Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A functional analysis of territorial behavior in breeding buffleheads Gauthier, Gilles 1985

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1985_A1 G38.pdf [ 6.16MB ]
Metadata
JSON: 831-1.0096666.json
JSON-LD: 831-1.0096666-ld.json
RDF/XML (Pretty): 831-1.0096666-rdf.xml
RDF/JSON: 831-1.0096666-rdf.json
Turtle: 831-1.0096666-turtle.txt
N-Triples: 831-1.0096666-rdf-ntriples.txt
Original Record: 831-1.0096666-source.json
Full Text
831-1.0096666-fulltext.txt
Citation
831-1.0096666.ris

Full Text

A FUNCTIONAL ANALYSIS OF TERRITORIAL BEHAVIOR IN BREEDING BUFFLEHEADS by GILLES GAUTHIER B.Sc. University of Montreal, Montreal, 1979 M.Sc. University Laval, Quebec, 1982 A'THESIS SUBMITTED IN PARTIAL FULFILLMENT OF 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 August 1985 © G i l l e s Gauthier, 1985 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 2 vO The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date OCT U i i ABSTRACT In this study, I investigate the adaptive significance and consequences of t e r r i t o r i a l behavior in buffleheads (Bucephala  albeola) . The aims aire: (1) to test the hypotheses that t e r r i t o r i a l males defend the food supply, the female or the nest s i t e , (2) to test the hypothesis that brooding females defend the food supply, and (3) to examine whether t e r r i t o r i a l behavior or nest s i t e a v a i l a b i l i t y l i m i t s breeding density. 1) Males defend a t e r r i t o r y from the pre-laying stage u n t i l late incubation. The size of male t e r r i t o r i e s was not related to food abundance, and food was poorly correlated with reproductive success. When the males of seven laying females were removed, four widowed females were evicted from th e i r t e r r i t o r y by neighboring males; widowed females also spent less time feeding and more time a l e r t . Females tended to s e t t l e on a t e r r i t o r y adjacent to their nest, and those that did not do so suffered a higher rate of nest parasitism. I suggest that protection of the female is a major function of male t e r r i t o r i e s and that protection of the nest s i t e is a secondary function. 2) Females become t e r r i t o r i a l after hatching the brood but they defend a di f f e r e n t t e r r i t o r y . The size of the brood t e r r i t o r y was inversely correlated with food density and the relationship was hyperbolic. Growth rate and s u r v i v a l of ducklings was negatively correlated with brood density in one year and survival was p o s i t i v e l y correlated with food density in another year. I suggest that brood t e r r i t o r i e s secure an adequate food supply for the young and that females adjust t e r r i t o r y size according to both food and brood den s i t i e s . 3) Natural nest s i t e s were not in short supply and the addition of nest boxes did not increase breeding density. Breeding density was stable over the four years of t h i s study. When I removed seven t e r r i t o r i a l males, four were replaced. I propose a model to explain the v a r i a b i l i t y of t e r r i t o r i a l behavior in ducks. This model i s based on the hypothesis that mate-guarding is a major function of t e r r i t o r i a l behavior, and predicts that the degree of t e r r i t o r i a l i t y i s correlated with habitat v a r i a b i l i t y . A review of the t e r r i t o r i a l status of 69 species of ducks i s consistent with the model. I conclude that breeding buffleheads exhibit two kinds of t e r r i t o r i a l behavior: males are t e r r i t o r i a l during the nesting season to protect the female and provide her with an undisturbed feeding area ( i . e . mate-guarding), and also to protect the nest s i t e ; females are t e r r i t o r i a l during the brood-rearing stage to secure food resources. T e r r i t o r i a l behavior of nesting pairs further appears to l i m i t breeding density. iv TABLE OF CONTENTS ABSTRACT i i LIST OF TABLES v i i i LIST OF FIGURES x ACKNOWLEDGEMENTS x i i Chapter I: general introduction 1 The study animal 3 Chapter I I : The adaptive significance of pair t e r r i t o r i e s . 4 Introduction 5 Study area . 11 Methods 14 Te r r i t o r y mapping 14 Nest monitoring 15 Invertebrate sampling 18 Removal experiment 19 Results 21 Food hypothesis 21 Ter r i t o r y size and food abundance 21 Reproductive success and food abundance 28 Mate-guarding hypothesis 33 Time spent feeding and a l e r t by widowed females 36 Sexual harassment of widowed females 39 Nest s i t e hypothesis 40 Ter r i t o r y and nest s i t e location 40 Te r r i t o r y location and change in nest s i t e 40 V T e r r i t o r y , nest s i t e location and nest parasitism ... 41 Discussion 44 Food hypothesis 44 Mate-guarding and nest s i t e hypotheses 46 T e r r i t o r i a l i t y and mate-guarding 48 Chapter I I I : The adaptive significance of brood t e r r i t o r i e s 51 Introduction 52 Methods 55 Te r r i t o r y mapping 55 Time-budget 56 Trapping and marking of young 59 Brood counts and computation of duckling survival 60 Invertebrate sampling and c o l l e c t i o n of young 63 Results 65 Food habits 65 Brood density and food abundance 68 Te r r i t o r y size and food abundance 68 Growth rate 75 Brood survival . . . . ; 82 Time budget 87 Discussion 89 Food habits of ducklings 89 T e r r i t o r y size, food abundance and brood density 90 Growth rate, survival and time budget 91 The adaptive value of brood t e r r i t o r i e s 94 Chapter IV: The role of t e r r i t o r i a l behavior and nest s i t e v i a v a i l a b i l i t y in l i m i t i n g breeding density 97 Introduction 98 Methods 100 Sampling of natural c a v i t i e s 100 Addition and monitoring of nest boxes 101 Pair and brood counts 102 Removal experiment .103 Results 104 Use of natural c a v i t i e s by buffleheads 104 A v a i l a b i l i t y of natural c a v i t i e s 107 Use of nest boxes 111 Size of the breeding population 116 Removal experiment 119 Discussion -.121 Nest s i t e a v a i l a b i l i t y 121 Are nest s i t e s limiting? 123 Is t e r r i t o r i a l behavior limiting? 124 Limitation of breeding populations in cavity-nesting • ducks 126 Chapter V: T e r r i t o r i a l behavior in ducks: a review and a model 129 The occurrence of t e r r i t o r i a l behavior in ducks 129 A model for the evolution of t e r r i t o r i a l behavior in ducks 132 Chapter VI: concluding remarks 139 L i t e r a t u r e c i t e d 141 Appendix I 157 v i i References 163 v i i i LIST OF TABLES Table I. Correlation between t e r r i t o r y size and pair density 29 Table I I . Correlation between pair density and food density 30 Table I I I . Food density in the t e r r i t o r y in r e l a t i o n to nesting success 34 Table IV. Outcome of the removal experiment 35 Table V. Relationship between the t e r r i t o r y , nest s i t e location and frequency of nest parasitism 42 Table VI. Correlation between brood density and food density . . . 69 Table VII. Regression of t o t a l amount of food in the t e r r i t o r y on t e r r i t o r y size and food density 73 Table VIII. Correlation and p a r t i a l correlation between t e r r i t o r y size and food density, brood size and brood density 74 Table IX. Correlation between duckling growth rate and food density, t e r r i t o r y size and brood density 80 Table X. P a r t i a l c orrelation between growth rate and food and brood density 81 Table XI. Correlation between duckling survival and food density, t e r r i t o r y size, brood size and brood density .. 86 Table XII. Correlation between time spent in d i f f e r e n t a c t i v i t i e s by ducklings and food density, t e r r i t o r y size and brood density 88 Table XIII. A v a i l a b i l i t y of natural c a v i t i e s 110 Table XIV. Nesting success of females using nest boxes ....112 Table XV. Use of nest boxes according to habitat 113 Table XVI. Density of t e r r i t o r i a l pairs on the pond in re l a t i o n to nesting success 115 Table XVII. Estimates of the number of nests i n i t i a t e d in the study area between 1982 and 1985 120 Table XVIII. Predictions of a model to explain the evolution of t e r r i t o r i a l behavior in ducks 138 X LIST OF FIGURES Figure 1. I l l u s t r a t i o n of the v a r i a b i l i t y in the degree of t e r r i t o r i a l i t y among ducks 6 Figure 2. Location of the study area 12 Figure 3. T e r r i t o r i e s of the four pairs that used pond I Soda from 28 A p r i l to 21 May 1 984 16 Figure 4. Diet of adult buffleheads 22 Figure 5. Relationship between t e r r i t o r y size of pairs and food density 24 Figure 6. Relationship between start of egg-laying and food density 31 Figure 7. Behavior of widowed females . . 37 Figure 8. T e r r i t o r i e s of the three broods that used pond E Soda from 13 June to 14 July 1984 57 Figure 9. Diet of bufflehead ducklings 66 Figure 10. Relationship between t e r r i t o r y size of broods and food density in the t e r r i t o r y 70 Figure 11. Relationship between duckling weight and age ... 76 Figure 12. Daily survival rate of ducklings 83 Figure 13. D i s t r i b u t i o n of cavity measurements for the sampled c a v i t i e s and for those used by buffleheads 105 Figure 14. D i s t r i b u t i o n of distances to water for the sampled c a v i t i e s and for those used by buffleheads 108 Figure 15. Total number of bufflehead breeding pairs using the study area from 1982 to 1985 117 Figure 16. A graphical model to explain the evolution of t e r r i t o r i a l behavior in ducks 1 x i i ACKNOWLEDGMENTS Many people helped me to complete th i s t h e s i s . I thank my supervisor, Jamie Smith, who provided encouragement and continuous help throughout t h i s study. I also thank my f i e l d assistants, Danielle Gauthier, Barbara Peterson', Simon Richards, and Linnie Nyland, and my lab assistant, Joyce Thomson. Jean-Pierre Savard introduced me to the bufflehead and also provided invaluable assistance. I thank the members of my supervisory committee, Lee Gass, Kim Cheng, Robin L i l e y and Geoffrey Scudder; The Institute of Animal Resource Ecology D i r t Lunch discussion group; and especially John Eadie, Jean-Pierre Savard and Peter Arcese for their ideas and c r i t i c a l discussions. Dave Z i t t i n , Susan E r t i s and A l i s t a i r Blanchford provided programming help. Rory Brown and Ron Boychuck of Ducks Unlimited provided some l o g i s t i c support and generously shared unpublished data on waterfowl surveys. J. Smith, L. Gass, K. Cheng, R. L i l e y , J. Eadie, J.-P. Savard, P. Arcese, D. Schluter and T. Erskine read part or a l l of this thesis. Above a l l , I thank my wife, Danielle, for her support throughout this work. I am grateful to W. Monical for allowing me to work on his property. This work was generously funded by the Canadian National Sportsmen's Fund, a Natural Sciences and Engineering Research Council of Canada (NSERC) grant to J. Smith, and the Canadian W i l d l i f e Service. I thank the NSERC and FCAC program of the ministere de 1'education du Quebec for their scholarships. 1 CHAPTER I: GENERAL INTRODUCTION T e r r i t o r i a l behavior i s a very widespread type of s o c i a l behavior in animals. Since i t was f i r s t described in birds by Howard (1920), t e r r i t o r i a l behavior has been documented in most groups of animals, including gastropods (Stimson 1973), insects ( F i t z p a t r i c k and Wellington 1983), spiders (Riechert 1981), fishes (van den Assem 1967), salamanders (Jaeger et a l . 1982), l i z a r d s (Rand 1967) and mammals (Armitage 1974). T e r r i t o r i a l behavior can be extremely diverse, and numerous reviews have described and c l a s s i f i e d d i f f e r e n t types of t e r r i t o r i e s (Nice 1941, Hinde 1956). I s h a l l follow Noble's (1939) and Emlen's (1957) d e f i n i t i o n of a t e r r i t o r y as a defended area where the owner is usually dominant over conspecific intruders. Since Brown (1964) introduced the concept of economic defendability of t e r r i t o r i e s , there has been considerable discussion of the adaptive significance of t e r r i t o r i a l behavior. In trying to understand the adaptive significance or functions (sensu Mayr 1983) of a t e r r i t o r y , one seeks to answer the question "why do animals defend a t e r r i t o r y ? " One approach to th i s question i s to relate t e r r i t o r y size or the occurrence of t e r r i t o r i a l i t y to ecological features on one hand, and to components of fit n e s s such as survival and reproductive success on the other hand. The need here i s to show that defense of resources (e.g. food, nest s i t e ) increases the reproductive success or survival of t e r r i t o r i a l i n d i v iduals. Numerous hypotheses have been proposed to explain the 2 adaptive significance of nesting t e r r i t o r i e s in birds (Wynne-Edwards 1962, Holmes 1970, Krebs 1971, MacLaren 1972, Verner 1977). In his review, Wilson (1975) considered that securing an adequate food supply for the parents and nestlings was the major function of nesting t e r r i t o r i e s . This view, however, has been challenged (e.g. Verner 1977, Weatherhead and Robertson 1980). The d i v e r s i t y in t e r r i t o r i a l behavior suggests that factors underlying t e r r i t o r i a l i t y are probably not the same in a l l species (Verner 1977). The family Anatidae is a good example of the v a r i a b i l i t y in t e r r i t o r i a l behavior (see chapter 2). Diversity among species i s in fact so great that workers have long argued whether ducks are t e r r i t o r i a l at a l l during the breeding season (Hochbaum 1944, Sowls 1955, Dzubin 1955). The purpose of t h i s thesis is to investigate the adaptive significance of t e r r i t o r i a l behavior, and i t s consequences in a species of duck, the bufflehead (Bucephala albeola). In chapter 2, I consider the male t e r r i t o r i a l i t y during the nesting stage and I test three hypotheses (food, mate-guarding and nest site) to account for i t s occurrence. In the t h i r d chapter, I show that brood t e r r i t o r i e s are d i f f e r e n t from male t e r r i t o r i e s and I test the hypothesis that brood t e r r i t o r i e s are defended for their food value to the young._ The fourth chapter evaluates whether t e r r i t o r i a l behavior or nest s i t e a v a i l a b i l i t y l i m i t s populations of breeding buffleheads. In chapter 5, I present a model which attempts to explain the v a r i a b i l i t y of t e r r i t o r i a l behavior in ducks. F i n a l l y , chapter 6 is a brief general discussion. 3 The study animal The bufflehead is the smallest species of North American diving ducks. It is also the most sexually dimorphic species of ducks (Livezey and Humphrey 1984). The l i f e history of buffleheads has been described in d e t a i l by Erskine (1972). They pair on the wintering ground or early during the spring migration. Sex r a t i o is biased toward males and there i s always a surplus of adult males on the breeding grounds. Pairs arrive on the breeding grounds in mid- or late A p r i l , and females start nesting shortly (7-10 days) afte r their a r r i v a l . The egg-laying period is r e l a t i v e l y long (10-14 days) because females often lay every other day. Females incubate for about 30 days, and i t is during this period that pair bonds break up and males leave for molting grounds. Broods are attended only by females u n t i l about a week or two before the young can f l y (50-55 days). The spacing behavior of breeding buffleheads has been extensively described by Donaghey (1975). Males are t e r r i t o r i a l from the pre-laying stage u n t i l the late incubation period. After hatching females become t e r r i t o r i a l , although they often select a t e r r i t o r y d i f f e r e n t from the one defended by t h e i r mate e a r l i e r in the season. Defense of two d i f f e r e n t t e r r i t o r i e s within the same season is unusual in birds. Donaghey (1975), however, suggested that the adaptive significance of both types of t e r r i t o r i e s was the same, namely to secure food resources. Donaghey's work was the s t a r t i n g point for t h i s thesis. CHAPTER II : THE ADAPTIVE SIGNIFICANCE OF PAIR TERRITORIES 5 INTRODUCTION The application of the concept of t e r r i t o r i a l i t y to the family Anatidae has been very confused (Hochbaum 1944, Sowls 1955, Dzubin 1955, McKinney 1965). Much of the confusion originated from early workers trying to apply a r i g i d d e f i n i t i o n of t e r r i t o r i a l i t y to ducks (e.g. Brown and Orians's 1970 d e f i n i t i o n of an exclusive and defended area fixed in time and space). It is now evident that d i f f e r e n t species of ducks vary greatly in their degree o f . t e r r i t o r i a l i t y (McKinney 1973, Titman and Seymour 1981). The following c l a s s i f i c a t i o n i l l u s t r a t e s the v a r i a b i l i t y in t e r r i t o r i a l i t y among ducks (Fig. 1). Type 1 species show l i t t l e or no t e r r i t o r i a l i t y ; males are not very, aggressive, there is l i t t l e or no defended area, and home , range are'large and overlapping (e.g. northern p i n t a i l , Anas acuta, McKinney 1973, Derrickson 1978; green-winged t e a l , A. crecca, McKinney and Stolen 1982). Males of type 2 species defend a "moving t e r r i t o r y " around the female (sensu Dzubin 1955) within their home ranges; males are only moderately aggressive, the defended area s h i f t s through time, and i t i s often not exclusive (e.g. mallard, A. platyrhynchos, Titman 1983; gadwall, A. strepera, Dwyer 1974). Type 3 species also defend a moving t e r r i t o r y around the female; however, males are much more aggressive, the defended area i s more s i t e - s p e c i f i c , exclusive and sometimes has well-defined boundaries, but pairs often feed away from the t e r r i t o r y (e.g. northern shoveler, A. clypeata, Poston 1974, Seymour 1 974a,b; blue-winged t e a l , A. discors, 6 Figure 1. I l l u s t r a t i o n of the v a r i a b i l i t y in the degree of t e r r i t o r i a l i t y among ducks (see introduction for d e t a i l s ) . Home range / \ / I \ Defended area ^ i • Type 1 No defended area I I n Type 2 Loosely defended "moving territory" Defended area Type 3 Strongly defended "moving territory" 0 Type 4 Highly site-specific territory 8 S t e w a r t and Titman 1980; American b l a c k duck, A. r u b r i p e s , Seymour and Titman 1978). F i n a l l y , type 4 s p e c i e s e x h i b i t a v e r y s t r o n g t e r r i t o r y : males a r e e x t r e m e l y a g g r e s s i v e and defend a s m a l l , w e l l - d e f i n e d a r e a where the p a i r r e s t r i c t s a l l i t s a c t i v i t i e s , and from where a l l c o n s p e c i f i c a r e e x c l u d e d ( e . g . b u f f l e h e a d , Donaghey 1975; common and Barrow's goldeneye, Bucephala c l a n g u l a and B. i s l a n d i c a , Savard 1982a, 1984; common she l d u c k Tadorna t a d o r n a , Young 1970, P a t t e r s o n 1982). Type 4 t e r r i t o r i a l i t y has been r e p o r t e d i n o n l y a s m a l l number of ducks and i s as y e t p o o r l y u n d e r s t o o d . In t h i s s t u d y , I t e s t t h r e e hypotheses t o e x p l a i n type 4 t e r r i t o r i a l i t y i n b u f f l e h e a d s : (1) the food h y p o t h e s i s , (2) the mate-guarding h y p o t h e s i s , and (3) the p r o t e c t i o n of the nest s i t e h y p o t h e s i s . Donaghey (1975) h y p o t h e s i z e d t h a t c o m p e t i t i o n f o r food among females was a major f a c t o r f a v o r i n g t e r r i t o r i a l b e h a v i o r i n b u f f l e h e a d s , and t h a t the t e r r i t o r y s e c u r e s a food s u p p l y f o r the female. An i n v e r s e r e l a t i o n s h i p between food abundance and t e r r i t o r y s i z e has c l a s s i c a l l y been taken as e v i d e n c e f o r a food-mediated t e r r i t o r y ( e . g . Holmes 1970, Stimson 1973, Slaney and N o r t h c o t e 1974, Simon 1975, Gass e_t a l . 1976, Kodric-Brown and Brown 1978, S e a s t e d t and MacLean 1979, F r a n z b l a u and C o l l i n s 1980). More r e c e n t l y , workers have s t u d i e d f e e d i n g t e r r i t o r i e s u s i n g c o s t - b e n e f i t models where t e r r i t o r y s i z e i s viewed as a compromise between the b e n e f i t s ( f o o d abundance i n the t e r r i t o r y ) and defense c o s t s ( i n t r u d e r p r e s s u r e ) ( e . g . Myers et a l . 1981). However, even the most r e c e n t models of f e e d i n g t e r r i t o r i e s a l s o p r e d i c t t h a t t e r r i t o r y s i z e w i l l d e c r e a s e as 9 food abundance increases (Schoener 1983, Lima 1984). An underlying assumption of the food hypothesis is that food is l i m i t i n g , because t e r r i t o r i a l i t y i s a form of interference competition (Brown 1964). We should thus expect a positive relationship between food abundance and reproductive success (MacLean and Seastedt 1979). Two predictions of the food hypothesis are therefore, (1) t e r r i t o r y size should be inversely related to food abundance in the t e r r i t o r y , and (2) reproductive success should be p o s i t i v e l y related to food abundance. In species exhibiting type 2 or 3 t e r r i t o r i a l i t y , several authors have suggested that the female i s the resource defended by the male (mate-guarding hypothesis, Seymour and Titman 1978, Stewart and Titman 1980, Patterson 1982, Titman 1983). The mate-guarding hypothesis includes two elements: f i r s t , male vigilance benefits the female because he protects her from conspecifics and predators, and thus provides her with undisturbed feeding time (Titman 1983, McKinney et a_l. 1983). This is especially important during the pre-laying and laying periods when the energy requirements of females are maximal (Krapu 1979, Owen and Reinecke 1979). Second, this i s the period when females are f e r t i l e . Forced copulations of paired females by strange males are common in several species of ducks (review by McKinney e_t a l . 1983). Paired males should thus attempt to prevent cuckoldry and insure their paternity through strong mate-guarding (Cheng et a l . 1983, McKinney et a l . 1983, Titman 1983). Two predictions of the mate-guarding hypothesis are therefore that, i f the male of a laying female is removed, (1) she should spend less time 10 feeding and more time a l e r t and scanning for disturbance (e.g. conspecifics or predators) and (2) she should be sexually harassed by neighboring males. Buffleheads nest in c a v i t i e s near water (Erskine 1972), and the s i t e - s p e c i f i c i t y of their t e r r i t o r y could be explained by the need to protect the nest s i t e . Protection of the nest could be achieved by excluding conspecifics from the water adjacent to the.nest s i t e . This hypothesis rests on the assumption that there is a cost to nest interference or to being parasitized by conspecifics (e.g. Jones and Leopold 1967, Andersson and Eriksson 1982, Hines and M i t c h e l l 1984). Three predictions of the nest s i t e hypothesis are that, (1) bufflehead t e r r i t o r i e s should be adjacent to their nest, (2) i f a female changes nest s i t e , the t e r r i t o r y should also move, and (3) the rate of' nest parasitism should be higher in nests farther from water or not adjacent to the t e r r i t o r y . 