Download the Full Ebook and Access More Features - ebooknice.

com

(Ebook) Avoiding attack : the evolutionary ecology

of crypsis, warning signals, and mimicry by
William L. Allen; Thomas N. Sherratt; Michael
Patrick Speed; Graeme D. Ruxton ISBN
9780199688678, 9780199688685, 0199688672,
https://ebooknice.com/product/avoiding-attack-the-
0199688680
evolutionary-ecology-of-crypsis-warning-signals-and-
mimicry-43638210

OR CLICK HERE

DOWLOAD EBOOK

Download more ebook instantly today at https://ebooknice.com

 Instant digital products (PDF, ePub, MOBI) ready for you
Download now and discover formats that fit your needs...

Start reading on any device today!

(Ebook) Biota Grow 2C gather 2C cook by Loucas, Jason;

Viles, James ISBN 9781459699816, 9781743365571,
9781925268492, 1459699815, 1743365578, 1925268497
https://ebooknice.com/product/biota-grow-2c-gather-2c-cook-6661374

ebooknice.com

(Ebook) Matematik 5000+ Kurs 2c Lärobok by Lena

Alfredsson, Hans Heikne, Sanna Bodemyr ISBN 9789127456600,
9127456609
https://ebooknice.com/product/matematik-5000-kurs-2c-larobok-23848312

ebooknice.com

(Ebook) SAT II Success MATH 1C and 2C 2002 (Peterson's SAT

II Success) by Peterson's ISBN 9780768906677, 0768906679

https://ebooknice.com/product/sat-ii-success-
math-1c-and-2c-2002-peterson-s-sat-ii-success-1722018

ebooknice.com

(Ebook) Master SAT II Math 1c and 2c 4th ed (Arco Master

the SAT Subject Test: Math Levels 1 & 2) by Arco ISBN
9780768923049, 0768923042
https://ebooknice.com/product/master-sat-ii-math-1c-and-2c-4th-ed-
arco-master-the-sat-subject-test-math-levels-1-2-2326094

ebooknice.com
(Ebook) Cambridge IGCSE and O Level History Workbook 2C -
Depth Study: the United States, 1919-41 2nd Edition by
Benjamin Harrison ISBN 9781398375147, 9781398375048,
1398375144, 1398375047
https://ebooknice.com/product/cambridge-igcse-and-o-level-history-
workbook-2c-depth-study-the-united-states-1919-41-2nd-edition-53538044

ebooknice.com

(Ebook) Piano adventures Performance 3b by Nancy and

Randall Faber

https://ebooknice.com/product/piano-adventures-performance-3b-52393612

ebooknice.com

(Ebook) Mimicry, Crypsis, Masquerade and other Adaptive

Resemblances by Donald L. J. Quicke ISBN 9781118931530,
111893153X
https://ebooknice.com/product/mimicry-crypsis-masquerade-and-other-
adaptive-resemblances-36894474

ebooknice.com

(Ebook) Ecology by William D. Bowman, Sally D. Hacker,■

Michael L. Cain ISBN 9781605356181, 1605356182

https://ebooknice.com/product/ecology-6835492

ebooknice.com

(Ebook) Experimental Design for the Life Sciences by

Graeme D. Ruxton, Nick Colegrave ISBN 9780199285112,
019928511X
https://ebooknice.com/product/experimental-design-for-the-life-
sciences-1954338

ebooknice.com
Avoiding Attack
Avoiding Attack
The Evolutionary Ecology of
Crypsis, Aposematism, and
Mimicry
Second edition

Graeme D. Ruxton
William L. Allen
Thomas N. Sherratt
Michael P. Speed

1
1
Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© Graeme D. Ruxton, William L. Allen, Thomas N. Sherratt, & Michael P. Speed 2018
The moral rights of the authors have been asserted
First Edition published in 2004
Second Edition published in 2018
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2018944296
ISBN 978–0–19–968867–8 (hbk.)
ISBN 978–0–19–968868–5 (pbk.)
DOI: 10.1093/oso/9780199688678.001.0001
Printed in Great Britain by
Bell & Bain Ltd., Glasgow
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
 OUP CORRECTED PROOF – FINAL, 09/07/18, SPi

Dedicated to
Katherine
Jioh
Frances, Finlay, Stuart, Kaelan, Nathan, Kyah, & Jakob
Yvonne
 Acknowledgements

Many colleagues helped with reading text, sug- The smartest thing the original three authors of
gesting references, and providing figures. In no Avoiding Attack did was bring in Will Allen as a co-
­particular order, we would like to acknowledge our author—his drive, energy, and scholarship have been
thanks to Nicholas Scott-Samuel, Anna Hughes, transformative—and every chapter of this book is bet-
Willem Bekkers, Jolyon Troscianko, Roger Hanlon, ter for his contributions; some simply would not exist
Sami Merilaita, Jennifer Kelley, Olivier Penacchio, at all without him. The joint smartest thing we did
Innes Cuthill, Sonke Johnsen, Malcolm Edmunds, was hire Rosalind Humphries as research assistant to
JayneYack, Takeo Kuriyama, Ullasa Kodandaramaiah, coordinate the four authors. Her friendly, unflappable
Carlos Cordero, Bree Putman, Dinesh Rao, Pierre- organization and very gentle chivvying were essential
Paul Bitton, Lynne Isbell, Chris Hassall, John Skelhorn, in completing this project and we all owe her a huge
Francis Gilbert, Thomas Reader, Krushnamegh Kunte, debt. Our final smart call was to stay with OUP—
Karl Loeffler-Henry, Thomas Aubier, Mathieu Cho­ Bethany Kershaw was exceptionally patient, always
uteau, Kyle Summers, David Kikuchi, Mikhail Kozlov, wise in her guidance, and everything we could
Kevin Arbuckle, Butch Brodie, Johanna Mappes, hope for in an Editor; the production and copyediting
Hannah Rowland, Bibi Rojas, Janne Valkonen, & side of the book (by Suryajeet Mullick and Julian
Elena Zvereva. We apologize to those that should Thomas particularly) was managed with collegiate
be on this list but slipped our mind. None of those efficiency and good grace. Julian in particular com-
named bear any responsibility for our errors and bines a very keen eye for detail with exceptional
omissions. intellectual engagement, professionalism, and wit.

vii
Contents

Introduction1

Chapter summary 3
The sequence of a predator–prey encounter and investment across
multiple defences 5

1 Background matching 9

1.1 Introduction, definition, mechanism, and chapter overview 9

1.2 Empirical evidence of background matching 10
1.3 The evolution of background matching 11
1.4 Co-evolutionary considerations 14
1.5 Ecological considerations 17
1.6 Unresolved issues and future challenges 23

2 Disruptive camouflage 24

2.1 Introduction and overview 24

2.2 Examples of disruptive camouflage 26
2.3 The multiple mechanisms of camouflage by disruption 28
2.4 Identifying and quantifying disruptive camouflage 33
2.5 The relationship between disruption and other forms of protective coloration 35
2.6 The ecology of disruption 36
2.7 Unresolved issues and future challenges 39

3 Countershading 41

3.1 What is countershading? 41

3.2 Examples and taxonomic distribution of countershading camouflage 43
3.3 Countershading camouflage mechanisms 45
3.4 Evolution 50
3.5 Costs of countershading and counterillumination camouflage 51
3.6 Developmental genetics of countershading 51
3.7 The evolutionary ecology of countershading 52
3.8 Countering countershading and counterillumination adaptations 54
3.9 Unresolved issues and future challenges 55

ix
x C o n te n t s

4 Transparency 57

4.1 Definition and introduction 57

4.2 The distribution of transparency across habitats 57
4.3 How transparency influences ease of detection 58
4.4 Evolutionary considerations 63
4.5 Ecological influences 68
4.6 Co-evolutionary considerations 70
4.7 Unresolved issues and future challenges 70

5 Secondary defences 72

5.1 Introduction and overview of the chapter 72

5.2 Evolution of secondary defence 72
5.3 Consequences of variation in costs of secondary defence 76
5.4 Ecology 78
5.5 Co-evolutionary considerations 80
5.6 Unresolved issues and future challenges 83

6 Aposematism 84

6.1 What is aposematism? 84

6.2 Characteristics of aposematic organisms 84
6.3 Evolution of aposematism 90
6.4 Maintenance of aposematic signalling 95
6.5 Ecology of aposematism 95
6.6 Co-evolution of aposematic signals and receiver psychology 102
6.7 Future work and conclusions 102

7 The evolution and maintenance of Müllerian mimicry 103

7.1 Introduction 103

7.2 Examples 104
7.3 Müller’s theory revisited 108
7.4 Evidence for Müller’s hypothesis 110
7.5 Questions and controversies 115
7.6 Overview 127

8 Advertising elusiveness 128

8.1 Introduction and definition 128

8.2 Empirical evidence of elusiveness signals 129
8.3 Evolution 138
8.4 Ecology 143
8.5 Co-evolutionary considerations 145
8.6 Unresolved issues and future challenges 146

9 Batesian mimicry and masquerade 148

9.1 Introduction and overview of the chapter 148

9.2 Examples of protective deceptive mimicry 150
 C o n te n t s xi

9.3 The origin of protective mimicry: selection or shared ancestry? 152

9.4 The evolution of protective deceptive mimicry 153
9.5 Ecological and phylogenetic considerations 169
9.6 Associated phenomena in the evolution of Batesian mimicry and masquerade 171
9.7 Overview 178

10 Startling predators 179

10.1 What do we mean by startle? 179

10.2 Empirical evidence for the defence 180
10.3 The evolution of startle defence 183
10.4 Ecological aspects of startle defences 186
10.5 Co-evolutionary considerations in startle defences 187
10.6 Unresolved issues and future challenges 188

11 Deflecting the point of attack 189

11.1 Overview 189

11.2 How deflecting traits work 189
11.3 The taxonomic distribution of deflecting traits 190
11.4 The evolution of deflective traits 198
11.5 Ecology 202
11.6 Co-evolutionary considerations 202
11.7 Future challenges 203

12 Dazzle camouflage 205

12.1 Camouflage in motion? 205

12.2 Examples of dazzle 206
12.3 How does dazzle camouflage work? 210
12.4 The evolution of dazzle 215
12.5 The costs and benefits of dazzle 215
12.6 The ecology of dazzle 215
12.7 Future challenges in dazzle camouflage research 216

13 Thanatosis 219

13.1 Introduction and overview of the chapter 219

13.2 Distribution 219
13.3 Form: the mechanisms involved 221
13.4 Evolutionary function: a cost/benefit approach 224
13.5 Ecological considerations 225
13.6 Co-evolutionary considerations 227
13.7 Unresolved issues and future challenges 228

Synthesis229

References 235
Index 277
(a) (b) (c)

(d) (e) (f)

(g) (h)

(i)

Plate 1 Examples of countershading, counterillumination, and reverse countershading (see page 42 for details).
 (a) (b) (c) (d) (e) (f) (g)

Plate 2 Treatment groups for experiment on countershading (see page 48 for details).

Plate 3 An aphid employing defensive secretions against a ladybird larva (see page 73 for details).

yellow-and-black
5

4
relative survival

YA BP YP YT YM YL

Plate 4 Relative survival of different types of artificial prey from an Plate 5 A moth displaying conspicuous warning signals (see page
experiment to investigate aposematism (see page 88 for details). 89 for details).
 (a) 1

Survival 0.8

0.6

0.4

20 25 30 35
Week
(b)
2,500

2,000
Count of nestlings ringed

1,500

1,000

500

0
17 18 19 20 2122 23 24 25 26 27 28 2930 31 32 33 34 35 Week
May June July August September

Plate 6 Seasonal variation in emergence of avian fledglings from nests and survival of different types of artificial prey in an experiment to explore
how naive individuals might affect selection for aposematism (see page 96 for details).
(a) (b)

Plate 7 A beetle and a moth that may be Müllerian mimics (see page 104 for detail).

Plate 8 A Heliconius butterfly sitting on a Heliconia flower (see page 105 for more on these butterflies).
Madrean

17 11 3 2 31 9 44 43

Desert

45 53 62 55 57 18 47

Texan

15 25 58 64 66 48 12 51 59 30 38

Tropical

27 7 22 21 5 8 42 41 1 26 10

Eastern

14 32 49 23 4 35 39 34 33 36 6 63

Western

28 20 40 37 67 29 24 50 16 46 61 65 60 19 13 54 52 56

Plate 9 Velvet ant mimicry rings (see page 106 for detail).

Plate 10 Above are three millipedes of the Apheloria clade; below are three co-mimics (see page 107).
(a) (b) (c)

H. sapho candidus

H. erato peterivana H. erato emma H. erato favorinus Polymorphic H. eleuchla eleusinus

forms of
H. cydno alithea

Plate 11 Three field experiments to test Müllerian mimicry in butterflies (see page 111).

(a) (b)

Plate 12 Right are the different morphs of artificial butterfly used in an experiment on Müllerian mimicry; left is an example of one as they were
deployed in the field (see page 112).

Plate 13 Below are hand-crafted clay models designed to resemble two conspicuous and one inconspicuous frogs shown above (see page 113
for detail).
 (a) Müller’s “fixed nk” (b) Optimal Sampling Strategy
species 1 species 2 species 1 species 2

Plate 14 Predictions of the spatial distribution of different prey colour morphs from models employing different assumptions about predator
behaviour (see page 121 for details).

Plate 15 A moth whose wing patterning is reminiscent of two flies feeding on a bird dropping (see page 151 for detail).

1.0
0.9
Proportion of individuals

0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
20 30 40 50 60
Latitude (degrees) of study sites

Proportion model (n = 356)

Proportion mimic morph (n = 348)

Plate 16 A transition zone between mimic and non-mimetic forms of a butterfly (see page 156 for detail).
(a) (b)

Plate 17 Left is a flesh-fly mimicking weevil; right is a potential flesh-fly model (see page 160 for details).

♀ ♂ (a) Original frame

f. cyrus

♀ ♂♀
f. polytes Pach. aristolochiae

(b) Motion map

♀
♂♀
f. theseus Pach. aristolochiae
black from

♀ Plate 19 Above is a frame from a video of several zebra moving in

♂♀ the same direction; below is the output of a computational model of
f. romulus Pachliopta hector motion detection applied to this image that implies a patchwork of
motion signals in different directions (see page 210 for details).
Papilio polytes (Batesian mimic)

Plate 18 Females of the non-mimic form of a butterfly look like the

males (top line); whereas three mimetic forms (first column) look like
distantly related species (second column – see page 168 for details).
 Introduction

In 2004, the first edition of ‘Avoiding Attack: The the ecological factors that the trait responds to, as
Evolutionary Ecology of Crypsis, Warning Signals, and well as co-evolutionary considerations. Each chapter
Mimicry’ by Ruxton et al. was published. The book then finishes with an open section where we outline
aimed to provide a systematic and up-to-date review what we see as key unresolved challenges and pro-
and synthesis of widespread anti-predator defences. pose some more-or-less easily achievable potential
In it, we focussed on sensorially mediated defences avenues for resolving these. In addition to our sub-
and the many factors that underpin these adapta- ject focus, we have broadened our author-base too,
tions, aiming to set out the state-of-understanding bringing Will Allen on board to compensate for sen-
in the fascinating world of anti-predator adapta- escence in the original three.
tions, and highlight which topics within the field As with the previous edition, our own interest in
seem most ripe for further investigation. predator–prey interactions remains as strong as
Since the publication of the first edition, many ever, but we would also argue that general scientific
research opportunities have been realized, and our interest in anti-predatory defences and signals has
understanding of the diverse and captivating strat- never been greater, as evidenced by recent high-
egies that have evolved in nature has developed profile reviews and special editions of well-known
significantly. In light of this, we here present a sec- journals (Cuthill et al., 2017; Caro et al., 2017). There
ond edition of Avoiding Attack. We have strived to are very few species for which perception by their
update and develop our previous work, particularly predators or prey is not a major influence on fitness.
in areas where scientific advance has been most Hence the phenomenon of predation and the many
radical in the past 13 years, including: mechanisms sensory adaptations surrounding it remain of great
of crypsis, understanding among-species variation ecological and evolutionary significance.
in anti-predator defences through evolutionary Interactions between predators and their prey
genomics and phylogenetics, and the causes and can often usefully be broken down into a sequence
consequences of variation in secondary defences. of stages, beginning with 1) encounter (spatial and
Since the first book, emphasis in the field of anti- temporal proximity), and leading through: 2) detec-
predatory adaptations has also shifted from a very tion, 3) identification, 4) approach, 5) subjugation
mechanistic perspective to an integrated under- and, ultimately, 6) consumption (see Endler, 1991;
standing including broad evolutionary and eco- Caro, 2005). In the literature, anti-predatory defences
logical consequences of adaptations; we hope that employed by prey individuals ahead of subjugation
this new book broadens our previous focus to con- (stages 1–4) are typically referred to as ‘primary
sider these consequences more systematically. To defences’, affecting the likelihood of the predator
do so, most chapters now follow a similar structure physically contacting the prey. While much of this
where appropriate, first discussing the natural his- book focusses on primary defences, we also consider
tory and key examples of a defence, before shifting ‘secondary defences’—those adaptations which act
to explain the mechanism via which it works to pro- once subjugation (or contact) has begun (stages 5 and
mote survival, followed by the evolutionary history 6). We think this view of a predator–prey interaction
and costs and benefits of the trait. We then review as a sequence of stages (Figure 0.1) is a powerful

Avoiding Attack: The Evolutionary Ecology of Crypsis, Aposematism, and Mimicry. Second Edition. Graeme D. Ruxton,
William L. Allen, Thomas N. Sherratt, & Michael P. Speed, Oxford University Press (2018). © Graeme D. Ruxton,
William L. Allen, Thomas N. Sherratt, & Michael P. Speed 2018. DOI: 10.1093/oso/9780199688678.001.0001
2 AV O I D I N G AT TA C K

conceptual tool for understanding co-evolution of make some predictions. A predation event can end
investment by the parties involved, and so we at different points in the process, so for a prey type
devote the second section of this chapter to a fur- that has several defences that act at different stages
ther exploration of this sequence. of the process, the earlier-acting defence will be
There are no perfect defences. No animal has employed more often. This does not necessarily
evolved a defence that gives 100 per cent protection mean that the earlier-acting defence will be more
against all possible predatory threats. This is not honed by evolution; it probably does mean, how-
surprising. Predators will often evolve to counter- ever, that if the two defences have costs that are
act defensive developments in their prey, and imposed every time they are engaged, then high
defences can be expensive to maintain. A suit of costs can be maintained for the late-acting defence
armour might offer you protection from surprise rather than the early-acting one. The stakes get
attack from a medieval swordsman; but you can higher for the prey the deeper into the sequence a
easily imagine how exhausting walking about in predation event gets as it becomes less likely that
that armour might be, and how it might limit your the prey will be bailed-out by a later-acting defence.
ability to go about all sorts of daily activities (not to Similarly, the deeper into a predation event we
mention how it might limit opportunities for repro- move the more time and effort the predator will
duction). Thus we would expect organisms to invest have invested in that attack and the less easily it
judiciously in anti-predator defences. In the context should be convinced to give up on this particular
of the sequence of a predation event, we might prey item. Thus, we can see obviously very expen-
expect that if prey invest strongly in being able to sive defences—such as autotomy of body parts—
flee effectively from predators then they do not also occurring as very late-acting defences.
invest in costly secondary defences too. Aside from Naturally occurring predation is unpredictable in
simultaneous heavy investment in two defences space and time, in large part because we would
being unattractive from a fitness perspective, it may expect predators to exploit any predictability in
be impossible from a design perspective. Our notional prey to increase predation rates (and for prey to exploit
suit of armour likely makes fleeing from an onrush- predator predictability to reduce predation). Further,
ing swordsman impossible. Similarly, design trade- naturally occurring predation can often be adversely
offs exist between trying to hide from predators using affected by the presence of human observers. This
camouflage and signal to predators using conspicu- creates real challenges studying predation in the
ous coloration. field and, as a generality, these challenges increase
In terms of general trends across defences acting as we advance through the sequence of a predation
at different stages of the predation process, we can event. If we switch from the field to the laboratory

