ASPECTS OF FORAGING

IN BLACK OYSTERCATCHERS

(AVES: HAEMATOPODIDAE)

by

SARAH GROVES

B. A. Biology, Harvard College, 1973

THESIS SUBMITTED IN PARTIAL FULFILMENT

THE REQUIREMENTS FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

in

THE FACULTY OF GRADUATE STUDIES

(Department of Zoology)

We accept this thesis as conforming

to the required standard

THE UNIVERSITY OF BRITISH COLUMBIA

April, 1982

© Sarah Groves, 1982 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.

Department of 7?:OOL.O Gf)/

The University of British Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5

DE-6 (2/79) i i

ABSTRACT

I studied foraging ecology of black oystercatchers

(Haematopus bachmani) in the rocky intertidal. The aims of this study were: 1) to analyze prey choice and patch choice by adult black oystercatchers and evaluate how well their foraging performance was predicted by foraging theory; 2) to study development of foraging in young oystercatchers; 3) to indirectly examine the relationship between parental foraging performance and fitness by measuring chick growth and chick production.

The following conclusions were reached: 1) Prey selection by oystercatchers was generally as predicted by theory, but birds showed partial preferences for prey. Patch choice followed general theoretical predictions, but profitabilities achieved within particular patch types were highly variable. Reasons for this are discussed. 2) Growth and physical maturation are important components in development of foraging. During the period of this study, chicks were heavily dependent on parental feeding, and the ability of chicks to forage independently developed after chicks left their natal area at about 50 days of age. 3) Chick growth varied between one-chick and two-chick broods, and this may be related to parental foraging performance. However, during this study chick production was constrained by weather and predation, and no relationship between parental foraging performance and fitness could "be def ined. TABLE OF CONTENTS

Abstract i i

List of Tables vi

List of Figures vii

Acknowledgements ix

General Introduction 1

Literature Cited 5

Exploiting a Patchy Environment: a Field Study of

Black Oystercatchers Foraging in the Rocky Intertidal ..6

Introduction 6

Study Area . 8

Study Animal 9

Methods 9

Distribution and Abundance of Intertidal Organisms ..9

Black Oystercatcher Foraging Behaviour 12

Results 16

Prey Distribution 16

Durations of Foraging Behaviours 21

Searching Behaviour 23

Prey Selection 25

Prey Profitability 32

Cumulative Prey Consumption 36

Allocation of Foraging Time between Zones 39

Discussion 43

Prey Choice 43

Patch Choice and Time Allocation between Zones 46 i v

Conclusions 49

Literature Cited 51

Development of Foraging Skills

in Young Black Oystercatchers 55

Introduction 55

Study Areas • 55

Study Animal 56

Methods . . 57

Intertidal Prey Organisms 57

Chick Growth -. 57

Foraging Behaviour 58

Results .' 59

Prey Distribution and Abundance 59

Chick Growth 59

Allocation of Foraging Time between Zones 62

Prey Selection by Chicks and their Parents 63

Searching Behaviour 72

Winter Foraging 74

' ' Prey Stealing ..." 75

Discussion 77

Chick Growth 78

Allocation of Foraging Time between Zones 78

Prey Choice 79

Searching Behaviour and Prey Handling 79

Prey Stealing 81

Important Processes in Development of Foraging Skills

81

Literature Cited • 83 V

Growth, Sibling Rivalry, and Chick Production

in Black Oystercatchers 86

Introduction 86

Study Area 86

Study Animal 87

Methods 88

Black Oystercatcher Territories 88

Chick Growth and Survival 88

Chick Feeding 89

Results ,.89

Territory Fidelity, Clutch Size, and Breeding Success

90

Brood Size and Chick Growth ..93

Weight Differences in Two-chick Broods ...98

Chick Survival 101

Clutch Size and Chick Production 107

Discussion . . 109

Chick Growth and Sibling Rivalry ..109

Chick Survival 111

Chick Production and Parental Investment 112

Literature Cited 114

General Conclusion 116

Literature Cited 120

Appendix A . 121

Literature Cited 122

Appendix B ' 123 vi

LIST OF TABLES

Table I: Correlations between Event Duration and Time Since

Start of a Foraging Bout 23

Table II: Durations of Search in Three Intertidal Zones ...25

Table III: Rate of Effective Search as Calculated by Multi-

species Disc Equation for 5 Prey Types in 3 Zones 30

Table IV: Average Bout Durations and Profitabilities in 3

Intertidal Zones 40

Table V: Allocation of Foraging Time between Zones 63

Table VI: Numbers of Different Prey Taken by Chicks of

Various Ages 72

Table VII: Initiation of Chases in Winter Foraging Flocks .77

Table VIII: Black Oystercatcher Clutch Size 91

Table IX: Chick Production on 11 Black Oystercatcher

Territories, 1975-1978 93

Table X: Resightings of Birds Color-banded as Chicks 107

Table XI: Chick Production from One and Two-egg Clutches ..108 vii

LIST OF FIGURES

Figure 1: Map of the Study Area 4

Figure 2: Structure of Behaviour Data 15

Figure 3: Abundance and Biomass of Invertebrates in

Quadrats 18

Figure 4: Abundance and Biomass of Invertebrates in Mussels

Beds 20

Figure 5: Relative Prey Availability and Prey Choice 28

Figure 6: Average Weight of Prey Taken in Each Zone 32

Figure 7: Profitability of Prey with respect to Prey Weight

.35

Figure 8: Cumulative Prey Intake during Foraging Bouts ....38

Figure 9: Bout Duration in Each Zone versus Profitability .42

Figure 10: Chick Bill Lengths and Weights versus Chick Age 62

Figure 11: Proportion of Chick Feedings from Parents and

Chicks 66

Figure 12: Average Weights of Prey Taken in Each

Zone versus Chick Age 68

Figure 13: Prey Types Selected under 3 Conditons ..71

Figure 14: Inter-peck Intervals for Chicks Foraging in each

Zone .74

Figure 15: Chick Growth 95

Figure 16: Sibling Growth in the Two Broods with Minimum

and Maximum Weight Differences 98

Figure 17: Sibling Chases and Parental Feedings 101

•Figure 18: Chick Survival to Fly and Weight at 20 Days ....103 viii

Figure 19: Chick Survival from Hatching 105 ix

ACKNOWLEDGEMENTS

It is a pleasure to acknowledge the many people who contributed time, ideas, and energy to this project. Jamie Smith supervised the project, and I am grateful for his patience, help in the field, and critical comments from start to finish of this work. Steve Borden was the principal designer of hardware and software for the event recorder, but more importantly I appreciate Steve's philosophic insights into the world. In the field, Mary Taitt and Patrick Michiel gave generously of their time and energy on many occasions during this project. Bristol

Foster of the British Columbia Ecological Reserves Unit provided financial support in 1977, and Wayne Campbell of the British

Columbia Provincial Museum suggested Cleland Island as a study site in the first place.

Members of my research committee - Larry Dill, Lee Gass,

Ray Hilborn, Charley Krebs - and Yoram Yom-Tov provided helpful suggestions at various stages of this research and critically read drafts of this thesis. Don Brandys provided the electronic expertise to debug and maintain the event recorder. Fergus

O'Har.a designed a weather-proof housing for the event recorder and frequently serviced .a chronically ailing outboard motor.

Teresa Tenisci and Bill Webb helped solve several major computing problems. During the last months of this project Maria

Weston acted as my agent at U. B. C, and E. E. Cudby and L. C.

Zerr made time 'available for me to complete • this work.

Many individuals came on field trips, helped with boat X handling, endured life-threatening attacks by gulls, and shared the unique pleasures and pains of life on Cleland Island. I thank my field assistants L. Dick, R. Jaremovic, and L. Paull, and I also appreciate the help at various times by A. J. Baker,

D. Dog, P. Groves, B. Hutchins, S. Krepp, P. Lee, K. Lindsay, S.

McCormack, S. McCoy, .J. Myers, R. Olenick, A. Peacock, C.

Redsell, L. Richards, B. Stiling, C. Whitney..

Finally, my parents Mary Groves and Laurence Groves contributed to this project in many ways. I am grateful for their understanding and continuing interest in black oystercatchers. 1

GENERAL INTRODUCTION

How animals forage in patchy environments is of considerable interest to ecologists. A large and growing body of work in this field (Krebs 1978, Pyke et al. 1977) addresses three major issues in foraging ecology: 1) function - what animals do in their search for food, 2) mechanism - how animals make decisions about foraging and how they locate, handle, and digest prey, and 3) consequences - the effects of foraging performance on an animal's fitness.

Functional description is the goal of most data and theoretical studies of foraging, but such studies are not concerned with mechanisms animals use in foraging (Krebs e_t al.

1981). Mechanisms and processes animals use in foraging have, with few exceptions (e. g. Ollason 1980, Waage 1979), been ignored by ecologists. However, the importance of understanding foraging mechanisms, especially those involved in decision making - learning, memory, and perception - has been stressed by

Orians (1981). Fitness consequences of foraging are important because foraging performance affects how much energy an animal can invest in reproduction (Schoener 1971). However, the use of foraging performance to evaluate fitness remains 'an -elusive goal.

Research I did on foraging ecology of black oystercatchers

(Haematopus bachmani) addresses each of these issues in foraging ecology - function, mechanism, and consequences. .Black oystercatchers on Cleland Island (Figure 1) near Tofino, British 2

Columbia, were the subjects of my research. Oystercatchers are large shorebirds with powerful, laterally compressed bills that enable them to exploit a variety of well-armored marine invertebrates. Foraging proficiency is acquired gradually with physical maturation and experience. Consequently, young chicks are unable to handle prey normally taken by adults and depend on parental feeding until they develop foraging proficiency.

This thesis is divided into three sections concerned with how animals forage in patchy environments. The first section,

"Exploiting a patchy environment: a field study of black oystercatchers foraging in the rocky intertidal", evaluates foraging performance (prey choice and patch choice) of adult black oystercatchers in terms.of foraging theory. Discrepancies between theory and performance of animals in their natural environment are discussed. The second section, "Development of foraging skills in young black oystercatchers"., analyzes the development of foraging behaviour in young oystercatchers and discusses processes that play important roles in behavioural development of young birds. The importance of these processes in adult foraging behaviour is also discussed. The third section,

"Growth, Sibling Rivalry, and Chick Production in Black

Oystercatchers", attempts to relate parental foraging performance to fitness indirectly by measuring chick growth and chick production. 3

Figure 1: Map of Cleland Island, British Columbia, Lat. 49°10'N;

Long. 126°05'W. CLELAND ISLAND

H GRASS

SHRUB

HI BEACH

• BARE ROCK

METERS

50 100 150 200 5

LITERATURE CITED

Krebs, J. R. 1978. Optimal foraging: decision rules for predators. In . J. R. Krebs and N. B. Davies (eds.). Behavioural ecology, an evolutionary approach. Blackwell Scientific Publications. Oxford, pp. 23-63.

Krebs, J. R., Houston, A. I., Charnov, E. L. 1981. Some recent developments in optimal foraging. In . A. C. Kamil and T. D. Sargent (eds.). Foraging behavior: ecological, ethological, and psychological approaches. Garland STPM Press, New York. pp.3-18.

Ollason, J. G. 1980. Learning to forage - optimally? Theor. Pop. Biol. 18:44-56.

Orians, G. H. 198.1. Foraging behavior and the evolution of discriminatory abilities. I_n . A. C. Kamil and T. D. Sargent (eds.). Foraging behavior: ecological, ethological, and psychological approaches. Garland STPM Press, New York. pp. 389-405.

Pyke, G. H., Pulliam, H. R., Charnov, E. L. 1977. Optimal foraging: a selective review of theory and tests. Q. Rev. Biol.52:137-154.

Schoener, T. W. 1971. Theory of feeding strategies. Ann. Rev. Ecol. Syst. 2:369-404.

Waage, J. K. 1979. Foraging for patchily-distributed hosts by the parasitoid, Nemer it i s canescens . J. Anim. Ecol. 48:353-371. 6

EXPLOITING A PATCHY ENVIRONMENT: A FIELD STUDY OF

BLACK OYSTERCATCHERS FORAGING IN THE ROCKY INTERTIDAL

INTRODUCTION

Ecologists view the world as a crazy-quilt of resource patches. Empirical and theoretical ecological investigations of foraging animals have been particularly concerned with how animals exploit prey in patchy environments. Three issues, prey choice, patch choice, and allocation of time between patches, have received considerable attention from theoreticians and empiricists. Pyke e_t a_l. (1977) and Krebs (1978) highlight some of these investigations.

Theoretical considerations (e. g. Schoener 1971; Charnov

1976a) have been useful in crystallizing key issues in foraging ecology and identifying important variables to be measured in the laboratory and field. However, theory is only as strong as the assumptions on which it is based (Ollason 1980), and real animals must be substituted for equations to test theory -and its underlying assumptions. Laboratories and carefully selected field situations are the testing grounds for foraging theory.

Laboratories are distillations of the real world. They provide controlled environments for presenting animals with foraging problems and evaluating their solutions to these problems. For instance, Smith -and Sweatman (1974) used a laboratory study to ' test predictions about how birds should 7

forage in patchy environments. Their results showed that birds learned which patch was most profitable, but they continued to sample other less profitable patches. Smith and Sweatman's study is a good example of the utility of laboratory environments as tools for studying foraging, but lab environments impose severe limitations on the scale, complexity, and, possibly, reality of feasible experiments. For example, in a carefully designed and executed laboratory study of optimal foraging, Cowie (1977) had to substitute increased handling time (prey containers that took birds a long time to open) for increased travel time between patches.

Field studies are the richest source of information about how animals actually forage in patchy environments. For instance, studies of redshank, Tringa totanus (Goss-Custard

1977a), provide a good picture of how these birds change their foraging behaviour and prey selection in response to variations in prey densities and availabilities. Major obstacles in most field studies of foraging are the immensity and complexity of natural environments. In real life, predators such as the great tit (Parus major), the laboratory rat of foraging studies, move through large three-dimensional volumes of space while foraging, distributions of prey and prey patches are often difficult to assess (Myers e_t a_l. 1980), and many predators are difficult to see and follow. Zach and Falls (1976a) solved some of the logistic problems of field studies by conducting foraging experiments with ovenbirds (Seiurus aurocapillus) in an outdoor laboratory, a set of walls enclosing a piece of forest floor where ovenbirds foraged. • Zach-and Falls ran their experiments in 8

the outdoor laboratory, but they used artificial patches of artificial prey because of the difficulty of obtaining and working with real prey.

