raised in the preceding section. It was possible for people to be enthusiasts
in both the new and the old sense - they were excited about their project
and worked hard at it while, at the same time, further consideration might
have shown that the program was unlikely to meet the need for which it
was designed.
Perhaps the moral is that good intentions are not enough and that we
have a duty to be as critical in planning a line of attack on a problem as
we will be when actually getting down to work on the selected project. One
difficulty here is the specialization of expertise. Food technologists, for ex¬
ample, may reasonably have said that they could not themselves arrive at
any responsible opinion as to the problems of the Third World, but that
they had read editorial pieces in Science, Nature and other respected journals
and understood that there was a need for new protein foods, and that that
was enough to justify their work on, say, the production of protein isolates
from a particular raw material or the attachment of an essential amino acid
to a protein source in which this compound was deficient. And in a culture
where enthusiasm was honored, and criticism looked down upon as “neg¬
ative,” there was little or no discussion of the value of such work, which

19. Poleman (1977), PP- 386~7-

231
 Protein and Energy

could continue for a decade before the researchers found that there was no
practical use for their product/0
Aaron Altschul, who spent many years in basic studies of the proteins in
oilseeds and later edited a series of volumes entitled New Protein Foods,
warned that some attempts at development in the Third World could lead
to tragedy, because the new technologies required so many fuel calories for
the production of each food calorie. These were commonly not available,
and the promise of greater food availability was an illusion/1
This brings us back to the complexity of things. We know the kinds of
technical problems encountered in our own narrow field, but feel that in
other areas, where we are not aware of pitfalls, they probably do not exist,
or could easily be overcome. As others have said already, it seemed a simple
thing for a protein concentrate that was cheap enough for poultry feeding
to be adapted as a human food. And where there was a “need,” surely that
was virtually the same thing as a “demand.” And so on, and so on. Perhaps
unconsciously scientists also felt that problems of manufacture, distribution
and marketing were relatively trivial once there had been a scientific analysis
of what was lacking in a community’s food supply.
It seems that, in practice, the problems were great and, moreover, that
the original analysis was unbalanced. There are therefore lessons to be
learned at all levels. Let us try to apply them, but also to remember in regard
to those who seem to have been in the wrong that “the ortly certain way
to avoid making a mistake is never to do anything.” We should also have
a special respect for the physicians who faced, day after day, the misery of
suffering and dying children, and the responsibility of immediately doing
the best they could for them.

Some Nutritional Best-Sellers

Just as health food stores sell products, such as amino acids, that orthodox
university department nutritionists and physicians are unlikely to recom¬
mend, so there is an “alternative” nutritional literature. Some of the best¬
selling nutrition books are of this kind. The authors convey an impression
of absolute certainty as to the correctness of their relatively simple advice
and of the almost immediate advantages enjoyed by readers who follow
their recommendations. And the public naturally finds these promises ap¬
pealing.
In many cases these books contain statements that are ludicrous to anyone
with even an elementary knowledge of physiology. For example: “It’s only
undigested food, food stuck in your body that accumulates and becomes

zo. Carpenter (1986b), pp. 1367-8.

2.1. Altschul (1976), pp. 294-5.

232
 In Retrospect

fat,” and “Once you have eaten a protein in the course of a day, there is
no way your body can then digest a carbohydrate,” and “Remember that
milk is a protein; so that little drop in your coffee will make all the car¬
bohydrates that follow it undigestible or fattening.” More than a million
copies of the book containing this extraordinary advice were sold, making
the author a rich woman/1 Clearly she knew nothing, or chose to appear
to know nothing, about the mechanisms of absorption through the wall of
the small intestine that are necessary before nutrients can reach the blood¬
stream and be circulated to the tissues.
Another best-seller explains that if proteins and carbohydrates are eaten
at the same meal, the digestive enzymes nullify one another, since the protein-
digesting enzymes require acidic and those digesting carbohydrate need
alkaline conditions/3 Do the authors really not know that a certain amount
of digestion occurs in the acidic conditions of the stomach but is completed
in the slightly alkaline conditions of the small intestine? The same authors
write that the protein in milk makes its calcium unavailable. Again one
wonders how they would explain the bone growth of suckling mammals,
of whatever species. They also state that the amino acids in egg proteins
are unavailable unless the eggs are eaten raw, which is contrary to all actual
evidence. Unfortunately, publishers bring such books to the market because
they satisfy a demand and yield considerable profit.
Scientists feel more obliged to restrict their claims to those for which there
is evidence - and to admit that, while some things are certain, others are
less so. They offer some of their recommendations, therefore, on the basis
of prudence rather than “proof.” Recommendations made with “ifs” and
“maybes” obviously have a less forceful impact than a straightforward
“Never do this and we guarantee that your skin will recover its youthful
appearance in 3 months.”

The Bottom Line

Having criticized these popular writings, what recommendations can we

offer? Is there a protein problem in the sense that we, and others, are in
danger of eating too much or too little? Given the swings in opinions on
this subject in the past, it is difficult to be certain that we now know the
unalterable truth. However, we do have a much larger body of evidence
than our predecessors did, both from direct experiments with human sub¬
jects and from field observations in many parts of the world with different
dietary traditions. One of the factors that has allowed the human race to

22. Mazel (1981).

23. Diamond and Diamond (1985).

233
 Protein and Energy

flourish and multiply in so many parts of the planet must be our ability to
thrive on many different kinds and combinations of foods.
In the United States 50 g protein per day is officially recommended as an
adequate allowance for a healthy woman of average weight (63 kg, or 138
lb.), as is 63 g for a man weighing 79 kg (or 174 lb.). Although there is
still controversy over adult requirements for individual amino acids, nearly
all the diets that have been studied so far have been found to supply total
protein in excess of the RDA and to meet even the highest estimates of
amino acid requirements. There seems to be no general need for concern
that adults in developed countries are short of protein, or in the other
direction that a large proportion are consuming nearly twice their RDA.
Recently there has been more concern whether some intakes are too high,
and this is a matter of debate. Certainly many people do consume more
than twice the RDA. This is definitely undesirable for people with kidney
disease. As discussed earlier, it also seems prudent for others not to exceed
this upper limit, even though many do so and seem to be unharmed. Here
we enter an area of degrees of risk. Just as we are acquainted with individuals
who smoke cigarettes without apparent harm, we know that the proportion
of their fellow smokers who have suffered from a number of diseases is very
much greater than that of otherwise comparable nonsmokers. The danger
from very high protein consumption is much less clear-cut than that from
smoking, but it still seems reasonable to recommend cautioti when there is
a possible risk of harm and there is no significant disadvantage to keeping
within the limits. Meanwhile, we know that the presently suggested upper
limit of “twice the RDA” is a more or less arbitrary one and may well
change considerably as knowledge accumulates.
Nutritional advice for relatively affluent adults in recent years has con¬
centrated on reducing fat and cholesterol intake and increasing the con¬
sumption of fruit and vegetables. An indirect effect on those following these
recommendations has been that their consumption of animal products has
declined and that these foods are now served more as flavoring agents in
dishes rather than as their major constituents. However, there seems no
reason for people to worry that this will leave them short of protein, pro¬
vided that they eat an otherwise varied diet. Women, who are more at risk
than men from osteoporosis in later life, will, in any case, usually want to
ensure adequate calcium intake by consuming low-fat milk products that
are also good protein sources; they will also need to ensure that they are
still receiving an adequate level of iron if they reduce their consumption of
meat and do not eat legumes.
Growing children need a slightly higher proportion of protein (to calories)
in their diet, but the difference is small because human growth is slower
than that of other species and children’s energy needs are also high. The
first priority is to ensure that they eat enough, and then that their fondness

234
 In Retrospect

for candy and other sweet things does not prevent them from having a
varied, balanced diet. For most children, dairy products again provide a
good and palatable source of protein and other nutrients. The small minority
who react badly to lactose or cow’s milk protein need to have more care
paid to their diet. Meat, of course, is a good protein source, and severe fat
reduction is not recommended for young children.
“Moderation” and “variety” are still sensible keywords. There is no evi¬
dence that we need to go to the extreme of eliminating whole classes of
foods or overemphasizing any one class. Our ancestors’ “peasant diet”
(when things were relatively good) consisted of whole grains, vegetables,
seasonable fruit, and a little meat, fish or cheese to impart flavor. The actual
choice of grain species and of vegetables depended on their location in the
world, but the general pattern reflects long experience and seems difficult
to improve upon.

235
 Appendix A: Chemical Structures of
Amino Acids

R R, R2
I1 1I 1I
CH CH CH
/ \ / \ / \
H—N C— OH -N C—N C-
1 1 1 II
H O H O H O
General formula for an The peptide linkage bettyeen
amino acid (R represents amino acids (with elimination of
a side chain) H—OH) as part of a protein chain
R groups for individual amino acids:

OH ch3 SH
I1 1
1

H CH3 ch2
V
CH—OH ch2
1
1
11 11 |1 11
Glycine Alanine Serine Threonine* Cysteine"

OH nh2 ch3
1I |1 1
OH nh2 c=o c=o s
1
1 I1 I1 11 1
c=o 0=0 ch2 ch2 ch2
1 1I 11 11 1
ch2 ch2 ch2 ch2 ch2
1 11 11 11 1
Aspartic Asparagine Glutamic Glutamine Methionine
acid acid

2.36
 Chemical Structures of Amino Acids

NH2

ch3 ch3 C=NH

1 1 11
ch3 chch3 ch2 nh2 NH
1 1 1 11 1
chch3 ch2 chch3 (CH2)4 (CH2)3
1 1 11 1 1
Valine8' Leucine* Isoleucine* Lysine* Arginine*

H H
OH C C
H 1 / \
C C HC CH
/ \ / \ \ /
HC CH HC CH HC—CH h2 h2
C—NH 1 1 \ C —C
1 1
\ NH / \
HC CH HC CH
CH
// \ / \ / / H2C CH
C C C=C \ / \
C—N
1 1 1 1 \ N C—OH
ch2 ch2 ch2 h 1 II
ch2
1 1 1 H O
Histidine* Phenylalanine* Tyrosine^ Tryptophan* Proline''

Note: Asterisks indicate amino acids that are “essential” (or indispensable) because
they cannot be synthesized by the body.
“Semiessential,” because it can be synthesized from methionine.
* Rates of synthesis limited, especially in infants.
“‘Semiessential,” because it can be synthesized from phenylalanine.
‘'Proline does not quite fit the general formula since the tail of the R group is coupled
to the NH group of the common structure.

2-37
 Appendix B: The Measurement of
Protein Quality

i\s described in chapter 7, it became possible to compare the nu¬

tritional values of different protein preparations after protein-free basal
mixtures had been developed that supplied all the other nutritional needs
of young rats for rapid growth. Osborne and Mendel began their studies
by supplying each test protein at different levels, giving their young rats
unlimited access to the powdered diets and measuring their weight gains
over a 4-week period.
One problem in interpreting the findings was that the diets that supported
faster growth had also typically been eaten in greater amounts. Had the
rats grown more because they had eaten more, or eaten more because they
had grown more? Preliminary trials indicated that rats of a particular size
tended to eat similar amounts of their dry diets regardless of the protein
source; and in further trials with fixed, limited quantities of food for rats
on different diets, growth rates still differed significantly. *
Their final procedure was to allow their rats to eat freely (ad libitum)
and then to calculate the response as grams weight gained per gram protein
eaten, which was later to be called the “protein efficiency ratio” (PER). The
value, as one would expect, depended on the percentage of protein in the
diet. At very low levels there could be no growth, so that the value had to
be zero. Then at very high levels, where more protein was being consumed
than was needed for growth, the ratio was again reduced. In general the
protein level at which the optimal PER was recorded for a particular protein
was higher for poorer proteins - one explanation for this being that more
was required before there was a surplus over the maintenance requirement.
Some of the relevant data from Osborne and Mendel’s paper, published in
1919, is illustrated in Figure B.i. Two protein fractions from cow’s milk -
lactalbumin and casein — were compared. Lactalbumin was much superior
at low levels, but at higher levels the difference almost disappeared.1

1. Osborne, Mendel and Ferry (1919).

Z38
 The Measurement of Protein Quality

Figure B.i. Mean results from Osborne and Mendel’s 1919 paper com¬
paring 4-week protein efficiency ratio (PER) values for casein (□) and
lactalbumin (O) when fed at different levels in the diet. (As explained in
the text, a 12% protein level here is equivalent, in relation to total calories,
to a 9% protein level in the later official procedure. The values also have
high standard errors, as many were mean results from only two or three
animals.)

At first, therefore, Osborne and Mendel recommended that each protein

(or mix) be tested at several levels and that the highest PER value be used
as the measure of its quality. However, in practice, workers later used a
single protein level (in the range of 9 to 10% of the dry weight of each
diet). Some work carried out 25 years later illustrates the general finding
that, although this changed absolute values, it did not usually alter the
ranking of different proteins (or mixes) for supporting growth in young
rats* (Table B.i).
Some 50 years after Osborne and Mendel’s original work, the PER pro¬
cedure became an official method of measuring protein quality in the United
States.3 It was specified that, because of the variability from animal to

2. Bames, Maack, Knights and Burr (1945).

3. Association of Official Agricultural Chemists (1965). PP- 785~6-

2-39
 Appendix B

Table B.i. Protein efficiency ratios of four materials

Test material

Protein Type of Whole Soy Sby Wheat

level feeding egg 1 2 gluten

Optimal level for each material Ad lib 3.8 (100) 2.1 (55) 1.8 (47) 0.8 (21)

10% for all Ad lib 3.8 (100) 2.0 (53) 1.5 (40) 0.3 (8)

10% for all Pair-feda 2.6 (100) 1.8 (69) 1.2 (46) 0.4 (15)

Note: Numbers in parentheses indicate percentage of the value for whole-egg protein.
Although “whole egg” was used as the standard in this trial, casein was preferred
for the official procedure, since it is more stable during storage.
"All rats were given the same weight of food.

animal, io individually caged rats should be used for each test diet. The
level of fat was also reduced to 8%, from the 26% in the early diets. This
meant that a diet containing 12% protein by weight had only 9% of the
total calories in the form of protein in the early work, and in the later official
method the standard 9% by weight also contributed 9% of the energy. This
has to be taken into account if one tries to compare old and new results.
The specifications also laid down that each trial should include a standard
type of casein as one of the test materials. The results with the other test
materials could then be standardized by multiplying each value by a number
that would bring the result for casein to a value of 2.5. It was thought that
this comparison would reduce variability among laboratories* that arose
from rats in some colonies growing more efficiently than in others.
The method was comparatively simple for a biological test, and useful
information was obtained with it, as we shall see later. However, from the
beginning the procedure was criticized on a number of grounds: The results
depended on an arbitrarily fixed protein level; the tissue gained by the
growing rats could vary in its protein content; and no credit was given for
a test protein’s capacity to provide for the maintenance needs of the animals.
Finally, there was no “ideal value” corresponding to a protein that could
be utilized with 100% efficiency in replacing worn-out tissue and building
new tissue.

The Concept of “Biological Value”

Some of the criticisms of the PER test were overcome in a different type of
rat experiment developed by H. H. Mitchell at the University of Illinois’s
College of Agriculture. Mitchell used a concept first applied by Karl Thomas,
 The Measurement of Protein Quality

working in Rubner’s laboratory in Berlin, to results from nitrogen balance

experiments with adult human subjects eating different protein foods.
Thomas defined the biological value of the protein in a food as the per¬
centage of the absorbed nitrogen it contributed that was retained — that is,
converted to body protein and not reexcreted into the urine — under certain
experimental conditions. These were that the protein was not required as
a source of energy and that it was not present in great excess of requirement.4 5
The method also required estimates of how much nitrogen the experimental
subjects would have lost (in feces and urine) if they had received a diet
containing no protein, but adequate levels of all other nutrients. Thomas’s
data are unsatisfactory because his experimental periods were too short to
eliminate the effect of the previous treatment. More rigorous data and a
discussion of the problems in doing this work with human subjects were
presented by another group in 1922.s
The calculations involved are best understood by following an example.
Imagine that an adult subject on a nitrogen-free diet had a mean fecal
nitrogen loss {f) of 1.0 g/day and a urinary nitrogen loss (u) of 4.0 g. Since
there was no nitrogen in the diet, the fecal nitrogen must have come from
cells rubbed off the wall of the gut and various secretions that were not
completely reabsorbed. When the same subject received protein in place of
some of the carbohydrate and/or fat in his previous diet and had a nitrogen
intake (/) of 7.5 g/day, the fecal loss (E) became 1.5 g nitrogen and the
urinary loss (U) 8.0 g nitrogen per day.
Without any dietary protein, the daily balance of incoming and outgoing
nitrogen was o - 1.0 - 4.0, that is, -5.0 g. With dietary protein it was
7.5 - 1.5 - 8.0, that is, - 2.0 g. The dietary protein had therefore spared
the body reserves by 5.0 - 2.0, that is, 3.0 g nitrogen per day, and it was
concluded that the 7.5 g dietary nitrogen had a “net” value of 3.0 g - a
concept comparable to that of a “net profit” calculated after all the accom¬
panying expenses of an operation have been deducted.
The efficiency with which protein has been used can be calculated in terms
of the total protein eaten, in which case it is 3.0 as a percentage of 7.5, that
is, 40%. (This is now termed the “net protein utilization” or NPU, value
of the protein.) Efficiency can alternatively be calculated in relation to the
digested nitrogen. Here, since fecal excretion of nitrogen increased by 0.5
g/day when the protein was fed, the absorbed nitrogen is estimated to have
been 7.0 g/day, and the 3.0 g net protein is 43% of that quantity. The term
“biological value” (BV) is now restricted to calculations made in this way.
For readers who prefer algebraic expressions:

4. Thomas (1909), pp. 238-47.

5. Martin and Robison (1922).

241
 Appendix B

/ - (U - u) - (F - f)
NPU X IOO
I

NPU depends both on how well a protein is digested and on how well
the digested protein is utilized. As one writer has put it, the first factor refers
to wastage in digestion and the second to wastage of protein in metabolism.
Thus,

