J. Insi. Brew.. January-February, 1983, Vol. 89,/v>.

41-46 41

CAMBRIDGE PRIZE LECTURE

DIMETHYL SULPHIDE—ITS ORIGIN AND CONTROL IN BREWING

By C. J. Dickknson

(Central Development Laboratory, Allied Breweries Limited, 107 Station Street. Burton-on-Trent. DEI4 1BZ)

Received 20 August 1982*

The fate of dimethyl sulphide (DMS) and its precursors, S-methylmethionine (SMM) and dimethyl
sulphoxide (DMSO), in the malting and brewing processes, has been studied on both a laboratory
and a production scale. The information gained has been used to show how the level of DMS in beer
may be controlled. The principal determinant of the beer DMS level was the SMM present in malt.
DMSO was not a significant precursor of DMS in normal lager production.

Key words: malt, wort, beer, dimethyl sulphide, sulphur Two other possible sources of DMS may be noted. DMS
compound, volatile compound. has been found in infected wort,17 but clearly this cannot be
responsible for the relative consistency of DMS levels in
This paper describes research done at Allied Breweries in products from many different breweries. Also, Anderson &
the last few years on the origin and control of dimethyl Howard2 found DMS in hops, but not at a level sufficient
sulphide (DMS) in malting and brewing. The results and to contribute to the flavour of the final product, particularly
conclusions are discussed in relation to previous work in since much of the DMS in hops would be expected to be lost
Allied Breweries and elsewhere, and to recent studies at the on boiling.
Brewing Research Foundation. There were a number of uncertainties preventing a clear
view of the origins of DMS in beer, and of the best way of
Introduction controlling its concentration. These were: the number and
In the last 10-20 years there has been much research into identity of sources of DMS in malt; how these sources
the origins and the flavour significance of beer constituents. behaved in wort production; and what controlled the DMS
British workers have been interested particularly in the formed in wort. In addition, it was particularly important to
characteristic flavours which distinguish various types of clarify how and to what extent DMS could be formed during
beer ale, lager and stout. Anderson el aP showed that many fermentation. Finally, it was not yet possible to define the
lager beers are characterised by flavours such as cabbagy- measures which maltings and breweries had to take in order
vegetable, sulphury and, in many cases, dimethyl sulphide. to control the DMS appearing in beer. These measures might
It has long been suspected that volatile sulphur compounds involve:
may contribute to lager flavour, but DMS is the one which 1. Choice of raw materials (barley varieties)
is now understood the most. 2. Germination conditions
DMS was first recognised as a component of beer in 3. Kilning conditions
1963.IIJ Subsequently, Sinclair et a/18'1* showed that different 4. Brewhousc procedure
levels of DMS were characteristic of certain types of beer. 5. Fermentation conditions
The flavour threshold of DMS is about 30 ug/litre.3 DMS 6. Choice of yeast strain
occurs in beer in the range 0-150 ug/litre and is therefore 7. Further beer treatments
probably significant in the flavour of European lager and in but the most appropriate means of control could not be
some British lagers, but not in ales or stout.19" chosen until the effects of the production processes on DMS
Allied Breweries1 involvement with research into DMS in and its precursors were understood better.
the flavour of lager stemmed from the desire to understand
the causes of flavours and potential flavours in beer, and Analysis methods
from the desirability of controlling the levels of significant DMS is volatile and can be analysed readily by head-space
compounds. To control DMS in beer it was necessary to gas chromatography.1'1 Although a sulphur-specific flame
determine the main sources of DMS in the raw materials photometric detector is commonly used, it was found that
and the fate of these compounds in the malting and brewing flame ionisation detection was adequate with a suitable sep
processes. aration.12 In beer or wort the sensitivity was less than
5 ug/litre, compared with the taste threshold of DMS of
Background about 30 ug/litre, and the reproducibility was good, with a
replication error of < 4 ug/litre when tested with a range of
White & Parsons" showed that a heat-labile precursor of
DMS, absent from barley, was formed during germination. beers.
A constituent of a precursor preparation was 5-methylmeth- For the analysis of DMS precursors) the method of White
ioninc (SMM), a compound known to release DMS on heat 6 Wainwright24 was used, employing hydrolysis in alkali
ing,25 and clearly the source of free DMS in wort. White & and determination of the DMS released using gas chromato
Wainwright showed that, under certain conditions, DMS graphy (GC). This method measures the amount of heat-
labile DMS precursor and was used routinely for malt and
could also be formed during fermentation.26 The extent of
wort, and in the isolation and identification of heat-labile
this metabolic production of DMS was related to the kilning
precursor(s).
which the malt had received. It seemed that the precursor
in green malt, which could not be metabolised to DMS
Identity of dms prucursors
by yeast, was converted during kilning to a form which
was metabolisable, but which could still release DMS by It had been established that there was a compound (or
hydrolysis. compounds) in green malt which released DMS on heating.
Whatever was the fate of this precursor when malt was
•Delivered on 5 July 1982 at the Cambridge Meeting of the Institute kilned, it was necessary first to establish the chemical
of Brewing. identity of the source of DMS in green malt.
42 mCKENSON: DIMETHYL SULPHIDE [J. Inst. Brew.

