UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The timing of breeding in double-crested cormorants (phalacrocorax auritus albociliatus) : its effects… Sullivan, Terrance Michael 1998

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

Item Metadata

Download

Media
831-ubc_1998-0310.pdf [ 5.29MB ]
Metadata
JSON: 831-1.0088450.json
JSON-LD: 831-1.0088450-ld.json
RDF/XML (Pretty): 831-1.0088450-rdf.xml
RDF/JSON: 831-1.0088450-rdf.json
Turtle: 831-1.0088450-turtle.txt
N-Triples: 831-1.0088450-rdf-ntriples.txt
Original Record: 831-1.0088450-source.json
Full Text
831-1.0088450-fulltext.txt
Citation
831-1.0088450.ris

Full Text

THE TIMING OF BREEDING IN DOUBLE-CRESTED CORMORANTS (Phalacrocorax auritus a l b o c i l i a t u s ) : ITS EFFECTS ON CLUTCH SIZE , NESTLING GROWTH, DIET AND SURVIVAL. By Terrance Michael S u l l i v a n B.Sc, University of B r i t i s h Columbia, 1987 A THESIS IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE STUDIES (Department of Animal Science) We accept the thesis as conforming to the required standards "hzk UNIVERSITY OF BRITISH COLUMBIA March 1998 © Terrance Michael S u l l i v a n 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 fgr reference and study. 1 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 /h\£YW$ ^C^lUntJZ^ The University of British Columbia Vancouver, Canada Date Z- M^A DE-6 (2/88) ABSTRACT I examined the effects of a delayed time of breeding on clu t c h size, n e s t l i n g growth and diet of Double-crested Cormorants {Phalacrocorax auritus albociliatus) at three colonies, Five Finger Island, Mandarte Island and the Fraser River, during the 1993 and 1994 breeding seasons. The e f f e c t of population changes was also monitored at these colonies and 4 additional colonies, Crofton, Chain Island, C h r i s t i e I s l e t and Hudson Rocks, a l l situated within the S t r a i t of Georgia As the breeding season progressed, asymptotic mass and condition index of the nestlings declined. In general, nestlings produced early i n the season were heavier, had a slower growth rate and required more time to achieve t h e i r f i n a l mass than those produced l a t e r i n the season. The l a s t birds to breed were on the Fraser River (1993) where successful egg laying occurred approximately 3 months af t e r the f i r s t colony (Five Finger Island) bred. The patterns of growth of the st r u c t u r a l components d i f f e r e d at t h i s colony from a l l others. The asymptotic length of both the culmen and tarsus, and the o v e r a l l size of the nestlings increased as the season progressed. These parameters were also larger than at a l l other colonies. ii Nestling cormorants on Five Finger Island were fed mainly-P a c i f i c Sandlance, nestlings on Mandarte Island were fed mainly blennies and those on the Fraser River were fed mainly P a c i f i c Staghorn Sculpin and Shiner Perch. Colony differences probably r e f l e c t colony locations i n the S t r a i t of Georgia but may be influenced by the time of breeding. The quantity or q u a l i t y of food delivered to the nestlings did not d i f f e r among colonies, except for the amount of l i p i d s . Differences i n nest l i n g growth could not be explained by differences i n the di e t . To determine the eff e c t of the timing of breeding on the changes i n cormorant numbers, I modeled the dynamics of the population. Both the predicted and observed changes i n the breeding population followed the same trend: at colonies where breeding occurred early i n the season, there was an o v e r a l l increase i n the breeding populations while those that bred l a t e r i n the season showed a decline. This trend was hypothesized to res u l t from poor survival of the nestlings r e s u l t i n g i n low recruitment into the breeding population. iii TABLE OF CONTENTS ABSTRACT i i TABLE OF CONTENTS iv LIST OF TABLES ix LIST OF FIGURES xi ACKNOWLEDGMENTS xii CHAPTER 1: INTRODUCTION, STUDY SPECIES AND STUDY SITES . . 1 Introduction 1 Aim of thesis 2 Study Species 3 D i s t r i b u t i o n 3 Breeding Biology 4 Non-breeding Season 6 Diet 7 Study Sites 7 Fraser River 8 Five Finger Island 8 Mandarte Island 11 Other Sites 11 CHAPTER 2. THE TIMING OF BREEDING OF DOUBLE-CRESTED CORMORANTS AND ITS EFFECTS ON CLUTCH SIZE . . . 13 Introduction 13 Methods 14 S t a t i s t i c a l analyses 14 Results 15 Timing of breeding 15 Clutch size 15 Discussion 19 The timing of breeding 19 Laying dates and clutch size . . . . . . . 19 Ecological determination of clutch size . . 20 Clutch s i z e : an alternative approach . . .21 Summary 22 iv CHAPTER 3. GROWTH OF NESTLING DOUBLE-CRESTED CORMORANTS IN RELATION TO THE TIME OF BREEDING 23 Introduction 23 Materials and Methods 24 Mass and str u c t u r a l size 25 Aging nestling cormorants 26 Growth rate 2 6 Length of the growth period 2 6 Structural Size 27 Condition index 27 Mass and environmental factors 2 7 S t a t i s t i c a l analyses 27 Results 30 Mass 31 Asymptotic mass 31 Lo g i s t i c growth rate for mass (LGRMass) . . 31 Time taken to grow from 10-90% of the asymptotic mass (t 1 0 . 9 0 M a s s) . .• 35 Culmen length 3 7 Asymptotic culmen length 3 7 L o g i s t i c growth rate for culmen ( L G R c u ^ n ) . 3 8 The time taken to achieve 10-90% of the culmen length ( t 1 0 . 9 0 cuimen) 4 2 Tarsal length 43 Asymptotic t a r s a l length 43 L o g i s t i c growth rate for tarsus (LGR T a r s u s) . 44 The time taken to achieve 10-90 % of the t a r s a l length ( t 1 0 _ 9 0 T a r s u s) 44 Structural size 44 Asymptotic s t r u c t u r a l size 44 Condition index 45 Growth and environmental conditions . . . .46 Discussion 46 The advantages and consequences of la t e breeding on the Fraser River (1993) . . . . 51 Laying Date and growth - possible mechanisms 52 Nestling mass and condition index: alternative explanations . . . . 53 Summary 54 CHAPTER 4. COMPOSITION AND NUTRITIONAL CONTENT OF THE DIETS FED TO NESTLING DOUBLE-CRESTED CORMORANTS 55 v Introduction 55 Materials and methods 56 Diet 56 U t i l i z a t i o n of prey species among colonies 57 Nu t r i t i o n a l analysis 58 Nutrient composition of the boluses . . 59 Provisioning, the timing of breeding and n e s t l i n g mass 60 Assumptions 60 S t a t i s t i c a l analyses 61 Results 61 Nestling diets 61 Occurrence of prey species among colonies 61 The quantity and qu a l i t y of ne s t l i n g diets and nestling growth 64 Nu t r i t i o n a l analysis 64 Nutrient composition of the boluses . . 64 Provisioning, the timing of breeding and nestli n g mass 68 Discussion 72 Diets fed to nestl i n g cormorants . . . . 72 Provisioning, timing of breeding and nestling mass 74 Prey composition and t h e i r e f f e c t s on growth 76 Summary 77 CHAPTER 5. THE TIMING OF BREEDING: ITS EFFECTS ON JUVENILE SURVIVAL AND POPULATION LEVELS OF DOUBLE-CRESTED CORMORANTS 79 Introduction 79 Materials and Methods 82 The population model 82 Adult survivorship 83 Assumptions 83 Number of nestlings produced 83 Assumptions 84 Juvenile survival rates 84 Assumptions 85 Number of juveniles surviving 85 Observed number of breeding pairs . . . . . . 85 A test of the predictions 85 S t a t i s t i c a l analyses 88 vi Results 88 The model data 88 The number of adults surviving 88 The number of young produced . . . . . 88 Juvenile s u r v i v a l rates and number of surviving young 91 The predicted number of breeding pairs . . . 91 Observed number of breeding pairs 94 Discussion 94 Adequacy of the data 94 Adult mortality rates 94 Number of young produced 97 Juvenile survival rates 97 Dispersion of Double-crested Cormorants . . . 98 Growth and Juvenile survival rates . . . . .99 The timing of breeding and numbers of breeding pairs 100 The timing of breeding and the metapopulation 101 Other factors a f f e c t i n g the number of breeding pairs 102 Summary 102 CHAPTER 6. GENERAL DISCUSSSION 104 Clutch size and the timing of breeding 104 Nestling growth and the timing of breeding . . .105 Diet of nestl i n g cormorants and the timing of breeding 106 Double-crested Cormorant populations and the timing of breeding 107 Future directions for research 108 Epilogue I l l LITERATURE CITED 112 APPENDIX I. ANOVA source of v a r i a t i o n tables, comparisons of least squared means and results of p a r t i a l F and Scheffe's multiple comparison tests . .120 APPENDIX II. Diet of nestl i n g Double-crested Cormorants at three colonies within the S t r a i t of Georgia 141 vii APPENDIX I I I . The frequency of occurrence of f i s h i n the boluses of nestl i n g Double-crested Cormorants i n 1993 142 APPENDIX IV. Diet of nestl i n g Double-crested Cormorants at Mandarte Island i n 1971 (Robertson 1974) and 1993 143 APPENDIX V. N u t r i t i o n a l composition of prey f i s h fed to f i s h eating birds 144 APPENDIX VI. Glossary of abbreviations 145 viii LIST OF TABLES Table 2.1. Clutch size of Double-crested Cormorants at three colonies within the S t r a i t of Georgia, B r i t i s h Columbia 18 Table 3.1. Summary table of growth data (± sd) for nes t l i n g Double-crested Cormorants 32 Table 3.2. Comparisons of l o g i s t i c growth rate for culmen among colonies i n broods from 1 to 4 young . .39 Table 3.3. A comparison of growth rates for ne s t l i n g Double-crested Cormorants 50 Table 4.1. The number of boluses, t o t a l mass, number of f i s h and mean bolus size c o l l e c t e d from n e s t l i n g Double-crested Cormorants at three colonies i n the S t r a i t of Georgia i n 1993 62 Table 4.2. The r e l a t i v e importance index (RII) of f i s h fed to n e s t l i n g Double-crested Cormorants at three colonies i n the S t r a i t of Georgia i n 1993 . . 63 Table 4.3. Nutrient analysis of f i s h fed to n e s t l i n g Double-crested Cormorants at 3 colonies within the S t r a i t of Georgia 65 Table 4.4. Nutrient composition of boluses (± sem) fed to nes t l i n g Double-crested Cormorants i n 1993 . . 66 Table 4.5. Total (preformed and metabolic) water produced from boluses fed to nestl i n g Double-crested Cormorants i n 1993 67 Table 4.6. The qu a l i t y of food (g nutrient dry matter/g bolus dry matter) delivered to ne s t l i n g Double-crested Cormorants i n 1993 (+ sem) . . . 69 ix Table 4.7. Comparisons i n the provisioning of the young: 1) the same quantity of food delivered to a l l nestlings; 2) the quantity of food delivered to the nestlings i n proportion to the length of daylight; and 3) the quantity of food delivered i n proportion to the length of daylight and the mean brood size at each colony 7 0 Table 4.8. The mean hatch date, brood size, the length of daylight at hatch, and the decline i n food provisioned to the young i f provisioning i s proportional to the amount of daylight at hatch 71 Table 5.1. The number of nests and number of adult Double-crested Cormorants breeding, dying and surviving (based on a mortality rate of 15% based on van der Veen 1973) 89 Table 5.2. The number of successful nests, brood size and predicted number of young surviving to breeding age at three colonies within the S t r a i t of Georgia 90 Table 5.3. Predicted juvenile survival rates based on the timing of breeding 92 Table 5.4. The number of breeding pairs i n 1993 and 1994 and the predicted number of breeding pairs projected to survive and breed i n 1996 and 1997 93 Table 5.5. Colony locations, number of nests observed and the annual change (%) i n nesting numbers between 1987 and 1994 i n the S t r a i t of Georgia, B.C 95 Table 5.6. The timing of breeding i n 1989, 1993 and 1994 and the observed change i n the number of breeding pairs of Double-crested Cormorants at 7 colonies within the S t r a i t of Georgia, B.C 96 x LIST OF FIGURES Figure 1.1. Location of Double-crested Cormorant colonies, within the S t r a i t of Georgia, where growth and diet s e l e c t i o n (closed c i r c l e s ) and changes i n the number of breeding pairs (open and closed c i r c l e s ) were studied i n 1993 and 1994 9 Figure 2.1. The mean (± sd) and range of laying dates of Double-crested Cormorants, i n 1993 and 1994, at 3 colonies within the S t r a i t of Georgia (FFI = Five Finger Island, MAND = Mandarte Island and FRSR = Fraser River) 16 Figure 3.1.Asymptotic mass of ne s t l i n g Double-crested Cormorants form Five Finger Island (FFI) i n 1993 (0) and 1994 (•), Mandarte Island (+), and the Fraser River i n 1993 (V) and 1994(X). . 33 Figure 3.2. Relationship between brood size and l o g i s t i c growth r a t e C u l m e n for nestlings: A) i n 1993 on Mandarte Island (0) and Fraser River ( + ) ; and B) i n 1994 on the Fraser River (X) and Five Finger Island (•) 40 Figure 3.3. Condition index of n e s t l i n g Double-crested Cormorants from Five Finger Island (FFI) i n 1993 (0) and 1994(D), Mandarte Island (+), and the Fraser River i n 1993 (V) and 1994(X) . 47 Figure 5.1. The post fledging survival estimates for Shags Phalacrocorax aristotelis surviving to 3 years of age, based on a model by Harris et al. (1994) 86 XI ACKNOWLEDGEMENTS I am indebted to many people. Without t h e i r help t h i s thesis would not be possible. My supervisor, Dr. Kim Cheng who believed i n me and has pa t i e n t l y helped me throughout t h i s thesis. Dr Maryanne Hughes who has shown me that physiology and ecology go hand i n hand and one cannot be considered without the other. Dr. Rob Butler, who has not only been my boss f o r many years but also my friend, has always been there f o r support and encouragement. Dr. Jamie Smith agreed to take time out of his busy schedule to act as the external. In the f i e l d , many friends shared i n t h i s study. Larry and Linda Smith and Jan and Diego Bittante were there i n the beginning(1989)to help band nest l i n g cormorants. Throughout t h i s thesis, these 4 have always been there to lend a hand, and a beer (or two . . .) and/or dinner. Lukas Ke l l e r , Ian Moul, Danielle McCall, B i l l Wareham, Karen Broom, Meg Hatch, Don, G a i l , Tanja, T r i c i a and Lana Su l l i v a n also provided invaluable assistance. Drs. Tony Kozak, John Smith and e s p e c i a l l y Val Lemay p a t i e n t l y helped with the s t a t i s t i c s i n t h i s t h e s i s . Dr. Dick Beames was always there to discuss aspects of the nutrient analysis. Dr. March always had an open door to t a l k about any aspect of t h i s thesis. I would also l i k e to thank Dr. Peter Arcese, Lukas K e l l e r and Meg Hatch for allowing me to work on Mandarte Island. I also owe a huge debt to Sy l v i a Leung for running (and re-running)SAS for me. Without the assistance of my parents, none of t h i s would have ever been possible. I do not know how to even begin to say thanks. My brother Sam penned the words on his CD, "an album i s never finished, only abandoned." In many ways, I f e e l the same. This thesis w i l l never t r u l y be "finished", however, i t i s now time to "abandon" i t . Then again, just another f i e l d season or two and I could . . . Finan c i a l support for t h i s thesis was provided by Environment Canada's Green Plan, s p e c i f i c a l l y , the Fraser River Action Plan. In addition, further funding was provided by the Canadian W i l d l i f e Service. xii Chapter 1. Introduction, study species and study s i t e s Introduction Why does a bi r d breed when i t does? It has been postulated that both the timing and the rate of reproduction (clutch size) are ultimately regulated by food abundance. Natural selection favours breeding individuals which time the greatest demands of growing chicks with peak levels of food a v a i l a b i l i t y (Lack 1954, Perrins 1970) and individuals which only produce as many young as they can support (Lack 1968). Since egg laying begins p r i o r to peak levels of food abundance, breeding birds use environmental cues to predict future food abundance. Thus proximate factors affect both the timing and rate of reproduction and greatly influence reproductive success. A seasonal decline i n clutch size occurs for many species (Klomp 1970) and i s thought to be an adaptation to declining food levels as the breeding season progresses (Perrins and Birkhead 1983, see also Daan et. al. 1988). The timing of breeding also affects the pr o b a b i l i t y of young surviving: young Shags (Phalacrocorax aristotelis) produced l a t e r i n the season had lower rates of survival than those produced early i n the season (Harris et al. 1994). Population crashes i n Shags have been attributed to late breeding (timing), lower production and poor survival of young (Aebischer and Wanless 1992) . In 1989, I f i r s t studied the breeding ecology of Double-1 crested Cormorants {Phalacrocorax auritus) i n the S t r a i t of Georgia, B r i t i s h Columbia. Due to prolonged periods of egg predation on Mandarte Island, breeding was delayed by a month and a half as compared to the 1970's (Robertson 1971). While there were fewer successful nests (21 % ) , clutch size did not d i f f e r s i g n i f i c a n t l y f rom the 1970's (Sullivan 1989). That breeding was delayed without a seasonal decline i n clutch size stimulated me to determine i f there were any associated cost with delayed breeding and, i f so, how would i t be manifested. Three colonies of Double-crested Cormorants, with dates of laying ranging from early May to early August, were studied i n 1993 and 1994. The effects of the timing of breeding on the rates of reproduction, growth of the nestlings and nestling diet were determined. Laying dates ranged from early May to early August. The extended time of breeding resulted from natural factors. Aim of the Thesis My aim here was to determine the effects of delayed breeding on Double-crested Cormorants at 3 colonies within the S t r a i t of Georgia. In chapter 2, I examine the effect of the timing of breeding on clutch size. In chapter 3, I examine the effects of the timing of breeding on nestling growth. In chapter 4, I examine the diet of nestlings at the 3 colonies and discuss i t s effects on the growth of the nestlings. In chapter 5, I consider the effects of the timing of breeding on recruitment of young into the breeding population. This i s done by modelling the breeding 2 population at individual colonies, and the metapopulation within the S t r a i t of Georgia. F i n a l l y , i n chapter 6, I discuss the effects of the timing of breeding on Double-crested Cormorants within the S t r a i t of Georgia based upon my findings. This study was i n i t i a t e d due to concerns about the effects of contaminant discharge from the Fraser River on the breeding biology of Double-crested Cormorants. It was f e l t that a study of a colony near the mouth of the Fraser River (Fraser River) and colonies further away (Mandarte Island and Five Finger Island) would provide necessary comparisons to determine i f there were any effects of contaminant exposure. Study Species D i s t r i b u t i o n - There are 4 subspecies of Double-crested Cormorants breeding i n North America: cincinatus, albociliatus, floridanus and auritus (Johnsgard 1993) . Cincinatus breeds i n Alaska and the Aleutian Islands; albociliatus breeds from southern B r i t i s h Columbia to Baja C a l i f o r n i a and i s the subject of t h i s thesis; floridanus breeds throughout the Gulf of Mexico and north to North Carolina and auritus breeds from the central i n t e r i o r of Canada and the United States east through the Great Lakes to the east coast. The f i r s t breeding record for Double-crested Cormorants i n the S t r a i t of Georgia was i n 1927 (Munro 1928). Yet t h e i r remains have been found i n archaeological sit e s , inhabited between 3 500 BC and 1800 AD, throughout the southern S t r a i t of Georgia (Hobson and 3 Driver 1989). Double-crested Cormorants have nested at 15 s i t e s within the S t r a i t of Georgia, (Vermeer et al. 1989) but only 9 are currently active (I. Moul pers. comm.). Throughout North America, populations of Double-crested Cormorants declined from the 1950's and into the early 1970's (Kury 1969, Henney et al. 1989). This was attributed to egg s h e l l thinning caused by high dichlorodiphenyltrichloroethane (DDT) leve l s (Henny et al. 1989). Since the banning of DDT, populations of Double-crested Cormorants have increased throughout t h e i r range (Scharf and Shugart 1981, Hatch 1984, Vermeer and Rankin 1984, Craven and Lev 1987, Findholt 1988, Hobson et al. 1989), including the S t r a i t of Georgia (Vermeer et al. 1989, Campbell et al. 1990). In 1989, the population of cormorants at several colonies within the S t r a i t of Georgia declined (Sullivan 1989). The reason for t h i s decline i s unknown. Breeding Biology - Within the S t r a i t of Georgia, Double-crested Cormorants arrive at the breeding colonies i n early A p r i l (Drent et al. 1964). They are colonial nesters that t y p i c a l l y nest on the ground on rocky islands or i s l e t s but w i l l also nest i n trees and on man-made structures (Lewis 1929). Previously used nests, which have not been destroyed by winter storms, are re-claimed by the males (Lewis 1929). Unused or undefended nests are quickly dismantled and incorporated into active nests. Nests are made up almost e n t i r e l y of intertwined s t i c k s and are cemented with faecal material. After a l l nests have 4 been claimed, new nests may be b u i l t . This requires about 4 days (Palmer 1962) . Once the male has a nest, he begins courting potential mates using vocalisations, combined with elaborate courtship displays (van Tets 1965) . The dramatic contrast of the cormorant's black plumage, white filoplumes, yellow-orange gular pouch and blue-coloured l i n i n g of the mouth and eye ring may be c r i t e r