One way to overcome the difficulties often encountered in

field studies of foraging in patchy environments is to select as

subjects highly visible predators foraging in environments where prey are relatively immobile, prey distributions can be characterized by an observer, and patches can be distinguished

in the field. The study presented here looks at how a conspicuous bird, the black oystercatcher (Haematopus bachmani), exploits multiple types of prey, marine invertebrates, in a

relatively simple patchy environment, the rocky intertidal.

Intertidal zonation presents a situation where patches (zones of

the intertidal) can be readily identified in the field.

Specifically, this study analyzes prey choice and patch choice

and evaluates how well ecological theory predicts foraging

behaviour of black oystercatchers in the rocky intertidal.

STUDY AREA

Cleland Island (Lat. 49°10'N.; Long. 126°05'W.) in

Clayoquot Sound off the west coast of Vancouver Island was the

site of- this study. Cleland is a low-lying, basalt island about

7.7 hectares in area. The tides are mixed semi-diurnal and range

between extremes of 0.1 metre and 3.9 metres. Low tide exposes

an extensive intertidal inhabited by a diverse community of

marine invertebrates (Campbell and Stirling 1968). A more

general survey .o.f the intertidal .zones of the British Columbia

coast is provided by Carefoot (1977). 9

STUDY ANIMAL

Black oystercatchers are large shorebirds (ca. 600 g) that inhabit rocky intertidal areas of the Pacific coast of North

America from the Aleutian Islands to Baja, California.

Oystercatchers are well-adapted for feeding on a variety of

intertidal invertebrates. The birds use their laterally compressed bills (6-7 cm long) to lever some prey items (chitons and limpets) off the substrate and to disable others (mussels, crabs, worms) by stabbing before eating them.

Each year about thirty pairs of black oystercatchers breed on Cleland Island. Birds maintain breeding territories from early spring until after their chicks are flying in late summer.

Territories include nest sites and contiguous intertidal

foraging areas used exclusively by the territory holders. A few birds fly to offshore reefs to forage at low tides, but most birds forage only on their territories during the breeding

season and can be readily observed. Hartwick (1976) also studied

this black oystercatcher population and took advantage of the prolonged period of chick dependence on parental feeding to test

Royama's (1970) profitability hypothesis.

METHODS

Distribution and Abundance of Intertidal Organisms

The distribution and abundance of organisms taken as prey

by oystercatchers were assessed by quadrat samples in intertidal 10

areas on 19 oystercatcher territories. On extreme low tides

between May and late August 1977, quadrats were sampled along

"randomly selected lines running from the high water to low water

line. On each line a 50 x 50 cm quadrat frame was laid down at

1.5 metre intervals and each quadrat was sampled non-

destructively. Percent cover by Fucus spp., coralline and

laminarian algae, surfgrass (Phyllospadix), barnacles

(Balanus spp.), gooseneck barnacles (Pollicipes polymerus),

mussels (Mytilus californianus) was estimated visually for each

quadrat before it was disturbed. Each quadrat was systemically

searched for limpets (Collisella digitalis, C. pelta, Notoacmea

scutum), chitons (Katharina tunicata, Tonicella 1ineata), worms

(Nereis vexillosa), and amphipod crustaceans (Orchestoidea

cali forniana, Liqia pallasi i). All individuals one centimetre or

greater in length were identified and measured to the nearest

millimetre. When crabs (Oedignathus inermi s) were found,

carapace width at the widest point was measured and recorded.

Specimens shorter than 1 cm were counted but not measured. A

total of 176 quadrats was sampled.

Data from intertidal quadrats were used to estimate the

numerical abundance and total biomass of various invertebrate

prey of oystercatchers. Length-wet weight regressions (Appendix

A) for limpets, chitons, worms, crustaceans, and crabs were used

to determine the wet weight biomass of prey items in each

quadrat.

Data on percent cover were used to assign quadrats to four

intertidal zones (patch types) between the high water and low

water lines: 1 •) spray zone (SPZO), 2) fucus zone (FUCU) , 3) 11

mytilus zone (MYTL), 4) laminaria-postelsia zone (LAPO). These zones could be readily identified at a distance by an observer.

The following criteria were used to assign quadrats to the four zones: 1) spray zone (adjacent to high water line) - only small barnacles (Balanus glandula) present; 2) fucus zone - cover greater than 50% Fucus spp.; 3) mytilus zone cover greater than 25% Myt ilus cali fornianus ; 4) laminaria-postelsia zone

(adjacent to low water line) - greater than 10% laminarians or kelps (Postelsia). (This classification scheme is similar to that of Kozloff (1973) and Ricketts and Calvin (1968).)

Intertidal quadrats were a good way of sampling sessile invertebrates and plants in the rocky intertidal, but highly mobile prey such as amphipod crustaceans and worms usually moved quickly from the sampling area when they were disturbed, so the abundance of these prey types was underestimated. This method of sampling was also poorly suited for estimating the abundance of more mobile invertebrates living in spaces between mussels in dense mussel beds. To assess the abundance of invertebrates in mussel beds, all of the mussels and associated invertebrates in eleven 25 x 25 cm quadrats were removed and measured. These data were then used to estimate the numerical abundance and total biomass of invertebrates in mussel.beds. Some invertebrates, especially worms, escaped from quadrat sites as collections were being made, so data from these quadrats represent minimum estimates of numbers and biomass present.

Neither sampling method was suitable for estimating abundance of crabs, Oedignathus inermis. This cryptic species spends low tide periods hiding in the holdfasts of kelp 1 2

(pers. obs.) and, in spite of persistent efforts to find it, was infrequently detected in the quadrats. Observation of prey taken by oystercatchers indicates that this crab is relatively abundant, but sampling techniques failed to detect it and provide a realistic estimate of its abundance. Myers e_t al.

(1980) have warned of the inherent dangers in using results of prey sampling schemes to estimate prey available to a predator.

The pitfalls they identify are, for the most part, avoided in this study because 1) most prey are visible and distributed across a rocky substrate, 2) most prey are sessile during low tide foraging periods, 3) a few prey types that are difficult to sample because they are very cryptic (crabs) or very mobile

(worms and crustaceans) have been explicitly identified and their abundances as determined by sampling recognized as underest imates.

Black Oystercatcher Foraging Behaviour

Individual oystercatchers on seven territories (a subset of the 19 territories used for quadrat sampling) were observed

foraging during low tide periods from 3 hours before until 3 hours after low tide. Continuous records of their foraging behaviour were made and included the following: bird

identification and zone of the intertidal (patch type); duration of prey-handling events, prey type, and prey size with respect

to the bird's bill length; durations of successful and unsuccessful searches (periods of continuous search ending with prey captures and without prey captures, respectively); durations of non-foraging behaviours such as sitting, standing, 13

and flying that occur during periods of foraging.

Prey sizes were estimated as fractions of a bird's bill length (7 size classes), and prey biomass was estimated using length-weight regressions. Measurements of the remains of prey eaten by oystercatchers were very similar to those recorded during foraging observations, and gave confidence in the size est imates.

Data were encoded with a digital event recorder and stored on magnetic tape (on a Uher tape recorder) for subsequent decoding by a PDP-11 computer. The event recorder was one I built with minor modifications to the event recorder described by Gass (1977). Data were recorded as a sequence of point events

(e. g. peck, pace, attack prey, swallow, sit, fly, etc.) and their.times of occurrence. Logical rules that defined events of biological interest with beginnings, endings, and durations were applied to the sequence of point events. The result was a

sequence of behavioural events (e. g. successful search, unsuccessful search, prey handling, etc.), their beginning

times, and durations. Figure 2 summarizes the relations between point events and behavioural events.

Behavioural events and their durations were analyzed

statistically with the MIDAS (Michigan Interactive Data Analysis

System) statistical package supported by the Computing Centre at

the University of British Columbia and by Fortran programs that

I wrote. The multi-species disc equation (Murdoch and Oaten

1975) was used to analyze prey choice.

The foraging data presented here are for adult . birds on

.breeding territories. Hartwick ( 197:6) .showed that compos it ion of 1 4

Figure 2: An example of a behavioural data record - sequence of point events encoded by event recorder and sequence of behavioural events defined by logical rules (see text for details). STRUCTURE of BEHAVIOUR DATA

SEQUENCE OF SEQUENCE OF TIME OF BEHAVIOURAL STARTING POINT EVENTS OCCURRENCE EVENTS TIME DURATION PREY TYPE

STAND V STAND PACE

PACE SUCCESSFUL PECK I "SEARCH PACE ATTACK PREY V < PECK LIMPET -PREY-HANDLING T6 T11-T6 LIMPET A1/3 T9 PECK SWALLOW < PACE Tl2 PECK T 13 UNSUCCESSFUL PACE T17- Tn Tl4 "SEARCH T11 PACE T15 PECK T16 SIT T17 y—SIT Tl7 Tift " T17 PACE %

LOGICAL \ ! RAW DATA RULES DATA FILE OF ENCODED BY BEHAVIOURAL EVENTS, EVENT RECORDER STARTING TIMES, AND DURATIONS 16

adult oystercatcher diets depended on whether or not birds had chicks and that the presence of chicks affected the frequency and length of foraging trips made by parent oystercatchers. The presence of chicks confounds the analysis of adult foraging, so only adults holding territories but without chicks were included in this analysis. These birds were completing clutches of eggs, laying replacement clutches, or incubating eggs.

RESULTS

Prey Distribution

The four zones (patch types) of the intertidal (spray zone, fucus zone, mytilus zone, laminaria-postelsla zone) differed markedly in numbers and total biomass (wet weight) of invertebrate prey present. Figure 3 summarizes the quadrat sample data according to zone of the intertidal and prey type

(except mussels). Numbers and biomass of prey increased dramatically between the high water (spray zone) and low water

(laminaria-postelsia zone) lines. In addition, certain types of prey such as chitons and crabs were found only in the lower zones of the intertidal.

The abundance and total biomass of prey (except mussels) sampled in 11 mussel bed quadrats are summarized in Figure 4.

The number of prey organisms living in mussel beds was high and quite variable. Most of these invertebrates were small (1 cm or smaller), and .some of them were probably not available as prey to oystercatchers as mussel beds may be 10-15 cm in depth, 1 7

Figure 3: Average numerical abundance and biomass of

invertebrates (except mussels) in 176 50 x 50 . cm intertidal quadrats on 19 oystercatcher territories. (Vertical lines are upper 95% confidence limits). NUMBER o o o o -+- LIMPETS II CP CRUSTAC. m> ssvwoia

Z oc II LIMPETS tn zp ssvwoia \

LIMPETS

II oz! CHJTO NS cn Zr me WORMS oo \ -Or- LIMPETS CRUSTAC. n O^- tn CHITONS mr:> CRABS >> SSVW0I8

+ ± CO 2 (swvao) o o SSVNOIB 19

Figure 4: Average numerical abundance and biomass of invertebrates (except mussels) in 11 25 x 25 cm mussel bed quadrats. (Vertical lines are upper 95% confidence limits). NUMBER ho tn o tn O o o o O

SEA CUCUMBERS

CRUSTAC.

CHITONS

LIMPETS

WORMS

SSVWOIB

o CP "I * o o o o O (swvy9) SSVH0I8 21

deeper than an oystercatcher's bill. The number of mussels (not included in Figure 3) in each quadrat averaged 355.8

(s.d. = 163.7) with an average length of 3.8 cm (s.d. = 0.75).

Because of the layering of mussels in mussel beds, only about one-third to one-half of the mussels were actually accessible to the birds, but no quantitative estimate of this was obtained.

Correlations between particular types of cover and the presence of prey may provide birds with information about the average quality of patches of the intertidal. Biomass of prey in quadrats was correlated with the amount of some types of cover.

Amounts of coralline algae and laminarian algae were positively correlated with the biomass of prey in quadrats (coralline algae, r = 0.55; laminarian algae, r = 0.61; df = 175, p<0.0l for both types of algae). Amounts of cover of Fucus spp., gooseneck barnacles, and surf grass, were not correlated with the biomass of prey in quadrats.

Durations of Foraging Behaviours

An animal's foraging behaviour may change after it has foraged for a period of time. For instance, satiation may effect foraging behaviour by causing an animal to be more selective in its choice of prey (Charnov 1976b) or less responsive to food

(Zach and Falls 1978). .If foraging behaviour changes with the amount of time an animal has been foraging, then the time of occurrence of various foraging behaviours with respect to the start of a foraging bout will be an important variable in any analysis of foraging behaviour. In black oystercatchers I examined the possibility that durations of foraging behaviours 22

vary systemically with the amount of time a bird has been foraging by testing for correlations between durations of different foraging behaviours and the amount of time" a bird had been foraging in an uninterrupted bout within one zone of the intertidal. Correlation coefficients for behaviour duration versus time of occurrence for successful search and unsuccessful search are shown for individual birds in Table I. No relations between amount of time a bird had been foraging and durations of foraging behaviours were detected statistically. In all subsequent analyses durations of events are treated independently of time of occurrence.

Searching Behaviour

Zonal differences in prey presence, abundance, and substrate might affect the way birds search for prey. Two components of searching behaviour were measured in this study: intervals between pecks and intervals between paces. Inter-peck intervals provide a measure of the frequency with which birds investigate potential prey items by touching their bills to the substrate and prey items. Average inter-peck intervals varied significantly between zones (fucus, 3.60 sees, s. d. = 3.21, N =

74; mytilus, 4.72 sees, s. d. = 4.67, N = 779; laminaria- postelsia, 3.72 .sees, s. d. = 3.86, N = 373 ; F = 7.67, df =

2,1225, p < 0.001). Inter-peck intervals were shortest in the fucus and laminarian-postelsia zones where birds often searched for prey tactilely by probing through layers of algae. Inter- pace intervals provide a measure of how fast birds moved while foraging. Inter-pace intervals were similar in all zones (fucus, TABLE I

Correlations between durations of successful

and unsuccessful search and amounts of time

individual birds had been foraging*

Succ. search Unsucc. search

BIRD ZONE £ df r df 1 1 0 mytl -0.02 18 0.47 10

1 1 1 mytl -0.34 1 7 -0.30 3

1 1-1 lapo -0.12 54 -0.18 28

202 mytl -0.02 46 -0.12 48

202 lapo 0.07 28 -0.17 1 5

203 mytl 0.02 1 28 0.08 49

218 lapo 0.03 1 4 -0.45 5

219 mytl 0.58 1 0.91 1

220 - lapo 0.35 1 7 , 0.30 9

222 lapo 0.07 27 -0.26 18

223 lapo 0.09 2 too few cases

224 f ucu 0.05 1 6 -0.69 1

224 mytl 0.09 1 6 0.24 17

225 fucu too few cases too few • cases

2.25 mytl 0.32 26 0.06 9

239 fucu 0.45 •3 too few cases

239 mytl 0.01 21 0.35 8

*None of these r-values are significant at the 0.05 level. 24

1.05 sees, s. d. = 0.65, N = 74; mytilus, 1.08 sees, s. d.