I-(F-f)
NPU = x BY
I

Mitchell’s Work with Rats

Human studies are slow and expensive. Mitchell, therefore, applied the
procedure to rats, with their small food consumption and potential for rapid
growth and reproduction.6 His protocol was to take five individually caged
rats (usually young ones weighing about 70 g, but sometimes older ones
weighing up to 190 g) and to maintain each of them on the same succession
of six diets for 10 days per diet. The first 3 days of each period were allowed
for adaptation, and the final 7 days were used to measure nitrogen balance.
The diets for periods 2 through 5 usually contained 8% crude protein
(nitrogen content x 6.25) from one of the materials or mixtures being
tested, 10% fat (8 parts butterfat, 2 parts cod-liver oil), 4% of Osborne
and Mendel’s salt mix, 10% sucrose and the remainder starch. In later trials
either 4% of cellulose flour or 2% agar was also included, at tfte expense
of starch, to provide roughage.7 Weighed amounts of food were moistened
and put daily into each cage, and the residue 24 hours later recovered, dried
and weighed. Usually either 8 or 10 g diet were given each day, depending
on the size of the rats at the beginning of each trial. In addition, each rat
received a small daily supplement of a yeast extract that was rich in B
vitamins, but contributed only 2 to 3 mg nitrogen.
It was the normal practice for the first and last diets to be either nitrogen-
free or, in later work, to contain only 4% protein from whole egg, whose
nitrogen was considered to be fully digestible, and at this level to replace
obligatory nitrogen losses with 100% efficiency, so that urinary nitrogen
levels did not increase.8
Then the biological value of the test material in each period could be

6. Mitchell (1924a). V
7. Mitchell and Carman (1926), pp. 192, 197.
8. Ibid., pp. 192-3.

242
 The Measurement of Protein Quality

determined as described earlier. The assumptions were that the fecal nitrogen
not coming from indigestible food was proportional to the weight of food
eaten and that the basal (or endogenous) level of urinary nitrogen was
proportional to the body weight of the rat. The appropriate values could
thus be obtained by adjusting the values obtained in the first and last periods.
Others have expressed doubt as to whether metabolic fecal nitrogen excre¬
tion really increased in proportion to food consumption. Since food con¬
sumption on the low-protein, or nitrogen-free, diet was always low, not
making the adjustment reduces estimated protein digestibility values by
about 3 percentage units.9
Mitchell and his colleagues carried out many tests of the procedure —
varying the order of the treatments, the size of the rats and so on, and
increased the preliminary period on each treatment from 3 days to 4. They
acknowledged that there were “considerable and consistent individual dif¬
ferences among individual rats in the utilization of dietary nitrogen in me¬
tabolism,” but believed that mean values based on five rats in each of two
trials had a satisfactory degree of precision.10
However, when the same materials were fed at levels contributing 5 % of
the protein in one diet and 10% in another, some very different results were
obtained. The biological values for whole milk were 93 and 75%, for corn
72 and 60%, and for potatoes 68 and 67%, respectively.11 Allowing for
maintenance did not therefore always eliminate the effect of “level” even
when the protein supplied was not sufficient to provide for optimal growth.
The proportions of the proteins used for maintenance and growth, respec¬
tively, depended on the level given, and Mitchell explained the higher values
obtained for some proteins at the 5 % level in terms of the rat not having
such critical amino acid requirements for maintenance.
Mitchell published an important review of the complete problem of meas¬
uring the nutritive value of proteins in 1924. He referred to 150 papers
existing on the subject, but wrote: “Many experimental results defy intel¬
ligent interpretation because food intake records are not reported.”IZ He
concluded:
The [determined] biological value of a food cannot be considered an absolute value
... it will depend upon the proportion of the absorbed nitrogen that is used for
growth ... [and] also upon the level of protein feeding-The search for a biological
constant seems a vain one. The best that can be hoped for is the determination of
values representing fairly the comparative worth of different proteins under certain
controlled conditions.13

9. Boas-Fixsen (1935), p. 449-

10. Mitchell and Carman (192.4), pp. 619-20.
11. Mitchell (1924b).
12. Mitchell (1924c), p. 462.
13. Ibid., p.443.

243
 Appendix B

Table B.z. Summary of results obtained with rats by Mitchell’s nitrogen

balance procedure

Biological Net protein

Materials Digestibility" value" utilization
tested (%, mean) (%, mean) (%, mean)

Eight animal foods 98 80 80

(milk, eggs, meat, offals) (95-100) (74-96) —
Protein fractions from animal foods
Casein (from milk) 98 69 68
Gelatin (from bones) 96 25 24 *
Seven legumes and oilseed flours 89 65 58
(84-97) (38-80) —
Seven whole cereal flours 90 66 59
(85-93) (54-78) —

"Numbers in parentheses indicate range.

Source: Block and Mitchell (1946).

Some years later, he seemed to reverse this view, writing, “The biological
values possess an absolute as well as a relative significance.”14 However,
from the context it seems that he merely meant that values for the same
material were fairly well reproducible from one trial to another, when the
conditions were the same.
After 20 years of work, Mitchell summarized the values obtained in his
laboratory and by others using his procedure.15 Some of the values, collected
in groups, are presented in Table B.2. We see that all of the wHble animal
foods tested are of high protein value, and among these whole eggs and
cow’s milk, designed by nature as foods for rapidly growing young, are at
the top of the list. In contrast, fractions obtained from animal foods can be
of lower value. Gelatin, as expected from the findings of early studies, is
particularly poor; and casein, obtained by the “curdling” of milk with acid,
is significantly poorer than the mixed proteins of whole milk. On average,
the vegetable foods tested also have both lower biological values and some¬
what lower digestibility than the whole animal foods. However, the mixtures
of proteins in whole cereals are superior to isolated proteins such as zein
extracted from corn or to the fraction of wheat proteins present in white
flour.

14. Mitchell, Burroughs and Beadle (1936), p. 272.

15. Block and Mitchell (1946), p. 263.

244
 Appendix C: Calculations of
Amino Acid Balance Using an
Isotope Label

This is an illustrative summary of the procedure followed and

results obtained in an experiment carried out by Meguid et al. to study
human requirements for the amino acid leucine.1 As stated in Chapter n,
after the subject had been on the test diet for 7 to 21 days, he received an
intravenous infusion of the amino acid under study with an isotope incor¬
porated into the molecule.
In this experiment, one subject was given 4 hours of intravenous infusion
after 7 days on the test diet. He received 38 mg/hour of leucine in which
the number 1 carbon atom (i.e., the carbon atom in the carboxylic acid
group) in almost all of the leucine molecules was the nonradioactive isotope
I3C. The rate of expiration of COz was measured and portions were analyzed
for their I3C content using a mass spectrometer. At the same time, blood
samples were taken and their content of free leucine analyzed, and again
portions of the leucine were also isolated and their content of I3C analyzed.
The analyses reached stable plateaus toward the end of 3 hours. It was
assumed that by this time the proportion of I3C in free leucine was the same
in tissues where leucine was being oxidized as it was in the blood and that
normal and labeled leucine were treated by the body in identical ways.
Simplifying the calculations a little for this run, in which the subject
continued to receive food, imagine that the plateau rate of expiration of
I3C was 316 |xg/hour more than one would expect the total quantity of
expired COz to obtain. (The carbon of food and tissues naturally includes
a small proportion of I3C, so that it is the “excess” that is calculated.) Other
studies had indicated that only 81 % of the COz produced from the oxidation
of amino acids labeled in this way is recovered in expired breath. Correcting
for this, the total I3C coming from [I3C]leucine is actually 390 p,g, which
corresponds to 390/13 = 30 \xmo\ of I3COz. Then if the plasma leucine
contained 3% excess I3C, the ratio of total leucine to labeled leucine was

1. Meguid et al. (1986).

245
 Appendix C

100:3. Then, since one would expect the labeled and unlabeled leucine to
be oxidized at the same rate, the total quantity of leucine oxidized (with 1
molecule of leucine yielding 1 molecule of COz from its carbon in position
1) is 30 x 100/3 = 1,000 p,mol/hour. Leucine has a molecular weight of
131, so that this corresponds to 131 mg/hour. '
In these studies the subjects, on infusion days, usually received their three
ordinary meals in the 12 daytime hours, divided up into equal hourly por¬
tions for the period of infusion. If their daily diet contained altogether 2,400
mg leucine, each hourly portion contained 200 mg. It could therefore be
concluded that at the time of the study the hourly intake of 200 mg exceeded
the hourly oxidation of 131 mg, by 69 mg, that is, a situation of positive
balance. On the assumption that this hour was typical of the situation during
the 12 daytime hours, the total retention in that period would be 69 x 12
= 828 mg.
However, to obtain a 24-hour balance, one must have an estimate of
nighttime losses when there is no food intake. This was obtained by con¬
tinuing the subject’s overnight fast through the following morning, but
otherwise repeating the same infusion experiment. If this yielded a leucine
oxidation rate of 115 mg/hour, and if it was assumed to be representative
of the 12-hour “night” period, the total oxidation in such a period would
be 1,380 mg. Since this exceeds the positive daytime balance of 828 mg,
the conclusion would be that the subject was losing 552 mg leucine per 24
hours and that to obtain balance a higher daily intake of leucine would be
required than the 2,400 mg provided during this trial.

246
 References

Abderhalden, E. (1905). Abbau und Aufbau der Eiweisskorper im tierischen Or-

ganismus. Z. physiol. Chem. 44:17-52.
(1909). Weiterer Beitrag zur Frage nach der Verwertung von tief abgebautem
Eiweiss im tierischen Organismus. Z. physiol. Chem. 61:194—9.
Abderhalden, E., &C Rona, P. (1905). Uber die Verwertung der Abbauprodukte des
Caseins im tierischen Organismus. Z. physiol. Chem. 44:198—205.
Abderhalden, E., &c Samuely, F. (1905). Beitrag zur Frage nach der Assimilation des
Nahrungseiweiss im tierschen Organismus. Z. physiol. Chem. 46:193—200.
Achar, S. T. (1950). Nutritional dystrophy among children in Madras. Br. Med. J.
1:701-3.
Advisory Committee on the Application of Science and Technology to Development
(1968). International action to avert the impending protein crisis. New York:
United Nations.
Aitken, G. A. (1892). The life and works of John Arbuthnot. London: Russell &
Russell.
Allen, L. H., et al. (1992). The interactive effects of dietary quality on the growth
and attained size of young Mexican children. Am. J. Clin. Nutr. 56:353—64.
Altschul, A. M. (1976). Limits of technology. In A. M. Altschul (Ed.), New protein
foods (Vol. 2, Part B, pp. 280-303). New York: Academic Press.
Altschul, A. M., Lyman, C. M., & Thurber, F. H. ( 1958). Cottonseed meal. In A.
M. Altschul (Ed.), Processed plant protein foodstuffs (pp. 469-534). New York:
Academic Press.
Anderson, A. (1847). On the recent differences of opinion as to the cause of scurvy.
Monthly J. Med. Sci. 8:176-81.
Anonymous (1837). Dietetic charlatanry or new ethics of eating. New York Rev.
1:336-51.
(1846). Reviews: Animal Chemistry by Baron Liebig. London Med. Gaz., n.s.
3:1018—24.
(1874). The late Edward Smith, F.R.S. Br. Med. J. 2:653-4.
(188 6). Food consumption: Quantities, costs and nutrients of food-materials (17th
Annual Report, pp. 237-326). Boston: Massachusetts Bureau of Statistics of
Labor.
(1887a). Kakke or Japanese beri-beri. Br. Med. J. 2:189-90.
(1887b). Robert Christison. Dictionary of national biography (Vol. 4, pp. 290-
1). Oxford: Oxford University Press.

247
 References

(1906). A discussion on over-nutrition and under-nutrition. Br. Med. J. 2:1100—

3-
(1911). Discussion on the merits of a relatively low protein diet. Br. Med. J.
2:656-67.
(1974). PAG’s name changed, scope widened. PAG Bulletin 4(3):!—2.
Arbuthnot, J. (1731). An essay concerning the nature of aliments. London: Tonson.
Armsby, H. P. (1906). The principles of animal nutrition (2d ed). New York: Wiley.
Aronson, N. (1982a). Nutrition as a social problem. Social Problems 29:474-87.
(1982b). Social definitions of entitlement: Food needs, 1885—1920. Media Culture
& Society 4:51—61.
Association of Official Agricultural Chemists (1965). Official methods of analysis
(10th ed.). Washington, DC: Association of Official Agricultural Chemists.
Atwater, W. O. (1883). Report of progress of an investigation of the chemical
composition and economic value offish and invertebrates used for foodi, Report
of the U.S. Commission of Fish and Fisheries for 1880, Part 8, pp. 231—67).
Washington, DC: U.S. Government Printing Office.
(1884). Percentages and costs of nutrients in foods (abstract). Proc. Am. Assoc.
Adv. Sci. (33rd meeting), 648—51.
(1887—8). Pecuniary economy of food. Century Magazine 35:437-46.
(1888). Ueber die Ausniitzung des Fischfleisches im Darmkanale im Vergleich mit
der des Rindfleisches. Z. Biol. 24:16—18.
(1891). American and European dietaries and dietary standards (4th Annual
Report, pp. 106-71). Middletown, CT: Storrs School Agricultural Experiment
Station.
Atwater, W. O., & Benedict, F. G. (1902). Experiments on the metabolism of matter
and energy in the human body (Bulletin No. 109). Washington,-'DC: U.S. Dept,
of Agriculture, Office of Experiment Stations.
Atwater, W. O., &c Bryant, A. P. (1900). Dietary studies of university boat crews
(Bulletin No. 75), Washington, DC: U.S. Dept, of Agriculture, Office of Ex¬
periment Stations.
Atwater, W. O., &c Langworthy, C. F. (1898). A digest of metabolism experiments
(Bulletin No. 5), Washington, DC: U.S. Dept, of Agriculture, Offi<^ of Exper¬
iment Stations.
Atwater, W. O., & Woods, C.D. (1896). Comments on the dietary studies at Purdue
University. Bulletin No. 32. Washington, DC: U.S. Dept, of Agriculture, Office
of Experiment Stations.
(1897). Dietary studies with reference to the food of the Negro in Alabama.
Bulletin No. 38. Washington, DC: U.S. Dept, of Agriculture, Office of Exper¬
iment Stations.
Aulie, R. P. (1970). Boussingault and the nitrogen cycle. Proc. Am. Phil. Soc.
nf435-79-
Austin, J. E., Ed. (1979). Global malnutrition and cereal fortification. Cambridge,
MA: Ballinger.
Autret, M. (1962). Preface to Encouraging the use of protein-rich foods (by J.
Fridthjof). Rome: Food &c Agriculture Organization of the United Nations.
Autret, M., & Behar, M. (1954). Sindrome policarencial infantil (kwashiorkor) and
its prevention in Central America (FAO Nutritional Studies No. 13). Geneva:
United Nations.
Aykroyd, W. R. (1935). Three philosophers (Lavoisier, Priestley and Cavendish).
London: Heinemann.

248
 References

Bailey, C. H. (1941). A translation of Beccari’s lecture “Concerning Grain.” Cereal

Chem. 18:555—61.
Balland, A. (1902). La chimie alimentaire dans 1’oeuvre de Parmentier. Paris: Bail-
liere.
Baly, W. (1843). On the prevention of scurvy in prisoners, pauper lunatic asylums,
etc. London Med. Gaz., n.s. 1:699-703.
Barkas, J. (1975). The vegetable passion. New York: Scribner’s. (The author may
also be catalogued as J. Yager.)
Barker, T. C., Oddy, D. J., &c Yudkin, J. (1970). The dietary surveys of Dr. Edward
Smith, 1862—3. London: Staples Press.
Barnes, R. H., Maack, J. E., Knights, M. J., & Burr, G. O. (1945). Measurement
of the growth-promoting quality of dietary protein. Cereal Chem. 22:273-95.
Beach, E. F. (1961). Beccari of Bologna: The discoverer of vegetable protein. J. Hist.
Med. 16:354-73.
Beaton, G. H., Calloway, D. H., & Waterlow, J. (1979). Protein and energy re¬
quirements: A joint FAO/WHO memorandum. Bull. WHO 57:65-79.
Beaumont, W. (1833). Experiments and observations on the gastric juice, and the
physiology of digestion. Plattsburgh, NY: Allen (rpt. 1959, New York: Dover).
Beccari, J. (1745). De frumento. De Bononiensi Scientarium et Artium Instituto
atque Academia Commentarii (Vol. 2, Part 2, pp. 122-7). Bologna: Laelius.
Benedict, F. (1906). The nutritive requirements of the body. Am. J. Physiol. 16:409-
37-
Berkel, J., &C de Waard, F. (1983). Mortality pattern and life expectancy of Seventh-
Day Adventists in the Netherlands. Internat. J. Epidem. 12:455-9.
Berman, A. (1974). Antoine-Augustin Parmentier. Die. Sci. Biogr. 10:325-6.
Berthollet, C. L. (1785a). Analyse de l’alkali volatil. Mem. Acad. Sci.,316-26.
(1785b). Suite des recherches sur la nature des substances animales et sur leurs
rapports avec les substances vegetales. Mem. Acad. Sci.,331-49.
(1786). Sur l’analyse animale comparee a l’analyse vegetale. Observ. Phys.
28:272-5.
Berzelius, J. J. (1813). Sur la composition des fluides animales. Ann. Chim. 88:26-
72, 113-41.
(1814). Experiments to determine the definite proportions in which the elements
of organic nature are combined. Ann. Phil. 4:323-31, 401-9.
(1842). Jahres-Bericht iiber die Fortschritte der physischen Wissenschaften, 22.
Bischoff, T. L. W., & Voit, C. (i860). Die Gesetze der Erndhrungdes Fleischfressers.
Leipzig: Winter.
Block, R. J, & Bolling, D. (1945). The amino acid composition of proteins and
foods: Analytical methods and results. Springfield, IL: Thomas.
Block, R. J., & Mitchell, H. H. (1946). The correlation of the amino acid composition
of proteins with their nutritive value. Nutr. Abstr. Rev. 16:249—78.
Boas, M. (1956). Acid and alkali in seventeenth century chemistry. Arch. Internat.
Hist. Sci. 9:13-28.
Boas-Fixsen, M. A. (1935). The biological value of the protein in nutrition. Nutr.
Abstr. Rev. 4:447-59.
Bodwell, C. E. (1981). Use of amino acid data to predict protein nutritive value for
adults. In C. E. Bodwell, J. S. Adkins, & D. T. Hopkins (Eds.), Protein quality
in humans (pp. 340—69). Westport, CT: Avi.
Boerhaave, H. (1715). Boerhaave’s aphorisms concerning the knowledge and cure
of diseases (J. Delacoste, Trans.). London: Cowse & Innys.

249
 References

(1719). A method of studying physick (Mr. Samber, Trans.), London: Rivington.