11 STUDY AREA The study was conducted from 1982 to 1985 in the Cariboo Parkland of B r i t i s h Columbia, 15 km north of 100 Mile House (Fig. 2). The study area covered 23 km2 and included 26 ponds and lakes. A l l ponds were permanent, and ranged in size from 1.5 ha to 61 ha, although 80 % of them were less than 8 ha. The habitat in t h i s area is a mosaic of rangeland, groves of aspen (Populus tremuloides), and montane forest dominated by Douglas f i r (Pseudotsuga menziesi i) and lodgepole pine (Pinus contorta). About two t h i r d of the ponds were located in the parkland habitat. Most ponds were shallow (less than 3 m deep), a l k a l i n e (pH 8-10), and ranged in s a l i n i t y from freshwater to moderately saline (conductivity 300-13000 umhos/cm), which i s t y p i c a l of the area (Topping and Scudder 1977). Ten ponds were created by Ducks Unlimited (Canada) in the early 1970's, and 6 others had control devices regulating their water l e v e l . The ponds were t y p i c a l l y open water surrounded by a fringe of emergent vegetation, mostly hard-stemmed bulrush (Scirpus acutus) and c a t t a i l (Typha l a t i f o l i a ) . 1 2 Figure 2. Location of the study area and of the ponds in the Cariboo Parkland (hatched area) of B r i t i s h Columbia. 14 METHODS Territory mapping Bufflehead t e r r i t o r i e s were mapped from the f i r s t week of May u n t i l the f i r s t week of June on seven ponds in 1983 and eight ponds in 1984. Observations were done from a high, distant vantage point using binoculars and a spotting scope. T e r r i t o r i e s were mapped once a week on each pond during a 3-h period (09:00 to 12:00 h PDT). Each ten to twenty minutes (depending on the density of b i r d s ) , the pond was scanned and the position and behavior of a l l buffleheads were plotted on large scale maps (1:1000). A l l aggressive interactions occurring during the 3-hour period were recorded, the birds involved were i d e n t i f i e d (see below), and their location plotted. Females were trapped on the nest (after mid-incubation to minimize desertion) and marked with color-coded nasal saddles (Doty and Greenwood 1974). Females are highly p h i l o p a t r i c to their nesting pond (Erskine 1961), and by 1984 at least 60 % of the nesting females were marked. Males are hard to catch and only three were trapped in 1983 using a mirror trap (Savard in press). A l l ponds observed had both unmarked pairs and pairs with at least one marked bird (usually the female). T e r r i t o r i a l pairs are very f a i t h f u l to their t e r r i t o r y (as confirmed by observations of marked pairs) and unmarked pairs could be i d e n t i f i e d with reference to the area they occupied. Areas of t e r r i t o r i e s were calculated from observation maps using the minimum area polygon method. Te r r i t o r y size of each 15 male was defined as being the point where the observation-area curve reached an asymptote (Odum and Kuenzler 1955). T e r r i t o r i e s determined by t h i s method always included at least 90 % of a l l locations. Each pond was observed for 10 to 13 hours and t e r r i t o r i e s were calculated only for pairs with at least 20 ( t y p i c a l l y 30-50) locations. F i g . 3 i l l u s t r a t e s the results obtained with this method on pond I Soda in 1984. Donaghey (1975) used aggressive interactions to map boundaries of bufflehead t e r r i t o r i e s . This method could not be used here because, on several ponds, too few aggressive interactions were recorded to map t e r r i t o r i e s accurately. Weekly censuses of buffleheads were conducted four times in May of 1983 and 1984 on a l l ponds. Goldeneyes were also surveyed because they are i n t e r s p e c i f i c a l l y t e r r i t o r i a l with buffleheads (Savard 1982a, 1984). The t o t a l number of pairs plus lone males and lone females was taken as an index of breeding pair density for each species, as suggested by Savard (1981). Censuses of the whole study area were- completed within two days, and were conducted between 08:00 and 12:00 h (PDT). Nest monitoring Bufflehead nests were found by intensively searching for natural c a v i t i e s around ponds,, or by following females f l y i n g to their nests during incubation. In 1982 and 1983, 180 nest boxes were erected on the study area (4-20 boxes per pond). These boxes were used by buffleheads, especially in 1984 when over a t h i r d of the population nested in boxes. During egg-laying, 16 Figure 3. T e r r i t o r i e s of the four pairs that used pond I Soda from 28 A p r i l to 21 May 1984. Each number i s a d i f f e r e n t location, and the lin e joining the outermost points i s the t o t a l area used by that pair during this period. The innermost l i n e encloses the intensively used area defined here as the t e r r i t o r y . The nest s i t e of pair no. 2 was not found. They disappeared the following week, and I presume that they did not breed in the study area. 17 18 nests were checked every 2-5 days. A l l eggs were measured (± 0.1 mm) and weighed with a spring scale (± 0.25 g). For nests found during incubation, the onset of incubation and of egg-laying were estimated by backdating from hatching (average incubation length: 30 days; -rate of egg-laying: one egg per day and a half, Erskine 1972), and egg weight was estimated by assuming a constant d a i l y weight loss of 0.5 % (Erskine 1972, Rahn and Ar 1974). Nests were checked every 7-10 days during incubat ion. Invertebrate sampling Buffleheads feed mainly on nektonic invertebrates (Erskine 1972). Invertebrates were sampled using a m o d i f i e d Gerkin device (Kaminski and Murkin 1-98J.) , with a trap opening of 25 x . 50 cm and a mesh of 0.5 mm. The trap was dropped v e r t i c a l l y from a rowboat, and i t could reach up to a depth of 1.5 m. The trap sampled invertebrates in the water column, from the surface down to the sediment, but not invertebrates in the sediments. The trap also sampled the invertebrate fauna of sparsely to moderately dense beds of submerged vegetation by cutting aquatic macrophytes. Organisms were transferred to p l a s t i c bags in the f i e l d and frozen for later analysis in the laboratory. I sampled a l l ponds where t e r r i t o r i e s were mapped within a 4-day period in the t h i r d week of May. A systematic sampling grid was l a i d out for each pond to cover a l l water areas used by bufflehead p a i r s . From 6 to 12 samples were c o l l e c t e d in each t e r r i t o r y , and the average of these samples gave an index of 19 food abundance for the t e r r i t o r y . About 300 samples were co l l e c t e d each year. In the lab, samples were sorted to the order or family l e v e l . Organisms within each taxa were separated by size according to t h e i r body length (either 5 or 10 mm) and counted. A sample of organisms (N > 20) was dried and weighed for each species and size c l a s s . Numbers were then transformed into biomass. Cladocerans were not counted: after a l l other organisms were sorted, the remaining cladocerans were rinsed, dried, and weighed. Results are presented in g dry weight/m2 (overall r e s u l t s were similar when expressed as either g/m2 or g/m3). Removal experiment Removal experiments were conducted in May 1984. Seven males were removed when the i r mate was about half-way through egg-laying. A l l widowed females had been banded on a nest in previous years and were at least 3 years of age. Males were shot between 06:00 and 07:00 h (PDT) and they were quickly removed from the pond. Behavior of the female was recorded on the day before removal (two days before removal for two females), again on the same morning that removal occurred, and f i n a l l y two days l a t e r . Behavioral observations were always made from 08:00-12:00 h. The time from removal to the start of observations (> 1 h) was long enough for the female to resume her normal a c t i v i t y . The 4-h period of observations was divided into 12 blocks of 15 min, each separated by a 5-min break, for a t o t a l of 3 h of observation. Females were kept under constant 20 observation within the 15-min periods, and their behavior was recorded at 5-sec intervals (instantaneous sampling, Altmann 1974). Seven behaviors were recorded: Feeding: mostly diving but sometimes (< 5 %) also picking at the surface Resting: b i l l i s tucked under the wing; eyes closed Preening: includes preening and bathing Swimming: a l l locomotor a c t i v i t y except f l y i n g A l e r t : includes both "extreme" a l e r t when the neck i s stretched (usually a response to a close disturbance), and "normal" a l e r t when the head is low but the bird i s attentive to i t s surroundings ( i . e . eyes open) Social interactions: a l l interactions with i t s mate (e.g. courtship display) or with other birds (e.g. aggression). Fly i n g . A l l aggressive interactions occurring at any time within the 4-h period were also recorded, and analyzed separately. Time-budgets were calculated as follows: for each morning of observation, the percentage of time spent in d i f f e r e n t a c t i v i t i e s was calculated within each 15-min period, and then averaged over the 12 periods giving equal weight to each period. A l l s t a t i s t i c a l analyses were performed using the SPSSx package. S t a t i s t i c a l tests were one-tailed (unless otherwise specified) because they dealt with a p r i o r i predictions. 21 RESULTS Food hypothesi s Adult buffleheads feed mostly on aquatic insects in spring and summer (Fig. 4). Hirudinea and Cladocera are never eaten by adults and they were excluded from the analysis. Amphipods were dominant in the invertebrate samples but their status as food for buffleheads i s unclear. Although they appear to be rarely eaten by buffleheads, amphipods are digested very rapidly (Swanson and Bartonek 1970). Therefore, they are probably eaten more than suggested by gizzard content analyses (see also Erskine 1972). If amphipods, ^cladocerans and Hirudinea are excluded from the invertebrate sampling, the remaining prey items are eaten by buffleheads roughly in proportion to their abundance (Fig. 4). However, because of the uncertainty over the status of amphipods as food, a l l comparisons of food abundance were done both with and without them. Predict ion J _ : T e r r i t o r y size w i l l be inversely related to food  abundance in the t e r r i t o r y In 1983 there was a barely s i g n i f i c a n t inverse relationship between food density and t e r r i t o r y size (Fig. 5a). In 1984, with a larger sample s i z e , no such r e l a t i o n s h i p was found (Fig. 5b), nor was there a s i g n i f i c a n t relationship for both years combined (r = -0.13, P = 0.19, N = 46). If amphipods were removed from the analysis, there was no r e l a t i o n s h i p between food and 22 Figure 4.- Diet of adult buffleheads in spring and summer (from Erskine 1972) and the proportion of d i f f e r e n t invertebrate taxa found in the sampling of bufflehead t e r r i t o r i e s (L = larvae, P = pupae). 23 E O > 20-10-Diet 60-20 10 -D) •35 T3 30-20 10 Invertebrate sampling Invertebrate sampling (without H P ) o JS O E CD C o "O o X •c o u E a CD o .c o o .t o J3 O CO o o CD T3 .is CO -C 24 Figure 5. Relationship between t e r r i t o r y size of pairs and food density (dry weight) in the t e r r i t o r y . (A) 1983: r_ = -0.43, P = 0.049. (B) 1984: r = 0.04, P > 0.1. 25 ( B L | ) eZI.S AJ01UJ91 ezis A J O I U J 9 J _ 27 t e r r i t o r y size in either year (1983, r = -0.27, P = 0.15; 1984, r = 0.21, P = 0.13). Invertebrate abundance tended to be as variable among ponds as within ponds. Hence, using the pond as the sampling unit rather than the t e r r i t o r y could result in a closer f i t between food abundance and t e r r i t o r y s i z e . However, the average t e r r i t o r y size per pond was not correlated with average food density when years were taken separately or together (P > 0.1; 1983, N = 5; 1984, N = 8). Food density in the t e r r i t o r y was, on average, nearly three times as abundant in 1984 (1.35 ± 0.31(SE) g/m2, N = 30) as in 1983 (0.46 ± 0.14 g/m2, N = 16; exclusion of amphipods did not change t h i s r e s u l t ) , yet the average t e r r i t o r y size (0.38 ± 0.03 ha vs 0.37 ± 0.04 ha) was i d e n t i c a l in both years. Models of feeding t e r r i t o r i e s predict that t e r r i t o r y size w i l l be influenced by intruder pressure (density) as well as by food abundance (Myers et_ a_l. 1981, Schoener 1983, Lima 1984). To test for t h i s , I examined the relationship between t e r r i t o r y size and pair density. Because bufflehead pairs are spaced out along the shoreline and the central area of the pond i s often unoccupied (e.g. F i g . 3), I used two indices of density: number of pairs per ha of water and per km of shoreline. Density i s viewed as a treatment, with d i f f e r e n t ponds having d i f f e r e n t densities of p a i r s . Thus, the pond i s here the sampling unit, because a l l pairs on a given pond experienced the same density (Hurlbert 1984). In each year, t e r r i t o r y size tended to be negatively correlated with density (Table I ) , and most 28 correlations were s i g n i f i c a n t when both years were combined. However, there was no relationship between pair density and average food density per pond (Table I I ) . Buffleheads are i n t e r s p e c i f i c a l l y t e r r i t o r i a l with goldeneyes (Savard 1982a, 1984). Goldeneyes were present on the study area although their density was about 10 times below that of buffleheads. It i s thus interesting that when goldeneye pairs were used in cal c u l a t i n g density, the f i t between density and t e r r i t o r y size was improved (Table I ) . Predict ion 2\ Some measures of reproduct ive success w i l l be  po s i t i v e l y related to food abundance Clutch size, mean- egg weight or clutch weight were not correlated with food density in the t e r r i t o r y in either year (P > 0.1; 1983, N = 9; 1984, N = 19), except for clutch weight in 1983 (r = 0.61, P = 0.04). There was, however, a s i g n i f i c a n t inverse relationship between start of egg-laying and food density in both years (Fig. 6), although the correlation disappeared i f amphipods were excluded from the analysis (P > 0.2). I also examined for a relationship between food den.sity and nesting success. I divided nests into successful nests (those having at least one young leaving the nest) and deserted ones (desertion accounts for over 80 % of nest f a i l u r e ) . In 1983, desertion rate tended to be higher on t e r r i t o r i e s with low food density (only i f amphipods were excluded), although the difference was not s i g n i f i c a n t (Table I I I ) . In 1984, contrary to my prediction, desertion tended to be higher on t e r r i t o r i e s with 29 Table I. Correlation c o e f f i c i e n t s of t e r r i t o r y size with pair density. Pair density i s expressed in buffleheads and buffleheads + goldeneyes per ha of water area and per km of shoreline (N = number of ponds). Year N buffleheads buffleheads + goldeneyes ha km ha km 1983 7 -0.51 -0.52 -0.55 -0.57 1984 8 -0;29 -0.64 * -0.49 -0.81 ** 1983+1984 15 -0.41 -0.58 ** -0.52 * • -0.69 ** P < 0.05, ** P < 0.01, otherwise P > 0.05 30 Table I I . Correlation c o e f f i c i e n t s of pair density with average food density per pond (including amphipods). Pair density i s expressed in buffleheads and buffleheads + goldeneyes per ha of water area and per km of shoreline. Results were similar i f amphipods were excluded (N = number of ponds). buffleheads buffleheads + goldeneyes Year N ha km ha km 1983 7 0.19 0.18 0.05 0.09 1984 8 0.37 0.13 0.20 -0.13 1983+1984 15 0.24 0.07 0.1.4 -0.06 A l l c o r r e l a t i o n s , P > 0.1 31 Figure 6 . Relationship between start of egg-laying and food density (dry weight) in the t e r r i t o r y . 1983: r = -0.57, P = 0.04, y = -5.501og(x) + 128 1984: r = -0.47, P = 0.02, y = -8.761og(x) + 139 32 33 more food (Table I I I ) , although sample sizes were small. I also looked at the relationship between time spent feeding and food density. Using time budgets of females before the removal experiment, I found that time spent feeding was not s i g n i f i c a n t l y correlated with food density (r - -0.01, P = 0.46, N = 6; when amphipods were excluded, r_ = -0.60, P = 0.10). To summarize the food hypothesis, in one year there was a weak inverse relationship between food and t e r r i t o r y size but this was found only i f amphipods were included. Food density was not correlated with any measure of reproductive success except for the start of egg-laying and clutch weight in one year, again only i f amphipods were included. Mate-guarding hypothesis Male removals were conducted when experimental females were about one half to two thirds through the egg-laying stage. Pair bonds and mate-guarding by males should be strongest at that time since females are f e r t i l e . A l l widowed females l a i d more eggs after removal of their mate, and a l l but one female successfuly raised their eggs to hatching (Table IV). Four out of seven widowed females were attacked shortly after the removal of their mate, and they were evicted from their t e r r i t o r y by neighboring males who took over their t e r r i t o r y . In one case (no. 581) the t e r r i t o r y of the widowed female was taken over by a new pair. Four females remated with a new male within 48 hours and a f i f t h one after 48 hours (Table IV). Two of the l a t t e r f i v e females were among those evicted from their t e r r i t o r i e s , 34 Table I I I . Food density (g dry weight/m2) in the t e r r i t o r y in rela t i o n to nesting success. Comparisons (t test) are between females that nested successfully and those that deserted their nest (mean ± SE) Food density Year Nest N With amphipods Without amphipods Successful 12 1.26±0.38 0.33±0.08 1983 P = 0.09 P = 0.49 Deserted 9 0.61±0.20 0.34±0.05 Successful 9 0.46±0.19 0.10±0.02 1984 • P = 0.12 P = 0.035 Deserted 2 1.06±0.51 0.20±0.04 Table I V . N e s t i n g s t a g e of widowed females and outcome o f the 1984 removal e x p e r i m e n t . A l l n e s t s were s u c c e s s f u l e xcept f o r female 1870 whi c h d e s e r t e d h e r n e s t two weeks l a t e r Female ^ . T , . . a Number o f F i n a l T e r r i t o r y Replacement Date I n c u b a t i o n ^ . J V number eggs l a i d c l u t c h s i z e t e n u r e male 1470 4 May -4 6 9 r e t a i n e d Yes 1670 6 May -4 5 7 r e t a i n e d Yes° 1570 7 May -10 6 12 l o s t Yes 581 8 May (-4)? ? 10 l o s t No 1680 12 May -6 7 9 l o s t d Yes 1870 18 May -2 4 6 l o s t No 1840 21 May -4 5 7 r e t a i n e d Yes° a Number o f days b e f o r e the s t a r t o f i n c u b a t i o n a t the time o f the removal b A t the time o f the removal c The replacement male o f these females was a n e i g h b o u r i n g male which became poly g y n o u s d The replacement male o f these females was a b l e t o r e g a i n t h e i r t e r r i t o r y w i t h i n 2 day 36 but their new mate was able to regain the t e r r i t o r y after several bouts of f i g h t i n g . The last two females apparently did not remate and they lost their t e r r i t o r y permanently. One of these two females deserted her nest during incubation. F i n a l l y , two of the replacement males were polygynous males (Gauthier in press). The other three replacement males were presumably unmated before. Predict ion j_: widowed females w i l l spend less t ime feeding and  more time a l e r t The behavior of six of the seven widowed females was recorded immediately after the removal, and four of them were observed again two days l a t e r . Time spent feeding by females decreased from 30.4 % to 22.3 % after the removal (P = 0.12, paired t test, F i g . 7a), but went back up to 31.3 % after they remated (P = 0.15, after removal v_s after remating). This included a female which was out of sight more than 50 % of the time (other females were out of sight 5 % of the time on average). Because feeding birds were more l i k e l y to become out of sight than those engaged in other a c t i v i t i e s , the feeding time of that female was probably underestimated. If she was removed from the analysis, the f i r s t test (before vs after removal) was s i g n i f i c a n t (P = 0.048). Time spent in ale r t increased from 24.1 % to 40.7 % after the removal (P = 0.048, paired t test, F i g . 7a), and went down to 29.6 % after females remated (P = 0.21 ) . 37 Figure 7. Behavior of female buffleheads before removal of their mate, after the removal, and afte r they remated. (A) Proportion of time spent feeding and a l e r t . (B) Frequency of aggressive interactions. (Mean and SE). 38 39 Predict ion 2: widowed females w i l l be sexually harassed by other  males The frequency of attacks on females by neighboring males increased from 0.14 per 4 h of observation to 2.17 after the removal, and went down to 0 after they remated (Fig. 7b). This increase was s i g n i f i c a n t (P < 0.01, Kruskal-Wallis analysis of variance) when polygynous males were excluded (as they quickly courted widowed females, they obviously did not react aggressively toward them). A l l interactions involving widowed females (excluding those with polygynous males) were agonistic: neighboring males always attacked females and t r i e d to evict them from their t e r r i t o r y through short a e r i a l pursuit. These males never attempted ..any forced copulations (Gauthier in press). In summary, widowed females were generally evicted from their t e r r i t o r y , they spent less time feeding, more time a l e r t , and they were harassed ( a l b e i t not sexually) by neighboring males. Thus, both predictions of the mate-guarding hypothesis were supported. 40 Nest s i t e hypothesis Prediction ]_: t e r r i t o r i e s w i l l be adjacent to the nest s i t e To define "adjacent", I used the shortest distance between a nest and any bufflehead t e r r i t o r y . The t e r r i t o r y was deemed to be adjacent i f the closest t e r r i t o r i a l pair to a nest was the owner of that nest (e.g. pairs 1 and 4 on F i g . 