Pre-contact predation sequence Post-contact predation sequence

Encounter Detection Identification Approach Subjugation Consumption

Example Anchoresis Transparency Disruptive Deflection Death

defence: (seek refuge) camouflage feigning

Primary defences Secondary defences

Figure 0.1 The classical predation sequence from search to subjugation (continuous lines) and examples of anti-predator behaviours that have
evolved to inhibit specific stages of the sequence (dashed lines). These behaviours can arise pre-contact (as primary defences) or post-contact (as
secondary defences) with the predator. Behavioural responses range from inhibiting search (by fighting back or seeking refuge, for instance) to
inhibiting consumption (by death feigning, for instance).
 INTRODUCTION 3

then ethical issues often arise, especially when perceptual processes such as edge detection and
studying anti-predator adaptations in vertebrates. figure-ground segregation to achieve camouflage.
Again, these challenges generally increase as we Chapter 3 goes on to explore countershading, a
move deeper into the predation sequence. So this term which describes the common colour pheno-
book is in large part a celebration of the fantastic type where the surfaces of an animal that orient
ingenuity of generations of researchers in inventing towards the sun are darker than surfaces that face
an extraordinary diversity of imaginative solu- away from the sun. The number of camouflage-
tions to the challenges of our field. It greatly helps related mechanisms that might select for counter-
that the researchers you will find in our reference shading is considerable. We list six, including three
list are exceptionally diverse in their background: that fall under the banner of self-shadow conceal-
you will find individuals that would be comfortable ment, the hypothesis that animals gain protection
describing themselves as one or more of the follow- from predation by hiding their self-cast shadows.
ing: naturalist, ethologist, animal behaviourist, ani- Thus a major task of this chapter is in clearly explain-
mal ecologist, evolutionary biologist, vision scientist, ing and distinguishing these alternative mechanisms.
psychologist, physicist, biochemist, or mathemat­ Our interest in countershading is about light and
ical biologist. vision. Countershading occurs because ambient
light is generally highly directional in nature—there
is no other sensory system that is likely to be
Chapter summary
­subverted by some equivalent of countershading
One widespread primary defence is crypsis, an because there is nothing analogous to the direc-
umbrella term for the various strategies organisms tional nature of ambient light in other modalities—
use to prevent detection. These are covered in Section although there is strong directionality in extrinsic
1 of the book: Chapter 1 discusses background thermal radiation and the earth’s magnetic field,
matching, an adaptation that remains relatively under- these are not used in predator–prey interactions.
studied, perhaps because as ‘standard’ camouflage Concluding Section 1, we discuss the cryptic
it is unfairly considered less interesting than other nature of transparency and silvering in Chapter 4.
defences. Among other things, we aim to show how Transparency adaptations are only common in pela-
understanding background matching addresses gic habitats and it is important to note that they are
important issues in evolutionary biology, such as not the same as invisibility—reflection and refrac-
the maintenance of polymorphisms. As achieving tion still occur, potentially creating visible features.
highly effective background matching in a complex Interestingly, transparency and polarization vision
environment is very hard to achieve, there is room might be linked phenomena, but recently there has
for other types of crypsis. In Chapter 2 we explore been more scepticism of this idea. Silvering, by con-
disruptive camouflage, a topic that has perhaps trast to transparency, is only common in mesopel-
seen the largest advances since the first edition, agic habitats. Silvering presents a similar solution to
which contained a four-page review that concluded transparency, but allows organs to be opaque with-
there was little empirical evidence for disruption. out reducing crypsis; we cover some of the down-
While background matching aims to minimize the sides of this adaptation.
signal sent to the receiver, disruptive camouflage In Section 2 of the book, we go on to consider
aims also to increase the sensory noise sent to adaptations through which prey can avoid attack
receivers by creating false edges and features that after they have been detected. Prey species may
make detection or recognition of an animal’s out- evolve traits that render themselves unprofitable to
line and shape difficult. Detecting an object and rec- predators and ways of advertising these defensive
ognizing what it is are conceptually distinct pro­cesses qualities. Chapter 5 first covers the nature of
but they are functionally related, and while a preda- ­secondary defences, such as armour, spines, and
tor can detect that something is present without ­toxins. These are the adaptations that underpin and
­recognizing what it is, the reverse cannot occur. Dis­ co-evolve with the defensive advertising described
ruptive camouflage exploits features of its viewer’s in Chapters 6 and 7 on aposematism and mimicry.
4 AV O I D I N G AT TA C K

We thus need to understand secondary defences to cuss exciting prospects for non-visual and multi-
understand defensive signalling fully. Further, the modal mimicry that are beginning to emerge. Finally,
existence of secondary defences will affect selection Section 2 concludes with a discussion of elusiveness
on other defences in camouflage. Secondary def­ences in Chapter 8, and how prey signal this to predators
can protect the bearer but also may offer ­protection through pursuit deterrence and perceptual adver-
to other individuals that the predator subsequently tisement. These strategies are distinguished from
encounters. Plants are discussed more in this chap- aposematism on the basis that prey are not signalling
ter than any other. Camouflage is hard for plants their difficulty of capture rather than the presence of
because of the demands of photosynthesis and a post-capture defence. Elusiveness signalling is an
­limits on mobility. Their lack of mobility also robs area where theory is strong and there are now many
them of fleeing as a defence, or seeking shelter in empirical examples of elusiveness signalling, espe-
protective environments like burrows, factors that cially from terrestrial vertebrates.
favour heavy investment in secondary defences. In The final section of the book, Section 3, concerns
Chapter 6, we look at aposematism, where a sec- anti-predatory defences that are focussed around
ondary defence combines with a warning signal misleading predators in one way or another. We
that ­advertises possession of the defence to pred- first look at the evolution and maintenance of
ators. Aposematic signals normally seem to have a Batesian mimicry and masquerade in Chapter 9.
visual component. Aposematism has evolved many Here we proceed with a healthy degree of caution,
times and there has been convergent evolution pausing before ascribing resemblance between two
towards some signal phenotypes such as repeating species as mimicry: it might be shared ancestry, or
bands of dark colour interspersed with yellow, the result of sharing a similar environment (conver-
orange, or red. In recent years fascinating new find- gent evolution). We also aim to avoid story-telling
ings have emerged and the chapter focusses on on the basis of perceived resemblance of the putative
these discoveries, for example evidence that selec- model and mimic to humans—who are unlikely to
tion for aposematism changes spatio-temporally— have similar visual and cognitive systems to agents
particularly in non-tropical regions aposematism is of selection that lead to mimicry. Masquerade has
less effective at the time of year when many pred- attracted a great deal of interest in the past decade,
ators are young of that year that have not yet and we now review a number of these recent stud-
learned to avoid aposematic signals. ies, as well as their implications. By contrast, Batesian
Chapter 7 turns to look at the occurrence, evolu- mimicry has long been understood theoretically,
tion, and maintenance of defence through Müllerian and its properties have been supported by lab and
mimicry. Müller’s original theory made a lot of sim- field studies. These studies not only demonstrate its
plifying assumptions about the way in which pred- occurrence, but also the fine-scale predictions of the
ators behave when faced with unfamiliar prey, but theory (such as the nature of the underlying fre-
recent experiments and theory have helped identify quency dependence). Nevertheless, since the first
how and why predators decide whether to sample edition there has been some exciting progress on a
these prey. In so doing, this work has helped reveal a range of topics including uncovering the genetics of
much greater range of situations in which we might some polymorphic Batesian mimics, laboratory
find Müllerian mimicry. Most examples we find of experiments on the ways in which natural predators
Müllerian mimicry are invertebrates, although it is classify profitable and unprofitable prey which dif-
also well known in vertebrates such as poison dart fer in several traits, and experimental and theoretical
frogs and catfish. We discuss how mimicry rings and work on how and why imperfect mimicry is not
spatial mosaics are two particularly important and improved upon by natural selection. Fascinatingly,
interesting features of the biogeography of coloration; evidence thus far suggests that different mimetic
hybrid zones between morphs are an excellent place species use very different underlying genetic archi-
to study selection and speciation processes. Previous tectures. The role this has in explaining differences
work has been heavily focussed on avian predators, in mimicry systems is an interesting avenue of dis-
but they are not the only visual predators and we dis- cussion and future research.
 INTRODUCTION 5

In Chapter 10, we consider startle defences, pro- defences; examining, for each, aspects of natural his-
viding a new definition and discussing such fascin- tory, of mechanism, of ecology and evolution. Before
ating adaptations like eye mimicry. We also present we do this in earnest, however, we first pause (briefly)
a discussion on whether distress calls in prey spe- to consider that prey defences are often multiple;
cies might be startle signals. Chapter 11 then looks camouflaged prey may also be good at rapid move-
into deflective signals which influence the position ments that lead to escape from nearby predators,
of the initial contact of a predator with the prey’s chemically defended, warningly coloured, and mim­etic
body in a way that benefits the prey. Here the inter- prey may also use camouflage or employ tough
esting questions are how this mechanism interacts integuments to protect themselves from predation.
with other defences and how predators allow them- Hence here we briefly pose two general questions
selves to be manipulated in this way. We next present that should in our view permeate our thinking
a new chapter, Chapter 12 on dazzle camouflage, a about prey defences: how many defences should a
putative anti-predator defence that was only briefly prey deploy, and what kind(s) should it use?
noted in the first edition, but that has now been sub- Perhaps the simplest explanation for the use of
ject to intensive experimental research such that we multiple defences is that they act at the same time,
felt it deserved its own chapter. Readers may be and tend to maximize survival of a prey from an
familiar with dazzle from photographs of warships encounter with a predator. Camouflage and noctur-
painted with high contrast geometric shapes. The nal activity may both contribute to reduced appar-
idea is that bold patterns on moving prey may cause ency of prey to diurnal predators, hence combining
predators to misjudge the speed and trajectory of them may decrease risk of detection, perhaps syner-
prey, hampering pursuit and capture. Experimental gistically. Prey that deploy chemical defences often
support for this is very mixed, and we are also not use multiple toxins that act simultaneously. Examples
yet clear whether dazzle effects occur in natural sys- of simultaneously acting defences are not then hard
tems. This chapter aims to review what we cur- to find, but they may not be the whole story.
rently do and do not know about ‘camouflage in A more complex reason for multiple defences
motion’. Finally, in another chapter new in this edi- could be that prey defences have a predictable
tion, Chapter 13 discusses the intriguing defence of sequential organization, designed around the struc-
death-feigning, or tonic immobility, in which prey ture of attacks. As discussed earlier in the chapter,
adopt a relatively immobile state that can often be John Endler (1991) argued that an attack by a predator
visually reminiscent of a dead individual. Like deflec- on its (animal) prey is typically composed of a
tion, the interesting questions are how this mechanism sequence of six stages:
interacts with other defences and how predators
1) encounter (spatial proximity)
allow themselves to be manipulated in this way.
2) detection
Before moving on to these issues, we want to fin-
3) identification
ish this introduction by returning to the concept of
4) approach
predation as a sequence of events. We feel that the
5) subjugation
sequential aspect of a predator–prey interaction is
6) consumption
fundamental to a holistic view of linkages between
all the defences that we are about to consider indi- At each stage in this sequence the prey organism
vidually in the rest of the book. can put up one or more lines of defence with the aim
of preventing, interrupting and stopping the attack.
Interested readers are recommended to read Malcolm
The sequence of a predator–prey Edmunds’ (1974) and Tim Caro’s (2005) texts which
encounter and investment across both develop the idea that defences evolve to reflect
the typical sequences of events during an encounter
multiple defences
between predators and prey.
As discussed in the previous section, this book will An animal prey may, for example, hide to prevent
review the varied forms of anti-predator sensory encounter (stage 1), and detection (2), use masquerade
6 AV O I D I N G AT TA C K

and cryptic coloration to prevent detection (2) and an attack. The maths here can, of course, be applied
identification (3), perhaps form aggressive defensive directly to sequentially acting organismal defences
groups to prevent approach (4). They may alterna- (including plants, microbes as well as animals).
tively have a startle display or use vigilance and Defensive layers can operate at any of the stages of
rapid escape behaviours to prevent approach (4). predation that Endler envisaged. There may, further-
They may violently retaliate, perhaps using stings, more, be several defensive levels within a stage of pre-
spines, or bites, and/or deploying irritating or toxic dation. Thus a prey animal could use several defences
chemicals (secondary defences), to prevent subju- in sequence to prevent subjugation—for example
gation (5) and consumption (6). At each stage in the sharp spines, tough exterior, and secretion of irritating
sequence that Endler identified, one or more defences chemicals. We argue here that a sequential organiza-
could be deployed by a prey animal, and they will tion of defences provides a useful general framework
often operate sequentially, as some defences are within which to think about the organization of pro-
typically used only if earlier-acting defences have tective phenotypes.
failed to stop the predation event. We can now ask Thus we can now ask: when should a prey dis-
whether there is one or more general framework(s) perse defensive investment across defensive levels
within which we can begin to evaluate how many or put all of its investment in a single level?
and what kinds of defences prey need. Equation 0.1 illustrates the diminishing nature of
There are interesting parallels between the organ- incremental survival benefits as more and more
ization of biological and human military defences. defensive layers are added to a set of defences. The
Both concern protection of valuable yet vulnerable first layer in our example makes a much bigger
targets, seeking optimal deployment of costly defen- contribution to survival (here 0.5) than the second
sive ‘assets’. A relevant military tactic is ‘layered (0.25), the third (0.125) and so on. If, as seems
defence’ in which sets of defensive resources, likely, there are costs associated with defences then
such as border security, naval warships, and inter-­ we can see limits on the number of defences that it
continental ballistic missiles, are deployed in is optimal for a prey to invest in. Imagine, for
sequence; when a first line of defence fails against an example, that sequentially acting defences are
incoming threat a second line of defence activates to ­rel­atively cheap to construct and maintain, but at
­minimize further risk, and after that perhaps a third their best each provides little protection, i.e. the
or fourth defence. In the military theory literature, survival benefit saturates at a low value. Here,
layered defence has been described as follows multiple levels of defence could be optimal to pro-
(Wilkening, 2000). Suppose that there are m defence vide high levels of protection in a cost-effective
levels and Li is the probability that a warhead passes manner. In contrast, a very effective (and indeed a
through the ith layer of an army’s defence. The prob- very cost-effective defence) may be sufficient on its
ability that the target is not destroyed, Kw, is own, and effectively block the evolution of any
later-acting defences (this has been called strategy-
K w = 1 − L1L2 ...Lm (0.1)
blocking, e.g. Britton et al., 2007).
The probability that the target survives an attack We suggest that this integrated approach to defences
then increases with the number of defensive levels that act in sequence can be important in directing
and the efficacy of each level. Suppose the defence research questions. For example, much of the early
at each layer is only moderately successful, Li = 0.5. part of this book focusses on mechanisms that gen-
In the case of only one defence layer, the probability erate camouflage and crypsis. We can now ask: how
that the target is not reached and destroyed is often is camouflage so effective that it blocks the
Kw =1−0.5 = 0.5; for two layers Kw=1−0.52 = 0.75; when need for later-acting (secondary) defences, such as
there are four sequentially acting defensive layers, repellent toxicity (and hence, how often are camou-
however, KW is now much greater 1–0.54=1−0.0625 = flaged prey edible)? Or, how often is camouflage
0.9375. Hence, layering of several moderately effective cheap but relatively ineffective and hence followed
defences leads to a high probability of survival from by a sequence of further defences that act after
 INTRODUCTION 7

camouflage has failed, such as rapid escape and ­ efences may also tend to be less effective than
d
repellent toxicity? later-acting defences.
An integrated-sequential view of defences can be The relative efficacy of multiple defences is per-
informative too in helping to determine what kinds haps more easily studied in plants than many ani-
of defence should be invested in, not merely the num- mals, because risks from enemies can be more easily
ber of defences. An organism that protects itself by assessed over a long period in sessile organisms.
reliably deterring enemies before they make con- Thus in a meta-analysis Carmona et al. (2011) found
tact, for example by camouflage and crypsis or by that variation in traits that act at the earlier stages of
defensive grouping or warning coloration, suffers attacks (varied flowering times, growth and morph-
no injuries to its tissues at an encounter. In contrast, ology, and physical defences) predicted susceptibil-
an organism that relied only on chemical defence ity to herbivory more strongly than variation in
may survive encounters but would more likely pay forms of chemical defence, which offer a late stage
costs that follow from having its tissues ruptured of defence. This makes sense in our framework of
during the attack, or from loss of fluids and other sequential defences (equation 0.1); since effective
resources. Hence, might we expect selection to favour chemical defences are then the last line of attack,
investment in the ‘before contact’ primary defences, variation in these traits will be less influential on
over those that operate as secondary defences, dur- survival than variation in earlier-acting defences.
ing contact between prey and their predators? An This interpretation would be consistent with John
additional point is that the ‘lifestyle’ of an organism Endler’s (1991) view that it is in the interests of vic-
is surely influential on the kinds of defences that are tims to interrupt predation at early stages in order
deployed. Sessile organisms, for example plants to save energetic and time costs, and to reduce risks
and some animals such as barnacles, cannot use of injury (and see Broom et al., 2010).
movement as a defence, so hiding, evasion or rapid There are relatively few theoretical treatments of
variation in group size is not possible. The intrinsic layered defence in the biology literature. Notably
apparency of the organism in its situation may deter- Frank (1993) considered the importance of ‘sequen-
mine the overall investment and type of defences tial defence’ models of hosts and their parasites.
used (cf. plant apparency hypothesis, see Stamp, Here a parasite must successfully cross m barriers
2003). For example, many forms of cactus and adult to gain any benefit from infection. Frank compares
barnacles are both exposed and highly vulnerable this to models of simultaneous defence in which
to predation, and both defend themselves with there is only one defence layer, but with more than
robust physical defences, shells or spines respect- one defence in operation. Broom et al. (2010), in
ively that will prevent attack by a broad range of contrast, separate defence investment into the two
predators. general layers of primary and secondary defences.
A final consideration with respect to sequential They use analytical and numerical solutions to
defences is that earlier-acting defences will, by def- determine the different scenarios that favour alter-
inition, be deployed more frequently than later-acting native outcomes (no defence, both primary and sec-
defences (equation 0.1, and Endler, 1991). For a ondary investment, primary or secondary investment
given effectiveness of defence, therefore, earlier- only). One intriguing prediction is that in some cir-
acting defences will reduce per capita mortality cumstances there may be very little fitness difference
more than later-acting defences and therefore be between alternative strategies, so that diversifica-
under stronger natural selection. Can we then expect tion in defensive organization within clades may be
selection to work harder to make earlier-acting strongly influenced by processes of drift. Species
defences more efficient and effective than later-acting that hide and use camouflage may be as persistent
defences? Selection on predators for adaptations that as those that use warning coloration and toxicity to
overcome early-acting defences should also be defend themselves (and see Stamp & Wilkens, 1993).
stronger, making defences better in absolute terms, Recently, Bateman et al. (2014) looked at a two-prey,
but relatively no more effective. Earlier-acting one-predator system where prey can differ in their
8 AV O I D I N G AT TA C K