By analogy with many other plant products it had been Anness el aldemonstrated that it was dimethyl sulphoxide
suggested that the methylated amino-acid S-methylmethio- (DMSO) which could be metabolised by yeast to give
nine could be the DMS precursor in green malt.16 White & DMS,46 but the reduction to DMS was not stoicheiometric.
Wainwright did show that SMM was a component of the DMSO was formed during malt kilning by oxidation of
precursor, but their work suggested that the precursor was DMS.6
not simply an amino acid, but was possibly a peptide." With Clearly, it was necessary to examine the significance of
this possibility in mind, separation methods were chosen DMSO in the brewing materials and processes used within
which were known to be useful in protein and peptide iso Allied Breweries, and therefore a convenient method of
lation, and which had the advantage of being conducted analysis was required. The method used to establish the pres
under mild conditions, that is, at ambient temperature, in ence of DMSO in malt6 (gas liquid chromotography, GLC,
neutral aqueous solutions. of an organic extract) was not applicable to wort and beer
The isolation of the heat-labile precursor in green malt10 because DMSO is very soluble in water. However, DMSO
is shown in Table I. At each stage the precursor was moni can be readily assayed as DMS after a reduction stage.
tored by its property of releasing DMS on alkaline hydroly Anness5 has used stannous chloride to reduce DMS. In the
sis,24 and the potential DMS in each fraction was compared present work two other methods have been examined.
with the concentration of amino acids and peptides, as Sodium mctabisulphitc can be used to generate DMS from
measured by their amino groups. The cation exchanger DMSO in the flask used for the head-space DMS measure
carboxymethyl (CM)-cellulose allowed the separation of the ment. Only a 50°C heating step is required." This method
precursor from neutral and anionic compounds. The first was used routinely. Reduction by the thiol group of cysteine
gel-filtration step separated the precursor from high molecu was less convenient than metabisulphite but was used to
lar weight proteins and peptides, and from low molecular confirm some of the results for DMSO levels in malt."
weight amino acids. Further resolution on the basis ofcharge
was obtained on a finer grade of CM-cellulose and the final Dms in malting
Sephadex column increased the separation from what at that In the malting process SMM is formed during the germin
stage was the major impurity. These gel-filtration results ation of barley.23 During kilning, some of this SMM is
showed that the precursor had the same elution volume as degraded to DMS and some oxidation to DMSO may occur.6
SMM and thus did not seem to be larger than a simple amino Thus DMS and both of its precursors may be present in malt.
acid. At this stage the ratio of potential DMS to amino In the course of this work many samples of malt have been
groups suggested that one major component was present and analysed, with results13 shown in Table II. The differences
that it would be useful to look at the preparation with an between lager and ale malts are a reflection of the lighter
independent method. Thin layer chromotography (TLC), in kilning oflager malt.
two different solvent systems, showed that the major
component had the same properties as SMM. When the TABLE II. Ranges of Levels of DMS and Its
preparation was hydrolysed by boiling, a compound was Precursors Found in UK Malt"
formed which had the same properties as homoserinc—the
hydrolysis product of SMM.