3 have t e r r i t o r i e s adjacent to their nest whereas pair 3 does not). This c l a s s i f i c a t i o n was easy to make because bufflehead nests are usually close to the water (75 % of the nests were within 50 m of the shoreline; see also Erskine 1972). Territory location was known for only 54 nesting females (out of 75 nests found): 72 % (39) were adjacent to the nests. In eleven cases pairs were separated from their nest by at least one other bufflehead or goldeneye (one case) t e r r i t o r y , and four pairs had their nest at the boundary of their t e r r i t o r y and that of another pair ( i . e . equidistant to 2 t e r r i t o r i e s ) . Thus, most bufflehead pairs defended t e r r i t o r i e s adjacent to their nests. Prediction 2: I_f a female changes nest s i t e , the t e r r i t o r y w i l l  also move Twelve females had known t e r r i t o r y locations in two consecutive years. Eight of them had a t e r r i t o r y adjacent to their nest and reused the same t e r r i t o r y and nest s i t e . Two females with t e r r i t o r i e s not adjacent to their nest moved to one adjacent to their nest the following year. F i n a l l y , two females 41 moved to a new nest s i t e and their t e r r i t o r y , which was adjacent to their nest in the f i r s t year, was also adjacent to their new nest s i t e the following year. Predict ion 3_: Nests that are not adjacent to the t e r r i t o r y w i l l  suffer a higher rate of nest parasitism A nest was judged to be para s i t i z e d i f more than one egg per day was l a i d and/or i f 13 or more eggs were l a i d with l i t t l e or no down present (Erskine 1972). Parasitism i s here used in a broad sense and refers to "i n t e n t i o n a l " parasitism as well as two females inadvertently using the same nest s i t e . Rates of nest parasitism are low in bufflehead: only 5.3 % (N 75) of a l l nests found in this study were p a r a s i t i z e d (see also Erskine 1972). Nests were never p a r a s i t i z e d when the pair t e r r i t o r y was adjacent to the nest s i t e (Table V; P = 0.02, Fisher exact prob. t e s t ) ; of three pa r a s i t i z e d nests, two were at the edge of two t e r r i t o r i e s and one had a goldeneye pair adjacent to i t . Furthermore, parasitized nests were deserted more often than normal nests (Table V, P = 0.01). However, nests that were not adjacent to the t e r r i t o r y (irrespective of whether they were parasitized or not) were deserted at the same rate as those adjacent (Table V, P = 0.33). Therefore, buffleheads usually defend t e r r i t o r i e s adjacent to their nest, and t e r r i t o r i e s of females that changed nest s i t e also moved to become adjacent to the new nest. Pairs that did not defend t e r r i t o r i e s adjacent to their nest tended to suffer a higher l e v e l of nest parasitism. A l l these r e s u l t s support the 42 Table V. Frequency of (a) nesting success in rel a t i o n to nest parasitism, (b) nest parasitism in relation to the sp a t i a l relationship of the nest to the t e r r i t o r y , and (c) nesting success in r e l a t i o n to the s p a t i a l relationship of the nest to the t e r r i t o r y . Location of the t e r r i t o r y was known for only 54 of the 75 bufflehead nests found. The group "non-adjacent" included nests at the edge of 2 t e r r i t o r i e s and nests separated from the t e r r i t o r y by at least one other t e r r i t o r i a l pair (Fisher exact prob. t e s t ) . (A) Successful Deserted Normal 60 1 1 P = 0.011 Parasitized (B) Adjacent Not adjacent Normal 39 1 2 P = 0.018 Parasit ized (C) Successful Deserted Adjacent 28 P = 0.334 Not adjacent nest s i t e hypothes 44 DISCUSSION Food hypothesis When food abundance i s sampled, there is always the problem of whether what was sampled is a good index of food available to the animal. Buffleheads are opportunistic feeders: the prey items found in the invertebrate samples were eaten by buffleheads roughly in proportion to their abundance, the only major discrepancy being amphipods. The average invertebrate biomass in my ponds ranged widely from 0.07 to 4.81 g dry weight/m3, as found in similar studies in the p r a i r i e s (Kaminski and Prince 1981, Murkin et a l . 1982). F i n a l l y , studies using similar sampling techniques (e.g. Murkin et a_l. 1 982), or cruder indices of food abundance (e.g. Joyner 1980, Scott and Birkhead 1983), have shown positive effects of food abundance on pair density or reproductive success. For these reasons, I believe that my invertebrate sampling was an adequate index of •food available to buffleheads. My data provided l i t t l e support for the hypothesis that pair t e r r i t o r i e s are food-mediated in buffleheads: there was no relationship between t e r r i t o r y size and food density in the t e r r i t o r y over a wide range of food density, and reproductive success was, at best, weakly affected by food abundance. In the boreal forest of north-central Alberta where productivity (and presumably food abundance) i s lower than in the Cariboo, one might expect t e r r i t o r y sizes to be larger. However, Donaghey (1975) reported t e r r i t o r y sizes that were similar to those found 45 in this study, ranging from 0.3 to 1.2 ha (no mean given). The only measure of reproductive success that was consistently correlated with food density was start of egg-laying (Fig. 6). An effect of food abundance on start of egg-laying has been reported in several- other species as well (e.g. Kallander 1974, Yom Tov 1974, Smith et a l . 1980, Dijkstra et a l . 1982, Ewald and Rowher 1982). This suggests that the relationship may be a general one in birds. Early laying females l a i d s l i g h t l y larger clutches, and thus food abundance could s t i l l have had an indirect effect on reproductive success. There was a weak positive r e l a t i o n s h i p between food abundance and clutch weight in 1983, but not in 1984. .The evidence presented above suggests that-food i s not a major resource defended by t e r r i t o r i a l buffleheads, at least over the range of food density'found in this study. The shelduck is the only other t e r r i t o r i a l duck for which food abundance has been investigated in d e t a i l : Buxton (i_n Patterson 1982) could not find any inverse relationship between t e r r i t o r y size and food abundance in the t e r r i t o r y , and he also concluded that food was not the resource defended by shelducks. Defense of food resources has also been invoked to explain t e r r i t o r i a l i t y in gadwall (Dwyer 1974), shoveler (Seymour 1974a), and steamer ducks (Livezey and Humphrey 1985). However, a relationship between food abundance and t e r r i t o r y size or reproductive success has yet to be shown in any of these species. Thus, the hypothesis that food per se is a major resource defended by t e r r i t o r i a l ducks i s not supported by any evidence. 46 Mate-guarding and nest s i t e hypotheses The removal experiment showed that the t e r r i t o r y provides an undisturbed feeding area for the female: widowed females were chased off the i r t e r r i t o r y , they spent less time feeding, more time being a l e r t , and were harassed by males much more often (Fig. 7). There was no evidence that widowed females were looking for the i r lost mate (e.g. by swimming around and c a l l i n g for t h e i r mate) and the a l e r t behavior observed after the removal was indistinguishable from before the removal. Males never attempted for.ced copulations with widowed females: males either attacked them or they paired with them. If males do not attempt forced copulations on paired females, there should be no risk of cuckoldry for.paired males. However, even i f protection of their paternity may not be c r i t i c a l for male buffleheads, mate-guarding i s s t i l l advantageous for them because, when they are absent (as in the removal experiment), females can be attacked and driven off their t e r r i t o r y . One of the two widowed females that lost her t e r r i t o r y permanently later deserted her nest, further suggesting that the male presence helps the female to nest successfuly. Thus, mate-guarding in buffleheads could be important for males, not so much to protect their paternity per se, but rather to protect the female from attacks by other males and to insure that she w i l l nest successfuly. Although mate-guarding influences t e r r i t o r i a l i t y in buffleheads, i t does not explain why their t e r r i t o r i e s are so s i t e - s p e c i f i c because, according to this hypothesis, males could 47 merely defend a moving t e r r i t o r y around the female (Type 2 or 3). The nest s i t e hypothesis offers a s u f f i c i e n t additional explanation for type 4 t e r r i t o r i a l i t y in buffleheads, i . e . t e r r i t o r i e s are fixed because females stay near the nest s i t e . The evidence (Table V) suggests that the t e r r i t o r y provides some protection for the nest (at least from parasitism) and that females prefer to s e t t l e in front of their nests. My data, though limited, also suggest that parasitized nests are more l i k e l y to be deserted than normal nests (Table V). This has also been reported in other ducks (Jones and Leopold 1967, Titman and Lowther 1975, Hines and M i t c h e l l 1984, but see also Clawson et a^. 1979, Heusmann et a_l. 1980 for contradictory evidence). Buffleheads very rarely renest and, although natural c a v i t i e s did not appear to be in short supply in my area (see chapter 4), they may be l i m i t i n g in other parts of the bufflehead breeding range (Erskine 1972). Thus, the risk of nest parasitism, and of having the nest taken over by another female, makes protection of the nest s i t e from other females advantageous. Protection of the nest s i t e i s , however, probably not s u f f i c i e n t to explain t e r r i t o r i a l i t y in buffleheads. For instance, wood ducks (Aix sponsa) also compete strongly for nest s i t e s and suffer a high rate of nest parasitism (Clawson et a l . 1979, Heusmann e_t a l . 1980), yet they are not t e r r i t o r i a l (Grice and Rogers 1965). Therefore, protection of the nest s i t e may only explain why bufflehead t e r r i t o r i e s are so s i t e - s p e c i f i c . The nest s i t e hypothesis can also explain why males behaved so aggressively toward widowed females, and also sometimes 48 toward paired females (especially n o n - t e r r i t o r i a l females, Donaghey 1975). Females select the t e r r i t o r y (Donaghey 1975, Savard 1984), and thus, by attacking other females, the male may insure that she and her mate w i l l not settle and compete with his mate for the same nest s i t e . Donaghey (1975) also reported male buffleheads attacking females f l y i n g to the nest of their mate. Goldeneyes (2 spp.) are closely related to buffleheads. They have a very similar t e r r i t o r i a l system and the three species are i n t e r s p e c i f i c a l l y t e r r i t o r i a l (Savard 1982a, 1984). However, pairs cannot obtain a t e r r i t o r y adjacent to their nest as often as buffleheads do (Savard J.-P., pers. comm.), because, being larger than buffleheads, they are r e s t r i c t e d to larger and scarcer- c a v i t i e s (Erskine 1972, Savard 1982b, in prep.). This could explain why nest parasitism i s more frequent in goldeneyes than in buffleheads, ranging from 25 % to 40 % (Savard J.-P., Eadie J. McA., pers. comm.). Int e r s p e c i f i c t e r r i t o r i a l i t y in the genus Bucephala could also be accounted for by competition for nest s i t e s (see also Erskine 1960, McLaren 1969). T e r r i t o r i a l i t y and mate-guarding There i s growing evidence that mate-guarding i s a major function of t e r r i t o r i a l i t y in waterfowl (Dwyer 1974, Seymour 1974a, Seymour and Titman 1978, Mineau and Cooke 1979, Stewart and Titman 1980, Patterson 1982, McKinney et a l . 1983, Titman 1983, Savard 1984, t h i s study). This conclusion apparently contradicts Beecher and Beecher (1979) and Power and Doner's 49 (1980) contention that strong mate-guarding should evolve mostly in species with extensive male parental care because of the higher cost of male cuckoldry in these species. However, in addition to prevention of cuckoldry, mate-guarding in ducks also provides an undisturbed feeding area for the female during egg-laying, a period of high energy demand for her (Krapu 1979, Owen and Reinecke 1979). Thus, the combination of these two factors ( i . e . prevention of male cuckoldry and provision of an undisturbed feeding area to the female) has probably promoted the evolution of mate-guarding and t e r r i t o r i a l i t y in ducks. Mate-guarding has recently been associated with t e r r i t o r i a l behavior in several other species of birds such as savannah sparrow (Passerculus sandwichensis, Weatherhead and Robertson 1980), b l a c k - b i l l e d magpie (Pica pica, Birkhead 1979), mountain bluebird ( S i a l i a currucoides,. Power and Doner 1980), and pied flycatcher (Ficedula hypoleuca, Bjorklund and Westman 1983). Dufty (1982) even suggested that mate-guarding is a major function of male t e r r i t o r i a l i t y in the brown-headed cowbird (Molothrus a t e r ) . A l l t h i s evidence suggests that mate-guarding may be a more important function of t e r r i t o r i a l i t y , at least in breeding birds, than has been t r a d i t i o n a l l y assumed (e.g. Hinde 1956, Wilson 1975). If food does not influence t e r r i t o r y s i z e , what does determine t e r r i t o r y size in buffleheads? The inverse relationship between density and t e r r i t o r y size (Table I) and the extension of neighboring male t e r r i t o r i e s a f t e r the removal suggest that density and intruder pressure may be one of the 50 proximal factors. The hypothesis that i n t r a s p e c i f i c interactions partly determine t e r r i t o r y size has recently gained considerable support (e.g. Ewald e_t a l . 1 980, Myers et_ a l . 1981, Norman and Jones 1984). The number of pairs attempting to s e t t l e in spring on a given pond may influence breeding density as well as t e r r i t o r y s i z e . I conclude that mate-guarding i s an important function of t e r r i t o r i a l i t y in buffleheads, and that protection of the nest s i t e may be a secondary function which can explain why their t e r r i t o r i e s are so s i t e - s p e c i f i c . I found l i t t l e evidence that t e r r i t o r i e s were defended for their, food value in this species. CHAPTER I I I : THE ADAPTIVE SIGNIFICANCE OF BROOD TERRITORIES 52 INTRODUCTION Spacing behavior of breeding ducks has recently received considerable attention (McKinney 1965, B a l l et a l . 1978, Stewart and Titman 1980, Titman and Seymour 1981, Patterson 1982, Titman 1983, Savard 1982a, 1984). Much of thi s work, however, has focused only on t e r r i t o r i a l behavior of males during the nesting period. In ducks, the pair bond usually breaks during incubation, and males are only t e r r i t o r i a l from the pre-laying to early incubation stage (McKinney 1965, Titman 1983, see chapter 2). Although females with broods may exhibit some spacing behavior (Haland 1983), they are t y p i c a l l y not t e r r i t o r i a l (McKinney 1965, Haland 1983). Thus, waterfowl t e r r i t o r i e s contrast sharply with the c l a s s i c avian breeding t e r r i t o r y defended from pre-laying to fledging- (Nice 1941). Some ducks, however, have been reported to defend brood t e r r i t o r i e s (brood t e r r i t o r y i s herein defined as a t e r r i t o r y defended by the female or both parents after hatching of the clutch). They include most species of ducks in which both parents attend the young, such as the common shelduck (Patterson 1982), the Australian shelduck (Tadorna tadornoides, Riggert 1977), steamer ducks (Tachyeres sp., Weller 1976), and the African black duck (Anas sparsa, B a l l et a l . 1978, McKinney et a l . 1978). T e r r i t o r i e s are also defended in a few species where broods are attended only by the female, such as the Barrow's goldeneye (Bengston 1971, Savard 1982a) and the bufflehead (Donaghey 1975). Most species that defend brood t e r r i t o r i e s are also strongly t e r r i t o r i a l during the nesting period (they 53 exhibit "type 4" t e r r i t o r i a l i t y ; see chapter 2). However, brood t e r r i t o r i e s are generally defended in d i f f e r e n t places from the pair t e r r i t o r i e s (Donaghey 1975, Patterson 1982; see discussion). Several hypotheses have been proposed to account for brood t e r r i t o r i e s in these species. T e r r i t o r i a l i t y could be an a n t i -predator strategy for the brood. Mortality rate of ducklings is usually high (often > 50%; e.g. Pienkowski and Evans 1982, Ringelman and Longcore 1982, Talent et a l . 1983) and predation can be a major mortality factor (Munro and Bedard 1977a, Talent et a l . 1983). Spacing out of broods through t e r r i t o r i a l i t y could enhance concealment and decrease chances of detection by predators (McKinney 1965, Tinbergen et a l . 1967, Krebs 1971). Brood t e r r i t o r i e s could also prevent brood-mixing (Patterson et a l . 1982). Brood-mixing or "creching" in ducks is often associated with high brood density (Williams 1974, Munro and Bedard 1977b, Pienkowski and Evans 1982). Brood-mixing apparently results from aggressive encounters between broods (Munro and Bedard 1977b, Patterson et a l . 1982) and from the lack of a strong imprinting by ducklings on their mother (Patterson 1982). In shelducks, duckling survival i s reduced both in creches and at high brood density (Williams 1974, Makepeace and Patterson 1980, Pienkowski and Evans 1982). Therefore, spacing out of broods could reduce brood-mixing and enhance duckling s u r v i v a l . The most widely accepted hypothesis to explain brood t e r r i t o r i e s i s that t e r r i t o r i e s secure a food reserve for the 54 growing d u c k l i n g s (Donaghey 1975, McKinney et a l . 1978, P a t t e r s o n 1982, L i v e z e y and Humphrey 1985). The growing p e r i o d i s a c r i t i c a l stage f o r d u c k l i n g s because energy requirements are very high ( R i c k l e f s 1974). There i s a l s o evidence that waterfowl broods respond to changes in food d e n s i t y and that food abundance can l i m i t brood d e n s i t y (Patterson 1976, E r i k s s o n 1978, D a n e l l and Sjoberg 1982). In t h i s chapter, I i n v e s t i g a t e the adaptive s i g n i f i c a n c e of brood t e r r i t o r i e s in b u f f l e h e a d s . S p e c i f i c a l l y , I t e s t the hypothesis that brood t e r r i t o r i e s secure an adequate food reserve f o r the d u c k l i n g s , as suggested by Donaghey (1975). A f i r s t p r e d i c t i o n of t h i s h y p o t h e s i s i s that t e r r i t o r y s i z e should be i n v e r s e l y r e l a t e d to food abundance in the t e r r i t o r y . T h i s p r e d i c t i o n has been supported by much e m p i r i c a l evidence-(e.g. Holmes 1970, Stimson 1973, Slaney and Northcote 1974, Simon 1975, Gass et a_l. 1976, Kodric-Brown and Brown 1978, Seastedt and MacLean 1979). Recent models of feeding t e r r i t o r i e s using c o s t - b e n e f i t approaches ( i n t r u d e r pressure vs food abundance; Myers et aJL. 1981) a l s o p r e d i c t that t e r r i t o r y s i z e should decrease as food d e n s i t y i n c r e a s e s (Schoener 1983, Lima 1984). A second p r e d i c t i o n of the food hypothesis i s t h a t , i f food i s a l i m i t e d resource f o r d u c k l i n g s , some measures of f i t n e s s (e.g. growth r a t e or s u r v i v a l ) should be c o r r e l a t e d with food abundance i n the t e r r i t o r y . I s h a l l a l s o d i s c u s s my r e s u l t s with respect to the other hypotheses that have been proposed to e x p l a i n brood t e r r i t o r i e s . 55 METHODS This study was conducted from 1982 to 1985 in the Cariboo Parkland of B r i t i s h Columbia, 15 km north of 100 Mile House. Details of the study area are given in chapter 2. T e r r i t o r y mapping Brood t e r r i t o r i e s were mapped in 1983 and 1984 from mid-June u n t i l mid-July on several ponds. Observations were made from a high, distant vantage point using binoculars and a spotting scope. T e r r i t o r i e s were mapped once a week on each pond during a 3-hour period (09:00 to 12:00 h PDT). Each five to f i f t e e n minutes (depending on the density of broods), the pond was scanned and the position and behavior of a 1-1 bufflehead broods were plotted on large sca'le maps ( 1:1000). A l l aggressive interactions occurring during the 3-hour period were recorded, the broods involved were i d e n t i f i e d , and their locations plotted. Females were trapped on their nests and marked with color-coded nasal saddles (Doty and Greenwood 1974). A few females were also marked when they were caught with their ducklings during brood-driving (see below). Females are highly p h i l o p a t r i c "to their nesting pond (Erskine 1961), and by 1984 about 60 % of the brooding females were marked. Thus, a l l ponds observed had both marked and unmarked females. However, a l l unmarked females could be e a s i l y i d e n t i f i e d by the combination of size and age of the ducklings in each brood. Areas of t e r r i t o r i e s were calculated using the method t 56 described in chapter 2. For a few late-hatched broods, t e r r i t o r i e s were mapped only when the brood was very young (< 10 days). These were excluded, because Donaghey (1975) reported that t e r r i t o r y size tends to increase with the age of the brood. F i g . 8 i l l u s t r a t e s the results obtained with t h i s method on pond E Soda in 1984. Time-budget In 1984, I collected time-budget data for some of the broods whose t e r r i t o r i e s were mapped and sampled for food. Each of these broods was observed for two mornings (except for two broods), usually two days apart. Because i t was impossible to keep'track of a l l ducklings in large groups, only broods with 8 ducklings or less were observed. Behavioral observations were always made between 08:00 and 11:00 h (PDT). The 3-h period was approximately divided into 8 blocks of 15 min, each one separated by a 5-min break, for a t o t a l of 2 h of observations. Ducklings were kept under constant observation within the 15-min periods, and the behavior of each duckling was recorded at 15-sec intervals (instantaneous sampling, Altmann 1974). Six behaviors were recorded: Feeding: t h i s behavior was divided into skimming ( i . e . jabbing at prey at the surface) and diving. Resting: b i l l i s tucked under the wing; eyes closed; includes brooding by the female. Preening: includes preening and bathing. Swimming: any locomotor a c t i v i t y on the water. 5 7 Figure 8. T e r r i t o r i e s of the three broods that used pond E Soda from 13 June to 14 July 1984. Each number i s a d i f f e r e n t location, and the l i n e joining the outermost points encloses the t o t a l area used by that brood. The innermost l i n e encloses the intensively used area defined here as the t e r r i t o r y . 59 Loafing: young are immobile but eyes are open and they are attentive to their surroundings. Social interactions: any interactions (e.g. aggression) with other birds. Time-budgets were calculated as follows: for each brood the percentage of time spent in d i f f e r e n t a c t i v i t i e s was calculated within each 15-min period, then averaged over the eight periods giving equal weight to each period, and f i n a l l y averaged over the two mornings of observations. Trapping and marking of young Buffleheads nested in both natural c a v i t i e s and nest boxes. Nests were found by intensively searching for natural c a v i t i e s around ponds, or by following females f l y i n g to their nests during incubation. Nests were checked upon hatching, and young were web-tagged (Haramis and Nice 1980) before they l e f t the nest, usually 24-48 h after hatching. Most ducklings (> 75 %) were caught once between 18 and 40 days of age using a drive trap (modified from Cowan and Hatter 1952). Ducklings were caught at various ages .because each pond had a number of broods of 'different ages, and the drive trap was set only once on each pond. Therefore, they were a l l caught on the same day, although broods were kept intact ( i . e . brood-mixing was avoided) by driving each brood separately. Each duckling was weighed (± 5 g), the culmen, tarsus (± 0.1 mm), and wing length (± 1 mm) were measured, and each was banded with a permanent metal band and.up to 3 p l a s t i c color 60 bands. Eversion of the phallus was found to be an unreliable method for sexing these ducklings, and therefore culmen and especially body weight were used to separate sexes. Buffleheads are strongly sexually dimorphic, and by 20 days of age males are already much heavier than females (Erskine -1972). When the frequency d i s t r i b u t i o n of duckling weights was plotted for each brood taken separately, two d i s t i n c t c l u s t e r s invariably showed up. These two clusters corresponded to males and females as established from Erskine's (1972) sex-specific growth curves. Brood counts and computation of duckling survival Counts of a l l bufflehead broods were conducted from the f i r s t week of June u n t i l the end of July. Goldeneye broods were also surveyed because they are i n t e r s p e c i . f i c a l l y t e r r i t o r i a l with buffleheads (Savard 1982a). Counts were made mostly in the morning hours (08:00 h to 12:00 h) from elevated vantage points along the shoreline. Intensively studied ponds (about 1/3 of the total) were surveyed every 3-4 days, and the remaining ponds weekly. For broods whose nests I had not found (about 40-50 % ) , hatching date was estimated to be the day midway between the f i r s t census that the brood was observed and the previous census, minus one day to account for the f i r s t 24 h spent in the nest. Thus, the maximum error on the hatching date of these broods was ± 3 days. A decrease in brood size between two consecutive censuses was taken as mortality, unless the size of a neighboring brood increased in the same period. Increases in brood size could 61 always be matched with corresponding losses of ducklings by other broods on the same pond. Exchange of ducklings between broods was r e l a t i v e l y uncommon, but did occur in a l l years of the study. Several of these exchanges were l a t e r confirmed by catching the web-tagged ducklings in the drive trap. Total brood disappearance sometimes occurred for small ducklings, and these losses were included in the calculation of mortality rates. In several cases, the females involved were marked and they were later resighted alone, thus confirming the loss of the entire brood. No other wetlands were present within 1.5 km of the study area and i t is highly unlikely that any of the broods that disappeared dispersed outside the study area. I used the Mayfield method (Mayfield 1961) to calculate brood mortality, as suggested by Ringelman and Longcore (1982). 