0.62, N = 779; laminaria-postelsia, 1.10 sees, s. d. = 0.83, N =

373; F = 0.239, df = 2,1225, NS). Foraging oystercatchers adjusted their searching behaviour to the zone in which they were foraging by changing their inter-peck intervals but not inter-pace intervals which were relatively constant between zones.

Two measures of duration of search were recorded: duration of successful search and duration of unsuccessful search

(uninterrupted periods of active search ending with a prey capture and without a prey capture, respectively). In all zones of the intertidal, the duration of unsuccessful search exceeded the duration of successful search -(Table II, F = 19.90, df = 728, p < 0.001). However, neither the duration of successful search (F = 0.30:, df ='466, P = 0.74) nor unsuccessful search (F = 2.11, df •= 2,260, P = 0.12) varied significantly between zones. This indicates that birds respond to conditions of prey abundance and substrate character in particular zones by varying the frequency of specific search behaviours (inter-peck intervals) rather than the duration of uninterrupted search.

Prey Selection

Predators should become more selective in their choice of prey as prey abundance increases (Schoener 1971). Relative abundances of six different prey types and frequencies with which each type was taken from each zone by foraging oystercatchers are shown in Figure 5. (These data are corrected TABLE II

Duration (seconds) of successful and unsuccessful

search in three intertidal zones*

Duration of Duration of

Zone Successful Search Unsuccessful Search

FUCU 15.9 (N=24, s.d.=12.1) 19.4 (N=6, s.d.=21.9)

MYTL 13.6 (N=289, s.d.=14.6) 18.1 (N=166, s.d.=19.6)

LAPO 13.4 (N=154, s.d.=14.6) 24.4 (N=89, s.d.=29.8)

*See text for details. 26

for different proportions of foraging time spent in each zone).

Prey selection reflects prey availability in each zone and may also reflect birds' prey preferences and prey detection abilities in each zone. For instance, the abundance of crabs and chitons increased from high to low water line as did the frequency with which these prey types were taken by oystercatchers. By contrast, the abundance of limpets was similar in the mytilus and laminaria-postelsia zones, but 37% of limpets eaten were taken from the mytilus zone while only 14% were taken from the laminaria-postelsia zone. The highest percentage of limpets eaten, 45%, was taken from the fucus zone where limpets had a very low absolute abundance.

Analysis of prey selection by black oystercatchers is complicated by changes in relative densities of different prey types between zones of the intertidal. To overcome this problem the multi-species disc equation (Murdoch and Oaten 1975) was used to analyze prey selection in more detail. The equation is

Nj = a; TP; 1 + £ahD

Where

Ni = number of ith prey eaten

a; = rate of effective search for ith prey

T = total foraging time

D; = density of ith prey

h = average handling time 27

Figure 5: Relative prey availability and prey selection by foraging oystercatchers in 4 zones of the intertidal. Each row of the graph represents one prey type. For each prey type

(rows), percent available and percent taken each sum to 100%.

Data are corrected for different proportions of foraging time spent in each zone. (Relative mussel availability was estimated from percent cover data. Intertidal quadrat data were used to determine relative availability of other prey types). SPRAY FUCUS MYTILUS LAM.-POST. ZONE ZONE ZONE ZONE 29

This equation can be solved for the rate of effective search, a measure of the probability that a prey type will be taken. Rate of effective search, a; , is a product of area searched, probability of detecting a prey, and probabi1ity• of attacking a prey.

The multi-species disc equation was solved for a; for each of five prey types in three intertidal zones (see Appendix B).

Values of rate of effective search are shown in Table III. These results show that rate of effective search for each prey type and, consequently, .the probability that a prey type would be taken varied between zones. For instance, across three intertidal zones, limpets were more likely to be taken by foraging oystercatchers in the fucus zone than in the mytilus or laminaria-postelsia zone. Within the laminaria-postelsia zone, where prey abundances and diversities were greatest, rate of effective search was highest for crabs. This suggests that oystercatchers were selecting crabs over other prey types.

Prey selection is affected by both prey distribution and prey preferences of oystercatchers. As a consequence of prey distribution, prey abundance, and selectivity by oystercatchers, the largest prey items were taken in the lowest zones of the

intertidal. Figure 6 shows average weights of prey taken in each zone of the intertidal. Average prey weights varied

significantly between zones (F = 5.18, df = 3,395, p < 0.01).

Prey Profitability

Prey profitability (grams of wet weight ingested/second of

handling time) varied with prey type and prey size. Even though TABLE III

Rate of effective search, a; , as calculated by multi-species disc equation for 5 prey types in 3 zones of the intertidal*

Zone

Laminar ia-

Prey type Fucus Mytilus postelsia limpet 0.15 0.02 0.006 mussel 0.00025 0.000014 0.000031 chiton 0.0043 0.00013 crab 0.108 worm 0.032

*See text for details. 31

Figure 6: Average weight of prey taken by foraging

oystercatchers in each zone of the intertidal. (Vertical lines

are upper 95% confidence limits). 32

10.0T

T

< g 1.0 I o Hi

>. 0.1 UJ cr F= 5.1ft df= 3,395 p<0.01

SPRAY FUCUS MYTILUS LAM.- POST. ZONE ZONE ZONE ZONE N = 1 N=21 N= 260 N = 117 33

most prey taken by birds fell into a very limited range of prey

sizes as indicated by sample sizes presented in Figure 7, there

are enough data to examine the relation between profitability

and prey size. Prey taken by black oystercatchers fall into two

groups with respect to profitabilities. The profitabilities of

some prey (limpets, mussels, crabs) increase with prey size over

the range of prey sizes taken by oystercatchers in the study

area (Figure 7a, 8b, 8c), and the profitabilities of other prey

(chitons, and perhaps also worms) remain relatively constant

over a range of prey sizes (Figure 7c, 8d). Differences in the

relationships between prey profitabilities and prey size reflect

difficulties birds have in handling some prey and the effect

these difficulties have on handling time.

Limpets, mussels, and crabs are relatively stationary prey,

and none of them can flee rapidly from a predator. When an

oystercatcher attacks, a limpet pulls its valve down flush with

the surface and clings tightly to the substrate, a mussel pulls

its valves firmly together by muscle contraction, and a crab

remains hidden in a cul-de-sac at the base of a kelp holdfast

and blocks the entrance of its hiding place with its well-

developed pincers (pers. obs.). All of these defenses are

passive, and once an oystercatcher has broken through the

defense, the prey is helpless and quickly consumed by an

oystercatcher.

Chitons and worms have more active defenses than other prey

types and are considerably more challenging for birds to handle.

Chitons cling tightly to the substrate and are often firmly

lodged in crevices where it is difficult for birds to gain 34

Figure 7: Profitability (grams/second of handling time) of

a) crabs, b) limpets, c) mussels, d) chitons, e) worms with

respect to prey weight (grams), (r-values were calculated using

a linear model. Quoted probability levels are only approximate

because of non-linearities in the data. Vertical bars are upper

95% confidence limits; numbers by each data point are sample

sizes) . 35

A-C RABS B- LIMPETS

I- 1.00' v'9 .''54 I--- 4V 0.10 38 97

0.01" r = 0.55 r=0.85 O 547 df=6A8 z df=102 o p< 0.01 p< 0.001 o + LU 0.0 36.0 72.0 0.0 ,3.0 6.0 00 \ 00 C-MUSSE L S D-CHI TONS E-W0RMS < cr o

1.0 0" l- 1 ,''3 1- 7--I I 2 4- 0.10 V-4

i / 0.01 r = 0.A1 r = -0.11 r=0.03 I t df = 44 df=6 df = 22 p

PREY WEIGHT (GRAMS) 36 sufficient purchase to lever them, off the substrate. Once chitons are captured, they contract on their ventral side and form a tight, rigid curl. Birds often struggle at length to flatten a captured chiton enough to be able to remove the flesh.

Worms are removed from the substrate, usually a mussel bed, by a tug-of-war between a worm and a bird. Often a worm pulls free of a bird several times before a bird successfully pulls it free of the substrate. After capture, worms usually thrash and writhe vigorously, so birds must subdue worms before they can eat them.

Prey profitabilities are affected by prey responses to attacks by oystercatchers. Profitabilities of relatively passive prey increased with prey size. Behaviour of prey such as chitons and worms with vigorous responses to attack may result in long handling times and relatively constant profitabilities over a range of prey sizes. However, more data are needed to confirm this, especially for chitons.

Cumulative Prey Consumption

The cumulative amount of prey obtained by a bird during an uninterrupted foraging bout varies between zones of the

intertidal. Figure 8 shows the relation between cumulative prey

intake (measured as grams wet weight) and time during a foraging bout within a zone for five individual birds. (Data are shown

for five long foraging bouts). Birds foraging for long periods of time in one zone generally obtained greater cumulative amounts of prey in the laminaria-postelsia zone than the mytilus zone. Different cumulative amounts of prey consumed in these two zones reflect zonal differences in prey abundance and size. 37

Figure 8: Cumulative prey intake during uninterrupted foraging bouts for individual birds during two bouts in mytilus and three bouts in laminaria-postelsia zones. Grams of prey (wet weight) are plotted against foraging bout duration (minutes). MYTILUS ZONE MYTILUS ZONE HO

CO

< 100 cr o x 60 o LU 20 >- 0 t LU 30 60 90 30 9'0 cr LAM.-P0ST. ZONE Q_ LAM.-POST. ZONE LAM.-POST. ZONE LU UO >

100

ZD O 60

20 to + co 30 60 90 0 30 60 90 MINUTES 39

Allocation of Foraging Time between Zones

Zonal variation in prey abundance and the availability of preferred prey types could influence the way birds allocate their foraging time between zones of the intertidal. During this study, oystercatchers spent 0.4% of their foraging time in the spray zone, 4.0% in the fucus zone, 54.0% in the mytilus zone, and 41.0% in the laminaria-postelsia zone. Prey abundances and availability of large and preferred prey were highest in the laminaria-postelsia zone, but foraging bouts in this zone were rarely longer than foraging bouts in other zones, and birds did not always achieve high profitabilities while foraging in the laminaria-postelsia zone. Data on foraging bout durations and associated average profitabilties in three intertidal zones are summarized in Table IV. Average profitabilities in mytilus and laminaria-postelsia zones were similar. Figure 9 presents a more detailed look at data in Table IV. Bout duration in a zone is plotted against average profitability achieved during the foraging bout. Low or high profitabilities were not consistently associated with a particular zone, and bout duration was not correlated with the profitability a bird achieved during a bout

(r = -0.11, df = 41, NS, untransf ormed data). Variability ,of profitabilities achieved in fucus, mytilus and laminaria- postelsia zones suggests within zone patchiness of prey. The step-like shapes of cumulative prey intake curves in Figure 8 also suggest this. TABLE IV

Average' bout duration (seconds) and profitability (grams/second) during foraging bouts in 3 intertidal zones

Average Average bout duration profitability Zone N (seconds) (grams/second) fucus 221.9 0.009 (s. d. = 270.9) (s. d. = 0.013) myt ilus 14 734.7 0.098 (s. d. = 1223.5) (s. d. = 0.21 ) laminaria- 26 632.8 0.022 postelsia (s. d. = 1058.3) (s. d. = 0.03) 41

Figure 9: Duration of foraging bouts in 3 intertidal zones plotted against profitability achieved during bouts. (Data were transformed by log-transformation for purposes of graphical presentat ion). 42

ZONE X=FUCUS • =MYTI LU S + = LA M I NARIA- POS TELSIA 4.0 I + +

3.OT . ; ; .••. * x

e

X .0T + 2 • +

X 1.0

° - ?4.0 TiS ^2D MX) ' OX)

GRAMS/SECOND (LOG1Q) 43

DISCUSSION

Animals foraging in patchy environments must solve three problems as they forage: what prey to take (prey choice), where to forage (patch choice), and how to allocate foraging time between patches. Ecological theory has addressed all three of these foraging problems, and insights into strengths and weaknessess of this theory may be gained by asking how well foraging behaviour of black oystercatchers in this study and of other animals reported in the literature is predicted by theory.

Prey Choice

Predators should become more selective in their choice of prey as prey abundances increase (Schoener 1971). In general, prey choice by black oystercatchers followed predicted trends.

In zones of the intertidal where prey biomass and abundance are relatively high oystercatchers took bigger, more profitable prey items than they did in zones where abundances and biomass were relatively low (Figure 6). This finding is fairly general, and similar results have been reported from field studies of redshank (Goss-Custard 1977a) and gray squirrels Sciurus carolinensis (Lewis 1980), and laboratory studies of bluegill sunfish Lepomis macrochirus (Werner and Hall 1974), and great tits (Krebs e_t al. 1977). Cases where predators do not select prey according to their relative profitabilities can probably be explained in terms of currencies besides biomass and kilocalories (e. g. Goss-Custard 1977b, Davies 1977) or unpalatable properties of -some prey types (e. g. Glander 1-981).. 44

A further prediction concerning prey choice and diet breadth is that predators should not show partial preferences; if a particular prey type is encountered it should always or never be taken, the "always or never" rule (MacArthur 1972;

Schoener 1971). In this study, the multi-species disc equation

(Murdoch and Oaten 1975) was used to calculate rates of effective search for five prey types in three zones of the intertidal. These calculations (Table III) indicate that probabilities of capture for different prey types may vary considerably between zones, but the results do not support the

"always or never" rule.

Results from Werner and Hall's (1974) study of diet breadth in bluegill sunfish suggest that relatively unprofitable prey are gradually dropped from the diet as abundance of big prey increases. In their experiments, different prey types were represented by different size classes of daphnia (Daphnia magna). As prey density was increased experimentally, size classes of daphnia were dropped sequentially from the bluegill diet. In another test of the "always or never" rule Krebs et al.

(1977) presented great tits with two different sizes of meal worms at different encounter frequencies. At high encounter frequencies birds showed strong selection for big prey, but, contrary to prediction, small unprofitable prey were only gradually eliminated from the diet.