i a used i n mate selection. Once a pair bond i s formed, copulations occur frequently throughout the day. The f i r s t egg i s l a i d between late A p r i l and mid May i n the S t r a i t of Georgia (Drent et al. 1964). Subsequent eggs are l a i d at one to two day intervals u n t i l a modal clutch size of four eggs i s reached. Double-crested Cormorants may be indeterminate layers since removal of eggs early i n the laying sequence results i n replacement eggs being l a i d (pers. obs.). Eggs are pale blue and are usually covered with large calcareous deposits. Eggs (n = 570) average (± sd) 6.21 + 0.28 cm i n length, 3.90 ± 0.14 cm i n width and have an average volume of 47.94 + 4.32 cc (Sullivan, unpubl. data). An average egg length of 6.29 and a width of 3.88 (n = 71) have been recorded for P. auritus albociliatus (Palmer 1962). Each egg i s approximately 2 % of the females body mass and a 4 egg clutch represents approximately 9 % of her mass. Incubation i s carried out by both sexes and l a s t s 25 - 2 8 days (Robertson 1971). Young emerge completely naked, b l i n d and helpless. Hatching i s asynchronous and usually resembles the 5 laying i n t e r v a l . Nestlings are fed by both parents and leave the nest approximately 50 - 55 days afte r hatching (Robertson 1971). They are t o t a l l y independent of parental care by day 70 (Lewis 192 9). A fter 2 to 3 years, young Double-crested Cormorants return to the colonies to breed. Double-crested Cormorants usually nest i n association with g u l l s {Larus spp). Disturbances within the colony, caused mainly by Bald Eagles {Haliaeetus leucocephalus) and/or humans (Verbeek 1982), are usually associated with egg predation by Glaucous-winged Gulls (Larus glaucescens) and Northwestern Crows (Corvus caurinus) (Drent et al. 1964, Verbeek 1982). Re-laying, following the loss of either single eggs or complete clutches, i s common (Drent et al. 1964). The mortality rate of adult Double-crested Cormorants (P. auritus albociliatus) was determined to be 15 % per year (van der Veen 1973) which translates to an average l i f e expectancy of 6.2 years. One Double-crested Cormorant, banded and re-sighted on Mandarte Island, was determined to be 19 years of age (Sullivan unpubl. data). Potts (1969) has recorded Shags up to 15 years of age. Non-breeding Season - East of the Rocky mountains, cormorants are migratory and winter i n the southern United States (Dolbeer 1991). Juvenile cormorants not only winter further away from t h e i r natal colonies than older birds, but they also return l a t e r i n the spring. On the west coast, l i t t l e i s known about the movements of adult or juvenile cormorants aft e r they leave the 6 breeding colonies, but Double-crested Cormorants are present year round i n the S t r a i t of Georgia (Campbell et al. 1990) . The migration/dispersal of Double-crested Cormorants nesting i n the S t r a i t of Georgia has not been systematically studied. In the winter months, I have observed adult Double-crested Cormorants near or on the breeding colonies at Crofton, B.C. and at the Fraser River estuary but never on either Mandarte Island or Five Finger Island. Therefore, a segment of the population may migrate/disperse while another segment of the population may remain at or near the colony year-round. Diet - Fish i s the main component of the diet of Double-crested Cormorants although invertebrates may be taken (Lewis 1929). On the west coast of North America, Ainley et al. (1981) suggested that these cormorants feed mainly on schooling f i s h found over f l a t substrates. Yet Robertson (1974) found that gunnels (Family Pholidae) made up over 50 % of the nestling diet by mass and over 45 % by number on Mandarte Island. Gunnels are benthic, c r y p t i c a l l y coloured and are usually s o l i t a r y (Lamb and Edgell 1986). Study si t e s I studied the effects of the time of breeding on clutch size, nestling growth and the diet of Double-crested Cormorants at three colonies (May to October 1993 and May to September 1994) within the southern S t r a i t of Georgia, B r i t i s h Columbia, Canada 7 (Fig 1.1). Cormorants were studied on the Fraser River, Five Finger Island and Mandarte Island for 3 reasons: i) the timing of breeding d i f f e r e d between the 3 s i t e s ; i i ) colonies were widely-di s t r i b u t e d throughout the S t r a i t of Georgia; and i i i ) the colonies could be ea s i l y accessed. I also made periodic v i s i t s to 4 other colonies within the s t r a i t to record the timing of breeding and band nestlings (June to September 1993 and 1994). These colonies were chosen to provide a more complete picture of the time of breeding of Double-crested Cormorants within the St r a i t of Georgia and to help develop a population model of Double-crested Cormorants within the S t r a i t of Georgia. Fraser River - Double-crested Cormorants nest on man-made structures at or near the delta of the south arm of the Fraser River, B.C. (Fig 1.1). There are two main nesting areas, one near Sandhead and the other between the Tsawwassen Ferry Terminal and Westport Coal Terminal. Between the j e t t i e s , cormorants nested on two towers. In 1993, there were 52 nests and only 26 nests i n 1994 due to netting being placed on portions of the towers. Only 12 pairs of cormorants nested on the platforms at Sandheads during t h i s study. Five Finger Island - Five Finger Island i s 8 km NE of Nanaimo, B.C. (Fig 1.1). Nesting occurs on the north-west side of the island. Breeding was f i r s t recorded here for Double-crested Cormorants i n 1959 (Campbell et al. 1990). Over the past 8 years, 8 Figure 1.1. Locations of Double-crested Cormorant colonies, within the S t r a i t of Georgia, where growth and diet selection (closed c i r c l e s ) and changes i n the number of breeding pairs (open and closed c i r c l e s ) were studied i n 1993 and 1994. 9 Kilometers 10 t h i s colony has increased from 138 nests i n 1989 to 378 nests i n 1994 . Mandarte Island - Mandarte Island i s 2 5 km ENE from V i c t o r i a , B.C. (Fig 1.1). Nesting occurs along the south side of the islan d (see Drent et al. 1964 for a s i t e description) . The colony increased i n size from a low of about 150 pairs i n 1960 and peaked at 1100 pairs i n 1983. Since this time, the colony size has declined: only 458 and 403 nests were found i n 1993 and 1994 respectively. Other Sites - Chain Island, C h r i s t i e I s l e t , Crofton and Hudson Rocks were also v i s i t e d during t h i s study (Fig 1.1). Breeding was f i r s t recorded for Double-crested Cormorants at these colonies i n 1959, 1941, 1987 and 1987, respectively (Vermeer et a l . 1989, Campbell et al. 1990) . Chain Island, an ecological reserve, i s a low rocky island where Double-crested Cormorant numbers showed a 3 f o l d increase between 1983 and 1987 (Vermeer et al. 1989). C h r i s t i e I s l e t , located i n Howe Sound, i s a migratory b i r d sanctuary. Nesting occurs along the top of the b l u f f s and along the north side of the island (Sullivan 1985). The number of nesting pairs has steadily declined from 718 nests i n 1978 (Sullivan 1985) to 145 nests i n 1989 (Vermeer et al. 1989) . The colonies at Crofton are located on 2 man-made structures approximately 1 km N of the town of Crofton. A maximum of 75 to 80 nests have been recorded for this colony due to a limited amount of nesting space. Hudson Rocks, 0.5 km SW of Five Finger Island, 11 i s a group of small rocky i s l e t s Nesting occurs mainly along the number of nesting cormorant pairs given year (I. Moul, pers. comm.). with l i t t l e or no vegetation, south side of the i s l e t . The i s usually less than 40 i n any 12 Chapter 2. The timing of breeding of Double-crested Cormorants and i t s e f f e c t s on clutch size Introduction There are a number of hypotheses which predict how many eggs a female lays: i) a female produces as many eggs as she i s phy s i o l o g i c a l l y capable of laying; i i ) a female produces as many eggs as she can incubate; iii) clutch size matches the mortality rate of the population; and iv) a female produced as many eggs which produce the maximum number of surviving young (Lack 1968). The la s t hypothesis, proposed by Lack (1954), i s currently considered to best r e f l e c t the determination of clutch size (Perrins and Birkhead 1983) . Lack (1968) believed that the a b i l i t y of the parents to provision t h e i r young was the ultimate factor a f f e c t i n g clutch size. As the number of young exceed the provisioning a b i l i t y of the parents, less food i s delivered to the nestlings. This results in decreased fitness of the surviving young. It has been shown that late breeding i s associated with a seasonal decline in clutch size in single-brooded species (Klomp 1970). This decline i s considered to be an adaptation to declining food supplies (Perrins and Birkhead 1983). Physiologically, there are 3 types of layers i n birds: determinate; semi-determinate; and indeterminate. The clutch size of determinate layers i s equal to the number of developing f o l l i c l e s , semi-determinate layers usually have one additional 13 d e v e l o p i n g f o l l i c l e w h i l e the number of d e v e l o p i n g f o l l i c l e s g r e a t l y exceeds the c l u t c h s i z e i n i n d e t e r m i n a t e l a y e r s (Haywood 1993) . The c l u t c h s i z e o f both semi-determinate and i n d e t e r m i n a t e l a y e r s i n r e g u l a t e d by e x t r i n s i c f a c t o r s , n o r m a l l y the t a c t i l e s t i m u l a t i o n from the eggs a l r e a d y l a i d . Thus, the d e t e r m i n a t i o n of c l u t c h s i z e can become v e r y c o m p l i c a t e d s i n c e both p h y s i o l o g i c a l and e c o l o g i c a l f a c t o r s p l a y a r o l e . I n t h i s c h a p t e r , I examine the c l u t c h s i z e o f D o u b l e - c r e s t e d Cormorants w i t h r e s p e c t t o the time of b r e e d i n g . Based upon the s e a s o n a l d e c l i n e shown by Klomp (1970), I p r e d i c t t h a t the c l u t c h s i z e o f D o u b l e - c r e s t e d Cormorants s h o u l d d e c l i n e as the b r e e d i n g season p r o g r e s s e s . Methods The e f f e c t of the time of b r e e d i n g on the c l u t c h s i z e of D o u b l e - c r e s t e d Cormorants was determined a t the F r a s e r R i v e r , F i v e F i n g e r I s l a n d and Mandarte I s l a n d i n 1993 and 1994. The number of eggs per n e s t was determined a t or near the time o f h a t c h . On the F r a s e r R i v e r , almost the e n t i r e c o l o n y was sampled. On F i v e F i n g e r I s l a n d and Mandarte I s l a n d , a l a r g e p o r t i o n o f the c o l o n y was sampled. Only n e s t s which' had eggs were used i n t h e a n a l y s i s . The time o f b r e e d i n g ( l a y i n g date) was c a l c u l a t e d by s u b t r a c t i n g 27 days from the time a t h a t c h . S t a t i s t i c a l analyses - A l l s t a t i s t i c a l a n a l y s e s were performed u s i n g SYSTAT 5.1 ( W i l k i n s o n 1990). C l u t c h s i z e s were 14 compared among pos t - h o c t e s t , was determined c o l o n i e s u s i n g a n a l y s i s of v a r i a n c e The e f f e c t of season ( l a y i n g date) u s i n g l i n e a r r e g r e s s i o n . and a Tukey's on c l u t c h s i z e Results Timing of breeding: The t i m i n g o f b r e e d i n g was d i f f e r e n t among a l l c o l o n i e s and a l l y e a r s ( F i g 2.1) . The mean l a y i n g date (± sd) o f eggs on F i v e F i n g e r I s l a n d was 10 May (± 3.6 d) i n 1993 and 22 May (± 2.4 d) i n 1994. On Mandarte I s l a n d , the mean l a y i n g date was 25 June (± 3.6 d) i n 1993 and t h e r e was almost a complete f a i l u r e i n b r e e d i n g i n 1994. On the F r a s e r R i v e r , the mean l a y i n g date was 9 August (± 2.7 d) i n 1993 and 25 J u l y (+ 3.6 d) i n 1994. O v e r a l l , t h e r e was a range o f 103 days from f i r s t t o l a s t l a i d egg. Clutch size: The c l u t c h s i z e of D o u b l e - c r e s t e d Cormorants d i f f e r e d s i g n i f i c a n t l y (F = 8.62, df = 4,606, p < 0.001) among c o l o n i e s and yea r s (Table 2.1), but no s e a s o n a l d e c l i n e was found (F = 2.60, df = 1,3, p = 0.21). The c l u t c h s i z e on Mandarte I s l a n d was s i g n i f i c a n t l y l a r g e r (p = 0.018) than a t a l l o t h e r c o l o n i e s (Table 2.1) even though the time of b r e e d i n g was median t o a l l c o l o n i e s ( F i g . 2.1). The c l u t c h s i z e on the F r a s e r R i v e r (1993) was s i g n i f i c a n t l y s m a l l e r (p < 0.001) than a t a l l o t h e r c o l o n i e s and was the l a t e s t b r e e d i n g attempt ( F i g 2.1). 15 Figure 2.1. The mean (± sd) and range of laying dates of Double-crested Cormorants, in 1993 and 1994, at 3 colonies within the S t r a i t of Georgia (FFI = Five Finger Island, MAND = Mandarte Island and FRSR = Fraser River). 16 FRSR 93 FRSR 94 >• O MAND 93 O O FFI 94 FFI 93 15 April 15 May 15 June 15 July 15 Aug 15 Sept Laying Date 17 4-1 O 4-> -H rO M 4-1 CO (U 4-1 4-> CO Q) - H c o rH O U CD CD U Xi 4-> 4-1 m CO 4-1 o e M O U TS CD 4-> CO • CD CO U -H ° r—I H CO 4-1 "H O 4-> - H CD M N PQ - H CO -"J i O 0> d o rH CD CM CD rH X! ro TJ to +1 CD N •H W .3 U +> H U C CD +> n) Q bi C •rl >i Hi C XI CM XI CM XI O 0 CM rH rH o rH rH +1 +1 +1 +1 +1 OO rH CM ro ro 00 CM co CM co CM co UO > 1 rc ro 2 CM CM CD C d ID LO CM O O CM CM u CO 00 oo rd CTl en cn cn (D CTl cn cn cx> rH rH rH rH rH TS -a a C (0 (0 rH rH T J CO CO 1—1 H (0 rH M H H M CO CD CD CD CD H > > cji 0 • H • H C C CD - H - H 4-> H H SH c ro CD CD 0 CD CD T5 CO CO H > > C (0 r  0 • H - H rO U M U DM S U-> d LO CM 4-> CO d 0i d co a CO CD 4-> CD 6 X! 4J - H H rO co (0 TS CD 4-> C CD CO CD H Q4 CD SH 'rO CO +J rO TS 4-1 CD SH CD 4-1 4-1 IT) O O V 4-> CO O - H 4H • H C 0i CD S-1 CO SH CD 4-1 4-1 CD 4-1 C CD M CD 4H 4H >i X) TS CD O O 4H CO C CO CD Discussion The timing of breeding: Egg l a y i n g was i n i t i a t e d between l a t e A p r i l and mid May on b o t h F i v e F i n g e r I s l a n d and Mandarte I s l a n d i n 1993 and 1994, s i m i l a r t o the t i m i n g of egg l a y i n g i n the 1960's (Drent e t al. 1964) and 1970's (Robertson 1971). On Mandarte I s l a n d , the c o m p l e t i o n of c l u t c h e s was d e l a y e d due t o d i s t u r b a n c e s by B a l d E a g l e s and humans. Cormorants were r e p e a t e d l y f l u s h e d from t h e i r n e s t s and eggs were depredated by both N o r t h w e s t e r n Crows and Glaucous-winged G u l l s . D i s t u r b a n c e s o c c u r r e d d a i l y throughout A p r i l and May and began t o t a p e r o f f i n l a t e June and e a r l y J u l y . In 1993, the c o m p l e t i o n o f c l u t c h e s was d e l a y e d by 45 days. In 1994, the p r e d a t i o n p r e s s u r e d i d not s u b s i d e and t h e r e was almost a complete b r e e d i n g f a i l u r e : o n l y 6 young were produced from 403 n e s t i n g a t t e m p t s . On the F r a s e r R i v e r , the cause of the d e l a y i n b r e e d i n g i s unknown. At t h i s c o l o n y , D o u b l e - c r e s t e d Cormorants began n e s t b u i l d i n g i n e a r l y May (pers. obs. ) but eggs were not found u n t i l l a t e J u l y (1994) and e a r l y August (1993) . No s i g n s o f depredated eggs were found and I d i d not w i t n e s s any d i s t u r b a n c e s i n e i t h e r y e a r . Laying dates and clutch size: O v e r a l l , no s e a s o n a l d e c l i n e was d e t e c t e d i n the c l u t c h s i z e o f D o u b l e - c r e s t e d Cormorants among c o l o n i e s (Table 2.1). C l u t c h e s on Mandarte I s l a n d were s i g n i f i c a n t l y l a r g e r than on F i v e F i n g e r 19 Island even though breeding began 46 days l a t e r . The lack of a seasonal decline i s further supported by the s i m i l a r i t y i n clutch size on Mandarte Island, between the 1970's (3.9 ± 0.5 eggs), when breeding occurred i n early May (Robertson 1971), and i n 1993 (3.8 ± 0.7) when breeding occurred i n mid June. The only seasonal effect detected was on the Fraser River when the clutch size i n 1993, the l a s t colony where breeding occurred, was s i g n i f i c a n t l y smaller than i n 1994. Thus the clutch size of Double-crested Cormorants did not follow my prediction of declining clutch size as the breeding season progressed. Ecological determination of clutch size: Lack (1954) postulated that clutch size i s ultimately determined by the adults' a b i l i t y to provision t h e i r young. Once the clutch size exceeds the provisioning a b i l i t y of the adults, undernourished young w i l l suffer higher rates of mortality and fewer w i l l survive to reproduce. While Double-crested Cormorants can lay many eggs within a single breeding season, the modal clutch size i s 4 eggs. Clutches of 5 eggs occur less than 3 % of the time at any one colony (T.Sullivan unpub. data). However, Robertson (1971) provides data which does not support Lack's hypothesis. Robertson (1971) enlarged Double-crested Cormorant broods and showed that they could successfully raise up to 7 young. He found that larger broods did not suffer higher post-fledging mortality rates u n t i l a brood size of 7 was reached. He concluded 20 that broods of 6 were the most productive and that these super-normal broods (5 to 7 young per nest) contributed more progeny to the breeding population than broods of 4 young. Thus, i t appears that Double-crested Cormorants could successfully provision more young than they currently do. In recent years, Lack's clutch size hypothesis had been altered to include the l i f e time reproductive success of the adults when considering the number of young produced. Thus, super-normal broods may be more productive but i t can be argued that the extra work load would reduce the chances of survival for the adults (Drent and Daan 1980). To date, this has not been tested i n Double-crested Cormorants. Clutch size: an alternative approach: Robertson's (1971) data shows that Double-crested Cormorants can successfully raise more young than they normally produce. The modal clutch size i s 4 eggs with few nests holding more eggs. In addition, the hatching, success of 5 egg clutches tends to be low resul t i n g i n brood sizes with 4 or less young produced (Robertson 1971, pers. obs.). This leads me to believe that the clutch size of Double-crested Cormorants i s limited not by the i r a b i l i t y to raise more young, but by their a b i l i t y to incubate more eggs. Double-crested Cormorants do not have a brood patch but incubate th e i r eggs on the tops of their feet (Drent 1975). Thus, the ultimate factor l i m i t i n g clutch size may involve the size of th e i r feet. Since these cormorants are foot propelled pursuit 21 divers (Ashmole 1971), increasing the incubation surface may affect t h e i r a b i l i t y to forage e f f i c i e n t l y . Thus, the lack of a seasonal decline in clutch size may not i n i t i a l l y r e f l e c t the provisioning a b i l i t y of the adults but instead r e f l e c t t h e i r i n a b i l i t y to incubate larger clutches. Since the incubation surfaces of Double-crested Cormorants are r e s t r i c t e d to the size of feet, I propose the following hypothesis: Double-crested Cormorants are limited i n the number of eggs that they can successfully incubate by the surface area of th e i r feet. Summary 1) While laying dates ranged from early May to early August, no seasonal decline i n clutch size was observed. 2) While clutch size may ultimately be determined by the a b i l i t y of the parents to feed their young (Lack 1954), I believe that the clutch size of Double-crested Cormorants i s also r e s t r i c t e d by thei r i n a b i l i t y to incubate larger numbers of eggs. 3) If the provisioning a b i l i t y of the adults does affect the clutch size of Double-crested Cormorants, i t i s not u n t i l the l a t e r stages of the breeding season. 