'The advantage of t h i s method is that a l l broods are placed on a comparable basis by using only the period a brood i s under observations. The period of time between two consecutive observations, measured in duckling-days, i s termed exposure. Thus, a brood of 6 ducklings resighted intact after 5 days is credited an exposure of 30 duckling-days. If ducklings are lost during that period, the d a i l y probability (P) of duckling loss is calculated according to the formula: P(loss)n = L x P(survive exactly n days) i - P(survive N days) where L = number of ducklings l o s t , N = number of days exposed, and n = nth day (from Klett and Johnson 1982). Following these authors, I assumed that mortality occurred at a constant rate of 62 5%/day in order to spread the duckling losses over the period at exposure. Thus, i f two ducklings are lost from our brood of 6 over 5 days, the exposure for the f i r s t day i s 6 duckling-days and the loss i s : P(loss)1 = 2 x (0.95)° (1 - 0.95) 1 - (0.95) 6 = 0.3775 For the second day, the exposure i s 5.623 duckling-days and the loss 0.3586, and so on u n t i l day 6. Daily losses and exposures are then added in each sample and da i l y survival rate was obtained by (1 - (tot a l l o s s e s / t o t a l exposure)). Variance of estimates was calculated according to Johnson (1979). An assumption of the Mayfield method i s that the d a i l y mortality rate remains constant through time, which i s very unlikely to be true for the whole brood-rearing period. Although the method i s f a i r l y robust to departure from this assumption (Johnson 1979), mortality rates were calculated separately for each week from hatching to 50 days (about fledging time for buffleheads; Erskine 1972). Each of the 7 weeks correspond roughly to age-classes Ia to III of Bellrose (1976). The f i r s t week (class Ia) was further divided into two periods: from hatching to the f i r s t sighting on the water (2.76 ± 0.25(SE) days, N = 41) and from the f i r s t sighting u n t i l the end of the f i r s t week (4.20 ± 0.19 days, N = 64). 63 Invertebrate sampling and c o l l e c t i o n of young Bufflehead young feed mainly on nektonic invertebrates (see r e s u l t s ) . Invertebrates were sampled . using a modified Gerkin device (Kaminski and Murkin 1981). Details of the method are given in chapter 2. Within a 4-day period at the peak of the brood-rearing season (3-6 July in 1983, 2-4 July in 1984), I sampled a l l ponds where t e r r i t o r i e s were mapped. From 4 to 12 samples were systematically c o l l e c t e d in each t e r r i t o r y , and the average of these samples gave an index of food abundance for the t e r r i t o r y . In the lab, samples were sorted to the order or family l e v e l , organisms were separated in size classes according to body length, and counted. Numbers were then transformed into biomass. Results are presented in g dry weight/m2. Data oh food habits of bufflehead young are scanty (Erskine 1972), and I therefore c o l l e c t e d some ducklings in both years to analyze their d i e t . Ducklings were observed feeding for at least 15 minutes before shooting them. Immediately after c o l l e c t i o n (ca 10-15 minutes), I removed the gut and preserved i t in 10 % formaldehyde solution. Prey items were later sorted to the order or family l e v e l , counted, dried overnight at 60° C and weighed. Only a small proportion of the gizzard content could be i d e n t i f i e d (ca 25 % ) , and thus only data from the oesophagus (including the proventriculus) i s presented. Because c o l l e c t i o n of birds was incompatible with t e r r i t o r y mapping and banding a c t i v i t i e s , ducklings were c o l l e c t e d on peripheral ponds where no intensive, invertebrate sampling took place. A few invertebrate samples were nonetheless co l l e c t e d on these ponds 6 4 using a c t i v i t y traps (Murkin et a l . 1983). These samples did not indicate any large differences in the type of invertebrates found between the ponds where ducklings were c o l l e c t e d and those where t e r r i t o r i e s were mapped. A l l s t a t i s t i c a l analyses were performed using the SPSSx package. S t a t i s t i c a l tests were one-tailed (unless otherwise sp e c i f i e d K because they dealt with a p r i o r i predictions. 65 RESULTS Food habits Thirteen ducklings were co l l e c t e d in 1983 and six in 1984. Only 15 of them had food in the oesophagus. Zygopteran larvae were by far the most abundant prey item (Fig. 9). Despite the fact that only oesophagi were analyzed, about 25% of the food was already digested beyond recognition. Munro (1942) summarized the food eaten by 20 ducklings c o l l e c t e d near my study area, and these data are also presented in F i g . 9. In that sample, Coleoptera (both adults and larvae), zygopteran larvae and amphipods were the dominant prey items. The abundance of Coleoptera and amphipods-. in Munro's data contrasts sharply with my data. This poses the problem of what should be considered food available to bufflehead young. The discrepancy for amphipods between the two studies is unfortunate because they were the most abundant aquatic invertebrate in my sampling (Fig. 9, bottom). However, because amphipods are digested very rapidly (Swanson and Bartonek 1970), they are l i k e l y to be underrepresented even in the oesophagus (see discussion). I therefore defined food available to the young as being a l l invertebrate taxa found in the sampling (including amphipods), with the exception of cladocerans (which were never found in bufflehead oesophagi, F i g . 9). The consequences of including amphipods in the food available to buffleheads w i l l be discussed l a t e r . 66 Figure 9. Diet of bufflehead ducklings (upper histogram: my data; middle histogram: from Munro 1942), and the proportion of d i f f e r e n t taxa found in the invertebrate sampling of brood t e r r i t o r i e s . 67 "D 30-20 10 • D i e t ( T h i s s t u d y N = 1 5 ) 30i 03 E 3 20 H o > 10 -D i e t ( M u n r o 1 9 4 2 , N = 2 0 ) "D 50-30-20' 10-I n v e r t e b r a t e s a m p l i n g ( N = 1 4 0 ) • E O CD a isoj X E o isoj a> O) c o c >> CO o a N 0) a c o o .3 © C ^ o ** ° o o a -o 3 C 3 68 Brood density and food abundance If food i s a c r i t i c a l resource for ducklings, females with broods should respond to changes in food abundance and concentrate on ponds with high food density. To test for t h i s , I examined i f the maximum density of broods recorded on each pond was related to the average food density per pond. Because broods are often spaced out along the shoreline (Fig. 8), I used two indices of density: number of broods per ha of water and per km of shoreline. In 1983, the corr e l a t i o n between brood density and food density was s i g n i f i c a n t with the f i r s t index of brood density and approached si g n i f i c a n c e with the second index (P = 0.07), but not in 1984 (Table VI). The reasons for this between-year difference is uncertain although the within-pond s p a t i a l v a r i a b i l i t y in food abundance may be a contributing factor. When both years were combined, both indices of density were s i g n i f i c a n t l y correlated- with food abundance (Table VI). This suggests that broods respond to changes in food density. T e r r i t o r y size and food abundance The f i r s t p r ediction of the food hypothesis is that t e r r i t o r y size should be inversely correlated with food density. If brooding females adjust t e r r i t o r y size (T) to food density (FD) in the t e r r i t o r y , these two variables should be related by a hyperbolic function (Gass et a l . 1976). I found that T was indeed negatively correlated with FD (r = -0.55, P < 0.001, N .= 31), and that the r e l a t i o n s h i p was hyperbolic (Fig. 10). The 69 T a b l e V I . C o r r e l a t i o n c o e f f i c i e n t s of b u f f l e h e a d brood d e n s i t y w i t h average fo o d d e n s i t y per pond. Two i n d i c e s of brood d e n s i t y were used, number of broods per ha of water and per km of s h o r e l i n e (N = number of pond s ) . Year N Brood d e n s i t y ha km 1 983 0.75 * 0.56 1 984 0.24 0.49 1983+1984 1 4 0.59 * 0.49 * * P < 0.05, o t h e r w i s e P > 0.05 70 Figure 10. Relationship between t e r r i t o r y size (T) of broods and food density (FD) in the t e r r i t o r y , both years combined. Log T = -1.14 log FD - 0.54 (P < 0.001; functional regression). The relationship expected i f t o t a l food i s constant regardless of t e r r i t o r y size (slope = -1) i s also presented (dashed l i n e ) . Food density (g/m2) 72 r e l a t i o n s h i p f u r t h e r h e l d f o r both years of the study (1983, r = -0.59, P < 0.01, N = 17; 1984, r = -0.58, P = 0.01, N = 14). I f bu f f l e h e a d s a d j u s t t h e i r t e r r i t o r y s i z e to maintain a constant food supply, the slope of the equation r e l a t i n g T to FD should be -1. The slope of the f u n c t i o n a l r e g r e s s i o n (Ricker 1973, 1975) r e l a t i n g these two v a r i a b l e s was not s i g n i f i c a n t l y d i f f e r e n t from -1 f o r both years combined (b = -1.14, -0.83 to -1.56 (95 % C I ) , F i g . 10), nor f o r each year taken s e p a r a t e l y (1983, b = -1.04, -0.68 to -1.60; 1984, b = -1.14, -0.70 to -1 .87) . If b u f f l e h e a d s maintain a constant food supply i n t h e i r t e r r i t o r y by a d j u s t i n g i t s s i z e , the t o t a l amount of food i n the t e r r i t o r y (FD x T) should remain s t a b l e as t e r r i t o r y s i z e i n c r e a s e s and, c o n v e r s e l y , as food d e n s i t y decreases. T o t a l food in the t e r r i t o r y d i d not change with FD but d i d i n c r e a s e with T (Table V I I ) . The l a t t e r r e l a t i o n s h i p was, however, s t r o n g l y a f f e c t e d by 3 un u s u a l l y l a r g e t e r r i t o r i e s (one of which was h e l d by a mixed-brood of 26 d u c k l i n g s ) . If these broods were excluded, t o t a l food in the t e r r i t o r y was no longer c o r r e l a t e d with T (Table V I I ) . Other f a c t o r s were a l s o c o r r e l a t e d with t e r r i t o r y s i z e : T was p o s i t i v e l y c o r r e l a t e d with brood s i z e (r = 0.53, P < 0.01, N = 34), and n e g a t i v e l y c o r r e l a t e d with brood d e n s i t y (BD) (Table V I I I ) . Although goldeneye broods were about ten times l e s s abundant than b u f f l e h e a d s , the s t r e n g t h of the c o r r e l a t i o n i n c r e a s e d when they were i n c l u d e d i n the d e n s i t y i n d i c e s . T h i s i s c o n s i s t e n t with the i n t e r s p e c i f i c t e r r i t o r i a l i t y shown by 73 Table VII. Regression of t o t a l amount of food in brood t e r r i t o r i e s on t e r r i t o r y size (T) and food density (FD) in the t e r r i t o r y . The second regression equation with t e r r i t o r y size excludes three unusually large t e r r i t o r i e s (see r e s u l t s ) . Total food vs N slope 95%CI intercept r P T 31 0.550 0.180 0.141 0.76 <0.01 28 0.383 0.663 0.180 0.23 0.12 FD 31 0.064 0.090 0.231 0.26 0.08 74 Table VIII. Correlation and p a r t i a l c o r r e l a t i o n c o e f f i c i e n t s of brood t e r r i t o r y size (T) with food density (FD) in the t e r r i t o r y , brood ' size (SIZE), bufflehead (BUFF) brood density (BD), and bufflehead + goldeneye (GOLD) brood density per ha of water. df log FD BUFF BD BUFF+GOLD BD Log T 30 -0.55 ** -0.54 ** -0.62 *** Log T con t r o l l i n g for SIZE 28 -0.57 ** -0.57 ** -0.65 *** SIZE & BUFF BD 27 -0.37 * SIZE & BUFF + GOLD BD 27 -0.2'4 SIZE & FD 27 -0.37 * -0.44 ** * P < 0.05, ** P < 0.01, *** P < 0.001, otherwise P > 0.1 75 these two species (Savard 1982a). In the analysis of brood density, each brood was considered to be an independent sample because hatching was asynchronous, and broods used the same ponds at d i f f e r e n t periods. To explore the interactions among a l l these variables (FD, BD, brood size and t e r r i t o r y s i z e ) , I used a p a r t i a l correlation analysis (e.g. Myers et. a l . 1979). This analysis showed that when the interaction of T with both brood size and BD (density of bufflehead and goldeneye broods combined) were controlled s t a t i s t i c a l l y , the correlation between T and FD was no longer s i g n i f i c a n t (Table VIII). However, when the interaction of T with brood size and FD were controlled, T was s t i l l s i g n i f i c a n t l y correlated with BD. Thus, the relationship between T and BD was stronger than between T and FD. To summarize, there was a s i g n i f i c a n t hyperbolic relationship between t e r r i t o r y size of broods and food density, and the slope of the regression was not d i f f e r e n t from -1. However, brood density accounted for s l i g h t l y more v a r i a b i l i t y in t e r r i t o r y size than food density. Growth rate Ducklings were caught only between 18 to 42 days of age (see methods for determination of age). The relationship between weight and age was best described by a linear regression (Fig. 11). Weights f a l l i n g above the regression l i n e were taken as evidence that these ducklings were growing at a faster rate than average and, conversely, those f a l l i n g below as evidence of 7 6 Figure 11. Relationship between duckling weight and age. Males (X), Females (0). (A) males, y =10.7x - 62.0, r = 0.90, P < 0.001; females, y = 6.95x - 3.50, r = 0.87, P <0.001 (B) 1984: males, y = 9.34x - 20.6, r = 0.82, P < 0.001; females, y = 5.70x + 33.4, r = 0.80, P < 0.001 WEIGHT (g) LL WEIGHT (g) 79 a slower growth rate. The residuals of the regression were thus used as an index of growth rate. However, the variance of residuals increased with age (Fig. 11). This result was not surprising because difference in growth rates among ducklings should lead to progressively larger differences in weight as they get older. To correct for t h i s , I s t a b i l i z e d the variance of the residuals using the following transformation: X' = 1/X; Y' = Y/X (Meter and Wasserman 1974). I used these residuals to derive three indices of growth rate: MEANWEIGHT and MAXWEIGHT were respectively the average and maximum residuals for each brood, and TOTWEIGHT was the product of MEANWEIGHT by brood size. If food l i m i t s duckling growth, there should be a positive relationship between growth rate and food abundance. I found no relationship between any indices of growth rate and t o t a l food in the t e r r i t o r y but, contrary to my prediction, these indices were negatively correlated with FD in both years (Table IX). However, growth indices were also negatively correlated with BD, at least in 1983. Surprisingly, correlations with BD were stronger in 1984 when goldeneye broods were excluded: MEANWEIGHT, r = -0.58, P < 0.05; MAXWEIGHT r = -0.54, P < 0.05; TOTWEIGHT r = -0.49 P = 0.05. Again using p a r t i a l c o r r e l a t i o n , I found that when BD was controlled for, correlations between growth indices and FD were no longer s i g n i f i c a n t (Table X). When I controlled for FD, MEANWEIGHT and MAXWEIGHT were s t i l l negatively correlated with BD in 1983, but not in 1984. 80 Table IX. Correlation c o e f f i c i e n t s of three indices of duckling growth rate (MEANWEIGHT, MAXWEIGHT, and TOTWEIGHT; see results) with t o t a l amount of food in the t e r r i t o r y , food density (FD) in the t e r r i t o r y , t e r r i t o r y size (T), and bufflehead + goldeneye brood density (BD) per ha of water and km of shoreline (1983, N = 16 broods; 1984, N = 12). Log t o t a l food Log FD Log T BD per ha BD per km Year MEANWEIGHT MAXWEIGHT TOTWEIGHT 1983 -0.18 0.00 -0.09 1984 -0.27 -0.27 -0.27 1983 -0.59 ** -0.62 ** -0.54 * 1984 -0.47 -0.47 -0.52 * 1983 0.39 0.57 * 0.43' * 1984 0.17 0.16 0.20 1983 -0.71 ** -0.78 *** -0.46 * 1984 -0.24 -0.21 -0.23 1983 -0.37 -0.49 * -0.37 * 1984 -0.37 -0.35 -0.46 * P < 0.05, ** P < 0.01, *** P < 0.001, otherwise P > 0.05 81 Table X. P a r t i a l c o r r e l a t i o n c o e f f i c i e n t s of three indices of duckling growth rate (see results) with food density (FD) in the t e r r i t o r y and bufflehead + goldeneye brood density (BD). Brood density i s expressed per ha of water in 1983 and per km of shoreline in 1984 (1983, df = 13; 1984, df = 9). Year MEANWEIGHT MAXWEIGHT TOTWEIGHT 1983 -0.16 -0.13 -0.34 Log FD (con t r o l l i n g for BD) 1984 -0.32 -0.33 -0.29 1983 -0.51 * -0.62 ** -0.12 BD (contr o l l i n g for FD) 1984 -0.03 -0.04' -0.13 * P < 0.05, ** P < 0.01, otherwise P > 0.1 82 Because weight was used to sex ducklings, the use of residuals based on the sex-specific regression could lead to c i r c u l a r i t y . I therefore repeated the previous analysis using the residuals based on the combined male-female regression l i n e . The results of this analysis were found to be i d e n t i c a l to those based on the sex-specific regression. Brood survival In both years, daily survival of ducklings increased with age (Fig. 12): i t increased s i g n i f i c a n t l y from hatching to f i r s t sighting on water, to age-class Ia, lb, and Ic, but remained high *and constant beyond class Ic (21 days). Overall duckling survival from hatching to fledging (50 days) was 0.4328. -Survival from class Ia to III did not d i f f e r between years, but survival from hatching to f i r s t sighting on the water was s i g n i f i c a n t l y higher in 1983 (P < 0.001, F i g . 12). Because survival was very high beyond 21 days, I retained only two variables for the following analysis: survival from hatching to f i r s t sighting (hatching survival) and from Ia to the end of Ic (class I s u r v i v a l ) . Weekly survival was f i r s t obtained for each of these periods from (daily survival)** where d = days at exposure." Total survival of class I was the product of the weekly survival of each class (Ia, lb and I c ) . If food l i m i t s duckling s u r v i v a l , there should be a posi t i v e relationship between food abundance and duckling survival and/or the number of young fledged. I found that duckling survival was not correlated with t o t a l food in the t 83 Figure 12. Daily survival rate of ducklings from hatching (week 0, hatching to f i r s t sighting on water) to age-class III (week 7). Mean(SE). 1983, N = 6036 duckling-days (34 broods); 1984, N = 5052 duckling-days (33 broods). Comparisons are between consecutive classes (* P < 0.05, ** P < 0.01, *** P < 0.001, NS P > 0.05). There was no difference between years for any class (P > 0.05) except for class 0 (P < 0.001). 1.00 CO 0.95 > • • CO 0.90 83 CO Q 0.85 84 0.80 0.751 NS NS , NS , Age (weeks) CO 85 t e r r i t o r y nor with t e r r i t o r y size in either year (Table XI). Both survival indices were also not correlated with FD in 1983, but class I survival was correlated with FD in 1984 (hatching survival was also weakly correlated with FD in 1984, P = 0.07, Table XI). Class I survival was not correlated with BD in either year but hatching survival was negatively correlated with BD in 1983. F i n a l l y , survival was not correlated with brood size in either year (Table XI). However, when both years were combined and two very large mixed-broods were excluded (26 ducklings each), class I survival was weakly correlated with brood size (r_ = 0.35, P = 0.03, N = 29) . Number of ducklings reaching class l i b (28 days) was taken as an index of the number of young fledged because a few broods could not be followed a l l the way to class I I I . Number, of young fledged was not correlated with either t o t a l food or FD in the t e r r i t o r y (P > 0.1). In 1983, however, the number of young fledged was negatively correlated with BD at hatching (r_ = -0.77, P < 0.01, N = 9) but not with BD at class I (r = -0.19, P > 0.1, N = 16). There was no such rel a t i o n s h i p in 1984 (P > 0.1). To summarize, in 1983 hatching survival and number of young fledged were negatively correlated with brood density at hatching, whereas in 1984 class I survival was p o s i t i v e l y correlated with food density. .86 Table XI. Correlation c o e f f i c i e n t s of hatching and class I survival of ducklings with t o t a l amount of food in the t e r r i t o r y , food density (FD) in the t e r r i t o r y , t e r r i t o r y size (T), brood size, and bufflehead + goldeneye brood density (BD) per ha of water (N = number of broods). Hatching survival Class I survival 1983 1984 1983 1984 N 9 7 17 14 Log t o t a l food 0.18 0.58 0.26 0.36 Log FD -0.13 0.62 -0.05 0.52 * Log T • 0.35 . 0.25 0.27 -0.15 Brood size 0.41 -0.49 0.28 0.38 BD -0.63 * 0.19 -0.41 -0.05 * P < 0.05, otherwise P > 0.05 87 Time budget Correlations between the four major behaviors and food, brood density, and t e r r i t o r y size are presented in Table XII. Only swimming was s i g n i f i c a n t l y correlated with FD and BD. Because swimming was predominantly associated with feeding, I combined swimming and feeding into "foraging" ( i . e . feeding related a c t i v i t i e s ) . Foraging was s i g n i f i c a n t l y correlated only with brood density (Table XII). 