TABLE I. Isolation and Identification of Heat-Labile DMS Precur

sor from Green Malt10

Ratio of precursor DMS to amino Concentrations expressed as ng DMS equiv

Stage nitrogen (mol/mol) alents per g malt.

Green malt extract 00034 The choice of barley variety as a factor to be examined in
CM22 CM-cellulose 0034 controlling DMS has been mentioned. Szlavko20 showed that
SephadcxGIS 010
CM52 CM-ccllulose
six-row varieties of Canadian barley gave malt with a higher
0-30
CM 52 CM-cellulosc 0-56 DMS content than did two-row varieties. However, as will
Sephadex G1S 0-63 be shown later, the content of SMM is a more important fac
(same eluiion volume as SMM) tor when comparing malts. Table III shows the results of
TLC (cthanol/NH, (same Rp as SMM, -»same Rp as micromalting ofdifferent barley varieties. The Canadian six-
or propan-2-ol/HCI) homoscrine after boiling) row variety Bonanza gave much SMM when malted in
the same way as some Canadian or British two-row varieties.
In a second experiment a British six-row variety was exam
ined, but the same effect was not observed. It seems from this
This work showed that the heat-labile DMS precursor in limited survey that choice of variety may not be a promising
green malt was SMM, a compound which cannot be metab means of varying the potential DMS content of malt, at least
olised to DMS by yeast.23-26 White & Wainwright had found for varieties available in the UK.
that, when malt was kilned at about 75°C, although some of
the heat-labile source of DMS was destroyed, there remained
TABLE III. Content of SMM in Lager Malt Produced from Various
a compound present in malt which could be metabolised by Varieties of Barley by Micromalting
yeast, resulting in DMS production during fermentation.26
The identity of this metabolisablc precursor in kilned malt
SMM
was not known. However, using a shorter version of the pro Variety (ug DMS equivalents per g malt)
cedure of Table I, it was shown that the only heat-labile
DMS precursor in kilned malt was that SMM which had
Experiment 1: Bonanza (6-row) 13-4
survived kilning.10 The malt examined was a lager malt Klages (2-row) 3-5
kilned under conditions expected to give much of the metab Betzcs (2-row) 5-7
olisablc precursor. Whatever is the source of DMS which Maris Otter (2-row) 5-3
may be produced during fermentation, it is not heat-labile
Experiment 2: Athene (6-row) 4-8
and therefore is not measured by the standard alkaline Maris Otter (2-row) 70
hydrolsis assay.24
Vol.89, 1983] dickenson: dimethyl sulphide 43

It was of greater interest to examine conditions which a substantial level of DMS in beer.3 Then it was concluded
affected the levels of DMS and its precursors in finished malt, that malt must be dried at a fairly low temperature, before
since these would affect the levels present at the start of the curing at a higher temperature. However, only the final beer
brewing process. It was already established that the amount DMS was measured—the intermediate details were not
of SMM in green malt increased as germination proceeded considered. The results here show that the earlier conclusion
and this has been confirmed on a number of occasions with does not apply as far as SMM is the source of DMS in the
production malting. Green malt contains negligible levels of beer.
free DMS and DMSO. A particular interest was the effect The results given here allow an estimate of the level of
of kilning on the degradation of the heat-labile SMM, and SMM in green malt which will survive the kilning process.
the stability of this compound over a range of temperatures This information is needed to control SMM in malt, so thai
used for kilning lager malt was examined.13 To control the raw materials consistent in potential DMS can be provided
variables, SMM levels were measured in various malts, for the brewing process.
which were held in sealed containers to ensure a constant During malt kilning, free DMS is formed from the SMM,
moisture level, and sampled at appropriate intervals. Results and some of the DMS remains in the malt. The conditions
at 65°C are shown in Fig. 1. In all the experiments SMM was governing free DMS levels in malt have not been exam
degraded in a first order manner (linear semi-log plot) for at ined since, as will be seen later, this free DMS is not relevant
least the duration of normal production kilning. to the control of DMS in beer, at least in normal production.
With DMSO, Anness el aP showed that the amount of
DMSO in malt depends on the kilning received. However,
1.0
detailed studies of the dependence of DMSO formation on
all the possible variables—moisture, DMS, SMM, oxygen—
have not yet been done.