22 Chapter 3. Growth of nestling Double-crested Cormorants i n r e l a t i o n to the time of breeding Introduction In chapter 2, I showed that the onset of successful breeding was delayed on both Mandarte Island and the Fraser River. In 1993, breeding was delayed by a month and a half on Mandarte Island and 3 months on the Fraser River compared to breeding on Five Finger Island. Clutch size did not decline seasonally, i n fact the clutch size on Mandarte Island was s i g n i f i c a n t l y larger than at a l l other colonies. In many species, young produced l a t e r i n the season have a lower mass than those produced e a r l i e r i n the season (Perrins and Birkhead 1983). Typically, these young suffer higher mortality rates than those produced e a r l i e r i n the season (Daan et al. 1988) . Natural selection may act on the development of body components to maximize the chances of survival (O'Connor 1977). Lightbody and Ankney (1984) found that ducklings of late breeding Lesser Scaup could f l y at an e a r l i e r age than those Canvasback ducklings, produced e a r l i e r i n the season. Unfortunately, they were unable to make comparisons within a single species. The structural size of a nestling may or may not affect i t s chances of survival. Large, heavy bodied Snow Geese do not have an advantage i n breeding over li g h t e r , smaller bodied birds (Cooke et al. 1995). Thus, i t i s important not only to measure the nestling 23 mass but also the development of the structural components. In t h i s chapter, I determine the growth (mass, culmen length, t a r s a l length, overall structural size and condition index) of nestling Double-crested Cormorants at a l l 3 colonies. I examine the effect of laying date on growth and correlate nestling mass with the amount of daylight and a i r temperature. Materials and Methods I measured the growth of nestling Double-crested Cormorants on Five Finger Island, Mandarte Island and the Fraser River i n 1993 and 1994. To minimize predation of eggs and young by Glaucous-winged Gulls, nestlings were measured at night on Five Finger Island and Mandarte Island. Nestlings on the Fraser River were measured during the day since depredation by g u l l s was not a problem. Nests were selected at each colony to provide a range of brood sizes and laying dates. Nests were selected throughout the colony including those at the centre and periphery. Nests along c l i f f faces were used only i f they could be safely accessed. Nestlings were i n d i v i d u a l l y marked with a small numbered web tag (#5 f i n g e r l i n g tag, National Band & Tag Company, Kentucky) and t h e i r mass, culmen and tarsus lengths were repeatedly measured throughout the breeding season. I attempted to measure nestlings every 3 days, a period deemed not to be too disruptive to the cormorants yet providing s u f f i c i e n t data points for growth curve analysis. In order to minimize c h i l l i n g of the nestlings, I did 24 not measure nestlings during high winds, rain, or on unusually-cold nights. Measurements followed methods outlined i n Mineau et al. (1982). B r i e f l y , the culmen was measured from the t i p of the b i l l to the edge of feathers on the top of the b i l l . The tarsus length was measured by folding the leg p a r a l l e l to the body and measuring from the d i s t a l end of the tarsometatarsus to the d i s t a l end of the t i b i o t a r s u s . Mass was measured with 100 ± 0.5, 500 + 5, 2500 ± 25 and 5000 ± 50 g pesola® scales. The culmen was measured with a vernier c a l l i p e r + 0.05 mm and the tarsus was measured with a ru l e r ± 0.5 mm. Laying dates were converted to Gregorian dates (1 = 1 January). Mass and structural size - Log i s t i c growth equations were used to describe the relationship between mass, structural size and nestling age. The l o g i s t i c equation was the best approximation of growth using the methods outlined i n Ricklefs (1968) . B r i e f l y , the sigmoid growth curves for known-aged young were converted into a straight l i n e using conversion factors for the l o g i s t i c , Gompertz and von Bertalannfy growth equations. The conversion factors which produce a li n e a r relationship indicate the best equation to describe the growth relationship. The l o g i s t i c growth equation i s : y = A/(l + {be-k')) (3.1) , U-i) and b = - (3.2) i where A i s the asymptotic mass (g) , culmen length (mm) or t a r s a l 25 length (mm) , i = the mass/size at hatching, k = the l o g i s t i c growth rate (day-1) and t = nestling age (d) (Ricklefs 1983) . Ageing nestling Cormorants - For those nestlings (n = 149) whose exact hatching date was not known, the age was calculated using culmen length measurements from 59 known aged chicks from a l l colonies. A l o g i s t i c growth curve was developed (equations 3.1 and 3.2) and the age at f i r s t measurement of a l l chicks was determined by comparing i t s culmen length against t h i s relationship. The equation of th i s relationship, from day 0 to day 15, was: ( A ) i , V Culmen J In Age = f (3.3) - k where A= 62.71 mm ; b= 5.74 and k= 0.129 (day"1) Growth rate - The l o g i s t i c growth rate (k i n equation 3.1) was determined for mass, culmen and t a r s a l lengths. Length of the growth period - The time i n t e r v a l for growth from 10% to 90% of the asymptote, a measure of the length of the growth period, was calculated as: ( C 9 0 - C 1 0 ) I0-90 (dW.Jdt) K ] where C 9 0 and C 1 0 are the l o g i s t i c conversion factors when the mass/size i s 90% and 10% of the asymptotic mass and dWi/dt i s the slope of the l i n e tangent to the growth curve at the point of i n f l e c t i o n (Ricklefs 1968). 26 Structural Size - The structural size was determined by-using a p r i n c i p l e component analysis of the culmen and tarsus lengths, multiplying each of these parameters by i t s loading factor and summing them. Condition Index - The condition index of nestlings was determined by dividing the asymptotic mass of each nestlings by i t s o v e r a l l structural size. Mass and environmental factors - At each colony, the mean asymptotic mass was compared to: i) day length (min) at time of hatch; i i ) day length at the time of fledging; i i i ) the mean monthly a i r temperature (°c) at hatch; and iv) the mean monthly a i r temperature at the time of fledging. Day length (min) was determined using sunrise/sunset tables for Vancouver, B.C. (Environment Canada). Mean monthly temperatures (°c) were determined using monthly summary tables (Environment Canada). S t a t i s t i c a l analyses: A l l s t a t i s t i c a l analyses were performed using SYSTAT 5.1 (Wilkinson 1990) or SAS 6.09 (SAS Institute Inc 1989). The l o g i s t i c growth equation (3.1) was f i t t e d for mass, culmen and tarsus using non-linear regression. The structural size of nestlings was determined using p r i n c i p a l component analysis of t a r s a l and culmen lengths. The time interval for growth from 10 to 90 % (t 1 0. 9 0) was determined using l i n e a r regression. A Pearson's cor r e l a t i o n was used to determine the association between the 27 mass, the amount of daylight and the a i r temperature at hatch and at fledging. The process for determining the differences i n the asymptotic parameters of growth, growth rates and time i n t e r v a l for growth from 10 to 90 % followed 4 steps: 1) The data were compared using an analysis of variance to determine the effect of colony, brood size, p o sition within the brood, laying date and any interactions among variables. The analysis of variance s t a t i s t i c a l model i s : Y = u + C +B +(B/P) +C*B +LD +C*LD+B*LD+e (3.5) ijkl i j j(k) ij ijkl i j ijkl i = 1,2,3,4,5; j = 1,2,3,4; k = 1,2,3,4 where Y = the growth parameter: mass (g) , culmen (mm) , tarsus (mm) , structural size (mm) and condition index (g*mm_1) , C colony, B = brood size, B/P = the position of the chick nested within the brood size, (C*B) = the interaction between colony and brood size, LD = the laying date of the chick as a covariate (d), C*LD = the interaction of colony and laying date and B*LD = the int e r a c t i o n of brood and laying date. The ANOVA model was modified for 3 comparisons: 1) for the length of the mass growth period (d) , both the mass and l o g i s t i c growth rate were included as covariates; 2) for the length of the culmen growth period (d), the asymptotic value was included as a covariate; and 3) for the length of the tarsus growth period (d), the asymptotic value was also included as a covariate. 28 If significance was detected for the main effects, a Tukey's post-hoc test was used for mean separation. If significance was detected only for the covariate, l i n e a r regression was used to determine the relationship between the variables. S i g n i f i c a n t interactions between main effects and the covariate could not be tested i n t h i s manner and the following method was used. 2) If s i g n i f i c a n t interactions were found, the regression approach to analysis of variance was used. B r i e f l y , t h i s method i s used to test, using a p a r t i a l F test, whether regression equations d i f f e r among the main ef f e c t s . For example, i f a s i g n i f i c a n t interaction was found between colony and laying date, t h i s method w i l l test whether the regression equations d i f f e r among individual colonies as laying date increased. The p a r t i a l F test i s : p t- (SSE (with colony)-SSE (withoutcolony))Ino. coeff'. res. MSE (with colony) where SSE = sum of square error, no. coeff. res. = the number of c o e f f i c i e n t s i n the model (colony and colony by laying date interactions) minus the number of c o e f f i c i e n t s i n the model without colony, and MSE = mean square error. 3) Since a regression equation has both a slope and an intercept, a s i g n i f i c a n t difference i n the p a r t i a l F may be due to differences i n either parameter. Using methods outlined by Neter 29 and Wasserman (1974), these parameters were tested among colonies using pairwise comparisons. Test s t a t i s t i c s for Scheffe's Multiple Comparison were calculated f i r s t for the slopes of the equations and then for t h e i r intercepts. If the slopes were significantly-d i f f e r e n t , the intercepts could not be tested. Test s t a t i s t i c s for Scheffe's Multiple Comparison test were calculated using the formula: s'bi + s'bj-lisbibj) K 1 where b± and ht = the c o e f f i c i e n t s to be tested, s 2 ^ and s 2bj = the variances for the co e f f i c i e n t s and s b ^ = i s the covariance for the two c o e f f i c i e n t s . The c r i t i c a l value S2C was calculated as: Sc2 =(*"!) »-*-.,.-« (3.8) where n = the sample size, k = the number of groups to be tested and F = the F test s t a t i s t i c . 4) If any of the differences among the slopes or intercepts could be attributed to one colony, i t was isola t e d from the rest of the colonies for each dependent variable. The data was then compared using the TANOVA model for both the single colony and the remaining colonies. Results A N O V A tables for the various s t a t i s t i c a l tests, the results 30 of the p a r t i a l F tests and the results of the Scheffe's Multiple Comparison tests for slopes and intercepts are presented i n Appendix I. Unless otherwise stated, comparisons of difference among colonies are based on least squared means presented i n Appendix I. Mass: Asymptotic Mass - The asymptotic mass of nestling Double-crested Cormorants (Table 3.1) was s i g n i f i c a n t l y and negatively related to laying date (p = 0.01) and was unaffected by any other factor (Table 1.1) . As the breeding season progressed, nestling mass declined (Fig 3.1). This relationship was best described by the l i n e : Asymptotic Mass = -5.96* Laying Date + 3466.58 (3.9) (F= 87.76, df= 1,201, p< 0.0001 r 2= 0.30) In 1993 and 1994, nestlings on Five Finger Island were 16.4% and 22.9% heavier, respectively, than nestlings produced l a t e r on the Fraser River (Table 3.1) . Lo g i s t i c Growth Rate for Mass (LGR^gJ - There was a si g n i f i c a n t difference (p = 0.0004) among colonies for LGRMass (Table 3.1). LGRMags also s i g n i f i c a n t l y regressed (p = 0.001) on the asymptotic mass of the nestlings. The i n i t i a l analysis indicated that there was a si g n i f i c a n t (p = 0.002) interaction between colony and laying date which may be responsible for the differences detected among colonies. The slope and intercept of 31 ro En Pi En ro cn cn H En cn cn H En En CN +1 CQ CJ) 5 r o ' + i r H m CN r o +1 co CJ i n CN CN +1 > i cd S CN CN r o +1 (0 S 0) 4J rd Q cn 0 • H cd CN CN CN +1 CO r o o o cn J 5 t o ro r- a ro O ra CN ro CN ro H s> r o o r o r o H r H H • o • • o CN O r o CN H o CM CN CN +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 o U> H r o o o o o r o <a H O L O CN r> CN i r o O in r o m CN o CN CN m CN m c n • ja CN r H i - l r H CN • O +1 +1 CN r-c n i - H m • c n o H CN a r H CN ^ O CN • o +1 +1 r o < CN c n CN o • r o o CN • ro r > c n CN o r o • o +1 + i i — l CN r- • r o O CN r-a 0 0 r H u> r H CN O +1 +1 c n CN r H • m o CN cd CN +1 U5 CN 5 03 CQ rd S u •rH 4-) o — CXi m CQ O < : J >1 rd 0 0 r H +1 o o +1 cn cn • r H ro ro ro ro CN ro r H O • • • « c n • CN o r H r H r H +1 +1 +1 +1 + i +1 r- r o o r o 0 0 • • CN • ^ r o • r o c n o CN *P r H r H ro l£> ro r o H o ro i n ro 0 0 ro r o o ro r> O r H o r o o o r o +1 +1 +1 +1 +1 +1 +1 C N o v o i n C N i n o r o r H 0 0 V O C N r H C N O C N 0 0 o C N c n ro r o ro i n o ro i n ro r -ro o rt r H r o r H r H o 0 0 o +1 +1 + i + i + i +1 +1 i n r> i n 0 0 r> r H 0 0 m r o o r H r o 0 0 C N o o C N ro i n ro V D a m o ro c n ro ro O ro 0 0 C N r H o r H o r o + i +1 + i +1 + i +1 +1 m H C N i n i n C N C N i n r H O ^ r o o 0 0 C N o H C N 4J Cn a CD CJ CD 3 u rd Di O rd T) Cn CJ CD r H CQ 0 CQ SH cd EH >i rd cd xi T j — +1 cn CN o o r H +1 r-r o cn H + i o o o CN r H +1 0 0 0 0 CN CN +1 0 0 r H H +1 m cn 6 B B B CD N CD - H T) CO CJ | | r H rd CJ JH 0 • H 4J 4J o •rH 0 T i SH a 4J 0 CO u L D O V ft 4-) CJ <D U CU M H iw • H Ti > i r H 4 J CJ rrj U TJ - H (U 4-1 4 J • H rrj Cn CQ • rH CQ CU CQ QJ • H U rd rH CD CQ JCJ rH 4 J d) 0 4-) 4 J CQ CU CQ r H 0) r H 4 J CJ ^3 a) rH CQ cu CJ rd 0 CQ • H e CJ rrj > i T S CD CU £ rH rd U • H cu CT4J CQ CU o B r H 4 J rC j r H CQ 4 J 0 rd • H 4 H cu rH r H rrj CQ CJ CQ CQ m rrj rrj (U T S T S CU CD 4 J 4-) CJ CJ CD CU CQ CQ CU CU r H U u 0 ft ft u CU CU CJ rH u - r H rrj rrj X J 4-) rrj rrj • H 4 J 4 J rrj rrj ja T S T S ro CN Figure 3.1. Asymptotic mass of nestling Double-crested Cormorants from Five Finger Island (FFI) i n 1993 (0) and 1994 (•) , Mandarte Island i n 1993 (+), and the Fraser River i n 1993 (V) and 1994(X). 33 2500 h Y- -5.96 X + 3466.6 x FRSR 1994 v FRSR 1993 + MAND 1993 • FFI 1994 O FFI 1993 15 April 15 May 15 June 15 July 15 Aug 15 Sept Laying Date the regression of LGRMass on laying date for Five Finger Island 1993 (with the fastest LGRMass) and Fraser River 1994 (with the slowest LGRMass) were s i g n i f i c a n t l y d ifferent from those of the other 3 colonies (Scheffe's multiple comparison tests, Tables 1.3 and 1.4) . While the LGRMass of nestlings from these two colonies (FFI 1993 and FRSR 1994) were affected d i f f e r e n t l y by laying date, the LGRMass did not regress s i g n i f i c a n t l y on laying dates when these two colonies were analysed i n d i v i d u a l l y (Tables 1.7 and 1.9) . At the 3 remaining colonies, LGRMass regressed s i g n i f i c a n t l y (p = 0.0001) on the asymptotic mass of the nestlings (Table 1.8). Heavier nestlings had a lower growth rate (LGRMass) than l i g h t e r nestlings. This relationship was best described by the l i n e : LGR =-0.0001* Asymptotic Mass + 0.33 (3.10) Mass F = 60.88, df = 1,120, r 2= 0.34 Since no data was collected on Mandarte Island i n 1994, the year effect on LGRMass could not be examined at a l l colonies. However, the results indicate that Double-crested Cormorants which hatched i n 1993 had a much faster LGRMass (0.64, 0.60 and 0.22 on Five Finger Island, Fraser River and Mandarte Island, respectively) than those hatched i n 1994 (0.19 and 0.17 on Five Finger Island and Fraser River) (Table 1.3). Time taken to grow from 10 - 90 % of the Asymptotic Mass (t 1 0. 90 Mass) " There was a s i g n i f i c a n t difference (p = 0.003) among 35 colonies for t 1 0 _ 9 0 M a s s (Table 3.1), and t 1 0 . 9 0 M a s s also s i g n i f i c a n t l y regressed on both LGRMass (p = 0.0001) and the asymptotic nestling mass (p = 0.009). As well, there was a s i g n i f i c a n t i n t e r a c t i o n between colony and laying date (p = 0.009). The slope and intercept of the regression of t 1 0 . 9 0 M a s s on laying date at Five Finger Island 1994 was s i g n i f i c a n t l y d i f f e r e n t (p < 0.05) than those from both Mandarte Island 1993 and Fraser River 1994 (Scheffe's multiple comparison test, Table 1.12). After removing Five Finger Island 1994 from the analysis, t 1 0 . 9 0 M a s s s i g n i f i c a n t l y regressed (p = 0.0001) on LGRMass and tended (p = 0.06) to regress on the asymptotic mass of nestlings (Table 1.13). As expected, as the rate of growth increased, the time taken to achieve 10 - 90% of the asymptote declined. This relationship was best described by the l i n e : t = -67.44 * LGR +34.80 (3.11) 10-90 Mass Mass (F = 303.11, df = 1,161, r 2 = 0.65) On Five Finger Island (1994) , t 1 0 . 9 0 M a s s was s i g n i f i c a n t l y affected by nestling position nested i n brood size (p = 0.005) and tio-so Mass s i g n i f i c a n t l y regressed (p = 0.0001) on LGRMass (Table 1.14) . The time taken to achieve t h i s period of growth i n the second nestling position i n broods of 3 was s i g n i f i c a n t l y shorter (p < 0.002) than i n a l l others (Table 1.15). No differences were detected among a l l other nestlings i n a l l other brood sizes nor 36 was there any distinguishable pattern exhibited. As the rate of growth increased, t 1 0 _ 9 0 M a s s declined as was found at a l l other colonies (equation 3.11). When a l l other factors are held constant, the relationship between LGRMass and t 1 0 _ 9 0 M a s s was best described by the l i n e : t = -85.51 *LGR +38.48 (3.12) 10-90 Mass Mass (F = 38.87, df = 1,29, r 2= 0.57) Culmen Length: Asymptotic Culmen Length - There was a s i g n i f i c a n t difference among colonies (p = 0.001) for culmen length (Table 3.1) and a s i g n i f i c a n t interaction between colony and laying date (p = 0.001). The slope and intercept of the regression of culmen length on laying date for the Fraser River 1993 (with the smallest culmen length) was s i g n i f i c a n t l y different than at a l l other colonies (Scheffe's multiple comparison test, Tables 1.17 and 1.18) . On the Fraser River 1993, culmen length s i g n i f i c a n t l y regressed (p = 0.007) on laying date (Table 1.20). As the laying date progressed, the asymptotic culmen length increased and was best described by the l i n e : Culmen Length = 01.87* Laying Date - 394.93 (3.13) (F = 31.92, df = l,26,r 2= 0.55) 37 At a l l remaining colonies, no s i g n i f i c a n t difference was detected (p > 0.05) among a l l factors (Table 1.19). Thus laying date was only related to the asymptotic culmen length on the Fraser River 1993. In the o r i g i n a l analysis, the least squared mean for the Fraser River 1993 showed a negative value, an adjustment due to the relationship between culmen length and laying date at thi s colony. The arithmetic mean (± sd) i s actually larger at thi s colony than at a l l others, 68.0 ± 6.9 mm and 63.1 + 4.5 mm respectively. Regardless of thi s difference, the outcome of thi s comparison i s that culmen length i s s i g n i f i c a n t l y affected by laying date. L o g i s t i c Growth Rate for Culmen (LGR^^) - There was a si g n i f i c a n t difference among colonies(p = 0.0001) and brood sizes (p = 0.01) for LGR^^en (Table 3.1). As well, s i g n i f i c a n t colony by laying date (p =• 0.0001) and brood size by laying date (p = 0.01) interactions were detected for LGR^ ,,,^ , which may explain differences found for both colony and brood size. LGR^^ also s i g n i f i c a n t l y regressed on the asymptotic culmen length of nestling Double-crested Cormorants (p = 0.0002). The slope and intercept of the regression of LGR^^ on laying date for Five Finger Island 1993 (with the fastest growth rate) was s i g n i f i c a n t l y different from a l l other colonies (Scheffe's multiple comparison test, Tables 1.22 and 1.24). After removing Five Finger Island 1993 from the analysis, neither colony 38 nor laying date were si g n i f i c a n t factors for the remaining 4 colonies. At these 4 colonies (Five Finger Island 1994, Mandarte Island 1993 and Fraser River 1993 and 1994), there was a si g n i f i c a n t (p = 0.03) colony by brood size interaction for L G R c u ^ n (Table 1.26). LGRCulmen also s i g n i f i c a n t l y regressed (p = 0.0001) on the asymptotic culmen length of the nestlings. In 1993, no s i g n i f i c a n t difference was found i n LGRCulmen among brood sizes but a s i g n i f i c a n t difference was found between years (Table 3.2, Fig 3.2A). -? Table 3.2. Comparisons of the l o g i s t i c growth rate for culmen among colonies i n broods from 1 to 4 young. Colony Year 1 Brood 2 Size 3 4 Fraser River 1993 0 16a 0 .21a 0 .14a 0 . 18a Mandarte Island 1993 0 14a 0.14a 0 .15a 0 . 14a Fraser River 1994 0 09b 0.15a 0 .10b 0 . 13a Five Finger Island 1994 0 l l a 0.04b 0 .13a 0 . 07b represent s i g n i f i c a n t differences among brood sizes In 1994, patterns of growth d i f f e r e d among brood sizes. On the Fraser River, nestlings• i n broods of 2 and 4 had s i g n i f i c a n t l y faster rates of growth than broods of 1 or 3 (Fig. 3.2B). On Five Finger Island, nestlings i n broods of 2 and 4 birds had slower rates of growth than broods of 1 and 3. The patterns of growth on the Fraser River (1993 and 1994) were similar between years (Fig 3.2A and 3.2B). 39 Figure 3.2. Relationship between r a t e c i r o e n (± sd) for nestlings: A) and the Fraser River (+); and B) and Five Finger Island (•) . brood size and l o g i s t i c growth i n 1993 on Mandarte Island (o) i n 1994 on the Fraser River (*) 40 0.30 > 0.25 • 0.20 a E | 0.15 O o » 0.10 0.05 MAND 1993 FRSR 1993 0.25 a a E O u O at o> o 0.15 h 0.05 -0.05 FRSR 1994 FFI 1994 41 At these 4 colonies, nestlings with a larger culmen had slower growth rates than those nestlings which had a smaller culmen, as was found on Five Finger Island 1 9 9 3 (equation 3 . 1 5 ) . This relationship was best described by the l i n e : LGR = -0.02 * Culmen Length + 0.29 ( 3 . 1 4 ) Culmen F = 1 3 5 . 2 1 , df = 1 , 1 3 8 , r 2= 0 . 5 0 On Five Finger Island 1 9 9 3 , L G R ^ ^ s i g n i f i c a n t l y regressed (p = 0 . 0 2 ) on the asymptotic culmen length of nestlings (Table 1 . 2 8 ) . Similar to the relationship between L G R ^ ^ n and culmen length for the other 4 colonies, nestlings with a larger culmen had slower growth rates ( L G R ^ ^ ) than those with a shorter culmen. This relationship was best described by the l i n e : LGR = -0.02 * Culmen Length + 0.28 ( 3 . 1 5 ) Culmen F = 8 . 8 1 , df = 1 , 3 0 , r 2 = 0 . 