88 Table XII. Correlation c o e f f i c i e n t s of time spent in different a c t i v i t i e s by ducklings with t o t a l amount of food in the t e r r i t o r y , food density (FD) in the t e r r i t o r y , t e r r i t o r y size (T) and bufflehead + goldeneye brood density (BD) per ha of water (N = 8 broods) Feeding Swimming Resting Loafing Feeding+ Swimming Total food 0.16 0.01 0.30 -0.61 0.09 Log FD 0.33 0.65 * -0.14 -0.52 0.55 Log T -0.08 -0.46 0.40 -0.23 -0.31 BD 0.50 0.65 * -0.16 -0..38 0.63 * * P < 0.05, otherwise P > 0.05 89 DISCUSSION Food habits of ducklirigs A l l the analyses presented here included amphipods in the index of food abundance to ducklings. When they were removed, most relationships were weaker and often not s t a t i s t i c a l l y s i g n i f i c a n t . Although I found only traces of amphipods in the oesophagi of ducklings (Fig. 9), several l i n e s of evidence suggest that consumption of amphipods was underestimated. F i r s t , the large proportion of unidentified prey items found in the oesophagi and p r o v e n t r i c u l i show that digestion is already occurring in these organs. Second, Swanson and Bartonek (1970) showed that, in blue-winged t e a l , amphipods were digested much faster than a number of other aquatic invertebrates. Therefore, some of the amphipods eaten by the ducklings may have already been digested beyond recognition. Third, Munro (1942) found amphipods to be nearly as important as odonates and coleopterans in the stomach of a sample of bufflehead ducklings c o l l e c t e d near my study area. The discrepancy between Munro's results and mine suggests that buffleheads are opportunistic, and can switch from one prey item to another depending upon their r e l a t i v e abundances. Amphipods have also been reported to be an important prey item for ducklings of other diving ducks (e.g. Sugden 1973) . Although the status of amphipods as a food for bufflehead ducklings remains uncertain, t h i s evidence suggests that they are a more important prey item than shown by my r e s u l t s . Thus, I 90 f e e l j u s t i f i e d to include amphipods in my index of food a v a i l a b i l i t y . Cladocerans were deleted from the invertebrate samples because they were not found in the oesophagi (Fig. 9). Although the arguments presented for amphipods could also apply to cladocerans (see also Erskine 1972), th e i r exclusion did not a f f e c t any of the relationships presented above. Te r r i t o r y size, food abundance and brood density The hyperbolic relationship between T and FD (Fig. 10) strongly supports the hypothesis that t e r r i t o r i e s are defended for their food value. The slope of that r e l a t i o n s h i p was further very close to -1 which suggests that buffleheads tend to adjust their t e r r i t o r y size to maintain food supply. Migrant hummingbirds also defend feeding t e r r i t o r i e s , and several studies have shown that they can maintain food supply to a constant l e v e l by regulating t e r r i t o r y size (Gass et a l . 1976, Kodric-Brown and Brown 1978, Gass 1979, Hixon et a l . 1983). Despite t h i s relationship, v a r i a b i l i t y in F i g . 10 was r e l a t i v e l y high, and several factors may have contributed to t h i s . For instance, large broods had bigger t e r r i t o r i e s , presumably because more young need more food. When brood size was controlled for, the co r r e l a t i o n c o e f f i c i e n t between T and FD did improve (Table VIII). Several other factors such as variation in i n d i v i d u a l aggressiveness (Watson and M i l l e r 1971, Wingfield 1984), or resource holding p o t e n t i a l (Petrie 1984) have also been shown to aff e c t t e r r i t o r y s i z e . None of these, however, were measured or could be c o n t r o l l e d f o r . 91 T e r r i t o r y size also decreased as brood density increased. If food is a more important determinant of t e r r i t o r y size than brood density, we would expect the c o r r e l a t i o n between T and FD to be stronger than between T and BD. The p a r t i a l c o r r e l a t i o n analysis, however, revealed the opposite. This suggests that brood density may be a more important proximal factor than food density in determining t e r r i t o r y s i z e . This conclusion, however, must be viewed with caution because of the dangers of i n f e r r i n g causation from cor r e l a t i o n s . This result i s , nonetheless, consistent with experimental evidence showing that density of t e r r i t o r i a l individuals and intruder pressure can determine t e r r i t o r y size (Norton et a l . 1982, Norman and Jones 1984). Therefore, although brooding females can adjust t e r r i t o r y size to maintain food supply, t e r r i t o r y size i s ultimately a compromise between the number of broods s e t t l i n g in a pond and food density in the t e r r i t o r y . Growth rate, survival and time budget Evidence on the second prediction of the food hypothesis (a positive r e l a t i o n s h i p between food, growth rate and survival) was contradictory: in both years, growth rate was negatively correlated with FD, but duckling survival was p o s i t i v e l y correlated with FD in one year. The negative relationship of growth with food density was unexpected, e s p e c i a l l y since Hunter et a l . (1984) showed that black duck and mallard ducklings grew more slowly on ponds where food had been experimentally reduced. 92 The p a r t i a l c o r r e l a t i o n analysis, however, showed that growth rate was more strongly correlated with BD than with FD, at least in 1984 (Table X). This suggests that the negative r e l a t i o n s h i p between growth and food density could result i n d i r e c t l y from an effect of brood density on growth rate. Frequency of aggressive interactions increased with brood density on the pond (Kendall's tau = 0.76, P < 0.01, N = 8), as also reported by Donaghey (1975) and, in shelducks, by Makepeace and Patterson (1980). Thus, increased aggressive interactions at high brood density may have caused the depressed growth rate observed in ducklings. Pienkowski and Evans (1982) also showed that shelduck broods grew faster and fledged e a r l i e r at isolated s i t e s than at more crowded ("colonial") s i t e s . Duckling survival was p o s i t i v e l y correlated with food abundance in 1984. Starvation of young is probably rare, and reduced food intake i s l i k e l y to affect survival through s u s c e p t i b i l i t y to c h i l l i n g and exposure to adverse weather (Makepeace and Patterson 1980). It i s interesting that in 1984, the only year with a s i g n i f i c a n t c o r r e l a t i o n between survival and FD, there was a cold and wet summer. Mean minimum temperature for June was 7.8° C in 1984 compared to 10.6° C in 1983. Erskine (1972) also reported that duckling survival decreased by 40 % under similar conditions. This suggests that food supply in the t e r r i t o r y i s p a r t i c u l a r l y important for duckling survival in adverse weather. As food density increases, young should meet their energy requirements faster, and should therefore spend less time 93 feeding. Kaminski and Prince (1981), however, found that foraging time of four species of dabbling ducks increased with food density. Schoener's (1983) models of feeding t e r r i t o r i e s further predict that time spent feeding can either increase or decrease with an increase in FD depending on the constraints. In th i s study, time spent swimming increased with FD and time spent "foraging" (feeding'+ swimming) tended to increase (Table XII), but c o r r e l a t i o n s were weak. In 1983, hatching survival and the number of young fledged were negatively correlated with brood density at hatching. Density dependent brood survival has been reported before in mallards (Titman and Lowther 1975, H i l l 1984) and shelducks (Makepeace and Patterson 1980, Pienkowski and Evans 1982). A density dependent predation rate onducklings could explain this pattern. However, neither Erskine (1972) nor ,1 found much evidence of predation on bufflehead ducklings. A l t e r n a t i v e l y , density dependent survival could result from i n t r a s p e c i f i c aggressions between broods. As mentioned e a r l i e r , the frequency of aggressive interactions increased with brood density. Brooding females sometimes attack ducklings of neighboring broods. Twice I saw ducklings (one bufflehead and one goldeneye) k i l l e d by attacking females, and similar k i l l i n g s or i n j u r i e s of ducklings have been reported in Barrow's goldeneye (Savard 1982a, pers. comm.) and shelducks (Makepeace and Patterson 1980). Newly-hatched broods that have not yet set t l e d on a t e r r i t o r y are l i k e l y to be e s p e c i a l l y susceptible to these attacks. 94 To summarize, high brood density was associated with depressed growth rate for ducklings, a lower su r v i v a l among young broods, and a reduction in the number of young fledged, probably as a result of intense aggressive interactions. However, duckling survival was also higher on t e r r i t o r i e s with high food density during a year of poor weather. Therefore, despite the costs associated with high brood density, there are some net benefits for females that defend a t e r r i t o r y in high food and brood density areas. The adaptive value of brood t e r r i t o r i e s My results are generally consistent with the hypothesis that securing a food reserve for the ducklings i s an important function of brood t e r r i t o r i e s . Brood density was correlated with food density, and overland movements of newly-hatched broods were generally toward ponds of higher food density, thus suggesting that these ponds were the preferred habitat. Furthermore, brooding females adjusted t e r r i t o r y size to maintain food supply in the t e r r i t o r y . Prevention of brood-mixing at high brood density could also be an important function of brood t e r r i t o r i e s . This assumes that brood-mixing i s costly to females, for example in terms of reduced growth rate or survival of ducklings. However, neither growth nor survival appeared to be affected by brood size and, on the contrary, large broods tended to survive better, as also found by Eadie and Lumsden (1985) in common goldeneye. Therefore, my results did not support t h i s hypothesis. 95 Brooding females select a t e r r i t o r y of their own a few days after hatching, and only about 40 % of them used the t e r r i t o r y (or part of i t ) defended by their mate e a r l i e r in the season. Shortly before hatching, females were observed to feed away from the t e r r i t o r y defended by their mate during the nesting period, sometimes even on d i f f e r e n t ponds. These females may have been sampling other areas in order to select an "optimal" brood t e r r i t o r y with respect to brood density (cost) and food abundance (benefit). Thus, selection of a t e r r i t o r y by female buffleheads may be an example of the ideal free d i s t r i b u t i o n as suggested by Fretwell (1972). This may explain why, o v e r a l l , duckling s u r v i v a l , growth rate and time budget were only weakly affected by food and brood density. Adjustments in t e r r i t o r y size by females may buffer the effects of food and brood density on ducklings. If brood t e r r i t o r i e s are adaptive to buffleheads, why are they so rare in other ducks? This may be related to the ephemeral brood-rearing habitat of many duck species. Brown's (1964) concept of economic defendability asserts that feeding t e r r i t o r i e s should evolve only i f food supply is predictable. For most dabbling ducks, seasonal and semipermanent wetlands are the optimum brood-rearing habitat (Talent et a l . 1982, Duebbert and Frank 1984) and, not s u r p r i s i n g l y , they do not defend brood t e r r i t o r i e s (Haland 1983). Conversely, species reported to defend brood t e r r i t o r i e s l i k e buffleheads, goldeneyes, shelducks, African black ducks, and steamer ducks occupy more predictable and stable environments such as estuaries, r i v e r s or 96 permanent wetlands (McKinney et a_l. 1978, Savard 1982a, 1984, Patterson 1982, Livezey and Humphrey 1985, this study). Further studies of spacing behavior in waterfowl broods are needed to c r i t i c a l l y evaluate t h i s hypothesis. 97 CHAPTER IV: THE ROLE OF TERRITORIAL BEHAVIOR AND NEST SITE AVAILABILITY IN LIMITING BREEDING DENSITY 9 8 INTRODUCTION The role of t e r r i t o r i a l behavior in l i m i t i n g population size has been discussed extensively (e.g. Wynne-Edwards 1962, Lack 1966, Brown 1969, Watson and Moss 1970, Patterson 1980). The best evidence that breeding densities can be limited by t e r r i t o r i a l behavior comes from studies of passerine birds (Kluyver and Tinbergen 1953, Krebs 1971) and grouse (Watson and Jenkins 1968, Watson and Moss 1970, 1980, Hannon 1983). In waterfowl, Dzubin (1969) and Patterson (1976) suggested that spacing behavior can l i m i t the size of breeding populations, although they did not s p e c i f i c a l l y refer to t e r r i t o r i a l behavior. Ducks vary greatly in their degree of t e r r i t o r i a l i t y (McKinney 1973, Titman and Seymour 1981, see also chapter 2): some species l i k e p i n t a i l s have no defended areas during the breeding season while others l i k e buffleheads are highly t e r r i t o r i a l . The three species of the genus Bucephala are probably the most t e r r i t o r i a l species of ducks in North America (Donaghey 1975, Savard 1982a, 1984). Population studies of other t e r r i t o r i a l species l i k e shelducks in England and Australia and African black ducks suggested that t e r r i t o r i a l behavior during the breeding season was l i m i t i n g population size (Riggert 1977, B a l l et a l . 1978, Evans and Pienkowski 1982, Patterson et a l . 1983). I therefore hypothesized that t e r r i t o r i a l behavior could l i m i t breeding density in buffleheads. Buffleheads are also among the few species of North American ducks to nest in tree c a v i t i e s . The a v a i l a b i l i t y of 99 natural c a v i t i e s has been shown to l i m i t breeding density of other cavity-nesting ducks l i k e wood ducks (McLaughlin and Grice 1952, Bellrose "et a l . 1964, Doty and Kruse 1972) and common goldeneyes (Eriksson 1982, Dennis and Dow 1984). I therefore hypothesized that a shortage of natural c a v i t i e s could also l i m i t breeding density in buffleheads. This study attempts to test whether t e r r i t o r i a l behavior or nest s i t e a v a i l a b i l i t y i s an important l i m i t i n g factor in breeding buffleheads. If nest s i t e s are l i m i t i n g , I predict that (1) suitable c a v i t i e s should be in short supply, and (2) the addition of a r t i f i c i a l nest s i t e s should increase the breeding population. If t e r r i t o r i a l behavior i s l i m i t i n g , I predict that (1) the breeding population should remain stable following the addition of nest s i t e s , and (2) when t e r r i t o r i a l pairs are removed, new pairs should s e t t l e and attempt to breed. 100 METHODS This study was conducted in the Cariboo Parkland of B r i t i s h Columbia, 15 km north of 100 Mile House. The study area of 23 km2 included 26 ponds. Much of the coniferous forest was regenerating following small scale logging over the past 50 years. More d e t a i l s of the study area are given in chapter 2. Sampling of natural c a v i t i e s Natural c a v i t i e s were found by intensively searching wooded areas around the ponds. A l l trees large enough to harbor a cavity were c a r e f u l l y inspected. Only c a v i t i e s > 5 cm in entrance diameter (about the minimum size of a f l i c k e r (Colaptes  auratus) cavity, McLaren 1963) were recorded. The radius of the area searched around the pond varied from 100 to 250 m depending on the topography and the type of vegetation. This radius was generally larger for ponds surrounded by rangeland or open forest, than for ponds surrounded by closed forest. The l i m i t of the searched area and the locations of a l l c a v i t i e s found were plotted on large scale maps (1:4000) drawn from a e r i a l photographs. For each pond, the t o t a l area searched and the proportion of rangeland, conifer, and aspen woodland within t h i s area were calculated using a d i g i t i z e r . I characterized the population of natural c a v i t i e s as follows. A sample of fiv e ponds located in the aspen parkland and five in the coniferous forest were randomly selected. For a l l c a v i t i e s found around these ponds, I measured: the average entrance diameter of the cavity (1/2 horizontal + 1/2 v e r t i c a l 101 diameter), the depth of the cavity from the bottom edge of the entrance to the fl o o r , the average diameter of the cavity (1/2 breadth + 1/2 width), and the distance to water. The breadth was measured from the inside edge of the entrance to the back wall and the width between the two side walls at the entrance l e v e l . Addition and monitoring of nest boxes At the start of t h i s study, 35 nest boxes were already present around some of the ponds. These boxes, however, were unattractive to buffleheads and only one female attempted to breed in them over a 4-year period (Gauthier, in prep.). I therefore considered that there were no a r t i f i c i a l nest sites available to buffleheads in the f i r s t year .(1982). Suitable boxes were added late in 1982 (N = 120), in 1983 (N = 60) and in 1984 (N = 20). Three sizes of boxes were provided (floor dimensions: 15 x 15 cm, 19 x 19 cm, 23 x 23 cm), a l l of them with a 6.5 cm entrance diameter and a depth of 30 cm. Nest boxes were spaced out along the shoreline of ponds, most of them within 15 m of the water. Most ponds had two sizes of boxes, and many'had a l l three siz e s . Buffleheads preferred the smallest size of box, but also used the two larger sizes (Gauthier, in prep.) By the end of 1983, 23 of the 26 ponds were saturated with nest boxes (from 4 to 20 per pond) by providing them in excess to the maximum density of breeding pairs recorded on each pond. Although other species including the European s t a r l i n g (Sturnus vulgaris) also used the boxes, at least 40 % 1 02 of them were unused each year. Thus, competition with other species should not have prevented buffleheads from using the boxes. Bufflehead nests were found by checking a l l natural c a v i t i e s and nest boxes of the study area, or by following incubating females f l y i n g to their nests. A nest s i t e was deemed to be used by buffleheads i f at least one egg was l a i d in that s i t e . Nesting success was defined as being the percentage of nests where at least one egg hatched successfully. Females were trapped on the nest after mid-incubation, and were i n d i v i d u a l l y marked with color-coded nasal saddles (Doty and Greenwood 1974). A l l natural c a v i t i e s used by buffleheads were also measured, and the results for the e a r l i e r years (1982-1983) have been reported elsewhere (Peterson and Gauthier 1985). . Pair and brood counts I estimated the t o t a l population of buffleheads breeding on my study area by conducting regular pair counts. Two counts were made in 1982, f i v e in 1983, four in f984 and one in 1985. In 1982, because only 19 of the 26 ponds were surveyed, the t o t a l breeding population was estimated by multiplying the number of pairs counted by the r a t i o t o t a l area of a l l ponds/area of the surveyed ponds in that year. From 1983 to 1985, surveys were completed within one or two days. In 1982, the surveys covered four days. This should have had a minimal effect on the count because breeding pairs rarely leave their t e r r i t o r i e s (Donaghey 1975). These counts were mostly conducted between 08:00 and 103 12:00 h (PDT) from elevated vantage points along the shore. The t o t a l number of pairs, lone males and lone females was taken as an index of breeding pair density, as suggested by Savard (1981). Group of birds (> 2) were considered to be non-breeders. Weekly brood counts of the whole study area were also conducted between early June and the end of July from 1982 to 1984. Methods were similar to the pair counts. At least f i v e counts were conducted each summer and a l l broods that reached the water were probably seen at least once. In 1985, only two broods counts were conducted in mid-July when a l l broods were already at least two weeks old. Removal experiment In 1984, seven t e r r i t o r i a l males, a l l of them from non-contiguous t e r r i t o r i e s , were removed during the egg-laying stage of their mates. The general methods employed in t h i s experiment are described in chapter 2. S t a t i s t i c a l tests were two-tailed, unless otherwise sp e c i f i e d . 104 RESULTS Use of natural c a v i t i e s by buffleheads To assess the a v a i l a b i l i t y of natural c a v i t i e s to buffleheads, I had f i r s t to determine what i s a suitable nest s i t e for them. In the Cariboo Parkland, the cavity dimensions have an important influence on the s u i t a b i l i t y of natural s i t e s for a species (Peterson and Gauthier 1985). I therefore tested i f buffleheads were selec t i v e in their use of c a v i t i e s by comparing the dimensions of c a v i t i e s used with the dimensions of a l l measured c a v i t i e s . The measurements of bufflehead c a v i t i e s used here included those reported by Peterson and Gauthier (1985), and 16 other c a v i t i e s measured in 1984. The d i s t r i b u t i o n of the average entrance diameter of bufflehead c a v i t i e s was not s i g n i f i c a n t l y d i f f e r e n t from the d i s t r i b u t i o n of sampled c a v i t i e s (Kolmogorov-Smirnoff (K-S), P > 0.1; F i g . 13A). The d i s t r i b u t i o n of bufflehead c a v i t i e s was, however, truncated to the l e f t as c a v i t i e s smaller than 6.0 cm in average diameter were not used. The d i s t r i b u t i o n of the depths of bufflehead c a v i t i e s were s i g n i f i c a n t l y d i f f e r e n t from the d i s t r i b u t i o n of sampled c a v i t i e s (K-S, P = 0.02; F i g . 13B), as they showed a preference to use deeper c a v i t i e s . There was no differences between the two d i s t r i b u t i o n s for the average diameter of the cavity (K-S, P > 0.1; F i g . 13C). F i g . 13 also shows that d i s t r i b u t i o n s of the sampled c a v i t i e s tended to be skewed to the l e f t . This i s because some of the larger c a v i t i e s were excavated by pil e a t e d 105 Figure 13. D i s t r i b u t i o n of cavity measurements for the sampled c a v i t i e s (open bars), and for the c a v i t i e s used by buffleheads (hatched bars) in the Cariboo Parkland. (A) average entrance diameter, (B) cavity depth, (C) average cavity diameter. 106 Depth 1 07 woodpeckers (Dryocopus p i l e a t u s ) . F i n a l l y , the distances to water of bufflehead c a v i t i e s were s i g n i f i c a n t l y shorter than those sampled (K-S, P = 0.025; F i g . 14). Eighty-eight percent of the nests were within 100 m of the water, although some were as far as 270 m. Av a i l a b i 1 i t y of natural c a v i t i e s The a v a i l a b i l i t y of suitable c a v i t i e s to buffleheads was determined using their nest s i t e preference. A s i t e was deemed to be suitable i f a l l three cavity measurements f e l l within the 95 % confidence interval of the average dimensions of c a v i t i e s used by that species. The ov e r a l l a v a i l a b i l i t y of suitable nest s i t e s was 2.34 ± 0.61(SE) cavities/ha of woodland (N = 10 ponds). A l l c a v i t i e s found in the parkland, and most of those found in the conifer forest, were in aspen trees, even though aspen were sparse in coniferous forest. The density of suitable c a v i t i e s , however, was five times higher in the aspen parkland than in the coniferous forest ( t - t e s t approx., t' = 4.56, P = 0.01; Table XIII). Although more of the parkland habitat was non-wooded ( i . e . rangeland), c a v i t i e s remained more abundant in the parkland when density was expressed per unit of t o t a l area (t = 2.62, P < 0.05; Table XIII). Furthermore, t h i s i s a conservative test because I searched, on average, farther from the water in the parkland than in the conifer (Table XIII). As the abundance of c a v i t i e s decreased with distance to water (Fig. 14), the number of c a v i t i e s found in the parkland may be 1 08 Figure 14. D i s t r i b u t i o n of distance to water for the sampled c a v i t i e s (open bars), and for the c a v i t i e s used by buffleheads (hatched bars) in the Cariboo Parkland. 