Dms in wort production

As SMM will be subject to thermal degradation in hot
wort it is clearly an important potential source of DMS in
pitching wort. Opposing this DMS formation will be the
tendency of the volatile DMS to be lost in boiling and other
heating processes. To predict the effects of brewhouse pro
cedure on the SMM derived from the malt it was necessary
to know its stability in wort at the high temperatures
involved in wort production.
It was found that SMM is hydrolyscd in hot wort in a first
order reaction.9 In boiling wort the half-life is about 40
-10
min—so that in the average 90 min copper boil about three-
so 100 quarters of the SMM is lost. At infusion mash temperatures
Timcflh) SMM is much more stable, and not much is expected to be
Fig. I. Thermal degradation or SMM in malt held at 65°C in sealed lost at that stage of wort production.
containers to ensure constant moisture content. (SMM concen The factors' found to affect the concentration of DMS in
trations in DMS equivalents, ng/g). O green malt, moisture the wort after cooling, immediately before pitching with
content 42% (w/w), • kilned malt, moisture content 5% (w/w) yeast, are summarised in Table V. The first four parameters
(normal production lager malt). The measured half-lives of govern the level of SMM which will be extracted from malt
SMM are 30 and 22 h respectively. into sweet wort, which depends on the amount of malt used
and its SMM content. SMM is hydrolyscd in boiling, accord
ing to the half-life determined, and it was found also that,
Fig. I shows that SMM in green malt is slightly more at the end of boiling, very little free DMS was present.
stable than in normal production kilned malt of about 5% During transfer of wort to a wort receiver, particularly the
moisture. At the other temperatures examined the same vigorous transfer associated with a whirlpool, it is expected
effect was seen, as summarised in Table IV. At the lower that free DMS will continue to be lost.
temperatures, the greater stability at high moisture content, When the hot wort is allowed to stand, and until it is
in malt which has had no previous heat treatment, may be passed to the cooler, SMM will continue to be hydrolysed,
due to some continuing synthesis of SMM. The measured according to the temperature, and more free DMS will be
half-lives then indicate the difference between thermal formed, depending on the amount of SMM which has sur
degradation and the continuing synthesis. At any moisture vived the boiling and transfer. However, in the hot wort
level raising the temperature gives a considerable increase in stand, Szlavko & Anderson found (confirmed in our brewery
the rate of breakdown of SMM.
The observed greater stability at high moistures contrasts
TABLE V. Factors Affecting the Concentration of DMS in Wort
with earlier work to find suitable malting conditions to give
at Pitching'

TABLE IV. Stability of SMM in Malt at Various Tem Parameter or process Effect on wort DMS level
peratures and Moisture Contents13
Malt: SMM content +

Half-life of SMM (h) Extract

Fraction of extract derived from mall +

Temperature 42% (w/w) 7-5% (w/w) 5% (w/w) Collection gravity 4__

(°C) moisture moisture moisture

Duration of boil -

Transfer to wort receiver

55 98 50
65 30 20 22 Duration of stand in wort receiver +
75 9-3 4-8 5-5 Duration of transfer to cooler +
85 2-4 1-9 Temperature in wort receiver +
44 dickenson: dimethyl sulphide [J. Inst. Brew.