investment in primary and/or secondary defences. deterred by these traits (see Table 1 of Carmona et al.).
They sought to identify the condition where one Specialist herbivores were particularly deterred
prey type could invade a system where the other from damaging plants by physical defences and life
initially dominated numerically, but found that this history variation, whereas generalists were not
was critically dependent on the fine detail of how ­det­erred by these variables.
costs and benefits of different types of defensive We suggest then that an important empirical ques-
investment were described. tion is the extent to which different defences in nat-
So far we have considered the simple scenario in ural systems evolved independently, each to protect
which prey defend themselves against a single type against different types of enemy, or because they are
of enemy. Nature is unlikely ever to be that simple, deployed in sequence against enemies. What seems
however, and defence systems must protect prey to lack systematic investigation is whether stages in
against multiple enemies which may have different the sequence of an attack (early or late) tend to be
modes of attack, and require different kinds of more generalized and function across sets of enemy
defence (see the excellent review in Greeney et al., types or more specific and function only against spe-
2012). Animal prey may have to defend, for example, cific enemies. Endler (1991), and Broom et al. (2010)
against predators of different sizes and levels of have argued that primary defences tend to be applied
mobility and against enemies such as ants or parasit- generally to sets of different predators, whereas sec-
oids which have different modes of attack to large ondary defences may be more specialized. Cryptic
predators. Rojas et al. (2017), for example, recently coloration may, for example, protect individual prey
showed targeting of alternative defensive secretions against visual detection by a range of predators, but
by wood tiger moth (Arctia plantaginis) toward differ- secondary defences such as toxins could operate
ent kinds of enemy (here ants and birds). There are more specifically. We think that this is an interesting
field-based measures of animal prey survival based area for further research.
on alternative forms of protection (see for example In summary, while this book is largely structured
Lichter-Marck et al., 2015; Carroll & Sherratt, 2013); around individual types of defence: different forms
how­ever, the plant literature provides quantitative of mimicry, disruptive camouflage etc., we do not
insight of relevance. As part of their meta-analysis, wish to suggest that defences are best viewed in iso-
Carmona et al. (2011), determined whether alterna- lation. Hence at the outset we emphasize that it is
tive defences offered protection to different kinds of common for prey to deploy more than one form of
herbivores. Their results show support for the idea defence (see Blanchard & Moreau, 2017 for a highly
that multiple types of defence evolved to protect integrative approach to the biology of defences).
against different types of herbivore. Chewing insects, Three reasons for multiple defences have been iden-
for example, were deterred from damaging plants tified: simultaneous action to maximize survival at
(expressed as per cent leaf area lost, biomass reduc- a stage of encounter, organization across sequen-
tion etc.) by variation in gross morphology (size, tially operating defences, and multiple types of
number of branches etc.) and by variation in life his- enemy. There are no doubt others, and we encour-
tory variables (flowering time, growth rates), whereas age readers to keep in mind the potential for inte-
herbivores that pierce and suck plant fluids were not gration of defences as they read the rest of the text.
 C H A PT ER 1

Background matching

1.1 Introduction, definition, mechanism, preventing recognition’, which include for example,
and chapter overview disruptive coloration (Chapter 2) and masquerade
(Chapter 9).
In this chapter we describe the phenomenon known Visual crypsis thus incorporates multiple strat­
as ‘background matching’ in which an animal’s col- egies, including countershading, transparency and
oration resembles the background against which it silvering (which we discuss fully in Chapters 3
is observed. The adaptive function of such matching and 4 respectively) and must also include aspects of
in colour patterns is to reduce the likelihood that a behaviour that contribute to reduced likelihood of
prey will be detected by its predators, or a predator detection.
detected by its prey, though we focus on the former. Most obviously, though, the term crypsis invokes
Here we will review evidence that background- the phenomenon known as background matching,
matching coloration works in this manner. We sub- which describes situations where ‘the appearance
sequently consider how natural selection causes the generally matches the colour, lightness and pattern
evolution of background matching, focussing on the of one (specialized) or several (compromise) back-
nature of frequency dependent selection and poly- ground types’ (Stevens & Merilaita, 2011). The
morphism, including the role of so-called ‘search authors also offer an elaboration of this definition
images’ in predators. Finally, we consider ecological which both clarifies aspects of the definition and
considerations that apply generally in the study of describes the underlying mechanism, stating:
background matching, notably how variation in ‘Comparison of local features over the retina is used
colour patterns of backgrounds limits the effective- in subsequent visual processing to distinguish an
ness of this defence mechanism. We argue however object from the background (called figure-ground
that visual background matching need not be perfect segregation). Thus if the appearance of an animal
in order for detection to be hindered. One of several does not match its background closely enough, a
reasons for imperfect background matching is that viewer will potentially detect a marked deviation
organisms can be viewed against a range of different in the local features between the animal surface
backgrounds and so adopt appearance traits that and its adjacent surroundings. This facilitates the
offer some degree of matching against several of detection of the animal as something that is not a
these. We start, however, with a brief discussion of part of the background. Background matching is
the terminology that relates the mechanism of back- therefore an adaptation that decreases the deviation
ground matching to the wider categories of crypsis in features between the appearance of the animal
and camouflage. and its background to counteract the figure-ground
In the recent edited volume ‘Animal camouflage: segregation.’
mechanisms and function’ Stevens & Merilaita (2011) These definitions seem clear and effective to us;
propose that the term crypsis is used for strategies however, we make the following points.
that hinder detection by predators, whereas camou- First, many authors use the terms crypsis and cam-
flage refers ‘to all forms of concealment including ouflage interchangeably. However, we agree with the
strategies to prevent detection . . . as well as those terminological framework of Stevens & Merilaita

Avoiding Attack: The Evolutionary Ecology of Crypsis, Aposematism, and Mimicry. Second Edition. Graeme D. Ruxton,
William L. Allen, Thomas N. Sherratt, & Michael P. Speed, Oxford University Press (2018). © Graeme D. Ruxton,
William L. Allen, Thomas N. Sherratt, & Michael P. Speed 2018. DOI: 10.1093/oso/9780199688678.001.0001
10 AV O I D I N G AT TA C K

(2011) and keep some distinction between the terms likely to be more fruitful that a focus on describing
in this book. appearance types, a perspective we aim to follow in
We feel there is utility in the definitions adopted all our chapters on camouflage.
in Stevens & Merilaita (2009a) where crypsis refers In the next section (1.2) we consider the empirical
to traits that hinder detection, whereas camouflage evidence for background matching in natural sys-
is a broader term that refers to traits that hinder tems and discuss the distribution of this adaptation.
detection and/or correct recognition of a prey indi- We then consider evolutionary and co-evolutionary
vidual. By this argument camouflage would encom- aspects of background matching (1.3 & 1.4), before
pass all the mechanisms considered to function in turning to how background matching is affected
crypsis, disruptive coloration plus masquerade. We by the ecology of predator–prey interactions (1.5).
note however that this definition could also encom- Finally we consider unresolved questions and oppor-
pass mimicry of conspicuous, warningly coloured tunities for further research (1.6).
prey, which is not usually considered a form of cam-
ouflage (Chapter 9). There may be merit in this use 1.2 Empirical evidence of background
of the term camouflage since it is in a sense a form
matching
of concealment of identity. Other readers may, how-
ever, prefer to limit the use of the term camouflage Organisms that seem to humans to offer uncanny
to colour patterns that do not draw a predator’s levels of background matching are staples of nat-
attention to the organism per se. ural history TV programmes. However, to us, the
A second point is that the concepts of crypsis and most persuasive evidence for background-match-
background matching can readily be extended to ing adaptations comes from species that behav-
modalities other than vision (as considered in depth iourally select their microhabitat and orientation
by Ruxton, 2009), and indeed could be applied so as to enhance similarity to features of the back-
multi-modally. Conceptually, if the environment in ground, and species that change aspects of their
the absence of the organism stimulates the senses of appearance in ways that enhance background
an observer in characteristic ways, then background matching. For example, many freshwater fish are
matching in another organism involves aspects of able to adjust their pigmentation to match the
the organism stimulating the senses of the observer background (Kelley et al., 2017; Kelley & Merilaita,
in ways that resemble the background and thus hin- 2015). Similar observations have been made for
der detection of the focal organism. cuttlefish (e.g. Buresch et al., 2011; Hanlon et al.,
A third point we make here is that although the 2009). Other bottom-dwelling fish actively choose
above description of background matching is substrates on which they achieve good back-
couched in terms of a cryptic animal, there is good ground matching (Tyrie et al., 2015). Lovell et al.
evidence that crypsis can apply to plants as well as (2013) demonstrated that birds that lay their eggs
animals (see Niu et al., 2017 for evidence of crypsis in simple scrapes can preferentially select sub-
related to local background matching in different strates that maximize background matching of
morphs of a plant), just parts of organisms, groups their eggs. Moths landing on bark often orientate
of organisms, or indeed to objects made by animals according to patterning of the background in a
(e.g. see Bailey et al., 2015 for evidence that some way that enhances background matching and hin-
avian nests are adapted for improved background ders detection (Kang et al., 2012).
matching). There is also experimental evidence that organisms
Finally we agree with Merilaita et al. (2017) that that appear to humans to be highly background
camouflage, including background matching, con- matching are also challenging for non-human
sists of a suite of adaptations to the perceptual and observers to find (e.g. Merilaita & Dimitrova, 2014).
cognitive systems of receivers that aims to reduce the The classic example of selection by predators driven
signal-to-noise ratio of stimuli utilized by searchers by the degree to which prey match the background
to detect, localize, and identify targets. We also agree remains the peppered moth Biston betularia, a poly-
with them that a focus on mechanistic functions is chromatic species with light and dark morphs. As
 B ackg r o u nd m atc h ing 11

coal burning increased throughout the industrial degree of background matching of different targets
revolution dark melanic morphs increased in fre- within them is difficult. Fully quantifying the
quency relative to the light morph, supposedly appearance of even a very simple background (e.g.
because they suffered lower predation at daytime the sky or a snowfield) is challenging. We can
resting sites on dark surfaces. Since 1970, with the describe the average colour and average brightness
introduction of pollution controls, there has been a relatively easily, for example—but quantifying spa-
rapid reversal with pale morphs becoming more fre- tial and temporal variation in appearance is much
quent, thought to have been caused by visual pred- more challenging (Allen & Higham, 2013). A com-
ators selecting against melanics at rest on today’s less plete quantification of the level of background
sooty surfaces. After decades of uncertainty about the matching of a single target for a single observer in a
underlying mechanism(s) driving morph fluctu- single snapshot would require measuring all of the
ations, Michael Majerus conducted a properly con- different visual parameters on which objects can be
trolled experiment to test the predator selection separated from the background (e.g. colour, lumi-
hypothesis, releasing 4864 moths of different morphs nance, edges, and conjunctions of these such as
then recording resightings of them over a six-year shape and texture), for both the target and the back-
period. This experiment, reported by Cook et al. ground, as they would be perceived by the obser-
(2012a) after the death of Majerus, confirmed that ver, and with reference to how parameter changes
differential bird predation against melanic pep- affect detection probability. This is hugely challen-
pered moths occurred at a rate sufficient in magni- ging, and yet it still comes short as it ignores the
tude and direction to explain the recent rapid decline differences between the sensory systems of the mul-
of melanism in post-industrial Britain. These data tiple observers organisms aim to hide from, and the
provide the most direct evidence yet implicating temporal variation in the relationship between fore-
camouflage through background matching and bird ground and background attributes (e.g. the sun
predation as the explanation for the rise and fall of passing behind a cloud, movement in the target or
melanism in moths. background, or target movement to a different
Not all organisms are camouflaged. Part of the background). The usual approach taken is to meas-
explanation for this is the costs of adopting a cam- ure just one or two attributes known to be relevant
ouflage strategy. First, since background matching is (e.g. colour difference) and use this as a proxy for
enhanced by adoption of certain microhabitats and the overall level of background matching. This has
behaviours (e.g. benefits through stillness, or adopt- the consequence of restricting the range of predic-
ing certain orientations relative to background elem- tions that can be tested, so it is exciting to see
ents) then there may be ‘loss of opportunity’ costs to approaches being developed that make collection of
background matching. For organisms that seek out richer measurements more straightforward (e.g.
microhabitats that enhance background matching Troscianko et al., 2017), though there are still con-
there may be additional locomotion costs in search- siderable methodological challenges to overcome
ing for and reaching these microhabitats. There are (Maia & White, 2018).
also physiological costs associated with production of
the pigmentation required for background matching.
These costs may be recurring for organisms that 1.3 The evolution of background
change appearance in order to track changing back- matching
grounds, or in organisms that shed and replace their
1.3.1 Polymorphism of background-matching
outer covering.
forms
From the argument above we can predict that
background matching should be most prevalent in We now consider how background matching can
species whose natural range offers predictability be reconciled with the frequent observation of poly-
and uniformity of background, but this is a difficult morphic coloration in cryptic species. Ford (1940)
hypothesis to test. In large part the problem here is defined genetic polymorphism as the occurrence
that characterizing backgrounds and measuring in the same locality of two or more discontinuous
12 AV O I D I N G AT TA C K

forms of a species in such proportions that the rar- Nevertheless, if p12 is greater than one, then this
est of them cannot be maintained only by recur- indicates an overall ‘preference’ for prey type 1 over
rent mutation from the other forms. This section prey type 2; if it is less than one then the preference
will consider a number of different mechanisms is reversed. In the boundary case, no preference is
via which selection for background matching in a apparent. Most importantly, since p12 is equivalent
population might lead to polymorphism. The to the ratio of the per capita mortality of prey type 1
main mechanisms produce frequency-dependent to the per capita mortality of prey type 2, it can be
per capita predation risk on prey morphs, and care- used as a simple measure of the relative fitness of
ful exposition of this requires a detour into the these two prey types in the context of predation.
complex maze of terminology used to describe Say in a series of experiments, the ratio of prey
frequency-dependent selection. availabilities (N1/N2) is varied, the ratio of consump-
tions (E1/E2) is noted, and p12 calculated. If p12 is
unaffected by the ratio of prey availabilities, then we
1.3.2 Definitions related to frequency-
have a frequency-independent preference. Otherwise,
dependent predation if p12 increases as N1/N2 increases, then the predator
There is a strong body of experimental evidence for displays a frequency-dependent preference, such
frequency-dependent predation: increased per cap- that per capita predation rate on prey type 1 com-
ita predation risk for the common morph increases pared to the per capita predation rate on prey type
relative to less common morphs (e.g. Allen, 1989a; 2 increases as prey type 1 becomes more common in
Reid, 1987). Whilst there is general agreement that the overall population. If p12 (as a measure of rela-
frequency-dependent predation often occurs, iden- tive fitness) rises from below one to above one as
tification of the mechanism behind it in any particular (N1/N2) rises, then when prey type 1 is rare it will
case can be challenging because possible mechanisms be selected for over prey type 2, and vice versa when
may act together and because different mechanisms it is common (see Figure 1.1). We call this form of
can produce very similar behaviours. Here we focus frequency-dependence (pro-) apostatic selection
on the potential mechanisms related to detection (selection for the rare form). Conversely, when the
of prey, though there are additional post-detection preference for a prey type decreases from above 1 to
mechanisms that can also lead to frequency- below 1 as it becomes more common then such an
dependent effects (Endler, 1988; Greenwood, 1984; outcome is referred to as anti-apostatic selection (see
Sherratt & Harvey, 1993). Figure 1.1). Following Greenwood (1984), we reserve
Let us consider a situation where a predator the term potential pro-apostatic selection for a subset
includes two prey morphs of the same species in its of cases analogous to pro-apostatic selection where
diet (labelled prey types 1 and 2). Over a certain p12 increases with increasing N1/N2 but one form is
time period, the predator eats E1 of type 1 and E2 of consistently at a selective advantage over the other
type 2. The densities of the two prey types in the (and potential anti-apostatic selection in a similar
environment are N1 and N2. We follow Murdoch way). Naturally, you could invent more names for
(1969) and Cock (1978) (amongst others) in defining other outcomes (e.g. when p12 falls over a range of
the predator’s preference between their two prey values of N1/N2 and then rises—as predicted in a
types (p12) by: theory paper of Sherratt & MacDougall (1995)) but
this is perhaps going too far.
E1 N
= p12 1 (1.1) Of course, it is possible to consider cases in which
E2 N2 the two prey types are separate species. Once again,
Of course p12 is a relatively uninformative measure preference for one prey type may cross the value
of preference for one morph over another: in prac- of one as a prey type gets more common. Here the
tice it is often difficult to separate a genuine ‘liking’ analogous form of positive frequency-dependent
for one prey type over another from circumstantial predation has been called ‘switching’ (Murdoch,
details such as the true availability of alternatives. 1969).
 B ackg r o u nd m atc h ing 13

Be warned when reading the literature that not The abundance of terminology reflects the general
everyone adopts the definitions given above. The importance of frequency-dependent predation, and
term ‘positive frequency-dependence’ can be particu- the fact that it has been observed in many different
larly confusing—it typically refers to cases where contexts. Hopefully, Table 1.1 will help (modified
relative fitness of a prey type is positively related to from Allen, 1989a).
relative frequency, but has sometimes been inappro-
priately interpreted in the converse: attack rates
1.3.3 Positive selection for polymorphism
increasing on a prey type as it becomes more com-
mon (this is negative frequency-dependence). Most In order to examine the role of frequency-dependence
authors do not differentiate between ‘potentially in the evolution of polymorphic crypsis, Bond &
­pro-apostatic selection’ and ‘pro-apostatic selec- Kamil (2002) designed an ingenious experiment
tion’, and refer to both as apostatic selection (e.g. using birds that were trained to search a computer
Endler, 1988). Bond (1983) has called particular forms screen and peck at images of moths in return for a
of apostatic selection ‘matching selection’ and anti- food reward. The population of moth images from
apostatic selection ‘oddity selection’. Pro-apostatic which individual images were randomly selected
selection has also been called ‘unifying selection’ by to appear on the screen was allowed to evolve over
Pielowski (1959). ‘Reflexive selection’ (Moment, 1962; time via a genetic algorithm, with offspring being
Owen & Whiteley, 1986) is also a synonym for pro- slightly mutated versions of their parents. Selection
apostatic selection (Allen, 1988a, b). Bewildered? for crypsis was introduced by allowing greater

Anti–apostatic Pro–apostatic

b<1 b=1 b>1

Frequency
independent
V>1 preference for
prey type 1
Proportion of prey type 1 in diet

V=1

Frequency
V<1 independent
preference for
prey type 2

Proportion of prey type 1 in environment

Figure 1.1 Graphs showing how the frequency-independent and frequency-dependent components of predator preference can combine to
influence the relationship between the proportion of a prey type in the environment and its proportion in a predator’s diet. All nine graphs have
limits 0.0–1.0. Here we have considered just two prey types (1 and 2) and used the Elton–Greenwood model (Greenwood & Elton,1979) to
generate the graphs. The parameter b in the Elton–Greenwood model reflects the degree of frequency dependence in the preference (b < 1
anti-apostatic, b > 1 apostatic), while the parameter V reflects the underlying frequency-independent selection (V > 1 preference for type 1,
V < 1 preference for type 2).
14 AV O I D I N G AT TA C K

Table 1.1 A glossary of terms relating to frequency-dependent predation. Adapted from Allen (1989a).