0 20 40 60 80 100 Distance to Water | | N = 123 N = 41 120 140 160 180 >200 (m) o 1 10 Table XIII. A v a i l a b i l i t y of natural c a v i t i e s to buffleheads in two habitats of the Cariboo Parkland. (Mean ± SE) Aspen parkland Coniferous forest N of ponds 5 5 Searched area (ha) 24.2±2.9 11.4±2.5 % of forest in the 24 % 61 % searched area Average radius of the 171±20 105±16 . searched area (m.) Total Number of c a v i t i e s 19.2±1.2 4.8±0.5 found Density of c a v i t i e s per ha 3.91±0.67 0.78±0.17 (wooded area only) Density of c a v i t i e s per ha 0.82±0.08 ( t o t a l area) 0.50±0.09 111 s l i g h t l y biased downward. Although c a v i t i e s were more abundant in the parkland habitat, cavity dimensions were similar in both habitats (K-S, P > 0.1 for d i s t r i b u t i o n s of a l l cavity measurements and distance to water). Using the density of c a v i t i e s and the average area of ponds in my study area (5.20 ± 0.86 ha, N = 26), I estimated the t o t a l number of c a v i t i e s available" to buffleheads within 100 m of water. If a l l the study area was in the aspen parkland, about 240 c a v i t i e s would be available, while i f the study area was e n t i r e l y coniferous, 146 would be available. Use of nest boxes •In 1983, the f i r s t year that suitable nest boxes were available, 9 of them were used by buffleheads. This was followed by a large increase in box use in 1984, and by a s l i g h t decline in 1985 (Table XIV). The low proportion of boxes used by buffleheads (7-13 %) was expected because a l l ponds were provided with a superabundance of boxes. The rate of use, however, varied s i g n i f i c a n t l y among habitats (Table XV), being higher in the coniferous forest (16.4 %) than in the aspen parkland (6.8 % ) . The nesting success of buffleheads using boxes varied widely among years, with desertion accounting for most of the losses (Table XIV). Nest losses were maximum in 1984, the second year that boxes were available, with 52 % of the nests deserted. In 1984, 17 nest-boxes were also used for the f i r s t time (twice the number of 1983), including three of them by known f i r s t time 1 1 2 Table XIV. Nesting success of buffleheads using nest boxes. 1983 1984 1985 Number of boxes available 121 187 200 Number of boxes used 9 26 21 Proportion of nests deserted 11 % 52 % 38 % Proportion of nests depredated 0 % 16 % 14 % Proportion of nests hatching eggs 89 % 32 % 48 % 113 Table XV. Number of nest boxes available and used by buffleheads according to the habitat. Aspen parkland Coniferous forest Boxes available 292 216 Boxes Used 20 36 1 1 4 breeders (females banded as ducklings). In goldeneyes, new nest boxes have been reported to be used mostly by inexperienced breeders (Eriksson 1982, Savard 1982b). Dow and Fredga (1984) also showed that f i r s t time breeders have a lower nesting success in that species. Therefore, the low nesting success observed in 1984 may have resulted from a high rate of nest box use by inexperienced breeders. In 1984, desertion rate was p o s i t i v e l y correlated with density of t e r r i t o r i a l pairs (Spearman's r = 0.63, P = 0.01, N = 13 ponds), but was not in 1983 nor in 1985 (both years, P > 0.1; 1983, N = 11; 1985, N = 12). This density-dependent e f f e c t could also be associated with a high proportion of f i r s t time breeders using nest boxes in 1984. If this i s true, the density-dependent effect should be much stronger on " f i r s t time users of boxes (presumably mostly young birds) than in birds reusing the same nest box or birds nesting in natural c a v i t i e s . I found that, among f i r s t time users of boxes in 1984, the density of t e r r i t o r i a l pairs was indeed s i g n i f i c a n t l y higher.on ponds where females deserted (Table XVI), but not among females reusing boxes or using natural c a v i t i e s , although the trend was si m i l a r . In 1985, there was no s i g n i f i c a n t difference in density of pairs between successful and f a i l e d breeders for either group (Table XVI) . To sum up, nest boxes received a substantial use in the second and t h i r d year after erection. In the second year, however, rate of desertion was high and density dependent among f i r s t time users of boxes, which suggests a high rate of use by 1 1 5 Table XVI. Density of t e r r i t o r i a l pairs on ponds where females successfully hatched a clutch and on those where they deserted. In 1984 and 1985, females were divided into f i r s t time users of nest boxes ("new") and into those reusing boxes or using natural c a v i t i e s ("old"). Comparisons (t test) are between successful breeders and those that deserted (Mean ± SE, N). Group Density per ha of water area Successful Deserted km of shoreline Successful Deserted 1.983 a l l 1.00±0.11 NS 0.94±0.23 9 4 3.94±0.49 NS 3.76±1.04 9 4 1 984 a l l 0.78±0.10 ** 1.22±0.06 3.30±0.37 18 14 18 * 4.33±0.24 1 4 new 0.63±0.16 ** 1.20±0.09 2.75±0.57 5 10 5 * 4.18±0.32 1 0 old 0.84±0.13 NS 1.24+0.03 1 3 4 3.5110.46 NS 4.68±0.27 13 4 1 985 a l l 0.74±0.08 NS 0.7810.10 22 9 3.50±0.39 NS 3.69+0.65 22 9 new 0.53+0.16 NS 0.72±0.15 7 5 2.43+0.78 NS 3.51+0.99 7 5 old 0.83+0.08 NS 0.8710.17 15 4 4.0010.40 NS 3.9010.94 15 4 NS P > 0.05, * P < 0.05, ** P < 0.01 1 1 6 f i r s t time breeders. Size of the breeding population Using the l a s t pair count in May (the only one for which I have data for a l l 4 years), the number of breeding pairs increased s l i g h t l y from 60 in 1982 and 59 in 1983 to 63 in 1984 and 65 in 1985 (Fig. 15). The repeated counts of 1983 and 1984 showed a decrease throughout the season in the number of breeding pairs (Fig. 15). This decline occurred mostly in the la s t week of A p r i l and f i r s t week of May, and was probably due to the presence of migrant pairs passing through the study area. If I use the three surveys conducted in May of 1983 and 1984 (Fig. 15), the increase in number of pairs in the second year approached significance (paired t - t e s t , t=3.0, 0.05 < P < 0.1). I also examined i f the t o t a l number of nests i n i t i a t e d increased over the same period. Several nests were not found each year, and the number of unmarked broods was used to estimate their number. From 1982 to 1984, 62.3 % of the nests found in natural c a v i t i e s (N = 57) produced broods that were seen at least once on the water (range: 58.3 % to 63.6 % over three years). In 1985, however, brood counts were conducted only in mid-July when broods were at least 2 weeks of age. Therefore, t o t a l brood losses from a r r i v a l on the water to 2 weeks of age had also to be accounted for in that year. From 1982 to 1984, 77.7 % of the broods from natural c a v i t i e s that reached water survived to 2 weeks of age (range: 71.4 % to 86.7 % ) . Using these conversion factors, I estimated the t o t a l number of nests 1 1 7 Figure 15. Total number of bufflehead breeding pairs using the study area from 1982 to 1985 according to the date of the season. N = 26 ponds except for 1982, N = 19 ponds. Number of breeding pairs was estimated for 26 ponds in that year (see methods). 118 s j j B d , 6 u ! p 9 9 j g jo jeqiunN 119 i n i t i a t e d in the study area. Table XVII shows that the t o t a l number of nests was stable in 1982 and 1983 but may have increased by about 10 % in 1984 and 1985. To sum up, from 1982 to 1985, the number of bufflehead pairs appears to have remained stable. The t o t a l number of nests i n i t i a t e d , however, may have increased s l i g h t l y after the addition of nest boxes. Removal experiment If t e r r i t o r i a l behavior prevents some pairs from breeding, I predicted that new pairs should s e t t l e following the removal. Seven male buffleheads were removed between 4 May and 21 May. The mean star t of egg-laying • for that year was 5 May, and therefore removals occurred after more than half, of the nests were already i n i t i a t e d . In three of the seven removals, neighboring males extended their t e r r i t o r i e s . Replacements occurred in the remaining four removals: in three cases, a new male, apparently unmated before the removal, mated with the widowed female, and in the other case the female was evicted from her t e r r i t o r y by a new p a i r . In the l a t t e r case, the new pair was not marked and i t s o r i g i n was unknown, although t h i s pair had not been a resident on that pond before because the other t e r r i t o r i e s remained stable after the removal. This pair presumably nested successfully because an unmarked brood later showed up in the same area of the pond. 120 Table XVII. Estimates of the number of nests i n i t i a t e d by buffleheads in the study area. Number of nests not found was estimated by: (number of unmarked broods x 0.623) from 1982 to 1984 and (number of unmarked brood x 0.484) for 1985 (see r e s u l t s ) . 1982 1983 1984 1985 Total number of broods Number of unmarked broods Number of nests not found 40 33 53 40 19 30 33 1 7 27 41 18 38 Number of nests found in c a v i t i e s 11 24 18 1 1 Total number of nests in c a v i t i e s 64 54 45 49 Number of nests in boxes 26 1 9 Total number of nests 64 63 71 68 a Two of the 21 boxes used in 1985 were located on a pond that was not surveyed for broods, .and these boxes were therefore excluded from t h i s analysis 121 DISCUSSION Nest s i t e a v a i l a b i l i t y The f i r s t prediction of the nest s i t e l i m i t a t i o n hypothesis is that c a v i t i e s should be in short supply. I estimated that 146 to 240 suitable c a v i t i e s were available on the study area, but how accurate was this estimate? The 95 % confidence interval of the dimensions of c a v i t i e s used by buffleheads should presumably give a representative sample of c a v i t i e s suitable to them. However, there was several possible sources of error. F i r s t , I could have missed some c a v i t i e s , but this would make the number of c a v i t i e s available a conservative estimate with respect to the prediction. Second, I used .a distance of 100 m from water ,to determine the number of c a v i t i e s available. If c a v i t i e s near water are not available, buffleheads can use nest s i t e s quite far from water (Erskine 1972, Donaghey 1975). However, in my area c a v i t i e s were more abundant near water than farther away (Fig. 14), possibly because th i s mesic environment produces more suitable trees for excavation (Conner e_t al. 1975). Therefore, although the distance of 100 m included 88 % of the c a v i t i e s used by buffleheads, i t also tends to make the number of ca v i t i e s available a conservative estimate. The t o t a l number of c a v i t i e s estimated for the study area (146-240) was much higher than the number of bufflehead pairs breeding in the same area (59-81; F i g . 15). Even in the coniferous habitat where much of the forest had been logged in the past 50 years, the number of c a v i t i e s estimated (146) was 1 22 r e l a t i v e l y high. Obviously, buffleheads are not the sole users of c a v i t i e s , and competition with other species may reduce the a v a i l a b i l i t y of nest s i t e s (Erskine 1959, 1960, 1964, McLaren 1969). St a r l i n g s , for instance, are the dominant users of c a v i t i e s in the aspen parkland and overlap extensively with buffleheads in the type of cavity used in t h i s area (Peterson and Gauthier 1985). However, in that study, a large proportion of c a v i t i e s (43 %) were s t i l l unused. The o v e r a l l density of c a v i t i e s found in t h i s study (2.3 cavities/ha of woodland) i s high compared to other studies. In wood ducks, Bellrose (1976) reported densities of 0.10 to 0.71 suitable cavities/ha. Buffleheads, but not the larger wood duck, use c a v i t i e s of the r e l a t i v e l y abundant f l i c k e r , and t h i s probably accounts for t h i s d i f f e r e n c e . Prince (1968) reported that in a v i r g i n floodplain hardwood forest, the highest density of suitable nest sites for wood ducks was 5.5 cavities/ha. This is only s l i g h t l y higher than the density I found in the aspen parkland (3.9 c a v i t i e s / h a ) . Therefore, a l l the evidence indicates that, in the Cariboo Parkland, suitable c a v i t i e s are not in short supply for buffleheads. 1 23 Are nest s i t e s limiting? The second prediction of the nest s i t e l i m i t a t i o n hypothesis i s that breeding density should increase following the addition of a r t i f i c i a l nest s i t e s . There was no control on my study area for the breeding density of buffleheads during t h i s experiment. However, during a study on Barrow's goldeneyes conducted 85 km northwest of t h i s study, the bufflehead population remained stable between 1982 and 1985 (Savard, in prep.). Buffleheads were quick to use nest boxes: in the second year that they were avail a b l e , 37 % of a l l nests i n i t i a t e d were in boxes. There was also a decrease in the number of natural s i t e s used during the same period, however (Table XVII). The use of boxes was therefore mostly compensatory: females preferred boxes over natural s i t e s , but many of these birds would presumably have nested in natural c a v i t i e s in the absence of boxes. The compensation, however, was not complete since the t o t a l number of nest i n i t i a t e d apparently increased (Table XVII). Although more nests may have been i n i t i a t e d , there was only a s l i g h t increase in the number of breeding pairs following the addition of boxes (Fig. 15). Further, i t i s d i f f i c u l t to determine i f t h i s increase was real or just due to the v a r i a b i l i t y of the pair counts. In wood ducks and goldeneyes, a two-fold increase in breeding density following the addition of boxes i s not uncommon (Savard 1982b, in prep., Dennis and Dow 1984, Haramis and Thompson 1985). This obviously did not occur here. The t o t a l number of broods did not increase either (Table 124 XVII). Therefore, I conclude that the addition of nest boxes did not s i g n i f i c a n t l y increase the density of breeding pairs in buffleheads. Since both predictions of the nest s i t e l i m i t a t i o n hypothesis were rejected, I also conclude that the abundance of c a v i t i e s does not l i m i t breeding density of buffleheads in the Cariboo Parkland. Is t e r r i t o r i a l behavior limiting? The s t a b i l i t y of the population following the addition of nest boxes is consistent with the t e r r i t o r i a l behavior l i m i t a t i o n hypothesis. However, while the number of pairs remained r e l a t i v e l y stable, the number of nests initiated-may have increased (e.g. from 63 to 71 between 1983 and 1984). This suggests that some of the pairs that established a t e r r i t o r y in 1983 did not nest because they lacked suitable nest s i t e s . This was confirmed in one instance, when the nesting tree of a female who bred successfully in 1982 f e l l before the 1983 season. In 1983, she and her mate secured a t e r r i t o r y on a neighboring pond but no nest was found and her behavior indicated that she did not breed. She came back in 1984 and nested in a nest box on a t h i r d pond. These results suggest that, in buffleheads, t e r r i t o r i a l behavior can prevent some pairs from breeding partly by creating an a r t i f i c i a l shortage of nest s i t e s . Although buffleheads generally defend only the water area adjacent to their nests, (chapter 2) they successfully exclude other pairs from that 125 area, and presumably prevent them from using other c a v i t i e s near the t e r r i t o r y (Donaghey 1975, chapter 2). V i l l a g e (1983) recently proposed a similar mechanism to explain population l i m i t a t i o n in the European kestrel (Falco tinnunculus). Numerous studies have shown that young birds are more l i k e l y to be prevented from breeding by t e r r i t o r i a l behavior than older birds (e.g. Hannon 1983, V i l l a g e 1983). Based on the low and density-dependent nesting success of f i r s t time users of boxes in 1984 (Table XIV, XVI), I suggested that new boxes were used mostly by young birds (see r e s u l t s ) . Therefore, boxes were probably most pr o f i t a b l e for young birds, and some of them may have been prevented from breeding without the addition of boxes. If so, why were the nest boxes substantially used only in their second- year (Table XIV)? First-time breeders apparently use si t e s that they have v i s i t e d the year before (Eadie and Gauthier 1985), and this may explain why few boxes were used and nesting success was high in 1983. The few females that used boxes in 1983 were probably experienced breeders. The second prediction of the t e r r i t o r i a l behavior l i m i t a t i o n hypothesis i s that new pairs should s e t t l e following the removal of t e r r i t o r i a l p a i r s . Four replacements occurred, three of them involving previously unmated males and only one involving a new pair. The l a t t e r replacement, however, occurred late in the season (most females had finished egg-laying) and i t is possible that t h i s pair would not have bred without the removal. Watson and Moss (1970) suggested that three conditions must 1 26 be met for t e r r i t o r i a l behavior to l i m i t populations: part of the population should not breed, the breeding population should not completely use some resources, and replacements should occur following removal of t e r r i t o r i a l i n d i v i d u a l s . The. f i r s t condition i s met for males as a large number of unmated adult males are present on the breeding grounds (Erskine 1972, Gauthier pers. observ.). However, i t i s uncertain whether some adult females do not breed, because I was unable to distinguish them from the non-breeding yearlings. My results also show that nest s i t e s are not in short supply and food i s not l i m i t i n g during the nesting period (chapter 2), thus f u l f i l l i n g the second condition. F i n a l l y , my results for the t h i r d condition are equivocal although some replacement -did occur following the removals. My results are therefore' consistent with the hypothesis that t e r r i t o r i a l behavior l i m i t s breeding density in buffleheads, although further experimental work i s needed to establish t h i s conclusively. Limitation of breeding populations in cavity-nesting ducks An interesting comparison can be drawn between wood ducks, buffleheads and Barrow's goldeneyes as a l l three are cavity nesters. The wood duck i s a n o n - t e r r i t o r i a l species (Grice and Rogers 1965) li m i t e d by the a v a i l a b i l i t y of nest s i t e s (McLaughlin and Grice 1952, Bellrose et a l . 1964). New nest boxes are quickly used, and i n s t a l l a t i o n of large number of boxes lead to a rapid population increase. Breeding populations 127 often exceed the a v a i l a b i l i t y of nest s i t e s , and increased levels of dump nesting, nest desertion, and a sharp decline in productivity result (Jones and Leopold 1967., Clawson et a l . 1979, Haramis and Thompson 1985). Jones and Leopold (1967) stated "lack of t e r r i t o r i a l defense of the nest s i t e by an established pair was primarily responsible for the i n e f f i c i e n c y of nesting in a high density population." My results show that nest boxes also received substantial use by buffleheads but, contrary to the case of wood ducks, my population remained f a i r l y stable. Nest parasitism did not increase and there was no evidence of nesting interference. Although nesting success of females using boxes was low in the second year that boxes were available, t h i s was probably because they were used by inexperienced breeders. The d i f f e r e n t responses of these two species further suggests that t e r r i t o r i a l behavior l i m i t s breeding density in buffleheads. F i n a l l y , Barrow's goldeneyes are, l i k e wood ducks, limited by the a v a i l a b i l i t y of nest s i t e s (Savard 1982b, in prep.), but they are also strongly t e r r i t o r i a l (Savard 1982a, 1984). Addition of nest boxes to goldeneye populations leads to an increase in both breeding density and, to a lesser extent, in nest parasitism, although t h i s does not decrease o v e r a l l productivity (Savard in prep.). Therefore, when nest s i t e l i m i t a t i o n i s removed in t h i s species, breeding density may increase to a l e v e l at which t e r r i t o r i a l behavior i s l i m i t i n g , thus preventing the kind of nest interference seen in wood ducks. 1 28 T e r r i t o r i a l behavior has been proposed to l i m i t breeding populations in other t e r r i t o r i a l ducks (Riggert 1977, Evans and Pienkowski 1982, Patterson et a l . 1983). My results suggest that t h i s i s also true in buffleheads. The Cariboo Parkland, however, is a prime habitat for cavity-nesting birds (McLaren 1963, t h i s study), unlike many other parts of the bufflehead breeding range (Erskine 1972). It therefore remains to be seen i f the absence of nest s i t e l i m i t a t i o n i s unique to t h i s area or widespread in buf fleheads. 129 CHAPTER V: TERRITORIAL BEHAVIOR IN DUCKS: A REVIEW AND A MODEL The occurrence of t e r r i t o r i a l behavior in ducks The v a r i a b i l i t y in soc i a l organization and t e r r i t o r i a l system in ducks (see chapter 2) has intrigued b i o l o g i s t s for a long time. McKinney (1965) was the f i r s t to point out differences in the ecology of t e r r i t o r i a l and n o n - t e r r i t o r i a l species by his c l a s s i c a l comparison of shovelers and p i n t a i l s . Shovelers are t e r r i t o r i a l and they use a ri c h food supply in permanent ponds, whereas p i n t a i l s are not t e r r i t o r i a l and they use a variable food supply in temporary ponds (McKinney 1973, 1975). Based on Brown's (1964) model, McKinney hypothesized that the degree of t e r r i t o r i a l i t y is function of the defendability of the food resources. The degree of v a r i a b i l i t y ( i . e . ephemeral vs permanent) of the food resources is the key point of McKinney's hypothesis. In chapter 2, I suggested a c l a s s i f i c a t i o n of the types of t e r r i t o r i a l systems in ducks based mostly on studies of North American species. Here I extend th i s c l a s s i f i c a t i o n to the whole sub-family Anatinae, and evaluate the hypothesis of a relati o n s h i p between the occurrence of t e r r i t o r i a l i t y and environment v a r i a b i l i t y . The information available i s s u f f i c i e n t to determine the type of t e r r i t o r i a l behavior exhibited by 69 of the 103 species of Anatinae (sheldgeese excluded). However, t e r r i t o r i a l status cannot be assigned with the same degree of confidence in a l l cases. The t e r i t o r i a l status and the preferred 130 habitat of each species are presented in Appendix I. Habitats were divided into permanent (estuaries, coast, r i v e r s , streams (but not floodplains), and permanent marshes, ponds and lakes) and temporary ones (floodplain and seasonal ponds and lakes). A l l but two t e r r i t o r i a l (type III + IV) species occur in permanent habitats such as deep lakes and estuaries (e.g. Tadorna sp.), coastal regions (e.g. Tachyeres sp.), mountain ri v e r s (e.g. Anas sparsa, Merganetta armata, etc.), and permanent ponds (e.g. Bucephala sp.; see Appendix I ) . The two t e r r i t o r i a l species that use temporary habitats are Sarkidiornis melanotos (a polygynous species, S i e g f r i e d 1979) and Aix g a l e r i c u l a t a , two perching ducks. Although no other species occurring in temporary habitats are t e r r i t o r i a l , many species using, permanent habitats also are non- or weakly t e r r i t o r i a l (type I + I I ) . For instance, t e r r i t o r i e s are conspicuously absent in the trib e Aythyini. Overall, 33 of the 35 t e r r i t o r i a l species occur in permanent habitats whereas only 20 of the 32 n o n - t e r r i t o r i a l species use similar habitats (P < 0.01, Fisher exact t e s t ) . Such a. test i s obviously preliminary, as this kind of review is always subject to many biases. For instance, the occurrence of t e r r i t o r i a l i t y i s s t i l l uncertain in some species (see Appendix I ) . Further, the number of t e r r i t o r i a l species may be overestimated in t h i s sample, because t e r r i t o r i a l displays are usually conspicuous and thus more l i k e l y to be reported in the l i t e r a t u r e . However, there appears to be some support for McKinney's (1973, 1975) contention that t e r r i t o r i a l i t y i s 131 associated with permanent habitats. By arguing the importance of a defendable (sensu Brown 1964) food supply in the evolution of t e r r i t o r i a l behavior, McKinney assumed that food was the major resource defended by t e r r i t o r i a l waterfowl. Several authors have suggested that food i s indeed a resource defended by ducks (Dwyer 1974, Seymour 1974a, Derricksson 1978, Nudds and Ankney 1982, Livezey and Humphrey 1985). However, the only two studies that have investigated the rel a t i o n s h i p between t e r r i t o r y and food abundance (Buxton i_n Patterson 1982, chapter 2 of t h i s study) found l i t t l e evidence to support the food hypothesis. On the other hand, there i s growing evidence that the female is a major resource defended by t e r r i t o r i a l males in most species of ducks (see chapter 2). Another important aspect of waterfowl s o c i a l organization is forced copulations. This behavior has been reported in over 25 % of a l l Anatidae, although i t appears to be much more common in weakly or n o n - t e r r i t o r i a l species than in t e r r i t o r i a l ones (McKinney et a l . 