Boorde, A. (1567). A compendyous regyment. London.
Bouchard, C. (1894). Lectures on auto-intoxication in disease (T. Oliver, Trans.)
Philadelphia: Davis.
Bouillon-Lagrange, E. F. B. (1805). Analyse de la glu. Ann. Chim. 56:24-36.
Boussingault, J. B. (1836). Recherches sur la quantite d’azote coiitenue dans les
fourrages, et sur leurs equivalents. Ann. Chim. (ser. 2) 63:225-44.
(1838a). Recherches sur la quantite d’azote contenue dans les fourrages et leurs
equivalents: Deuxieme memoire. Ann. Chim. (Ser. 2) 67:408—21.
(1838b). Recherches chimiques sur la vegetation, entreprises dans le but d’ex-
aminer si les plantes prennent de l’azote a l’atmosphere. Ann. Chim. (Ser. 2)
67:5-54; 69:353-67.
(1839a). Analyses comparees des aliments consommes et des produits rendus par
une vache laitiere: Recherches entreprises dans le but d’examiner si les Snimaux
herbivores empruntent de l’azote a l’atmosphere. Ann. Chim. (Ser. 2) 71:113—
27.
(1839b). Analyses comparees des aliments consommes et des produits rendus par
un cheval soumis a la ration d’entretien: Suite des recherches entreprises dans
le but d’examiner si les herbivores prelevent de l’azote a l’atmosphere. Ann.
Chim. (Ser. 2) 71:128—36.
(1845). Rural economy (G. Law, Trans.) New York: Orange Judd.
Bowie, H. C. (1879). Ueber den Eieweissbedarf eines mitteleren Arbeiters. Z. Biol.
15:459—84.
Boyle, R. (1684). Memoirs for the natural history of humane blood. London: Smith.
Bras, G., Berry, D. M., & Gyorgy, P. (1957). Plants as aetiological factor in veno-
occlusive disease of the liver. Lancet 1:960—2.
Bricker, M., Mitchell, H. H., & Kinsman, G. M. (1945). The protein requirements
of adult human subjects in terms of the protein contained in individual foods
and food combinations. J. Nutr. 30:269—83.
Brock, J. F. (1961). Dietary proteins in relation to man’s health. Proc. Internat.
Congr. Nutr., 1-5.
(1974). Protein requirement. Lancet 2:712—3. ^
Brock, J. F., & Autret, M. (1952). Kwashiorkor in Africa (W.H.O. Monograph
Series No. 8). Geneva: United Nations.
Brock, W. H. (1975). William Prout. Diet. Sci. Biogr. 11:172-4.
Brock, W. H., & Stark, S. (1990). Liebig, Gregory and the British Association,
1837—1842. Ambix 37:134-47.
Broussais, F. J. V. (1826). A treatise on physiology (J. Bell, trans.). Philadelphia:
Carey & Lea.
Brouwer, E. (1952). Gerrit Jan Mulder (1802—1880)./. Nutr. 46:3-11.
Brown, S. C. (1976). Benjamin Thompson (Count Rumford). Diet. Sci. Biogr.
13:350-2.
Brown, T. M. (1968). The mechanical philosophy and the “animal oeconomy.”
Ph.D. dissertation, Princeton University.
Browne, C. A. (1942). Justus von Liebig - man and teacher. In F. R. Moulton (Ed.),
Liebig and after Liebig (pp. 1-9). Washington, DC: American Association for
the Advancement of Science.
Bunge, G. (1902). Textbook of physiological and pathological chemistry (F. A.
Starling, trans.). Philadelphia: Blakiston’s.
Bylebyl, J. (1977). Nutrition, quantification and circulation. Bull. Hist. Med.
35:470-5-

250
 References

Calloway, D. H., 8c Margen, S. (1971). Variation in endogenous nitrogen excretion

and dietary nitrogen utilization as determinants of human protein requirement.
J. Nutr. 101:205—16.
Calloway, D. H., Murphy, S. P., 8c Beaton, G. ft. (1988). Food intake and human
function: A cross-project perspective (Report to US AID). Berkeley: University
of California.
Calloway, D. H., et al. (1992). Village nutrition in Egypt, Kenya and Mexico:
Looking across the CRSP Projects. Washington, DC: U.S. Agency for Inter¬
national Development.
Calloway, D. H., Odell, A. C. F., 8c Margen, S. (1971). Sweat and miscellaneous
nitrogen losses in human balance studies. /. Nutr. 101:775—86.
Calloway, D. H., & Spector, H. (1954). Nitrogen balance as related to caloric and
protein intake in active young men. Am. J. Clin. Nutr. 2:405-11.
Carman, J. A. (1935). A nutritional disease of childhood. Trans. Roy. Soc. Trop.
Med. 28:665—6.
Carpenter, K. J. (1973). Damage to lysine in food processing: Its measurement and
its significance. Nutr. Abstr. Rev. 43:423—51.
(1986a). The history of scurvy and vitamin C. Cambridge: Cambridge University
Press.
(1986b). The history of enthusiasm for protein. J. Nutr. 116:1364—70.
(1990). The history of a controversy over the role of inorganic iron in the treatment
of anemia. J. Nutr. 120:141—7.
(1991). Edward Smith, 1819—1874./. Nutr. 121:1515—21.
(1992). Protein requirements of adults from an evolutionary perspective. Am. J.
Clin. Nutr. 55:913—17.
Carpenter, K. J., Ed. (1981). Pellagra. Stroudsburg, PA: Hutchinson Ross.
Carpenter, K. J., 8c Lewin, W. J. (1985). A re-examination of the composition of
diets associated with pellagra./. Nutr. 115:543—52.
Carson, G. (1957). Cornflake crusade. New York: Rinehart.
Cathcart, E. P. (1912). The physiology of protein metabolism. London: Longman,
Green.
Chambers, W. H. (1952). Max Rubner, 1854-1932. J. Nutr. 48:3-12.
Champagnat, A., Vernet, C., Laine, B., & Filosa, J. (1963). Biosynthesis of protein-
vitamin concentrates from petroleum. Nature, London 197:13-4.
Chan, H., 8c Waterlow, J. (1966). The protein requirement of infants at the age of
about 1 year. Br. J. Nutr. 20:775-82.
Chapman, C. B. (1967). Edward Smith (?i818-1874), physiologist, human ecologist,
reformer. J. Hist. Med. 22:1—26.
Charleton [or Charlton], W. (1659). Natural history of nutrition, life and voluntary
motion. London: Herringham.
Cheyne, G. (1740). An essay on regimen. London: Rivington 8c Leake.
Chibnall, A. C. (1939). Protein metabolism in the plant. New Haven, CT: Yale
University Press.
Chittenden, R. H. (1895). On digestive proteolysis. New Haven, CT: Tuttle, Mo¬
rehouse 8c Taylor.
(1904). Physiological economy in nutrition. New York: Stokes.
(1907). The nutrition of man. New York: Stokes.
(1911). The merits of a relatively low protein diet. Br. Med. J. 2:656—62.
Chorley, R. J., 8c Haggett, P. (1967). Models in geography. London: Methuen.
Christison, R. (1847). Account of an epidemic of scurvy which prevailed in the
general prison at Perth in 1846. Monthly J. Med. Sci. 7:873-91.

251
 References

Clark, J. T. (1959). The philosophy of science and the history of science. In M.

Clagett (Ed.), Critical problems in the history of science (pp. 103-40). Madison:
University of Wisconsin Press.
Clarke, H. T. (1948). The use of isotopes in biology and medicine (pp. 3-15).
Madison: University of Wisconsin Press.
Cohnheim, O. (1901). Die Umwandlung des Eiweiss durch die Darmwand. Z. phys¬
iol. Chem. 33:451-65.
Coleman, W. (1977). Biology in the nineteenth century. Cambridge: Cambridge
University Press. (
Committee on Diet and Health (1989). Diet and health. Washington, DC: National
Academy Press.
Cosman, M. P. (1983). A feast for Aesculapius. Ann. Rev. Nutr. 3:1-33.
Costa, A. B. (1976). Johannes Wislicenus. Diet. Sci. Biogr. 14:454-5.
Cowgill, G. R. (1964). Jean Baptiste Boussingault: A biographical sketch//- Nutr.
84:2.-9.
Craddock, S. (1983). Retired except on demand: The life of Dr. Cicely Williams.
Oxford: Green College.
Cravens, W. W., & Sipos, E. (1958). Soybean oil meal. In A. M. Altschul (Ed.),
Processed plant protein foodstuffs (pp. 353-97). New York: Academic Press.
Crawford, J. (1724). Cursus medicinae, or a complete theory of physic. London:
Taylor & Osborn.
Crosland, M. (1978). Gay-Lussac: Scientist and bourgeois. Cambridge: Cambridge
University Press.
Cullen, W. (1789). A treatise of the Materia Medica. Philadelphia: Crukshank &
Campbell.
Cuthbertson, D. P. (1931). The distribution of nitrogen and sulphur in the urine
during conditions of increased catabolism. Biochem. J. 25:236-44.
(1942). Post-shock metabolic response. Lancet, 1:433-7.
Dally, A. (1968). Cicely: The story of a doctor. London: Gollancz.
Dalton, J. (1808). A new system of chemical philosophy (rpt. 1964, New York:
Philosophical Library).
Darby, W. J. (1973). Cicely D. Williams: Her life and influence. Nutr. Rsv. 31:331—
3-
(1975). Beginnings of PAG. In The PAG compendium (Vol. A, v-xxviii). New
York: Worldmark Press.
Davies, J. N. P. (1948). The essential pathology of kwashiorkor. Lancet 1:317-20.
Davis, A. B. (1973). Circulation, physiology and medical chemistry in England,
1650-1680. Lawrence, KA: Coronado Press.
Davis, K. S. (1966). The cautionary scientists. New York: Putnam.
Dean, R. F. A. (1949). Ein neuer Vorschlag zur Entwicklung der Kinderernahrung.
Monatsschr. Kinderheirlk. 97:471-7 (Abstr. 819, Nutr. Abstr. Rev. 20
[1950:1]).
(1952). The treatment of kwashiorkor with milk and vegetable proteins. Br. Med.
J. 2:791-6.
(1953a). Plant proteins in child feeding (Medical Research Council Rep. Special
Ser. No. 279). London: H.M. Stationery Office.
(1953b). Treatment and prevention of kwashiorkor. Bull. World Health Org.
9:767-83.
(1958). Use of processed plant proteins as human food. In A. A. Altschul (Ed.),
Processed plant protein foodstuffs (pp. 205-47). New York: Academic Press.

252
 References

Debus, A. G. (1978). Man and nature in the Renaissance. Cambridge: Cambridge

University Press.
De Muelenaere, H. J. H. (1969). Development, production, and marketing of high-
protein foods. In M. Milner (Ed.), Protein-enriched cereal foods for world needs
(pp. 2.66—77). St. Paul, MN: American Association of Cereal Chemists.
Descartes, R. (1649). A discourse on a method for the well guiding of reason and
the discovery of truth in the sciences. London: Newcombe.
Diamond, H., and Diamond, M. (1985). Fit for life. New York: Warner.
Dimino, A. (1969). Incaparina in Colombia. In M. Milner (Ed.), Protein-enriched
cereal foods for world needs (pp. 341—3). St. Paul, MN: American Association
of Cereal Chemists.
Dodart, D. (1731). Memoire pour servir a l’histoire des plantes. Mem. Acad. Roy.
Sci. depuis 1666 jusqu’a 1699 4:121-231.
Dombrowski, D. A. (1984). The philosophy of vegetarianism. Amherst: University
of Massachusetts Press.
Donoso, G., Maccioni, A., & Monckeberg, F. (1967). The use of fish flour as sole
source of dietary protein in normal infants. Proc. yth Internat. Congr. Nutr.,
Hamburg 4:572.
Draper, J. C. (1856). Is muscular motion the cause of the production of urea? New
York J. Med. (Ser. 2) 16:147—67.
Drummond, J. C., & Wilbraham, A. (1939). The Englishman’s food. London:
Jonathan Cape.
Dubos, R. J. (1950). Louis Pasteur, free lance of science. Boston: Little, Brown.
Dumas, J. B., & Cahours, A. (1842). Memoire sur les matieres azotees neutres de
l’organisation. C.R. Acad. Sci. Paris 75:976-1000.
Edman, M., Forbes, R. M., &c Johnson, B. C. (1968). Harold Hansen Mitchell: A
biographical sketch. J. Nutr. 96:1—9.
Edwards, W. F., & Balzac, _. (1833). Extrait de recherches sur les proprietes de
la gelatine. Ann. Sci. Nat. 26:318-27.
Eliot, T. (1541). The castel of health. New York: Scholars’, 1937.
Fantini, B. (1983). Chemical and biological classification of proteins. Hist. Phil. Life
Sci. 5:3—32.
FAO Committee on Protein Requirements (1957). Protein requirements (FAO Nu¬
tritional Studies No. 16). Rome: United Nations.
FAO/WHO/UNU (1985). Energy and protein requirements (World Health Orga¬
nization Tech. Rep. Ser. No. 724). Geneva: United Nations.
Fick, A., & Wislicenus, J. (1866). On the origin of muscular power. Phil. Mag.
London (4th ser.) 31:485-503.
Fidanza, F. (1979). Diets and dietary recommendations in ancient Greece and Rome
and the school of Salerno. Progr. Food Nutr. Sci. 3:79-99.
Finlay, M. R. (1991). The rehabilitation of an agricultural chemist: Justus von Liebig
and the seventh edition. Amhix 38:155-67.
(1992). Quackery and cookery: Justus von Liebig’s extract of meat and the theory
of nutrition in the Victorian age. Bull. Hist. Med. 66:404—18.
Fischer, E., & Abderhalden, E. (1903). Uber die Verdauung einiger Eiweisskorper
durch Pankreasfermente. Z. physiol. Chem. 39:81—94.
Fleitman, S. (1986). Walter Charleton (1620-1707): “Virtuoso.” Frankfurt: Peter
Lang.
Flodin, N. W. (1956). The philosophy of amino acid fortification of foods. Cereal
Sci. Today 1:165-70.

253
 References

Florkin, M. (1972). A history of biochemistry. Vol. 30 of Comprehensive biochem¬

istry (M. Florkin 8c E. H. Stotz, Eds.). Amsterdam: Elsevier.
Food and Nutrition Board. (1958). Recommended dietary allowances, revised 1958
(Publ. No. 589). Washington, DC: National Research Council.
Forward, C. W. (1898). Fifty years of food reform: A history of the vegetarian
movement in England. London: Ideal.
Fourcroy, A. F. de (1789a). Recherches pour servir a l’histoire du gaz azote ou de
la mofette, comme principes des matieres animales. Ann. Chim. 1:40—6.
(1789b). Memoire sur l’existence de la matiere albumineuse dans les vegetaux.
Ann. Chim. 3:252-62.
(1792). Axiomes chimiques. In Encyclopedic methodique (Vol. 2, pp. 455—89).
Paris: Panckoucke.
Fowler, O. S. (1871). Physiology, animal and mental (5th ed.). New York: Fowler
8c Wells. *
Frankland, E. (1866a). On the origin of muscular power. Phil. Mag. London (4th
ser.) 32:182-99.
(1866b). On the source of muscular power. Roy. Inst. Proc. 4:661-85.
Fruton, J. S. (1972). Molecules and life. New York: Wiley-Interscience.
(1979). Early theories of protein structure. Ann. N.Y. Acad. Sci. 325:1-15.
Funk, C. (1912). The etiology of deficiency diseases. J. State Med. 20:341—68.
(1913). Studies on beri-beri: 7. Chemistry of the vitamine-fraction from yeast and
rice-polishings. J. Physiol. 46:173—9.
Garrison, F. FI. (1929). An introduction to the history of medicine (4th ed.). Phil¬
adelphia: Saunders.
Garrow, J. S., Smith, R., 8c Ward E. E. (1968). Electrolyte metabolism in severe
infantile malnutrition. Oxford: Pergamon.
Gelbart, N. R. (1971). The intellectual development of Walter Charleton. Ambix
18:149—68.
Gelfand, M. (1946). Kwashiorkor. Clin. Proc.J. Cape Town Post-Grad. Med. Assoc.
5:135—53. (Abstr. 4744, Nutr. Abstr. Rev. 16 [1947].)
Gillman, J., 8c Gillman T. (1951). Perspectives in human malnutrition. New York:
Grune 8c Stratton. ^
(1945). Hepatic damage in infantile pellagra, and its response to vitamin, liver
and dried stomach therapy as determined by repeated liver biopsies. J. Am.
Med. Assoc. 129:12—9.
Glas, E. (1975). The protein theory of G. J. Mulder (1802-1880). Janus 62:289-
308.
Golden, M. H. N. (1988). The role of individual nutrient deficiencies in growth
retardation of children as exemplified by zinc and protein. In J. C. Waterlow
(Ed.), Linear growth retardation in less developed countries (pp. 143—63). New
York: Raven Press.
Golden, M. H. N., Golden, B. E., 8c Jackson, A. A. (1980). Albumin and nutritional
oedema. Lancet 1:114—6.
Goodman, D. C. (1971). The application of chemical criteria to biological classi¬
fication in the eighteenth century. Med. Hist. 15:23-44.
Gopalan, C. (1967). Major nutritional problems of India and Southeast Asia. Proc.
yth Internat. Congr. Nutr., Hamburg 3:320—6.
(1968). Kwashiorkor and marasmus: Evolution and distinguishing features. In R.
A. McCance 8c E. M. Widdowson (Eds.), Calorie deficiencies and protein de¬
ficiencies (pp. 49—58). London: Churchill.