In these laboratory tests of the "always or never" rule different size classes of a single organism were used to represent different prey types. Data from field studies where different prey types actually are different organisms.do not 45

support the "always or never" rule either. In this study of black oystercatchers limpets were not a preferred prey type, but they were never completely dropped from the diet. Similarly in redshank, a small amphipod crustacean (Corophium volutator) was a preferred prey, but less preferred nereid worms were never eliminated from the diet even when Corophium densities were high

(Goss-Custard 1977b).

Exceptions to the "always or never" rule seem to be general, and suggest that the rule rarely or never applies in laboratory or field situations. Schluter (1981) recently reviewed diet breadth of 44 species reported in the literature and concluded that diet theory has generally failed to predict prey choices of predators foraging for multiple prey. A number of alternatives may explain violations of the "always or never" rule and shortcomings of diet theory: 1) predators may forage simultaneously for several different currencies (e. g. kilocalories and trace nutrients; for examples see Davies

(1977), Goss-Custard (1977b)); 2) during a foraging bout predators may switch from foraging as specialists to foraging as generalists (Heller 1980); 3) . predators foraging in variable environments must sample to evaluate and monitor prey abundances and profitabilities (e. g. Krebs et al. 1977); 4) predators in patchy environments may .minimize the risk of going hungry by taking less profitable or less preferred prey as they are encountered because considerable time may elapse before the next preferred prey is encountered (e. g. Caraco 1980). 46

Patch Choice and Time Allocation between Zones

Predators' patch choices should be governed by the same criteria that control prey choice (Pyke et a_l. 1977). Predators should concentrate their foraging efforts in the best patches.

Field and laboratory studies (Lewis 1980; O'Connor and Brown

1977) indicate that predators generally forage according to this prediction, but they always spend some time foraging in less profitable patches.

In this study of black oystercatchers, intertidal zonation was used to define patches. Birds spent over 95% of their foraging time in the mytilus and laminaria-postelsia zones where prey abundance and biomass were greatest. However, birds also spent time foraging in the fucus zone and sometimes achieved profitabilities similar to those attained in the mytilus and laminaria-postelsia zones (Figure 9). Results from other studies

(Smith and Sweatman 1974; Zach and Falls 1976a,1976b) indicate that although predators quickly identify the most profitable patches and concentrate their foraging activities there, they continue to visit less profitable patches.

Predators use information besides average prey profitability to select patches where they will forage.

Recently, it has been shown both theoretically (Caraco 1980) and experimentally (Caraco ejt al. 1980, Real 1981) that resource variability is an important determinant of patch preferences. In variable environments foraging behaviour may reflect the variability in prey profitabilities in addition to average prey profitabilities, so mean prey profitabilities may not be sufficient to characterize predators' patch preferences. 47

A predator should forage in a patch as long as its rate of food intake remains higher than the average rate for the environment (Charnov 1976a). This rule for patch switching, the marginal value theorem, is deterministic and assumes: 1) random movement between patches by a predator, 2) prey depletion by a predator while foraging in a patch, and 3) the ability of a predator to monitor its marginal capture rate as it forages. The marginal value theorem has been tested under controlled laboratory conditions (Cowie 1977), but a rigorous field test is lacking.

Field tests of the marginal value theorem present some serious difficulties. For a number of reasons black oystercatchers in this study would not be suitable subjects for a test of the marginal value theorem: 1) black oystercatchers are territorial and their familiarity with territory topography and resource distribution affect where they search for prey; 2) no good data exists to address the issue of prey depletion in patches, but lack of correlation between duration of successful searches and amounts of time birds foraged in uninterrupted bouts (Table I) suggests birds do not significantly deplete prey during foraging bouts; 3) the lack of correlation between foraging bout duration and achieved.profitability (Figure 9) suggests that black oystercatchers either cannot or do not calculate marginal capture rates while they forage. Other vertebrate predators, bluegill sunfish (Werner et §_1. 1981), ovenbirds (Zach and Falls 1976b, 1976c), great tits (Smith and

Sweatman 1974), and squirrels (Lewis 1980), also forage in ways that violate assumptions of the marginal value theorem. The 48

foraging behaviour of all of these predators and their patterns of search were influenced by previous experience in an area.

Increasingly detailed studies of how vertebrate predators forage suggest that the assumptions of the marginal value theorem are so restrictive that a rigorous field test with a vertebrate is unlikely.

Several patch switching models exist as alternatives to

Charnov's model. Cowie and Krebs (1979) suggest that predators simply use "rules-of-thumb" to decide when to switch patches.

The performance of an animal foraging according to "rules-of- thumb" may closely approximate the solution predicted by an optimal foraging model such as the marginal value theorem.

Ollason's (1980) learning model for foraging in a patchy environment is based on such a "rule-of-thumb". In Ollason's model a predator simply continues foraging in a particular patch as long as it is feeding faster than it remembers doing. The length of memory is a variable in Ollason's model, and for long values of memory the results from the model are close approximations of predicted "optimal" solutions.

The problem of when to switch patches is closely related to the problem of when and how much to sample among patches. In fact the behaviours of switching and sampling are so similar that there is no a priori way to distinguish between them. (An a posteriori way to distinguish between sampling and switching would be to define a brief visit to a new patch as sampling and a long visit as switching, but this approach is not predictive and of no theoretical interest). Perhaps this ambiguity explains why the question of how much to sample has received so little 49

attention.

Krebs et al. (1978) have dealt theoretically and experimentally with the question of how much to sample. They used a mathematical model to predict the optimal amount of sampling a bird should do during a specified number of trials.

The results of their meticuluously designed two-patch laboratory experiment approximated the predicted sampling frequency, but the birds (great tits) sampled more often than expected and there was considerable variation in performance among individual birds. A problem arises in interpreting results that approximated a predicted optimal solution. Any deviation from a prediction may be attributed either to experimental error or to failure of a particular model to explain a behaviour of interest

(Ollason 1980). So, for philosophical reasons it is difficult to accept or reject the hypothesis of optimal sampling. Problems encountered testing optimal sampling in the laboratory are compounded in the field where it is not possible to distinguish functionally between switching and sampling.

Conclusions

Theoreticians find it difficult to ' predict optimal solutions for animals foraging in complex natural environments

(Zach and Smith 1981), and it is unlikely that animals can do this either. Most natural environments are relatively complex and variable, and, as a result, the adaptive solutions predators employ for foraging in variable environments are likely to be stochastic in nature. It follows that memory and learning,

stochastic processes (Bateson 1979), are of critical importance 50

in understanding how animals forage in patchy environments.

Recently, a number of workers have developed models of foraging in variable environments (e. g. Caraco 1980, Green 1980, Oaten

1977), and some specific solutions animals might employ while foraging in variable environments have also been proposed. Krebs and Cowie (1979) have suggested that "rules-of-thumb" form the bases of animals' foraging decisions, and Ollason (1980) has proposed a learning model that explains patch choice.

In the future new insights into how predators exploit prey in patchy environments will depend on: 1) further development and refinement of stochastic foraging models; 2) better understanding of the roles of memory and learning in predator behaviour; 3) more laboratory and, especially, more field investigations of foraging models and their underlying assumptions. 51

LITERATURE CITED

Bateson, G. 1979. Mind and nature. E. P. Dutton. New York.

Campbell, R. W. and Stirling, D. 1968. Notes on the natural history of Cleland Island, British Columbia, with emphasis on the breeding bird fauna. B. C. Prov. Mus. Report for 1967. Victoria, B. C. pp.25-43.

Caraco, T. 1980. On foraging time allocation in a stochastic environment. Ecology 6:119-128.

Caraco, T., Martindale, S., and Whittam, T. S. 1980. An empirical demonstration of risk-sensitive foraging, preferences. Anim. Behav. 28:820-830.

Carefoot, T. 1977. Pacific Seashores. J. J. Douglas Ltd., Vancouver. 208 p.

Charnov, E. L. 1976a. Optimal foraging, the marginal value theorem. Theor. Pop. Biol. 9:129-136.

Charnov, E. L. 1976b. Optimal foraging: attack strategy of a mantid. Amer. Nat. 110: 141-151.

Cowie, R. J., 1977. Optimal foraging in Great Tits (Parus major). Nature 268:137-139.

Cowie, R .J. and Krebs, J. R. 1979. Optimal foraging in patchy environments. I_n . R. M. Anderson, B. D. Turner, L. R. Taylor (eds.). Population dynamics: the 20th Symposium of the British Ecological Society, London, 1978. Blackwell Scient. Publ. Oxford, pp. 183-205.

Davies, N. B. 1977. Prey selection and the search strategy of the spotted flycatcher (Muse icapa striata) , a field study on optimal foraging. Anim. Behav. 25:1016-1033.

Gass, C. L. 1977. A digital encoder for field recording of behavioral, temporal, and spatial information in directly computer-accessible form. Behav. Res. Methods Instrum. 9:5- 11. 52

Glander, K. E. 1981. Feeding patterns in mantled howling monkeys. I_n . A. C. Kamil and T. D. Sargent (eds.). Foraging behavior: ecological, ethological, and psychological approaches. Garland STPM Press. New York. pp. 231-257.

Goss-Custard, J. D. 1977a. Optimal foraging and the size selection of worms by redshank Tringa totanus. Anim. Behav. 25:10-29.

Goss-Custard, J. D. 1977b. The energetics of prey selection by redshank, Tr inga totanus ' (L.), in relation to prey density. J. Anim. Ecol. 46: 1-19.

Green, R. F. 1980. Bayesian birds: a simple example of Oaten's stochastic model of optimal foraging. Theor. Pop. Biol. 18:244-256.

Hartwick, E. B. 1976. Foraging strategy of the black oystercatcher(Haematopus bachmani Audubon). Can. J. Zool. 54:142-155.

Heller, R. 1980. On optimal diet in a patchy environment. Theor. Pop. Biol. 17:201-214.

Kozloff, E. N. 1973. Seashore life of Puget Sound, the Strait of Georgia and the San Juan Archipelago. Univ. of Wash. Press. Seattle. 226 p.

Krebs, J. R. 1978. Optimal foraging: decision rules for predators. I_n . J. R. Krebs and N. B. Davies (eds.). Behavioural ecology, an evolutionary approach. Blackwell Scientific Publications. Oxford, pp. 23-63.

Krebs, J. R., Erichsen, J. T., Webber, M. I., and Charnov, E. L. 1977. Optimal prey selection in the Great Tit (Parus major). Anim. Behav. 25:30-38.

Krebs, J. R., Kacelnik, A. and Taylor, P. J. 1978. Test of optimal sampling by foraging great tits. Nature 275:27-31.

Lewis, A. R. 1980. Patch used by gray squirrels and optimal foraging. Ecology 61:1371-1379. 53

MacArthur, R. H. 1972. Geographical ecology. Harper and Row, New York. 269 p.

Murdoch, W. W. and Oaten, A. 1975. Predation and population stability. Adv. Ecol. Res. 9: 2-131.

Myers, J. P., Williams, S. L., and Pitelka, F. A. 1980. An experimental analysis of prey availability for sanderlings (Aves: Scolopacidae) feeding on sandy beach crustaceans. Can. J. Zool. 58:1564-1574.

Oaten, A. 1977. Optimal foraging in patches: a case for stochasticity. Theor. Pop. Biol. 12:263-285.

O'Connor, R. J. And Brown, R. A. 1977. Prey depletion and foraging strategy in the oystercatcher Haematopus ostralegus . Oecologia 27:75-92.

Ollason, J. G. 1980. Learning to forage - optimally? Theor. Pop. Biol. 18:44-56.

Pyke, G. H., Pulliam, H. R.,. Charnov, E. L. 1977. Optimal foraging: a selective review of theory and tests. Q. Rev. Biol. 52:137-154.

Real, L. A. 1981. Uncertainty and pollinator-plant interactions: the foraging behavior of bees and wasps on artificial flowers. Ecology 62:20-26.

Ricketts, E. W. and Calvin, J. 1968. Between Pacific Tides, 4th ed., revised by J.W. Hedgepeth. Stanford Univ. Press, Stanford, Calif. 614 p.

Royama, T. 1970. Factors governing the hunting behaviour and selection of food by the Great Tit (Parus major L.). J. Anim. Ecol. 39:619-668.

Schluter, D. 1981. Does the theory of optimal diets apply in complex environments? Amer. Nat. 118:139-147.

Schoener, T. W. 1971. Theory of feeding strategies. Ann. Rev. Ecol. Syst. 2:369-404. 54

Smith, J. N. M. and Sweatman, H. P. 1974. Food searching behaviour of titmice in patchy environments. Ecology 55:1216-1232.

Werner, E. E. and Hall, D. J. 1974. Optimal foraging and the size selection of prey by the bluegill sunfish (Lepomis macrochirus). Ecology 55:1042-1052.

Werner, E. E., Mittelbach, G. G., and Hall, D. J. 1981. The,role of foraging profitability and experience in habitat use by the bluegill sunfish. Ecology 62:116-125.

Zach, R. and Falls, J. B. 1976a. Ovenbird (Aves: Parulidae) hunting behavior in a patchy environment: an experimental study. Can. J. Zool. 54:1863-1879.

Zach, R. and Falls, J. B. 1976b. Foraging behavior, learning, and exploration by captive ovenbirds (Aves: Parulidae). Can. J. Zool. 54:1880-1893.

Zach, R. and Falls, J. B. 1976c. Do ovenbirds (Aves: Parulidae) hunt by expectation? Can. J. Zool. 54: 1894-1903.

Zach, R. and Falls, J. B. 1978. Prey selection by captive ovenbirds (Aves: Parulidae). J. Anim. Ecol. 47:929-943.

Zach, R. and Smith, J. N. M. 1981. Optimal foraging in wild birds? I_n . A. C. Kamil and T. D. Sargent (eds.). Foraging behavior: ecological, ethological, and psychological approaches. Garland STPM Press, New York. pp. 95-109. 55

DEVELOPMENT OF FORAGING SKILLS

IN YOUNG BLACK OYSTERCATCHERS

INTRODUCTION

Young animals often have few or poorly developed foraging skills (e. g. Buckley and Buckley 1974, Dunn 1972, Groves 1978,

Orians 1969, Recher and Recher 1969, Verbeek 1977a) and may rely on parental feeding until they develop foraging proficiency.

Growth, physical maturation, and early experience are all components in the development of foraging in young animals.