Term Examples of use Meaning

Frequency-dependent Fisher (1930), Ayala & Campbell (1974) Selection that results in a positive or negative relationship between relative
selection (fds) fitness and relative frequency
Positive fds Levin (1988) Fds where relative fitness of prey is positively related to relative frequency
Negative fds Partridge (1988), Antovics & Kareiva Fds where relative fitness of prey is negatively related to relative frequency
(1988), O’Donald & Majerus (1988)
Apostate Clarke (1962) Rare morph maintained by apostatic selection
Apostatic selection Clarke (1962) Negative fds by predators in the absence of Batesian mimicry
Apostatic polymorphism Clarke (1962) Polymorphism maintained by apostatic selection
Pro-apostatic selection Greenwood (1985) = apostatic selection
Anti-apostatic selection Greenwood (1985) Positive fds by predators in the absence of Batesian mimicry
Potential apostatic Greenwood (1984) When the prey type taken is taken more often than expected by chance at all
selection frequencies, and this selection increases with frequency
Switching Murdoch (1969) = apostatic selection (especially when prey are different species)
Matching selection Bond (1983) Apostatic selection (in matching backgrounds)
Oddity selection Bond (1983) = anti-apostatic selection
Reflexive selection Moment (1962) = apostatic selection (on massive polymorphisms)
Reflexive polymorphism Owen & Whiteley (1989) Massive polymorphism maintained by reflexive selection

representation of the least detected ‘prey’ in suc- detect, then this will come to dominate the popula-
ceeding generations. Selection by the birds led to tion, reducing diversity. These experiments are a
increased crypsis and greater phenotypic variance convincing demonstration of predator-induced
compared to control populations that were sub- selection for prey polymorphism of appearance
jected to no selection. Bond & Kamil interpret their (albeit in an artificial system). Further, this poly-
results as a suggestion that polymorphism is morphism seems to be driven by pro-apostatic
caused by the birds displaying pro-apostatic selec- selection.
tion. That is, a given morph is less likely to be One of the most popular explanations for pro-
detected by the birds if the birds have little recent apostatic selection is the formation of search images
experience of that morph because it is rare in the (Tinbergen, 1960). The term ‘search image’ has been
population. This interpretation was strongly sup- used to describe an enormous host of activities
ported as both phenotypic variance and crypsis related to foraging. This has led some (e.g. Dawkins,
increased to higher levels in the experiments with 1971) to suggest that the term is too imprecise to be
avian predators than in experiments with com- useful. Whilst we have some sympathy with this
puter-controlled selection that eliminated any fre- viewpoint, we use the term here, but clearly define
quency-dependent effects. what we mean by a search image in the following
However, an earlier experiment using a similar section.
set-up demonstrated that polymorphism was not
an inevitable outcome of this paradigm (Bond &
1.4 Co-evolutionary considerations
Kamil, 1998). If all phenotypes are too easy for the
birds to detect then there will be no selection, since Simply speaking there are a range of counter-adap-
all phenotypes are always found when they are tations predators can adopt to combat background
presented to the birds. Conversely if a morph is matching by prey. Firstly, and least interestingly, they
introduced that is impossible for the birds to could switch to focus on prey that have poorer back-
 B ackg r o u nd m atc h ing 15

ground matching, or focus on foraging in places or of search images) are more likely to induce and
under lighting conditions (e.g. specific times of day) maintain a search image for that prey type. This will
when background matching is less effective. More lead to prey types being subject to disproportionately
interestingly, they could search the environment high predation when common.
more carefully (likely at a cost of how quickly they It is important to remember that formation of
can search the environment) and/or they could refine search images need not necessarily lead to stable
their search strategy to focus on the search for par- polymorphism. Formation of a search image for a
ticular prey characteristics, forming a ‘search image’ common prey type may reduce the per capita likeli-
for particular types of prey. We focus on these issues hood that a rare prey type is detected, but this reduc-
in this section. tion need not necessarily reduce this likelihood
below that of the common morph, and so may simply
slow rather than reverse a decline in the relative fre-
1.4.1 Search image formation as a means of
quency of the rare morph. However, search image for-
enhancing detection of cryptic prey
mation, and more generally pro-apostatic selection by
Following Dukas & Kamil (2001), we define a search predators on cryptic prey can provide an evolution-
image as 1.) a process of transitory attentional spe- ary pressure for the development and maintenance
cialization that results in enhanced detection ability of polymorphism in prey species.
for particular cryptic prey types or characteristics
(e.g. ‘a triangular shape on a wing-like feature’),
1.4.2 Control of search rate to enhance
and that 2.) follows from repeated visual detection
detection of cryptic prey
of an item over a relatively short timescale. This is
predicted to result in increased capture success for Frequency-dependent selection on cryptic prey can
prey that match the search image, but decreased also arise from predators’ control of their rate of
capture success for prey which do not match the searching the environment. There is likely to be a
search image. While this definition remains some- trade-off between how rapidly an area is searched for
what imprecise about an exact mechanism, a key prey and the efficiency of detection (fraction of prey
criticism of Dawkins (1971), it does allow us to be in the scanned area that is detected: for empirical
explicit about the types of processes underlying our support for this conjecture see Gendron & Staddon
conception of a search image. Equally importantly, (1983)). Further, cryptic prey may require a slower
we are explicit about the types of processes that do search rate to obtain a specified detection efficiency,
not lead to search images under our definition. These compared to a less cryptic prey type. Gendron &
include behavioural changes such as modification of Staddon (1983) argued that it would be optimal for
search paths, and strategic decision-making about predators to reduce their rate of search when cryptic
which detected prey items to accept or reject. prey types are common in the environment. However,
A search image is formed after repeated detec- when cryptic prey types are uncommon, compared
tions of a particular prey type (Pietrewicz & Kamil, to less cryptic types, the optimal strategy is to
1979), and requires further detections of that prey increase search rate. Although this will lead to a
type in order to be maintained (Plaisted & Mackintosh, greater fraction of cryptic prey being missed, this
1995; and we refer readers to the reviews of Ishii & will be more than compensated for by the larger
Shimada, 2010 and Skelhorn & Rowe, 2016). Though number of less cryptic prey discovered. Hence,
a feature of individual cognition, search image for- optimal control of search rate leads to a reduction in
mation may be influenced by observation of peers the per capita risk of detection to individuals of the
in social groups (White & Gowan, 2014). Search more cryptic morph when this morph is rare. In this
images should lead to relative protection for prey way, predator strategies involving control of the rate
types when they are rare—promoting heterogen- at which they search their environment can often
eity of prey types. This occurs because common explain empirical results that had originally been
prey types are more likely to be encountered so (pro- attributed to a search image effect (Allen, 1989b;
viding prey types are equally cryptic in the absence Guilford & Dawkins, 1989a,b).
16 AV O I D I N G AT TA C K

1.4.3 Comparing search image and search rate monomorphic population to a polymorphic popu-
mechanisms lation where all the morphs are equally cryptic. The
search-image mechanism suggests that frequency-
Optimal control of search rate leads to a reduction dependent selection will occur in the polymorphic
in the per capita risk of detection to individuals of case, whereas the search rate hypothesis does not.
the more cryptic morph when this morph is rare. The search rate model predicts entirely the same
However, per capita detection risk is always lower search rate in the two cases, and predicts that indi-
for the more cryptic morph than for the less cryptic viduals of all prey morphs will be equally vulnerable
morph (in the absence of search image effects). Thus to detection. Whereas the search image hypothesis
(in the absence of any other differences between the suggests that polymorphism will reduce the preda-
morphs), frequency-dependent predation acts to tor’s performance, search images will either take
increase the rate at which the relative frequency of longer to arise or will be less successfully main-
the more cryptic morph increases. However, its fre- tained because the encounter rate of individual
quency would still increase in the absence of this morphs decreases compared to the monomorphic
effect. When the less cryptic morph is rare, it does case. Hence, in this special case where the morphs
not receive the same protection. In such circum- have equal crypsis, polymorphism is advantageous
stances the predator should slow up its search rate so under the search image mechanism, but not the
as to improve its detection of the more cryptic prey. search rate one (Guilford & Dawkins, 1987; Knill &
This will not decrease its ability to detect the less Allen, 1995). Knill & Allen (1995) found that human
cryptic morph, indeed it may increase it. Thus, in ‘predators’ were less effective at detecting prey in the
contrast to search image formation, optimal search polymorphic case. Similarly, Glanville & Allen (1997)
speed does not lead to protection for the less cryptic found that human subjects were slower to detect
morph (which will always be discovered dispro- computer-generated prey displayed on a screen in
portionately often), and so polymorphism may not trials where prey were polymorphic compared to
be maintained. This difference allows distinction monomorphic trials. The prey types were assumed
between the search rate control and search image to be equally cryptic, although this was not dem-
mechanisms, even though both provide the ability onstrated explicitly. However, these results tend
to produce frequency-dependent selection. Under to support the suggestion that in these studies
the search image mechanism, the rank order of per frequency-dependent selection occurred through
capita detection risk of different morphs can change; search image formation rather than search rate
this is not true of the search rate mechanism. Hence, control.
other things being equal, in a situation where search There is no reason why search image mechanisms
images do not occur, then the more cryptic of the and optimal control of search rate cannot occur sim-
two morphs will always have a fitness advantage ultaneously. Such a situation was explored theoret-
and will steadily increase its frequency in the popu- ically by Dukas & Ellner (1993). Their model does
lation. Should a random perturbation to the fre- not explicitly use the phrase ‘search image’; how-
quencies of morphs in the population or change in ever, such a phenomenon is implicit in the assump-
selection pressures mean that the more cryptic morph tions of the model. They assumed that prey detection
is at a low level, then control of search rate will requires information processing, and that the preda-
act to strengthen the competitive advantage of this tor has a finite capacity for processing, which it must
morph, and so may reduce the likelihood of sto- apportion to detecting different prey types. Increasing
chastic extinction. In this very specialized set of the processing ability devoted to a given prey type
circumstances, optimal control of search rate act- increases the chance that an encountered individual
ing in isolation from other mechanisms (such as of that type will be detected. However, because of
search image formation) may play a role in the the finite processing ability available, increasing
maintenance of polymorphism. ability to detect one prey type can only be bought at
An interesting comparison between the two mech- the cost of reduced ability to detect other prey types.
anisms considered above involves comparing a The other key assumption of the model is that the
 B ackg r o u nd m atc h ing 17

more cryptic a prey type is, the more information fixed in ontogeny before turning to those organisms
processing capacity is required to achieve a specified that have the ability to change appearance. Finally,
detection level. These assumptions are motivated by we consider the effects of ecology on combining
consideration of neurobiological experiments mostly crypsis with other selection pressures.
on humans (see references in Bernays & Wcislo, 1994;
Dall & Cuthill, 1997; Dukas & Ellner, 1993; Dukas &
Kamil, 2001). This model predicts that when prey 1.5.1 Optimizing of background matching for
are very cryptic, then the predator should devote a single appearance in visually variable
all its processing capacity to detection of one type. backgrounds
However, as the conspicuousness of prey increases,
so the diet of the predator should broaden, as it A significant drawback to background matching as
becomes advantageous for it to spread its informa- a strategy for crypsis is that almost all organisms will
tion processing capacity across a wider range of prey be viewed against a variety of backgrounds with
types. When the predator divides its ‘attention’ varied appearances. Even if the habitat is physically
between several prey types, this division should not homogeneous, then temporal change in light condi-
necessarily be even. If prey are cryptic, then the tions will change the nature of the background that
predator should give more attention to the least the organism must attempt to match. Indeed, an
cryptic of the prey types; whereas if the prey are organism can be viewed against two backgrounds
generally less cryptic then the predator should attend simultaneously when viewed by two organisms
most to the least detectable type. with different visual sensory systems. This raises the
In summary, much heated debate has surrounded question, should the organism specialize by maxi-
the concept of search images over the last 30 years. mizing its matching to one particular background,
This debate has arisen because many other mechan- or should it seek a compromise that provides rea-
isms (most notably control of search rate) can also sonable crypsis against more than one background
produce very similar behaviour to search image for- but which is not maximally effective against any
mation. This had led some to overstate evidence of one of them. This was addressed using a simple
search image formation, without logically exclud- model by Merilaita et al. (1999). We outline a very
ing plausible alternative explanations. However, slight generalization of their analysis below.
more recent studies do seem to have demonstrated We assume two backgrounds (a and b). Let the
the existence of this mechanism, and it does appear probability of being viewed by a predator against
that search image formation (perhaps working in background a be Va. The corresponding probability
concert with other mechanisms, such as control of that the potential prey is not detected whilst in the
search rate) could potentially select and maintain predator’s view is Ca. The prey is always viewed
polymorphism in cryptic populations. This has against background a or b, so
been demonstrated in the laboratory, but not yet in Va + Vb = 1. (1.2)
the field.
Hence, the overall probability of being detected by
a predator is
1.5 Ecological considerations
D = Va (1 − Ca ) + Vb (1 − Cb ) . (1.3)
The key ecological issue for background matching
is the overwhelming majority of organisms will be The probability of escaping detection is
viewed against a range of different backgrounds.
E = 1 − D = 1 − Va (1 − Ca ) − Vb (1 − Cb ). (1.4)
Even species that have very narrow microhabitat
use (e.g. living on the bark of a particular tree spe- In some cases, there is likely to be a trade-off between
cies) will be viewed by observers from a diversity of crypsis against the two backgrounds, and improved
distances and angles, and under a range of different crypsis against one background can only be
light conditions. There are two options here: we con- bought with reduced crypsis against the other.
sider organisms whose appearance is more-or-less Mathematically, Cb is a declining function of Ca.
18 AV O I D I N G AT TA C K

Cb = f (Ca ) found, or the trade-off line is concave, like the broken

line in the figure, then the optimal strategy is to
df (Ca )
< 0. (1.5) specialize and maximize crypsis against one of the
dCa
backgrounds. In the figure, the optimal strategy is to
It is trivial to show that E has a turning point (i.e. maximize crypsis against a providing
either a maximum or a minimum) if there is a value
AVa > BVb , (1.8)
of Ca that satisfies
df (Ca ) −Va where B is the value of Cb corresponding to Ca = 0,
= . (1.6)
dCa Vb i.e. the maximal probability of not being detected
whilst in the predator’s view against background b.
Further, this point is a maximum if, at that point,
Otherwise crypsis against b should be maximized.
d 2 f (Ca )
< 0. (1.7) That is, crypsis should be maximized in the environ-
dC 2 a
ment where the product of the probability of being
Figure 1.2 describes this situation graphically. We viewed by a predator and the probability of that view
assume that the prey individual is free to adopt any not leading to detection is maximized.
value of Ca from 0 (which maximizes crypsis against Hence, both background specialization and com-
b but provides no crypsis against a), to a value A promise can be predicted, a key determining factor
(which maximizes crypsis against a but provides no being the shape of the trade-off between crypsis in
crypsis against b). If the shape of the trade-off curve one environment and another. We now look to give
is ‘convex’ like the solid line in the figure, and if a more ecological description of this shape. Imagine
there is a point where that line has gradient –Va/Vb, a prey organism that specializes in crypsis against
then the optimal strategy for minimizing predator background b, and so adopts strategy Ca = 0. If it can
detection is a compromise value that provides increase its crypsis against a by losing a relatively
some crypsis in both environments. Increasing Va smaller amount of its crypsis against b, then we
(or decreasing Vb) moves this compromise towards have the type of convex shape of trade-off curve
improving crypsis against background a, as we that can lead to evolution of an intermediate level of
would expect. However, if no such point can be crypsis that provides some protection in both envir-
onments. However, if a little crypsis against a can
only be bought with a relatively large decrease in
1 crypsis against b, then we have a concave shape and
Gradient = –Va background specialization is favoured. Merilaita
B Vb
et al. (1999) give simple examples of abstract back-
ground combinations that might lead to these two
different types of situation; these are reproduced in
Cb Figure 1.3.
Of course, the situation can become much more
complex than the simple example considered here.
The costs associated with being detected need not
be the same for detection against both backgrounds
(since, for example, the predator may be more effect-
0 ive at capturing prey against one background). The
0 Ca 1 trade-off curve can have both convex and concave
A
segments, or can incorporate straight line segments
Figure 1.2 The probability of not being detected when viewed and/or discontinuities, and many more backgrounds
against background b (Cb ) as a function of the equivalent probability than two can be considered. In such circumstances,
against background a (Ca ). If the function is convex (like the solid
line) and has a point where its gradient is −Va /Vb , then a compromise
the adaptive landscape will be much more complex
level of crypsis is optimal, otherwise the organism should maximize and there will be considerable potential for evolution-
crypsis against one of the backgrounds. arily stable polymorphisms. However, the essential
 B ackg r o u nd m atc h ing 19

(a) mal strategy is to maximize crypsis against back-

ground a. However, if there is a small change in one
of the parameters, such that AVa becomes slightly
smaller than BVb then this has dramatic consequences
and now the optimal strategy is to maximize crypsis
against b. This has important ramifications:

• Two types of organisms could have very similar

ecologies but adopt very different appearances
(b) because the difference in their ecologies means
that they lie on different sides of the knife-edge
described above and so adopt colorations that maxi­
mize their crypsis against different backgrounds.
• If the environment inhabited by a population
changes (perhaps only slightly), such that it moves
to the other side of this knife-edge, then individ-
uals in that population that previously had the
optimal choice of coloration can now find them-
Figure 1.3 Two hypothetical examples of heterogeneous habitats
and animals relying on crypsis through background matching. In (a) selves with a coloration that is far from the opti-
the habitat consists of two different microhabitats, one with circular mal. Evolution towards the new optimum may
and the other with square elements. The two outermost of the four be particularly challenging as colorations only
animals have adapted to the microhabitats with respective patterns slightly different from the ‘old’ optimal may still
only. The two animals in the middle, one with a circle and a square to
be selectively worse than the current situation.
the left and one with two hexagons to the right, represent
compromised adaptations for crypsis in both microhabitats. An example of this is shown in Figure 1.4. Here
Successfully compromised colorations give the trade-off between we have a concave trade-off function shown in
crypsis in these microhabitats a convex form. In (b) one microhabitat Figure 1.4a. This means that the turning point in
is black and the other is white. Again the outermost animals represent D is a maximum, and so the optimal strategy is
adaptations to one microhabitat only. However, this time the
one of the extremes. Imagine, first, that we are in
compromised colorations in the middle are apparently very poor,
making the trade-off between crypsis in the two habitats concave. a situation represented by the solid line in Figure
1.4b, Ca = A is the optimal strategy. However,
imagine now that Va is changed such that we
points of our analysis above will remain unchanged,
move to the broken line: now Ca = 0 is the best
specifically:
strategy. However a small mutation away from
1) The optimal cryptic strategy for an organism may Ca = A produces an increase in detection rate D.
be one that does not maximize its crypsis against Only a macromutation producing a Ca value less
any one of the backgrounds against which it is than β in the diagram would be selected over the
viewed, but rather provides some measure of strategy Ca = A.
crypsis against a suite of the backgrounds that it • Polymorphism within a population may be main-
is viewed against. tained by even slight fluctuations in the propor-
2) In some cases, specialization against one back- tions of different backgrounds that individuals
ground will be favoured, in which case this will experience, if those fluctuations continually move
be the background against which the organism the system across the knife-edge.
can maximize the rate of occurrence of viewings
In a follow up paper, Merilaita et al. (2001) reported
of it by a predator that do not lead to detection.
the results of experiments with captive birds search-
The case where specialization is predicted war- ing for artificial prey. Birds were faced with one of
rants further scrutiny. Let us return to our simple two background types (differing in the size of pat-
case of the two environments a and b. Let us imagine terning). There were three types of prey: one which
that AVa is only slightly bigger than BVb. The opti- matched the small background pattern, one which
20 AV O I D I N G AT TA C K

(a) between the two extremes of background matching

would be optimal. These results can be seen as sup-
1 portive of the theory presented in this section; how-
ever, further empirical study of this theory would
B be very welcome.
Gradient = –Va Houston et al. (2007) extended the modelling
Vb
approach of Merilaita et al. (1999) by allowing pred-
Cb
ators control of which background they chose to
search against. However, the key result of Merilaita
remained—under some combinations of parameter
values background specialism was selected in the
prey, under other parameter value combinations
0
0 Ca A 1 compromise solutions flourished.
An interesting situation arises when backgrounds
themselves can evolve in response to predator–prey
interactions occurring against them, as is the case
(b)
of flowers on which predators lurk to prey on the
flower’s pollinators. This situation was modelled by
Abbott (2010), with results predicting that flowers
D
should match the appearance of predators when
predator densities are high to prevent predators
overexploiting the plant’s pollinators, and when
pollinators are less discriminating as they are will-
ing to land on a flower even if it might contain a
predator. Floral polymorphisms could evolve that
both concealed and revealed ambushing predators.
The arguments in this section have many paral-
0 lels with the arguments of Edmunds (2000) on why
0 β Ca A 1
some mimicry systems display poor mimicry such
that humans can easily tell model and mimic apart.
Figure 1.4 (a) The trade-off between Ca and Cb is concave so
specialization is predicted. (b) When Va is such that detectability (D ) is He suggests that under some circumstances evolu-
given by the solid line then Ca = A is the optimal choice, but if Va tion will drive a mimic species to specialize on one
changes slightly so we move to the broken line, then Ca =0 is optimal. particular model species, whereas in other circum-
stances generalized mimicry of a number of differ-
matched the larger patterned background, and an ent models (necessitating imperfect mimicry of each
intermediate-sized pattern. On the small-patterned one) will evolve (see Barnard (1984) for a similar
background, small-patterned prey were most cryp- idea). Further discussion of the evolutionary pres-
tic (measured by mean time between prey discover- sures on the ‘perfection’ of mimicry can be found in
ies), followed by the intermediate pattern, which in Chapter 7.
turn was more cryptic than the large pattern. On the Empirical demonstration of a natural organism
large-patterned background, the small-patterned having compromise appearance is challenging. We
prey was least cryptic, but there was no difference have already discussed why quantification of the
between intermediate and large-patterned prey. degree of background matching to one particular
This means that in a situation where both back- background in challenging. In this case, we must
grounds were encountered with equal frequency, add that exploration of compromise appearance
the intermediately patterned prey would be the requires identification of the backgrounds involved
best protected of the three, and generally that cir- and the frequencies at which the organism is viewed
cumstances exist where some intermediate type by the observer against these different backgrounds.
 B ackg r o u nd m atc h ing 21

However, we have valuable empirical studies utiliz- ing the most-common background is an effective
ing more artificial situations. Merilaita & Dimitrova strategy for prey, but once predators respond to this
(2014) trained wild-caught birds to search for artifi- by forming a search image for that prey the condi-
cial prey against two different backgrounds in an tions exist for prey with alternative appearances
aviary and found that a compromise design offered to become advantaged via apostatic selection. The
as much protection as prey designed to provide a experiments of Bond and Kamil have another inter-
good match against one of the backgrounds (but esting aspect; when the background was changed to
sometimes found against the other background). a more coarse-grained structure where it was homo-
Sherratt et al. (2007) explored a situation where geneous on the spatial scale of targets, then pred-
humans were tasked with detecting prey presented ators switch from a strategy of extensively searching
on examples of two different backgrounds. They the whole environment to one where they focussed
found that in general habitat specialists performed primarily on looking for prey on the parts of the
better than compromise prey, but the extent of this environment with the currently most-profitable
benefit varied considerably depending on the nature background. This behaviour reduced the effects of
of the two backgrounds and the difference between apostatic selection and led to a population that
them, and there were some situations where a com- mainly just shifted in the relative frequency of
promise appearance could gain very similar protec- morphs that offered a good match to one of the
tion to a specialist. In a very similar study, Toh & background types.
Todd (2017) found situations where compromise It is also worth remembering the issue that some
prey had lower net detectability than background backgrounds are fundamentally more challenging
specialists. to detect targets against than others, regardless of
While most studies have considered very simple background matching issues, as demonstrated ele-
situations in which prey can be found on one of two gantly by Merilaita & Lind (2005) in experiments
backgrounds, Michalis et al. (2017) considered a with wild-caught birds searching an aviary for arti-
more realistic situation where there is continuous ficial prey against control backgrounds. Further,
variation in background appearance. They argue prey can preferentially select such habitats over
that, logically, in such a situation the optimal choice those that would confer a benefit through back-
for prey to maximize protection through background ground matching (Kjernsmo & Merilaita, 2012).
matching should be to match the background that
they are viewed against most frequently. They sup-
1.5.2 Changing appearance to enhance
port this prediction with evidence using artificial
background matching
prey pinned to trees and predated by wild-living
birds, and with experiments using human search- Some organisms (e.g. cuttlefish) are known to change
ing for targets on a computer screen. They note that appearance in ways that seem to improve back-
Bond & Kamil (2006) demonstrated that when birds ground matching on timescales of seconds. Other
were trained to search for cryptic prey against species can take minutes, hours, or days like some
backgrounds that showed fine-grained variation in fish (Akkaynak et al., 2017; Smithers et al., 2017)
background (i.e. such that the background changed and frogs (Kang et al., 2016). Others, such as crust-
on a spatial scale similar to that of individual targets), aceans and insect larvae, change ontologically in a
birds selected for a prey population composed of a way influenced by environment but controlled by
shifting portfolio of different morphs through apos- the timescale of their moults; still other species (such
tatic selection. Initially this seems counter to the as many Arctic vertebrates) show seasonal variation
arguments of Michalis et al.; however, they suggest (Stevens, 2016). With the exception of cuttlefish, this
that the difference results from the importance of form of crypsis has been very understudied, as
predator learning in Bond & Kamil’s experiments emphasized by a recent review by Duarte et al.
whereas theirs were designed to minimize learning (2017). From our perspective, there is much similar-
by having prey at low densities and in different loca- ity between issues raised by such within-individual
tions. In the absence of learning, therefore, match- change and within-population change discussed
22 AV O I D I N G AT TA C K

elsewhere in the chapter. However, additional ques- impairing crypsis, if the cryptic species has higher
tions worthy of more consideration relate to costs. acuity than its predators or if signalling to conspe-
First, exploring the costs of maintaining the ability cifics occurs under higher ambient light levels than
to sense background (or some correlate—such as predation events. For example, conspecific signal-
available food), and changes in appearance in ling can involve small multicoloured spots that do
response to this. Second, the costs of a time lag not match the pattern of the background but which
between a change in the background and a change blend over the coarser spatial frequency of predator
in responsive appearance; or the costs of decreased acuity to provide a good match to the background
reliability of correlates of background (such as the (for fuller discussion see Endler, 1991; Marshall, 2000).
link between time of year and snow cover for verte- It has also been demonstrated that background
brates that turn white in winter). Lastly, once these matching can be combined with aposematic signal-
costs are better understood, we can ask how unpre- ling by using striped patterns that blend to a cryptic
dictable change in background has to be for such colour at a distance, but are effective warning pat-
within-individual change in appearance to bring a terns when viewed close up (Barnett & Cuthill, 2014).
net advantage. As the review of Duarte et al. empha- Similarly, it has been suggested that certain avian
sizes, we are only just beginning to scratch the sur- alarm calls decay rapidly over distance, so nearby
face of these questions. conspecifics can be warned about an overflying bird
Eacock et al. (2017) describe how larvae of the of prey without alerting the predator to the caller’s
peppered moth (Biston betularia) respond to match position (Klump et al., 1986; Klump & Shalter, 1984).
the appearance of the twigs that they rest on when Some deep-water organisms may use red light for
not feeding. They observed that larvae that were sub- intraspecific signalling without giving their pos-
ject to exposure to a diversity of twig backgrounds ition away to their predators (a ‘private communi-
evolved a good match towards the appearance of cation channel’), since red light rapidly attenuates
one twig type rather than a compromise. The flexi- in water and (because of this) many deep-water
bility in appearance is suggested by the authors as organisms are not receptive to red light. Cummings
being linked to wind-based dispersal and ability to et al. (2003) have suggested that UV may have simi-
feed on a range of tree species, which means that the lar advantages to aquatic organisms as a short-range
background an individual will experience cannot be intraspecific signal, because of high signal degrad-
predicted over ontological timescales. However, the ation over longer distances due to UV’s high pro-
lack of compromise might be linked, we suggest, to pensity to scattering by water molecules. In such
reasonably strong fidelity to a given tree and thus situations, we must explain why the predator has
individuals uncommonly experiencing a range of not evolved its visual system to detect intraspecific
backgrounds (twig colours) on shorter timescales. signals between prey individuals. This is relatively
easy for generalist predators, since a diversity of
intraspecific signals combined with physical trade-
1.5.3 Combining background matching with
offs in the visual system would prevent evolution of
other functions
a system that could effectively detect the different
Since effective background matching depends on signals of the predator’s suite of prey types. Prey
the visual abilities of the viewer, this can mean that individuals on the other hand have no need to
an organism that does not appear cryptic to us is detect the signals of other species that share the same
cryptic to its predators. One obvious example arises predator, and so do not have the same constraint
from inter-specific differences in colour vision. If an on their visual system. Hence, for these predators
organism’s main predators are colour-blind, then being a generalist presents sensory constraints that
matching intensity of light is important to crypsis may prevent enhanced detection of ‘private chan-
but spectral matching is not. Similarly, detail in the nels of communication’. In contrast, explaining why
background that falls below the acuity threshold of a specialist predator has not developed the ability
predators need not be matched. This can present an to detect the intraspecific signals of its prey is more
opportunity for signalling to conspecifics without challenging, but not impossible.
 B ackg r o u nd m atc h ing 23

One reason that specialist predators do not enhance rainbowfish (Melanotaenia australis) who darken
their capacity to detect ‘private channels of commu- their colour to improve their degree of background
nication’ is that prey generally have shorter gener­ matching receive more aggressive interactions from
ation times than their predators, producing different rival males because skin darkness is also used to
rates of evolutionary change. Furthermore, intraspe- signal dominance. Subdominant males must there-
cific signalling may be essential for reproduction, fore pay a cost in aggression from rivals to achieve
where a prey individual that does not signal does better background matching. As subordinates were
not reproduce. In contrast, predator individuals observed to change colour less than dominant males
that can detect intraspecific signals would do bet- when placed on a dark background, this cost is
ter than those that could not, but those that cannot apparently more severe than the cost of increased
would still be able to contribute to the next generation predation risk.
(if the prey can also be detected by other means).
Both this ‘life-dinner’ argument and the difference 1.6 Unresolved issues and future
in generation times might allow the prey to evolve
challenges
signals faster than their specialist predator can evolve
to detect them. It is also worth remembering that An outstanding basic question is how the visual
specialist predators are uncommon, and generaliza- appearance of the background selects for different
tion over a number of prey types is the norm for anti-predator strategies. Dimitrova & Merilaita
predators (Brodie & Brodie, 1999a,b). (2009, 2011) show experimentally in a lab environ-
While the ways organisms have managed to com- ment that background matching is more effective
bine crypsis and conspicuousness are fascinating against more visually complex backgrounds, a result
and illuminating, the more usual situation is for con- confirmed in field experiments (Xiao & Cuthill,
flict. Many studies have shown that more exuberant 2016). This implies that simple backgrounds should
intraspecific signalling leads to higher predation select for alternative strategies such as warning
rates (e.g. Roberts et al., 2006). A neat example of the coloration. Whether this predicted relationship
trade-offs organisms often make to balance camou- between background complexity and defence strat-
flage and signalling was recently demonstrated by egy is seen in across-species patterns remains to be
Kelley et al. (2016) who showed that male western investigated.
 CH A PT ER 2

Disruptive camouflage

2.1 Introduction and overview a visual scene that contain features corresponding
to the object class of interest (where is it?), and rec-
Background matching coloration has several limita- ognition involves determining if a set of visual fea-
tions in protecting prey from detection and recog- tures corresponds to an object in a particular class
nition. For example the alignment of patterning (what is it?).
between the organism and background may be off- Like other anti-predator coloration strategies dis-
set, shadows may give away its form, and the out- cussed in this book, the concept of disruptive cam-
line of an animal may therefore remain quite visible. ouflage originated in the work of Poulton (1890)
This creates opportunities for alternative strategies and Thayer (1909), and went on to be substantially
for improving crypsis, one of which is known as dis- developed by Cott (1940). Writing about the cater-
ruptive camouflage. Disruptive camouflage involves pillar of the privet hawk moth Sphinx ligustri,
using coloration to hinder detection or recognition Poulton noted that its ‘stripes increase protection by
of an object’s outline, or other conspicuous features breaking up the large green surface of the caterpil-
of its body. A classic example of disruption is color- lar into smaller areas.’ (Poulton, 1890, p.42). Thayer
ation patterns that extend across the folded legs of a (1909) observed that objects with a very high level
frog (Cott, 1940; Figure 2.1). By forming a continu- of background matching are often still easily detect-
ous pattern between the upper and lower legs and able by their outline. For example, shapes cut from
then across the foot, disruptive theory proposes that a patterned wallpaper can be picked up off a sheet
the frog is harder to find or recognize. This anti- identical to the one that they were cut from fairly eas-
predatory benefit arises because the high-contrast ily because they have a conspicuous cut edge and
patterns make the ‘true’ interior and exterior edges pattern elements will likely misalign. There was early
that visual predators use to find and recognize prey appreciation of the critical importance of animals’
less apparent. outlines to detection and recognition, and there-
The most recent formal definition of disruptive fore the advantage that might be gained by using
camouflage is ‘a set of markings that creates the coloration to interfere with the perceptual processes
appearance of false edges and boundaries and hin- involved in finding object outlines (Cott, 1940). While
ders the detection or recognition of an object’s, or the task of finding and picking out an object with a
part of an object’s true outline and shape’ (Stevens clear outline seems quite straightforward to us as act-
& Merilaita, 2009b, p.1). This definition highlights ors, in terms of visual computation it is enormously
the dependence of disruption on the perceptual and complex (Marr, 1982), and it is no great surprise that
cognitive processing of receivers. Whereas counter- the colour and pattern of the object potentially makes
shading can be defined on the basis of phenotype the task more difficult in a number of distinct ways.
alone, and background-matching camouflage by the A major topic in this chapter is therefore explaining
relationship between the target and background, dis- the multiple mechanisms by which colour markings
ruptive camouflage must be explained with respect can create ‘false’ information and obfuscate ‘true’
to how it works to prevent predator detection or information to improve camouflage.
recognition. Detection involves finding locations in

Avoiding Attack: The Evolutionary Ecology of Crypsis, Aposematism, and Mimicry. Second Edition. Graeme D. Ruxton,
William L. Allen, Thomas N. Sherratt, & Michael P. Speed, Oxford University Press (2018). © Graeme D. Ruxton,
William L. Allen, Thomas N. Sherratt, & Michael P. Speed 2018. DOI: 10.1093/oso/9780199688678.001.0001
Discovering Diverse Content Through
Random Scribd Documents
bist und weil du mir den Brief verschafft hast. Auch möchte ich
damit deinem Vater eine Freude machen, der den Brief geschrieben
hat und Herrn Vanderbilt, an den er gerichtet war. Und nun stoßt alle
mit mir an, auf das Wohl des großen Mannes! Hoch soll er leben,
dreimal hoch!«
Die Tabakspekulation hatte einen glänzenden Erfolg und so war
auch hier der rechte Mann zur rechten Zeit erschienen, trotzdem er
erst hatte von weither kommen müssen und nur mit Hilfe eines
Schabernacks entdeckt worden war.
 Achtundzwanzigstes Kapitel.
Die Natur hat der Heuschrecke ein
Verlangen nach Pflanzenkost verliehen;
hätte der Mensch die Heuschrecke
geschaffen, so würde sie nichts als
Wüstensand fressen.
Querkopf Wilsons Kalender.