254
 References

Gopalan, C., Sc Patwardhan, V. N. (1951). Some observations on the “nutritional

oedema syndrome.” Ind. J. Med. Sci. 5:312-7.
Gopalan, C., Swaminathan, M. C., Krishna Dumari, V. K., Hanumantha Rao, D.,
Sc Vijayaraghavan, K. (1973). Effect of calorie supplementation on growth of
undernourished children. Am. J. Clin. Nutr. 26:563—6.
Gordon, H. H., Sc Ganzon, A. F. (1959). On the protein allowances for young
infants. J. Pediatr. Sci. 54:503—28.
Graham, G., Cordano, A., Sc Baertl, J. M. (1963). Studies in infantile malnutrition:
2. Effect of protein and calorie intake on weight gain. J. Nutr. 81:249-54.
Graham, G. G., Sc Morales, E. (1963). Studies in infantile malnutrition: 1. Nature
of the problem in Peru. J. Nutr. 79:479—87.
Graham, S. (1839). Lectures on the science of human life (2 vols.). Boston: Marsh,
Capen, Lyon Sc Webb.
Grande encyclopedic (1899). Vol. 25, pp. 1177-8.
Greenstein, J. P., Sc Winitz, M. (1961). Chemistry of the amino acids (3 vols.). New
York: Wiley.
Grmek, M. D. (1974). Francois Magendie. Diet. Sci. Biogr. 9:6—11.
Guerlac, H. (1961). Lavoisier — the crucial year: The background and origin of his
first experiments on combustion in 1772. Ithaca, NY: Cornell University Press.
Guggenheim, K. Y. (1981). Nutrition and nutritional diseases: The evolution of
concepts. Lexington, MA: Heath.
(1991). Rudolf Schoenheimer and the concept of the dynamic state of body con¬
stituents. J. Nutr. iz1:1701—4
Guibourt, M., Sc Depaul, M. (1867). Sur un lait artificiel. Bull. Acad. Med. Paris.
32:803—8.
Hall, T. S. (1969). Ideas of life and matter (2 vols.) Chicago: University of Chicago
Press.
Halle, J. N. (1791). Essai de theorie: Sur l’animalisation et Passimilation des alimens.
Ann. Chim. 11:158-74.
Haller, A. (1754). Dr. Albrecht Haller’s physiology (2 vols.). London: Innys Sc
Richardson.
Hammond, E. G. (1987). Sylvester Graham: America’s first food reformer. Iowa St.
J. Res. 62:43-61.
Hammond, W. A. (1855). The relations existing between urea and uric acid. Am.
J. Med. Sci. 29:119—23.
Hansen, J. D. L. (1968). Features and treatment of kwashiorkor at the Cape. In R.
A. McCance Sc E. M. Widdowson (Eds.), Calorie deficiencies and protein de¬
ficiencies (pp. 33—47). London: Churchill.
(1974). Protein requirement. Lancet 2:713-14.
Hansen, J. D. L., Howe, E. E., Sc Brock, J. F. (1956). Amino-acids and kwashiorkor.
Lancet 2:911—13.
Hanson, L. A., et al. (1988). Antiviral and antibacterial factors in human milk.
In L. A. Hansen (Ed.), Biology of human milk (pp. xx-x). New York: Raven
Press.
Harper, A. E. (1974). Effects of disproportionate amounts of amino acids. In A. E.
Harper Sc D. M. Hegsted (Ed.), Improvement of protein nutrition (pp. 138-
66). Washington, DC: National Academy of Sciences.
Harris, H. F. (1919). Pellagra. New York: Macmillan.
Hegsted, D. M. (1952). A study of the minimum calcium requirements of adult men.
J. Nutr. 46:181-99.

2-55
 References

(1963). Variation in requirements of nutrients: Amino acids. Fed. Proc. 22:1424-

3°.
Hegsted, D. M., Tsongas, A. G., Abbott, D. B., & Stare, F. J. (1946). Protein
requirements of adults./. Lab. Clin. Med. 31:261—84.
Heilbron, J. (1991). The contribution of Bologna to galvanism. Histor. Stud. Phys.
Biol. Sci. 22:57-85.
Helmholtz, H. (1861). On the application of the law of force to organic nature.
Roy. Inst. Proc. 3:347-57.
Helmont, J. B., van. (1662). Oriatrike or physick refined (J. Chandler, Trans.).
London: Loyd.
Henriques, V., &c Hansen, C. (1904). Ueber Eiweisssynthese im Thierkorper. Z.
Physiol. Cbem. 43:417-46.
Henry, C. J. K., Donachie, P. A., &c Rivers, J. P. W. (1985). The winged bean: Will
the wonder crop be another flop? Ecol. Food. Nutr. 16:331—8. *
Hindhede, M. (1913). Protein and nutrition. London: Ewart, Seymour.
Hirschfeld, F. (1887). Untersuchungen iiber den Eiweissbedarf des Menschen. Pfliig-
ers Arch. 4:533-66.
Holemans, K., Lambrecht, A., & Martin, H. (1956). Resultats d’une campagne
d’alimentation supplemental par proteines vegetales (arachides). Mem. Acad.
Sci. colon. Classe Sci. nat. med. (w.s/4 (6).
Holmes, F. L. (1963). Elementary analysis and the origins of physiological chemistry.
Isis 54:50—81.
(1964). introduction to Justus Liebig’s “Animal Chemistry” (pp. vii-cxvi). New
York: Johnson Reprint.
(1971). Analysis by fire and solvent extractions: The metamorphosis of a tradition.
Isis 62:129—48.
(1:973). Justus von Liebig. In Diet. Sci. Biogr. 8:329-50.
(1974). Claude Bernard and animal chemistry. Cambridge, MA: Harvard Uni¬
versity Press.
(1975). The transformation of the science of nutrition. Isis 8:135—44.
(1979). Early theories of protein metabolism. Ann. N.Y. Acad. Sci. 325:171-85.
(1985). Lavoisier and the chemistry of life. Madison: University <jf Wisconsin
Press.
(1987). The intake-output method of quantification in physiology. Hist. Stud.
Phys. Sci. iy:239—70.
Homberg, M. (1702). Essais de chymie. Mem. Acad. Sci., 33—52.
Hopkins, F. G., & Cole, S. W. (1902). A contribution to the chemistry of proteids:
I. A preliminary study of a hitherto undescribed product of tryptic digestion.
J. Physiol. 27:418—28.
Hudson, R. P. (1989). Theory and therapy: Ptosis, stasis and autointoxication. Bull.
Hist. Med. 63:392—413.
Hull, D. L. (1988). Science as a process. Chicago: University of Chicago Press.
Icaza, S. J. (1976). Case histories in mass communications: The Incaparina project.
PAG Bull. 6(i):35~7.
Idler, D. R. (1969). The Halifax isopropanol process for the manufacture of FPC.
In Production of fish protein concentrate (pp. 107—15). New York: United
Nations.
Ihde, A. J. (1964). The development of modern chemistry. New York: Harper &c
Row.
INCAP. (1962). Incaparina. Guatemala City: Institute of Nutrition of Central Amer¬
ica and Panama.

256
 References

Irwin, M. I., &c Hegsted, D. M. (1971a). A conspectus of research on protein

requirements of man. J. Nutr. 101:385—430.
(1971b). A conspectus of research on the amino acid requirements of man. /.
Nutr. 101:539—66.
Jackson, R. W., &c Block, R. J. (1932). The availability of methionine in supple¬
menting a diet deficient in cystine. J. Biol. Chem. 98:465-77.
Jagerroos, B. H. (1902). Ueber die Folgen einer ausreichenden, aber eiweissarmen
Nahrung. Skand. Arch. Physiol. 13:375—418.
Jaya Rao, K. S. (1974). Evolution of kwashiorkor and marasmus. Lancet 1: 709—
11.

Jaya Rao, K. S., Srikantia, S. G., & Gopalan, C. (1968). Plasma cortisol levels in
protein-calorie malnutrition. Arch. Dis. Child. 43:365-7.
Jelliffe, D. B. (1959). Protein—calorie malnutrition in tropical preschool children./.
Pediatr. 54:227-56.
(1967). Approaches to village-level infant feeding. (1) Multimixes as weaning
foods. /. Trop. Pediatr. 13:46-8.
Jelliffe, D. B., &c Jelliffe, E. F. P. (1978). Human milk in the modern world. Oxford:
Oxford University Press.
Jevons, F. R. (1962). Boerhaave’s biochemistry. Med. Hist. 6:343—62.
(1963). Boerhaave’s teaching in relation to Beccari’s identification of gluten as an
animal substance./. Hist. Med. 18:174-5.
Johnston, V. J. (1985). Diet in workhouses and prisons, 1835—1893. New York:
Garland.
Joint FAO/WHO Ad Hoc Expert Committee (1973). Energy and protein require¬
ments (WHO Tech. Rep. Ser. No. 522). Geneva: World Health Organization.
Joint FAO/WHO Expert Committee on Nutrition (1950). Report on the first session
(WHO Tech. Rep. Ser. No. 16). Geneva: World Health Organization.
(1951). Report on the second session (WHO Tech. Rep. Ser. No. 44). Geneva:
World Health Organization.
(1953). Report on the third session (WHO Tech. Rep. Ser. No. 72). Geneva:
World Health Organization.
(1955). Report on the fourth session (WHO Tech. Rep. Ser. No. 97). Geneva:
World Health Organization.
(1958). Fifth report (WHO Tech. Rep. Ser. No. 149). Geneva: World Health
Organization.
(1962). Sixth report (WHO Tech. Rep. Ser. No. 245). Geneva: World Health
Organization.
(1967). Seventh report (WHO Tech. Rep. Ser. No. 377). Geneva: World Health
Organization.
(1971). Eighth report: Food fortification, protein-calorie malnutrition (WHO
Tech. Rep. Ser. No. 477). Geneva: World Health Organization.
(1976). Ninth report: Food and nutrition strategies in national development
(WHO Tech. Rep. Ser. No. 584). Geneva: World Health Organization.
Joint FAO/WHO Expert Group (1965). Protein requirements (WHO Tech. Rep.
Ser. No. 301). Geneva: World Health Organization.
Joint FAO/WHO Informal Gathering of Experts (1975). Energy and protein re¬
quirements. Food. Nutr. i{z):u—19.
Kapoor, S. C. (1971). Jean-Baptiste-Andre Dumas. Diet. Sci. Biogr. 4:242-8.
Keill, J. (1728). Medicina Statica Brittanica (4th ed., J. Quincy, Ed.). London:
Osborn.
Kellogg, J. H. (1877). A household manual. Battle Creek, MI: Good Health Publ.

257
 References

(1893). Ladies guide in health and disease. Battle Creek, MI: Modern Medicine
Ptibl.
(19x1). The new dietetics. Battle Creek, MI: Modern Medicine Publ.
(1922). Autointoxication or intestinal toxemia. Battle Creek, MI: Modern Med¬
icine Publ.
(1923). The natural diet of man. Battle Creek, MI: Modern Medicine Publ.
Kennedy, E. T., El Lozy, M., & Gershoff, S. (1979)- Nutritional need. In J. E. Austin
(Ed.), Global malnutrition and cereal fortification (pp. 15-34)- Cambridge, MA:
Ballinger.
Kertesz, Z. I. (1968). Development, formdlation and testing of protein mixtures:
Past experience. In Protein foods for the Caribbean (pp. 71—2). Kingston: Ca¬
ribbean Food & Nutrition Institute.
Kesselmeyer, J. (1759). Dissertatio Inauguralis Medica de Quorumdam Vegetabi-
lium Principis Nutriente. Strasbourg: Kiirsner. *
Kies, C. V., Shortridge, L., & Reynolds, M. S. (1965)- Effect on nitrogen retention
of men of varying the total dietary nitrogen with essential amino acids kept
constant. J. Nutr. 85:260-4.
Kilbourne, E. M. (1992). Eosinophilia - myalgia syndrome: Coming to grips with
a new illness. Epidemiol. Rev. 14:16—36.
Kohlrausch, O. (1844). Physiologie und Chemie in ihrer geigenseitigen Stellung
beleuchtet durch eine Kritik von Liebigs Thierchemie. Gottingen: Dieterich.
Konig, J. (1876). Der Gehalt der menschlichen Nahrungsmittel an Nahrungstoffen
im Vergleich zu ihren Preisen. Z. Biol. 12:497—512.
Kopple, J. D., & Swendseid, M. E. (1975). Evidence that histidine is an essential
amino acid in normal and uremic men. J. Clin. Invest. 55:881—91.
Kuhn, T. S. (1970). The structure of scientific revolutions (2d ed.‘). Chicago: Uni¬
versity of Chicago Press.
Kutscher, F., & Seemann, J. (1902). Zur Kentniss der Verdauungsvorgange im
Diinndarm, 1. Z. physiol. Chem. 34:528-43.
Ladenburg, A. (1900). Lectures on the history of the development of chemistry since
the time of Lavoisier (L. Dobbin, Trans.) Edinburgh: Alembic Club.
Land, G. (1986). Adventism in America: A history. Grand Rapids, ML Erdmanns.
Landman, J., &c Jackson, A. A. (1980). The role of protein deficiency in the aetiology
of kwashiorkor. West Indies Med. J. 29:229—38.
Langran, M., Moran, B. J., Murphy, J. L., &C Jackson, A. A. (1992). Adaptation to
a diet low in protein: Effect of complex carbohydrate upon urea kinetics in
normal man. Clin. Sci. 82:192—8.
Lantigua, R. A., Amatruda, J. M., Biddle, T. L., Forbes, G. B., & Lockwood, D.
H. (1980). Cardiac arhythmias associated with a liquid protein diet for the
treatment of obesity. N. Engl. J. Med. 303:735-8.
Laskowski, N. (1846). Ueber die Protein theorie. Ann. Chem. Pharm. 58:
129-66.
Lavoisier, A. L. (1778). Rapport sur les bles et farines gates. Republished in Oeuvres
de Lavoisier (Vol. 4, pp. 317—22). Paris: Imprimerie Imperiale, 1868.
(1790). Elements of chemistry (R. Kerr, Trans.). Edinburgh: Creech (rpt 1965,
New York: Dover).
Lawes, J. B. (1853). Agricultural chemistry: Pig feeding. J. Roy. Agric. Soc. Engl.
i4;459-544-
La wes, J. B., & Gilbert, J. H. (1859). Experimental inquiry into the composition of
some of the animals fed and slaughtered as human food. Phil. Trans. Roy. Soc.
London 149: 493-680.

258
 References

Lea, A. S. (1890). A comparative study of artificial and natural digestions./. Physiol.

n:zz6-63.
Lehmann, C. G. (1842). Untersuchungen iiber den menschlichen Harn. J. Prakt.
Chem. 27:257—74.
Leitch, I., Sc Duckworth, J. (1937). The determination of the protein requirements
of man. Nutr. Abstr. Rev. 7:257—67.
Lemery, L. (1721). Quatrieme memoire sur les analyses ordinaires des plantes et des
animaux. Mem. Acad. Sci., 22-44.
Leroux, A., et al. (1814). Rapport sur un travail de M. d’Arcet, ayant pour objet
l’extraction de la gelatine des os, et son application aux differens usages econ-
omiques. Ann. Chim. 92:300—10.
Lewis, J. S. (1986). In the family way. New Brunswick, NJ: Rutgers University Press.
Liebenow, R. C. (1966). Conference introduction. In E. T. Mertz and O. E. Nelson
(Eds.), Proceedings of the High Lysine Corn Conference (pp. 5-7). Washington
DC: Corn Industries Research Foundation.
Liebig, J. (1833). Sur la composition de l’asparamide et de l’acide aspartique. Ann.
Chim. (Ser. z) 53:416-21.
(1840). Organic chemistry in its applications to agriculture and physiology. (L.
Playfair, Trans.) London: Taylor Sc Walton.
(1841). Ueber die stickstoffhaltigen Nahrungsmittel des Pflanzenreichs. Ann.
Chem. Pharm. 39:129—60.
(1842a). Animal chemistry or organic chemistry in its application to physiology
and pathology (W. Gregory, Trans.). Cambridge, MA: Owen.
(1842b). Antwort auf Hrn. Dumas’s Rechtfertigung wegen eines Plagiats. Ann.
Chem. Pharm. 41:351—6.
(1846a). Ueber das Protei'nbioxyd. Ann. Chem. Pharm. 57:129—31.
(1846b). Animal chemistry (3d ed.; W. Gregory, Trans.) London: Taylor Sc Wal¬
ton.
(1847). Researches on the chemistry of food (W. Gregory, Trans.). London: Taylor
Sc Walton.
(1867). Food for infants: A complete substitute for that provided by nature (2d
ed.; E.V. Lersner-Ebersburg, Trans.). London: Walton.
(1870a). The source of muscular power. Pharm. J. Trans. (Ser. 3) 1:161—3, 182—
5, 201—4, 2,21—4.
(1870b). Ueber Gahrung, iiber Quelle de Muskelkraft und Ernahrung. Leipzig:
Winter.
(1870c). The process of nutrition. Pharm. J. Trans. (Ser. 3) 1:261-3, 281—5.
(1872). Liebig’s extract of meat. The Times (London), Oct. 1, p. 6.
(1873). Extract of meat. Pharm. J. Trans. (Ser. 3) 3:561—3.
Liener, I. E., ed. (1980). Toxic constituents of plant foodstuffs (2d ed.). New York:
Academic Press.
Loeser, H. A. (1912). Diet of mine natives. Transvaal Med. J. 7:152-63.
Lund, C. C., Sc Levenson, S. M. (1948). Protein nutrition in surgical patients. In
M. Sahyun (Ed.), Proteins and amino acids in nutrition (pp. 349-63). New
York: Reinhold.
Lunn, P. G., Whitehead, R. G., Hay, R. W., Sc Baker, B. A. (1973). Progressive
changes in serum cortisol, insulin and growth hormone concentrations and their
relationship to the distorted amino acid pattern during the development of
kwashiorkor. Br. J. Nutr. 29:399-422.
Lusk, G. (1922). A history of metabolism. In L. F. Barker (Ed.), Endocrinology and
metabolism (Vol. 3, pp. 3-78). New York: Appleton.