Surprisingly, studies of development of foraging skills are rare (e. g. Cadrnan 1980, Davies 1976, Davies and Green 1 976,

Norton-Griffiths 1968, Rabinowitch 1968, 1969, Vince 1960). I studied development of foraging skills in black oystercatchers and attempted to identify important processes in the development of foraging skills. I studied morphological and behavioural components of development of foraging in young birds and analyzed prey choice, patch choice, searching behaviour, and prey handling by young birds.

STUDY AREAS

The summer portion of this study was conducted on Cleland

Island, an Ecological Reserve off the west coast of Vancouver

Island, B. C. The winter .portion of this study was conducted on 56

tidal mudflats in Lemmens Inlet (Lat. 49°10' N., Long. 125°50'

W. ) near Tofino, B. C. During the winter months mixed semi• diurnal tides 'in the area range from 0.2 metre to 3.9 metres.

Low tide exposes extensive mudflats where flocks of black oystercatchers come to feed.

STUDY ANIMAL

Breeding black oystercatchers and their chicks were the subjects of this study. From late April until the end of August breeding oystercatchers on Cleland Island defended territories that included nest sites and intertidal feeding areas. Black oystercatcher chicks are precocial, but they remain on their parents' territories until they can fly (approximately 35 days of age) and usually longer (50 or more days of. age). During this period most prey items eaten by chicks are provided for them by their parents. Shortly after hatching, all chicks on Cleland

Island were individually marked with unique combinations of colored plastic leg bands. Through at least their first year, young birds can be readily distinguished from adults by their dark brown bill tips and dull iris color. Adults have brilliant

red bills and chrome yellow irises.

By late August, most adults and flying chicks spent very

little time on Cleland. During the fall and winter months oystercatchers foraged on sheltered reefs and tidal flats in the area of Tofino, B. C. Some oystercatchers foraged on tidal mudflats in Lemmens Inlet, and these birds were the subjects of the winter foraging observations reported here. 57

METHODS

Intertidal Prey Organisms

The distribution and abundance of organisms taken as prey by black oystercatchers on Cleland Island were assessed by 176

50 x 50 cm quadrat samples made between high and low water line during extreme low tides. The abundance of small invertebrates living in spaces between mussels in mussel beds was assessed by eleven 25 x 25 cm quadrats. (Details of the sampling scheme are outlined in the previous chapter).

The abundance of mussels (Mytilus edulis) in the winter foraging area, Lemmens Inlet, was also assessed by quadrat sampling. Mussels occurred in large patches (up to 50 m long) on the mudflats. Fifteen quadrats (50 x 50 cm) in mussel patches were randomly selected, and the numbers and lengths of all mussels within the quadrats were recorded. A length-weight regression (Menge 1972) was used to estimate mussel biomass.

Chick Growth

Chicks were weighed and measured about every third day from hatching until they could no longer be caught (about 35 days of age when they began to fly). Bill length (exposed culmen) was measured in millimetres. Weight was recorded to the nearest gram. 58

Foraging Behaviour

Individual adults and chicks (0-55 days of age) in eight oystercatcher families were observed foraging during low tide periods from 3 hours before until 3 hours after low tide.

Continuous records of foraging behaviour were encoded using a digital event recorder. Records of foraging behaviour included the following: bird identification and zone of the intertidal; duration of prey-handling and chick-feeding events, prey type, and prey size with respect to the bird's bill length; duration of successful searches (periods of continuous search ending with a prey capture) and unsuccessful searches (periods of continuous search ending in a non-searching behaviour without a prey capture); duration of non-foraging behaviours such as sitting, standing, and flying that occur during periods of foraging. Data were analyzed by Fortran programs that I wrote and by the MIDAS statistical package. (Details of data collection are summarized in the previous chapter).

Foraging birds, especially chicks, sometimes selected prey items so small that the prey type could not be identified by an observer. These prey items were classified as "unknown" and assigned a wet weight of 0.02 g, the weight of the smallest prey items measured in the quadrat samples.

Data on foraging behaviour presented here were collected in two different places: 1) on Cleland Island breeding territories from late June through August, 1977, when chicks were 0-55 days old; 2) on Lemmens Inlet mudflats during January and February,

1-978, -when chicks were approximately 6 months old. 59

RESULTS

Prey Distribution and Abundance

Prey abundance, diversity, and range of sizes were quite high on the Cleland Island study site and very low at the

Lemmens Inlet site. On Cleland Island four intertidal zones were

identified between the high and low water lines. In order from high to low water they are: 1) spray zone (SPZO), 2) fucus zone

(FUCU), 3) mytilus zone (MYTL), 4) laminaria-postelsia zone

(LAPO). The abundance.and biomass of prey present in each zone

increased markedly between the high and low water lines (see

Figures 3 and 4). The greatest abundance and -greatest biomass of

prey occurred in the lowest zones of the intertidal. Certain

profitable prey types (e. g. chitons, Katharina tunicata, and

crabs, Oedignathus inermis) were recorded only in the lowest

zones.

In Lemmens Inlet, mussels (Mytilus edulis) were the most

abundant invertebrates occurring on the mudflats and the only

invertebrates that birds were observed to eat. The number of

live mussels in each of the fifteen 50 x 50 cm quadrat was

highly variable. An average of 100.4 mussels (s. d. = 102.6) was

found in each quadrat. Live mussels were relatively uniform in

size with an average length of 40.0 mm (s. d. = 3.3).

Chick Growth

•Black oystercatcher chicks grew from 30-35 grams at 60

hatching to 350 grams or more by the time of their first flight.

A dramatic change in bill length accompanied this change in body weight. Chicks' bills grew from about 15 mm at hatching to 45 mm or more by 35 days of age. Average chick bill lengths and weights for 10 day increments of chick age are shown in Figure

10. Average adult weights and bill lengths are included for compar i son.

Allocation of Foraging Time between Zones

Chicks and their parents foraged together and, consequently, spent similar proportions of their foraging time in each zone of the intertidal. Data on proportions of foraging time 'spent in each zone of the intertidal by adults and chicks are summarized in Table V. Chicks fed slightly but not significantly more often in upper intertidal zones than did their parents, and parents never foraged in the uppermost zone, the spray zone. For comparison, data on proportion of their foraging time adults without chicks spent in each zone are also included in Table V. Adults with and without chicks did not differ significantly in their allocation of foraging time between zones.

Prey Selection by Chicks and their Parents

Food delivered by parents and small "unknown" prey items that chicks picked out of mussel beds were frequently included in the chick diet. Young oystercatchers have short bills and therefore lack effective tools for procuring and handling marine 61

Figure 10: Average chick bill lengths and weights over ten day intervals from . hatching to fledging. (Vertical bars are 95% confidence limits; numbers beside data points are sample sizes). 62 TABLE V

Average proportion of foraging time spent

in each zone of the intertidal*

Adults 1 Adults6 Chicks with Chicks without Chicks Zone (N = 22) (N = 25) (N = 12)

SPZO 0.05 0.002 0.037

FUCU 0.19 0.053 0.03s

MYTL 0.58 0.70" 0.529

LAPO 0.18 0.245 0.4210

* Arcsin transformation was applied to all data before analysis of variance.

1 F-values for comparison of chicks with their parents

2 F = 2.12, df = 1,45, NS; 3 F = 1.91, df = 1,45, NS ;

• F = 2.24, df = 1,45, NS; 5 F = 0.79, df = 1,45, NS.

6 F-values for comparison of adults with chicks and adults without chicks:

7 F = 3.65, df = 1,35, NS; 8 F = 0.34, df = 1,35, NS;

9 F = 2.56, df = 1,35, NS; 10 F = 1.16, df = 1,35, NS 64

invertebrates (limpets, chitons, mussels, crabs, worms) eaten by adult black oystercatchers in the study area. Consequently, chicks depend heavily on their parents for food early in life.

The proportions of feedings chicks obtained on their own and the proportion of total food biomass these feedings represent are plotted in Figure 11. As chicks got older and bigger they obtained an increasing proportion of prey for themselves but continued to receive some food from their parents for an extended time. Flying chicks 50 days of age and older received

15% of their feedings and about 57% of their total food biomass

from their parents. In each zone, the average biomass of

individual prey selected by: 1) chicks, 2) parents, and 3) parents feeding chicks are shown in Figure 12 • for ten day

increments of chick age. The largest prey items in the chick diet were those items fed to chicks by their parents. Prey items

taken by chicks were generally much smaller than prey items

taken by adults either for themselves or for their chicks.

The largest prey items taken by chicks were obtained in the mytilus zone where the abundance of small prey items is greatest

(see Figure 4). (Average prey weights, with standard deviations,

taken by chicks in each zone are: SPZO - 0.02g (0.01), N = 26;

FUCU - 0.05g (0.14), N = 22; MYTL - 0.43g (1.55), N = 78; LAPO -

0.02g (0.07), N = 86;, F = 3.09, df = 3,211, p.< 0.05). Small

and weakly attached or unattached prey items such as amphipod

crustaceans, very small sea cucumbers and very small chitons

were often taken by chicks pecking among the mussels. Often,

these small, soft-bodied prey items lacking distinctive hard

parts were difficult to identify.accurately in the field. These 65

Figure 11: Proportions of feedings and food biomass procured chicks. CHICK AGE (DAYS) 67

Figure 12: Average biomass of individual prey 1) taken by chicks, 2) taken by adults, and 3) fed to chicks by adults in each zone of the intertidal. (Numbers below histogram bars are sarple sizes. Vertical bars are 95% confidence limits). 68

SPRAY ZONE 0.1 OT

. 0.01- 10 1 1 12 FUCUS ZONE 1.00

^ 0.1 0' to < 0.01' cr 82 31 1 7 3 19 1 99 2 4 44 o MYTILUS ZONE h- 10.0 0 X o UJ 1.0 0

>" 0.10 LU cr 0.01 5tifaif-2 2 ' 2 3 8 2 2 16 74 26 ' 5 40 53 47 49 118 8i 6 10 T LAM INAR IA-POSTELSIA ZONE

• ^ -> ^ ' ir-

1.0 0

0.10

0.01 13 2 1 1 93 20 * 1 1 7 5 " 2 9 8 1 2 20 86 ' 76 6 1 -9 10-19 20- 29 30-39 AO- 49 50-55 CHICK AGE (DAYS) = PREY TAKEN BY CHICKS rZ PREY TAKEN BY ADULTS JX, PREY FED TO CHICKS BY ADULTS 69

"unknown" prey items were common in the chick diet. Figure 13 shows the total frequencies with which each prey type was taken by 1) chicks, 2) adults, 3) adults feeding chicks. Chicks seldom took limpets or mussels, the most profitable items in their diet that adults provided for them. The frequencies of prey types taken, by chicks and those taken by adults for themselves differed significantly (X2 = 238.7, df = 6, p < 0.001). A further breakdown of prey types taken by chicks is presented in

Table VI. From.10 through 55 days of age, small "unknown" prey items are the most common prey items taken by chicks.

Searching Behaviour

Chicks' searching behaviour changed as they grew. Inter- peck intervals (seconds between'pecks) measure how frequently foraging birds investigated potential prey items. Inter-peck intervals of young chicks foraging in each zone of the intertidal were significantly longer than those of adults foraging in the same zone. Data on inter-peck intervals for chicks (averaged over 10 day increments of age) foraging in each zone of the intertidal are shown in Figure 14 along with average inter-peck intervals of adults foraging wrth their chicks. As chicks became older their average inter-peck intervals shortened and approached adult performance.

Winter Foraging

By mid-winter young and adult oystercatchers did not differ much in their foraging behaviour. During the winter, young and 70

Figure 13: Frequencies with which 7 prey types were 1) taken chicks, 2) taken by adults, and 3) fed to chicks by adults. 1.0

>- LU cr Q_ o

O 0.5 t— cr o Q_ o I acr.

0.0 BARNACLE CHITON CRAB LIMPET MUSSEL WORM UNKNOWN N=164 N=15 N=68 N=688 N=104 N=52 N=333

PREY TYPE = PREY TAKEN BY CHICKS, N= 212 "PREY TAKEN BY ADULTS, N=495 ^ >^PREY FED TO CHICKS // BY ADULTS, N = 717 TABLE VI

Numbers of various prey types taken by chicks of different ages

Chick age (days)

Prey Type 10-19 20-29 30-39 40-49 50-55

barnacles 0 0 0 1 7 4

crabs 0 3 1 0 1

amphipod crustac. 0 0 0 2 0

1impets 1 5 2 8 1

mussels 0 0 0 2 0

sea- cucumbers 0 1 1 1 0

worms 0 1 0 0 0

unknown 1 2 10 22 43 82

Hours* 0.29 1 .09 1 .35 4.09 1 .22

Total observation time of chicks foraging on their own. 73

Figure 14: Average duration of inter-peck intervals of chicks over ten day intervals of age in each intertidal zone. Average

inter-peck intervals of adults foraging in each zone are

included for comparison with chicks. (Vertical bars are 95% confidence limits). INTER-PECK INTERVALS (SECONDS)

O oo o o O, o, L 10-19 $

20-29 w en 30-39 "0 X) 40-49 £ 50-55 & M Oz ADULT m

10-19

20-29 $ -n c o o X 30-39 S cz CO o 40-49 ^ M 50-55 Oz > ADULT S m o m cn oo a 10"19 ^ 1 >-< 20-29 S CD 30-39 £ cr to 40-49 o fiiiiit* NI 50-55 ^ O ADULT » z o < m > 10-19 £ o 20-29 cn (/TTI < 30-39 g 3 CO Q z.0-49 * a. $ 50-55 8 XJ /\ ADULT gfc> C3 O m 75

adult oystercatchers foraged together in flocks on tidal mudflats in Lemmens Inlet. Uniformly sized mussels distributed across the mudflats were the only prey items birds took in this area. By midwinter, January and February, average inter-peck intervals for young birds did not differ statistically from average inter-peck intervals of adults foraging in the same area. (adults: average inter-peck interval = 4.39 sec, s. d. = 4.1, N = 512; young: average inter-peck interval = 4.55 sec, s. d. = 4.8, N = 596; F = 0.33, df = 1,1107, NS) . Young birds took slightly longer to handle mussels than did adults, but the difference is not statistically significant (adults: average handling time = 29.5 sec, s. d. = 17.9, N = 89; young: average handling time = 33.5 sec, s. d. = 22.3, N = 104;

F = 1.90, df'= 1,192, NS).

Prey Stealing

Prey stealing was exhibited by some young oystercatchers.