2 3 The time taken for culmen length to grow from 10 - 90% of i t s f i n a l length ( t 1 0 . 9 0 - The t 1 0 . 9 0 did not d i f f e r among colonies (Table 3 . 1 ) but s i g n i f i c a n t l y regressed on both laying date (p = 0 . 0 4 ) and L G R C u l m e n (p = 0 . 0 0 0 1 ) (Table 1 . 2 9 ) . As the laying date progressed, the time taken to achieve 10 - 90% of the asymptotic culmen length increased. This relationship was best 42 represented by the l i n e : t 10-90 Culmen = 0.02* Laying Date + 27.45 (3.16) (F = 1.61, df = 1,140, r = 0.01) As the LGR, "Culmen increased, the time taken to achieve t h i s period of growth declined. This was best represented by the l i n e : t = -237.65* LGR + 64.76 (3.17) 0-90 Culmen Culmen (F = 319.80, df = 1,140, r 2= 0.70) Tarsal Length: Asymptotic Tarsal length - There was a s i g n i f i c a n t difference (p = 0.01) among colonies (Table 3.1) as well as a si g n i f i c a n t colony by laying date interaction (p = 0.009). The slope and intercept of the regressions of asymptotic t a r s a l length on laying date d i f f e r e d s i g n i f i c a n t l y between the Fraser River 1993 (with the smallest tarsus) and a l l other colonies (Scheffe's multiple comparison test, Tables 1.31 and 1.32). When analysed separately, no s i g n i f i c a n t differences were found i n t a r s a l lengths among birds at the remaining colonies, however, t a r s a l length tended (p = 0.09) to regress on laying date on the Fraser River 1993 (Table 1.34). As the laying date progressed, the asymptotic length of the tarsus also increased i n 43 size. This relationship was best described by the l i n e : Tarsal Length = 0.64* Laying Date - 71.96 (3.18) (F = 9.26, df = l,26,r 2= 0.26) The arithmetic means (± sd) of the asymptotic t a r s a l length of nestlings on the Fraser River 1993 was larger (86.2 ± 3 . 4 mm) than at a l l other colonies (85.9 +2.8 mm). Lo g i s t i c Growth Rate for Tarsus (LGRTarsus) - The LGRTarsus did not d i f f e r among colonies (Table 3.1) but s i g n i f i c a n t l y regressed (p = 0.0001) on the asymptotic t a r s a l length of the nestlings (Table 1.35). Nestlings with a larger t a r s a l length had slower growth rates (LGRTarsus) than those with smaller t a r s a l lengths. This relationship was best represented by the l i n e : LGR = -0.07* Tarsal Length + 0.83 (3.19) Tarsus (F = 39.20, df = 1,165, r 2= 0.19) The time taken to achieve 10 - 90% of the t a r s a l length ( t 1 0 . 9 0 T a r s u s ) - T n e tio-go Tarsus did not d i f f e r s i g n i f i c a n t l y (p > 0.05) among colonies (Table 3.1) (Table 1.36). Structural s i z e : Asymptotic Structural Size - There was a s i g n i f i c a n t difference (p = 0.0001) among colonies for the structural size 44 (Table 3.1), as well as a s i g n i f i c a n t interaction (p = 0.0001) between colony and laying date. The slope and intercept of the regression l i n e r e l a t i n g structural size to laying date for the Fraser River 1993 (with the smallest size) was s i g n i f i c a n t l y d i f f e r e n t (p < 0.05) from those at a l l other colonies (Scheffe's multiple comparison test, Tables 1.38 and 1.39). On the Fraser River (1993), size s i g n i f i c a n t l y regressed (p = 0.007) on laying date (Table 1.41). Here, the asymptotic size of the nestling increased as the laying date progressed. This relationship was best described by the l i n e : Structural Size = 1.99* Laying Date - 3 63.5 0 (3.20) (F = 42.16, df = l,25,r 2= 0.63) On a l l remaining colonies, no s i g n i f i c a n t differences were detected among nestlings (Table 1.40). The timing of breeding was only related to the structural size of nestlings on the Fraser River colony (1993). The arithmetic means (± sd) of the asymptotic size of nestlings on the Fraser River 1993 was larger (131.2 ± 7.0 mm) than at a l l other colonies (127.5 ± 5 . 4 mm). Condition Index: The condition index of nestling Double-crested Cormorants (Table 3.1) s i g n i f i c a n t l y regressed (p = 0.006) on laying date and 45 was unaffected by a l l other factors (Table 1.42). As laying date progressed, the condition index declined (Fig 3.3). This relationship was best described by the l i n e : Condition Index = -0.04* Laying Date + 27.65 (3.21) (F = 52.59, df =1,139, r 2= 0.27) In 1993 and 1994, the condition index of nestlings on Five Finger Island were 19.0 and 24.4 % larger, respectively, than nestlings produced l a t e r on the Fraser River (Table 3.1). Growth and environmental conditions: The nestling mass was highly correlated with day length at hatch (r = 0.961, p = 0.028) but not at the time of fledging (r = 0.862, p = 0.181). Nestling mass was not correlated with a i r temperature at hatch (r = - 0.469, p = 1.0) nor at the time of fledging (r = 0.684, p = 0.610). Discussion: Laying date asserted a s i g n i f i c a n t influence on many of the parameters measured. In general, young produced e a r l i e r i n the season were s i g n i f i c a n t l y heavier, had a slower l o g i s t i c growth rate and required more time to achieve t h e i r f i n a l mass than young produced late i n the season. In 1993 and 1994, nestlings on Five Finger Island were 16.4% and 22.9% heavier than nestlings on the Fraser River (Table 3.1). 46 Figure 3.3. Condition index of nestling Double-crested Cormorants from Five Finger Island (FFI) i n 1993(0) and 1994 (•) , Mandarte Island i n 1993 (+), and the Fraser River i n 1993 (V) and 1994(X). 47 x FRSR 1994 v FRSR 1993 + MAND 1993 • FFI 1994 O FFI 1993 15 April 15 May 15 June 15 July 15 Aug 15 Sept Laying Date 48 Furthermore, laying date appears to be r e l a t i v e l y more important as the laying date progresses. On the Fraser River (1993), nestling mass declined as the laying date progressed, but these nestlings grew more rapidly. Their asymptotic culmen length, t a r s a l length and overall size were greater than those nestlings at a l l other colonies. While no seasonal effect was detected among a l l other colonies, both the culmen length and ov e r a l l size increased s i g n i f i c a n t l y as the season progressed on the Fraser River (1993) . The ta r s a l length also increased but i t was 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 (p = 0.09). As the culmen length increased, the LGR^^ declined and i t took longer to achieve 10-90% of the asymptote. A larger tarsus also had a lower LGR T a r s u d but the time taken to achieve 10-90% of the asymptote did not d i f f e r . There was a seasonal decline i n the condition index of nestlings at a l l colonies. In 1993 and 1994, nestlings on Five Finger Island had, on average, a condition index 19.0% and 24.4% greater than nestlings on the Fraser River (Table 3.2). The growth rates of nestling Double-crested Cormorants within the S t r a i t of Georgia are higher than those found i n other studies (Table 3.3), presumably because albociliatus i s a larger sub-species of P. auritus. In addition, these values appear to be higher than those predicted by Ricklefs (1973). He also found that as the asymptotic mass increases, the growth rate declines. 49 Table 3.3: A comparison of growth rates for nestling Double-crested Cormorants. Subspecies K1 (day"1) t -10-90 (Days) Asymptote (g or mm) Reference Mass auritus 0 214 20.2 1900 Dunn 1975 auritus 0 1762 nd 1889 Leger and McNeil 1987 auritus 0 235 nd 1650 Cleary 1977 auritus 0 196 22 . 5 1900 Palmer 1962 a l b o c i l i a t u s 0 175 nd 2406 Robertson 19713 a l b o c i l i a t u s 0 207 20 . 5 2328 Five Finger Island 19944 Mandarte Island 1993 Fraser River 1993 a l b o c i l i a t u s 0 216 20.3 2373 Five Finger Island 19934 a l b o c i l i a t u s 0 269 16 . 9 1959 Fraser River 19944 Culmen auritus 0 124 nd 61. 0 Dunn 1975 a l b o c i l i a t u s 0 136 31.3 64 .1 Five Finger Island 19944 Mandarte Island 1993 Fraser River 1993 Fraser River 1994 a l b o c i l i a t u s 0 . 145 31.4 63 .3 Five Finger Island 19934 Tarsus auritus 0 .201 nd 80 . 8 Dunn 1975 a l b o c i l i a t u s 0 .230 21.5 85 . 9 Five Finger Island 19934 Five Finger Island 1994 Mandarte Island 1993 Fraser River 1993 Fraser River 1994 1 K i s given as l o g i s t i c K, 2 K was determined using Gompertz equation and converted to l o g i s t i c K 3 parameters estimated from graph, best estimate of growth estimated by eye 4 data from th i s study nd = no data 50 The advantages and consequences of late breeding on the Fraser River (1993): Nestlings produced on the Fraser River 1993 weighed 22% less than nestlings produced early i n the season on Five Finger Island i n 1993 (Table 3.1), probably because the nestlings at the Fraser River colony stored less f a t . The asymptotic lengths of structural components of nestling at the Fraser River colony 1993 increased as the season progressed, a pattern not observed at other colonies. More importantly, this pattern was not observed at the Fraser River colony i n 1994 when breeding occurred e a r l i e r i n the season. This strongly suggests that these patterns result not from colony differences but due to the lateness of breeding. Seasonal differences i n growth have also been recorded i n other studies. Calageros (1996) found a seasonal decline i n the f i n a l mass of nestling Mallards, which had been fed ad libitum. Janiga (1992) also found a seasonal decline i n mass of pigeons and suggested that i t allowed nestlings to leave the nest sooner than e a r l i e r hatched young. Smart (1965) found that the primaries of late hatched Redheads {Aythya americana) emerged a week e a r l i e r and allowed nestlings to fledge i n a shorter time than early hatched young. As well, Lightbody and Ankney (1984) found that late hatched Lesser Scaup fledged 6 days, faster than early hatched Canvasback. The differences observed i n the growth of both the mass and the structural components suggests that these may be adaptations 51 for late breeding. Loading c o e f f i c i e n t s are determined by dividing the mass of the b i r d by i t s structural size, thus a lower mass divided by an even larger structural size produces a much lower loading c o e f f i c i e n t which may f a c i l i t a t e f l i g h t at an e a r l i e r age. East of the Rockies, the mortality rates of juveniles are highest i n December and January. Assuming that the timing i s si m i l a r west of the Rockies, this i s 2 months afte r the fledging date on the Fraser River (1993) and 6 months afte r the fledging on Five Finger Island. Therefore, i t may be advantageous to nestlings produced late i n the season to fledge e a r l i e r and learn to forage before the onset of adverse winter conditions. A reduction i n mass i s usually considered maladaptive since i t r e f l e c t s a decrease i n stored energy reserves which the b i r d could use i n times of poor food a v a i l a b i l i t y or adverse weather conditions (Montevecchi et al. 1984). There may, however, be benefits to reduced body mass. Loworn and Jones (1991) showed that high body fat raised the cost of diving by increasing work against drag, buoyancy and increased acceleration. Therefore, l a t e r fledged birds may benefit from reduced costs associated with diving and/or f l y i n g . Whether the reduction i n f l i g h t and foraging costs outweighs the benefits of higher energy stores i s unknown. Laying date and growth - possible mechanisms: Growth i s a complex process i n which many factors play a part (Ricklefs 1983). These include genetic and environmental factors and interactions among them. Environmental factors include 52 n u t r i t i o n , temperature, and effect of photoperiod on hormone concentrations. Nestling mass and the length of daylight at hatch were highly correlated. The seasonal decline i n nestling mass pa r a l l e l e d a seasonal reduction i n day length. The seasonal decrease i n daylight may l i m i t the amount of time that adults are able to forage (see chapter 4) . Alternatively, there may be a hormonal response which affects the mass of the nestling. A seasonal decline i n the concentration of growth hormone (GH) was found i n male Peking Ducks Anas platyrhynchos (Scanes et al. 1980) and male and female Red Grouse Lagopus lagopus scoticus (Harvey et a l . 1982); a seasonal decline i n thyroid hormone (TH) concentration was found i n male Peking Ducks (Scanes et al. 1980). Both growth and thyroid hormones are important for growth (Ricklefs 1983) and reductions i n plasma concentrations of either GH or TH results i n reduced growth (GH: Scanes and Lauterio 1984; TH: King and May 1984, McNabb 1988). The amount of nutrients that a nestling receives can a l t e r i t s rate of growth and f i n a l mass and size (Pond et al. 1995) . Nutrients include l i p i d s , protein (amino acids), carbohydrates, minerals, vitamins and water. The role of diet and nutrient composition, as factors affecting growth, are considered i n chapter 4. Nestling mass and condition index: alternative explanations There are two alternative explanations for the observed 53 differences i n nestling mass, which affects condition index: 1) asymptotic nestling mass i s s i t e s p e c i f i c ; and 2) Double-crested Cormorants at each colony are genetically d i f f e r e n t . The lack of difference between years i n nestling mass on the Fraser River and Five Finger Island suggests that there may be a s i t e s p e c i f i c factor. This may r e f l e c t genetic differences among colonies or differences i n the l o c a l environments. Summary 1) A seasonal decline i n the asymptotic mass of nestling Double-crested Cormorants was found. 2) There were no difference i n the structural size as the breeding season progressed except for the colony which bred l a t e s t i n the season, Fraser River 1993. 3) Relative to other colonies i n the S t r a i t of Georgia, nestling at the Fraser River 1993 decreased i n mass but increased i n t h e i r s t ructural size as the season progressed. This may be a response to allow nestlings to fledge at an e a r l i e r date by reducing the physical loading on the bird. 4) Nestling mass was highly correlated with length of daylight but not with a i r temperature. 54 Chapter 4. Composition and n u t r i t i o n a l content of the d i e t s fed to n e s t l i n g Double-crested Cormorants Introduction In chapter 2, I showed that clutch size did not decline seasonally, although the l a s t colony to breed (Fraser River 1993) had a s i g n i f i c a n t l y smaller clutch size than a l l other colonies. In chapter 3, I showed that the mass and condition index of n e s t l i n g Double-crested Cormorants declined seasonally so that nestlings produced early i n the season (Five Finger Island) were 22.0% and 16.9% heavier, i n 1993 and 1994 respectively, than nestlings produced late i n the season (Fraser River) . As the season progressed, the amount of daylight decreased from 968 (Five Finger Island) minutes to 790 minutes (Fraser River) . While the cause of the seasonal decline i n mass (Fig 3.1) i s unknown, the decline was highly correlated (r = 0.96, p < 0.03) with the amount of daylight at hatch (Chapter 3) . This suggests that foraging time, or some other factor influencing growth, i s affected by declining photoperiod. Nestling growth or growth patterns can be modified by changing either the quantity (Batchelor and Ross 1984, Bertram et al. 1991, Barrett and Rikardsen 1992) or the q u a l i t y of food (Batchelor and Ross 1984, Heath and Randall 1985) fed to the chick. Thus the composition of the prey available to parents foraging at the d i f f e r e n t colonies could be responsible for the 55 observed growth patterns. L i t t l e i s known about the diet of nestl i n g Double-crested Cormorants or the n u t r i t i o n a l composition of the prey species i n general. Robertson (1974) provides some q u a l i t a t i v e information on t h e i r diets on Mandarte Island. B r i e f l y , he found that nestlings were fed mainly gunnels (Family Pholidae) and Shiner Perch {Cymatogaster aggregatta). In t h i s chapter, I: 1) describe the diets fed to nestl i n g Double-crested Cormorants at three colonies within the S t r a i t of Georgia; 2) describe the n u t r i t i o n a l composition of the prey species; and 3) compare the n u t r i t i o n a l composition of the diets and comment on i t s effects on growth. Materials and Methods Diet: The diet of nestling Double-crested Cormorants {Phalacrocorax auritus albociliatus) was determined i n 1993 at colonies on Five Finger Island, Mandarte Island and the Fraser River using bolus analysis. I used bolus analysis for 3 reasons: i) i t i s a non-destructive technique; ii) i t represents the diet of n e s t l i n g at that time (Hobson et al. 1989); and i i i ) i t has l i t t l e or no perceived effect on the nestlings i f ca r r i e d out late i n the breeding season (pers. obs.). The disturbance generated by my entering the colonies during daylight hours caused the nestlings to regurgitate t h e i r food (referred to as a bolus) and was coll e c t e d for analysis. 56 On both Five Finger and Mandarte Islands, boluses were co l l e c t e d i n a single day when the nestlings were approximately 30 days old. On the Fraser River, boluses were c o l l e c t e d over a 3 day period, when nestlings were approximately 20 to 30 days old. This time was chosen since the nestling cormorants were large enough to defend themselves against predatory g u l l s and crows. Each bolus was i n d i v i d u a l l y bagged, rinsed i n sea water to remove surface digestive enzymes and debris, and stored at -20°C. Prior to analysis, samples were rinsed i n fresh water. Each bolus was weighed on a Mettler PE 3600 e l e c t r o n i c balance (± 0.005 g) and teased apart. The number and t o t a l mass of each prey species was determined. Fish fragments were i d e n t i f i e d to species and the number and mass was determined. Gunnels, including Penpoint Gunnels (Apodichthys flavidus) , Crescent Gunnels [Pholis laeta) and Saddleback Gunnels (Pholis ornata), were grouped together since many were p a r t i a l l y digested and sometimes d i f f i c u l t to i d e n t i f y to species. U t i l i z a t i o n of prey species among colonies - Each prey species was expressed as percent of the t o t a l mass, t o t a l number and frequency of occurrence at each colony. While each of these measurements could be used separately to determine the most important prey item, each has i t s own inherent bias (Hyslop 1980) . Therefore, I used the r e l a t i v e importance index (RII) to minimize the biases (Hyslop 1980) . The RII was determined for each colony using: 57 RU,= All, (4.1) n 1=1 and All, = %TN + % TWM + % FO (4.2) where RII = the r e l a t i v e importance index for species i through n; A l l = the absolute importance index for species i ; n = the number of f i s h species; % TN = the percent of the t o t a l number of f i s h ; % TWM = the percent of the t o t a l wet mass for each species and % FO = frequency of occurrence (the percentage of boluses i n which the species was found). N u t r i t i o n a l analysis: The n u t r i t i o n a l composition ( l i p i d , protein, ash, and gross energy) was determined for 13 samples of f i s h , representing 7 species, from the 3 colonies. Fish were chosen from a l l samples so that the analyses of the diet would be as representative as possible. Each f i s h species was homogenized at a high speed for 5 minutes i n a blender, s h e l l freeze-dried (Labconco Freeze Dryer No. 18, Kansas City, Mo.) for 24 h, and then ground to a powder using a coffee grinder. Duplicate sub-samples were used to determine percent dry matter, and content of l i p i d , ash and gross energy. Since f i s h contain less than 1 % carbohydrate (R. Beames, Dept of Animal Science, UBC, pers. comm.), protein 58 content was determined by subtracting the percent l i p i d and ash from the t o t a l sample (% protein = 100% - % l i p i d - % ash). The energy content was determined using an isothermal bomb calorimeter (Leco Automatic Calorimeter model AC-300), standardized with benzoic acid and corrected for nitrogen content. The l i p i d content was determined using Goldfisch extraction (Labconco Corp., Kansas City, Mo.) with anhydrous ethyl ether. The ash content was determined by heating the samples at 600 °c i n a muffle furnace (Thermolyne Furnatrol 133, Syborn Corp.) for 12 h. Data are presented as means. The water composition of each bolus was also calculated. The fresh water content of each species was calculated by multiplying the mass of each prey species by i t s dry matter estimate. This value was then subtracted from the o r i g i n a l mass. The metabolic water (mis) of each bolus was calculated by multiplying the l i p i d content (g) by 1.071 and the protein •content (g) by 0.396 (Schmidt-Nielsen 1964). The t o t a l dietary water included both the preformed water i n the f i s h and water produced by metabolism. Nutrient composition of the boluses - The nutrient composition of the boluses was compared among colonies. The composition of each bolus was determined by multiplying the mass of each f i s h species i n the bolus by i t s dry matter, and multiplying t h i s product by i t s n u t r i t i o n a l composition, i . e . % protein, % l i p i d , % ash and gross energy content. For those colonies where the species of f i s h was not analyzed, data from 59 another colony was used. If more than one analysis was performed on the species, the mean value for a l l analyses was used. If no analysis was performed on the f i s h species, the average of a l l f i s h at that colony was used. Provisioning, the timing of breeding and n e s t l i n g mass: The e f f e c