259
 References

Magendie, F. (1816). Sur les proprieties nutritives des substances qui ne contiennent
pas d’azote. Ann. Chim. (Ser. 2) 3:66—77, 408—10.
(1831). An elementary compendium of physiology (E. Milligan, Trans.), pp. 478-
9). Edinburgh: John Carfrae.
(1841). Rapport fait a l’Academie des Sciences au nom de la Commission dite
“de la gelatine.” C.R. Acad. Sci. Paris, 2.37-83.
Manatt, M. W., 6c Garcia P. A. (1992). Nitrogen balance: Concepts and techniques.
In S. Nissen (Ed.), Modern methods in protein nutrition and metabolism (pp.
9-66). San Diego, CA: Academic Press.
Martin, C. J., 6c Robison, R. (1922). The minimum nitrogen expenditure of man
and the biological value of various proteins for human nutrition. Biochem. J.
16:407—47.
Mateles, R. I., 6c Tannenbaum, S. R., Eds. (1968). Single-cell protein. Cambridge,
MA: MIT Press.
Matthews, D. M. (1991). Protein absorption. New York: Wiley-Liss.
Mauron, J. (1980-1). Have single-cell proteins a future? Nestle Research News, 70-
9.
(1986). Food, mood and health: The medieval outlook. Internat. J. Vit. Nutr.
Rev. 56 (suppl. 29)19-26.
Mayer, J. R. (1845). Die organische Bewegung in ihren Zusammenhange mit dem
Stoffwechsel. Heilbronn: Drecksler.
Mayhew, H. J., 6c Binny, J. (1862). The criminal prisons of London and scenes of
prison life (rpt. 1968, London: Cass).
Maynard, L. A. (1962). Wilbur O. Atwater: A biographical sketch. J. Nutr. 78:3-
9* *

Mayow, J. (1674). Medico-physical works (trans. 1908). Chicago: University of

Chicago Press.
Mazel, J. (1981). The Beverly Hills Diet. New York: Macmillan.
McCance, R. A. (1968). The two syndromes. In R. A. McCance 6c E. M. Widdowson
(Eds.), Calorie deficiencies and protein deficiencies (pp. 1-4). London: Chur¬
chill.
McCay, D. (1912). The protein element in nutrition. London: Edward Arnold.
McCollum, E. V. (1957). A history of nutrition. Boston: Houghton Mifflin.
McCosh, F. W. J. (1984). Boussingault, chemist and agriculturalist. Dordrecht:
Reidel.
McCoy, R. H., Meyer, C. E., 6c Rose, W. C. (1935)- Feeding experiments with
mixtures of highly purified amino acids, 8. J. Biol. Chem. 112:283—302.
McKie, D. (1944). Wohler’s “synthetic” urea and the rejection of vitalism: A chem¬
ical legend. Nature 153:608-10.
(1965). Introduction to Lavoisier’s “Elements of Chemistry” (1790). New York:
Dover.
McLaren, D. S. (1966). A fresh look at protein-calorie malnutrition. Lancet 2:485-
8.
(1967). A criticism of the rationale for the development of vegetable food mixtures
for the prevention of protein—calorie malnutrition in young children. Proc. 7th
Internat. Congr. Nutr., Hamburg 3:149.
(1968). Vitamin deficiencies complicating the several forms of protein—calorie
malnutrition, with special reference to vitamin A. In R. A. McCance 6c E. M.
Widdowson (Eds.), Calorie deficiencies and protein deficiencies (pp. 191-9).
London: Churchill.
(1972). Expert committees. Nature 237:119.
 References

(1973). Energy-protein malnutrition. Lancet 1:994.

(1974). The great protein fiasco. Lancet 2:93-6, 107-9.
(1978). Nutrition planning day dreams at the United Nations. Am. J. Clin. Nutr.
31:1x95-9.
Medsger, T. A. (1990). Tryptophan-induced eosinophilia-myalgia syndrome. N.
Engl. J. Med. 322:926—8.
Meguid, M. M., Matthews, D. E., Bier, D. M., Meredith, C. N., Soeldner, J. S., Sc
Young, V. R. (1986). Leucine kinetics at graded leucine intakes in young men.
Am. J. Clin. Nutr. 43:770—80.
Mendel, L. B. (1923). Nutrition: The chemistry of life. New Haven, CT: Yale
University Press.
Mendelsohn, E. (1964). Heat and life. Cambridge, MA: Harvard University
Press.
Mertz, E. T. (1992). Quality protein maize. St. Paul, MN: Am. Assoc. Cereal Chem¬
ists.
Mertz, E. T., Bates, L. S., Sc Nelson, O. E. (1964). Mutant gene that changes protein
composition and increases lysine content of maize endosperm. Science 145:279-
80
Meyer, E. V. (1891). A history of chemistry from the earliest times to the present
day. London: Macmillan.
Michaud, L. (1909). Beitrag zur Kenntnis des physiologischen Eiweissminimum. Z.
physiol. Chem. 59:405-91.
Millward, D. J. (1985). Human protein requirements: The physiological significance
of changes in the rate of whole-body protein turnover. In J. S. Garrow Sc D.
Halliday (Eds.), Substrate and energy metabolism (pp. 135-44). London: Lib-
bey.
Millward, D. J., Price, G. M., Pacy, P. J. H., Sc Halliday, D. (1991). Whole-body
protein and amino acid turnover in man: What can we measure with confidence?
Proc. Nutr. Soc. 50:197—216.
Mitchell, H. H. (1924a). A method of determining the biological value of protein.
J. Biol. Chem. 58:873-903.
(1924b). The biological value of proteins at different levels of intake. /. Biol.
Chem. 58:905-22.
(1924c). The nutritive value of proteins. Physiol. Rev. 4:424-78.
(1942). The metabolism of protein and amino acids. Ann. Rev. Biochem. 11:2.57-
82.
Mitchell, H. H., Sc Block, R. J. (1946). Some relationships between the amino acid
contents of proteins and their nutritive values for the rat. J. Biol. Chem.
163:599—620.
Mitchell, H. H., Burroughs, W., Sc Beadles, J. R. (1936). The significance and
accuracy of biological values of proteins computed from nitrogen metabolism
data. J. Nutr. 11:257-74.
Mitchell, H. H., Sc Carman, G. G. (1924). The biological value for maintenance
and growth of whole wheat, eggs, and pork. J. Biol. Chem. 60:613-20.
(1926). The biological value of the nitrogen of mixtures of patent white flour and
animal foods./. Biol. Chem. 68:183-215.
Money, J. (1985). The destroying angel. Buffalo, NY: Prometheus.
Moo-Young, M., Sc Gregory, K. F., eds. (1986). Microbial biomass proteins. Lon¬
don: Elsevier.
Moore, B. (1898). Chemistry of the digestive processes. In E. A. Schafer (Ed.),
Textbook of physiology (Vol. 1, pp. 312-474). Edinburgh: Pentland.
 References

Mortimer, C. (1745). The natural heat of animals. Phil. Trans. Roy. Soc. 43:473-

Mueller, J. H. (192.3). A new sulfur-containing amino-acid isolated from the hy¬

drolytic products of protein. J. Biol. Chem. 56:57-69.
Mulder, G. J. (1839). Uber die Zusammensetzung einiger thierischen Substanzen.
J. Prakt. Chem. 16:1x9-52.
(1846). Liebig’s question to Mulder (P. F. H. Fromberg, Trans.). London: Black¬
wood.
Muller, M. (1906). Untersuchungen uber die bisher beobachtete eiweisssparende
Wirkung des Asparagins bei der Ernahrung. Arch. ges. Physiol. 112:245-91.
Munday, P. (1990). Social climbing through chemistry: Justus Liebig’s rise from the
Niederer Mittelstand to the Bildungsburgertum. Ambix 37:1-6.
(1991). Liebig’s metamorphosis: From organic chemistry to the chemistry of ag¬
riculture. Ambix 38:135—53.
Munk, I. (1893). Ueber die Folgen einer ausreichenden, aber eiweissarmen Nahrung.
Virchow’s Arch. path. Anat. Physiol. 132:91—157.
Munro, H. N. (1964). Regulation of protein metabolism. In H. N. Munro &C J. B.
Allison (Eds.), Mammalian protein metabolism (Vol. 1, pp. 382—481). New
York: Academic Press.
(1985). Historical perspectives on protein requirements: Objectives for the future.
In K. L. Blaxter & J. C. Waterlow (Eds.), Nutritional adaptation in man (pp.
155-67). London: John Libbey.
Naismith, D. J. (1973). Kwashiorkor in Western Nigeria: A study of traditional
weaning foods, with particular reference to energy and linoleic acid. Br. J. Nutr.
30:567-76.
National Nutrition Research Institute (1959). Food enrichment in South Africa.
Pretoria: South African Council for Scientific and Industrial Research.
National Research Council (1989). Recommended dietary allowances (10th ed.).
Washington, DC: National Academy Press.
Naylor, M. V. (1942). Sylvester Graham, 1794-1851. Ann. Med. Hist. (Ser. 3)
4:236-40.
Needham, D. M. (1971). Machina carnis: The biochemistry of muscular^contr action
in its historical development. Cambridge: Cambridge University Press.
Newton, J. F. (1811). The return to nature or a defence of the vegetable regimen.
London: Cadell & Davies.
Nicol, B. M. (1962). The utilization of protein-rich foods in the prevention of
protein—calorie deficiency diseases. In G. Blix (Ed.), Mild—moderate forms of
protein-calorie malnutrition (pp. 152-8). Uppsala: Swedish Nutrition Foun¬
dation.
(1971). Protein and calorie concentration. Nutr. Rev. 29:83-8.
Nissenbaum, S. (1980). Sex, diet, and debility in Jacksonian America. Westport,
CT: Greenwood Press.
Nitti, F. S. (1896). The food and labour power of nations. Econ. J. 6:30-63.
Nutrition Expert Group (1968). Recommended daily allowances of nutrients and
balanced diets. Hyderabad: Indian Council of Medical Research.
ODA Advisory Committee on Protein. (1975). British aid and the relief of malnu¬
trition (Overseas Development Paper No. 2). London: H.M. Stationery Office.
Oddoye, E. A., & Margen, S. (1979). Nitrogen balance studies in humans: Long¬
term effect of high nitrogen intake on nitrogen accretion. J. Nutr. 109:363—
77-

2 62.
 References

O’Hara May, J. (1977). Elizabethan dyetary of health. Lawrence, KA: Coronado

Press.
Orr, E. (1972). The use of protein-rich foods for the relief of malnutrition in de¬
veloping countries: An analysis of experience. London: Tropical Products In¬
stitute.
(1977). The contribution of new food mixtures to the relief of malnutrition: A
second look. Food Nutr. 3(z):z—10.
Orr, E., & Adair, D. (1967). The production of protein foods and concentrates from
oilseeds. London: Tropical Products Institute.
Orr, J. B. (1937). Food, health and income (2d ed.). London: Macmillan.
Osborne, T. B., &c Jones, D. B. (1910). A consideration of the sources of loss in
analyzing the products of protein hydrolysis. Am. J. Physiol. 26:305-28.
Osborne, T. B., &c Mendel, L. B. (1911). Feeding experiments with isolated food-
substances (Publ. No. 156). Washington, DC: Carnegie Institution of Wash¬
ington.
(1914). Amino-acids in nutrition and growth. J. Biol. Chem. iy:325—49.
Osborne, T. B., Mendel, L. B. & Ferry, E. L. (1919). A method of expressing
numerically the growth-promoting value of proteins. J. Biol. Chem. 37:223—9.
Pagel, W. (1956). Van Helmont’s ideas on gastric digestion and the gastric acid.
Bull. Hist. Med. 30:524-36.
Panel of Experts on the Protein Problem Confronting Developing Countries (1971).
Strategy statement on action to avert the protein crisis in the developing coun¬
tries. New York: United Nations.
Paris, J. A. (1826). A treatise on diet. London: Underwood.
Pariser, E. R., Wallerstein, M. B., Corkery, C. J., &C Brown, N. L. (1978). Fish
protein concentrate: Panacea for protein malnutrition? Cambridge, MA: MIT
Press.
Parkes, E. A. (1871). Further experiments on the effect of diet and exercise on the
elimination of nitrogen. Proc. Roy. Soc. Fondon 19:349—61.
Parmentier, A. (1773). Examen chymique des pommes de terre. Paris: Didet.
Parpia, H. A. B. (1969). Protein foods of India based on cereals, legumes, and
oilseeds. In M. Milner (Ed.), Protein-enriched cereal foods for world needs (pp.
129—39). St. Paul, MN: American Association of Cereal Chemists.
Partington, J. R. (1961-4). A history of chemistry (Vols. 2-4). London: Macmillan.
Paton, D. N. (1895). Muscular energy: The present state of our knowledge in regard
to its source. Edinburgh Med. J. 40:1081—91.
Pereira, J. (1843). A treatise on food and diet. London: Longmans.
Peterson, G. E. (1964). The New England college in the age of the university.
Amherst, MA: Amherst College Press.
Petit, et al. (1831). Rapport des medecins, chirurgiens et pharmaciens de l’Hotel-
Dieu. C.R. Acad. Sci. Paris, 13 (i84i):286-95.
Pettenkoffer, M., & Voit. C. (1866). Untersuchungen iiber den Stoffverbrauch des
normalen Menschen. Z. Biol. 2:459-573.
Pitcairn, A. (1718). The philosophical and mathematical elements of physick. Lon¬
don: Bell & Osborn.
Playfair, L. (1843). Second lecture on the rearing and feeding of cattle. J. Roy. Agric.
Soc. England 4:237—66.
(1853). On the food of man under different conditions of age and employment.
Proc. Roy. Inst. Great Brit. 1:313-16.
Poleman, T. T. (1977). World food: Myth and reality. World Develop. 5:383-94.

263
 References

Poppel, J. (1869). Erfahrungen iiber die Liebig’sche Malzsuppe. Berliner Klin. Woch.
6:494-6.
Prahlad Rao, N., Singh, D., & Swaminathan, M. C. (1969)- Nutritional status of
pre-school children of rural communities near Hyderabad City. Ind. J. Med.
Res. 57:2132-46.
Pretorius, P. J., & Smit, Z. M. (1968). The effects of various skimmed milk formulae
on the diarrhea, nitrogen retention and initiation of cure in kwashiorkor. J.
Trop. Pediatr. 4:50—60.
Protein Advisory Group (1970). Guideline, for preparation of edible cottonseed
protein concentrate. Republished in The PAG compendium (1975) (Vol. Ci,
pp. 1137-40). New York: Worldmark Press.
(1973). PAG statement (no.20) on the “protein problem.” PAG Bull. 3 (1):
4—10.
Proust, L. J. (1802). Essai sur la fecule des plantes vertes. J. Phys. Chim. Hist. nat.
56:97-113.
Prout, W. (1819). Proprietes chimiques et composition de quelques-uns des principes
immediats de l’urine. Ann. Chim. Phys. (Ser. z) 10:369-88.
(1845). Chemistry, meteorology and the function of digestion (3d ed.). London:
Churchill.
Quesnay, F. (1747). Essai physique sur Poeconomie animale (3 vols.). Paris: Cavelier.
Rand, W. M., Uauy, R., &C Scrimshaw, N. S., eds. (1984). Protein-energy require¬
ment studies in developing countries: Results of international research. Food
Nutr. Bull., Suppl No. 10.
Rao, K. S., Swaminathan, M. C., Swarup, S., &c Patwardhan, V. N. (1959). Protein
malnutrition in South India. Bull. World Health Org. 20:603—39.
Rappaport, R. (i960). G.-F. Rouelle: An eighteenth century chemist and teacher.
Chymia 6:68-101.
Ratner, S. (1979). The dynamic state of body proteins. Ann. N. Y. Acad. Sci.
325:189-209.
Reeds, P. J. (1990). Amino acid needs and protein scoring patterns. Proc. Nutr. Soc.
49:489-97.
Rhodes, K. (1957). Two types of liver disease in Jamaican children, 3~»West Ind.
Med. J. 6:145—78.
Ritson, J. (1802). An essay on abstinence from animal food as a moral duty. London:
Richard Phillips.
Ritthausen, H. (1872). Die Eiweisskorper der Getreidarten, Hulsenfriichte und Ol-
samen. Bonn: Max Cohen &c Sohn.
Rivers, J. P. W. (1979). The profession of nutrition: An historical perspective. Proc.
Nutr. Soc. 38:225-31.
Roe, D. A. (1981). William Cumming Rose: A biographical sketch. J. Nutr.
hi: 1313—20.
Roels, O. A. (1972). Fish protein concentrate: History and trends in production. In
Production of fish-protein concentrate (Part 2). New York: United Nations.
Rolleston, H. (1940). Walter Charleton, D.M., F.R.C.P., F.R.S. Bull. Hist. Med.
8:403-16.
Rose, W. C. (1931). Feeding experiments with mixtures of highly purified amino
acid: 1. The inadequacy of diets containing nineteen amino acids./. Biol. Chem.
94’ 155-65.

(1936). The significance of the amino acids in nutrition. In The Harvey Lectures
(Vol. 30, pp. 49-65). Baltimore: Wilkins.
(1949). Amino acid requirements of man. Fed. Proc. 8:546-52.
 References

(1957). The amino acid requirements of adult man. Nutr. Abstr. Rev. 27:631-
47-
Rose, W. C., Oesterling, M. J., &c Womack, M. (1948). Comparative growth on
diets containing ten and nineteen amino acids, with further observations upon
the role of glutamic and aspartic acid. J. Biol. Chem. 176:753—62.
Rose, W. C., & Wixom, R. L. (1955). The amino acid requirements of man: 16.
The role of nitrogen intake. J. Biol. Chem. 2x7:997-1004.
Rosen, E., Buchanan, N., & Hansen, J. D. L. (1974). Evolution of kwashiorkor and
marasmus. Lancet 2:458.
Rosen, G. D. (1958). Groundnuts (peanuts) and groundnut meal. In A. A. Altschul
(Ed.), Processed plant protein foodstuffs (pp. 419-68). New York: Academic
Press.
Rosenheim, T. (1893). Weiterer Untersuchungen iiber die Schadlichkeit eiweissarmer
Nahrung. Pfliiger’s Arch. Ges. Physiol. 54:61—71.
Rossiter, M. W. (1975). The emergence of agricultural science: Justus Liebig and
the Americans. New Haven, CT: Yale University Press.
Rothschuh, K. E. (1971). Adolf Eugen Fick. Diet. Sci. Biogr. 4:614-17.
(I975)- Max Rubner. Diet. Sci. Biogr. 11:585—6.
Rouelle, H. M. (1773a). Experiences. J. Med. Chir. Pharm. 39:250—66.
(1773b). Observations sur les fecules ou parties vertes des plantes et sur la matiere
glutineuse ou vegeto animale. J. Med. Chir. Pharm. 40:59—67.
(1773c). Sur l’urin humain, et sur celles de vache et cheval. /. Med. 40:451—68.
Rubner, M. (1879). Ueber die Ausniitzung einer Nahrungsmittel im Darmcanale
des Menschen. Z. Biol, ij: 115—202.
(1908). Ernahrungsvorgange beim Wachstum des Kindes. Arch. Hyg. 66:81-126.
(1911). Verluste und Wiedererneuerung im Lebensprozess. Arch. Anat. Physiol.,
39-60.
Rumford, B. (1800). Essays, political, economical, and philosophical. London: Cod¬
ell & Davies.
Rush, B. (1812). Medical inquiries and observations upon the diseases of the mind.
Philadelphia: Kimber & Richardson.
Rutishauser, I. H. E., &C Whitehead, R. G. (1972). Energy intake and expenditure
in 1—3 year old Ugandan children living in a rural environment. Br. J. Nutr.
28:145-52.
Salkowski, E. (1880). Weitere Beitrage zur Theorie der Harnstoffbildung. Z. physiol.
Chem. 4:100—32.
Sanchez, L., Calvo, M., & Brock, J. H. (1992). Biological role of lactoferrin. Arch.
Dis. Child. 67:657—61.
Santorio, S. (1614). Insensible perspiration. Translation in J. F. Fulton &c L. G.
Wilson (Eds.), Selected readings in the history of physiology (1966) (pp. 162-
3). Springfield, IL: Thomas.
Schatan, J. (1977)- Termination of the PAG. PAG Bulletin 7 (3—4): 1-20.
Scherer, J. (1841). Chemisch-physiologische Untersuchungen. Ann. Chem. Pharm.
40:1—64.
Schoenheimer, R. (1942). The dynamic state of body constituents. Cambridge, MA:
Harvard University Press.
Schoenheimer, R., Ratner, S., & Rittenberg, D. (1939)- The metabolic activity of
body proteins investigated with /(-)-leucine containing two isotopes. J. Biol.
Chem. 230:703-32.
Scrimshaw, N. S. (1973). The “protein problem.” Nutrition Notes (American In¬
stitute of Nutrition) Sept.:3~4.