investigation) that the free DMS formed was mostly retained on DMS. SMM is taken up by yeast" but is not metab
in solution in the wort. Thus in the absence of any evapor olised to DMS.23 The rapid disappearance of SMM
ation there is little tendency for the DMS to escape and it from fermenting wort has been confirmed.9
is at this stage that the DMS arises which is present in wort 3. There may be some reduction of DMSO to DMS, by
at pitching. the action of the yeast. Brewing yeasts have some abil
Knowing the temperatures and durations of the processes, ity to convert DMSO to DMS, up to -20% conversion
and using the appropriate half-life for SMM, the level of in synthetic media but less in malt wort. The signifi
DMS expected in wort for any particular brew can be calcu cance of this route to DMS in beer will depend on the
lated. The results of such an exercise1*15 with a number of levels of DMSO normally present in brewing wort and
brews in a lager brewery, are shown in Table VI. There was on the efficiency of the conversion under the fermen
a satisfactory agreement between measured and predicted tation conditions used in the brewery.
DMS levels. The effect of extending the hot wort stand in Thus the effect of fermentation on DMS levels will be the
the whirlpool by I h (in brews 4 and 5) is also shown. This net result of two opposing processes—loss of DMS with the
resulted in the expected higher wort DMS concentration, as evolved CO2 and further formation by reduction of DMSO.
it allowed a more complete breakdown of the SMM which A number of fermentations in lager breweries were examined
survived boiling. in order to determine the relative importance of these two
processes. A typical set of results for two brews13 is shown
TABLE VI. Predicted and Measured DMS Levels in in Fig. 2. Free DMS levels fell as the fermentations pro
Production Lager Brews'"
ceeded but the final filtered and packaged lager generally
contained the same concentration of DMS as the beer at the
Wort DMS (ug/litrc) end of fermentation.
Brew Malt SMM
number (fig DMS/g) Predicted Measured During fermentation and cold storage DMSO levels were
roughly constant, and reflected the amounts present in malt.
Similar levels were found in packaged beers.11 Since the
1 3-2 36 38
2 4-6 61 61 DMSO levels were much higher than those of DMS there
3 4-6 61 50 was no indication of whether any DMS had been formed
4* 5-6 103 93 from DMSO by reduction. However, in these fermentations,
5* 6-4 117 115 there was no net production of DMS within the experimental
error of the analyses.
•Hoi wort stand extended by I h. Fig. 3 shows the relationship of free DMS in the final beer
to that present at collection, for several different production
The foregoing work showed how wort DMS levels brews.13 There was generally a greater proportionate loss
depended on raw materials and wort production procedures. with the higher initial DMS levels. These fermentations all
With DMSO it was found that there was no significant followed similar temperature profiles, and all were in cylin-
change in concentration in samples drawn at various stages droconical vessels. It would be expected that fermentation at
during wort production. The concentration of DMSO in different termperatures or in other types of vessel would alter
wort at pitching corresponded to the amount present in mall. the extent of DMS loss, but not the general shape of the
graph.
Dms in fermentation
Although net formation of DMS was not observed in the
The following effects on DMS and its precursors may be brewery, it was important to see whether the production
expected to occur during fermentation and subsequent mat lager yeast was capable of reducing DMSO, that is, whether
uration and cold storage periods: any DMS had been formed in fermentation, but accom
1. Free DMS, present in wort at pitching, or formed sub panied by a greater loss of DMS than was apparent. If the
sequently, may be subject to losses along with the yeast were capable of reducing DMSO, then it would be im
evolved carbon dioxide.1421 portant to control the level of this DMS precursor appearing
2. SMM remaining in the wort will have no further effect in wort. Therefore, the formation of DMS from DMSO was

600

z
o

IS
Days from pitching

Fig. 2. DMS (•) and DMSO (O) levels during brewery fermentations (at 11-I6"C)
and subsequent cold storage.11
Vol. 89, 1983] dickenson: dimethyl sulphide 45

BO

40

40 80 120

DMS of co'lcciion Ijig/litrcl

Fig. 3. Comparison of DMS levels in finished beer with levels at collec

tion.13 Results from production lager brews with fermentation at
I I-16°C followed by cold storage. line of unit slope.