1983). There i s also evidence that forced copulations may be part of a mixed reproductive strategy of paired males in several s-pecies of waterfowl (Mineau and Cooke 1979, McKinney et a l . 1983, Afton 1985). Therefore, any model wishing to explain the v a r i a b i l i t y of s o c i a l organization and t e r r i t o r i a l system in ducks has to account for (1) the degree of s t a b i l i t y of the environment, (2) the fact that females rather than food appear to be the resource defended by males, and (3) the high incidence of forced copulations in waterfowl and i t s 132 inverse relationship with t e r r i t o r i a l i t y . In the following section I propose such a model. A model for the evolution of t e r r i t o r i a l behavior in ducks I hypothesize that v a r i a b i l i t y in the nesting success of females, and therefore in the expected fitness gain of paired males, determines whether males w i l l be t e r r i t o r i a l or not. This model can be shown graphically (Fig. 16). I define ESM(x) as the average expected reproductive success of paired males with their mate. ESM(x) i s d i r e c t l y related to the degree of mate attendance (Fig. 16A). The reason i s that, as a male spends more time away from his mate, she i s more susceptible to forced copulations, and his degree of paternity decreases. Furthermore, the slope o f E S M ( x ) is shallower in an ephemeral habitat (Fig. 16A), because of a higher p r o b a b i l i t y of nest f a i l u r e (e.g. because of drought). Males can also s i r e some offspring through forced copulations (Burns et a_l. 1980). I then define ESF(x), the average expected reproductive success of paired males with other females ( i . e . through forced copulations). This function i s inversely related to the degree of mate attendance (Fig. 16B), i. e . as a male spends more time away from his mate, he has more opportunities to engage in forced copulations. The slope of ESM(x) i s generally steeper than ESF(x) because the pro b a b i l i t y of f e r t i l i z a t i o n for males is higher with their mate than in forced copulations (Burns et a_l. 1980, Cheng et a l . 1983). The slope of ESF(x) is. also shallower in an ephemeral habitat for 133 Figure 16. A graphical model to explain the evolution of t e r r i t o r i a l behavior in ducks . • (A) Expected reproductive success of mated males with their mate, (B) Expected reproductive success of mated males through forced copulations, (C) Total expected reproductive success of mated males in a permanent habitat, (D) Total expected reproductive success of males in an ephemeral habitat. Forced copulation (ESF(x)) High Degree of mate attendance LOW High Degree of mate attendance LOW 136 the same reason as in ESM{x). The t o t a l expected reproductive success of males is therefore the summation of ESM(x) and ESF(x). In, a permanent habitat, the p r o b a b i l i t y for a male that his mate w i l l produce some offspring i s high in most years. Under these conditions, any gain in reproductive success obtained by leaving his mate and a c t i v e l y seeking forced copulations i s more than offset by the cost of being cuckolded. The t o t a l expected reproductive success of males (ESM(x) + ESF(x) on F i g . 16C) i s maximal i f they show a high degree of mate attendance, and therefore the model predicts the evolution of strong t e r r i t o r i a l i t y and the absence of forced copulations. This si t u a t i o n i s well i l l u s t r a t e d by buffleheads. In a highly ephemeral .habitat, both curves (ESM(x) and ESF(x)) are shallower (Fig. 16D). However, and t h i s i s the key point, the r a t i o of the slopes ESM(x)/ESF(x) decreases ( i . e . the slope of ESM(x) approaches the one of ESF(x)) as the habitat becomes more ephemeral. The rationale for this is that, in a sit u a t i o n where the r i s k of nest f a i l u r e is high, males that inseminate several females w i l l have, on average, a higher reproductive success ( i . e . chance of producing some offspring) than i f they inseminate only one female ( i . e . their mate). In other words, because the variance of male's expected gain with th e i r mate is very high in an ephemeral environment, d i v e r s i f y i n g t h e i r investment w i l l reduce t h i s variance. Rubenstein (1982) presented a good discussion of t h i s problem. Under these conditions, the t o t a l expected reproductive success 1 37 of males in r e l a t i o n to their degree of mate attendance passes by a maximum before declining (Fig. 16D). Therefore, the model predicts that males w i l l engage in forced copulations, show a low degree of mate attendance, and hence a loose t e r r i t o r i a l system w i l l evolve. This situation i s well i l l u s t r a t e d by p i n t a i l s . Several predictions can be made from t h i s model (Table XVIII). However, the data available are i n s u f f i c i e n t to test any of them. Furthermore, some of the f i e l d data required to test this model may be d i f f i c u l t to c o l l e c t (e.g. success of forced copulations). However, the model also predicts that, in monogamous ducks, t e r r i t o r i a l behavior should not evolve in a highly ephemeral habitat. Therefore, evidence for that w i l l argue against t h i s model. I have shown before that a l l monogamous species of ducks that are highly t e r r i t o r i a l are found in r e l a t i v e l y permanent habitats, although many other species that occur in similar environments are not t e r r i t o r i a l (e.g. most of the t r i b e Aythyini). This suggests that s t a b i l i t y of the environment is a necessary but not always s u f f i c i e n t condition to explain the evolution of t e r r i t o r i a l i t y . Nest predation and the p o s s i b i l i t y of renesting or of multiple broods are two factors that may also be important. For instance, i f nest predation i s very high, t h i s w i l l decrease the reproductive success even in permanent habitats, whereas high potential for renesting w i l l have the opposite e f f e c t . C learly, more data are needed, and the study of species using ephemeral habitats should be a promising area. 138 Table XVIII. Predictions of a model to explain the evolution of t e r r i t o r i a l behavior in ducks based on the degree of v a r i a b i l i t y (permanent vs ephemeral) of the environment during the breeding season. Permanent Ephemeral Reproductive success(RS) Variance in RS high low low high Forced copulations rare f requent Cuckoldry rare frequent Mate-guarding strong weak RS with mate(ESM(x)) vs by forced copulations(ESF(x)) ESM(x) > ESF(x) ESM(x) = ESF(x) a under these conditions, males should not only be opportunists but they should a c t i v e l y seek forced copulations. 1 39 CHAPTER VI: CONCLUDING REMARKS The adaptive significance of t e r r i t o r i a l behavior in birds is s t i l l controversial (Verner 1977, MacLean and Seastedt 1979, Weatherhead and Robertson 1980). In ducks, t e r r i t o r i a l behavior has been poorly studied u n t i l recently. Many species of ducks only defend a moving t e r r i t o r y around the female and several studies have suggested that the female i s the resource defended by males (e.g. Seymour and Titman 1978, Stewart and Titman 1980, Titman 1983). According to this hypothesis, t e r r i t o r i a l males protect their mates and insure their paternity. Some species of ducks l i k e buffleheads, however, are strongly t e r r i t o r i a l and defend a very s i t e - s p e c i f i c t e r r i t o r y . Donaghey (1975) f i r s t described t h i s in buffleheads, and he proposed that both males and females defend a t e r r i t o r y for i t s food value. Thus, he suggested that the adaptive significance of bufflehead t e r r i t o r i e s d i f f e r e d from most other ducks. This thesis has made three s i g n i f i c a n t contributions. First", I have challenged some of Donaghey's conclusions. My results provided l i t t l e support for the hypothesis that pair t e r r i t o r i e s are defended for their food value in buffleheads. Instead, they support the hypothesis that the t e r r i t o r y protects the female and provides her with an undisturbed feeding area, and also protects the nest s i t e . Thus, although buffleheads are more strongly t e r r i t o r i a l than most other species of North American ducks, the adaptive significance of t e r r i t o r i a l i t y appears to be similar to most other species. Based on t h i s conclusion, I have formulated a model to explain the v a r i a b i l i t y 1 40 of t e r r i t o r i a l behavior among ducks. This model i s consistent with available data but further work is needed to test i t rigorously. Second, I have shown that there are two di f f e r e n t types of t e r r i t o r i e s in breeding buffleheads. After hatching the brood, females start to defend a t e r r i t o r y which i s d i f f e r e n t from the t e r r i t o r y defended by males e a r l i e r in the season. My results suggest that t h i s t e r r i t o r y secures a food reserve for the ducklings, as previously proposed by Donaghey (1975). F i n a l l y , population l i m i t a t i o n by t e r r i t o r i a l behavior i s a controversial topic that has received considerable attention in the l i t e r a t u r e (e.g. Lack 1966, Brown 1969, Watson and Moss 1970, Patterson 1980). Although my evidence is broadly •consistent with the hypothesis that t e r r i t o r i a l behavior l i m i t s breeding density in buffleheads, the results of the removal experiment were equivocal. Future studies should attempt to conduct a more conclusive experiment by removing both nesting females and t e r r i t o r i a l males. The importance of mate-guarding as an explanation for bufflehead t e r r i t o r i a l i t y was unexpected, especially given Donaghey's (1975) conclusions. There has been growing evidence in the past decade that mate-guarding i s strongly associated with t e r r i t o r i a l behavior in birds in general (Birkhead 1979, Power and Doner 1980, Weatherhead and Robertson 1980, Dufty 1982, Bjorklund and Westam 1983). Future studies on the adaptive significance of t e r r i t o r i a l behavior in birds should not overlook the importance of mate-guarding. 141 LITERATURE CITED Afton, A.D. 1985. Forced copulation as a reproductive strategy of male lesser scaup: a f i e l d test of some predictions. Behaviour 92:146-167. Altmann, J. 1974. Observational study of behavior: sampling methods. Behaviour 49:227-267. Andersson, M. and M.O.G. Eriksson. 1982. Nest parasitism in goldeneyes Bucephala clangula: some evolutionary aspects, Am. Nat. 120:1-16. Armitage, K.B. 1974. Male behaviour and t e r r i t o r i a l i t y in the yellow-bellied marmot. J. Zool. (London) 172:233-265. Assem, J.V.D. 1967. Te r r i t o r y in the three-spined sticklebacks Gasterosteus aculeatus L. Behaviour suppl. No 16. B a l l , I.J., P.G.H. Frost, W.R. S i e g f r i e d , and F. McKinney. 1978 T e r r i t o r i e s and l o c a l movements of African black ducks. Wildfowl 29:61-79.. Beecher, M.D. and I.M. Beecher. 1979. Sociobiology of bank swallows: reproductive strategy of the male. Science 197:788-789. Bellrose, F.C. 1976. Ducks, geese and swans of North America. Stackpole Press, Harrisburg. Bellrose, F.C, K.L. Johnson and T.U. Meyers. 1964. Relative value of natural c a v i t i e s and nesting houses for wood ducks. J. Wi l d l . Manage. 28:661-676. Bengston, S.-A. 1971. Habitat selection of duck broods in lake Myvatn area, north-east Iceland. Ornis Scand. 2:17-26. Birkhead, T.R. 1979. Mate guarding in the magpie Pica pica. Anim. Behav. 27:866-874. 142 Bjorklund, M. and B. Westman. 1983. Extra-pair copulations in the Pied Flycatcher (Ficedula hypoleuca). Behav. Ecol. Sociob. 13:271-275. Brown, J.L. 1964. The evolution of d i v e r s i t y in avian t e r r i t o r i a l systems. Wilson B u l l . 76:160-169. Brown, J.L. 1969. T e r r i t o r i a l behavior and population regulation in birds. Wilson B u l l . 81:293-329. Brown, J.L. and G. Orians. 1970. Spacing patterns in mobile animals. Ann. Rev. Ecol. 1:239-262. Burns, J.T., K.M. Cheng and F. McKinney. 1980. forced copulation in captive mallards. I. F e r t i l i z a t i o n of eggs. Auk 97:875-879. Cheng, K.M., J.T. Burns and F. McKinney. 1983. Forced copulation in captive mallards I I I . sperm competition. Auk 100:302-310. Clawson, R.L., G.W. Hartman and L.H. Fredrickspn. 1979. Dump nesting in a Missouri wood duck population. J. Wi l d l . Manage. 43:347-355. Conner, R.N., R.G. Hooper, H.S. Crawford and H.S. Mosby. 1975. Woodpecker nesting habitat in cut and uncut woodlands in V i r g i n i a . J. W i l d l . Manage. 39:144-150. Cowan, I. McT. and J. Hatter. 1952. A trap and technique for the capture of diving waterfowl. J. Wildl. Manage. 16:438-441. Danell, K. and K. Sjoberg. 1982. Successional patterns of plants, invertebrates and ducks in a man-made lake. J. Appl. Ecol. 19:395-409. Dennis, R.H. and H. Dow. 1984. The establishment of a population of goldeneyes (Bucephala clangula) breeding in Scotland. Bird Study 3:217-222. 143 Derrickson, S.R. 1978. The mobility of breeding p i n t a i l s . Auk 95:104-114. Di j k s t r a , C , L. Vuursteen, S. Daan, and D. Masman. 1982. Clutch size and laying date in the kestrel Falco tinnunculus: ef f e c t of supplementary food. Ibis 124:210-213. Donaghey, R.H. 1975. Spacing behavior of breeding buffleheads (Bucephala albeola) on ponds in the southern boreal forest. M.Sc. thesis, Univ. of Alberta, Edmonton. Dow, H. and S. Fredga. 1984. Factors affecting reproductive output of the goldeneye Bucephala clangula. J. Anim. Ecol. 53:679-692. Doty, H.A. and R.J. Greenwood. 1974. Improved nasal-saddle marker for mallards. J. Wi l d l . Manage. 38:938-939. Doty, H.A. and A.D. Kruse. 1972. Technique for establishing l o c a l breeding populations of wood ducks. J.. Wildl. Manage. 36:428-435. Duebbert, H.F. and A.M. Frank. 1984. Value of p r a i r i e wetlands to duck broods. W i l d l . Soc. B u l l . 12:27-34. Dufty, A.M. 1982. Response of brown-headed cowbirds to simulated conspecific intruders. Anim. Behav. 30:1043-1052. Dwyer, T.J. 1974. Social behavior of breeding gadwalls in North Dakota. Auk 91:375-386. Dzubin, A. 1955. Some evidence of home range in waterfowl. Trans. N. Am. Wildl. Conf. 20:278-298. Dzubin, A. 1969. Comments on carrying capacity of small ponds for ducks and possible e f f e c t s of density on Mallard production. In Saskatoon Wetland Seminar, Can. Wildl. Ser., Rep. Ser. no 6, pp 138-171. 1 44 Eadie, J. McA. and G. Gauthier. 1985. Prospecting for nest s i t e s in cavity-nesting ducks of the genus Bucephala. Condor (in press) . Eadie, J. McA. and H.G. Lumsden. 1985. Is nest parasitism always deleterious to goldeneye? Am. Nat. (in press). Emlen, J.T. 1957. Defended area? - a c r i t i q u e of the t e r r i t o r y concept and of conventional thinking. Ibis 99:352. Eriksson, M.O.G. 1978. Lake selection by goldeneye ducklings in rela t i o n to the abundance of food. Wildfowl 29:81-85. Eriksson, M.O.G. 1982. Differences between old and newly established goldeneye (Bucephala clangula) populations, Ornis Fenn. 59:13-19. Erskine, A.J. 1959. A joint clutch of Barrow's goldeneye and bufflehead eggs. Can. Field-Nat. 73:131. Erskine, A.J. 1960. Further notes'on i n t e r - s p e c i f i c competition among hole-nesting ducks. Can. Field-Nat. 74:161-162. Erskine, A.J. 1961. Nest-site tenacity and homing in the bufflehead. Auk 78:389-396. Erskine, A.J. 1964. Nest-site competition between bufflehead, mountain bluebird and tree swallow. Can. Field-Nat. 78:202-203. Erskine, A.J. 1972. Buffleheads. Can. W i l d l . Serv. Monogr. Ser., no 4. Evans, P.R. and M.W. Pienkowski. 1982. Behaviour of Shelducks Tadorna tadorna in a winter flock: does regulation occur? J. Anim. Ecol. 51:241-262. Ewald, P.W., G.L. Hunt and M. Warner. 1980. T e r r i t o r y size in western g u l l s : importance of intrusion pressure, defence investments and vegetation structure. Ecology 61:80-87. 145 Ewald, P.W. and S. Rohwer 1982. E f f e c t s of supplemental feeding on timing of breeding, clutch-size and polygyny in red-winged blackbirds Agelaius phoeniceus. J. Anim. Ecol. 51:429-450. F i t z p a t r i c k , S.M. and W.G. Wellington. 1983. Insect t e r r i t o r i a l i t y . Can. J. Zool. 61:471-486. Franzblau, M.A. and J.P. C o l l i n s . 1980. Test of a hypothesis of t e r r i t o r y regulation in an insectivorous b i r d by experimentally increasing prey abundance. Oecologia 46:164-170. Fretwell S.D. 1972. Populations in a seasonal environment. Monogr. Popul. B i o l . no. 5, Princenton University Press, Pr incenton. Gass, C L . 1979. T e r r i t o r y regulation, tenure, and migration in rufous hummingbirds. Can. J . Zool. 57:914-923. Gass, C.L., G. Angehr and„J. Centa. 1976. Regulation of food supply by feeding t e r r i t o r y in the rufous hummingbird. Can. J. Zool. 54:2046-2054. Gauthier, G. 1986. Experimentally-induced polygyny in buffleheads: evidence for a mixed reproductive strategy? Anim. Behav. (in press). Grice, D. and J.P. Rogers. 1965. The wood duck in Massachusetts. Massachusetts Div. Fi s h . Game, Report W-19-R. Haland, A. 1983. Home range use and spacing in mallard Anas  platyrhynchos broods. Ornis Scand. 14:24-35. Hannon, S.J. 1983. Spacing and breeding density of willow ptarmigan in response to an experimental a l t e r a t i o n of sex ra t i o . J. Anim. Ecol. 52:807-820. Haramis, G.M. and D.M. Nice. 1980. An improved web-tagging technique for waterfowl. J. Wi l d l . Manage. 44:898-899. 146 Haramis, G.M. and D.Q. Thompson. 1985. Density-production c h a r a c t e r i s t i c s of box-nesting wood ducks in a northern greentree impoundment. J. W i l d l . Manage. 49:429-436. Heusmann, H.W., R.W. B e l l v i l l e and R.G. B u r r e l l . 1980. Further observations on dump nesting by wood ducks. J. Wildl . Manage. 44:908-915. H i l l , D.A. 1984. Population regulation in the mallard (Anas  platyrhynchos). J. Anim. Ecol. 53:191-202. Hinde, R.A. 1956. The b i o l o g i c a l significance of the t e r r i t o r i e s of birds. Ibis 98:340-369. Hines, J.E. and G.J. M i t c h e l l . 1984. P a r a s i t i c laying in nests of gadwalls. Can. J. Zool. 62:627-630. Hixon, M.A., F.L. Carpenter and D.C. Paton. 1983. T e r r i t o r y area, flower density, and time budgeting in hummingbirds: an experimental and theore t i c a l analysis. Am. Nat. 122:366-391. Hochbaum, H.A. 1944. The canvasback on a p r a i r i e marsh. American W i l d l i f e Institute, Washington, D.C. Holmes, R.T. 1970. Differences in population density, t e r r i t o r i a l i t y and food supply of Dunlin on a r c t i c and subarctic tundra. In Animal populations in r e l a t i o n to their food resources, A. Watson ed., Symp. B r i t . Ecol. S o c , Blackwell S c i e n t i f i c Publications, Oxford, pp. 303-320. Howard, H.E. 1920. Te r r i t o r y in bird l i f e . Murray, London. Hunter, M.L., J.W. Witham and H. Dow. 1984. E f f e c t s of a carbaryl-induced depression in invertebrate abundance on the growth and behavior of American black duck and mallard ducklings. Can. J . Zool. 62:452-456. Hurlbert, S.H. 1984. Pseudoreplication and the design of ecological experiments. Ecol. Monogr. 54:187-211. 147 Jaeger, R.G., D. Kalvarsky and N. Shimizu. 1982. T e r r i t o r i a l behaviour of the red-backed salamander: expulsion of intruders. Anim. Behav. 30:490-496. Johnson, D.H. 1979. Estimating.nest success: the Mayfield method and an al t e r n a t i v e . Auk 96:651-661. Jones, R.E. and A.S. Leopold. 1967. Nesting interference in a dense population of wood ducks. J. Wildl. Manage. 31:221-228. Joyner, D.E. 1980. Influence of invertebrates on pond selection by ducks in Ontario. J. Wildl. Manage. 44:700-705. Kallander, H. 1974. Advancement of laying of Great T i t Parus  major by provision of food. Ibis 116:365-367. Kaminski, R.M. and H.R. Murkin. 1981. Evaluation of two devices for sampling nektonic invertebrates. J. Wi l d l . Manage. 45:493-496. Kaminski, R.M. and H.H. Prince. 1981. Dabbling duck a c t i v i t y and foraging responses to aquatic macroinvertebrates. Auk 98:115-126. Klett, A.T. and D.H. Johnson. 1982. V a r i a b i l i t y in nest survival rates and implications to nesting studies. Auk 99:77-87. Kluyver, H.N. and L. Tinbergen. 1953. Ter r i t o r y and the regulation of density in Titmice. Arch. Neerlandaises Zool. 10:265-287. Kodric-Brown, A. and J.H. Brown. 1978. Influence of economics, i n t e r s p e c i f i c composition, and sexual dimorphism on t e r r i t o r i a l i t y of migrant rufous hummingbirds. Ecology 59:285-296. Krapu, G.L, 1979. Nutrition.of female dabbling ducks during reproduction. Proceed. 1977 Symp.: Waterfowl and Wetlands - an Integrated Review, T.A. Bookhout ed., The W i l d l i f e Society, Madison, pp. 59-70. 1 48 Krebs, J.R. 1971. T e r r i t o r y and breeding density in the great t i t , Parus major L. Ecology 52:2-22. Lack, D. 1966. Population studies of birds. Clarendon Press, Oxford. Lima, S.L. 1984. T e r r i t o r i a l i t y in variable environments: a simple model. Am. Nat. 124:641-655. Livezey, B.C. and P.S. Humphrey. 1984. Sexual dimorphism in continental steamer-ducks. Condor 86:368-377. Livezey, B.C. and P.S. Humphrey. 1985. T e r r i t o r i a l i t y and i n t e r s p e c i f i c aggression in steamer-ducks. Condor 87:154-157. MacLaren, I.A. 1972. Polygyny as the adaptive function of breeding t e r r i t o r y in birds. Trans. Connecticut Acad. Arts S c i . 44:191-210. MacLean, S.F. and T.R. Seastedt. 1979. Avian t e r r i t o r i a l i t y : s u f f i c i e n t resources or interference competition? Am. Nat. 1 14:308-312. Makepeace, M. and I.J. Patterson. 1980. Duckling mortality in the shelduck, in r e l a t i o n to density, aggressive i n t e r a c t i o n and weather. Wildfowl 31:57,72. Mayfield, H.F. 1961. Nesting success calculated from exposure. Wilson B u l l . 73:255-261. Mayr, E. 1983. How to carry out the adaptationist program? Am. Nat. 121:324-334. McKinney, F. 1965. Spacing and chasing in breeding ducks. Wildfowl Trust Ann. Rep. 16:108-121. McKinney, F. 1973. Ecoethological aspects of reproduction. In Breeding Biology of Birds, D.S. Farner ed, Nat. Acad. S c i . , Washington, D.C. pp. 6-21. 1 49 McKinney, F. 1975. The evolution of duck displays. In Function and evolution in behaviour, G. Baerends, C. Beer and A. Manning eds., Clarendon Press, Oxford, pp. 331-357. McKinney, F., S.C. Derrickson and P. Mineau. 1983. Forced copulation in waterfowl. Behaviour 86:250-294. McKinney, F. , W.R. Siegfried, I.J. B a l l and P.G.H. Frost. 