Young birds foraging on mudflats sometimes attempted to steal opened mussels from feeding birds. Typically, a young bird would approach a feeding individual and suddenly lunge towards it in an apparent attempt to steal a partially eaten mussel. Sometimes a young bird succeeded in- stealing a mussel, but often the feeding bird would run or f,ly away carrying the mussel. -A total of 22 chases were observed during seven low tide observation periods; 19 of these chases were initiated by young birds (13 chases towards adults, 6 chases towards other young birds). Six of the 13 times young birds chased adults, young birds 'succeeded in stealing partially eaten mussels. (On one occasion when a 76

young bird chased a young bird, the chaser succeeded in stealing a mussel). Three of the observed chases were initiated by adults

(2 chases towards young birds, 1 chase towards an adult), apparently as defensive responses to intrusions into feeding adults' individual distances. On none of these occasions did adults steal prey. Observed and expected numbers of chases based on the age composition of foraging flocks are summarized in

Table VII. Young birds inititated chases, apparent attempts to

steal food, more often than adults.

I made observations of uninterrupted foraging bouts by 21 adults and 22 young birds on the mudflats. Among the young birds

10 were observed to initiate chases, attempts to steal prey, and

the other 12 were not. Young birds in these two groups, chasers and non-chasers, did not differ in prey-handling times

(F =0.76, df = 1,103, NS) or duration of successful search

(F =0.06, df = 1,103, NS), but the two groups did differ in

duration of unsuccessful search (chasers: 40.1 sec,

s. d. = 36.3; non-chasers: 19.3 sec, s. d. = 16.4; F = 8.41,

df = 1,67, p < 0.01). The long durations of unsuccessful search

for chasers reflect the time these individuals spent in pursuit.

DISCUSSION

Growth and physical maturation are important in the

development of foraging skills in young black oystercatchers.

Other studies suggest that early experience and critical periods

are also important in behavioural development of young animals

(Bateson 1976). TABLE VII

Chases on winter foraging area* (Numbers in parentheses are proportions)

Group Chases initiated by: Composition Adult Young

Date Adults Young Obs. (Exp.) Obs. (Exp.)

29/1/78 19(0.56) 15(0.44) 2 (6.16) 9 (4.84)

30/1/78 17(0.65) 9(0.35) 0 (3.25) 5 (1.75)

13/2/78 10(0.50) 10(0.50) 1 (3.00) 5. (3.00)

*Results of Chi-square one-tailed test: 1 X2 = 6.38, df = 1, p < 0.01 2 X2 = 9.29, df = 1, p < 0.005 3 X2 = 2.67, df = 1, NS 78

Chick Growth

Black oystercatcher chicks undergo great morphological change from hatching at 35 grams to their first flight about 35 days later at 350 grams or more. Elongation of the bill into a more versatile foraging tool and increased strength are two important aspects of maturation and development of foraging proficiency in oystercatcher chicks. However, even after an oystercatcher's bill has reached adult length, a bird may lack the physical strength of an adult until at least two years of age due to incomplete skull ossification and, probably, less muscle mass (Cadman 1980). Furthermore, juvenile oystercatchers' pointed bill tips hinder their ability to handle prey such as mussels. Norton-Griffiths (1968) amputated the pointed bill tips of two 54 day old oystercatcher chicks and recorded- 11% and 15% decreases in the time birds took to handle mussels (Mytilus edulis). Both a well-developed, blunted bill and physical strength are necessary before a bird can proficiently handle the full array of prey types taken by adult oystercatchers.

Allocation of Foraging Time between Zones

Early exposure to particular habitat types appears important in developing habitat preferences that persist in later life. The early experience of chicks foraging with their parents and being fed in different zones of the intertidal may be important in establishing patterns of zone use by chicks.

Oystercatcher chicks foraged with their parents and allocated their foraging time between zones in similar proportions to 79

their parents. (These patterns of zone use were similar to those of adult oystercatchers without chicks). Klopfer (1963) has shown experimentally that early experience does affect habitat preference in later life. In his experiments, captive chipping sparrows (Spizella passerina) preferred laboratory habitats that they were exposed to from early life to novel habitats.

Prey Choice

Early experience of young oystercatcher chicks is probably very important in the development of prey recognition, prey preferences, and prey handling techniques. Although young chicks did not take many of the profitable prey types common in the adult diet (Figure 13), searching for and handling small prey in mussel beds gave chicks experience in recognizing and manipulating prey. As chicks grew and began to obtain some prey on their own, the parental contribution of food declined in a pattern similar to that noted by Davies (1976) in spotted flycatchers (Muse icapa striata) . However, even as the frequency with which chicks obtained prey for themselves increased, parents continued to provide the largest prey items consumed by chicks (Figure 12). Parental feedings probably gave chicks experience in handling prey, especially when adults brought them prey such as limpets and mussels that had not had the edible and

inedible parts separated.

Searching Behaviour and Prey Handling

Searching behaviour changed as.chicks matured. Young chicks 80

investigated prey items (as measured by inter-peck intervals) at much slower rates than did their parents foraging with them in the same zones (Figure 14). As chicks got older, the rate at which they searched increased and approached adult rates shortly after chicks began flying. Similar patterns of development of searching behaviour in young birds have been reported by Davies and Green (1976) in young reed warblers (Acrocephalus sc i rpaceus) and by Davies (1976) in spotted flycatchers. In these two species, number of movements per minute and capture attempts per minute, respectively, increased with age and approached adult performance.

By mid-winter, searching and prey-handling behaviour of young black oystercatchers on the mudflats was similar to that of adult oystercatchers. Young birds in mid-winter took slightly, but not significantly, longer to handle prey items

(mussels) than adults. This result is surprising in view of

Norton-Griffiths' (1968) claim that oystercatchers take up to 3 years to attain adult proficiency handling prey items. Data from a population of oystercatchers (H. palliatus) in Virginia indicate that prey size is the key factor in comparing prey handling times of adult and immature birds. Cadman (1980) found that adults and immatures had similar handling times for small oysters (Crassostrea virginica) less than 1 ml in volume, but that adults took significantly less time to handle larger oysters. Quinney and Smith (1980) obtained similar results in a study of adult and juvenile great blue herons (Ardea herodias) feeding on fish. Similarities in handling times of adult and first winter black oystercatchers foraging on mussels in Lemmens 81

Inlet probably occurred because the available mussels were relatively small and uniform in size. There were no large mussels at the Lemmens Inlet site to allow evaluation of proficiency of first winter birds over a broad range of prey sizes.

Prey Stealing

Young birds sometimes initiated chases in apparent attempts to steal prey from other foraging birds, mostly adults. Prey stealing by foraging oystercatchers also occurred in winter in

Virginia, and there, also, adults were more often the victims

(adults lost 1 of 5 prey items, immatures lost 1 of 14 prey items; Cadman 1980). Prey stealing has also been reported in gulls (Verbeek 1977b) and herons, egrets, and ibises in winter foraging aggregations (Kushlan 1978).

Birds may chase and steal prey for several reasons. 1) Some young birds may not obtain sufficient rations on their own

(e. g. Dare 1977) and may chase feeding birds to make up food deficits. 2) Young birds may still recognize feeding adults as providers of food, and chases may result because an adult does not offer food as a parent would. In this and other studies of oystercatchers (Cadman 1980; Norton-Griffiths 1969) young from 4 to 10 months of age successfuly begged from their parents.

Important Processes in Development of Foraging Skills

In oystercatchers, development of foraging skills that

•enable birds to exploit difficult-to-handle -prey items results 82

from morphological and behavioural changes. Young chicks are ill-equipped morphologically to handle prey types common in diets of adults, but physical maturation alone does not bestow a full complement of foraging skills on young birds. Learning and early experience probably play important roles in development of foraging behaviour as chicks grow.

The importance of learning in the development of feeding behaviour in oystercatcher chicks was demonstrated by Horlyk and

Lind (1978). They concluded from experiments with newly hatched oystercatchers (H. ostralegus) that learning ability and early experience explained the development of chick responses during parental feeding. In another experiment they demonstrated that a

30 hour-old chick could distinguish between edible and inedible objects. The chick developed a clear preference for edible objects by the end of the first five minute test period (Horlyk and Lind 1978). The importance of learning and experience in all phases of development of foraging behaviour in young oystercatchers is further demonstrated in Norton-Griffiths'

(1967) study of mussel-feeding techniques in oystercatchers.

Studies of other species suggest that early experience

(e. g. Weigl and Hanson 1980) and critical periods (e. g.

Rabinowitch 1968, 1969; Smith 1972; Vince 1960) are also

important in the development of foraging behaviour in young animals. However, these areas of development were not addressed

in this study. An important developmental period apparently

occurs between the time chicks are about 50 days old and leave

the natal area and mid-winter when chicks are observed foraging

on their own on tidal mudflats. 83

LITERATURE CITED

Bateson, P. P. G. 1976. Rules and reciprocity in behavioural development. In. P. P. G. Bateson and R. A. Hinde (eds.). Growing points in ethology. Cambridge University Press. Pp.401-421.

Buckley, F. G. and Buckley, P. A. 1974. Comparative feeding ecology of wintering adult and juvenile Royal Terns (Aves:Laridae, Sterninae). Ecology 55:1053-1063.

Cadman, M. D. 1980. Age-related foraging efficiency of the American Oystercatcher (Haematopus palliatus). M.Sc. thesis, University of Toronto.

Dare, P. J. 1977. Seasonal changes in body-weight of oystercatchers Haematopus ostralequs. Ibis 119:494-506.

Davies, N.. B. 1976. Parental care and the transition to independent feeding in the young "spotted flycatcher (Museicapa striata). Behaviour 59:280-295.

Davies, N. B. and Green, R. E. 1976. The development and ecological significance of feeding techniques in the Reed Warbler (Acrocephalus scirpaceus). Anim. Behav. 24:213- 229.

Dunn, E. K. 1972. Effect of age on fishing ability of Sandwich Terns (Sterna sandvicensis). Ibis 114: 360-366.

Groves, S. 1978. Age-related differences in ruddy turnstone foraging and aggressive behavior. Auk 95:95-103.

Horlyk, N.-O. and Lind, H. 1978. Pecking response of artificially hatched Oystercatcher Haematopus ostralequs young. Ornis Scand. 9:138-145.

Klopfer, P. 1963. Behavioral aspects of habitat selection: the role of early experience. Wilson Bull. 75:15-22.

Kushlan, J. A. 1978. Nonrigorous foraging by robbing egrets. Ecology 59:649-653. 84

Menge, B. A. 1972. Foraging strategy of a starfish in relation to actual prey availability and environmental predictability. Ecol. Monogr. 42:25-50.

Norton-Griffiths, M. 1967. Some ecological aspects of feeding behaviour of the Oystercatcher (Haematopus ostralegus) on the edible mussel Mytilus edulis. Ibis 109:412-424.

Norton-Griffiths, M. 1968. The feeding behaviour of the oystercatcher (Haematopus ostralegus). Ph.D. thesis. Oxford.

Norton-Griffiths, M. 1969. The organisation, control, and development of parental feeding in the oystercatcher (Haematopus ostralegus). Behaviour 34:55-114.

Orians, G. H. 1969. Age and hunting success in the Brown Pelican (Pelecanus occ identali s). Anim. Behav. 17:316-319.

Quinney, T. E. and Smith, P. C. 1980. Comparative foraging behaviour and efficiency of adult and juvenile great blue herons. Can. J. Zool. 58:1168-1173.

Rabinowitch, V. E. 1968. The role of experience in the development of food preferences in gull chicks. Anim. Behav. 16:425-428.

Rabinowitch, V. E. 1969. The role of experience in the development and retention of seed preferences in Zebra Finches. Behaviour 33:222-236.

Recher, H. F. and Recher, J. A. 1969. Comparative foraging behaviour of adult and immature little blue herons (Florida caerulea). Anim. Beh. 17:320-322.

Smith, S. M. 1972. The ontogeny of impaling behaviour in the "Loggerhead Shrike Lanius ludovici-anus L. Behaviour -42:232- 242.

Verbeek, N. A. M. 1977a. Age differences in the digging frequency of Herring Gulls on a dump. Condor 79:123-125.

Verbeek, N. A, M. 1977b. Comparative feeding behavior of immature -and adult gulls. Wilson Bull. 89:415-421. 85

Vince, M. A. 1960. Developmental changes in responsiveness of the Great Tit (Parus major). Behaviour 15:219-243.

Weigl, P. D. and Hanson, E. V. 1980. Observational learning and the feeding .behavior of the red squirrel Tamiasciurus hudsonicus: the ontogeny of optimization. Ecology 61:213- 218. 86

GROWTH, SIBLING RIVALRY, AND CHICK PRODUCTION

IN BLACK OYSTERCATCHERS

INTRODUCTION

Growth and survival of young animals are affected by how parental investment is partitioned among offspring and how much energy in invested in each offspring (Smith and Fretwell 1974).

Animals without parental care partition energy among eggs, and this energy allocation affects subsequent growth and survival of offspring. Energy is also partitioned among offspring through parental care, and its allocation affects offspring growth and survival (Brockelman 1975). When offspring are heavily dependent on parental parental feeding, parental foraging performance and

fitness (number or quality of offspring) should be related. In principle, it should be possible to relate parental foraging performance to reproductive output. In the case analyzed here,

the relationship between parental foraging performance and

fitness was examined indirectly by studying chick growth and

chick production in black oystercatchers (Haematopus bachmani).

STUDY AREA

Cleland Island near Tofino, British Columbia was the site

of this study. The climate is mild and wet with an average

•annual temperature of 9°C.and average annual precipitation-..of 87

309 cm recorded for Tofino (Climate of British Columbia, 1976).

Major storms sometimes occur during the summer. During storms, heavy seas continually wash the intertidal zone, even at low tide, and poor drainage on the island results in flooding and accumulation of water that evaporates slowly.

STUDY ANIMAL

About 30 pairs of black oystercatchers hold breeding territories on Cleland. Birds build simple nest cups lined with bits of broken shell in basalt crevices or lay their eggs in scrapes made on beaches of broken shell. Clutch size is 1-3 eggs. Incubation begins when the clutch is complete and lasts about 26 days. Eggs hatch within a few hours of each other. (The longest hatching interval observed during this study was of 12-

18 hours). First hatched chicks are brooded while the rest of the clutch hatches.