An der Mündung des Derwent liegt, ganz von Laubwäldern

umschlossen, die Hauptstadt Tasmaniens, das freundliche Hobart,
auf Hügeln, die allmählich zum Hafen abfallen und in deren
Hintergrund der Wellingtonberg zu majestätischer Höhe emporsteigt.
Die Gegend ist himmlisch schön; sie bietet mit ihrem Reichtum an
Formen und Gruppen, ihrer Farbenpracht, dem üppigen Grün, den
Buchten und Vorgebirgen, den anmutig welligen Hügeln, dem
leuchtenden Sonnenglanz und den mattschimmernden Fernsichten,
ein entzückendes Landschaftsbild.
Mir ist noch keine Stadt vorgekommen, in der alles so von
Sauberkeit und Ordnung strahlt, wie in Hobart; nirgends sieht man
hier baufällige, schäbige Häuser oder eingefallene Zäune; kein
Unkraut wächst in den Vorgärten; im Hinterhof der Armen liegen
weder alte Blechbüchsen noch zerrissene Stiefel oder leere Flaschen,
kein Kehricht ist im Rinnstein, kein Schmutz auf dem Bürgersteig.
Selbst die bescheidenste Hütte sieht aus wie gestriegelt und
gebügelt; jede hat ihre Blumen, ihre Schlingpflanzen, die sie
umranken, einen sauberen Zaun mit guter Tür, und auf dem
Fensterbrett liegt die wohlgepflegte Katze und schläft.
Der Kurator des Museums, ein Herr aus Amerika, war so
freundlich, uns die Sammlungen zu zeigen. Wir sahen dort
wenigstens ein halbes Dutzend verschiedener Marsupialia[7], unter
andern den ›Tasmanischen Teufel‹, der zu dieser Gattung gehört,
wie ich glaube. Auch ein Fisch war da, der durch Lungen atmet und
im Schlamm weiterlebt, wenn der Fluß austrocknet. Am
merkwürdigsten ist aber ein Papagei, der den Schafen nachstellt. Auf
einer großen Schafweide hat er einmal in einem einzigen Jahre
tausend Stück umgebracht. Da er aber nicht das ganze Schaf frißt,
sondern nur das Nierenfett, so ist die Ernährung des Vogels sehr
kostspielig. Er hackt das Fett mit dem Schnabel heraus und das
Schaf stirbt an der Wunde. Die Geschichte dieses Papageis ist für die
Entwicklungslehre von Wichtigkeit; sie zeigt, daß unter veränderten
Bedingungen eine wesentliche Neubildung stattfinden kann. Als man
zuerst die Schafzucht einführte, wurden gewisse Würmer vertilgt, die
des Papageis Hauptnahrung gebildet hatten. Der Hunger trieb ihn
dazu, Fleischreste zu verzehren, die er noch an den Schaffellen fand,
welche zum Trocknen auf den Zäunen hingen. Bald schmeckte ihm
das Nierenfett der Schafe am allerbesten, aber die Form seines
Schnabels hinderte ihn es sich zu verschaffen. Da kam ihm die Natur
zu Hilfe und bildete seinen Schnabel um, so daß er sich jetzt nach
Herzenslust vom Fett seiner Mitgeschöpfe nähren kann.
Wir fuhren durch ein blühendes, duftendes Zauberland nach dem
Armenasyl, einem geräumigen, bequem eingerichteten Heim mit
Krankenhäusern und dgl. für Männer und Frauen. Dort waren
Scharen der ältesten Leute beisammen, die mir je vorgekommen
sind. Man sah sich plötzlich in eine andere Welt versetzt – eine
unheimliche Welt, aus der die Jugend verbannt war, und in der nur
das Alter mit seinen zahllosen Runzeln gebückten Ganges
umherwandelte. Von den 359 dort untergebrachten Personen waren
223 frühere Deportierte; sie hätten ohne Zweifel aufregende
Geschichten erzählen können, wären sie mitteilsam gewesen.
Zweiundvierzig hatten das achtzigste Lebensjahr überschritten und
einige waren nahe an neunzig; das durchschnittliche Sterbealter ist
dort sechsundsiebzig Jahre. Nein, in einem so gesunden Orte
möchte ich nicht leben. Siebzig ist ganz alt genug – später wird die
Sache zu ungewiß. Jugend und Heiterkeit könnten verschwinden,
ehe man sich’s versieht – und was bleibt dann noch übrig? Nur ein
Tod bei lebendigem Leibe, der weder Wohltäter noch Befreier ist. –
Unter den 185 Frauen in jenem Asyl waren 81 vormalige
Strafgefangene.
Das Dampfboot machte uns einen Strich durch die Rechnung; wir
hatten gedacht, es würde, wie gewöhnlich, lange in Hobart
verweilen, statt dessen fuhr es nach kurzem Aufenthalt weiter, so
daß wir Tasmanien nur wie im Fluge zu sehen bekamen.
Den Abend und die darauffolgende Nacht brachten wir auf dem
Meere zu und erreichten am frühen Morgen die Stadt Bluff am
Südende von Neuseeland. Eigentlich hätten wir nun quer nach der
Westküste hinüberwandern müssen, die reich an den herrlichsten
Naturschönheiten ist; hohe Schneeberge und mächtige Gletscher,
wundervolle Seen und tiefe Buchten, die den norwegischen Fjords
nicht an malerischem Reiz nachstehen, daneben ein Wasserfall, der
neunzehnhundert Fuß tief hinabstürzt; man nennt diese Gegend die
Neuseeländer Schweiz. Doch die Umstände nötigten uns, diesen
Besuch auf unbestimmte Zeit zu verschieben, so leid es uns tat.

6. November. Ein prächtiger Sommermorgen mit glänzend blauem

Himmel. Hinter Invercargill kamen wir durch meilenweite grüne
Ebenen, die mit Schafen wie beschneit waren – ein hübscher
Anblick. Die Bewohner von Dunedin sind Schottländer. Auf ihrem
Wege von der Heimat in den Himmel haben sie hier Wohnung
genommen, weil sie glaubten, sie wären am Ziel ihrer Wallfahrt. Die
Einwohnerzahl der Stadt wird von dem Journalisten Malcolm Roß auf
40 000 geschätzt; ein Parlamentsmitglied behauptet dagegen, sie
betrüge 60 000 – aber Journalisten können nicht lügen.
Wir besuchten Dr. Hockin in seiner Wohnung. Er hat eine gute
Sammlung Bücher über Neuseeland und sein Haus ist ganz mit
Altertümern der Maori und Erzeugnissen ihrer Kunstfertigkeit
angefüllt. Von vielen früheren Häuptlingen der Eingeborenen besitzt
er Porträts in farbigen Holzschnitten. Es sind auch geschichtlich
berühmte Leute darunter. Sie tragen nichts von Barbaren an sich;
ihre schönen regelmäßigen Züge, der kluge Gesichtsausdruck, ihre
männlich edle Haltung berühren den Beschauer aufs angenehmste.
Die Ureinwohner von Australien und Tasmanien sehen wie Wilde aus,
aber diese Neuseeländer Häuptlinge gleichen römischen Patriziern.
Sogar die Tätowierung der Bildnisse ändert daran nichts. Die Linien
sind so schön, so leicht und anmutig gezeichnet, daß man sie nur
mit Wohlgefallen betrachtet. Nach den ersten zehn Minuten hat man
sich an die Farben gewöhnt, nach weiteren zehn Minuten möchte
man es gar nicht anders haben, und von da ab macht jedes
unbemalte, europäische Gesicht einen häßlichen, unedlen Eindruck.

9. November. Der Präsident des Künstlervereins führte uns durch

das Museum und die öffentliche Bildergalerie. Unter den Bildern, die
teils geschenkt sind, teils durch Kauf erworben, gibt es einige sehr
schöne. Von dort gingen wir in die Kunstausstellung, welche eben
eröffnet war und alljährlich stattfindet. Daß eine verhältnismäßig
kleine Stadt zwei solche Kunstanstalten und einen Künstlerverein
hat, ist in Australien etwas ganz Gewöhnliches. Letzterer besitzt ein
eigenes Gebäude, das die Mitglieder auf ihre Kosten errichtet haben.
Ich würde dies Gedeihen der Kunst begreiflich finden, wenn man
es mit einer absoluten Monarchie zu tun hätte, die sich die Mittel zu
ihren Zwecken nicht erst von den Abgeordneten bewilligen läßt,
sondern einfach das Geld benutzt, wozu sie will. Aber die Kolonien
haben republikanische Verfassung und allgemeines Stimmrecht – in
Neuseeland sogar auch für die Frauen. Sonst pflegen in Republiken
weder die Regierungen noch die reichen Bürger sehr geneigt zu sein,
zur Ausbreitung der Kunst beizutragen; in ganz Australien werden
jedoch Gemälde berühmter europäischer Künstler für die öffentlichen
Galerien angeschafft. Man kauft sie entweder aus Staatsmitteln oder
es sind Geschenke von wohlhabenden Bürgern, welche diese
obendrein bei ihren Lebzeiten machen – nicht erst wenn sie tot sind.
 Neunundzwanzigstes Kapitel.
Die Sündhaftigkeit des Fluchens liegt
nicht in den Worten, sondern in dem
Zorngeist, der sie gebiert. Das Kind fängt
schon an zu fluchen, ehe es noch
sprechen kann.
Querkopf Wilsons Kalender.
11. November. Auf der Eisenbahn. Die fahrplanmäßige Zeit
unseres Schnellzugs ist nur zwanzig und eine halbe Meile die
Stunde; aber man möchte gar nicht rascher fahren, um die
wechselnde Aussicht auf Meer und Land nach Wunsch genießen zu
können, wozu die behaglich eingerichteten Wagen die beste
Gelegenheit bieten. Sie sind weder englisch noch amerikanisch,
sondern ein Mittelding zwischen beiden, wie in der Schweiz. An der
Seite hin läuft ein schmaler Vorbau, der mit einem Geländer
versehen ist, so daß man während der Fahrt auf und ab gehen kann;
auch gehört zu jedem Waggon eine Wasch- und Toiletteeinrichtung.
Das nenne ich Fortschritt! Da zeigt sich der Geist des neunzehnten
Jahrhunderts! –
Die sogenannten Schnellzüge fahren in Neuseeland zweimal die
Woche. Es ist gut, wenn man das weiß, denn wer mit
Zwanzigmeilenschritten das Land durchmessen will, könnte leicht an
einem der fünf falschen Tage abfahren und in einen Zug geraten, der
nicht einmal mit seinem eigenen Schatten Schritt hält.
Mich erinnerten die angenehmen Bahnzüge in Neuseeland, um des
Gegensatzes willen, an die Zweigbahn von Maryborough auf dem
australischen Festland und die Bemerkungen, welche mir ein
Mitreisender auf der Fahrt über die Bahn und das dortige Hotel
gemacht hat.
 Ich war unterwegs eine Weile in das Rauchcoupé gegangen, wo
sich noch zwei Herren befanden, die beide rückwärts fuhren und an
den äußersten Enden des Wagens Platz genommen hatten. Ich
setzte mich dem einen gegenüber ans Fenster; dem Anzug nach
hielt ich ihn für den Prediger einer Dissentergemeinde, er hatte ein
gutes, freundliches Gesicht und mochte etwa fünfzig Jahre alt sein.
Unaufgefordert zündete er ein Streichholz an und hielt die Hand
davor, bis meine Zigarre in Brand war. Das weitere entnehme ich
meinem Tagebuche:
Um die Unterhaltung in Gang zu bringen, fragte ich ihn einiges
über Maryborough, worauf er mit sehr angenehmer, wohltönender
Stimme die ruhige und bestimmte Antwort gab:
»Es ist eine reizende Stadt, aber das Hotel ist ein wahrer
Höllenpfuhl.«
Ich sah ihn verwundert an; es kam mir sehr merkwürdig vor, daß
ein Prediger solchen Ausdruck gebrauchte.
»Jawohl,« fuhr er gelassen fort, »ein schlechteres Hotel gibt es in
ganz Australien nicht.«
»Schlechte Betten?«
»Nein – gar keine. Nur Sandsäcke.«
»Auch die Kopfkissen?«
»Versteht sich. Nichts als Sand – und obendrein kein guter; er
backt zusammen und ist nicht durchgesiebt, sondern grober Kies.
Man schläft wie auf Haselnüssen.«
»Gibt es denn keinen feinen Sand?«
»Die Hülle und Fülle. Man findet in hiesiger Gegend einen so
vorzüglich losen und lockern Sand wie sonst nirgends; aber, den
kauft der Wirt nicht. Er nimmt nur Sand, der sich zusammenbackt
und hart wird wie Stein.«
»Wie sind denn die Zimmer?«
 »Acht Fuß groß im Viereck; der Boden ist mit eiskaltem Wachstuch
bedeckt, auf das man treten muß, wenn man am Morgen aus dem
Sandloch kommt.«
»Und die Beleuchtung?«
»Eine Erdöllampe.«
»Die hell brennt?«
»Nein, so düster wie möglich.«
»Ich lasse meine Lampe gern die ganze Nacht brennen.«
»Das geht bei dieser nicht; man muß sie früh auslöschen.«
»Aber man könnte doch nachts Licht brauchen und sie im Finstern
nicht wieder finden.«
»O, man findet sie leicht – sie stinkt ganz abscheulich.«
»Ist ein Schrank da?«
»Nur zwei Nägel an der Tür, um sieben Anzüge aufzuhängen –
falls man so viele hat.«
»Eine Klingel?«
»Ist nicht vorhanden.«
»Was tut man denn, falls man Bedienung braucht?«
»Man ruft, aber es kommt niemand.«
»Wenn nun aber das Zimmermädchen den Wassereimer
ausgießen soll?«
»Dergleichen gibt es nicht. Außer in Sydney und Melbourne findet
man nirgends Wassereimer in den Hotels.«
»Ach ja, das weiß ich; es ist eine Eigentümlichkeit von Australien
und kommt mir sehr komisch vor. – Noch eins: ich muß morgen früh
im Dunkeln aufstehen und mit dem Fünfuhrzug weiter reisen. Wenn
nun der Hausknecht –«
»Den gibt es nicht.«
 »Oder der Portier –«
»Es ist keiner da.«
»Aber, wer will mich denn wecken?«
»Kein Mensch. Sie müssen von selber aufwachen und sich auch
hinunterleuchten. Es brennt kein Licht, weder im Gang noch
anderswo. Auf der Treppe würden Sie im Dunkeln den Hals
brechen.«
»Wer wird mir helfen mein Gepäck hinuntertragen?«
»Niemand. Aber, ich will Ihnen einen Rat geben. In Maryborough
wohnt ein Amerikaner, schon seit einem halben Menschenalter, ein
stattlicher Mann, auch wohlhabend und allgemein beliebt. Machen
Sie dessen Bekanntschaft; dann sorgt er für alles und Sie können in
Frieden schlafen. Morgens weckt er Sie zur rechten Zeit und bringt
Sie auf den Zug. – Wo haben Sie denn Ihren Geschäftsführer
gelassen?«
»In Ballarat. Er macht dort Sprachstudien; auch muß er nach
Melbourne fahren, um alles für Neuseeland vorzubereiten. Es ist
mein erster Versuch, mich allein durchzusteuern, und mir scheint,
das ist gar nicht sehr leicht.«
»Leicht? Wo denken Sie hin – besonders auf dieser Strecke, der
schwierigsten in ganz Australien. Dazu muß man ein ganz
ungewöhnlich praktischer Mensch sein. Aber, das sind Sie wohl
auch?«
»Ich – hm – ich glaube – indessen –«
»Ja, dann werden Sie es nun und nimmermehr allein fertig
bringen. Aber der Amerikaner wird Ihnen helfen, verlassen Sie sich
darauf. Haben Sie Ihr Billet?«
»Ja, zur Rundreise bis nach Sydney.«
»Aha! dacht’ ich mir’s doch! Sie fahren um 5 Uhr morgens über
Castlemaine – zwölf Meilen – weil der 7.15 Zug über Ballarat zwei
Stunden länger unterwegs bleibt. Aber Ihr Billet lautet auf Ballarat –
auf den zwölf Meilen ist es ungültig, und die Regierung gestattet
nicht –«
»Was kümmert es denn die Regierung, welchen Weg ich wähle?«
»Das weiß der liebe Himmel. In der Eisenbahnverwaltung läßt sie
sich nicht drein reden. Zuerst übergab man sie einer Gesellschaft
von Blödsinnigen, das war schlimm; dann rief man Franzosen ins
Land, die verstanden noch weniger davon; zuletzt übernahm die
Regierung die Verwaltung selbst, und nun geht alles rückwärts. Um
die Gunst eines Wählers nicht zu verscherzen, der vielleicht zwei
Schafe und einen Hund besitzt, baut man ihm eine Bahn wohin er
will. So kommt es, daß wir zum Beispiel in der Kolonie Viktoria
achthundert Bahnhöfe haben, darunter achtzig, wo an der Kasse
kaum zwanzig Schillinge die Woche einkommen.«
»Fünf Dollars? Sie scherzen wohl!«
»Es ist buchstäblich wahr.«
»Aber auf jedem Bahnhof sind doch drei oder vier Beamte
angestellt, die ihr Gehalt beziehen.«
»Natürlich. Glauben Sie mir, die Bahnlinie verdient nicht einmal
den Zucker in ihren Kaffee. Und gefällig sind die Beamten! Man
braucht nur mit irgend einem Lappen zu wehen, so hält der Zug
mitten in der Wildnis und läßt den Reisenden einsteigen. Das macht
alles Kosten. Und wenn in einer Stadt viele stimmberechtigte Bürger
sind, die sich einen schönen Bahnhof wünschen, so wird er gebaut.
Sehen Sie sich nur den Bahnhof von Maryborough einmal an!
Sämtliche Einwohner haben Platz darin, jeder kann ein Sofa für sich
haben und es bleibt noch Raum übrig. Und eine Uhr ist dort – ich
sage Ihnen, so etwas hat kein Bahnhof in ganz Europa aufzuweisen.
Zum Glück schlägt sie nicht und hat auch keine Glocken, denn in
Australien hört das Gebimmel und das ewige dong-dong Tag und
Nacht nicht auf. Na, bei so vielen Prachtbauten für die Eisenbahn
und Verlust beim Betrieb, können Sie sich schon denken, daß die
Regierung irgendwo sparen muß. Das Betriebsmaterial muß
herhalten. Achtzehn Güterwagen und für die Passagiere zwei elende
Hundelöcher, die so schäbig und liederlich eingerichtet sind wie nur
möglich, ohne hygienische Vorkehrungen, ohne Trinkwasser, mit aller
nur denkbaren Unbequemlichkeit – das ist der Zug von Maryborough
– und langsam geht er, wie eine Schneckenpost. So spart die
Regierung ihr Geld. Für Tonnen Goldes baut sie Paläste auf, in denen
man ein paar Minuten warten muß, und deportiert einen dann sechs
Stunden lang wie den gemeinsten Verbrecher, um die vergeudeten
Summen wieder einzubringen. Jeder vernünftige Mensch fühlt sich
gern unbehaglich im Wartezimmer, weil ihm dann die Fahrt in einem
hübschen, bequemen Zug eine angenehme Abwechslung bietet.
Aber gesunden Menschenverstand sucht man bei der
Eisenbahnverwaltung vergebens. Und dann steckt sie noch für die
zwölf Meilen eine Extravergütung ein, erklärt ihre eigenen Fahrkarten
für ungültig und – –«
»Nun aber, jedenfalls kann ich wenigstens –«
»Warten Sie nur, ich bin noch nicht fertig. Ohne den Amerikaner
würden Sie überhaupt nicht fortkommen. Zuerst sieht niemand Ihr
Billet an, der Schaffner läßt es sich erst zeigen, wenn der Zug schon
im Abfahren ist. Dann können Sie kein Extrabillet mehr kaufen, der
Zug wartet nicht, und Sie müssen wieder aussteigen.«
»Aber, kann ich es denn nicht dem Schaffner bezahlen?«
»Er ist nicht berechtigt Geld anzunehmen, und nimmt auch keins.
Es bleibt Ihnen nichts übrig – Sie müssen heraus. Der
Eisenbahnbetrieb ist hier das einzige, was ganz nach europäischem
Muster eingerichtet ist – nicht nach englischem, nein, wie auf dem
Festland. Alles wird genau so gemacht, sogar das Wiegen des
Gepäcks, diese Erzplackerei, fehlt nicht.«
Jetzt hielt der Zug an der Station meines Mitreisenden. Beim
Aussteigen sagte er noch: »Jawohl, in Maryborough wird es Ihnen
gefallen. Sie treffen dort sehr gebildete Leute. Eine reizende Stadt!
Aber das Hotel ist ein wahrer Höllenpfuhl!«
Als er fort war, wandte ich mich an den andern Herrn:
»Ihr Freund ist wohl Geistlicher?«
»Nein, aber er treibt theologische Studien.«
 Dreißigstes Kapitel.
Zeit und Flut warten auf niemand! Ein
prahlerisches, dünkelhaftes Sprichwort,
das sich eine Billion Jahre erhalten hat,
aber heutzutage nicht mehr gilt. Wir, mit
unsern elektrischen Drähten und den
Schiffen mit Wasserballast, kehren es um
und sagen: Niemand wartet auf Zeit und
Flut. –
Querkopf Wilsons Kalender.