8990
too

Days from Ditching

Fig. 4. Reduction of DMSO to DMS in EBC tube fermentations13 of brewery lager

wort at I4°C, using production yeast Saecharomyces uvarum AB 140. The
total initial DMSO concentration (ug DMS equivalents per litre) is shown
against each curve. The dashed line shows the DMS concentration during a
similar fermentation with the lowest initial DMSO level, but with DMS
added to give the initial DMS concentration shown.

examined in laboratory fermentations of brewery wort, using Results of fermentations with added DMSO,13 at both 14°
EBC tubes.13 Fig. 4 shows results of a typical experiment, at and 8-5°C, are summarised in Table VII. Apparently only
a temperature similar to that of the brewery fermentations. about 1% of the DMSO present was reduced to DMS by the
Parallel fermentations, with increasing additions of DMSO, production lager yeast AB 140 at I4°C. Slightly more DMS
were examined, and in all cases the conversion to DMS was formation was observed with an ale yeast NCYC 240, stud
apparently only -1%. Separate control experiments were ied eleswhere.4*' At 8-5°C the observed conversion was
also performed, in which DMS (but not DMSO) was added about twice that found at the higher temperature, but free
in various concentrations, in order to demonstrate the loss DMS was lost more slowly. When the loss of free DMS was
of free DMS which would have occurred when there was a taken into account, the results showed that the production
negligible formation of DMS from DMSO. Over a wide yeast AB 140 converted only about 2% of the DMSO to
range of initial DMS concentrations, DMS was lost from DMS. It was concluded, therefore, that with wort
EBC tube fermentations with half-lives of ~ 3-5 days and > 7 DMSO concentrations of < 500 ug/litre, reduction of this
days at 14" and 8-5°C respectively. The control experiments precursor to DMS did not make a significant contribution
at I4°C suggested that, in the fermentations with added to the DMS in beer. A wort DMSO concentration of
DMSO (Fig. 4), if no losses of DMS had occurred, there 500 ug/litre corresponds to a level in malt of about
would have been no more than twice the observed conver 5 ug/g—considerably higher than that found routinely in
sion of DMSO to DMS. lager malt (Table II).
46 dickhnson: dimethyl sulphide [J. Inst. Brew.

TABLE VII. Reduction of DMSO to DMS in EBC Tube Fermentations"

Apparent conversion
DMS formed in fermentation ofDMSO to DMS
(Hg/litrc)
(%)

Saccharomyces Saccharomyces Saccharomyces Saccharomyces

Temperature Wort DMSO Added DMSO uvarum cerevisiae uvarum cerevisiae
<C°) (ug DMS/litrc) (HB DMS/litrc) ABI40 NCYC 240 AB 140 NCYC 240

0 ~3 -5 ~l-2 -20
14 250 874 11 16 10 1-4
2623 31 47 II 1-6
8744 98 126 II 1-4

0 ~5 ~6 -2-8 -3-3
8-5 180 874 18 23 1-7 2-2
2623 54 74 1-9 2-6
8744 180 260 2-0 2-9