Egg laying begins in mid-May, and replacement clutches may be laid if eggs are lost. In many seasons whole clutches are lost due to flooding during storms, and every year northwestern crows (Corvus brachyrhynchos) destroy clutches of oystercatcher eggs. Glaucous-winged gulls (Larus glaucescens) are chick predators, but they were never observed to take oystercatcher eggs. Many birds replace the first clutch if it is lost, and some birds relay after losing their second clutch. In no year were birds observed to lay replacement clutches after losing chicks nor has egg-iaying continued after 15 July.

Black oystercatcher chicks are precocial but dependent ,on parental feeding for an extended period (Figure 11). Parent 88

oystercatchers feed their chicks during low tide periods. Very young chicks remain near the nest, and parents take turns guarding the chicks and carrying food item by item from the intertidal to the chicks. As they become bigger and more agile, chicks follow their parents into the intertidal to be fed.

Around the time they begin to fly (35+ days), chicks begin to capture small prey items for themselves, but the most profitable prey items are still provided by their parents (see Figure 13).

METHODS

Black Oystercatcher Territories

Black oystercatcher territories and nest sites were mapped

from 1975 through 1978. Territories were delineated by recording

locations of boundary disputes. In 1976 and 1977 I censused the

island daily (weather permitting) from May through August to gather data for territory mapping. In 1975, territory and nest data were gathered in censuses made during two four-day visits

to the island in mid-July and mid-August, and 1978 data were

gathered over 5 trips between the end of June and the end of

August. Three breeding birds banded on their territories in 1971

by E. B. Hartwick (pers. comm.) were present on the same

territories from 1975 through 1978.

Chick Growth and Survival

.Detail-s of chick .growth and survival were recorded in 1-976 89

and 1977. Nests were located as clutches were commenced and checked every three days to obtain nest histories. After eggs pipped, nests were checked daily to determine the date of hatching. Unique combinations of colored plastic leg bands and numbered metal wildlife service bands were put on each chick to ensure positive identification. Chick weights and exposed culmen lengths were measured every third day (as weather permitted) until chicks flew or disappeared. All weighings were done within

1 1/2 hours of high tide, when most intertidal feeding areas had been under water for a few hours. This minimized the contribution of recently filled crops to chick weights.

Chick Feeding

Records of parents feeding chicks were made in 1977.

Observations were made with a 20 X spotting scope over low tide periods. When a chick was fed the prey type, prey size, chick

identification, parent identification, and time of feeding were recorded. Interactions between siblings were also recorded. The data presented here focus on four oystercatcher families with two-chick broods. They were observed for 43.2 hours during 14 observation periods. The chicks observed ranged in age from 14

to 53 days (day of hatching is day 0).

RESULTS 90

Territory fidelity, clutch size, and breeding success

Black oystercatchers on Cleland Island showed a high degree of territory fidelity. From 1975 through 1978 no shifts in territory boundaries were observed, and in many cases, the same nest sites were used in two or more years. Three adults color- banded in 1971 occupied the same territories in 1975-78 that they had occupied at the time of color-banding. Four adults banded on territories in 1976 defended the same territories through 1978. Another color-banded bird and its mate occupied the same territory in 1976 and 1977, but in 1978 they had disappeared and the territory was vacant. A sixth color-banded bird left its territory, after being banded in 1976, and the territory was unoccupied until 1980 when the banded bird reappeared and bred on the same territory. The territorial fidelity of these marked birds, constancy of territorial boundaries of unmarked birds, long lifespan, and frequent reuse of nest sites in successive years suggest long term fidelity by birds to territories.

Black oystercatcher clutch size ranges from 1 to 3 eggs.

Data on 310 clutches gathered in three seasons of this study and three seasons of Hartwick's (1974) study are presented in Table

VIII. The mean size of all 310 clutches was 2.04 eggs ts. d. = 0.66).

Breeding success of individual pairs of black oystercatchers was highly variable between years. From nest records and territory maps it is possible to identify 32 territories on which birds attempted breeding in more than one year from 197 5-T'978. For eleven of these territories presence of TABLE VIII

Black oystercatcher clutch size in 6 years, Cleland Island

Number of Clutches Mean Total • clutch Year T eqq 2 eqqs 3 eggs clutches size s.d.

1 970* 1 4 38 9 61 1 .92 0.61

1 97 1 * 8 26 1 4 48 2.13 0.67

1 972* 11' 34 1 3 58 2.03 0.65

1 976 1 1 25 18 54 2.13 0.73

1 977 1 2 38 1 0 60 1 .97 0.61

1978** 5 1 5 9 29 2.14 0.69

Total 61 176 73 310 2.04 0.66

* Data from E. B. Hartwick (1974) ** Data based on 5 visits to the island. 92

color-banded adults and annual reuse of the same nest site indicate that the same pairs bred on these territories for the four years of observations. Data on chick production of these pairs is shown in Table IX. These data show that: 1) four pairs did not successfully raise any young in 4 years; 2) one pair successfully raised young in all 4 years; 3) six pairs were successful in some years and failed in other years; 4) there was no tendency for success of a pair to increase from one year to the next over the four years of observation.

Brood size and chick growth

Black oystercatcher chick growth from hatching to time of

first flight depends on brood size and a chick's rank in the brood (Figure 15). Analysis of variance was used to analyze chick weights. Data were analyzed according to brood size for

three day intervals of age as shown in Figure 15. Weights of chicks in one-chick and two-chick broods did not differ

significantly until chicks were .14-16 days old. At that time,

average weights of chicks in one-chick broods were greater than

weights of chicks in two-chick broods (F = 8.89, df = 1,22,

p < 0.005), and these differences persisted until chicks could

fly. These differences persisted at least until chicks began

flying. Average weights of siblings in two-chick broods did not

differ statistically, but in some broods, the larger chick

weighed up to 48% more when chicks began to fly. The average

weight difference at the time siblings began flying was 84.2

grams (N = 4, s. d. = 53.2 g; this includes one brood in which

the smaller •sib never flew). The two extreme cases of sibling TABLE IX

Number of chicks produced on 11 black oystercatcher territories, 1975-1978

Territory 1975 1 976 1 977 1 978 Total

1 (B) 1 1 1 2 5

2(B) ' 1 0 1 0 2

3 3 0 . 1 0 4

4 3 0 o' 1 4

5 0 0 0 ' 0 0

6(B) 1 0 -0 - X 1

7 1 1 0 2 4

8(2B) 0 0 0 0 0

9(B) 0 0 0 0 0

10 1 0 0 1 2

1 1 0 0 0 0 0

B .= 1 banded parent 2B = 2 banded parents X = adults' absent from territory 94

Figure 15: Average chick weights (log scale) are-plotted against chick age in days. Data are plotted according to brood-size and chick rank in brood (for two-chick broods). (Vertical bars are

95% confidence limits). 95

500T

7t 9T8 to < tr '100 t i o

"t 1 I 20 ' 0-3 ' 5-9 ' 10-12 'u-16 ' 17-19 ' 20-22 ' 2 3- 2 5 26-29 34-37

AGE (DAYS)

X SINGLE CHICKS BIGSIBS * LITTLE SIBS 96

weight differences are illustrated in Figure 16. The chick that was 152 grams lighter than its sibling never flew and died of apparent starvation at 49 days of age, nine days after its sib began flying. The small sib was unable to follow its parents and flying sib to nearby offshore reefs to forage and be fed.

Weight Differences in Two-chick Broods

Weight differences between siblings in two-chick broods were possibly affected by dominance relations between sibs and resulting unequal distribution of food. On one occasion I observed the establishment or reinforcement of dominance between

sibs. The 15 day old chicks weighed 158 grams and 128 grams.

During previous observations no. conflict between, these sibs had

been observed.

The siblings were sitting near the nest site with one

parent. The second parent had been foraging and returned

carrying a piece of food. Both chicks approached the returning

parent and reached it together. The big sib turned and began

pecking the little sib on the head and neck. The little sib ran

away pursued by the big sib. After chasing for several meters,

the big sib returned to the parent and took the food. The little

sib stayed away. Subsequently, whenever both chicks approached a

parent to be fed, a lunge by the big sib towards the little sib

sent the latter running away from the parent with food.

Sometimes lunging and chasing were overt and vigorous, but often

merely the big sib turning towards the lttle sib resulted in the

little -sib moving away or staying away from a parent with food.

In this and other two-chick broods, dominance relations 97

Figure 16: Growth of siblings in the two two-chick broods with a) minimum and b) maximum weight differences at the time chicks began flying.• AGE (DAYS) 99

persisted after the chicks were flying and still being fed by their parents.

Occasionally little sibs in broods chased big sibs, but 36 of 38 overt chases were initiated by big sibs (X2 = 30.42, df = 1, p < 0.005). During observation sessions when chasing was observed, little sibs were fed much less often than big sibs

(Figure 17, X2 = 85.78, df = 1, p < 0.001). However, during sessions when chasing did not occur, little sibs were fed more often than big sibs (X2 = 40.03, df = 1, p < 0.005). As a result of chasing and dominance big sibs could control access to parental feedings.

Chick survival

Heavier oystercatcher chicks had a better chance than lighter chicks of surviving to take their first flight. By, twenty days of age, weight differences related to brood size were well developed (Figure 15). Chicks that survived to fly were heavier at 20 days of age than chicks that did not (Figure

18, t=2.36, df=21, p<0.05). All chicks heavier than 200 grams at

20 days of age eventually flew except one that was probably eaten by a river otter, Lutra canadensis (pers. obs.).

Chick survival from hatching varied with brood size and rank in brood (Figure 19). Mortality was highest during the

first week after hatching. In one-chick broods all of the mortality occurred during this period. In two-chick broods mortality occurred until chicks began flying.

Sometimes chicks vanished, and the cause of chick mortality could not be determined. Chicks that disappeared during big 100

Figure 17: Frequency of parental feedings and sibling chases in two-chick broods. 320

LITTLE SIB 1 02

Figure 18: Chick survival to fly and weight at 20 days of age. 103

.1

151-175 175-200 200-225 225-250 250-275 275-300 300-325

GRAMS

SURVIVED TO FLY T DID NOT FLY 1 04

Figure 19: Chick survivorship from hatching versus chick age in days. Data are plotted according to brood-size and chick rank

(in two-chick broods). 105

1.0

0.8

X to 0.6 cr o > > cr 0.4 Xi ZD 00

—+

0.2

0.0' o U 21 2 8 35 kl 49 AGE (DAYS)

X SINGLE CHICKS N = 10 • BIG SIBS N = 16 + LITTLE SIBS N= 16 106

storms were probably swept away by storm surge and rain run-off.

Crows preyed on oystercatcher eggs and probably also took some very young chicks that were still being brooded by their parents. Gulls were observed to attack chicks, and parent oystercatchers responded by persistently flying towards and striking attacking gulls. However, chicks were still lost to gulls. In August, 1978, a color-banded leg of a chick about 4 weeks old and over 300 grams in weight was found in a gull pellet.

Survival after chicks could fly was similar for chicks from different brood sizes. Similar proportions of color-banded black oystercatchers from one-chick and two-chick broods were resighted as one and two year olds (Table X). There was no difference in weight at the time of first flight between birds resighted one or two years later and those not resighted

(resighted: 351 g, s. d. = 45.1; not resighted: 350 g,

s. d. =68.7; t = 0.32, df = 15).

Clutch size and chick production

One-egg and two-egg clutches produced similar numbers of

chicks per clutch. Survival of chicks to fly from one and two-

egg clutches in which all eggs hatched is shown in Table XI. The

number of chicks surviving to fly did not differ significantly

with clutch- size (X2 = 0.57, df = 1, NS). An average of 0.50 (±

0.53) chicks per clutch survived to fly from one-egg clutches

while 0.75 (± 0.77) chicks per clutch survived to fly from two-

-egg clutches. TABLE X

Resightings of black oystercatchers

banded as chicks in 1976 and 1977*

Number Number of chicks resighted Brood size fledged by Sept.1978

1 chick 7 3(0.43)

2 chicks 10 6(0.60)

* P = 0.30, Fisher Exact Test (Siegel 1956). TABLE XI

Black oystercatcher chick production (proportion in parentheses)

ONE-EGG CLUTCHES

1 976 1 977 Total

Hatching 1 chick 2(1.00) 8(1.00) 10(1.00)

Fledging 1 chick 1(0.50) 4(0.50) 5(0.50)

Fledging 0 chicks 1(0.50) 4(0.50) 5(0.50)

TWO-EGG CLUTCHES

1 976 1 977 Total

Hatching 2 chicks 7(1.00) 9(1.00) 16(1.00)

Fledging 2 chicks 2(0.29) 1(0.11) 3(0.19)

Fledging 1 chick 1(0.14) 5(0.56) 6(0.38)

Fledging 0 chicks 4(0.57) 3(0.33) 7(0.44) 109

DISCUSSION

Chick growth and sibling rivalry

Dependence of black oystercatcher chick growth on brood size (Figure 15) suggests that availability of food provided by parents limits chick growth. Two-chick broods are more demanding of parents' time and energy than one-chick broods, and differences in chick weights related to brood size appeared by the time chicks were two weeks old. Chick growth patterns indicate that parental investment in chicks increases with brood size but that the increase is not proportional to brood size.

Less than proportional increase in parental investment with increasing brood size has been reported in numerous other bird species and these reports are summarized by Klomp (1970). An hypothesis that is consistent with the less than proportional increase in parental feeding with increased brood size was proposed by Norton-Griffiths (1969). He hypothesized that parental feeding of oystercatcher chicks is motivated by parental hunger and that chick feeding will only occur during periods when parents are motivated to feed themselves.

Data on chick growth provide indirect evidence that parental feedings-are a limited resource for chicks in two-chick broods. Sibling rivalry is a response to this food limitation, and it may affect chick growth and survival. Prey availability to chicks is controlled by parental delivery of prey, and chicks, especially large chicks, are probably rarely satiated.

Hunger and the sight of a parent carrying food are likely 1 10

stimuli for the occurrence of sibling rivalry. Hunger alone does not stimulate dominance behaviour in black oystercatcher chicks since siblings showing well-established dominance relations in the presence of a parent with food will spend hours crouched side-by-side in hiding places between feedings. However, in cases of severe food limitation hunger alone may stimulate dominance. An extreme example of this is cannibalism in nestling short-eared owls (Asio flammeus). Ingram (1959) reported that cannibalism between siblings was most frequent in years of food shortage when parents were unable to provision their nest by caching food within reach of bigger chicks. A similar situation was reported for South Polar skuas (Catharacta maccormicki).