Auf der Fahrt bis Christchurch glaubt man im heutigen England

zu sein – die Gegend sieht aus wie ein Garten. Christchurch selbst ist
eine englische Stadt, mit englischen Parkanlagen und einem
englischen Fluß, der sich gleich dem Avon durch die Landschaft
schlängelt – er heißt auch Avon, aber nach einem Manne, nicht nach
Shakespeares Fluß. An seinen grünen Ufern stehen die stattlichsten
und denkwürdigsten Trauerweiden der ganzen Welt. Sie machen
ihrer hohen Abkunft alle Ehre, denn sie sind aus Ablegern der Weide
gezogen, die einst Napoleons Grab in Sankt Helena beschattete. In
dem guten alten Städtchen finden sich alle Reize und
Annehmlichkeiten, alles Behagen und alle Heiterkeit eines idealen
Familienlebens beisammen. Man könnte sich einbilden, drüben in
England zu sein; es fehlt nichts als die Staatskirche und der
Unterschied der Stände.
Das Museum enthält Raritäten. Unter anderm sahen wir ein
schönes Haus der Eingeborenen aus alter Zeit; ganz getreu nach der
Natur, sowohl die grellen Farben als alle Einzelheiten. Jedes Ding war
am richtigen Platz, die hübschen Matten und Teppiche, die
künstlichen, wundervollen Holzschnitzereien, deren Muster und feine
Ausführung mit Recht unser Staunen erregen, wenn wir bedenken,
daß die Eingeborenen dazu keine bessern Werkzeuge hatten, als
Feuerstein, Nephrit und Muscheln sie ihnen lieferten. Auch Totems
waren da, Pfähle, auf denen sämtliche Ahnherren des Stammes,
einer über dem andern, abgebildet sind. Die abscheulichen,
häßlichen Teufel waren mit großem Geschick und vieler Liebe
geschnitzt; sie streckten alle die Zunge heraus und falteten die
Hände behaglich über dem Bauche, in dem sie die Ahnherren
anderer Leute begraben hatten. In dem Haus waren auch
ausgestopfte Eingeborene als Staffage, jeder wo er hingehörte, als
ob sie leibten und lebten, sowie ihr ganzes Hausgerät; dicht dabei
lag ein geschnitztes und reich verziertes Kriegskanoe.
Wir sahen auch kleine Götzenbilder aus Nephrit oder Beilstein, die,
aber nur von Eingeborenen hohen Ranges, um den Hals getragen
wurden; ferner Waffen und allerlei Schmuck, aus dem gleichen,
ausnehmend harten Stein ohne jedes eiserne Werkzeug verfertigt.
Durch einige Stücke waren kleine runde Löcher gebohrt; niemand
weiß wie das gemacht wurde, es ist ein Geheimnis, eine verlorene
Kunst. Wer jetzt in ein Stück Nephrit ein Loch gebohrt haben will,
muß erst zu einem Steinschneider nach London oder Amsterdam
schicken, wie man mir sagt.
Auch ein ganzes Skelett der Riesen-Moa bekamen wir zu sehen. Es
war zehn Fuß hoch und muß einen netten Anblick gewährt haben,
als es noch ein lebendiger Vogel war. Wie der Strauß brauchte das
Tier beim Kampf nicht den Schnabel, sondern die Füße. Einen
solchen Fußtritt zu bekommen, mag kein schlechtes Vergnügen
gewesen sein, etwa wie wenn man von Windmühlenflügeln in die
Luft geschleudert wird.
In den alten, längst vergessenen Zeiten, als das Geschlecht der
Moas noch auf Erden wandelte, müssen sie in großer Menge
vorhanden gewesen sein. Man findet ihre Knochen haufenweise in
ungeheuren Gräbern dicht beisammen liegen, und zwar nicht in
Felsenhöhlen, sondern in der Erde. Niemand weiß, wer sie dort
eingescharrt hat. Da es wirkliche Knochen und keine
Versteinerungen sind, können die Moas noch nicht so gar lange
ausgestorben sein. Merkwürdig, daß sie die einzigen Tiere
Neuseelands sind, welche in den so zahlreichen Legenden der
Eingeborenen gar keine Rolle spielen. Dieser Umstand ist sehr
bedeutsam und kann als Indizienbeweis dafür dienen, daß die Moas
seit vierhundert Jahren von der Erde verschwunden sind, denn der
Maori selbst ist nach der Ueberlieferung seit dem Ende des 15.
Jahrhunderts in Neuseeland ansässig. Der erste Maori kam aus
einem unbekannten Lande, ruderte in seinem Kanoe dahin zurück
und kehrte in Begleitung aller Stammesangehörigen wieder; sie
drängten die Ureinwohner in das Meer oder brachten sie unter den
Boden und nahmen die Inseln in Besitz. So sagt die Ueberlieferung.
Die Ankunft jenes ersten Maori ist begreiflich, denn jeder kann
schließlich unversehens an einen Ort kommen, wohin er nicht will;
aber wie der Entdecker ohne Kompaß den Weg nach der Heimat
zurückfand, ist ein Rätsel, das er mit sich ins Grab genommen hat.
Aus seiner Sprache ging hervor, daß er aus Polynesien stammte. Er
hat auch gesagt, woher er kam, aber da er den Namen des Ortes
nicht richtig buchstabieren konnte, fand man ihn nicht auf der Karte,
welche von klugen Leuten gemacht ist, die den Namen ganz anders
buchstabieren. Vor allem müssen doch die Namen richtig auf der
Karte stehen, so daß man sich auf sie verlassen kann; das übrige ist
Nebensache.
In Neuseeland haben die Frauen das Recht, die Mitglieder des
gesetzgebenden Körpers zu wählen, sie selbst dürfen aber nicht
Abgeordnete sein. Das Gesetz, welches ihnen das Stimmrecht
verleiht, wurde 1893 erlassen, als die Einwohnerzahl von
Christchurch sich nach dem Census von 1891 auf 31 454 belief. Die
erste Wahl, nachdem das Gesetz in Kraft getreten war, fand im
November statt. An der Wahlurne erschienen 6313 Männer und 5989
Frauen. Diese Zahlen liefern den Beweis, daß die Frauen der Politik
nicht so gleichgültig gegenüberstehen, wie man uns glauben
machen will. Die Gesamtzahl der stimmfähigen weiblichen
Bevölkerung Neuseelands betrug 139 915; hiervon zeichneten
109 461 ihre Namen in die Wahllisten ein – 78,23 Prozent von allen.
90 290 erschienen wirklich an der Wahlurne und gaben ihre
Stimmen ab – 85,18 Prozent. Eifriger erfüllen die Männer wohl kaum
ihre politischen Pflichten, weder in Amerika noch in andern Ländern.
Der amtliche Bericht spricht sich auch noch in anderer Beziehung
günstig über jene Wahl aus:
»Besonders bemerkenswert,« heißt es, »war der Geist der
Ordnung und Nüchternheit, der unter den Leuten herrschte. Die
Frauen wurden auf keine Weise belästigt.«
Bei uns in Amerika wird als Hauptgrund gegen das weibliche
Stimmrecht regelmäßig der Satz aufgestellt, daß die Frauen nicht zur
Wahlurne gehen könnten, ohne sich tätlichen Beleidigungen
auszusetzen. Dergleichen Weissagungen sind äußerst bequem. Seit
im Jahre 1848 die Frauenbewegung begann, haben die Propheten
nicht aufgehört zu verkünden, daß allerlei Unheil daraus entstehen
würde; aber in einem Zeitraum von fünfzig Jahren ist noch nichts
davon eingetroffen. Im Gegenteil, es ist den Frauen gelungen, aus
dem amerikanischen Gesetzbuch eine große Anzahl ungerechter
Verordnungen zu entfernen. Die Männer sollten alle Achtung vor
ihren Müttern, Frauen und Töchtern haben und sie mit ganz andern
Augen betrachten, nachdem sie sich in einer verhältnismäßig so
kurzen Frist im wesentlichen aus ihrer Leibeigenschaft befreit haben.
Ohne Blutvergießen hätten die Männer das nicht zustande gebracht;
wenigstens haben sie es bis jetzt noch nicht getan, vermutlich, weil
sie nicht wußten, wie sie es machen sollten. Aber selbst durch die
friedliche und höchst wohltätige Umwälzung, welche die Frauen
bewirkt haben, sind die Männer im Durchschnitt noch nicht zu
überzeugen gewesen, daß die Frau Mut, Kraft, Ausdauer und
Seelenstärke besitzt. Es gehört eben viel dazu, um den
Durchschnittsmann von irgend etwas zu überzeugen. Vielleicht wird
nichts imstande sein, ihn jemals zu der Erkenntnis zu bringen, daß
die Frau ihm im Durchschnitt überlegen ist, und doch scheint das in
mancher wichtigen Einzelheit wirklich der Fall zu sein, wie sich aus
Tatsachen beweisen läßt. Seit Anbeginn der Welt hat der Mann das
Menschengeschlecht regiert, und er wird zugeben müssen, daß die
Welt bis zur Mitte dieses Jahrhunderts im allgemeinen recht träge,
dumm und unwissend war. Jetzt sieht es schon weit weniger
unerfreulich darin aus, und es wird mit der Zeit fort und fort besser.
Die Frau hat gegenwärtig Gelegenheit zu zeigen was sie kann – und
zwar zum allererstenmal. Wo wird wohl der Mann nach abermals
fünfzig Jahren sein – das möchte ich wissen.
Im Gesetz von Neuseeland steht folgende Anmerkung:
»Wo in den Verordnungen das Wort Person vorkommt, ist die Frau
mit inbegriffen.«
Das nenne ich doch eine Beförderung! Durch solche Erweiterung
des Begriffs wird die weise Matrone, der eine fünfzigjährige
Erfahrung zur Seite steht, mit einem Schlage in politischer Hinsicht
gleichberechtigt mit dem einundzwanzigjährigen Bürschchen, das
kaum trocken hinter den Ohren ist. Die Einwohnerzahl der Weißen in
der Kolonie beläuft sich auf 626 000, die der Maori auf 42 000. Die
Weißen wählen siebzig Mitglieder in das Abgeordnetenhaus und die
Maori vier; auch die Maori-Frauen stimmen mit bei der Wahl ihrer
vier Abgeordneten.

16. November. Nach einem viertägigen angenehmen Aufenthalt in

Christchurch verlassen wir den Ort heute um Mitternacht. Mr. Kinsey
hat mir einen Ornithorhynchus geschenkt, den ich zähmen will.

Sonntag den 17. Gestern abend fuhren wir in der ›Flora‹ von
Lyttleton ab.
Ja, wahrlich! Und wer die Fahrt in der ›Flora‹ mitgemacht hat,
wird sie sein Lebtag nicht vergessen, wie alt er auch werden mag.
Das Schiff war auf wahnsinnige Art überfüllt. Wäre es in jener Nacht
gesunken, so hätte man für die Hälfte der Leute, die an Bord waren,
keinerlei Rettungsmittel auftreiben können. Wenn auch die
Schiffseigentümer in technischer Hinsicht keinen Mordversuch
gemacht hatten – von der moralischen Schuld konnte man sie nicht
freisprechen.
Mir wurde ein Verschlag des großen Viehstalls zugeteilt, in
welchem eine lange, doppelte Reihe Schlafkojen angebracht war,
immer zwei übereinander. Ein Kattunvorhang diente als
Zwischenwand; auf einer Seite befanden sich zwanzig Männer und
Knaben, auf der andern zwanzig Frauen und Mädchen. Das Loch war
so finster wie die Seele der Unionsgesellschaft und es roch darin wie
in einem Hundestall. Als das Schiff aufs hohe Meer kam und anfing
zu rollen und zu stampfen, fielen die Gefangenen in ihrer dunkeln
Höhle sofort der Seekrankheit zum Raube. Durch das was nun
folgte, wurden alle meine früheren Erfahrungen dieser Art völlig in
den Schatten gestellt. Und dann das Heulen und Stöhnen, das
Geschrei und Gekreische, die Stoßseufzer aller Art – es spottet
jeglicher Beschreibung.
Die Frauen und Kinder, auch einige Männer und Knaben
verbrachten die Nacht in dem Raume, weil sie zu krank waren, um
sich zu rühren; wir übrigen aber standen allmählich auf und begaben
uns nach dem Sturmdeck.
Nie zuvor war ich an Bord eines so abscheulichen Fahrzeugs
gewesen. Als wir uns im Frühstückssalon zwischen den auf dem
Boden und den Tischen dicht aneinander gelagerten, schwitzenden
Passagieren hindurchwanden, ließ der Geruch an Kräftigkeit nichts
zu wünschen übrig.
Beim ersten Anlegeplatz stiegen viele von uns aus, um ein anderes
Schiff zu benutzen. Nach dreistündigem Warten fanden wir denn
auch in dem ›Mahinapua‹ gute Unterkunft. Es war ein schmuckes
kleines Fahrzeug von nur 205 Tonnengehalt, reinlich, bequem, gute
Bedienung, gute Betten, ein guter Tisch und kein Gedränge. Die
Wellen warfen es hin und her wie eine Ente, aber es war sicher und
seetüchtig.
 Ganz früh am Morgen kamen wir an den ›französischen Paß‹, eine
enge Durchfahrt zwischen steilen Felswänden, die nicht viel breiter
aussah wie eine Straße. Die Strömung schoß mit rasender Gewalt
hindurch und das Schiff flog dahin wie ein Telegramm. In einer
halben Minute lag der Engpaß hinter uns und wir kamen an eine
breite Stelle, wo großartige Wasserwirbel in der Nähe von Untiefen
fort und fort ihre stolze Runde machten. Ich fragte mich, wie es dem
kleinen Fahrzeug wohl dabei ergehen würde. Das sollte ich nur
allzubald erfahren. Die Wirbel hoben es auf, warfen es mit
Leichtigkeit herum und landeten es ganz behutsam auf einer
weichen, festen Sandbank. So sanft war die Berührung, daß wir sie
kaum fühlten und gerade nur merkten, wie das Schiff erbebte, als es
zum Stillstand kam. Das Wasser war hell wie Glas, man sah deutlich
den Sand auf dem Grunde, und die Fische schienen im leeren Raum
umherzuschwimmen. Rasch wurden die Angeln herausgeholt, aber
noch bevor wir den Köder am Haken befestigen konnten, war das
Schiff wieder flott und segelte auf und davon.
 Einunddreißigstes Kapitel.
Laßt uns Adam, unserm Wohltäter,
dankbar sein, daß er den ›Segen‹ des
Müßiggangs von uns genommen und den
›Fluch‹ der Arbeit über uns gebracht hat.
Querkopf Wilsons Kalender.

Am 20. November erreichten wir Auckland und hielten uns einige

Tage in dieser schönen, sehr hoch gelegenen Stadt auf; man hat
dort einen Ausblick über das Meer, an dem man sich gar nicht satt
sehen kann. Wir machten wundervolle Spazierfahrten mit Bekannten
in der Umgegend. Von dem grasbewachsenen Kratergipfel des
Mount Eden schweift das Auge über eine weite und wechselvolle
Landschaft: dicht belaubte Wälder, grüne Hügel, blumige Wiesen
und dahinter lange ebene Strecken, auf denen sich hier und da
hohe, ausgebrannte Krater erheben. Weiterhin glänzen und funkeln
blaue Buchten, und in traumhafter Ferne schimmern die Berge
gespenstisch durch ihre Nebelschleier.
Gewöhnlich fährt man von Auckland aus nach Rotorua zu den
berühmten heißen Seen und Springquellen, welche als große
Merkwürdigkeit Neuseelands gelten; aber ich war nicht wohl genug,
um den Ausflug zu unternehmen. Die Regierung hat dort ein
Sanatorium errichtet, wo für den Touristen sowohl wie den Kranken
aufs angenehmste gesorgt wird. Der daselbst angestellte Arzt drückt
sich stets sehr mäßig aus, wenn er von dem Heilerfolg der Bäder bei
Rheumatismus, Gicht, Lähmung und ähnlichen Uebeln spricht;
dagegen preist er die Wirkung des Wassers als unvergleichlich in
Fällen von Trunksucht. Der Trinker wird unfehlbar geheilt, selbst
wenn die Krankheit bei ihm noch so chronisch geworden ist, ja, er
verliert sogar die Begierde nach berauschenden Getränken für alle
Zeiten. Sobald es nur erst allgemein bekannt sein wird, daß für die
Opfer des Alkoholismus hier sichere Rettung zu finden ist, werden
die Leute scharenweise aus Europa und Amerika herbeiströmen.
Diese ganze, wegen ihrer Thermalquellen berühmte Gegend
Neuseelands umfaßt eine Landstrecke von über 600 000 Morgen
oder etwa 1000 Quadratmeilen. Rotorua ist am besuchtesten; es
bildet den Mittelpunkt des schönen, gebirgigen Seedistrikts und dient
den Reisenden zum Standquartier bei ihren Ausflügen. Die Zahl der
Kranken ist groß und wächst beständig. Rotorua ist das Karlsbad
Australiens.
In Auckland wird auch der Kauri-Kopal verschifft, hauptsächlich
nach Amerika. Man bringt durchschnittlich 8000 Tonnen des Jahrs
zur Stadt. Unassortiert hat er einen Wert von 300 Dollars die Tonne;
die feinsten Sorten erzielen aber oft einen Preis von 1000 Dollars.
Das Harz kommt in Stücken vor, ist hart und glatt und gleicht dem
Bernstein; auch ist es, wie dieser, hellgelb bis dunkelbraun. Unter
den hellfarbigen Stücken hätte man einige für ziemlich gute
Nachahmungen roher südafrikanischer Diamanten halten können, sie
waren so wundervoll glänzend, glatt und durchsichtig. Der Kauri-
Kopal dient zur Bereitung von Lack und Firnis, er stammt von der
Kaurifichte und wird meist aus dem Boden gegraben, wo das Harz
seit Jahrhunderten liegt, da viele von den Bäumen, aus denen es
geflossen ist, der heutigen Vegetation nicht mehr angehören. Dr.
Campbell in Auckland erzählte mir, er habe schon vor fünfzig Jahren
eine Ladung nach England geschickt, aber ohne Erfolg. Niemand
wußte etwas damit anzufangen und man verkaufte es für fünf Pfund
Sterling die Tonne zum Feueranzünden.