Control of dms in beer References

This examination of the fate of DMS and its possible pre 1. Ahrenst-Larscn. B. & Hanscn, H. L.. Brauwissenschafi. 1963,
cursors during the brewing process has shown that, in nor 16, 393.
2. Anderson, R. J. & Howard, G. A., Journal of the Institute of
mal lager production, DMSO is not an important source of Brewing. 1974,80,357.
the DMS in beer, which arises principally by thermal degra 3. Anderson, R. J., Clappcrton, J. F., Crabb. D. & Hudson, J. R.,
dation of SMM during wort production. Measures to control Journal ofthe Institute ofBrewing, 1975, 81, 208.
beer DMS levels must, therefore, concentrate on the content 4. Anness, B. J., Journal ofthe Institute ofBrewing. 1980,86,134.
5. Anness, B. J., Journal of the Science of Food and Agriculture.
of SMM in malt, on brewing conditions which affect SMM 1981,32,353.
hydrolysis and DMS losses, and on fermentation conditions 6. Anncss, B. J., Bamforth, C. W. & Wainwright, T., Journal of
which may cause further losses of DMS. the Institute of Brewing. 1979, 85, 346.
With fixed brewing and fermenting conditions the ultimate 7. Bamforth, C. W. & Anncss, B. J.. Journal of the Institute of
Brewing. 1981,87,30.
control required is on the SMM content of malt. Germina
8. Booer, C. D. & Wilson, R. J. H., Journal ofthe Institute ofBrew
tion conditions are constrained by the extent of modifica ing. 1979, 85, 35.
tion required and also the potential of barley to form SMM 9. Dickenson, C. J.. Journal of the Institute of Brewing. 1979.85,
varies somewhat from year to year (C. J. Dickenson & P. A.
10. Dickenson, C. J., Journal of the Institute of Brewing. 1979. 85,
Brookes, unpublished results). Therefore, provided sufficient 329.
SMM can be formed during germination, the best method 11. Dickenson, C. J., Journal ofthe Institute of Brewing. 1981. 87,
is to adjust kilning conditions in order to control the extent 16.
of SMM survival from green malt. The results in Table IV 12. Dickenson, C. J. & Martin, P. A., Journal of the Institute of
Brewing. 1978.84, 143.
show how kilning affects SMM degradation.
13. Dickenson, C. J. & Anderson, R. C, European Brewery Con
Earlier uncertainty about the relative significance of the vention, Proceedings of the 18th Congress, Copenhagen, 1981,
two routes to DMS in beer (i.e. thermal degradation of SMM 413.
and enzymic reduction ofDMSO) led to criticism of brewers 14. Kavanagh, T. E., Steward, S. R.. Hildcbrand, R. P., Clarke,
B. J. & Meeker, F. J., Journal ofthe Institute ofBrewing. 1975,
specifying the SMM content of malt in order to give the 81, 322.
DMS level desired in beer.8 The present work has shown that 15. Kcpner, R. E., Strating, J. & Weurman, C, Journal ofthe Insti
the low levels of DMSO found in production lager brewing tute ofBrewing. 1963, 69, 399.
do not appreciably affect final beer DMS concentrations. 16. Nicfind, H. J. & Snath, G., Proceedings ofthe American Society
of Brewing Chemists. 1975, 33, 54.
Therefore, to produce beer with any particular desired level 17. Priest, F. G. & Hough, J. S., Journal ofthe Institute ofBrening,
of DMS, the appropriate content of SMM in malt, which 1974,80.370.
depends on the brewing and fermenting conditions 18. Sinclair, A., Hall, R. D. & Thorbum Burns, D., European
employed, should be specified. Brewery Convention, Proceedings of the 12th Congress. Inter-
Men. 1969,427.
19. Sinclair. A., Hall, R. D., Thorburn Burns. D. & Hayes, W. P.,
Journal ofthe Science ofFood and Agriculture, 1970,21,468.
20. Szlavko, C. M., Journal of the Institute of Brewing. 1975, 81,
438.
21. Szlavko, C. M. & Anderson, R. J., Journal of the American
Acknowledgements—I thank the Institute of Brewing and the Society of Brewing Chemists, 1979, 37, 20.
London Section for the invitation to give the Cambridge 22. Whjte, F. H., Brewers Digest, 1977, 52 (5), 38.
23. White, F. H. & Parsons, R., European Brewery Convention,
Prize Lecture. I am grateful to Mrs L. J. Blood, for assistance Proceedings ofthe 15th Congress, Nice, 1975, 721.
throughout this work, and to staff of Ind Coope Burton 24. White, F. H. & Wainwright, T., Journal of the Institute of
Brewery Ltd, Wrexham Lager Beer Company Ltd and Brewing. 1976,82,46.
Albrew Maltsters Ltd, for their co-operation during pro 25. White, F. H., & Wainwright, T., Journal of the Institute of
Brewing. 1976, 82, 292.
duction investigations. I thank the Directors of Allied 26. White, F. H., & Wainwright, T., Journal of the Institute of
Breweries Limited for permission to publish the work. Brewing. 1977, 83, 224.