26 5
 References

(1976). Shattuck lecture: Strengths and weaknesses of the committee approach.

N. Engl. J. Med. 294:136-42, 198-203.
(1980). The background and history of Incaparina. Food Nutr. Bull. 2:1-2.
Scrimshaw, N. S., Behar, M., Wilson, D., Viteri, F., Arroyave, G., & Bressani, R.
(1961). All-vegetable protein mixtures for human feeding: 5. Clinical trials with
INCAP mixtures 8 and 9 and with corn and beans. Am. J. Clin '. Nutr. 9:196-
205.
Scrimshaw, N. S., &C Dillon, J.-C. (1979). Allergic responses to some single-cell
proteins in human subjects. In S. Garattini, S. Paglialunga, &c N. S. Scrimshaw
(Eds.), Single-cell protein: Safety for human and animal feeding (pp. 171-8).
Oxford: Pergamon.
Scrimshaw, N. S., Hussein, M. A., Murray, E., Rand, W. M., & Young, V. R.
(1972). Protein requirements of man: Variations in obligatory urinary and fecal
nitrogen losses in young men. J. Nutr. 102:1595-604.
Scrimshaw, N. S., & Young, V. R. (1989). Adaptation to low protein and energy
intakes. Human Org. 48:20-30.
Sebrell, W. H., et al., eds. (1961). Progress in meeting protein needs of infants and
preschool children (Publ. No. 843). Washington, DC: National Research Coun¬
cil.
Seguin, A., & Lavoisier, A. L. (1789). Premier memoire sur la respiration des ani-
maux. Mem. Acad. Sci., 566—84.
Sharp, N. A. D. (1935). A note on a nutritional disease of childhood. Trans. Roy.
Soc. Trop. Med. Hyg. 28:411—12.
Shaw, R. L. (1969). Incaparina in Central America. In M. Milner (Ed.), Protein-
enriched cereal foods for world needs (pp. 320-33). St. Paul, MN: American
Association of Cereal Chemists.
Shelley, P. B. (1813). A vindication of natural diet. London: Callow.
Shenstone, W. A. (1901). Justus von Liebig: His life and work. London: Cassell.
Sherman, H. C. (1920). Protein requirement of maintenance in man and the nutritive
efficiency of bread protein. J. Biol. Chem. 41:97-109.
Smeaton, W. A. (1962). Fourcroy: Chemist and revolutionary. Cambridge: Heffer.
Smith, A. H. (1956). Lafayette Benedict Mendel. J. Nutr. 60:3—12. ^
Smith, A. K., & Circle, S. j. (1972). Soybean: Chemistry and technology. Westport,
CT: Avi.
Smith, E. (1857a). On the principles involved in a scheme of prison dietary. Trans.
Nat. Assoc. Promotion Soc. Sci., 293-306.
(1857b). The influence of the labour of the treadwheel over respiration and pul¬
sation. Med. Times Gaz. 14:601—3.
(1859). Experimental inquiries into the chemical and other phenomena of res¬
piration and their modification by various physical agencies. Phil. Trans. Roy.
Soc. London 149:681—714.
(1862). On the elimination of urea and urinary water. Phil. Trans. Roy. Soc.
London 15-7:747—834.
(1864). Report on the food of the poorer laboring classes in England. British
Parliamentary Papers for 1864 28:216—329.
(1874). Foods (3d ed.). London: King.
Smuts, D. B. (1935). The relation between the basal metabolism and the endogenous
nitrogen metabolism, with particular reference to the estimation of the main¬
tenance requirement of protein. J. Nutr. 9:403—33.
Snelders, H. A. M. (1982). The Mulder-Liebig controversy elucidated by their
correspondence. Janus 69:199-221.

266
 References

Snell, E. E. (1979). Lactic acid bacteria and identification of B-vitamins: Some

historical notes, 1937-40. Fed. Proc. 38:2690-3.
Sokolow, J. A. (1983). Eros and modernization: Sylvester Graham, health reform,
and the origins of Victorian sexuality in America. London: Associated Univer¬
sities Press.
Stannus, H. S. (1934). A nutritional disease of childhood associated with a maize
diet and pellagra. Arch. Dis. Child. 9:115-18.
(1935). Kwashiorkor. Lancet 2:1207-8.
(1936). Pellagra and pellagra-like conditions in warm climates. Trop. Dis. Bull.
33:729-41, 815-25, 885-901.
Stroup, A. (1990). A company of scientists: Botany, patronage and community at
the seventeenth-century Parisian Royal Academy of Sciences. Berkeley and Los
Angeles: University of California Press.
Sukhatme, P. V. (1966). The world’s food supplies. J. Roy. Stat. Soc. (Ser. A)
129:222-43.
(1970a). Incidence of protein deficiency in relation to different diets in India. Br.
J. Nutr. 24:477-87.
(1970b). Size and nature of the protein gap. Nutr. Rev. 28:223—6.
Takaki, K. (1885). On the cause and prevention of Kak’ke. Sei-i-kai Med. J. 4:2.9—
37-

Tannenbaum, S. R., & Wang, D. I. C., Eds. (1975). Single-cell protein (Vol. 2).
Cambridge, MA: MIT Press.
Teich, M. (1965). On the historical foundations of modern biochemistry. Clio Med-
ica 1:41—57.
(1992). A documentary history of biochemistry, iyjo—1940. Rutherford, NJ:
Fairleigh Dickinson University Press.
Temkin, O. (i960). Nutrition from classical antiquity to the baroque. In I. Goldston
(Ed.), Human nutrition, historic and scientific (pp. 78—97). New York: Inter¬
national Universities Press.
Thomas, J. L. (1965). Romantic reform in America, 1815—1865. Am. Quart.
17:656—81.
Thomas, K. (1909). Uber die biologische Wertigkeit der Stickstoffsubstanzen in
verschiedenen Nahrungsmitteln. Arch. Physiol. Leipzig, 219-302.
Thouvenel, P. (1806). Melanges d’ histoire naturelle, de physique et de chimie (3
vols.). Paris: Bertrand &C Colnet.
Tomlinson, M. H. (1978). “Not an instrument of punishment”: Prison diet in the
mid-nineteenth century. J. Consum. Stud. Home Econ. 2:15-26.
Torun, B., Pineda, O., Viteri, F. E., & Arroyave, G. (1981). Use of amino acid
composition data to predict nutritive value for children with specific reference
to new estimates of their essential amino acid requirements. In C. E. Bodwell,
N. S. Adkins &c D. T. Hopkins (Eds.), Protein quality in humans (pp. 374-93).
Westport, CT: Avi.
Torun, B., Young, V. R., & Rand, W. D., eds. (1981). Protein-energy requirements of
developing countries: Evaluation of new data. Food Nutr. Bull., Suppl No. 5.
Toulmin, S., & Goodfield, J. (1962). The architecture of matter. New York: Elarper
& Row.
Traube, M. (1861). Ueber die Beziehung der Respiration zur Musckelthatigkeit und
die Bedeutung der Respiration uberhaupt. Arch Path. Anat. 21:386-414.
Tremolieres, J. (1975). A history of dietetics. Prog. Food Nutr. Sci. 1:65-114.
Trowell, H. C. (1940). Infantile pellagra. Trans. Roy. Soc. Trop. Med. Hyg. 33:389-
404.
 References

(1944). Malnutrition in Bantu of Central Africa: Syndrome of malignant mal¬

nutrition. Clin. Proc. 3:381-401.
(1946). The kwashiorkor syndrome. Trans. Roy. Soc. Trop. Med. Hyg. 39:272.
(1949). Malignant malnutrition. Trans. Roy. Soc. Trop. Med. Hyg. 42:417—33.
(1982). The beginning of the kwashiorkor story in Africa. Preface to.H. C. Trowell,
J. N. P. Davies & R. F. A. Dean, Kwashiorkor (pp. xxi-xxviii). New York:
Academic Press.
Trowell, H. C., Davies, J. N. P., Sc Dean, R. F. A. (1954). Kwashiorkor. London:
Edward Arnold (republished with additional prefaces in 1982, New York:
Academic Press).
Trowell, H. C., & Muwazi, E. M. K. (1945). A contribution to the study of mal¬
nutrition in Central Africa: A syndrome of malignant malnutrition. Trans. Roy.
Soc. Trop. Med. Hyg. 39:229-43.
Underhill, F. P., & Mendel, L. B. (1928). A dietary deficiency canine disease: Further
experiments. Am. J. Physiol. 83:589-633.
United Nations (1975). Report of the World Food Conference, Rome, 1974. New
York: United Nations..
U.S. Department of Agriculture (1983). Nationwide Food Consumption Survey,
1977-78 (Report No. H-6). Washington, DC: U.S. Government Printing Office.
(1985). Nationwide Food Consumption Survey. 1977—78, (Report No. H-n).
Washington, DC: U.S. Government Printing Office.
Vaghefi, S. B., Makdani, D. D., &c Mickelsen, O. (1974). Lysine supplementation
of wheat proteins: A review. Am. J. Clin. Nutr. 27:1231-43.
Van-Bochaute, _ (1786). Memoire sur l’origine et de la nature de la substance
animale. Obs. physique 28:109-15.
Vauquelin, L. N., &c Fourcroy, A. N. (1806). Memoir upon the germination and
fermentation of grains and farinaceous substances. Phil. Mag. 25:176—82.
Venel, G.-F. (1755). Essai sur l’analyse des vegetaux: Premier memoire. Mem. Acad.
Sci. 2 319—32.
Vickery, H. B. (1945). Russell Henry Chittenden, 1865-1943. Nat. Acad. Sci. Biogr.
Mem. 24:59-104.
(1950). The origin of the word protein. Yale J. Biol. Med. 22:387-9^.
(1956). Thomas Burr Osborne. /. Nutr. 59:3—26.
Vickery, H. B., & Osborne, T. B. (1928). A review of hypotheses of the structure
of proteins. Physiol. Rev. 8:393—446.
Vickery, H. B., & Schmidt, C. L. A. (1931). The history of the discovery of the
amino acids. Chem. Rev. 9:169-318.
Voit, C. (i860). Untersuchungen iiber den Einfluss des Kochsalzes, des Kaffee’s und
der Muskelbewegungen auf den Stoffwechsel. Munich: Cotta.
(1867). Der Eiweissumsatz bei Ernahrung mit reinem Fleisch. Z. Biol. 3:1-85.
(1869). Ueber den Eiweissumsatz bei Zufuhr von Eiweiss und Fett, und fiber die
Bedeutung des Fettes ffir die Ernahrung. Z. Biol. 5:329—68.
(1870). Ueber der Lehre von der Quelle der Muskelkraft und einiger Theile der
Ernahrung seit 25 Jahren. Z. Biol. 6:303-401.
(1876). Ueber die Kost in offentlichen Anstalten. Z. Biol. 12:1-59
(1881). Physiologie des allgemeines Stoffwechsels und der Ernahrung. In L. Her¬
mann (Ed.), Handbuch der Physiologie (Vol. 6, Part 1, pp. 1-575). Leipzig:
Vogel.
(1889). Ueber die Kost eines Vegetariers. Z. Biol. 25:232-88.
Waslien, C. I., Calloway, D. H., & Margen, S. (1969). Human intolerance to bacteria
as food. Nature 221:84—5.

268
 References

Waterlow, J. C. (1948a). Comment on kwashiorkor. Trop. Dis. Bull. 45:

637-8.
(1948b). Fatty liver disease in infants in the British West Indies (Medical Research
Council Spec. Rep. Ser. No. 263). London: H.M. Stationery Office.
(1961). The rate of recovery of malnourished infants in relation to the protein
and calorie levels of the diet. J. Trop. Fed. 7:16—22.
(1974). Evolution of kwashiorkor and marasmus. Lancet 2:712.
(1984). Kwashiorkor revisited: The pathogenesis of oedema in kwashiorkor and
its significance. Trans. Roy. Soc. Trop. Med. Hyg. 78:436—41.
(1985). What do we mean by adaptation? In K. L. Blaxter 8c J. C. Waterlow
(Eds.), Nutritional adaptation in man (pp. 1—10). London: John Libbey.
(1990). Protein requirements of infants: an operational assessment. Proc. Nutr.
Soc. 49:499—506.
(1992). Protein energy malnutrition. London: Edward Arnold.
Waterlow, J.C., ed. (1955). Protein malnutrition (FAO/WHO/FAO Nutrition Meet¬
ings Rep. No. 12). Geneva: United Nations.
Waterlow, J. C., 8c Payne, P. R. (1975). The protein gap. Nature, London 258:113—
17-
Waterlow, J. C., 8c Stephen, J. M. L., Eds. (1957). Human protein requirements
and their fulfilment in practice (FAO Nutritional Meeting Rep. Ser. No. 12).
Rome: United Nations.
Waterlow, J. C., 8c Vergara, A. (1956). Protein malnutrition in Brazil. (FAO Nu¬
tritional Studies No. 14). Rome: United Nations.
Waterlow, J. C., Sc Wills, V. G. (i960). Balance studies in malnourished Jamaican
infants: 1. Absorption and retention of nitrogen and phosphorus. Br. J. Nutr.
14:183—98.
Welbourn, H. (1955). The danger period during weaning. J. Trop. Fed. 1:34-46,
98—111, 161—73.
Weller, L. A., Calloway, D. H., & Margen, S. (1971). Nitrogen balance of men fed
amino acid mixtures based on Rose’s requirements, egg white protein and serum
free amino acid patterns. J. Nutr. 101:1499-508.
Wharton, B. A. (1968). Difficulties in the initial treatment of kwashiorkor. In R. A.
McCance Sc E. M. Widdowson (Eds.), Calorie deficiencies and protein defi¬
ciencies (pp. 147—53). London: Churchill.
Wharton, B. A., Howells, G. R., Sc McCance, R. A. (1967). Cardiac failure in
kwashiorkor. Lancet 2:384-7.
Whitehead, R. G. (1969). Factors which may affect the biochemical response to
protein-calorie malnutrition. In A. V. Muralt (Ed.), Protein-calorie malnutri¬
tion. (pp. 38-47). Berlin:Springer.
(1973). The protein needs of malnourished children. In J. W. G. Porter Sc B. A.
Rolls (Eds.), Proteins in human nutrition (pp. 103-17). London: Academic
Press.
Whitehead, R. G., Sc Alleyne, G. A. O. (1972). Pathophysiological factors of im¬
portance in protein—calorie malnutrition. Br. Med. Bull. 28:72—8.
Whitehead, R. G., Coward, W. A., Lunn, P. G., Sc Rutishauser, I. (1977). A com¬
parison of the pathogenesis of protein-energy malnutrition in Uganda and the
Gambia. Trans. Roy. Soc. Trop. Med. Hyg. 71:189—98.
Whorton, J. C. (1982). Crusaders for fitness: The history of American health re¬
formers. Princeton, NJ: Princeton University Press.
Widdowson, E. M., Sc McCance, R. A. (1954). Studies on the nutritive value of
bread and on the effect of variations in the extraction rate of flour on the growth

269
 References

of undernourished children (Medical Research Council Special Rep. Ser. No.

Z87). London: H.M. Stationery Office.
Willcock, E. G., St Hopkins, F. G. (1906). The importance of individual amino-
acids in metabolism./. Physiol. 35:88—102.
Williams, C. D. (1933). A nutritional disease of childhood associated with a maize
diet. Arch. Dis Child. 8:4Z3-33.
(i935)- Kwashiorkor, a nutritional disease of children associated with a maize
diet. Lancet 2:1151—z.
(1938). Child health in Gold Coast. Lancet 1:97-102.
(1940). What is pellagra in children? Trans. Roy. Soc. Trop. Med. Hyg. 34:85-
9°- . ,
(1973). The story of kwashiorkor. Nutr. Rev. 31:334—40.
Williams, R. R. (1961). Toward the conquest of beriberi. Cambridge, MA: Harvard
University Press.
Willis, T. (1684). A medical-philosophical discourse on fermentation. London:
Dring, Harper St Leigh.
Wise, R. P. (1980). The case of Incaparina in Guatemala. Food Nutr. Bull. 2:3-8.
Wohler, F. (1828). On the artificial production of urea. Quart. J. Sci. Art. 25:491-
z.
Yates, F. E. (1983). Contribution of statistics to the ethics of science. Am. J. Physiol.
244.^3—5.
Young, V. R. (1986). Nutritional balance studies: Indicators of human requirement
or of adaptive mechanisms. J. Nutr. 116:700—3.
Young, V. R., Bier, D. M., St Pellet, P. L. (1989)- A theoretical basis for increasing
current estimates of the amino acid requirements in adult man, with experi¬
mental support. Am. J. Clin. Nutr. 50:80—92.
Young, V. R., Gucalp, C., Rand, W. M., Matthews, D. E., St Bier, D. M. (1987).
Leucine kinetics during three weeks at submaintenance-to-maintenance intakes
of leucine in men: Adaptation and accommodation. Hum. Nutr. Clin. Nutr.
4jC:i-i8.
Young, V. R., 8t Pellet, P. L. (1988). How to evaluate dietary protein. In C. A.
Barth St E. Schlimmer (Eds.), Milk proteins: Nutritional, clinical^functional
and technological aspects (pp. 7-36). New York: Springer.
Young, V. R., Puig, M., Queiroz, E., Scrimshaw, N. S., St Rand, W. M. (1984).
Evaluation of the protein quality of an isolated soy protein in young men:
Relative nitrogen requirements and effect of methionine supplementation. Am.
J. Clin. Nutr. 39:16-24.
Zello, G. A., Pencharz, P. B., St Ball, R. O. (1993). Dietary lysine requirement of
young adult males determined by oxidation of L-[i-I3C]phenylalanine. Am. J.
Physiol. 264--E677-85.
Zezulka, A. Y., 8t Calloway, D. H. (1976). Nitrogen retention in men fed varying
levels of amino acids from soy protein with or without added L-methionine. J.
Nutr. 106: ziz-zx.
Zuntz, N. (1897). The metabolism of nutrients in the animal body and the source
of muscular energy. USDA Exp. Station Rec. 7:538—50.
 Index