26. November. 3 Uhr nachmittags abgesegelt. Der Hafen ist schön

und ungeheuer groß; noch stundenweit hat man Land ringsumher.
Der Tangariwa ragt empor – das ist der Berg, der, wie in Auckland
allgemein behauptet wird, überall gleich aussieht, man mag ihn
betrachten von welcher Seite man will. Ganz richtig – von welcher
Seite man will – mit dreizehn Ausnahmen …
Herrliches Sommerwetter. In der Ferne tummelt sich eine große
Schar Walfische. Der Staubregen, den sie emporspritzen, sieht zart
und luftig aus in dem rosigen Schein der untergehenden Sonne oder
wenn er sich abhebt gegen den dunkeln Hintergrund einer Insel, die
im tiefblauen Schatten von Sturmwolken ruht … Zur Linken erhebt
sich ein großer Felsblock mitten im Meer. Vor einiger Zeit stieß ein
Schiff, das im Nebel zwanzig Meilen aus seinem Kurs gekommen war,
bei voller Fahrt dagegen; 140 Menschen ertranken in den Wellen.
Der Kapitän nahm sich auf der Stelle selbst das Leben, ohne sich
erst zu besinnen. Mochte er schuld an dem Unfall sein oder nicht, er
wußte, daß die Gesellschaft, der das Schiff gehörte, ihn jedenfalls
sofort entlassen würde, um mit ihrer Sorge für die Sicherheit der
Passagiere Reklame zu machen. Dadurch wurde es ihm aber fast
unmöglich gemacht, seinen ferneren Lebensunterhalt zu verdienen.

27. November. Heute kamen wir bis Gisborne und ankerten in

einer großen Bucht, eine Meile weit vom Ufer. Die See ging hoch und
wir blieben an Bord. Ein kleiner Schleppdampfer, der vom Lande auf
uns zufuhr, war ein Gegenstand atemlosen Interesses. Er klomm bis
zum Gipfel einer Welle, schwankte dort einen Augenblick als ein
grauer Schatten wie betrunken hin und her, in dem vom Sturm
zerstäubten Wasser, tauchte plötzlich in die Tiefe und blieb so lange
unsichtbar bis man ihn für verloren hielt; dann schoß er auf einmal
wieder in ganz schräger Richtung empor, während das Wasser in
Strömen vom Vorderdeck herabstürzte. So trieb es der Dampfer die
ganze Zeit über, bis er unser Schiff erreicht hatte. In seinem Bauche
befanden sich fünfundzwanzig Passagiere, die zu uns an Bord
wollten – Männer und Frauen, meistens Mitglieder einer reisenden
Schauspielertruppe. Die Mannschaft war auf Deck in Südwestern,
wasserdichten Anzügen von gelbem Segeltuch und hohen Stiefeln.
Das Deck stand so schräg wie eine Leiter und schwankte auf und ab;
große Wellen sprangen fortwährend an Bord und rollten darüber hin.
Wir befestigten ein langes Seil an der Raanocke, hingen einen
höchst kunstlosen Korbstuhl daran und ließen ihn im weiten
Himmelsraum wie einen Pendel auf gut Glück hin und her schaukeln.
Im geeigneten Moment wurde er von geschickten Händen
hinabgelassen und drüben fingen zwei Männer am Vorderdeck die
gut gezielte Leine auf. Ein junger Bursche von unserer Mannschaft
saß im Korbstuhl, um den weiblichen Passagieren herüber zu helfen.
Sofort erschienen einige Damen aus der Kajüte, setzten sich ihm auf
den Schoß, und wir zogen sie über uns in den Himmel hinauf. Einige
Augenblicke warteten wir noch, bis das Schlingern des Schiffes sie
herüberbrachte, dann ließen wir das Seil herab und erfaßten den
Korb, als er eben das Deck erreichte. So brachten wir alle
fünfundzwanzig an Bord und schafften fünfundzwanzig unserer
eigenen Passagiere in den Schleppdampfer – darunter mehrere alte
Damen und eine Blinde – noch dazu ohne den geringsten Unfall. Es
war ein schönes Stück Arbeit.
Wir sind mit unserm Schiff sehr zufrieden, es ist hübsch und
geräumig, auch alles darin bequem und gut in Ordnung. In einem
Hotel kommt es wohl vor, daß man auf eine Ratte tritt, aber an Bord
haben wir lange nichts von Ratten gespürt, außer vielleicht auf der
›Flora‹; aber da waren wir mit wichtigeren Dingen beschäftigt. Es ist
mir aufgefallen, daß man nur noch auf Schiffen und in Hotels Ratten
findet, wo die abscheulichen chinesischen Gongs gebraucht werden.
Der Grund kann nur sein, daß die Ratten nicht nach der Uhr zu
sehen verstehen, um zu erfahren, in welcher Tageszeit sie leben,
und einen Ort fliehen, an dem sie nie wissen, wann das Essen fertig
ist.

2. Dezember. Montag. Von Napier nach Hastings benutzten wir

den Schnellzug, der in Neuseeland zweimal die Woche fährt. In
Waitukurau war zwanzig Minuten Aufenthalt und wir nahmen einen
Imbiß. Ich saß oben am Tisch, so daß ich die rechte Wand sehen
konnte, während meine Frau, meine Tochter und Mr. Carlyle Smythe,
mein Geschäftsführer, der Wand den Rücken zukehrten. Auf dieser
Wand hingen einige Bilder ziemlich weit von mir, so daß ich sie nicht
deutlich erkennen konnte, aber nach der ganzen Gruppierung der
Gestalten nahm ich an, daß eins derselben die Ermordung von
Napoleons des Dritten Sohn durch die Zulus in Südafrika darstellte.
Ich unterbrach das Gespräch, welches sich eben um Poesie,
Kohlköpfe und bildende Kunst drehte und wandte mich an meine
Frau mit der Frage:
»Weißt du noch, wie die Nachricht in Paris ankam –«
»Daß der Prinz ermordet wäre?«
(Ich hatte genau diese Worte im Sinn gehabt.) »Welcher Prinz
denn?«
»Napoleon – Lulu.«
»Wie kommst du eben jetzt darauf?«
»Ich weiß nicht.«
Höchst sonderbar! – Wir hatten uns auf keine Weise miteinander
verständigt. Die Bilder waren nicht erwähnt worden und meine Frau
konnte sie nicht sehen. Vor sieben Monaten waren wir nach einem
mehrjährigen Aufenthalt von Paris abgefahren, um diese Reise zu
unternehmen und meine Frau hätte an irgend eine Nachricht denken
müssen, die in jüngst vergangener Zeit nach Paris gekommen war.
Statt dessen dachte sie an ein Erlebnis bei unserm kurzen Besuch in
Paris vor sechzehn Jahren.
Es war ein deutliches Beispiel von Gedankentelegraphie, von
geistiger Wechselwirkung. Ich hatte die Idee aus meinem Hirn an sie
telegraphiert. – Woher ich das so bestimmt weiß? – Nun einfach
deshalb, weil es ein Irrtum war. Es ergab sich nämlich, daß jenes Bild
weder die Ermordung Lulus darstellte, noch überhaupt etwas, das
sich irgendwie auf Lulu bezog. Ich mußte ihr den Irrtum
telegraphiert haben, denn außer in meinem Kopfe war er nirgends
vorhanden.
 Zweiunddreißigstes Kapitel.
Der Selbstherrscher von Rußland hat
mehr Macht als irgend ein Mensch auf
Erden; aber das Niesen kann er doch
nicht zurückhalten.
Querkopf Wilsons Kalender.

Wanganui 3. Dezember. Unsere gestrige Fahrt war sehr

angenehm und dauerte vier Stunden. Meinetwegen hätte der Zug sie
auf acht Stunden ausdehnen können. Wenn man sich behaglich fühlt
und kein Grund zur Eile vorhanden ist, liegt mir gar nichts an
übergroßer Schnelligkeit. Nun kenne ich aber kein bequemeres
Beförderungsmittel als einen Neuseeländer Zug. Nirgends findet man
außerhalb Amerikas so vernünftig eingerichtete Eisenbahnwagen.
Rechnet man dazu noch den unausgesetzten Anblick des reizendsten
Landschaftsbildes und den fast gänzlichen Mangel an Staub, so kann
man nur jedem raten, der damit noch nicht zufrieden ist, er soll
aussteigen und zu Fuße gehen. – Würde er dadurch aber anderer
Meinung werden? Ganz gewiß. Nach Ablauf einer Stunde träfe man
ihn sicherlich bescheiden wartend neben dem Schienenstrang, und
er wäre froh mit dem Zug weiter fahren zu dürfen.
In der Stadt und Umgegend sieht man viele Leute zu Pferde und
hübsche junge Mädchen in luftigen Sommerkleidern, auch die
Heilsarmee und eine Menge Maori; Gesicht und Körper der älteren
sind meist sehr geschmackvoll bemalt. Jenseits des Flusses liegt das
Rathaus der Maori, ein großes, festes Gebäude, das von einem Ende
zum andern mit Matten ausgelegt und mit reichen, kunstvoll
verfertigten Holzschnitzereien geschmückt ist. Die Maori sind auch
sehr höfliche Leute.
 Einer der Volksvertreter gab mir die Versicherung, daß die
eingeborene Rasse nicht abnimmt, sondern sich im Gegenteil
langsam vermehrt. Dies ist ein neuer Beweis, daß die Maori, als
Wilde, auf einer hohen Stufe stehen. Ich weiß mich an keine
anderen Eingeborenen irgend welcher Rasse zu erinnern, die so gute
Häuser oder so starke, zweckentsprechende Festungswerke errichtet
hätten, die den Ackerbau so eifrig betrieben oder bei denen die
Kriegswissenschaft und Verteidigungskunst fast bis zu einer
Vollkommenheit gediehen wäre, wie man sie sonst nur bei den
Weißen findet. Zählt man hierzu noch ihre große Geschicklichkeit in
der Anfertigung von Booten und ihre Begabung für die dekorativen
Künste, so kann man sie höchstens noch als Halb- oder Dreiviertel-
Barbaren betrachten und ihnen eine gewisse Zivilisation nicht
absprechen.
Es ist schmeichelhaft für die Maori, daß die britische Regierung,
statt sie auszurotten wie die Australneger und die Tasmanier, sich
mit ihrer Unterwerfung begnügt hat. Auch nahmen die Engländer
ihnen nicht alles gute Land fort, sondern ließen ein großes Stück in
ihrem Besitz und schützten sie sogar vor den Landwucherern, wie es
die Regierung von Neuseeland noch heutigen Tages tut. Am
schmeichelhaftesten für die Maori ist jedoch, daß sie ihre
eingeborenen Vertreter sowohl im Ministerium wie im
gesetzgebenden Körper haben, und daß auch das weibliche
Geschlecht wahlberechtigt ist. Die Regierung ehrt sich selbst durch
diese Einrichtungen, denn bisher war es in der Welt nicht Sitte, daß
ein Eroberer mit den Besiegten auf so weitherzige Art verfuhr.
Die gebildetsten Weißen, die zuerst unter den Maori lebten, hatten
wirkliche Zuneigung für sie und eine hohe Meinung von ihrer
Gesittung. Ich will nur den Verfasser des ›Alten Neuseeland‹
anführen, sowie Dr. Campbell von Auckland. Letzterer hatte mit
mehreren Häuptlingen große Freundschaft geschlossen und wußte
viel Gutes von ihrer Treue, Hochherzigkeit und Großmut zu
berichten. Er erzählte auch, was sie sich für wunderliche Begriffe von
der Zivilisation der Weißen machten und wie komisch sie dieselben
beurteilten. Einer von ihnen meinte unter anderm, der Missionar
greife alles am unrechten Ende an und kehre das Unterste zu oberst.
»Hat er doch sogar gesagt, wir sollten aufhören, die bösen Götter
anzubeten und um ihren Schutz zu flehen; wir brauchten uns mit
Verehrung und Bitten nur noch an den guten Gott zu wenden. Das
hat ja weder Sinn noch Verstand! Ein guter Gott wird uns doch
keinen Schaden tun!«
Unter den Maori herrschte das ›Tabu‹ in so großartigem Umfang,
wie es für Polynesien paßte. Von einigen Verboten hätte man
glauben können, sie stammten aus Indien oder Judäa. Weder bei
den Maori noch bei den Indern durfte der gemeine Mann an einem
Feuer kochen, das von einem Mitglied der höheren Kasten benutzt
worden war. Ebensowenig gestattete man dem vornehmen Maori
oder dem vornehmen Inder sich des Feuers zu bedienen, an dem
der gemeine Mann seine Speise bereitet hatte. Trank ein Maori oder
ein Inder niederen Ranges aus dem Gefäß, das einem
Höhergestellten gehörte, so war das Gefäß verunreinigt und mußte
zerschlagen werden. Auch noch in mancher Beziehung erinnert das
Maori-Tabu an den Kastengeist der Inder.

8. Dezember. Hier in Wanganui stehen ein paar sonderbare

Kriegerdenkmäler. Das eine ist zum Andenken an die Weißen
errichtet, die bei »der Verteidigung von Ordnung und Gesetz gegen
Barbarei und Fanatismus« gefallen sind. – Fanatismus – das Wort
sollte unverweilt entfernt werden, denn es ist sicher nur aus Irrtum
und Mangel an Ueberlegung auf das Denkmal gesetzt worden;
wenigstens möchte ich das zur Ehre der uns stammverwandten
englischen Nation annehmen. Man verpflanze einmal die Inschrift:
»welche zur Verteidigung von Gesetz und Ordnung gegen den
Fanatismus gefallen sind,« an den Ort, wo Winkelried starb, an die
Thermopylen oder auf das Bunker-Hill-Monument – da wird man
einsehen, was das Wort bedeutet und wie verkehrt es in jenem Fall
angewendet ist. Patriotismus bleibt Patriotismus. Nichts kann ihn
herabwürdigen, mag man ihn auch Fanatismus nennen, so viel man
will. Selbst wenn er vom politischen Standpunkt aus tausendmal im
Irrtum ist, so ändert das nichts an der Sache. Der Patriot ist und
bleibt ehrenhaft, edel und groß, er darf getrost das Haupt erheben!
Mit Recht preist man die tapfern Weißen, die im Maorikrieg gefallen
sind – sie verdienen alles Lob. Aber das Wort ›Fanatismus‹ stellt die
Sache so dar, als hätten sie ihr Blut in keinem würdigen Kampfe
vergossen, in einem Kampfe gegen unedle Feinde, die des Opfers
nicht wert waren. Und doch standen sie wackern Männern
gegenüber, mit denen zu fechten keine Schande war, Männern, die
für ihre Heimstätten und ihr Vaterland als tapfere Feinde stritten und
einen ehrenvollen Tod fanden. Es würde dem Ruhm der braven
Engländer, die unter dem Denkmal liegen, keinen Abbruch tun,
sondern ihn nur erhöhen, wenn die Inschrift besagte, daß sie in
Verteidigung des englischen Gesetzes und ihrer englischen Heimat
gefallen sind, im Kampf mit Gegnern, die aller Achtung wert waren –
mit den für ihr Vaterland sterbenden Maori.
An dem zweiten Denkmal läßt sich nichts verbessern – außer mit
Dynamit. Es ist ein Irrtum durch und durch und ein Beweis von
großer Gedankenlosigkeit. Die Engländer haben es zur Erinnerung an
die Maori aufgestellt, die auf Seiten der Weißen gegen ihre eigenen
Landsleute kämpften. »Dem Andenken der wackeren Männer
gewidmet, die am 14. Mai 1864 gefallen sind etc.,« lautet die
Inschrift. Auf der Rückseite stehen die Namen von ungefähr zwanzig
Maori.
Es ist kein Phantasiegebilde von mir; das Denkmal steht wirklich
da und ich habe es gesehen. Welche Lehre für die kommenden
Geschlechter! Es fordert mit dürren Worten zu Verrat, Untreue und
Verleugnung des Patriotismus auf: »Verlasse deine Fahne,« ruft es,
»erschlage deine Landsleute, verbrenne ihre Häuser, sei eine
Schande für dein Volk – solchen Leuten erweisen wir Ehre!«
 12. Dezember. Nach zehnstündiger Eisenbahnfahrt von Wanganui
aus, erreichten wir Wellington … Eine schöne Stadt, in stolzer Lage;
reger Handel, viel Leben und Bewegung. Ich habe hier drei Tage
teils mit Spazierengehen und angenehmem geselligem Verkehr
zugebracht, teils bin ich in dem herrlichen Garten von Hutt
umhergeschlendert, der eine Strecke weiter am Ufer liegt.
Dergleichen sehen wir gewiß sobald nicht wieder!
Wir packen heute abend unsere Koffer zur Rückreise nach
Australien. Der Aufenthalt in Neuseeland ist zu kurz gewesen, doch
sind wir froh, daß wir es wenigstens flüchtig sehen durften.
Die tapfern Maori haben es den Weißen ziemlich schwer gemacht,
sich im Lande anzusiedeln. Anfangs jedoch empfingen sie die
Engländer freundlich und machten gern Geschäfte mit ihnen.
Besonders kauften sie Flinten, denn sie führten oft zum Zeitvertreib
Krieg unter einander, und die Waffen der Weißen gefielen ihnen weit
besser als ihre eigenen. Ich brauche den Ausdruck ›Zeitvertreib‹ mit
gutem Bedacht; sie kamen wirklich häufig zusammen, ohne daß
irgend ein Streit vorlag und erschlugen einander bloß zum
Vergnügen. Der Verfasser des ›Alten Neuseeland‹ erwähnt einen
Fall, wo das siegreiche Heer nur seinen Vorteil auszunutzen
brauchte, um den Feind zu vernichten und es gleichwohl unterließ.
»Denn,« lautete die naive Erklärung, »täten wir das, so gäbe es
keinen Kampf mehr.« Ein andermal ließ die eine Armee dem Heer,
das ihr feindlich gegenüberstand, sagen, ihr sei das Pulver
ausgegangen und sie müsse das Schießen einstellen, falls sie nicht
neuen Vorrat erhalte. Man schickte ihr, was sie brauchte, und die
Schlacht nahm ihren Fortgang.
Wie gesagt, als die Engländer sich zuerst in Neuseeland
niederlassen wollten, ging alles gut. Die Eingeborenen verkauften
ihnen Strecken Landes, ohne die Bedingungen des Vertrages zu
verstehen, und die Weißen kauften von ihnen und kümmerten sich
nicht darum, daß die Maori nicht wußten, was sie taten. Als dann
letztere allmählich einsahen, daß ihnen unrecht geschah, begannen
die Zwistigkeiten. Kein Maori hätte ein Unrecht, das ihm widerfuhr,
ruhig erduldet oder sich mit Klagen begnügt. Das Volk besaß
dieselbe Ausdauer wie die Tasmanier und daneben noch allerlei
militärische Kenntnisse; es stand gegen seine Bedrücker auf und die
tapfern ›Fanatiker‹ entfachten einen Krieg, dessen schließliche
Entscheidung erst erfolgte, nachdem mehrere Generationen zu
Grabe gegangen waren.
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebooknice.com