Procter (1975) demonstrated that nutritional condition of skua chicks (measured as percent of standard weight at a particular age) was positively correlated with the occurrence and intensity of aggression between sibs in naturally and artificially created

two-chick broods.

Once established, dominance relations between siblings

persisted. In this study small sibs (in four broods) chased

dominant big sibs in only 2 of 38 overt chases and in no case

was the dominance order observed to reverse. Studies of

dominance in Cactus wrens, Campy1orhynchus brunneicapillus

(Ricklefs and Hainsworth 1967) and eaglets (Meyburg 1978)

suggest that dominance is quickly and irreversibly decided in

one fight. Dominance is usually based on weight or age

differences attributed to asynchronous hatching (e. g. Bryant

1978, Ingram 1962, Meyburg 1973, Miller 1973, Procter 1975).

However, dominance has been reported•between eagle •chicks of the 111

same weight (Meyburg 1978), and Ricklefs and Hainsworth (1967) report that the younger of two cactus wrens emerged from a fight as the dominant. In black oystercatchers heavier chicks were dominant.

Chick survival

Brood size and chick weight affect survival until chicks are able to fly. In one-chick broods all chick losses occurred during the first week after hatching while in two-chick broods chick losses continued through the time chicks began flying

(Figure 19). These patterns of chick loss may be due to behaviour of parent oystercatchers. Chicks are frequently brooded by their parents during the first week or longer after hatching, and predators, crows and gulls, are able to locate chicks by observing behaviour of parents. As chicks grow older and are brooded less, predators are no longer able to locate chicks directly by observing their parents (Y. Yom-Tov, pers. comm.). Instead, predators must search for cryptically colored chicks. A predator probably has a better chance of finding a chick on a territory with a two-chick brood than on a territory with a one-chick brood.

Heavier chicks had a better chance of surviving to fly.

Twelve of thirteen chicks weighing over 200 grams at ,20 days of age survived to fly while only 5 of 10 chicks weighing 200 grams or less eventually flew (Figure 18). There were no differences in weight at time of first flight between birds that were and were not resighted as one and two-year olds, nor were there any differences related to brood size in the proportion of birds 1 1 2

resighted one and two years later (Table X). This suggests the critical life history period for black oystercatchers is survival from hatching to first flight. Once a bird can fly its prospects for survival are apparently quite good although no data is available on eventual recruitment of flying young to the breeding population.

Chick production and parental investment

Chick production by black oystercatchers was low and varied between years. One hundred and fourteen clutches of eggs survived at least to the beginning of incubation in 1976 and

1977 (Table VIII). Only 23% (26 clutches) survived to hatch chicks, and just 12% (14 clutches) produced flying chicks (Table

XI). Chick production by any pair of oystercatchers varied considerably between years, and over four seasons some pairs did not produce any flying chicks at all (Table IX). Predation and unpredictable bouts of severe weather were critical factors affecting chick production. These factors, unrelated to parental

foraging performance, placed constraints on chick production.

Breeding success and chick production for Cleland Island

black oystercatchers were much lower than Harris (1967) reported

for oystercatchers (Haematopus ostralequs) on Skokholm Island,

-Wales. There, 37% and 59% of eggs produced flying young in two

consecutive years. On Cleland Island Hartwick (1974) reported

that 13%, 9%, and 13% of eggs produced flying young in three

years (1970-1972), and during this study only 5% and 9% of eggs

produced flying young.in 1976 and 1977.

One and two-chick broods during the period of this study 1 1 3

produced similar numbers of flying young (Table XI). In two- chick broods lower chick growth and occurrence of sibling rivalry suggest that chicks in two-chick broods were not as well provisioned by their parents as chicks in one-chick broods. In only one brood in this study (see "Brood size and chick growth") did these factors contribute to starvation of a chick. However, in seasons of unusually stormy weather that severely limits access to feeding areas, weight differences maintained by sibling rivalry could effect chick production. Siblings of different sizes represent different amounts of parental investment (Trivers 1972), and sibl-ing rivalry can maintain or increase these differences (O'Connor 1978). Differences in sizes of siblings facilitate rapid responses to sudden changes in food availability through brood reduction (Ricklefs 1965). Heavier chicks have more energy reserves to support them, especially through periods of food shortage (O'Connor 1976). For instance,

in skua broods that fledged 2 chicks the bigger chicks were heavier than the bigger chicks in two-chick broods that eventually fledged only one chick (Procter 1975). Sibling

rivalry is a mechanism that may preserve some chick production

under conditions of severe food shortage. 1 1 4

LITERATURE CITED

Brockelman, W. Y. 1975. Competition, the fitness of offspring, and optimal clutch size. Amer. Nat. 109: 677-699.

Bryant, D. M. 1978. Establishment of weight hierarchies in the broods of house martins Delichon urbica. Ibis 120: 16-26.

Climate of British Columbia, 1976: Tables of temperature, precipitation, and sunshine. British Columbia Ministry of Agriculture. Victoria, B. C, 82 pp.

Harris, M. P. 1967. The biology of oystercatchers Haematopus ostralequs on Skokholm Island, S. Wales. Ibis 109: 180-193.

Hartwick, E. B. 1974. Breeding ecology of the black oystercatcher (Haematopus bachmani Audubon). Syesis .7:83- 92.

Ingram, C. 1959. The importance of juvenile cannibalism in the breeding biology of certain birds of prey. Auk 76: 218-226.

Ingram, C. 1962. Cannibalism by nestling Short-eared Owls. Auk 79: 715.

Klomp, H. 1970. The determination of clutch-size in birds: a review. Ardea 58: 1-151.

Meyburg, B. 1973. Sibling aggression and mortality among nestling eagles. Ibis 116: 224-228.

Meyburg, B. 1978. Sibling aggression and cross-fostering of eagles. In. Endangered Birds: Management Techniques for Preserving Threatened Species. S. A. Temple (ed.). Univ. of Wisconsin Press, Madison.

Miller, R. S. 1973. The brood size of cranes. Wilson Bull. 85: 436-441.

Norton-Griffiths, M. 1969. The- organisation, control, and development of parental feeding in the oystercatcher (Haematopus ostr-alegus) . 'Behaviour 34 : 55-1 1 2 . 1 1 5

O'Connor, R. J. 1976. Weight and body composition in nestling Blue Tits Parus caerulus. Ibis 118: 108-112.

O'Connor, R. J. 1978. Brood reduction in birds: selection for fratricide, infanticide, and suicide. Anim. Behav. 26:79- 96.

Procter, D. L. C. 1975. The problem of chick loss in the South Polar Skua Catharacta maccormicki . Ibis 117: 452-459.

Ricklefs, R. E. 1965. Brood reduction in the Curve-billed Thrasher. Condor 67: 505-510.

Ricklefs, R. E. and F. R. Hainsworth. 1967. The temporary establishment of dominance between two hand-raised juvenile Cactus Wrens (Campylorhynchus brunneicapi1lus) . Condor 69: 528.

Siegel, S. 1956. Nonparametric Statistics for the Behavioral Sciences. McGraw-Hill Book Co., New York.

Smith, C. C. and Fretwell, S. D. 1974. The optimal balance between size and number of offspring. Amer. Nat. 108:499- 506.

Trivers, R. L. 1972. Parental investment and sexual selection. In. B. Campbell (ed.). Sexual selection and the descent of man. Aldine, Chicago, pp. 139-179. 1 16

GENERAL CONCLUSION

Curiosity about how animals forage in patchy environments has generated extensive theoretical, laboratory, and field work in foraging ecology. Work on foraging ecology addresses three main issues - function, mechanism, and consequences.

Functional description of foraging has been of considerable interest to theoreticians and empiricists. Laboratory and field studies have been the testing grounds for foraging theory. These studies have shown that animals are very good at solving foraging problems they confront in the laboratory and field, but

1) discrepancies exist between observed and theoretically predicted performance, and 2) behaviour of foraging animals violates assumptions of foraging theory in all but the most restrictive laboratory set-ups. This study reports discrepancies between observed foraging performance of black oystercatchers and theoretical predictions concerning prey choice, patch choice, and allocation of foraging time between patches. These discrepancies occurred because the rocky intertidal is a variable environment, birds foraged for multiple different prey species, and birds probably foraged simultaneously for several different currencies (e. g. calories, trace nutrients, etc.).

Furthermore, because oystercatchers are territorial, assumptions about random movement through the foraging area are probably violated.

Mechanisms of foraging have received relatively little attention from ecologists.- Laboratory experiments that 1 1 7

manipulate single variables have been critical in determining

the range of discriminatory abilities of animals (e. g. large

vs. small prey, Goss-Custard 1970; rich vs. poor patches, Smith

and Sweatman 1974), but these experiments have provided little

insight into processes important in foraging animals, especially

processes concerned with making decisions about foraging. These

experiments do not aid in the understanding of mechanisms of

foraging largely because the problems they pose are

unrealistically simple compared to the problems foraging animals

solve daily in nature. For instance, black oystercatchers forage

as though simultaneously solving problems of prey choice and

patch choice, but neither theory nor laboratory studies suggest

what protocol (if any) an oystercatcher foraging in the rocky

intertidal uses to solve these problems. Better understanding of.

foraging mechanisms will provide useful insights for improving

functional descriptions of foraging.

Studies of behavioural development in young animals are one

way to gain an understanding of important processes in foraging

animals. Young, unskilled animals are good subjects for studying

the acquisition and refinement of foraging skills and

techniques, identifying critical developmental periods for

foraging skills, and assessing the development of discriminatory

capacities. Results from this study of young oystercatchers

suggest that growth and physical maturation are important in

early behavioural development, and it is suggested that learning

and early experience are also important. In adults, learning and

recent experience are probably important processes in the fine-

tuning of foraging behaviour. 1 18

Relating fitness to foraging performance is a significant challenge for theoreticians, laboratory and field workers.

Foraging performance is an important determinant of an animal's fitness because it affects how much energy an animal has available to invest in offspring and parental care (Schoener

1971). However, the translation of foraging performance into fitness is fraught with difficulties.

To evaluate foraging performance in terms of fitness it is first necessary to relate some measure of foraging performance to reproductive output of an individual and its progeny, but this is particularly difficult in long-lived birds such as oystercatchers (breeding birds of 27-34+ years of age have been reported from Europe, Nice 1962). Furthermore, many constraints affect' reproductive success and fitness (McFarland 1976).

Unpredictable environmental conditions and hazards to eggs and young (e. g. crows and gulls) result in low and variable annual production in Cleland Island oystercatchers and have nothing to do with a ..bird's foraging performance. Territory availability and quality may affect reproductive output, but they are also unrelated to foraging performance. Finally, foraging performance may not be a good measure of fitness-because intense selection acting on foraging performance may be infrequent (e. g. Boag and

Grant 1981). Because fitness is subject to so many constraints,

it may not be possible in most cases to use foraging performance as a measure of fitness.

In future studies of foraging ecology, the interface between function and mechanism should be a particularly productive area of research. Understanding mechanisms -and 119

processes in foraging animals will provide new foundations for functional descriptions of foraging. The translation of foraging performance into fitness is likely to remain an elusive goal because fitness is the integration of all facets of an animal's existence, not merely foraging performance, but in principle it is possible to relate foraging performance to fitness. 1 20

LITERATURE CITED

Boag, P. T. and Grant, P. R. 1981. Intense natural selection in a population of Darwin's Finches (Geospizinae) in the Galapagos. Science 214:82-85.

Goss-Custard, J. D. 1970. Factors affecting the diet and feeding of the redshank (Trinqa totanus). In . A. Watson (ed.). Animal populations in relation to their food resources. Blackwell Scientific Publications, Oxford, pp. 101-110.

McFarland, D. J. 1976. Form and function in temporal organization of behaviour. I_n . P. P. G. Bateson and R. A. Hinde (eds.). Growing points in ethology. Cambridge University Press, Cambridge, pp. 55-93.

Nice, M. M. 1962. Oystercatcher 34 years old - the oldest ringed bird to date of the Vogelwarte Helgoland. Bird-Banding 33:205.

Smith, J. N. M. and Sweatman, H. P. 1974. Food searching behaviour of titmice in patchy environments. Ecology 55:1216-1232. 121

APPENDIX A

Length(x)-weight(y) regressions used to calculate

wet weights of invertebrates measured in intertidal quadrats

and consumed by black oystercatchers

SPECIES EQUATION N

Nereis vexillosa y = 0. 199x - 0.398 1 6

Katharina tunicata log(y) = 2.451og(x) - 3.71 22

Tonicella 1ineata log(y) = 0.03x - 1.11 11

Collisella digitalis

and C^ Pelta log(y) = 2.771og(x) - 4.33 86

(less than 20mm)

Notoacmea scutum log(y) = 5.861og(x) - 8.31 44

(greater than 20mm)

Hemigrapsus nudus1 y = 1.5(1 . 11x - 0.42)2 1 4

Orchestoides californiana2 y = (9.37x - 2.66)2 1 5

Mytilus cali forianus: y = (0.233x - 0.27)

1 Although oystercatchers were never observed eating Hemigrapsus nudus, this species was used to obtain an approximation of the wet weight of Oediqnathus inermis because it was not possible to collect enough of the later species for use in determining a length-weight regression.

2 Unpublished data from L. J. Richards.

3 Harger (1970) 1 22

LITERATURE CITED

Harger, J. R. E. 1970. The effect of wave impact on some aspects of the biology of sea mussels. Veliger 12:401-414. 1 23

APPENDIX B

Values used for variables in multispecies disc equation

Prey Type (i)

Limpet Mussel Chiton Worm crab

Average 8.3 34.9 53.5 6.8 19.1 handling s.d.=4.4 s.d.=32.5 s.d.=35.9 s.d.=3.0 s.d.= time (s) N = 282 N=48 N = 4 N=9 N=49

Prey Density (Dj).and Number Eaten (N; )

in Three Intertidal Zones

(Prey densities are per 50 x 50 cm quadrat)

Limpet Mussel Chiton Worm Crab

Zone N- D- D- N- D- N 9-i Mi P-i —i —i M| — l —i —i fucus 0 .20 20 0,. 1 3 1 a. .00 0 0..0 0 0 0..0 0 0 mytilus 1 .32 202 0..7 2 37 0,.0 3 1 0..0 2 5 0,.0 0 12

lam.post. 1 .42 60 0,.1 0 1 0 3,.3 4 3 0,.0 0 4 0,.0 5 37

Total foraging time, seconds (T )

fucus 853

mytilus 10779

lam.post. 8594