1978, Behavioral speci a l i z a t i o n s for river l i f e in the African black duck (anas sparsa Eyton). Z. Tierpsychology 48:349-400. McKinney, F. and P. Stolen. 1982. extra-pair-bond courtship and forced copulation among captive green-winged tea l (Anas crecca c a r o l i n e n s i s ) . Anim. Behav. 30:461-474. McLaren, W.D. 1963. A preliminary study of nest-site competition in a group of hole-nesting birds. MSc. thesis, Univ. of B r i t i s h Columbia, Vancouver. McLaren, W.D. 1969. Further data on i n t e r s p e c i f i c competition at a joint bufflehead-goldeneye nest s i t e . Can. Field-Nat. 83:59-60. McLaughlin, C L . and D. Grice. 1952. The effectiveness of large-scale erection of wood duck boxes as a management procedure. North Am. Wildl. Conf. 17:242-259. Mineau, P. and F. Cooke. 1979. Rape in the lesser snow goose. Behaviour 20:280-291. Munro, J . and J, Bedard. 1977a. Gull predation and creching . behaviour in the common eider. J. Anim. Ecol. 46:799-810, Munro, J . and J. Bedard. 1977b. creche formation in the common eider. Auk 94:759-771. Munro, J.A. 1942. Studies of waterfowl in B r i t i s h Columbia. Buffle-head. Can. J. Res. 20D:133-160. 1 50 Murkin, H.R., R.G. Abbott and J.A. Kaldec. 1983. A comparison of a c t i v i t y traps and sweep nets for sampling nektonic invertebrates in wetlands. Freshwat. Inverter. B i o l . 2:99-101. Murkin, H.R., R.M. Kaminski and R.D. Titman. 1982. Responses by dabbling ducks and aquatic invertebrates to an experimentally manipulated cattai-l marsh. Can. J . Zool. 60:2324-2332. Myers, J.P., P.G. Connors and F.A. P i t e l k a . 1979. T e r r i t o r y size in wintering sanderlings: the effects of prey abundance and intruder density. Auk 96:551-561. Myers, J.P., P.G. Connors and F.A. P i t e l k a . 1981. Optimal t e r r i t o r y size and the sanderling: compromise in a variable environment. In Foraging behavior, A.C. Kamil and T.D. Sargent eds, Garland, New York, pp. 135-158. Neter, J.G. and W. Wasserman. 1974. Applied linear s t a t i s t i c a l . models. Irwin-Dorsey, Georgetown. Nice, M.M. 1941. The role of t e r r i i o r y in bird l i f e . Am. Midi. Nat. 26:441-487. Noble, G.K. 1939. The role of dominance in the so c i a l l i f e of birds. Auk 56:263-273. Norman, M.D. and G.P. Jones. 1984. Determinants of t e r r i t o r y s i z e ' i n the pomacentrid reef f i s h , Parma v i c t o r i a e . Oecologia 61:60-69. Norton, M.E., P. Arcese and P.W. Ewald. 1982. Effect of intrusion pressure on t e r r i t o r y size in black-chinned hummingbirds (Archilochus alexandri). Auk 99:761-764. Nudds, T.D. and CD. Ankney. 1982. Ecological correlates of t e r r i t o r y and home range size in North American dabbling ducks. Wildfowl 33:58-62. 151 Odum, E.P. and E.J. Kuenzler. 1955. Measurement of t e r r i t o r y and home range size in birds. Auk 72:128-137. Owen, R.B. and K. Reinecke. 1979. Bioenergetics of breeding dabbling ducks. In Proceed. 1977 Symp.: Waterfowl and Wetlands - an Integrated Review, T.A. BookhoUt ed, The W i l d l i f e Society, Madison, pp. 71-92. Patterson, I.J. 1980. T e r r i t o r i a l behavior and the l i m i t a t i o n of population density. Ardea 68:53-62. Patterson, I.J. 1982. The shelduck. Cambridge University Press, Cambridge. Patterson, I.J., A. Gilboa and D.J. Tozer. 1982. Rearing other peoples' young; brood-mixing in the shelduck Tadorna tadorna. Anim. Behav. 30:199-202. Patterson, I.J., M. Makepeace and M. Williams. 1983. Limitation of l o c a l population size in the shelduck. Ardea 105-116. Patterson, J.H. 1976. The role of environmental heterogeneity in the regulation of duck populations. J. Wildl. Manage. 40:22-32. Peterson, B. and G. Gauthier. 1985. Nest s i t e use by cavity-nesting birds of the Cariboo Parkland, B r i t i s h Columbia. Wilson B u l l . 97:319-331. Petrie, M. 1984. Ter r i t o r y size in the moorhen (Gallinula chloropus): an outcome of RHP asymetry between neighbours. Anim. Behav. 32:861-870. Pienkowski, M.W. and P.R. Evans. 1982. Breeding behaviour, productivity and survival of co l o n i a l and non-colonial shelducks Tadorna tadorna. Ornis Scand. 13:101-116. Poston, H.J. 1974. Home range and breeding biology of the shoveler. Can. Wildl. Serv. Report Ser., no. 25. 1 52 Power, H.W. and C.G. Doner. 1980. Experiments on cuckoldry in the mountain bluebird. Am. Nat. 116:689-704. Prince, H.H. 1968. Nest s i t e s used by wood ducks and common goldeneyes in New Brunswick. J . Wildl. Manage. 32:489-500. Rahn, H. and A. Ar. 1974. The avian egg: incubation time and water los s . Condor 76:147-152. Rand, A.S. 1967. The adaptive significance of t e r r i t o r i a l i t y in iguanid l i z a r d . In l i z a r d ecology: a symposium, W.W. Milstead ed., pp.109-115. Ricker, W.E. 1973. Linear regresions in fishery research. J . Fish. Res. Board Can. 30: 409-434. Ricker, W.E. 1975. A note concerning professor Jolicoeur's comments. J. Fish. Res. Board Can. 32:1494-1498. Ric k l e f s , R.E. 1974. Energetics of reproduction in birds. In Avian energetics, R.A. Paynter j r . ed, Nuttal Ornithological Club, Cambridge, pp. 152-297. Riechert, S.A. 1981. The consequence of being t e r r i t o r i a l : spiders, a case study. Am. Nat. 117:871-892. Riggert, T.L. 1977. The biology of the mountain duck on Rottnest island, Western A u s t r a l i a . W i l d l . Monogr. 52. Ringelman, J.K. and J.R. Longcore. 1982. Survival of juvenile black ducks during brood rearing. J. W i l d l . Manage. 43:622-628. Rubenstein, D.I. 1982. Risk, uncertainty and evolutionary strategies. In Current problems in sociobiology, King's college sociobiology group ed, Cambridge University Press, Cambridge, pp. 91-112. Savard, J.P. 1981. V a r i a b i l i t y of waterfowl counts in the Cariboo parkland, B r i t i s h Columbia. In Symposium on census and inventory methods for population and habitats, S.R. Heib ed, The W i l d l i f e Society, Moscow, pp. 127-132. 153 Savard, J.P. 1982a. Intra- and i n t e r - s p e c i f i c competition between Barrow's goldeneye (Bucephala islandica) and bufflehead (Bucephala albeola). Can. J . Zool. 60:3439-344'6. Savard, J.P. 1982b. Barrow's goldeneye nest box u t i l i z a t i o n in the Cariboo Parkland, B r i t i s h Columbia: Year 1. Can. Wildl. Serv., Progress Note, no 131. Savard, J.P. 1984. T e r r i t o r i a l behaviour of common goldeneye, Barrow's goldeneye and bufflehead in areas of sympatry. Ornis Scand. 15:211-216. Savard, J.P. 1985. Use of a mirror trap to capture t e r r i t o r i a l waterfowl. J . F i e l d Ornith. (in press). Schoener, T.W. 1983. Simple models of optimal feeding-territory s i z e : a r e c o n c i l i a t i o n . Am. Nat. 121:608-629. Scott, D.K. and M.E. Birkhead. 1983. Resources and reproductive performance in mute swans Cygnus ol o r . J. Zool. (London) 200:539-547. Seastedt, T.R. and S.F. MacLean. 1979. T e r r i t o r y size and composition in relation to resource abundance in lapland longspurs breeding in a r c t i c Alaska. Auk 96:131-142. Seymour, N.E. 1974a. T e r r i t o r i a l behavior of wild shovelers at Delta, Manitoba. Wildfowl 25:49-55. Seymour, N.E. 1974b. Site attachment in the northern shoveler. Auk 91:423-427. Seymour, N.E. and R.D. Titman. 1978. Changes in a c t i v i t y patterns, agonistic behaviour and t e r r i t o r i a l i t y of black ducks (Anas rubripes) during the breeding season in a Nova Scotia t i d a l marsh. Can. J. Zool. 56:1773-1785. Sie g f r i e d , W.R. 1979. Social behavior of the African comb duck. Living Bird 17:85-104. 1 54 Simon, CA. 1975. The influence of food abundance on t e r r i t o r y size in the iguanid l i z a r d Sceloporus j a r r o v i . Ecology 56:993-998. Slaney, P.A. and T.G. Northcote. 1974. Effects of prey abundance on density and t e r r i t o r i a l behavior of young rainbow trout (Salmo gairdneri) in laboratory stream channels. J. Fish. Res. Board Can. 31:1201-1209. Smith, J.N.M., R.D. Montgomerie, M. T a i t t and Y. Yom-Tov. 1980. A winter feeding experiment on an island sparrow population. Oecologia 47:164-170. Sowls, L.K. 1955. P r a i r i e ducks. W i l d l i f e Management Instute, Washington, D.C. Stewart, G.R. and R.D. Titman. 1980. T e r r i t o r i a l behaviour by p r a i r i e pothole blue-winged t e a l . Can. J. Zool. 58:639-649. Stimson, J . 1973. The role of the t e r r i t o r y in the'ecology of the i n t e r t i d a l limpet L o t t i a qiqantea (Gray). Ecology 54:1020-1030. Sugden, L.G. 1973. Feeding ecology of p i n t a i l , gadwall, American widgeon and lesser scaup ducklings. Can. W i l d l . Serv. Report Ser., no. 24. Swanson, G.A. and J.C. Bartonek. 1970. Bias associated with food analysis in gizzards of blue-winged t e a l . J . W i l d l . Manage. 34:739-746. Talent, L.G., R.L. J a r v i s and G.L. Krapu. 1983. Survival of mallard broods in south-central North Dakota. Condor 85:74-78. Talent, L.G., G.L. Krapu and R.L. J a r v i s . 1982. Habitat use by mallard broods in south central North Dakota. J. Wildl. Manage. 46:629-635. 155 Tinbergen, N., M. Impekoven and D. Franck. 1967. An experiment on spacing-out as a defence against predation. Behaviour 28:307-321. Titman, R.D. 1983. Spacing and three-bird f l i g h t s of mallards breeding in pothole habitat. Can. J. Zool. 61:839-847. Titman, R.D. and J.K. Lowther. 1975. The breeding behavior of a crowded population of mallards. Can. J. Zool. 53:1270-1283. Titman, R.D. and N.R. Seymour. 1981. A comparison of pursuit f l i g h t s by six North American ducks of the genus Anas. Wildfowl 32:11-18. Topping, M.S. and G.G.E. Scudder. 1977. Some physical and chemical features of saline lakes in central B r i t i s h Columbia. Syesis 10:145-166. Verner, J. 1977. On the adaptive s i g n i f i c a n c e of t e r r i t o r i a l i t y . Am. Nat. 111 :769-775. Vi l l a g e , A. 1983. The role of ne s t - s i t e a v a i l a b i l i t y and t e r r i t o r i a l behaviour in l i m i t i n g the breeding density of kestrels. J . Anim. Ecol. 52:635-645. Watson, A. and G.R. M i l l e r . 1971. T e r r i t o r y size and aggression in a fluctuating red grouse population. J. Anim. Ecol. 40:367-383. Watson, A. and R. Moss. 1970. Dominance, spacing behaviour and aggression in re l a t i o n to population l i m i t a t i o n in vertebrates. In Animal populations in rel a t i o n to their food resources, A. Watson ed., Symp. B r i t . Ecol. S o c , Blackwell S c i e n t i f i c Publications, Oxford, pp. 167-218. Watson, A. and R. Moss. 1980. Advances in our understanding of the population dynamics of red grouse from a recent fluctuation in numbers. Ardea 68:103-111. 1 56 Weatherhead, P.J. and R.J. Robertson. 1980. Sexual recognition and anticuckoldry behaviour in savannah sparrows. Can. J. Zool. 58:991-996. Weller, M.W. 1976. Ecology and behaviour of steamer ducks. Wildfowl 27:45-53. Williams, M. 1974. Creching behaviour of the shelduck Tadorna tadorna L. Ornis Scand. 5:131-143. Wilson, E.O. 1975. Sociobiology. Harvard University Press, Cambridge. Wingfield, J.C. 1984. Androgens and mating systems: testosterone-induced polygyny in normally monogamous birds. Auk 101:665-671. Wynne-Edwards, V.C. 1962. Animal dispersion in r e l a t i o n to so c i a l behaviour. Oliver and Boyd, Edinburgh. Yom-Tov, Y. 1974. The. ef f e c t of food and predation on breeding density and success, clutch s i z e , and laying date of the crow (Corvus corone). J. Anim. Ecol. 43:479-498. Young, CM. 1970. T e r r i t o r i a l i t y in the common shelduck Tadorna tadorna. Ibis 112:330-335. 157 APPENDIX I T e r r i t o r i a l system and preferred habitat of the 103 species of the sub-family Anatinae (excluding sheldgeese). Spec ies Zoogr, Habitat T e r r i t . Ref TRIBE TADORNINI Tadorna ferruginea PA brackish lakes (steppe) IV 8 Tadorna cana ET lakes, reservoirs, pools ? Tadorna tadornoides AU brackish lakes, estuaries IV 26 Tadorna variegata AU streams, lakes IV 42 Tadorna tadorna PA estuaries, coast IV 24,43 Tadorna radjah AU coastal mangroves, mudflats IV 1 3 TRIBE TACHYERINI Tachyeres patachonicus NO lakes, estuaries IV 41 Tachyeres pteneres NO coast IV 18 Tachyeres brachypterus NO coast IV 1 8 TRIBE CAIRINI Plectropterus gambensis ET lakes, reservoirs, marshes ? Cairina maschata NO slow-moving r i v e r s , marshes II 1 5 Cairina scutulata OR slow-moving r i v e r s , flooded forest Sarkidiornis melanotos NO/OR/ET temporary ponds in open woodlands I l l 33 Pteronetta hartlaubi ET small streams in rain forest III 1 6 Nettapus pulchellus AU permanent lagoons, lakes ( t r o p i c a l forest) IV 1 3 158 Nettapus coromandelianus AU/OR permanent lagoons, r i v e r s ? (forest) Callonetta leucophrys OR seasonal marshes, flooded forest II? 16 Aix sponsa NA slow-moving r i v e r s , floodplains II 1 4 Aix q a l e r i c u l a t a PA slow-moving r i v e r s , ponds III 7 Chenonetta jubata AU seasonal ponds, creeks (open woodlands) II? 1 3 Amazonetta b r a s i l i e n s i s NO small lagoons (t r o p i c a l forest) ? TRIBE MERGANETTINI Merqanetta armata NO mountain r i v e r s IV 23 TRIBE ANATINI Hymenolaimus malacorhynchos AU mountain r i v e r s IV 21 Anas waigiuensis AU mountain r i v e r s , lakes IV 1 7 Anas sparsa ET fast-flowing r i v e r s IV 19 Anas penelope PA ponds, lakes(boreal forest) II 8 Anas americana NA/PA ponds, lakes (parkland) II 4 Anas s i b i l a t r i s NO permanent lakes, lagoons Anas falcata PA lakes (boreal forest) II? 9 Anas strepera NA/PA seasonal ponds (mixed p r a i r i e ) II 12,36 Anas formosa PA ponds, r i v e r deltas ? Anas crecca NA/PA permanent ponds (parkland, boreal forest) I 20 Anas f l a v i r o s t r i s NO ponds, lagoons, lakes •? Anas capensis ET seasonal ponds, saline ,pools ? Anas gibber ifrons AU seasonal lagoons,floodplains I? 13 159 Anas castanea AU estuaries, coastal lagoons 7 Anas aucklandica AU coastal streams, estuaries IV 39 Anas platyrhynchos NA/PA seasonal ponds (mixed p r a i r i e , parkland) II 35,36 Anas rubripes NA permanent ponds, coastal marshes III 30 Anas melleri ET streams, forested ponds III? 1 6 Anas undulata ET ponds, flooded f i e l d s , estuar ies II? 1 6 Anas poec ilorhyncha PA/OR/AU permanent ponds, marshes, estuaries ? Anas luzonica OR ponds, r i v e r s , t i d a l creeks ? Anas specular i s NO fast-moving r i v e r s IV? 1 6 Anas specularioides NO mountain lakes, coast IV 38 Anas acuta NA/PA seasonal ponds (prairie) I 1 1 Anas qeorgica NO ponds, lakes, estuaries II 40 Anas bahamensis NA/NO brackish ponds, mangroves III 21 Anas erythrorhyncha ET seasonal ponds, lakes 7 Anas versicolor NO shallow ponds, marshes 7 Anas hottentota ET shallow ponds I? 1 6 Anas querquedula PA shallow ponds (steppe, forests) I l l 8 Anas discors NA shallow marshes (parkland) III 34,36 Anas cyanoptera NA/NO shallow ponds II? 1 6 Anas platalea NO brackish lagoons 7 Anas smithii ET seasonal ponds, marshes II 31 Anas rhynchotis AU permanent swamps, marshes III? 1 6 Anas clypeata NA/PA permanent ponds ( p r a i r i e , I l l 25,29 parkland) 160 Malacorhynchus  membranaceous Marmaronetta  a n g u s t i r o s t r i s AU temporary s a l i n e ponds PA a l k a l i n e ponds, f l o o d l a n d s I I ? (step p e ) TRIBE AYTHYINI N e t t a r u f i n a PA N e t t a e r y t h r o p t h a l m a NO/ET N e t t a peposaca NO Aythya v a l i s i n e r i a NA Aythya f e r i n a A ythya americana  Aythya c o l l a r i s  A ythya a u s t r a l i s  A ythya b a e r i  Aythya n y r o c a  Aythya i n n o t a t a  Aythya f u l i g u l a PA NA NA AU PA PA ET PA Aythya n o v a e - s e e l a n d i a e AU Aythya m a r i l a PA/NA Aythya a f f i n i s NA l a r g e a l k a l i n e ponds permanent l a k e s s h a l l o w s e a s o n a l ponds l a k e s , ponds, marshes ( p a r k l a n d ) l a k e s , a l k a l i n e marshes (ste p p e ) permanent a l k a l i n e l a k e s a c i d i c marshes, bogs permanent swamps, l a k e s -deep 'lakes s h a l l o w ponds (st e p p e ) marshes, l a k e s l a g o o n s , deep l a k e s ( b o r e a l f o r e s t ) I I 8 ? ? I I 3,15 I I 8 I I I ? ? ? ? ? I I ? deep l a k e s , l agoons ? s h a l l o w ponds, bogs ( t u n d r a ) ? marshes, ponds, l a k e s I I ( p a r k l a n d ) TRIBE MERGINI Som a t e r i a m o l l i s s i m a NA/PA c o a s t , e s t u a r i e s S o m a t e r i a s p e c t a b i l i s NA/PA ponds i n c o a s t a l t u n d r a S o m a t e r i a f i s c h e r i NA/PA ponds i n c o a s t a l t u n d r a P o l y s t i c t a s t e l l e r i NA/PA ponds i n c o a s t a l t u n d r a I I I ? 1 5 22 1,16 15 161 Histrionicus h i s t r i o n i c u s NA/PA mountain rivers Clanqula hyemalis NA/PA Melanitta nigra NA/PA Melanitta p e r s p i c i l l a t a NA Melanitta fusca NA/PA Bucephala albeola NA Bucephala islandica NA/PA Bucephala clanqula NA/PA Merqus cucullatus NA Merqus a l b e l l u s PA Merqus actosetaceus NO Merqus serrator NA/PA Merqus squamatus PA Merqus merganser NA/PA TRIBE OXYURINI Heteronetta a t r i c a p i l l a NO Oxyura dominica NO Oxyura jamaicensis NA/NO Oxyura leucocephala PA Oxyura maccoa ET Oxyura v i t t a t a NO Oxyura austral is AU Biziura lobata AU coast, tundra lakes tundra lakes and ponds lakes and ponds (boreal forest) lakes and ponds (boreal forest) permanent ponds (parkland, boreal forest) permanent ponds, lakes (parkland, boreal forest) permanent ponds, lakes (boreal forest) III III II? ? III? IV 5,6 2 8 1 1 IV 27,28 IV 28 wooded streams, lakes ? slow-moving r i v e r s , lakes ? fast-moving streams (forest) ? lakes and streams (forest) I mountain streams ? lakes, ponds (boreal forest) I permanent marshes II tro p i c a l marshes ? permanent marshes(parkland) II marshes, ponds (steppe) I marshes, ponds IV marshes, permanent lakes ? permanent marshes III? permanent marshes IV 37 32 1 6 32 1 6 16 1 62 a Zoogr.: zoogeographical region inhabited by the species. NA = Nearctic, NO = Neotropical, PA = Pale a r c t i c , ET = Ethiopian, OR = Oriental, AU = Australian. b T e r r i t . : Type of t e r r i t o r i a l system. I = no t e r r i t o r y , II = loosely defended moving t e r r i t o r y around the female, III = strongly defended moving t e r r i t o r y around the female, sometimes s i t e - s p e c i f i c , IV = strongly defended s i t e - s p e c i f i c t e r r i t o r y (see chapter 2 and Fig. 1 for more d e t a i l s ) . 163 REFERENCES 1. Afton, A.D. 1985. Forced copulation as a reproductive strategy of male lesser scaup: a f i e l d test of some predictions. Behaviour 92:146-167. 2. Alison, R. 1975. Breeding biology and behavio-r of the oldsquaw (Clangula hyemalis L.). Am. Ornithol. Monogr. 18. 3. Anderson, M.G. 1984. Parental investment and pair-bond behavior among canvasback ducks (Aythya v a l i s i n e r i a , Anatidae). Behav. Ecol. Sociob. 15:81-90. 4. Bellrose, F.C. 1976. Ducks, geese and swans of North America. Stackpole Press, Harrisburg. 5. Bengston, S.-A. 1966. F i e l d studies on the harlequin duck in Iceland. Wildfowl Trust, 17th Ann. Rep., pp. 79-94. 6. Bengston, S.-A. 1972. Breeding ecology of the harlequin duck His t r i o n i c u s h i s t r i o n i c u s (L.) in Iceland..Ornis Scand. 3:1-19. : 7. Bruggers, R.L. 1979. Nesting patterns of captive mandarin ducks. Wildfowl 30:45-54. 8. Cramp, D. and K.E.L. Simmons. 1977. Handbook of the birds of Europe, the Middle East and North A f r i c a . Vol.1. Oxford University Press, Oxford. 9. Dementiev, G.P. and N.A. Gladkov. 1967. Birds of the Soviet Union. Vol. 2. Trans, by Israel Program for S c i e n t i f i c Translations, U.S. Dept. Interior and Nat. S c i . Found., Washington D.C. 10. Derrickson, S.R. 1978. The mobility of breeding p i n t a i l s . Auk 95:104-114. 11. Donaghey, R.H. 1975. Spacing behavior of breeding buffleheads (Bucephala albeola) on ponds in the southern boreal forest. M.Sc. thesis, Univ. of Alberta, Edmonton. 1 64 12. Dwyer, T.J. 1974. Social behavior of breeding gadwalls in North Dakota. Auk 91:375-386. 13. F r i t h , H.J. 1967. Waterfowl in A u s t r a l i a . East-West Center Press, Honolulu. 14. Grice, D. and J.P. Rogers. 1965. The wood duck in Massachusetts. Massachusetts Div. Fish. Game, Report W-19-R. 15. Johnsgard, P.A. 1975. Waterfowl of North America. Indiana University Press, Bloomington. 16. Johnsgard, P.A. 1978. Ducks, geese and swans of the world. University of Nebraska Press, Lincoln. 17. Kear, J . 1975. Salvadori's duck of New Guinea. Wildfowl 26:104-111. 18. Livezey, B.C. and P.S. Humphrey. 1985. T e r r i t o r i a l i t y and i n t e r s p e c i f i c aggression in steamer-ducks. Condor 87:154-157. 19. McKinney, F., W.R. Sie g f r i e d , I.J. B a l l and P.G.H. Frost. 1978. Behavioral s p e c i a l i z a t i o n s for river l i f e in the African black duck (Anas sparsa Eyton). Z. Tierpsychology 48:349-400. 20. McKinney, F. and P. Stolen. 1982. Extra-pair-bond courtship and forced copulation among captive green-winged teal (Anas crecca carol inensi s) . Anim. Behav. 3.0:461-474. 21. McKinney, F. and D.J. Brugger. 1979. Status and breeding behavior of the Bahama p i n t a i l and the-New Zealand blue duck. In Proc. Jean Delacour/IFCB symposium on breeding birds in c a p t i v i t y , Hollywood, pp. 211-221. 22. Mendall, H.L. 1958. The ring-necked duck in the northeast. Univ. of Maine B u l l 60(16). 165 23. Moffett, G.M., j r . 1970. A study of nesting torrent ducks in the Andes. Living Bird 9:5-28. 24. Patterson, I.J. 1982. The shelduck. Cambridge University Press, Cambridge. 25. Poston, H.J. 1974. Home range and breeding biology of the shoveler. Can. Wildl . Serv. Report Ser., no. 25. 26. Riggert, T.L. 1977. The biology of the mountain duck on Rottnest island, Western A u s t r a l i a . W i l d l . Monogr. 52. 27. Savard, J.P. 1982. Intra- and i n t e r - s p e c i f i c competition between Barrow's goldeneye (Bucephala islandica) and bufflehead (Bucephala albeola). Can. J . Zool. 60:3439-3446. 28. Savard, J.P. 1984. T e r r i t o r i a l behaviour of common goldeneye, Barrow's goldeneye and bufflehead in areas of sympatry. Ornis Scand. 15:211-216. 29. Seymour, N.E. 1974. T e r r i t o r i a l behavior of wild shovelers at Delta, Manitoba. Wildfowl 25:49-55. 30. Seymour, N.E. and R.D. Titman. 1978. Changes in a c t i v i t y patterns, agonistic behaviour and t e r r i t o r i a l i t y of black ducks (Anas rubripes) during the breeding season in a Nova Scotia t i d a l marsh. Can. J. Zool. 56:1773-1785. 31. S i e g f r i e d , W.R. 1965. The Cape shoveler Anas smithi i (Hartert) in southern A f r i c a . Ostrich 37:155-198. 32. S i e g f r i e d , W.R. 1976. Social organization in ruddy and maccoa duck. Auk 93:560-570. 33. S i e g f r i e d , W.R. 1979. Social behavior of the African comb duck. Living Bird 17:85-104. 34. Stewart, G.R. and R.D. Titman. 1980. T e r r i t o r i a l behaviour by p r a i r i e pothole blue-winged t e a l . Can. J. Zool. 58:639-649. 166 35. Titman, R.D. 1983. Spacing and three-bird f l i g h t s of mallards breeding in pothole habitat. Can. J. Zool. 61:839-847. 36. Titman, R.D. and N.R. Seymour. 1981. A comparison of pursuit f l i g h t s by six North American ducks of the genus Anas. Wildfowl 32:11-18. 37. Weller, M.W. 1968. The breeding biology of the p a r a s i t i c black-headed duck. L i v i n g Bird 7:169-207. 38. Weller, M.W. 1972. Ec o l o g i c a l studies of Falkland Island's waterfowl. Wildfowl 23:25-44. 39. Weller, M.W. 1975. Ec o l o g i c a l studies of the Auckland Islands f l i g h t l e s s t e a l . Auk 92:280-297. 40. Weller, M.W. 1975. Ecology and behavior of the South Georgia p i n t a i l Anas c[. georgica. Ibis 117:217-231. 41. Weller, M.W. 1976. Ecology and behaviour of steamer ducks. Wildfowl 27:45-53. 42. Williams, M. 1979. The s o c i a l structure, breeding and population dynamics of paradise shelduck in the Gisborne-East Coast d i s t r i c t . Notornis 26:213-272. 43. Young, CM. 1970. T e r r i t o r i a l i t y in the common shelduck Tadorna tadorna. Ibis 112:330-335. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0096666/manifest

Comment

Related Items