t s of nestling provisioning on nest l i n g mass was determined. F i r s t , assuming equal amounts of food were delivered to the nestlings i t s ef f e c t on nestling mass was compared. Second, the effects of available foraging time, i . e . the amount of daylight at hatch, and the quantity of food delivered was compared with nestling mass. Lastly, the ef f e c t s of both the amount of foraging time and the number of nestlings (brood size) on the quantity of food delivered was compared with nes t l i n g mass. The amount of foraging time was calculated by determining the length of daylight at the time of hatch using sunrise/sunset tables for Vancouver, B.C. (Environment Canada). Day length at hatch was used since i t was highly correlated with n e s t l i n g mass (Chapter 3). Since growth requires energy and most animals must feed to meet energetic requirements (Pond et al. 1995), only gross energy was used i n t h i s analysis. The t o t a l gross energy was compared among colonies by expressing i t as a percentage of the colony with the largest gross energy content. Assumptions - In t h i s analysis, I have made 4 assumptions: 60 i) each bolus represents a single feeding; ii) the reduction i n foraging time i s proportional to the decrease i n the length of daylight; i i i ) the prey provide a l l nutrients required for normal growth and iv) the bolus samples are representative of the entire season. S t a t i s t i c a l Analyses: A l l s t a t i s t i c a l analyses were performed on SYSTAT 5.1 (Wilkinson 1991). An analysis of variance (ANOVA) was used to compare the n u t r i t i o n a l components of the boluses. A Tukey's post-hoc comparison was used for separation of groups. Significance was accepted at a = 0.05. Results Nestling d i e t s : A t o t a l of 1,139 i n d i v i d u a l f i s h were c o l l e c t e d from nestl i n g Double-crested Cormorants i n 17.4 kg of f i s h , representing 305 boluses (Table 4.1). Occurrence of prey species among colonies - Nestling cormorants were fed almost i d e n t i c a l prey species among colonies (Appendix 2), however, prey composition d i f f e r e d dramatically at each colony (Table 4.2): Gunnels were found 8.5 times more often on Mandarte Island than at any other colony, P a c i f i c Sandlance were found 2.3 times more often on Five Finger Island and P a c i f i c Staghorn Sculpins were found 2.1 times more often on the 61 T S CU 4-1 o cu rH U H rH O O U 4J - H CD CT5 N M - H 4-> CO CO CO 0> rH 4-> O XI £ -H rrj co CD CD e - H c T S O G rH rd o O x: CO CD •H CD 4H H X 4H 4-> O 4-> SH m CD X! CO e 4J C m SH v o CO g CO M rO O E O rH T S ro CD 4-> 4-1 O co 4-> CD M - U CO 1 CD CD CO rH 3 X) rH 3 O O . X Q oo cn 4-1 tnCTi O C rH •H M rH C CD 4-1 -H X) co £ CD rO -H c tn g M CD O O XI M CD E H 4-) O CD rH X) rO 0) ftJ co H O PQ 1 m m rrj S to H O pq a m I <P s +> CD +> CO rd S +i Q +) tn •H t (11 ro rd r - 00 LO cn CO o o rH +1 +l +1 rH cn rH CM CM LO LO rH rH rH m ro ro 00 VO CO o CO 00 00 LO +1 +1 +1 rH cn rH o LO 00 CO LO CO LO LO LO W •H 0 55 to w rd B -tn H X +> O C 0 •H -p rd U O H >1 O H 0 U cn o o »H 00 00 0 ro VO rH o cn LO • • • CO cn VO cn LO T S C CO rH T S CO d. 1—1 rO rH M SH CO CD CD H > tn •H £ CD - H 4-> b-> H SH rO 0 CD T S co > C (0 -H ro H S LO o o A a, CO CD - H C o rH o O tn o CO T S CD 4-1 U CD 4-1 CD T S CO CD O a CD SH CD 4H 4H 4-> C rO O -H 4H -H C tn O CM VO TS 0 - P CO 0) u O I CU rH X! d O • Q ro CTi trim d rH -H rH iH 4-1 -H CO cu rd G -H Cn O SH -P O CU T3 C J 0 m m o CO 4-1 •H -H m ro SH o CO H 0) > •rl DI rd M b (U |P TJ _ rd nj , , fa H S M CD Cn. ^1 CO o cn L O CN P O L O L O L O co co C M C M o o o L O o CD C M co C M ro ro co CD co r H ° <-* L O r H C O ro r H ^ r H C M C O r H r-co co ro ro o C M — 0 ^ ^ H X! CO CO 1—1 4-> 3 rd 3 CC 4-J + J -LJ — c CQ 4 J <a -H CD •H X -H tn .—. SH CD CO 3 fcn CO rd .—. X) CD U rd 3 CO G -H rd CO SH CO 3 -H G ra CD 3 4-J O CO -U CO 4-J 4-J rd CD rH 3 rd 3 ^ 4-J -H O O CU fcn G rd 0 G O CO OH 4-J CD CU X U CD rd CO SH CU O 4 J 4-1 CD fcn CO O £; - G -U CO H CD G 3 O SH CD CO CD ro CO CO a x : H 3 4 J — CD g - P J2 -d CO SH 4-J - G -H ttq CJ rd CD ^ + J - P d CJ 4 J u d - G CU rd rd — CO rd 0 -H CJ > CD -d rd X! ft •H -H CO SH O tn rH rH -U 4-> 4-> 3 o rd 0 CD CO d rd CO G •~H 0 X! -u M • — u -H rH rd CD rd U to CO CO CM CU H •—- O rH £• •H CD CD -—-H O — O SH CD u 4 J d g O CU rd d CO SH SH 0 SH d d •H — ' T5 rd rd o CD X! O •H 0 4-1 CD •H rH si TS EH CJ SH g CO si M - H TS CD d SH rH o rd 0 d CO rd d CU rd CU SH d r? rd — 4-1 0 CM DC CO d -H Pe CO CM CO si CO Fl CO U O a o 0 CO u CD -H •H to SH •H rH •H •H •H >! CD •rl m m CU CU m CU 4H m 4H SH rH 0 -H -H CD d -H d -H ^ •H SH X) CD u o SH •H u d U o O rd rd Oi rd rd si si rd d rd o rd - P EH w CM EH CO D-i a CM ce CU CO TS d rd CD 4H 4 H O O O d rd SH > , - P 0 U O g CD ft d d g d tr •H CD r H SH CD rd MH d O CD rH +-> Si O - P CO 0 X! Si || rd - P LLJ O Si ° - P ^ cAP II § T J M SH ^ ( H 0 < ^ 0 c 73 0 a, • & » " P E H U SH rd r d o y 0 0 • v SH •H CO O 0 4H CO 0 o 0 ^ r^ co si co SH o 4H O CO 0 CO a rd co g -p 0 0 TS d 0 u d 0 rd si -p -p SH O II ft g d -H 0 > CO 4-1 0 rd -H rH O . 0 0 ft. rd - P O - P 0 si -p 4 H o -p d 0 u SH 0 ft 0 Si -p II co S II 0 S SH EH ce - P O d 14H o\o re 0 II SH SH 0 Si d 0 i—i TS CO U Si i—i d - H O 3= cc - H 4H O ro VD Fraser River. The u t i l i z a t i o n of each prey species by mass, number and frequency of occurrence are l i s t e d i n Appendix 2 and 3. The quantity and qu a l i t y of ne s t l i n g diets and n e s t l i n g growth: There was no s i g n i f i c a n t difference among colonies i n either the wet (ANOVA F = 0.109, df = 2, 305, p = 0.9) or dry mass (ANOVA F = 0.289, df = 2, 305, p = 0.7) of the boluses (Table 4.1). N u t r i t i o n a l Analysis - The n u t r i t i o n a l composition of f i s h fed to nest l i n g Double-crested Cormorants i n the S t r a i t of Georgia ranged widely (Table 4.3). In general, most species had a r e l a t i v e l y low l i p i d content (1.23 - 7.45 %) with the exception of P a c i f i c Salmon (11.99 %) and Shiner Perch (23.68 %) co l l e c t e d at the Fraser River colony. The protein content ranged from 62.71 to 85.13 %, gross energy ranged from 19.58 to 25.81 KJ/g and ash ranged from 11.48 to 20.7 %. Nutrient composition of the boluses - There was no s i g n i f i c a n t difference among colonies i n the amount of gross energy (F = 0.202, df = 2, 305, p > 0.8), protein (F = 0.416, df = 2, 305, p > 0.6) or ash (F = 2.806, df = 2, 305, p > 0.06) i n the boluses (Table 4.4). Boluses from the Fraser River had s i g n i f i c a n t l y more l i p i d (F = 5.414, df = 2,305, p = 0.005) than those from Five Finger Island (p = 0.003) but did not d i f f e r s i g n i f i c a n t l y from Mandarte Island (p = 0.144). There was no s i g n i f i c a n t difference (F = 0.246, df = 2, 305 p = 0.8) i n the t o t a l amount of water generated from boluses among colonies (Table 4.5). 64 00 4-> rO cn 4-> d rrj H O e u o o T5 0) 4-1 co CD SH U I CD rH X! d O Q Cn d -H rH 4-1 CO CD d O 4-) CD 4H CO -H UH 4H O ro -H Cn M O 0 O m rO SH 4-1 CO CD r H X! CO 4-1 -H co d >i-H rH X! (0 4-1 d -H rO IS 4-J CO C CD CD -H H 4-> 2 d o rH O o ro 0 rH X ro EH M CD — ti t n CO w o Cn VO ro vo cn 00 o CO rH rH ID O CN 00 cn cn rH VD ro r -CO < dP CM cn J j CM rH rH ro ro o rH rH rH rH rH rH rH rH rH CM n .—. a, dp H v kn-ap §2 co CD •H U CD a, w rH CM O CM ro LO o rH rH OO rH 00 O rH ro r - LO CO o LO rH rH rH co CM CM CM ro LO CM cn CM CM CM CM CM CM CM CM CM CM rH a o •rl -P rd 0 0 h3 CO 1 <~~\ 00 t—s ro O 00 LO U ) t—> vo CO rH CM CM vo VO ro LO r -ro vo rH rH CO rH LO LO r~ CO CO O LO vo o 00 rH CM rH CO r~ CO CO CO 00 CM cn CM VO VO O r - ro rH 00 VO o CM vo LO r - cn CM CM CM CM CM CM CM d o -H rO a X! rH 0 rH o M & CO O OH 0 -H CO o d SH d SH CM C ro O O rH X 0 & X T5 Cn rH U d to rO rO SH to CO 4-> d S^H CO <D CO CO CO OH Cu X! CO O o CO U U •H SH •H -H -H -H rH m 0 m UH 4H VM 0 d -H M -H -H d u -H o U U O d rO X CO O rO rO d P-i CO Cu Oil OH CM o Q Q H H H H H z PM Cu PM PM UH <C PM PM UH PM UH S s cn cn o cn oo vo co CM CM vo oo vo rH r -o ro 0 , CO d o e o -H m •H u to cu X u H 0 cu M 0 d -H X CO ro CO X CO •H m o o CM ro CM vo co vo o CM 00 co r~ CM CM r~ cn co CM CM Oh rH d o co d u o x DH Cn tO 4-> CO o o ro Cu CC cC c£ PC CO CO CO CO CM Pi CC CC PM PM PM MM CO 0 4J rO -H 4-1 CO 0 SH 0 4-> 4-1 to e >1 SH X i d O TS 0 co fO X) co 0 d rH rO > LO vo X! CO rO o\° TS -H DM •H rH o\° I SH 0 o\° 4-> O 4-> O tO rH II SH d Q - H 0 II 4-1 O 2 M Q Cu TS d tO rH TS co d I—I tO rH SH S-1 CO 0 0 H > Cn -iH d 0 Cd •H 4-1 PM SH SH CO 0 0 TS CO > d ro •H rrj H PM 2 PM H Z CO PM < Pi PM 2 PM & u o ti w (0 W o M Cn 43 to Cn C •rl CD -P O M Cn •rl •r l o •rl +) rd O O rH >l ti 0 H 0 CJ ra ro r— LO VD CM LO O CM CO CTl CM rH CM +1 +1 +1 00 CO CO CO CO VD r~ CM rH ro 00 00 CO 00 ro ro ro CM O rH rH co o O o +1 +1 +1 oo r- CM o o VD CM CM CM ro ro ro CTi r-U3 rH O o rH +1 +l +1 CM CTl r-LO rH rH rH CM rH rH rH rH ro ro XI CO CD O o rH O o o +1 +1 +1 CTi LO CO VO CTi rH O O rH Ti a rd rH TS CO c M rd rH H H CO CU CU M > Cn •H a a) CC •H -p &H H SH rd d) CU TS CO > c rd •H rd SH PLI S CO CD • P rd e -H P CO CU SH CU P - P rd >i SH Ti G O TS CD co rd Xi co CU rH rd > 4-1 rd 4-) G CU SH CU m m -H T S >1 rH 4-J G rd o •H m -H G Cn -H CO cu SH rd co SH CD 4-1 JJ CD 4-1 c CD SH CD m 4H -H TS >. X! TS CU o o 4H CO G rd CD C 6 rH 0 U J G U rd LO V ft CD CD t n 4-> co 0) C O 4-> TS 0) m co 0) co rH O X ) g o n U H TS CU o 1 3 O M OH SH d) 4 J CO . cn o 2 o X> (0 4-1 CU g co 4-> C (0 s g fO CU £ T 3 M CD O 4-1 H H CO CU 0) H M D-i u CD CO XJ 4 J d O O EH Q LO CD X ) cO EH U <D •p H rd +J O EH C M H — <D <D rH +> -P g rt) O — o •rl H 0 -p S Cd r j PM g H CD •P g_ CD g B~ o CD U ro rd ro CM CO CM LO CO ro CM +1 +1 +1 CO CM rH rH CM r~ cn VO 00 rH VD VD cn CM o O o +l +1 +1 VD CO CM LO 00 -vr VD o o rH o o o +l +1 +1 ^ rH r~ O CM o rH rH CO rH rH CO CM CM CM +1 +1 +1 CO CO rH CO ro O co o 00 TS C (0 c rH TS 0 co C •H M rO -P rH M rd SH CO CD 0 CD 1—1 > 0 tn •H C CD CC •H 4-1 py M SH a rO CD 0 CD TS CO H > C rO 0 •H rO SH u Cu S CXJ 00 o II Cu co CD -H O rH o U tn C 0 g rO T S CD > H CD CO X ) O CO CD u C CD SH CD UH UH 4-1 C rO U -H C t n -H co O VD The q u a l i t y o f t h e b o l u s e s (grams n u t r i e n t / g r a m s d r y m a t t e r ) showed the same p a t t e r n . There was no s i g n i f i c a n t d i f f e r e n c e among c o l o n i e s i n the amount of g r o s s energy (F = 0.202, df = 2,305, p < 0.8), p r o t e i n (F = 0.42, d f = 2,305, p > 0.6) o r ash (F = 2.81, d f = 2,305, p > 0.06) (Table 4.6). Provisioning, the timing of breeding and n e s t l i n g mass: Assuming n e s t l i n g s were f e d e q u a l amounts of f o o d a t each c o l o n y , t h e o b s e r v e d d i f f e r e n c e s i n n e s t l i n g mass cannot be e x p l a i n e d by f o o d p r o v i s i o n i n g (Table 4.7). N e s t l i n g s on Mandarte I s l a n d would have r e c e i v e d more g r o s s energy t h a n a t any o t h e r c o l o n y , w h i l e n e s t l i n g s on t h e F r a s e r R i v e r and F i v e F i n g e r I s l a n d would have r e c e i v e d 1.5 and 6.2 % l e s s f o o d r e s p e c t i v e l y (Table 4.7). As t h e b r e e d i n g season p r o g r e s s e d , th e amount of d a y l i g h t d e c l i n e d . I n 1993, a d u l t cormorants on F i v e F i n g e r I s l a n d had 968 minutes of d a y l i g h t i n which t o p r o v i s i o n t h e i r young. As t h e season p r o g r e s s e d , the amount of d a y l i g h t d e c l i n e d t o 926 m inutes f o r t h e Mandarte I s l a n d c o l o n y and 790 m i n u t es f o r t h e F r a s e r R i v e r c o l o n y (Table 4.8). I f t h e q u a n t i t y o f f o o d d e l i v e r e d t o t h e n e s t l i n g i s p r o p o r t i o n a l t o t h e amount of d a y l i g h t , t h e r e would be a r e d u c t i o n i n p r o v i s i o n i n g by 4.3 % on Mandarte I s l a n d and 18.4 % on t h e F r a s e r R i v e r (Table 4.8). Due t o t h e h i g h g r o s s energy c o n t e n t of b o l u s e s on Mandarte I s l a n d (Table 4.4), t h e n e s t l i n g s 68 Xi (0 EH U Q s § CO •rl CD +) o rl •rl •rl Hi 2 Q Cn Cn 2 Q Cn Cn Q Cn — Cn CD O ro CM r— ro CTl CD Csl rH CM +1 +1 +1 co CO Csl rH CD CTl CM rH CO O CO <o rH O ro rH O n) CO o O O o +1 +1 +1 CTl rH CTl rH Csl O o O ro r— o ro CD O ro o rH o O O +l +1 +1 O CO rH 00 o rH rH i-H ro rH o XI ro rH O XI rH o o O o +1 +1 +1 CD O CTl O rH rH O O o d rrj c rH TS 0 co d •rl rH rd •P rH SH rrj H co 0 0 CD H > 0 •H Hi d 0 Pi •H p >i SH SH C CO 0 0 CD TS CO H > d CO 0 •H rd SH O [14 p CO P d 0 SH 0 4H P >i rH p d rO u •H P -H d tn •iH CO 0 SH CO CO SH 0 P P 0 P d 0 SH 0 P P •H TS >i X! TS 0 O o p CO d cO 0 d g d rH O U 4d o CO LO 0 o 4d p v •H 2 ft CD CO T > CT CU CJ H -H CU rH > TS O LLj CO i—I CD CU 4-1 G "O ° CD Ti > ^ o -P -P CJ rrj CT CD e cO CO CD xi 4-> 4-> O 4-1 O -P 4H O Ti CD >i H 4-> CD -H > -P -H CJ (0 CT . . M—i ro CT ° CJ 3 T S >i CJ 0 r-l > i T J 4-> CO >1 CJ O rH O o xi o ro CD 4-1 ro CD N -rH CO Ti O O H XI CJ ro CD cu XJ -P TJ rd CU XJ -P 4H O 4-1 -P xi xi CT CT CT-H -H CT rd rd TS Ti to Cu 10 Cu 3 Cn C •rl H •P to CD 55 CT CN CJ — 4H 4H -H CM O o CN LO CJ O O - xi xi H CO • • -rH CO -P -P o rH CO CT CT CT 1 1 -H CJ CJ CJ > -H cu 0 O rH rH rH H -P Cu CO 0 0 CU XJ Xi CU CJ -p -P xi in t o all to to ar ro cn ro cn ro cn in t o all on on Ye cn rH cn rH cn rH CO -P -H -H CJ -P -P O T) H SH CO CD o o -H M T3 SH CD 0 o CJ rd > SH SH ro CVH CU CU CJ rH Ti 6 rH 0 CO CJ O CD CJ CJ •rl i—i rd U Ti •H •H +> rH H m SH CO 0 • • 0 0 H > 0 CT •H • Hi CJ 0 CC -H -P >1 Cu SH SH CU C! rd 0 rH 0 CD TS CO X! H > CJ rd rd . 0 -H rd u EH CJ Cu S UA o o CM LO CM 00 cn rH I CN CM I ^ cn i cn co CM o CN I o o CN I CO CO 00 ro oo cn o o cn LO rH I 0 -P Ti xi co 0 4-i 0 SH CT 0 SH > cd -H O -H R - ' SH rO 0 CJ xi 0 4-1 o CO ,, 0 Ti CO -P CJ 0 rd H ^ J C T > CJ ^ CJ H g - H 0 « CO ^ > 0 SH CJ . 0 o o o -P SH n « Ct.(j> Ti >i CT"H >i SH rd CT Q) cJ SH a o CJ 0 r-l w 0 0 -H H +J » o CO X ^ 4 H 0 Ti § CJ 0 O ^ CO ^ CO " CO H & ® ^ C T c o •y rd cu m '"I Xi M 0 o 0 xi CO O * rd -P < >i Ti Ti CT 0 0 -P SH CJ rd 0 a <u SH 0 CJ e 0 CO • 0 CO CO O CT SH CJ CT-H CO 0 SH 0 0 SH SH rd ^ 'jj o co CO 0 co 0 >i CJ 0 -p -p 3 rO -H 0 rH g 4-1 Xi rd -H CJ -P > -P rd CO 3 O rl 0 CT-P o TS CJ rd rH co SH 0 CT CJ •H Cu 0 > -H Cu Cu Cu 0 X -p o 4-1 0 X rH +J f0 G -a o G "rH fC 4-1 M v O x a o 4-1 rO a . x: cn 4-1 -H ro Cn G Cn  •H O rH -H >i co ro -H -a > o U H SH O C M X! UH 4-1 -H Cn G Cn 0) G rH d O 0) >i X! 4-> CD - 4J 0 N O -rH 4-1 CO Tl X! "d CD u G 4-> O ro -H x! CO -H 4-> > ro 0 O 4-1 M 4-1 ro a x TS Cn TS -H X O rH O O >i 4-) U H rO rO TS X G - H UH G O rO 0 0 G 4-) g -rH G rH 3 0 u o X 0 g H Tl CO OO 0 r  X) C O E H O SH X! a •H (1) c •rl H u CD Q CD TJ •H > _ 0 o\° u — a. o o 43 +) H g & Q o o M PQ Ci rd i M rd CD rH OO VD o VD VD CM LO cn CTl cn cn CO >1 Ci 0 H O U CD +> rd a 45 0 +> rd C rd I! r -o I r o I co rH CO I I rH CM co r- CM 00 00 oo CM CM 00 00 00 cn cn cn cn cn cn cn cn cn cn rH rH rH rH rH T S T S G G (0 ro rH rH T S CO CO G i—i i—i to rH SH SH M SH CO 0 0 0 0 M > > Cn Cn -H -H G G 0 OS Pd -H •rH 4-1 P M P M SH SH SH rO 0 0 0 0 Ti CO CO > > G rO rO -H •H rO SH SH P M P M s P M P M SH 4-) 0 0 >1 CO X) G rH G g G Cn 0 h^ l rO G 4-1 1 J < SOM r—. cn oo 0 1 rH CM VD rH LO 00 CTl cn CM + 0 4J (0 TS Cn G -H >i rO 0 4-) to TS X! O 4-) rO DC co cn cn G -H TS G C O rH CO M SH 0 Cn G -H P M 0 > •H P M 4-1 rO 4J X Cn >i rO TS UH O 4-1 G O 0 X 4-1 G o TS 0 CO rO X TS 0 G SH 0 4-) 0 T S 0 M 0 ro 4-) rO a Tl G (0 rH CO SH 0 Cn G - H P M 0 > 4-> rO co Cn G 4-) CO 0 G UH o CO CO ro 0 X 4-> G O TS 0 CO tO X! TS 0 G -H £ SH 0 4-1 0 TS 0 CO 4-1 ro Q would s t i l l have received the greatest amount of gross energy (Table 4.7) . If the amount of gross energy provided to the young i s proportional to the amount of foraging time and the mean brood size (Table 4.8), nestlings on the Fraser River would have received more gross energy than a l l other nestlings (Table 4.7). Nestlings on Mandarte Island would have received the least amount of gross energy. Provisioning, based upon these factors, does not explain the observed differences i n nes t l i n g mass. Discussion To date, t h i s i s the most complete quantitative and q u a l i t a t i v e analysis of the diets fed to nes t l i n g Double-crested Cormorants i n the S t r a i t of Georgia. As well, t h i s i s also the f i r s t q u a l i t a t i v e analysis for the majority of prey species found i n the boluses. Diets fed to n e s t l i n g cormorants - While the f i s h species fed to nes t l i n g Double-crested Cormorants were si m i l a r among colonies, t h e i r proportions d i f f e r e d . Nestlings on Five Finger Island were fed mainly P a c i f i c Sandlance and Shiner Perch, nestlings on Mandarte Island were fed mainly gunnels while nestlings on the Fraser River were fed mainly P a c i f i c Staghorn Sculpin and Shiner Perch (Table 4.1). The quantity of food delivered, whether measured as wet or dry mass of the boluses (Table 4.1), were not s i g n i f i c a n t l y 72 d i f f e r e n t among colonies. Nor were the amounts of gross energy, protein, ash, (Table 4.4) or water content (Table 4.5) of t h i s food s i g n i f i c a n t l y d i f f e r e n t , based either the t o t a l mass per bolus or on the quality of food (g nutrient/g dry matter) . The amount of l i p i d s i n the boluses did d i f f e r s i g n i f i c a n t l y (Table 4.4), nestlings on the Fraser River received more l i p i d s than nestlings on Five Finger Island. At t h i s l e v e l of analysis, i t appears that neither the quantity nor the quality of the diets explain the seasonal decline observed i n nestling mass (Chapter 3) . If each bolus i s considered as a single feeding, i t may be that the frequency of feedings rather than the quantity of i n d i v i d u a l feedings explains the observed differences i n nestling mass. The diet of these nestlings was higher i n protein, lower i n l i p i d and s i m i l a r i n both ash and gross energy than diets of other f i s h eating birds (Table 4.3, Appendix 5). The n u t r i t i o n a l values of prey species, obtained i n t h i s study, are similar to the limited information available. Vermeer and Devito (1986) , i n the only analysis of prey f i s h from the west coast of B r i t i s h Columbia, found the gross energy of P a c i f i c Salmon was 20.9 KJ/g and values of 19.7 and 22.5 KJ/g for f i r s t and second year old P a c i f i c Sandlance. Their values are s i m i l a r to those found i n t h i s study, 21.14 and 23.50 KJ/g for P a c i f i c Salmon and 21.02 KJ/g for P a c i f i c Sandlance (Table 4.3) . 