Abbreviations: FAO, Food and Agriculture Organization of the United Nations; FPC, fish pro¬
tein concentrate; FI, hydrogen; INCAP, Institute of Nutrition for Central America and Panama;
N, nitrogen; PAG, Protein Advisory Group; PCM, protein-calorie malnutrition; RDA, recom¬
mended dietary allowance; UN, United Nations Organization; UNICEF, United Nations Inter¬
national Children’s Emergency Fund; WHO, World Health Organization of the United Nations

Abderhalden, E., 128 amino acids: acid digestion of protein to

accommodation (adaptation at the expense obtain, 124; chemical structures, 236—7;
of function), 215 effect of adding to the diet, 126; imbal¬
acescent foods, 8, 9, 12, 23, 86-7 ance effects, 204; in intestinal juice,
acid-base balance, 8,9; value of vegetables, 125-6, 223; methods of analysis, 133—4,
12 204; nutritional value, 125-6; replace¬
acridity, as a cause of disease, 9 ment of protein by, in treating kwa¬
activity, physical: and carbonic acid produc¬ shiorkor, 158-9; use by “muscle”
tion, 20, 51, 63, 72.; and protein needs in builders, 207-8; see also lysine; methio¬
proportion to, 51, 62; and urea produc¬ nine; tryptophan(e)
tion, 59, 63, 70, 72 amino acids, essential: adult needs for, 205;
adaptation, to low protein intakes, 93-4, discovery of threonine, 132; for humans,
214-15 135-6, 204-6, 211-12; levels in poor
aflatoxins, 170 diets, 194; marketing of, as dietary sup¬
albumin: analyses, 55-6, 73; in eggs an^ plements, 162; for rats, 131—3
blood, 23; modification to form casein ammonia: chemical composition, 22, 27,
and fibrin, 24, 49, 57; absence in acescent 40—1; production from “animal sub¬
plants, 23 stance,” 23; value of salts as fertilizers,
albumin, serum: chemical formula, 43, 73; 36, 39; see also volatile alkali
levels in kwashiorkor, 148, 181, 186—7; analytical methods: for amino acids, 133-4,
major component of serum, 49 204; Boussingault’s problem, 38; proxi¬
albumin, vegetable, 23, 33, 49; antiscorbu¬ mate system, 100—1; use of distillation,
tic, 61 11

albuminoids, 56 Animal Chemistry (Liebig): first edition

alkalescent foods, 8, 9; responsibility for in¬ (1842), 48, 55, 56, 225; 1846 edition,
testinal putrefaction, 86-7 56—7; see also Liebig, Justus von
Altschul, A. M., 232 animal glue, 10, 12; see also animal sub¬
amino acid turnover, 137-8; as indicator of stance
nutritional status, 212-14, 2.45-6, 223; animal heat, source of: combustion, 3, 16,
variation with protein intake, 212 20; fermentation, 7; friction, 5, 7

271
 Index

animal kingdom, lack of major chemical 123, 220; prevention by traces of thia¬
synthesis within, 38-9, 53-4, 113 mine, 123
animal substance, 10, 15, 22, 33, 219; and Bernard, Claude: discovery of blood sugar,
nutritional value of food, 33; isolation 225; as early critic of balance studies,
from green plants, 15; origin in vegetable
139
kingdom, 24—5 Berthollet, Claude: composition of ammo¬
animalization (of digested plant foods), 9, nia, 22; nitrogen in animal substance, 23,
24—6; special “power” of herbivores, 12 2-5
anorexia (reduced appetite): following infec¬ Berzelius, J. J, 24, 41, 43, 45-6, 47
tions, 181; of kwashiorkor patients, 154, biological value (of proteins), 136, 240—4
180, 181, 187, 222; in Jamaican “sugar Bischoff, Theodor, 69, 73
babies,” 156, 181, 226 blood: complexity, 224—5; “dissimilar”
appetite loss, see anorexia parts, 12
applied scientists, responsibilities of, 229—31 blood plasma: conversion to muscle, 49; de¬
Arbuthnot, John, 7 tection of sugar, 225; two principal com¬
Aronson, Naomi, 104 ponents, 49; see also albumin, serum
athletes: high protein consumption of, 106; Boerhaave, Hermann, 5, 9
study of metabolism in, 112 Boussingault, Jean Baptiste, 3 2—3; challenge
atomic theory, 40—1, 59 to Liebig, 48, 53; criticism of, by Lawes
Atwater, W. O., 100—7; dietary standards, and Gilbert, 60; N balance trials, 37—8;
105, no, 112; dietary surveys, 103, 105— N fixation by plants, 36—7
6; digestibility study, 101—2; economic Bowie, H. C., 89-91, 93
value of foods, 101, 104; recommenda¬ Boyle, Robert, 8
tion of high protein intakes, 106, 113, bread, see wheat bread
215; tables of food composition, 100—1, breakfast cereals, 87
108 breast feeding: advantages’of, 202; composi¬
autointoxication: and constipation, 87; and tion of milk, 117; decline of, and maras¬
high-protein diets, 86-7 mus epidemic, 184, 198; need to prolong
Autret, M.: as FAO representative investi¬ in absence of animal milks, 155, 182
gating kwashiorkor, 149—53; milk pro¬ breast milk, see milk, human
duction as a priority, 158; protein British Medical Association, 117
deficiency as the world’s most serious Brock, J. E.: on incidence of marasmus vs.
problem, 160 kwashiorkor, 199; “protein decade,”
160; treatment of kwashiorkor with
amino acids, 158-9; as WHO representa¬
bacteria: in ruminant digestion, 127; as tive investigating kwashiorkor, 149—52
source of food, 162; toxicity problems, Broussais, F. J. V., 81
86, 177; use in vitamin and amino acid bulk, dietary: limiting catch-up growth,
assay, 134 192, 197; in weanling diets, 180-1, 183,
beans: Beccari’s failure to detect gluten in, 196
11; fixation of atmospheric N, 36; high Bureau of Commercial Fisheries, FPC pro¬
N content, 34-5; in prevention of kwa¬ duction, 165—8
shiorkor, 157; see also soybeans burns, as cause of severe N losses, 136—7
Beaumont, William, 82
Beccari, Jacopo: discovery of gluten in Cahours, A., 38—9, 44
wheat, 10-11, 14, 23, 219; questioning Calloway, D. H., 208—9
of his views, 26 calorie, 66, 150
Behar, Moise, 152-3 calorie requirements, see energy require¬
Benedict, Francis, 113, 116, 120 ments
beri-beri: vs. kwashiorkor syndrome, 146; calorimeters: bomb, 68; direct, 107; indi¬
prevention by high-protein foods, 120, rect, 71

272
 Index

carbohydrates: in treatment of kwashiorkor, cow’s milk, see milk, cow’s

181; reduction of protein need by, 96 creatine: failure to confirm value of, 128;
casein, 15, 25, 56-7; comparison with Liebig’s belief in nutritional value of,
whole-milk protein in rats, 238—9; as in¬ 74
gredient of “health food” supplements, “crude protein”: analysis for, 100-1
206; as modified form of albumin, 24; cycles of nature, 3 8-9
see also vegetable casein Cullen, William, 26
cassava: association with kwashiorkor, 159;
bulk, 183; N content, 35, 151; undesira¬ Dalton, John, 40
bility as a staple food, 155 Da Vinci, Leonardo, 77
catch-up growth, nutritional needs for, Dean, R. F. A., 153-4, x56
183-4, 192, 193, 197 de le Boe, Francois, 9
Charl(e)ton, Walter, 3-6, 11 de Muelenaere, H., 172
chemical score (of proteins): for adults, Descartes, Rene, 3
193; for rats, 134; relation to biological diarrhea, 153, 198; see also infections
value, 135; relevance to humans, 190, dietary surveys: low-income U.K. families
194, 206 (1863), 62—3; New England workers
Chittenden, Russell, 107-18; confirmation (1886), 102; U.S. regions (1896), 105-6
of results, 140; criticism of results, 115- digestion: as a change in state of cohesion,
17; digestion of protein, 126; improve¬ 12, 24-5, 49, 124; effect of fermentation
ment of health by eating less, 108; low- on, 7-8; effect of heat and air pressure
protein trials with volunteers, 108—12; on, 6; efficiency of, in cold-blooded spe¬
studies with dogs, 119—20; “unknown cies, 7; radical changes, 223; see also
nutrients” concept, 119-20, 221 protein, digestion of
Christison, R., 61 disease; morbid excitement as a cause of, 81
circulating protein; Voit’s concept of, 91-2, distrofia pluricarencial (Brazilian name for
215, 223 condition considered identical to kwa¬
cirrhosis, in adults; incidence ip Brazil, 156; shiorkor), 155-6
relation to earlier kwashiorkor, 149-50, dogs: as experimental animals, 28—32, 91—
156-7 2, 102, 123, 125, 127, 143; need for con¬
cobalamin (vitamin B,*), 122-3, ZI7 siderate care of, 119—20
combustion: eighteenth-century studies, 17, Draper, John C., 59
18; source of animal heat, 20 Dumas, Jean Baptiste, 33, 38—9, 44; chal¬
committees aiming for scientific consensus: lenge to Liebig, 48, 53; trivial changes in
227—8; criticism of, 192, 228 protein on digestion, 125
complementation, between proteins, 131
conservation of energy: Joule’s factor, 66; economic index value of foods, 101
Mayer’s contribution, 52, 66 edema: as characteristic of kwashiorkor,
controversy, conduct of, 56, 228-9 143, 146, 148; and protein deficiency,
corn: associated with kwashiorkor in W. 146, 201
Africa, 143, 180; biological value of pro¬ egg proteins: albumin as protein standard
tein, 243; in treatment of kwashiorkor for human trials, 208; efficiency of use by

patients, 153—4 humans, 190; quantity needed by hu¬

corn, opaque-2, 177 mans, 191—2, 193, 210—11; as standard
corn flakes, as a nonputrefying food, 87 for chemical scoring, 134
cortisol (hydrocortisone), possible role in elements: Lavoisier’s “simple substances,”
kwashiorkor, 187—8 19; inconvertibility, 26, 27
cottonseed: gossypol problem, 174; as a endogenous losses (of N on an N-free diet),
source of dietary protein, 161, 173-5; use 190, 209-10; relation to protein needs,
in Incaparina, 174-5 190
Cowherd, Rev. William, 70—1 energy: comparison with strength, 114; in-

2-73
 Index

takes in N balance trials, 211—12, 214; in vent extraction, 164-5; Third World pro¬
kwashiorkor diets, 156, 183, 195—7; law duction, 179
of conservation of, 52—3, 66; as mechani¬ flame, vital, 3
cal equivalent of heat, 66-7, 76 Fletcher, Horace, 108
energy requirements: of adults, 45, 62, 103, Food and Agriculture Organization (FAO),
hi, 115-16, 191; of children, 152, 180, 149
191, 197; of dogs, 120 Fourcroy, Antoine, 23, 24, 35
energy value of foods: vs. combustible en¬ Frankland, Edward, 67-8, 73, 225
ergy in living systems, 70, 73; gross heats Funk, Casimir, 121—2
of combustion, 67—8; metabolizable, 68
enthusiasm, 140, 231—2
Galen, 1—2
essential fatty acids; in kwashiorkor diets,
gases, 17, 18
i97> 2-01
Gay-Lussac, Joseph Louie, 46 -
gelatin: albumin with added oxygen, 24;
FAO/WHO Joint Committee on Nutrition,
difference from albumin, 50; grouping
meetings of: (1949), 149! (1951), 151;
with protein, 45; low nutritive value, 30—
(1952.), 154; (i954), 156; (i957), 159,
2, 130, 140, 244; methods of prepara¬
161; (1961), 181; (1966), 162, 185;
tion, 23, 30—1; use in “liquid protein”
(1970), 196; (1975), 200
diets, 207
fat, body; characterization as not essential,
genetically improved plants, 162, 177
48; formation from vegetable oil, 25;
Gilbert, J. H., 60
functions, 12; synthesis from carbohy¬
Glisson, Francis, 5
drates, 50, 53—4; role in protecting nitro¬
gluten: absence from vegetable foods other
genous tissues, 51
than wheat, 11, 14, 219; as “animal sub¬
fat, dietary: absorption in kwashiorkor pa¬
stance” in wheat, 10— 11; comparison
tients, 154, 157; as an auxiliary nutrient,
with animal protein, 61, 74; lack of crea¬
35; as chief aliment of vegetables, 26;
tine, 74; nutritional importance, 11, 22,
conversion to body fat, 25; in kwashior¬
33, 219; soluble form in green plants, 15
kor diets, 183-4, x97> 2.01; as food for
glycine: in dogs’ diets, 125; synthesised by
dogs, 28; intakes in U.S. vs. Europe, 103;
animals, 129
as source of animal heat, 51; as source of
Gopalan, C.: energy and protey} intakes in
energy for muscular work, 69
poor children, 156, 186; role of infection
fermentation: in digestion, 8, 25, 49; as
in kwashiorkor, 158; similarity of diet in
source of animal heat, 7
kwashiorkor and marasmus, 186
ferments, see fermentation
Graham, G. G., 184, 185
fertilizers, 48; nitrogenous, 60
Graham, Sylvester, 79—84
fiber, dietary: inertness, 35; role in diet, 29,
groundnuts, see peanuts
82, 83
guano, as fertilizer, 36
fibrin: abundance in muscles, 23; analyses,
gums, 28; auxiliary nutrients, 35; see also
43, 56; conversion to albumin, 49
vegetable jelly
fibrin, vegetable, 49
Fick, Adolf, 65—8, 221
fish, value relative to meat, 101—2 Halle, J. N., 26
fish farming, 152, 157 Hammond, W. A., 59
fish flour, see fish protein concentrate (FPC) Hansen, J. D. L.: association of kwashior¬
fish protein concentrate (FPC), 161, 163-8; kor and marasmus with protein-deficient
as additive to bread, 157, 164; fluorine diets, 185; criticism of McLaren, 199;
content of bones, 167; intestinal “filth,” potassium depletion in kwashiorkor, 157
167; manufacture in Chile, 163; motives Harvey, William, 2
of promoters, 230; production on a large Hatch Act, 105
scale, 158; protein quality of, 167-8; sol¬ health foods, 206—8

274
 Index

heat of combustion, see energy value of 186, 201; toxic factor in foods, 148, 201;
foods vitamin deficiency, 146, 148, 158-9, 222
Hegsted, D. M., 215 kwashiorkor, character and occurrence: in
Helmholtz, H., 67 adults, 147; association with early wean¬
herbivores: effect of acescent diet on, 12- ing, 187; in breast-fed infants, 157, 196;
13; source of N for, 28, 37, 49 diarrhea, 143; difference from pellagra,
high-lysine corn, 177 146; endocrine changes, 187-8; in ethnic
Hindhede, M., 97—8 groups, 148, 152, 156; first descriptions,
Hopkins, F. G.: isolation of tryptophan, 142—7; as form of pellagra, 142, 143,
127; need for tryptophan in mice, 128 145—6; low serum albumin, 186; and
horses, use in experiments, 37—8, 126 marasmus, 182; meaning of “kwashior¬
kor,” 143, 147; reproducibility in ani¬
mals, 188
immunoproteins, in milks, 202
INCAP, 156; development of Incaparina,
laborers: high-carbohydrate diet, 52; “in¬
174-5
stinct of,” to eat fat, 60; meat require¬
Incaparina (economic weaning food): devel¬
ment, 6, 10; need for coarser foods, 6;
opment by INCAP, 174; commercial pro¬
protein requirement, 89; see also activity,
duction in Central America, 174—5
physical
infections: measles, 197; protein and resis¬
lactoferrin, in human milk, 201
tance, 198; triggering factors for kwa¬
lactose, and diarrhea in kwashiorkor pa¬
shiorkor, 149, 156-8, 181, 187-8, 198
tients, 153
insensible perspiration, 1, 2
Lavoisier, Antoine, 17—22, 25, 42; increase
insulin, serum levels in kwashiorkor, 187-8
in carbon dioxide with activity, 20, 51
isomerism, 42
law of multiple proportions, 40
isotopes, in study of protein metabolism,
Lawes, J. B., 60
137-9, 2.13-M, M5-6
leaf proteins, 15
legume plants: ability to fix atmospheric N,
Jaya, Rao, 187-8 36; presence of protein in, 44
Jelliffe, D. B.: introduction of term “pro¬ leukotrienes, levels in kwashiorkor, 203
tein-calorie malnutrition” (PCM), 181; Leverton, Ruth, 205
on importance of breast feeding for infant Liebig, Justus von: career, 42, 44, 46-8, 57;
health, 201; replacement of bulky staples character, 56, 58, 73-4, 228-9; creatine’s
with oil and sugar, 184 role in nutrition, 74, 128; on digestion,
Joule, James, 66, 68, 76 57, 124-5; malt soup, 57, 74; meat ex¬
joule (J): unit of energy, 76 tract, 57, 74-5, 12.8; move to animal
chemistry, 48, 224; on protein as the only
Keill, James, 6 true nutrient and muscle fuel, 48—50, 52,
Kellogg, John Harvey, 84-7, 98 57, 61, 65, 73, 221, 223-4; rejection of
Kesselmeyer, Johanne, 14 the “protein radical,” 55-7; vegetable
kidney stones: relation to diet, 28 casein, 44, 49, 223
kilocalorie (kcal), 66, 150 Lind, James, 13
Konig, J., 101 liver: in kwashiorkor, 188; as site of protein
Kuhn, T. S., 222 metabolism, 70, 126
kwashiorkor, causes of: complexity, 148, liver, fatty: absence of, in marasmus, 186;
156, 158, 201, 220; deficiency of energy, and infant malnutrition in Brazil, 156; in
153, 156, 180-1, 195-8; deficiency of kwashiorkor, 143, 148, 149; production
protein, 146, 148, 154—5, 156, 185, 195, by methionine deficiency in animals, 150
201, 222; dysadaptation to dietary stress, liver damage: in kwashiorkor, 148; relation
186-9; infections, 156-7, 158, 186; po¬ to earlier kwashiorkor, 149-50; from
tassium and magnesium deficiency, 157, toxins, 148