73 Provisioning, timing of breeding and n e s t l i n g mass Nestling mass was highly correlated with length of daylight at hatch (Chapter 3). As the season progressed, day length declined and n e s t l i n g mass declined. The reason(s) for t h i s may be from a reduction i n foraging time, increased energetic costs associated with late hatching or changes i n hormonal concentrations of the adults or young (Chapter 3) . I w i l l only discuss foraging and the provisioning of young since I do not have any information on the energetic costs or hormonal concentrations. If colonies did not d i f f e r i n the amount of food each breeding pair brought to the nestlings, nestlings on Mandarte Island and Fraser River would have received more gross energy than those on Five Finger Island (Table 4.8). This i s d i r e c t l y opposite to the pattern found i n nestling mass (Table 4.7). Nestlings were on average 6.4% and 21.9% l i g h t e r on the Fraser River and Mandarte Island, respectively, than on Five Finger Island. As day length decreased, the amount of time available to forage may have declined (Table 4.8). If provisioning of the young and the amount of daylight are proportionate, then nestlings on Mandarte Island and the Fraser River would have received 4.3% and 18.4% less food than those on Five Finger Island (Table 4.7). Late breeding cormorants could maintain the amount of food delivered to t h e i r young by either increasing t h e i r foraging e f f i c i e n c y and/or the time spent foraging (Table 4.7). Assuming 74 parents at a l l colonies had the same foraging e f f i c i e n c y , i n 1993, increasing t h e i r e f f o r t by 4.3% on Mandarte Island and 18.4% on the Fraser River would have provided provisioning s i m i l a r to that on Five Finger Island (Table 4.8). Gremillet et al. (1995) reported that Great Cormorants (Phalacrocorax carbo), r a i s i n g downy chicks, spend at least 25 % of t h e i r time resting at the nest. If parents i n the present study spent t h i s time foraging, food delivery to the young at a l l colonies could be si m i l a r . If provisioning of the young were reduced i n proportion to the length of daylight, delivery of gross energy would have been least on the Fraser River and greatest on Mandarte Island (Table 4.8). Gross energy a v a i l a b i l i t y may explain the lower n e s t l i n g mass on the Fraser River, but not on Mandarte Island. Compared to nestlings on Five Finger Island, nestlings on Mandarte Island received more gross energy but had a smaller mass. If provisioning were proportional to both the length of daylight and brood size, nestlings on the Fraser River would receive s u b s t a n t i a l l y more food than a l l other nestlings (Table 4.8). In fact, nestling mass was smallest at the Fraser River colony. There i s no clear pattern between the provisioning and nes t l i n g mass, th i s suggests that other n u t r i t i o n a l or environmental factors are involved. The timing of feeding, the location and the effect of prey densities on foraging rate have not been studied. The d i s t r i b u t i o n and abundance of prey species 75 also changes as the season progresses. How t h i s a f f e c t s foraging i s unknown. Low tides occur during the day i n summer with a gradual t r a n s i t i o n to low tides at night i n winter. If cormorants forage during s p e c i f i c t i d a l heights, the amount of food that a ne s t l i n g receives may have been further reduced proportionate to the amount of daylight that coincides with a s p e c i f i c t i d a l cycle. Prey composition and t h e i r e f f e c t s on growth: The analysis of f i s h fed to nes t l i n g Double-crested Cormorants provides a gauge to the quality of the diet but 2 assumptions are made: i) the metabolizable portion of the diet i s s i m i l a r among f i s h species and i i ) es s e n t i a l f a t t y acids, amino acids and minerals, necessary for normal growth, are found i n the prey items. If either of these assumptions i s incorrect, growth w i l l be affected. Batchelor and Ross (1984) found that r e s t r i c t i n g the diets of Cape Gannets (Sula capensis) , fed diets with high and low metabolizable energy c o e f f i c i e n t (MEC) diets, did not e f f e c t the s t r u c t u r a l size but did affect the mass. This p a r a l l e l s growth results i n the present study. When gannet nestlings were fed low MEC diets ad libitum, they ate sub s t a n t i a l l y more food (26 %) than nestlings on high MEC diets but s t i l l f a i l e d to reach the asymptotic mass of these nestlings. The MECs of Double-crested Cormorants were inversely 76 related to the ash content of the f i s h eaten (Brugger 1993). However, the MECs were calculated on adult birds which regurgitated p e l l e t s containing the bony portions of the f i s h . Nestling cormorants do not produce such p e l l e t s u n t i l they reach fledging age (Ainley et al. 1981). The ingestion of high ash species, i . e . P a c i f i c Staghorn Sculpin and Shiner Perch which predominate i n the diet of nestlings on the Fraser River, may aff e c t the amount of energy available for growth. There may be increased energetic costs associated with processing a high ash species. To date, t h i s has not been investigated. Likewise, a high ash content slows the passage rate, which could reduce the quantity of food that may be ingested. The nutrient requirements for Double-crested Cormorants and most other avian w i l d l i f e are unknown. In growing animals, a reduced intake of any of the essential nutrients retards growth rate and/or body size (Pond et al. 1995). Whole f i s h are almost the sole food source of both adult and nestling Double-crested Cormorants and probably ' provide a diet with s u f f i c i e n t nutrients. U n t i l either the essential requirements for growth are determined or the nestlings are raised on s p e c i f i c prey species, the ef f e c t of the prey species on growth cannot be f u l l y assessed. Summary: 1) Nestling Double-crested Cormorants were fed almost i d e n t i c a l 77 prey species but the proportions of these species d i f f e r e d among colonies. 2) Most prey f i s h had r e l a t i v e l y low l i p i d content (1.2 - 7.5%), high protein content (62.7 - 85.1%) and gross energy ranging from 19.6 to 25.8 KJ/g. The ash content ranged from 11.5 20.7%. •3) As the breeding season progresses, adult Double-crested Cormorants must either increase t h e i r provisioning rate i n order to provide t h e i r nestlings with a set amount of food or the quantity of food would decrease by 4.3% on Mandarte Island and 18.4% on the Fraser River. In 1993, nestlings on Mandarte Island and the Fraser River were 6.4% and- 23% l i g h t e r respectively than those on Five Finger Island (Chapter 3). 4) The diet of nestlings, within the S t r a i t of Georgia, support the findings of others that the main prey of Double-crested Cormorants are non-commercial species (Lewis 1929, Robertson 1971, Ainley et al. 1981, Craven and Lev 1987 and Hobson et al. 1989. 78 Chapter 5. The timing of breeding: i t s e f f e c t s on juvenile s u r v i v a l and population l e v e l s of Double-crested Cormorants Introduction Within the S t r a i t of Georgia, the number of Double-crested Cormorants increased from 203 breeding pairs i n 1959 to 1,981 pairs i n 1987 (Vermeer et al. 1989). This increase was f e l t to be a response to a decline i n persecution and the cessation of egg c o l l e c t i o n . A l t e r n a t i v e l y , or concomitantly, a decline i n contaminant loading may also be responsible for the r i s e i n cormorant populations (J. E l l i o t t , Canadian W i l d l i f e Service, pers. comm.) as has been suggested for cormorants i n the Great Lakes area (Craven and Lev 1987). By 1989, the population had declined to . 1, 326 breeding pairs (Sullivan 1989) and continues to decline (Sullivan unpubl. data) . With l i t t l e persecution or egg c o l l e c t i n g and a further reduction i n contaminant loading (Whitehead 1989), the cause of t h i s decline i s unknown. From 1987 to 1989, changes i n the number of breeding pairs followed d i f f e r e n t patterns at d i f f e r e n t colonies: the number of breeding pairs increased on Five Finger Island and declined on both Mandarte Island and Chain Island (Sullivan 1989). The reason for these changes i s unknown. Three parameters af f e c t population s i z e . These are rates of: 1) reproduction; 2) adult and juvenile mortality; and 3) dispersal (Perrins and Birkhead 1983). Changes i n one or more of 79 these parameters can af f e c t the number of adults returning to breed and/or the number of young recruited into the breeding population. Approximately 70 - 80% of a l l Double-crested Cormorant nesting attempts are successful (pers. obs.) and clutch sizes ranging from 2.2 to 3.8 eggs (Chapter 2). Thus a large number of young are produced each year (pers. obs.). Regardless, the number of breeding pairs continues to decline at most colonies. The qu a l i t y of these young may be a c r i t i c a l factor a f f e c t i n g t h e i r recruitment into the breeding population. In chapter 3, I found a seasonal decline i n both the mass and condition index of nestling cormorants. It may be argued that the f i t n e s s of the young declines as the breeding season progresses and survival rates are adversely affected. In many species, survival rates of the young are highly correlated with both the timing of breeding and the mass at fledging. Nestlings from early clutches or are heavier at fledging survive better than those from clutches that are late or are l i g h t e r at the time of fledging. This has been demonstrated i n the Shag (Phalacrocorax aristotelis) (Harris et al. 1994), Cape Gannet (Sula capensis) (Jarvis 1974), Guillemot {Uria aalge) (Harris et al. 1992), Kittiwake {Rissa tridactyla) (Coulson and White 1958), Manx Shearwater {Puffinus puffinus) (Perrins 1966) Herring Gull (Larus argentatus) (Nisbet and Drury 1972) and Western Gull {Larus occidentalis) (Spear and Nur 1994) . 80 Delayed breeding and/or the seasonal decline i n mass could predispose these young to higher rates of mortality. Thus, the number of breeding pairs could be affected. Delays i n the timing of breeding have occurred on both Mandarte Island (Sullivan 1989) and Chain Island (I. Moul, pers. comm.). My hypothesis i s that the number of Double-crested Cormorant breeding pairs i s affected by the timing of breeding at i n d i v i d u a l colonies. It i s my contention that the time of breeding i s r e l a t i v e l y more c r i t i c a l than the rate of adult mortality, the number of young produced or the rate of dispersal in determining the population regulation at these colonies. The demographics of a species can be studied i n one of two ways, either as a population (a single colony) or as a metapopulation, a larger subset of populations (many colonies), joined by some l e v e l of dispersion (Pulliam 1988). The metapopulation approach has the advantage that factors regulating population levels may vary at d i f f e r e n t locations due to differences i n either b i o t i c or a b i o t i c conditions. Thus further insight may be gained instead of looking at a single s i t e . Colonies can be classed either as sources or sinks: source colonies are those where the annual production of surviving young exceed the annual adult mortality whereas the annual production of surviving young i s less than the annual adult mortality i n sink colonies (Pulliam 1988). As a whole, populations at ind i v i d u a l colonies can be stable, increasing or 81 decreasing while the metapopulation may remain unaffected. In t h i s chapter, I model the population dynamics of Double-crested Cormorants to determine: 1) the effects of the timing of breeding on the changes i n the breeding population at i n d i v i d u a l colonies; and 2) the effects of the timing of breeding within a metapopulation, a l l colonies within the S t r a i t of Georgia. I compare the differences i n the predicted changes i n breeding numbers, based on the timing of breeding, with observed changes i n breeding numbers. Materials and Methods The population model: The model used i n t h i s analysis includes the timing of breeding at i n d i v i d u a l colonies, the reproductive rates of cormorants at these colonies and the rates of adult and juvenile mortality. Population = NAS + (NYP*JSR) (5.1) where NAS = the estimated number of adults surviving from one breeding season to the next; NYP = the number of young produced; JSR = the proportion of juveniles surviving to 3 years of age (%) estimated from a model by Harris et a l . (1994) (equation 5.2), based on the timing of breeding; and (NYP*JSR) = the t o t a l number of juveniles recruited into the breeding population. Early breeding i s defined as egg laying i n late A p r i l to 82 e a r l y May and t h e s u c c e s s f u l r a i s i n g o f young (as o b s e r v e d by Drent e t al. (1964) r e c o r d e d i n t h e e a r l y 1960's). L a t e b r e e d i n g i s t h e s u c c e s s f u l c o m p l e t i o n o f c l u t c h e s and r a i s i n g o f young a month and a h a l f t o t h r e e months a f t e r t h i s p e r i o d . Adult survivorship: At each c o l o n y , n e s t s was counted l a t e i n t h e b r e e d i n g season a f t e r any unused n e s t s had been d i s m a n t l e d and i n c o r p o r a t e d i n t o a c t i v e n e s t s . The t o t a l number o f b r e e d i n g a d u l t s was t h e number of n e s t s m u l t i p l i e d by 2 o f which 15 % were p r e d i c t e d t o p e r i s h over the upcoming y e a r . The a n n u a l a d u l t m o r t a l i t y r a t e (15 %) was e s t i m a t e d f o r a d u l t Double-c r e s t e d Cormorants on Mandarte I s l a n d (van der Veen 1973) . Assumptions - I assumed t h a t : 1) the d i s t r i b u t i o n s o f age c l a s s e s was s i m i l a r among the c o l o n i e s ; 2) a l l a d u l t s a t t e m p t e d t o b r e e d a t l e a s t once; 3) a l l a d u l t s have t h e same p o t e n t i a l m o r t a l i t y r a t e ; 4) t h e a d u l t m o r t a l i t y r a t e was t h e same as e s t i m a t e d i n 1973; and 5) t h e r a t e s o f d i s p e r s i o n a r e s i m i l a r among c o l o n i e s . Number of nestlings produced: R o b e r t s o n (1971) found t h a t the g r e a t e s t n e s t l i n g m o r t a l i t y o c c u r r e d b e f o r e n e s t l i n g s were 14 - 21 days of age. A f t e r t h i s t i m e , t h e number of young produced was e s t i m a t e d . T h i s was done by m u l t i p l y i n g the number of s u c c e s s f u l n e s t s by t h e mean brood s i z e . A s u c c e s s f u l n e s t was a n e s t which h e l d eggs when sampled 83 for clutch size data i n Chapter 2. Assumptions - I assumed a nestling mortality rate a f t e r 14 - 21 days of age to be < 5% (Robertson 1971) , and to be si m i l a r among a l l colonies. Juvenile s u r v i v a l rates: The number of nestling Double-crested Cormorants surviving to 3 years of age was calculated using a model developed for Shags (Harris et al. 1994). B r i e f l y , s u rvival from post fledging to breeding age was calculated for 25 cohorts of Shags i n r e l a t i o n to the time of breeding. Estimates were based on band re-sightings and birds found dead. I used Harris et ai.'s (1994) model for 4 reasons. Shags and Double-crested Cormorants belong to the same genus, breed at the same time of the year, have similar ranges of breeding dates and have sim i l a r l i f e h i s t o r i e s . The model they propose has been documented for many species (Perrins and Birkhead 1983). Harris et al. (1994) estimated survival based on the timing of breeding and brood size. While I did f i n d s i g n i f i c a n t differences i n nestling mass as the season progressed, I did not fi n d any differences i n mass among broods (chapter 3) . For t h i s reason, I calculated survival estimates from t h e i r data based s o l e l y on the timing of breeding. Ringing dates were converted to laying dates by subtracting 21 days to give hatch date and subtracting a further 27 days to give laying date.. The post fledging survival to breeding age was best described by the 84 l i n e : Probability of Survival = -0.07* Laying Date + 22.8 (5.2) r 2 = 0.587, d f = 10, p = 0.004 ( F i g 5.1) Assumptions - I assumed t h a t : 1) the s u r v i v a l model, based on Shags, i s s i m i l a r i n D o u b l e - c r e s t e d Cormorants; and 2) t h e p a t t e r n o f m o r t a l i t y i s f i x e d among b r e e d i n g a t t e m p t s . Number of juveniles surviving: The number o f j u v e n i l e s s u r v i v i n g t o 3 y e a r s o f age was de t e r m i n e d by m u l t i p l y i n g t h e number o f j u v e n i l e s produced by the s u r v i v a l e s t i m a t e s based on the t i m i n g o f b r e e d i n g ( H a r r i s e t al. 1994) . Observed number of breeding p a i r s : The number of b r e e d i n g p a i r s was d e t e r m i n e d u s i n g n e s t count d a t a i n c l u d i n g d a t a from Vermeer e t al. 1989, S u l l i v a n 1989, Campbell e t al. 1990 and I . Moul (unpubl. d a t a ) . The s l o p e o f t h e l i n e d e s c r i b i n g t h e change i n p o p u l a t i o n was d e t e r m i n e d a t each c o l o n y f o r n e s t counts from 1987 t o 1994. Data f o r Mandarte I s l a n d o n l y i n c l u d e d the y e a r s 1989 t o 1994 s i n c e t h e t r e n d i n n e s t numbers was c o n s i s t e n t o n l y d u r i n g t h e s e y e a r s . A t e s t of the predictions: The p r e d i c t e d and observ e d changes i n t h e number o f 85 Figure 5.1. The post fledging survival estimates for Shags Phalacrocorax aristotelis surviving to 3 years of age, based on a model by Harris et al.(1994). 86 87 breeding pairs at ind i v i d u a l colonies was compared. The ef f e c t of the timing of breeding and the changes i n the number of breeding pairs was also compared on the l e v e l of a metapopulation. This was considered to be a l l breeding pairs at 7 Double-crested Cormorant colonies within the S t r a i t of Georgia. The above methods were used at each of these colonies to determine the predicted and observed changes i n the number of adults and juveniles recruited into the breeding population. On Chain Island, the proportion of nests which were successful was considered to be 85%. and a brood size of 2.9 young per successful nest. S t a t i s t i c a l analyses - A l l s t a t i s t i c a l analyses were performed using SYSTAT 5.1 (Wilkinson 1990). The p r o b a b i l i t y of sur v i v a l of Shags and the changes i n breeding pairs of Double-crested Cormorants were determined using l i n e a r regression. Results The model data: The number of adults surviving - The number of adults estimated to survive to breed again r e f l e c t s the number of adults at each colony and the annual mortality rate of 15% (Table 5.1). The number of adults predicted to perish ranged from 138 on Mandarte Island to 8 on the Fraser River. The number of young produced - With the exception of Mandarte Island (1994), breeding success ranged from 76.9% to 86.3% and brood sizes ranged from 2.2 to 3.3 young per nest (Table 5.2). As a result, each colony produced at least 5.4 88 C •rl > •H > H W O 55 Cn G •rl Q O 55 Cn C •rl •d cu cu M pq o 55 w •p 0) 0) 5 o 55 H rd (I) C 0 •H -P rd O O hH >i Ci 0 H 0 U o CM CO LO CM CO LO CM CT CO rH rH co CM rH rH rH CD CO 00 00 o r - LO o ro CO •sT ro ^J1 CO •vT CTl CTl CTl CT cn CTl CT CT rH rH rH rH Xi X ! G G rd rd rH rH X! X i co co G G H H rd rd rH rH H H CO CO CD CD 1—1 t—i Cn Cn G G CD CD -H -H -p -P Cu Cu SH S H rd rd CD CD XS Xi > > C G -H -rH rd rd Cu Cu S 2 CO CO CM CD CD CD rH LO rH O O kD r - CT CO rH CM LO 00 CT CT S H CD > S H CD co rd S H Cu CM LO <x> CM CT CT rH S H CD > CC CC S H CD CO rd S H Cu G 13 O Xi x i CD Xi G G o S H CD S H rd CO CD G rH rd > CT 00 tn rd d -H d tn o SH >i 0 CD MH CJ 0 U H SH o 0 4-> -H d rd a SH 4-1 TS CO CD 4-1 CD U X! -H 4-> TS OJ d SH •H a X 4-1 TS -H d fO co 0 CD N -H - r H d U l o r H TS O O u O M CD X CD SH x; cn 4-) 4-> cn 4-1 0) rd d CD r H tn d CO M-4 cn tn cn d CD -H 0 TS O CD d CD cn SH X ! m o O 4-1 S-4 CD tn d •H d > d •H > CD SH X d EH CO 0» •rl > M CO CM CO O rH rH CM O LO rH rH rH o T3 0 0 M PQ CO +) 0) a) c o •H +> rrj O O Hi >i O H 0 U C •H % Q) ro M Hi is CM rH 00 00 OO co 00 r - cn VD rH cn CO CO "— "— ^ CD CD o VD CM 00 00 CM co LO cn cn co VD u oo 00 rrj cn CX> cn a> cn cn cn >H rH rH rH >1 rO 2 cn cn (0 2 CM CM 0 d d LO CM VD CO 00 cn cn LO LO o CM oo CM CM LO .—. o cn 00 o • • • VD o o CO —* — — CM o rH CM 4-1 CO d tn d < CM d LO CM cn cn TS TS d d rO (0 r H r H TS TS CO CO d d H H CO rd r H r H SH SH S-1 SH CO cn 0 0 CD 0 M i—i > > tn D i • r H • H d d 0 0 (X 0C • H • H 4-1 4-1 Cu Cu SH u SH SH rd ro 0 0 CD 0 T5 TS cn cn > > d d ro ro - H - H CO rO SH SH Cu Cu 2 2 Cu Cu CO d tn d < cn d O TS TS 0 TS d d o SH 0 SH co co 0 d r H ro > 0 tn to 4H O co SH ro 0 >i oo O 4-1 rO > -H > SH d co m o >i 4-1 X! rO X! O SH a 0 X! 4-1 d o T S 0 CO rO X 0 SH ro co 0 d r H rO > o cn times as many nestlings (Table 5.2) as adults predicted to perish (Table 5.1), with the exception of Mandarte Island (1994). Juvenile s u r v i v a l rates and number of surviving young -Based on the model of survival for Shags (Fig 5.1), the s u r v i v a l rates decline seasonally from a high of 13.4% to a low of 6.7% (Table 5.3). Based upon these estimates, the number of young reaching the age of breeding was s i g n i f i c a n t l y reduced as the season progressed (Table 5.2). In 1993, the colony on Mandarte Island produced 347 more young than on Five Finger Island. Due to the l a t e r timing of breeding on Mandarte Island, and hence poorer sur v i v a l rates, only 8 more young were predicted to reach breeding age than on Five Finger Island (Table 5.2). The predicted number of breeding p a i r s : The number of predicted breeding pairs at colonies where birds bred early i n the season was larger compared to colonies that bred l a t e r i n the season (Table 5.4). The number of pot e n t i a l breeding pairs produced was greatly affected by the timing of breeding. The predicted number of surviving young (Table 5.2) exceeded the number of adults predicted to perishing (Table 5.1) only Five Finger Island (Table 5.4), since s u r v i v a l rates of the young at other colonies declined as the season progressed. 91 w cu -p TJ rd CU Pi -p o •H TJ CU U CM rd o\° > — •rl > rl CO >1 C! O H O CJ LO rH CTl CT 00 rH CM rH O rH UJ u oo 00 M< 00 rd CT CT CT CT CT CT CD CT CT <T CT CT CT rH rH rH rH rH rH T i T i G G CO to rH rH T i T i CO CO G G 1—1 1—1 (0 rO rH rH SH SH SH SH CO CO CD CD 0 CD H 1—1 > > Cn CJ) •rH • H G G CD CD PS PS •H -rH -P -P [u P M SH SH SH SH CO rd CD CD CU CD T i T S CO CO > > G G rO rO - H •rH CO to SH M Pn 2 S CD •P rd Q Cn >i >1 G rO rd •rl S S rd o CM r l rH CM a rd CU 2 -P CD CO >! G G rH G CJ) G rO G < rO LO LO CM CM rH CM •p CO G CJ) G C T CO CJ) rO XC c/3 SH O m 0 CJ) rd m o co SH rO 0 >1 oo o •p CO > -H > SH G CO CJ) G -H CD TS 0 rH m I -p CO o a. m o 0 TS o e rO G O — rj CT CT TS <H 0 CO ' ro "H X! <0 •u 0 CO • H SH SH ro DC CM CT 4-1 O CD Cn C rrj , G U XI CD -P O •rl X) CD u CU w U •rl rd CM 4-1 0 o 5 cn C ! •rl TJ CD cu u CQ O VO O CTl rH CO 00 CO Cn ci •rl a rd CD 2 co LO o o r— oo o CM LO r-+ + 1 rH i CM oo CM CO O CO 00 u C O CO CO C O CM •rl o r- LO o C M LO rd 00 CO Oi M 00 00 co rd CTl cn cn cn cn cn CD CTl cn cn cn cn m rH rH rH rH rH rH TJ TJ C c rrj rrj C rH rH TJ TJ 0 CO CO C a •rl rH HH rrj rrj +> rH rH SH SH rd SH SH CO CO CD CD 0 CD CD 1—1 i—i > > 0 Cn Cn •H •H a C CD CD O S O S •H -H 4-> -P >1 Cu SH SH SH SH ci rrj rrj CD CD 0 CD CD T3 TJ CO CO H > > C C rrj rrj 0 -H -H rrj rrj SH SH U Cu Cu S S Cu Cu >1 >1 rrj rrj S S O CM rH CM CD C5 l-J LO CM 4-1. CO CJ Cn es CM ID LO CM 4-1 CO CJ Cn CJ < cn oo o CM LO O O o >1 rH c o 6 o H co 4-1 rH CJ CO CD SH CO SH -H rrj Cn C -H TJ CD CD SH £} CD C •H rH O CD T3 TJ • CD Cn 4-> C O CJ CD O f i >i CD Cn C ro O ^ -2 4-> TJ O * rl CD C i J3, Cn co •H P X ! w CD CD C x: oo cn 4-J CD x: 4-1 Observed number of breeding p a i r s : The only colony that substantially increased i n the number of breeding pairs between 1987 and 1994 was Five Finger Island (Table 5.5), although the two colonies at Crofton and Hudson Rocks showed small increases. A l l other colonies showed a decline i n the number of breeding pairs. When the timing of breeding was matched to the changes i n the number of breeding pairs, the same trend was found (Table 5.6). In colonies where breeding occurs early, the number of breeding pairs increased while the number of breeding pairs declined i n colonies which either bred l a t e r or f a i l e d to breed. Discussion Adequacy of the data: This i s the f i r s t attempt at modeling the factors influencing the population of Double-crested Cormorants within the S t r a i t of Georgia. There are 3 parameters used i n t h i s model: 1) adult mortality rates; 2) the number of young produced; and 3) the survival rates of these young. While I could not use dispersion rates i n t h i s model, t h i s factor i s important i n population studies (Perrins and Birkhead 1983). In order to provide r e a l i s t i c results, the data used or omitted (dispersion) i n t h i s model must be examined. Adult mortality rates - van der Veen (1973) estimated 15 % of t h i s population of adults perished between breeding seasons. Since 1973, the number of Bald Eagles nesting within the S t r a i t 94 LO Csl . u CQ 0 -Cn rrj C -H rrj Cn X U O O 0 rH O rO C a rO +J 4H O 0 rOX M 4-> 4-1 CO TS a 0 rO X! 4-> T S 0 C > -H SH 0 ^ co cn X! cn O rH CO TS 4-> C co ro 0 C r-co m cn O rH SH C 0 0 Xi 0 e s 3 4-> C 0 X CO CO C 4J O CO -H 0 4J C rO O MH O O rH CO >i SH C 0 O X! O O LO LO 0 rH X! rO S H H rd 0) CD M CD (3 cn co cn a t CM cn cn cn cn o cn cn LO co co ro cn TS vo cn I ro o VD 00 O LO ro o co CM ro ro LO r-rH cn LO CO rH ro cn rH H CO CM CO CO r -cn rH cn H CM O I VD CM CM LO cn CM CM CM + cn co LO r-r-oo LO VD cn co 00 VD CO VD o LO LO CO I VD LO cn CM o VD LO CM O co [ ro VD CM CM CTi VD O LO CM VD LO ro cn co ro VD co Ci ro rH TS 0 CO C 4-> •H H rO 0 •P rH SH TS rH CO rH rd SH CO 0 C CO CO 0 0 H > (0 t—t U 4-> 0 cn -H rH O O tH" C •rH te CC C Is 0 -rH Pi fH r>1 SH SH O 4-> a Ci ro 0 4-1 CO o 0 0 T3 CO 4H -H •H CO H > d. ro O CO SH T5 0 -rH rO SM SH xi x u PM 2 PM CJ CJ CJ 4-> 0 ro cn rO CQ TS C rO cn cn TS G ro oo cn cn co 4-> co 0 C co ro co oo >, rH TS ro C C rO ro O 0 rH x: — 4-1 on co on TS C C! - H CO rH TS M CO 0 ro M TS T J c > C o o 4-) ro -H C 0 4-) 4-) -rH 2 C O CO 0 £ co •H 9? _p C r H « •H O 0 m ° • p 4-1 a> S m on co cn 3 • CO H e e o o S-1 (H 4H 4H CO CO 4-) 4-> C C 3 3 cn cn O 4-1 TS 0 cn 4-) oo co cn -H S co O 0 SH -H 4H C o CO rH 4-> O rO O T! cn d. 0 H x: CO 4-1 co 3 SH SH 0 T> O D, 0 4H — 4-1 CO >i rH rH rH 3 3 C O O O 2 rH rO 0 • O M H ro CO H TS 0 C 4J 0 C > 0 H CO cn 0 SH 0 a O O Cn O O C co rO rH 4-> 4-) x! ro CO CO O 4-> - m vj V J \ J +-> X £ 0 0 0 O 4-1 G C C C °\° EH LO on \ Table 5.6: The timing of breeding i n 1989, 1993 and 1994 and the observed change i n the number of breeding pairs of Double-crested Cormorants at 7 colonies within the S t r a i t of Georgia, B.C. Timing or Breeding 1 Colony 1989 1993 1994 Observed Change2 Five Finger Island 1 1 1 + 34.5 Mandarte Island 2 2 4 -9.6 Fraser River 2 3 3 - 0.2 Crofton 1 1 1 + 2.4 Chain Island 2 3 3 -18.9 C h r i s t i e I s l e t 4 nd 4 - 8.5 Hudson Rocks 4 4 4 + 0.2 1 The timing of breeding grouped as: 1 = Successful laying and incubation of eggs i n late A p r i l to early May 2 = Successful laying and incubation of eggs delayed by 1 to 2 months 3 = Successful laying and incubation of eggs delayed by 2 to 3 months 4 = Colony f a i l e d (no young produced or < 10 % of nests successful) 2 Observed change i n colony size, as determined i n Table 5.5 nd= no data 96 of Georgia have increased (Vermeer et al. 1987) but environmental contamination by toxic chemicals have declined (Whitehead 1987). No evidence exists to suggest that the rate of adult mortality has changed. Therefore, t h i s i s the best available estimate of the adult mortality rate for Double-crested Cormorants. Number' of young produced - With the exception of Mandarte Island i n 1994, which only produced 6 young (Table 5.2), a l l colonies had a high rate of successful breeding attempts and r e l a t i v e l y large brood sizes (Table 5.2) r e s u l t i n g i n a high reproductive output. The 15% loss from the adult population during the year (Table 5.1) i s almost n e g l i g i b l e compared to the number of nestlings produced each year (Table 5.2), which, with the exception of Mandarte Island i n 1994, was at least 5 times more than the a t t r i t i o n to the population (Table 5.2). Juvenile s u r v i v a l rates - Not a l l the nestlings produced survive to breed. Juvenile mortality rates far exceed adult mortality rates (Perrins and Birkhead 1983). Harris et al. (1994) estimated that at least 86% of a l l juvenile Shags perished before reaching breeding age. Survival rates for juvenile Double-crested Cormorants estimated from the model developed for Shags (Harris et al. 1994) were dependent on the time of breeding. As the breeding season progressed, the predicted survival rate declined from a high of 13.4% to a low of 6.7% (Table 5.3). 97 The seasonal decline i n the mass of nes t l i n g Double-crested Cormorants (Chapter 3) was similar to that reported i n other studies (Coulson and White 1958, Perrins 1966, Nisbet and Drury 1972, Jarvis 1974, Harris et al. 1992, Harris et al. 1994 and Spear and Nur 1994) that also showed a seasonal decline i n juvenile s u r v i v a l . Thus there i s strong support that the seasonal decline i n the mass of nestling Double-crested Cormorants, the timing of breeding or a combination of the two also results i n a seasonal decline i n juvenile s u r v i v a l rates. Dispersion of Double-crested Cormorants: The rates and distances of dispersion by Double-crested Cormorants are unknown and could not be included i n t h i s population model. It was assumed that the rates of immigration and emigration were constant at a l l colonies. It could be argued that those individuals from early breeding colonies do not move to other colonies, due to a low frequency of disturbance, while those from colonies which breed late have higher rates of dispersion. It may be more b e n e f i c i a l for late breeders to have higher rates of dispersal i f either t h e i r attempts to breed f a i l repeatedly or i f predation pressure i s high. However, the loss of breeding pairs at late breeding colonies i s far less than the increases i n breeding pairs at early breeding colonies. The rates of dispersal are not" known for Double-crested Cormorants, but these rates are extremely low i n Shags. Of 1288 98 shags banded, 1246 (96.7%) were found breeding at t h e i r natal colony (Harris et al. 1994). Potts (1969) found that less than 1 % of cormorants that previously bred emigrated to other colonies. If dispersal rates are similar i n Shags and Double-crested Cormorants, dispersal probably does not a f f e c t the number of breeding pairs. The o v e r a l l decline i n the number of breeding pairs suggests that i f Double-crested Cormorants are dispersing to other colonies, these are not within the S t r a i t of Georgia. Dispersal to and from colonies i n northwest Washington i s un l i k e l y to explain observed changes since the number of breeding pairs has also declined i n t h i s area (Henny et al. 1989). Small scale dispersal to other colonies within the S t r a i t of Georgia may occur but i t does not detract from the fact that the population as a whole i s declining. Growth and Juvenile Survival Rates: The timing of breeding appears to be the most important factor a f f e c t i n g the number of breeding pairs within the S t r a i t of Georgia. While adult mortality rates and the number of young produced do a f f e c t the number of the number of breeding pairs, i t i s juvenile survival rates which l i m i t the population when many young are produced. In situations where the number of young produced i s r e s t r i c t e d , i . e . due to high predation pressure (Mandarte Island 1994), the number of nestlings produced w i l l play a much larger role i n determining the number of breeding 99 p a i r s . After fledging, juvenile cormorants must learn how to forage, a s k i l l that requires time to learn (Wunderle 1991). The highest l e v e l of adult and juvenile mortality occurs i n December and January i n populations east of the Rockies (Dolbeer 1991). Assuming that the timing of mortality i s s i m i l a r west of the Rockies, t h i s i s only 2 months after fledging on the Fraser River (1993) but 6 months after fledging on Five Finger Island. Juveniles fledged early not only have a higher mass and/or condition index, but they also have more time to master t h e i r foraging s k i l l s before the period of high fledging mortality. Their greater mass may be used i n times of poor foraging giving them a greater potential for s u r v i v a l . The timing of breeding and numbers of breeding p a i r s : At i n d i v i d u a l colonies, the predicted and the observed changes i n the number of breeding pairs followed a s i m i l a r pattern: colonies that bred early i n the season increased i n the number of breeding pairs while those that bred l a t e r i n the season declined (Tables 5.4 and 5.6). This supports my hypothesis that the number of breeding pairs i s affected by the timing of breeding. These patterns from predicted and observed changes d i f f e r e d q u a n t i t a t i v e l y possibly because: 1) the rate of adult or juvenile mortality d i f f e r e d in magnitude as the season progressed or; 2) dispersal of cormorants from one colony to 100 another may occur at a higher rate than predicted. Any difference i n the rate or the pattern of decline between Double-crested Cormorants and Shags would a l t e r the differences between predicted and observed changes i n the number of breeding pairs of cormorants. Mortality rate of juveniles i s not fixed. Most studies have found large fluctuations among years (Perrins 1966, Potts 1969, Aebischer 1986, Harris et al. 1992, Harris et al. 1994, Spear and Nur 1994). None the less, my model, a l b e i t s i m p l i s t i c , appears to predict short term changes in the size of breeding population at in d i v i d u a l colonies. The timing of breeding and the metapopulation: The number of breeding pairs i n the metapopulation, consisting of a l l colonies within the S t r a i t of Georgia, has declined (Table 5.5). Breeding has been delayed i n 5 out of the 7 colonies (Table 5.6) suggesting that, due to the timing of breeding, recruitment of young i s not replacing the adults which perish i n any given year. Thus, these colonies are acting as sink colonies. The 2 remaining source colonies, Five Finger Island and Crofton, do not produce enough young to replace the d e f i c i t of young from a l l sink colonies. The l i k e l y scenario i s that as long as t h i s trend i n delayed breeding continues, the breeding population of Double-crested Cormorants w i l l continue to decline. The rate of the decline depends on the timing of reproduction and the frequency of breeding f a i l u r e s . Eventually, Double-crested Cormorants may 101 no longer breed i n t h i s area unless the time of breeding occurs e a r l i e r or there i s recruitment from source colonies outside the S t r a i t of Georgia. Other factors a f f e c t i n g the number of breeding p a i r s : The number of breeding birds may also r e f l e c t the number of birds which "choose" not to breed. Aebischer and Wanless (1992) showed that many Shags do not attempt to breed every year. Two reported population crashes of Shags were probably caused by late breeding, poor reproductive success and many birds choosing not to breed (Aebisher and Wanless 1992). There are no data on the number of breeding versus non-breeding Double-crested Cormorants i n the S t r a i t of Georgia. If there i s a pool of potential breeders present, the question remains would they delay breeding as most birds were observed to do i n t h i s study. The prediction of the model i s that regardless of the number of breeding pairs or the number of young produced, the poor survival rates of late produced young w i l l not replace the number of adults perishing during the year. Summary 1) The predicted (Table 5.4) and observed (Table 5.6) changes in the number of breeding pairs followed the same trend: colonies where breeding occurred early showed a net increase i n the number of breeding pairs whereas those with late breeding showed a decline. 102 2) The observed number of breeding pairs of Double-crested Cormorants i n the S t r a i t of Georgia i s declining at approximately 49 nests (2.6%) per year or 343 nests (18.4%)over a 7 year period (Table 5.5). 3) Based on the timing of breeding and number of breeding pairs, I predict that the population of Double-crested Cormorants within the S t r a i t of Georgia w i l l continue to decline. 103 Chapter 6. General Discussion My aim i n t h i s thesis was to determine the e f f e c t s of the time of breeding on the clutch size, growth and diet of n e s t l i n g Double-crested Cormorants. Using results and data from t h i s and other studies, I was also able to model the population dynamics of Double-crested Cormorants and determine the e f f e c t s of the timing of breeding on the population within the S t r a i t of Georgia. Here, I summarize and integrate the r e s u l t s . In t h i s discussion, delayed breeding i s defined as the delay i n successful breeding within a single breeding season. This d i f f e r s from the more common usage of delayed breeding, i n the l i t e r a t u r e , that describe individuals which delay the onset of t h e i r f i r s t breeding attempt from one year to the next. Clutch s i z e and the timing of breeding: There are 4 main conclusions from t h i s study. F i r s t , the timing of breeding does not af f e c t clutch size greatly i n Double-crested Cormorants as has been reported for other species (Klomp 1970) . The only exception was on the Fraser River where clutches completed l a t e r i n the season were smaller than those completed e a r l i e r i n the season. There was l i t t l e or no difference detected i n clutch sizes within colonies which bred early i n the season (Five Finger Island) or within colonies which bred early on Mandarte Island i n 1971 and late i n 1993. Clutch sizes on Mandarte Island were also consistently larger than on Five Finger Island and the Fraser River. The reason for 104 colony differences within and between years i s unknown. Lack's (1954) clutch size hypothesis states that the provisioning a b i l i t y of the parents i s the ultimate factor c o n t r o l l i n g clutch size. If t h i s hypothesis holds for cormorants, abundance or a v a i l a b i l i t y of prey species could be i n f e r r e d from the clutch size. However, I believe that the absence of a seasonal decline i n clutch size i n these cormorants does not r e f l e c t the provisioning a b i l i t y of the adults but r e f l e c t s t h e i r i n a b i l i t y to incubate larger clutches. My attempts to measure the e f f e c t of increasing clutch sizes was unsuccessful, mainly due to egg predation. The hypothesis that clutch size, i n cormorants, i s determined by the i n a b i l i t y of the adults to incubate larger clutches s t i l l needs to be tested. Nestling Growth and the timing of breeding: My second main conclusion i s that the timing of breeding i s paramount i n determining the mass and condition index of the nestlings. Early breeding produces young which are heavier and possess higher condition indices than late breeding. Calogeros (1996) also found a seasonal reduction i n n e s t l i n g mass of Mallards even though they were fed ad libitum. This suggests that the decline i n mass i s not related d i r e c t l y to food intake, but may instead be a response to other factors such as a decline i n the amount of daylight. Nestling mass and the time of fledging have been shown to 105 a f f e c t juvenile survival for many species of birds (Perrins and Birkhead 1983) . Yet differences i n the s t r u c t u r a l components of nestlings on the Fraser River (1993), the l a s t colony to breed, suggest that there may be some adaptation to growth to minimize the time spent i n the nest. On the Fraser River (1993), nestlings had larger culmen and t a r s a l lengths than nestlings at any other colony. Lower mass and larger s t r u c t u r a l size should decrease the physical loading allowing the birds to f l y at an e a r l i e r age, a pattern observed i n other species of birds (Smart 1965, Lightbody and Ankney 1984, Janiga 1992). Diet of nestlings cormorants and the timing of breeding: My t h i r d main conclusion i s that the prey fed to ne s t l i n g Double-crested Cormorants were q u a l i t a t i v e l y s i milar, but qua n t i t a t i v e l y d i f f e r e n t among colonies. Ainley et al. (1981) report that Double-crested Cormorants feed on schooling f i s h over f l a t substrates. My results show that within a small geographic location, these cormorants may feed predominantly on schooling f i s h ( P a c i f i c Sandlance at Five Finger Island), benthic f i s h (blennies at Mandarte Island) or a combination of the two (Pa c i f i c Staghorn Sculpin and Shiner Seaperch at the Fraser River). It i s unknown i f the prey species, fed to ne s t l i n g cormorants, r e f l e c t t h e i r abundance or a v a i l a b i l i t y . While the same range of species was selected at a l l colonies, the abundance or a v a i l a b i l i t y of these species may change 106 s e a s o n a l l y . For example, s e x u a l l y mature P a c i f i c S t a g h o r n S c u l p i n move i n t o i n t e r t i d a l a r e a s i n September and O c t o b e r t o b r e e d , when the a d u l t s a t t h e F r a s e r R i v e r c o l o n y were f e e d i n g young. The d i e t o f young on the F r a s e r R i v e r was composed p r e d o m i n a n t l y of l a r g e s c u l p i n . At a l l o t h e r c o l o n i e s , s c u l p i n were s h o r t e r and l e s s abundant i n the n e s t l i n g s d i e t . No s i g n i f i c a n t d i f f e r e n c e was d e t e c t e d i n t h e b o l u s mass o r i n t h e amount o f n u t r i e n t s d e l i v e r e d t o t h e n e s t l i n g s . U s i n g t h e amount o f g r o s s energy i n each d i e t , I was u n a b le t o r e l a t e p r o v i s i o n i n g and t h e o b s e r v e d d i f f e r e n c e s i n n e s t l i n g mass u s i n g t h r e e s c e n a r i o s : i ) n e s t l i n g s r e c e i v e d e q u a l amounts o f f o o d ; ii) n e s t l i n g s r e c e i v e d f o o d i n p r o p o r t i o n t o t h e amount o f d a y l i g h t ; and i i i ) n e s t l i n g s r e c e i v e d f o o d i n p r o p o r t i o n t o t h e amount o f d a y l i g h t and the number of young i n t h e n e s t . Double-crested Cormorant populations and the timing of breeding: My f o u r t h main c o n c l u s i o n i s t h a t t h e t ime of b r e e d i n g i s a p r i n c i p l e f a c t o r d e t e r m i n i n g t h e p o p u l a t i o n s i z e o f Double-c r e s t e d Cormorants w i t h i n the S t r a i t o f G e o r g i a . P r e d i c t e d and o b s e r v e d changes f o l l o w e d t h e same p a t t e r n : c o l o n i e s i n c r e a s e i n s i z e where b r e e d i n g o c c u r s e a r l y i n t h e season and d e c l i n e i n s i z e a t c o l o n i e s where b r e e d i n g i s d e l a y e d . The number of young produced a t t h e m a j o r i t y o f c o l o n i e s g r e a t l y exceeded th e p r e d i c t e d number of a d u l t s which would p e r i s h i n any g i v e n y e a r . Based upon my p o p u l a t i o n model, the r e a s o n f o r the r e d u c t i o n i n the number of b r e e d i n g p a i r s was due 107 to high rates of mortality among fledged young. The only exception was at Mandarte Island (1994) where only 6 young were produced from 403 nests. My data suggest that Double-crested Cormorants that breed early have a better chance of producing offspring that survive to become breeding adults, compared to birds that breed l a t e r i n the season. Furthermore, th i s discrepancy could be further exacerbated i f early fledged juveniles are able to breed at an e a r l i e r age than late fledged cormorants. Future d i r e c t i o n s f o r research: The future directions for research include: 1) There appears to be increasing body of data suggesting that the seasonal decline i n nestling mass does not r e s u l t from differences i n food intake but from some other factor. I suggest hormonal differences among the nestlings may be responsible for the patterns observed. Maternal deposits of hormones i n the eggs may decline due to the seasonal declines i n the female. Hormones such as growth hormone, thyroid hormone (T3 and T4) and testosterone are known to af f e c t growth and they also decline seasonally. 2) Determine the a b i l i t y of Double-crested Cormorants to incubate clutches larger than 4 eggs. This can be accomplished by increasing the normal 4 egg clutch by one egg and monitoring 108 hatching success. 3) For "normal" growth to occur, nestlings require e s s e n t i a l amino acids, f a t t y acids, minerals and vitamins. These must a l l be found i n the prey that they consume. Thus i t i s important to conduct further analyses of the prey species fed to nestl i n g cormorants. In addition, the metabolizable energy c o e f f i c i e n t s of the various prey f i s h may provide further insight into the eff e c t of the diets of the nestling growth. It i s unknown whether ash content of prey species, high i n P a c i f i c Staghorn Sculpin and low i n blennies, affect growth of nestl i n g cormorants. 