275
 Index

lysine: as antiviral agent, 207; breeding tion of the albuminous substance,” 140;
corn high in, 177; as limiting amino acid on need for lysine and tryptophan by
in grains, 135, 177; requirement for adult young rats, 128; as one of Chittenden’s
rats, 128, 135; requirement for growing subjects, 108—10; protein efficiency ratios,
rats, 128, 135; requirement for humans, 129—30, 238-40; as teacher of W. C.
135—6; testing of, as supplement to hu¬ Rose, 131
man diets, 177-8; unavailability in milk Metcalfe, Rev. William, 79
powders, 204 methionine: adult human needs, 211; defi-
, ciency as cause of fatty liver, 150; human
Magendie, Frangoise, 27-32, 39, 53, 82 need for, 211; as limiting amino acid in
magnesium deficiency in kwashiorkor, 186, soy protein, 154, 170, 211; as possible
201 limiting factor in kwashiorkor, 150; re¬
maintenance requirements: isotope studies, quirement of rats for, 132
139, 141; “wear and tear,” 3, 128, 216 mice, in experiments, 128
maize, see corn microbiological assays, 134
malaria, see infections micronutrients: discovery, 120—3; impor¬
malignant malnutrition, 147, 149 tance of, 200
malt soup, see Liebig, Justus von milk, cow’s: allergic reaction to, among in¬
marasmic kwashiorkor, 155, 182 fants, 202; comparison with human milk,
marasmus: as adaptation to protein-energy 182; digestion of, 183; lack of use in
deficiency, 186—8; association with early some cultures, 143; limits to production,
weaning, 196, 198; as a condition of 152; need for increased production of,
“simple” starvation, 144, 155, 185; com¬ 158; in prevention and treatment of kwa¬
parison with kwashiorkor, 156, 184, 185, shiorkor, 143, 148, 153; protein quality,
186—7, 198; role of high-energy intakes 131, 135—6, 183, 243; as rich source of
in recovery, 184 many nutrients, 122, 22b
Margen, S., 208—9 milk, human: composition of, 151—2; im-
Mayer, J. R., 52-3, 66 munoproteins and lactoferrin in, 201; low
Mayow, John, 5 protein content, 117, 151—2
McCance, R. A., 185 milk powder: distribution by UNICEF, 157;
McCay, D., 114—7, 221 importing of, 152, 163; need for new
McLaren, D. S.: “daydreams at the UN,” substitutes for, 150, 152, i6*t; as treat¬
200—1; “great protein fiasco,” 198, 228; ment of choice for kwashiorkor, 150, 157
on marasmus vs. kwashiorkor, 184; on Mitchell, H. H.: character, 131; concept of
need for energy vs. protein, 198; on vita¬ “chemical scores,” 134-5, 190; human
min A deficiency, 186 studies, 136; meaning of “turnover,”
measles, see infections 140; protein evaluation, 240—4
meat: alkalescence of, 9; contribution of Mulder, Gerrit: discovery of the “protein
unknown factors, 32, 60; digestion of, 8, radical,” 43—4, 54, 124, 229; dispute
82, 102; expense of, 104; high biological with Liebig, 56—7
value for humans, 136; and improvement muscles: albuminoids and water as constitu¬
of military morale, 79; need for, by the ents, 225; development of tension during
inactive, 6, 10; overstimulation caused rest, 72-3; formation from blood, 49;
by, 51, 81—2; as preferred food, 113, mechanical efficiency, 53, 68; mechanism
224; responsibility for autointoxication, of action, 49, 72, 76; protein as fuel, 49,
86—7; as supplement to bread, 60; value 67, 69, 221; use of sugar and fat, 53, 67,
for choleric types, 2; value for dogs, 32, 69, 221
120
meat extract,see Liebig, Justus von net protein utilization (NPU), 241—2
mechanical physiology, 5—7, 16 Newton, John, 78
Mendel, L. B.: education, 129; “glorifica¬ niacin, see vitamins

276
 Index

Nicol, B. M., 183—4, 196 PCals% (percentage of energy in a food or

nitrogen: in ammonia, 22; analyses for, 34, diet supplied by protein), 151; of mixes
36; double molecule of, 41; as an ele¬ used to treat kwashiorkor, 183, 184; safe
ment, 20, 27; exhalation by herbivores, allowance for humans, 191-2
38; in respiration, 20, 25; urinary losses, peanuts, 169—70: aflatoxin contamination,
63—4; utilization from atmosphere, 20, 170; as flour for children, 161; as a
26, 36, 49; see also urea “milk” in Asia, 169; as protein source
nitrogen balance: criticism of technique, available in large quantities, 158, 169;
106, 115, 212, 214; at different N in¬ use in Indian “multipurpose food,” 169
takes, 90-4, 94, 113, 114, 225; first peas, see legume plants
trials, 37—8, improvement with extra en¬ pellagra: investigation of, using dogs, 143;
ergy intake, 95-7, 109, hi, 193, 211- vitamin deficiency as cause of, 122, 224
12, 225; longer trial periods, 214; as pellagra, infantile: equation with kwashior¬
measure of protein adequacy, 140; reten¬ kor, 142-3, 145—6; in New Orleans,
tion on high-protein diets, 210; reexami¬ 146
nation of procedure, 208—12 pepsin, 125
nitrogen content: of foods, 34; as measure peptones, 125—6, 223
of nutritive value, 34-5, 72; of tissues, 23 Pettenkoffer, M., 71
nitrogen-free diets: failure to support life, Pitcairn, Andrew, 5
27—30, 33; use in measuring obligatory N plantains: association with kwashiorkor,
losses, 190, 241 147, 159; excessive bulk of, 184, 196—7;
nucleic acids, and single-cell proteins, 176 as a low-protein staple food, 35, 151
nutrition: complexity of requirements, 30; plants, see vegetable foods
knowledge of in 1800, 27; as subject of Playfair, Sir Lyon, 51-2, 61-2
conjectures (1816), 28 Popper, K., 220
nutritional equivalents, 33—4 potassium: deficiency in kwashiorkor, 157,
186, 201
oil, see fat potatoes: biological value of protein, 243;
oilseeds, 168—9; see also cottonseed; pea¬ lack of gluten, 14, 219; N content, 34; in
nuts; soybeans; vegetable proteins prevention of scurvy, 61; as substitute for
opaque-2 corn, 177 grains, 13-14, 34
Osborne, T. B.: education, 129; on need for poverty, and malnutrition, 198—9, 200
tryptophan and lysine in growing rats, prisoners: stress of hard labor, 62—4; sus¬
128; protein efficiency ratio, 129—30, ceptibility to scurvy, 61; traditional ra¬
238-40 tions of, 60, 62
overstimulation: as cause of disease, 81-2; Pronutro (proprietary protein supplement
as cause of exhaustion and debility, 83 marketed in S. Africa), 172—3
oxidation, role of animal kingdom, 3 8 protein: chemical composition, 124-9, 133-
oxygen: in Arctic air, 51, 58; damage by, 5; dietary importance, 60; dynamic state
50, 52; decomposition of muscle fibrin in tissues, 137—40, 212—16, 245-6; en¬
by, 51; in digestion, 50; Lavoisier’s ele¬ ergy from metabolism of, 67—8; exclusive
ment, 19; measurement of uptake, 20; synthesis by the vegetable kingdom, 44,
vital function, 50 223; “liquid,” in slimming diets, 207;
manufacture from petroleum, 50, 69;
pancreas: damage in kwashiorkor, 148, Mulder’s radical, 43, 46, 54, 55; as the
157; function in protein digestion, 125, only true nutrient, 50, 69; origin of the
127 word, 43
Parkes, E. A., 68 protein, dietary: danger of excess, 86-8,
Parmentier, Antoine, 13-14, 22 217; effect of low intakes, 112; functions,
paupers, dietary standards for (1853), 62 102; new sources of, 161—79; and ner¬
vous energy, 106, 114, 215; and physical
Payne, P. R., i99>

277
 Index
activity, 49, 62; responsibility for autoin¬ controversy over, 182; warning against
toxication, 86—7 excessive protein intake, 217; see also en¬
protein, dietary standards for adults: ergy requirements; protein, dietary stan¬
(1853), 62; (1875), 89; (1891), 105 dards for adults; protein, dietary
(1904), 112; (1952), 151; (1955)1 !58; standards for infants
(1957-85), 189-95, 208-12; questioning reference protein, 205—6
of values from N balance, 214—15 relative protein values, 191
protein, dietary standards for infants: RDA respiration: cooling function, 2, 15; relation
(1958), 182; UN (1952), 150-1; UN < to combustion of tissues, 21; see also
(1955-85), 184, 189-95 oxygen
protein, digestion of: minimal changes dur¬ Rhodes, Katerina, 156, 181, 226
ing, 124, 223; comparison of fish and rice: association with lack of vigor, 114; as
meat, 101-2; efficiency of, in Third replacement for cassava in youqg chil¬
World diets, 192, 193; human studies, dren, 183; unsuitability as sole food, 34;
91, 192; vs. putrefaction, 8; release of white, deficiency of thiamine, 220
amino acids, 125, 223 rice polishings, in prevention of beri-beri,
Protein Advisory Group (PAG): change of 121
name to Protein-Calorie Advisory Group rickets, 5; vitamin deficiency, 122
(1974), 163; encouragement of new pro¬ Ritson, Joseph, 78
tein production, 178; establishment by Rittenberg, David, 138
WHO (1955), 159; joint responsibility of Ritthausen, Heinrich, 127
FAO, WHO and UNICEF (1961), 162; Rivers, J. P. W., 230
recruitment of food scientists and market¬ Rockefeller Foundation, 159
ing experts, 162; reputed association with root vegetables, 183, 192
disastrous events, 198, 228; response to Rosa, E., 107
criticism, 198; termination of (1977), Rose, W. C.: on need for amino acids by
163; World Bank as additional sponsor, rats, 131—2; discovery of threonine, 132;
163 on human requirements for amino acids,
protein—calorie malnutrition (PCM) or pro¬ 135-6, 193, 2-05, in
tein-energy malnutrition (PEM), 181, Rothamsted Experiment Station, 60
183, 185, 200 Rouelle, M. H., 14-15, 23, 24, 41
protein deficiency, effects of: beri-beri, 120; Rousseau, Jean-Jacques, 77 ^
cirrhosis of the liver, 150; idleness, 97, Rubner, Max: need for protein in infants,
113, 155; kwashiorkor, 146, 148, 155; 117; measured digestibility of foods, 90,
scurvy, 61; worldwide, 157, 160, 161—2 93, 94, 96
protein efficiency ratio, 238-40 Rumford, Count, 26
protein isolates, 169, 231 ruminants, 127; see also herbivores
protein quality, see biological value; chemi¬ Rush, Benjamin, 81, 84
cal score; net protein utlization; protein
efficiency ratio “safe practical allowances” for protein,
Prout, William, 41, 59, 225 189-90
proximate analysis, 100-1 Sanctorius, see Santorio
Santorio, 2, 6
Quesnay, Francois, 7 Schoenheimer, Rudolf, 138—40
Scrimshaw, N. S.: criticism of committee
radicals, as stable organic groupings, 42 consensus, 192, 228; doubts over protein
rats: growth on purified diets, 128—9; needs standards, 190, 192; N balance work at
of, for amino acids, 129—33, 154; protein MIT, 209; promotion of Incaparina, 175
efficiency ratio, 238—40; use as model for response to McLaren, 198; studies of
humans, 242-4 kwashiorkor, 156, 157
recommended dietary allowance (RDA): scurvy: alkalescent diet as cause of, 9, 13,
 Index

25, 84; bad air as cause of, 17; on bread- Takaki, K., 121, 220
and-water diet, 84; infantile, 145; protein thiamin(e), see vitamins
or potassium deficiency as cause of, 61, Thomas, Karl, 136, 240—1
224; as vitamin deficiency, 122 Thompson, Benjamin, see Rumford, Count
Seguin, Armand, 20, 25 tofu, 172
Seventh Day Adventists: belief in vegetari¬ transpiration, see insensible perspiration
anism, 84-5, 112; good health records, trauma, as cause of N loss, 136—7
87 Trowell, H. C.: criticism of Stannus, 223;
sexual activity: as a cause of debility, 83-4, on kwashiorkor, 145, 148
86; stimulation by animal food, 84 tryptophan(e): destruction by acids, 127; as
Shelley, Percy Bysshe, 78 hormone precursor, 128; synthetic, 207;
Sherman, H. C., 189 nutritional importance, 128
sickness as a sin, 79 tubers, see root vegetables
sindrome policarencial infantil (Central turnover (of body tissues): complete change
American name for condition later de¬ every three years, 12; as explanation for
scribed as PCM), 152 nutritional needs, 16; study of, with iso¬
single-cell proteins (SCP), 162, 175-7; ad¬ topes, 138-9
verse effects on humans, 177; hydrocar¬
bons as energy sources, 176; safety limits UNICEF: joint sponsorship of the PAG,
on nucleic acid intake, 176, 230-1 159, 162; support for manufacture of
sleep, as period of protein synthesis, 50 protein supplement in Nigeria, 170
Smith, Edward: criticism of Liebig’s meat United Nations Organization (UN), 149;
extract, 75; early dietary survey, 62; approval of “Protein Crisis” paper in
effect of labor on metabolism, 62—4, 68 1968, 162, 185
soybean protein: amount needed to meet UN University World Hunger Programme:
human needs, 191, 210-11; methionine sponsorship of balance experiments, 192
first-limiting, 170 unknown nutrients, 32, 117-18, 119—24
soybeans: bitterness and antinutrients, 153, urea: as component of kidney stones, 29;
170- 3; growth in large quantities, 158, effect of work on excretion of, 59, 64,
170; processing methods, 153—45 l6l> 67, 70, 71; heat of combustion, 68; as
171- 2; value in kwashiorkor, 153-4 major N metabolite, 42, 45, 49; recycling
spiced foods as cause of disease, 81 of through the colon, 214; synthesis in
Stannus, H. S., 142, 143, 223 1828, 41
starch: conversion to protein, 43-4; as fuel uric acid: as cause of gout, 176, 231; excre¬
for the “vital lamp,” 11; as source of en¬ tion of, in relation to dietary N, 28-9; in
ergy for muscular work, 69; and reduc¬ protein breakdown, 59; stress factor, 112
tion of need for protein 50, 96-7; sugar
as digestion product of, 50; as support of Van-Bochaute, 2.4-5
respiration, 50; as “the true nutrient,” Van Helmont, Jean-Baptiste, 7
14, 22, 219 Vauquelin, L. N., 35
sugar(s): auxiliary nutrient, 35; in blood, vegetable casein, 44, 49, 223
225; as “chief aliment” in vegetables, 26; vegetable foods: acescence of, 8, 9, 11; con¬
convertibility to body fat, 50, 53-4; fail¬ version to animal flesh, 5, 9-10; cooling
ure as sole food for dogs, 28; lack of N, character of, 12, 81, 84; poor/slow diges¬
22, 26; as support of respiration, 50 tion of, 82, 95; as economical, 101; eval¬
sugar baby, 154, 181, 226 uation of, by their N content, 79; jelly,
Sukhatme, P. V., 195-6 12, 23; as “metaphysical,” 84; promotion
sunflower seeds, 154 of bowel movements, 82, 83, 87; sugar
sweet potatoes: association with kwashior¬ and oil as “chief aliments,” 26; use in
kor, 147; protein content of, 151 treating kwashiorkor, 149, 153-4* 1573
Sylvius, see de le Boe, Francois see also animalization

2-79
 Index

vegetable jelly, 23, 26 wheat: acescence of, 9; “animal glue” and

vegetable kingdom, exclusive capacity for albumin content, 10, 23; high nutritional
complex synthesis, 38, 53, 222-3 value of, 14, 22; inability to fix atmo¬
vegetable proteins: comparison with animal spheric N, 36; inferiority of wheat pro¬
protein, 49, 57, 124, 127, 128, 223; use tein to that of milk, 135, 136; nitrogen
for dogs and rats, 120, 128, 130, 244 content, 34; nutritional component of,
vegetarianism: advantages, 77—88, 97—8; 10, 14, 219; see also gluten
disadvantages, 96—7; religious basis, 77—8 wheat bran, 83
vegeto-animal substance, see animal sub¬ wheat bread: as good protein source for
stance children, 137; insufficiency as sole food
ventriculin, 148 for dogs, 30; as major food of Victorian
vital flame, see animal heat poor, 62—3; praise of, by Sylvester Gra¬
vitalism, 42 ham, 82—3
vitamin A: deficiency in Chittenden’s dogs, wheat gluten, see gluten
123, deficiency in Third World, 154, 186, Whitehead, R. G.: endocrine changes in
200; discovery, 122; need for worldwide kwashiorkor, 187—9; energy as appar¬
supplementation, 200 ently limiting factor in kwashiorkor, 197
vitamins: discovery, 121—3; “health food” Widdowson, E. M., 185
supplements, 206; in treatment of kwa¬ Williams, Cicely: reports on kwashiorkor,
shiorkor 145, 148—9, 152, 158-9, 222 142-3, 146, 226; transfer to Malaysia,
Voit, Carl (or Karl): arguement for high MS
protein intakes, 215, 225; attack by Lie¬ Willis, Thomas, 8
big, 73; balance experiments with dogs winged bean, 230
and humans, 69—72; concept of “circulat¬ Wislicenus, Johannes, 65-8, 221
ing protein,” 91-2, 225; dietary stan¬ Wohler, Friedrich, 41, 42
dards of, 89, 97, 102, 103; on relation of women, scientific contributions of, 226—7
energy to N balance, 212, 225; view of World Health Organization (WHO), 149
vegetarianism, 94—7 world protein gap: acceptance by UN as
volatile alkali, 8, 10, 15; identification as major problem in 1960s, 162, 185; omis¬
ammonia, 22; procedure for obtaining, sion from 1975 World Food Conference,
42- 200; questioning of existence of, 196; un¬
von Haller, Albrecht, 11—12 desirable effects of the concept of, 200

Waterlow, J. C.: as chair of 1970 FAO/ yeasts, see single-cell proteins

WHO Expert Committee, 196; concepts Young, V. R.: criticism of N balance tech¬
of adaptation and accommodation, 214- nique, 212, 214-15; use of amino acid
15; digestion of cow’s milk in kwashior¬ turnover as measure of nutritional ade¬
kor, 182—3; kwashiorkor in Brazil, 155- quacy, 212-13, 245—6
6; “protein gap” as a myth, 199, 228; re¬
vised ideas, 201; treatment of kwashior¬ zein: as “incomplete” protein, 128; low
kor with low-protein mixes, 183, 184, value of, for rats and mice, 128, 140, 244
222 zinc, 123, 178, 201
Welbourn, H., 180, 226 Zuntz, N., 106

280
*
■

C
 LIBRARY

Amazing Research.
Amazing Help.

http://nihlibrary.nih.gov

10 Center Drive
Bethesda, MD 20892-1150
301-496-1080