4) On many breeding colonies, the. timing of breeding i n Double-crested Cormorants depends on the extent of disturbances caused mainly by Bald Eagles. The breeding population of Bald Eagles has doubled i n the l a s t 2 decades within the S t r a i t of Georgia (Vermeer et al. 1987). Bald Eagles are drawn to many of the cormorant breeding colonies since cormorants usually nest i n association with Glaucous-winged Gulls, a favorite prey of Bald Eagles (Vermeer et al. 1987). As I have shown, delayed breeding can be equated to a reduction i n the breeding population. Thus, i t i s important to study the dynamics in t e r a c t i n g among eagles, g u l l s and cormorants with the aim of reducing the amount of disturbances at the colonies. 109 5) It i s important to understand the abundance and d i s t r i b u t i o n of prey f i s h . L i t t l e i s known about the basic biology of most prey species yet they are an important part of the food web upon which f i s h eating birds rel y . Since many of the prey species for Double-crested Cormorants are s o l i t a r y and benthic, scuba surveys may provide the best r e s u l t s . 6) Double-crested Cormorants are a sentinel species used i n toxic chemical research. Egg analysis does not provide clear cut answers to toxic loading since toxins deposited i n the yolk could come from l i p i d s accumulated i n wintering locations. Analysis of blood and/or l i p i d samples as well as bolus samples may provide toxic p r o f i l e s of the l o c a l environment and the prey species which are the pathway for contamination. 7) The location, timing and duration of foraging by Double-crested Cormorants i s unknown. The time of foraging may d i f f e r among colonies due to the type or a v a i l a b i l i t y of prey species. The differences i n prey fed to the nestlings may simply r e f l e c t the t i d a l height on a v a i l a b i l i t y and abundance of prey species. 8) Develop l e g i s l a t i v e and enforceable guidelines for boaters, kayakers and tour-boat operators on the safe distance that they can approach colonies to minimize disturbances and maximize viewing opportunities. With the increase i n the human population l i v i n g i n and around the S t r a i t of Georgia, the po t e n t i a l for 110 c o n f l i c t between birds and humans increases. Epilogue I began t h i s research to inquire whether the breeding biology of Double-crested Cormorants was adversely affected by contaminant loading from the Fraser River. As i t turned out, nesting success i s jeopardized by disturbances, rather than contaminants, which postpones the successful breeding and consequently the number of young cormorants produced each year. The reason for delayed breeding at coastal s i t e s appears to be the r e s u l t of eagles, p r i n c i p a l l y hunting g u l l s which nest on the same i s l a n d as cormorants. While the reason for delayed breeding on the Fraser River i s unknown, i t appears that colony differences are related to the time of breeding. It i s clear, however, that Double-crested Cormorant colonies i n the S t r a i t of Georgia are unsustainable under current lev e l s of disturbance. I l l Literature Cited Aebischer, N.J. 1986. Retrospective investigation of an ecological disaster i n the Shag, Phalacrocorax a r i s t o t e l i s : a general method based on long-term marking. J. Anim. Ecol.55: 613 -629. Aebischer, N.J. and S. Wanless. 1992. Relationships between colony-size, adults non-breeding and environmental conditions for Shags Phalacrocorax a r i s t o t e l i s on the Isle of May, Scotland. Bird Study 39: 4 3 - 5 2 . Ainley, D.G., D.W. Anderson and P.R. Kelly. 1981. Feeding ecology of marine cormorants i n southwestern North America. Condor 83: 120 - 130. Ashmole, N.P. 1971. Seabird ecology and the marine environment, in Farner, D.S. and J.R. King (eds). Avian Biology Vol 1: 224 -271, New York Academic Press. Barrett, R.T. and F. Rikardsen. 1992. Chick growth, fledging periods and adult mass loss of A t l a n t i c Puffins Fratercula a r c t i c a during years of prolonged food stress. Colonial Waterbirds 15: 24 - 32. Batchelor, A.L. and G.J.B. Ross. 1984. The diet and implications of dietary change of Cape Gannets on Bird Island, Algoa Bay. Ostrich 55: 45 - 63. Bayer R.D. 1985. Shiner Perch and P a c i f i c Staghorn Sculpin i n Yaquina Estuary, Oregon. Northwest Science 59: 230 - 240. Bennett, D.C. and L. E. Hart. 1993. Metabolizable energy of f i s h when fed to captive Great Blue Herons (Ardea herodias). Can. J. Zool. 71: 1767 - 1771. Bertram, D.F., G.W. Kaiser and R.C. Ydenburg. 1991. Patterns i n the provisioning and growth of nestling Rhinoceros Auklets. Auk 108: 842 - 852. Brugger, K.E. 1993. D i g e s t i b i l i t y of three f i s h species by Double-crested Cormorants. Condor 95: 25 - 32. Calogeros, A. 1996. Optimum outbreeding: mate choices and consequences of inbreeding i n Mallards. Msc Thesis. Dept. of Animal Science, University of B r i t i s h Columbia, Vancouver, B.C. Campbell, R.W., N.K. Dawe, I McTaggart-Cowan, J.M. Cooper, G.W. Kaiser and M.C.E. McNall. 1990. The birds of B r i t i s h Columbia, Vol 1. Non-passerines. Royal B r i t i s h Museum, V i c t o r i a . 112 Cleary, L. 1977. Succes de reproduction du Cormoran a aigrettes, Phalacrocorax auritus a. sur 3 l i e s du St-Laurent, en 1975 et 1976. These de maitrise, Universite Laval, Quebec, in Leger and McNeil 1987. Cooke, F., R.F. Rockwell and D.B. Lank. 1995. The Snow Goose of La Perouse Bay: Natural selection i n the wild. Oxford University Press. Coulson, J.C. and E. White. 1957. Mortality rates of the Shag estimated using two independent methods. Bird Study 4: 166 -171.. Craven, S.R. and E. Lev. 1987. Double-crested Cormorants i n the Apostle Islands, Wisconsin, USA: Population trends, food habits and fishery depredations. Colonial Waterbirds 10: 64 - 71. Daan, S., C. Dijkstra, R. Drent and T. Meijer. 1988. Food supply and the annual timing of avian reproduction. Acta Congressus Internationalis Ornithologici 19: 392 - 407. Dolbeer, R.A. 1991. Migration patterns of Double-crested Cormorants east of the Rocky Mountains. J. F i e l d Ornithol. 62: 83 - 93. Drent, R.H. 1975. Incubation, in: Farner, D.S., J.R. King and K.C. Parks (eds). Avian Biology 5: 333 - 420. Drent, R.H. and S. Daan. 1980. The prudent parent: energetic adjustment i n avian breeding. Ardea 68: 225 - 252. Drent, R., G.F. van Tets, F. Tompa and K. Vermeer. 1964. The breeding birds of Mandarte Island, B.C. Can. F i e l d Nat.78: 208 - 263. Duffy, D.C. and L.J.B. Laurenson. 1983. Pellets of Cape Cormorants as indicators of diet. Condor 85: 305 - 307. Dunn, E.H. 1975. Growth, body component and energy content of nestling Double-crested Cormorants. Condor 77: 431 - 438. E l l i s o n , L.N. and L. Cleary. 1978. Effects of human disturbance on breeding of Double-crested Cormorants. Auk 95: 510 - 517. Findholt, S.L. 1988. Status, d i s t r i b u t i o n and habitat a f f i n i t i e s of Double-crested Cormorant nesting colonies i n Wyoming. Colonial Waterbirds 11: 245 - 251. Gremillet, D., D. Schmid, and B. Culik. 1995. Energy requirements of breeding Great Cormorants Phalacrocorax carbo sinensis. Mar. Ecol. Prog. Ser. 121: 1 - 9 . 113 Harris, M.P., S.T. Buckland, S.M. Russel and S. Wanless. 1994. Post fledging survival to breeding age of Shags Phalacrocorax aristotelis i n re l a t i o n to year, date of fledging and brood size. J. Avian B i o l . 25: 268 - 274. Harris, M.P., D.J. Halley and S. Wanless. 1992. The post-fledging survival of young Guillemots Uria aalge i n r e l a t i o n to hatching date and growth. Ibis 134: 335 - 339. Harvey, S., C.G. Scanes and P.J. Sharp. 1982. Annual cycle of plasma concentration of growth hormone i n Red Grouse {Lagopus lagopus scoticus). Gen. Comp. Endocrinol. 48: 411 -414 . Hatch, J.J. 1984. Rapid increase of Double-crested Cormorants nesting i n southern New England. American Birds 38: 984 -988. Heath, R.G.M. and R.M. Randall. 1985. Growth of Jackass Penguin chicks (Spheniscus demersus) hand reared on di f f e r e n t diets. J. Zool. Lond. 205: 91 - 105. Henny, C.J., L.J. Blus, S.P. Thompson and U.W. Wilson. 1989. Environmental contaminants, human disturbance and nesting of the Double-crested Cormorant i n northwestern Washington. Colonial Waterbirds 12: 198 - 206. Hobson, K.A. and J.C. Driver. 1989. Archaeological evidence for use of the S t r a i t of Georgia by marine birds, in Vermeer, K. and R.W. Butler (eds). 1989. The ecology and status of marine birds and shoreline birds i n the S t r a i t of Georgia, B r i t i s h Columbia. Spec. Publ. Can. Wildl. Serv. Ottawa. Hobson, K.A., R.W. Knapton and W. Lysack. 1989. Population, diet and reproductive success of Double-crested Cormorants breeding on Lake Winnipegosis, Manitoba, i n 1989. Colonial Birds 12: 191 - 197. Hyslop E.J. 1980. Stomach contents analysis - a review of methods and th e i r application. J. Fish B i o l . 17: 411 - 429. Janiga, M. 1992. Intraspecific ontogenic allometry and growth responses of the f e r a l pigeon, Columba l i v i a . Ekologia Polska 40: 577 - 587. Jarvis, M.J.F. 1974. The ecological significance of clutch size i n the South African Gannet (Sula capensis {lichtenstein)). J. Animal Ecol. 43: 1 - 17. Johnsgard, P.A. 1993. Cormorants, Darters and Pelicans of the World. Smithsonian Institute Press, Washington 114 King, D.B. and J.D. May. 1984. Thyroidal influence on body growth. J. Exp. Zool. 232: 453 - 460. Klomp, H. 1970. The determination of clutch size i n birds: A reveiw. Ardea 58: 1 - 124. Kury, CR. 1969. Pesticide residue i n a marine population of Double-crested Cormorants J. Wildl. Manage. 33: 91 - 95. Kury, CR. and M. Gochfeld. 1975. Human interference and g u l l predation in cormorant colonies. B i o l . Conserv. 8: 23 - 34. Lack, D. 1954. The natural regulation of animal numbers. Clarendon Press, Oxford. Lack, D. 1968. Ecological adaptations for breeding i n birds. Methuen, London. Lamb, A. and P. Edgell. 1986. Coastal fishes of the P a c i f i c Northwest. Harbour Publishing, Madeira Park, B.C. Canada Leger, C. and R. McNeil. 1987. Brood size and chick position as a factor influencing feeding frequency, growth and survival of nestling Double^crested Cormorants, Phalacrocorax auritus. Can. Field-Nat. 101: 351 - 361. Lewis, H.F. 1929. The natural history of the Double-crested Cormorant {Phalacrocorax auritus auritus ( (Lesson)) . RU-MI-LOU Books, Ottawa, Canada. Lightbody J.P. and CD. Ankney. 1984. Seasonal influence on the strategies of growth and, development of Canvasback and Lesser Scaup ducklings. Auk 101: 121 - 133. Lovvorn, J.R. and D.R. Jones. 1991. Effects of body size, body fat and change i n pressure with depth on buoyancy and costs of diving ducks {Aythya spp.). Can. J. Zool. 69: 2879 - 2887. Martin, K. 1995. Patterns and mechanisms for age-dependent reproduction and survival in birds. Amer. Zool. 35: 340 -348. McNabb, F.M.A. 1988. Peripheral thyroid hormone dynamics i n precocial and a l t r i c i a l avian development. Amer. Zool. 28: 427 - 440. Melrose, W.D. and S.C Nicol. 1992. Haematology, red c e l l metabolism and blood chemistry of the Black-faced Cormorant Leucocarbo fuscescens. Comp. Biochem. Physiol. 102A: 67 -70. 115 Mineau, P., G.E.J. Smith, R. Markel and C.S. Lam. 1982. Aging Herring Gulls from hatching to fledging. J. F i e l d Ornithol. 53: 394 - 402. Montevecchi, W.A., R.E. Ricklefs, I.R. Kirkham and D. Gabaldon. Growth energetics of nestling Northern Gannets {Sula bassanus). Auk 101: 334 - 341. Munro, J.A. 1928. Cormorants nesting on Bare Island, B r i t i s h Columbia. Condor 30: 327 - 328. Nacarro, R.A. 1992. Body composition, fat reserves and fasting c a p a b i l i t y of Cape Gannet chicks. Wilson B u l l . 104: 644 -655. Neter, J. and W. Wasserman. 1974. Applied linear s t a t i s t i c a l models: regression, analysis of variance and experimental design. Irwin-Dorsey Ltd., Georgetown, Ottawa. Nisbet, I.C.T. and W.H. Drury. 1972. Post fledging survival i n Herring Gulls in re l a t i o n to brood size and date of hatching. Bird Banding 43: 161 - 240. O'Connor, R.J. 1977. D i f f e r e n t i a l growth and body composition i n a l t r i c i a l species. Ibis 119: 147 - 166. Palmer, R.S. 1962. Handbook of North American Birds, Volume 1. Yale University Press, New Haven, USA. Perrins, CM. 1966. Survival of young Manx Shearwaters Puffinus puffinus i n re l a t i o n to th e i r presumed date of hatching. Ibis 108: 132 - 135. Perrins, CM. 1970. The timing of birds' breeding season. Ibis 112: 242 - 255. Perrins, CM. and T.R. Birkhead. 1983. Avian Ecology. Chapman and Hall, New York. Pond, W.G., D.C Church and K.R. Pond. 1995. Basic animal n u t r i t i o n and feeding, fourth edition. John Wiley and Sons, New York Potts G.R. 1969. The influence of eruptive movements, age, population size and other factors on the survival of the Shag {Phalacrocorax a r i s t o t e l i s (L)). J. Animal Ecol. 38: 85 - 102. Pulliam, H.R. 1988. Sources, sinks and population regulation. Amer. Nat. 132: 652 - 661. 116 Ricklefs, R.E. 1967. A graphical method of f i t t i n g equations to growth curves. Ecology 48: 978 - 983. Ricklefs, R.E. 1968. Patterns of growth i n birds. Ibis 110: 419 -451. Ricklefs, R.E. 1973. Patterns of growth i n birds. II. growth rate and mode .of development. Ibis 115: 177 - 201. Ricklefs, R.E. 1983. Avian postnatal development, in: Farner, D.S., J.R. King and K.C. Parkes (eds). Avian Biology 7: 1 -83. Robertson, I. 1971. The influence of brood size on reproductive success i n 2 species of cormorant, Phalacrocorax auritus and P. pelagicus, and i t s re l a t i o n to the problem of clutch size. Msc thesis, University of B r i t i s h Columbia, Department of Zoology. Robertson, I. 1974. The food of nestling Double-crested and Pelagic Cormorants at Mandarte Island, B.C., with notes on feeding ecology. Condor 76: 346 - 348. SAS Institute Inc. 1989. SAS® user's guide: S t a t i s t i c s , Version 6.09 Edition, Cary, NC, USA. Scanes, C.G., M. Jallageas and I. Assenmacher. 1980. Seasonal variation i n the c i r c u l a t i n g concentrations on growth hormone i n male Peking Duck (Anas platyrhynchos) and Teal (Anas crecca); correlations with thyroid function. Gen. Comp. Endocrinol. 41: 7 6 - 79. Scanes, C.G. and T.J. Lauterio. 1984. Growth hormone: Its physiology and control. J. Exp. Zool. 232: 443 - 452. Scharf, W.C. and G.W. Shugart. 1981. Recent increases i n Double-crested Cormorants i n the United States Great Lakes. American Birds 35: 910 - 911. Schmidt-Nielsen, K. 1964. Desert Animals. Oxford University Press, London. Scolaro, J.A. 1990. Effects of nest density on breeding success i n a colony of Magellanic Penguins (Spheniscus magellanicus) . Colonial Waterbirds 13: 41 - 49. Smart, G. 1965. Development and maturation of primary feathers of Redhead ducklings. J. Wildl., Manage. 29: 533 - 336. Snow, B. 1960. The breeding biology of the Shag Phalacrocorax aristotelis on the island of Lunby, B r i s t o l Channel. Ibis 102: 554 - 575. 117 Spear, L. and N. Nur. 1994. Brood size, hatching order and hatching date: effects on four l i f e - h i s t o r y stages from hatching to recruitment i n Western Gulls. J. Animal Ecol. 63: 283 - 298. Sullivan, T.M. 1985. A survey of Chr i s t i e I s l e t Migratory Bird Sanctuary, B.C. Canadian W i l d l i f e Service, Unpublished Report, Delta, B r i t i s h Columbia, 8 pp. Sullivan, T.M. 1989. Double-crested and Pelagic Cormorants i n the St r a i t of Georgia. Canadian W i l d l i f e Service Contractors Report, 24 pp. van der Veen, H.E. 1973. Some aspects of the breeding demography of the Double-crested Cormorant Phalacrocorax auritus on Mandarte Island. PhD. diss., Zool. Lab. Der R i j k s u n i v e r s i t i t e i t t e Groningen, Netherlands, in: Palmer 1973. van Tets, G.F. 1965. A comparative study of some s o c i a l communication patterns i n the Pelecaniformes. Ornithol. Monogr. No.2. Verbeek, N.A.M. 1982. Egg predation by Northwestern Crows: i t s association with human and Bald Eagle a c t i v i t y . Auk 99: 347 - 352. Vermeer, K. 1970. Some aspects of the nesting of Double-crested Cormorants at Cypress Lake, Saskatchewan i n 1969, a plea for protection. Blue Jay 28: 11 - 13. Vermeer, K. and L. Rankin. 1984. Population trends i n nesting Double-crested and Pelagic Cormorants i n Canada. Murrelet 65:1 - 9. Vermeer, K. and K. Devito. 1986. Size, c a l o r i c content and association of prey fishes i n meals of nestling Rhinoceros Auklets. The Murrelet 67: 1 - 9. Vermeer, K., K.H. Morgan and G.E. Smith. 1989. Population trends and nesting habitat of Double-crested and Pelagic Cormorants i n the S t r a i t of Georgia, in Vermeer, K. and R.W. Butler (eds). 1989. The ecology and status of marine birds and shoreline birds in the St r a i t of Georgia, B r i t i s h Columbia. Spec. Publ. Can. Wildl. Serv. Ottawa. Whitehead, P.E. 1989. Toxic chemicals i n Great Blue Heron {Ardea herodias) eggs i n the St r a i t of Georgia, in Vermeer, K. and R.W. Butler (eds).1989. The ecology and status of marine birds and shoreline birds in the St r a i t of Georgia, B r i t i s h Columbia. Spec. Publ. Can. Wildl. Serv. Ottawa. 118 Wilkinson, L. 1990. SYSTAT: The system for s t a t i s t i c s . Evanston IL, SYSTAT Inc. Wunderle, J.M. 1991. Age-specific foraging proficiency i n birds. Current Ornithology 8: 273 - 324. Ydenberg, R.C. and D.F. Bertram. 1989. Lack's clutch size hypothesis and brood enlargement studies on col o n i a l seabirds. Colonial Waterbirds 12: 134 - 137. 119 Appendix I: ANOVA Source of v a r i a t i o n tables, comparisons of least square means and results of the p a r t i a l F and Scheffe's Multiple Comparison t e s t s . Significance was accepted at a= 0.05 i n a l l tests. Table 1.1: ANOVA source of v a r i a t i o n table for n e s t l i n g asymptotic mass at a l l colonies: Sources of Variation df SS MS F value P Colony 4 308708. 9 77177 . 2 1.18 0 .32 Brood S i z e 3 168508. 4 56169. 5 0.86 0 .46 P o s i t i o n ( B r o o d S i z e ) 6 • 334258. 5 55709. 8 0.85 0 .53 L a y i n g Date 1 401727. 7 401727 .7 6.16 0 .01 Colony*Brood 12 421500. 4 35125. 0 0.54 0 .89 L a y i n g Date*Colony 4 274922. 2 68730. 6 1.05 0 .38 L a y i n g Date*Brood S i z e 3 176309. 9 58770. 0 0.90 0 .44 E r r o r 202 11019712 .8 65205. 4 T o t a l 169 22217216 .4 Table 1.2: ANOVA source of var i a t i o n table for the l o g i s t i c growth rate for mass (LGRM) at a l l colonies: Sources of Variation df SS MS F value P Colony 4 0.05 0.01 5.46 0.0004 Brood S i z e 3 0.007 0.002 1.05 0.37 . P o s i t i o n ( B r o o d S i z e ) 6 0.01 0.002 0.93 0.47 L a y i n g Date 1 0.002 0.002 0.93 0.34 Colony*Brood S i z e 12 0.01 0.001 0.53 0.89 L a y i n g Date*Colony 4 0.05 0.01 5.72 0.0002 L a y i n g Date*Brood 3 0.007 0.002 1.03 0.38 Asymptotic Mass 1 0.03 0.03 10.66 0.001 E r r o r ' 165 0.39 0.002 T o t a l 199 0.64 120 Table 1.3: Comparisons of adjusted least square means of the l o g i s t i c growth rate (day"1) for colony: FFI 1993 FRSR 1993 MAND 1993 FFI 1994 FRSR 1994 0. 64 a 0.60 a b 0.22 b 0. 19 b 0.17 b Table 1.4: Results of p a r t i a l F test and Scheffe's multiple comparison tests f o r the interaction between colony and laying date at a l l colonies: P a r t i a l F Test Parameter Value SSR (with colony) SSR (without colony) no. c o e f f . Res. MSR (with colony) Scheffe's Multiple Comparison Test Colony 1 Colony 2 Slope Sf Intercept S2 FFI 93 FFI 94 18.54 a 19.09 a FFI 93 MAND 93 22.26 a 19.36 a FFI 93 FRSR 93 7.09 3.84 FFI 93 FRSR 94 4.63 1.86 FFI 94 MAND 93 0.03 0.03 FFI 94 FRSR 93 0.00 0.01 FFI 94 FRSR 94 3.49 3.94 MAND 93 FRSR 93 0.02 0.03 MAND 93 FRSR 94 3.77 3.67 FRSR 93 FRSR 94 1.46 1.23 S c r i t i c a l - 9 - 7 6 a d i f f e r e n c e s s i g n i f i c a n t between c o l o n y 1 and 2, p> 0.05 0.25 P a r t i a l F = 2.14 0.13 F c r i t i c a l = 2.01 8 0.007 121 Table 1.5: ANOVA source of v a r i a t i o n table for LGRj, at a l l colonies except Five Finger Island 1993: Sources of Variation df SS MS F value P Colony 3 0.007 0.002 3.17 0.03 Brood S i z e 3 0.003 0.0009 1.16 0.33 P o s i t i o n ( B r o o d Size) 6 0.001 0.0002 0.30 0. 94 L a y i n g Date 1 0.0007 0.0007 0.85 0.36 Colony*Brood S i z e 9 0.008 0.0008 1.08 0.38 L a y i n g Date*Colony 3 0.007 0.002 3.15 0. 03 L a y i n g Date*Brood S i z e 3 0.003 0.0009 1.15 0.33 Asymptotic Mass 1 0.03 0.03 35.35 0.0001 E r r o r 102 0.08 0.0008 T o t a l 131 0.22 Table 1.6: Results of the p a r t i a l F test and Scheffe's multiple comparison tests for the interaction between colony and laying date at a l l colonies except Five Finger Island 1993: P a r t i a l F Test Parameter Value SSR (with colony) 0.14 P a r t i a l F = 1.67 SSR (without colony) 0.10 F = ? 1 9 •"• critical " no. c o e f f . Res. 6 MSR (with colony) 0.004 Scheffe's Multiple Comparison Test Colony Colony Slope Intercept 1 2 S2 S2 FFI 94 MAND 93 1.26 1.19 122 FFI 94 FFI 94 MAND 93 MAND 93 FRSR 93 FRSR 93 FRSR 94 FRSR 93 FRSR 94 FRSR 94 0.62 2.61 0.01 8.67a 3.63 0.40 4.08 0.02 9.78a 3.06 S c r i t i c a l ~ 8.13 d i f f e r e n c e s s i g n i f i c a n t between c o l o n y 1 and 2, p> 0.05 Table 1.7: ANOVA source of v a r i a t i o n table for LGR„ at Five Finger Island 1993 Sources of Variation df SS MS F value P Brood S i z e 2 0. 003 0.001 0.84 0 .45 P o s i t i o n ( B r o o d S i z e ) 5 0.0008 0.0001 0.06 0 .81 L a y i n g Date 1 0.0001 0.0001 0.10 0 .99 L a y i n g Date*Brood S i z e 2 0.003 0.001 0.84 0 .45 Asymptotic Mass 1 0.00009 0.00009 0.05 0 .83 E r r o r 19 0.03 0.002 T o t a l 31 0.04 Table 1.8: ANOVA source of v a r i a t i o n table for LGR„ with Five Finger Island 1993 and Fraser River 1994 removed from data set Sources of Variation df SS MS F value P Colony 2 0.0002 0.0001 0.22 0.80 Brood S i z e 3 0.0003 0.0001 0.18 0. 91 P o s i t i o n ( B r o o d Size) 6 0.001 0.0002 0.29 0.94 L a y i n g Date 1 0.0004 0.0004 0. 65 0.42 Colony*Brood S i z e 6 0.002 0.0003 0.49 0.82 L a y i n g Date*Colony 2 0.0002 0.0001 0.22 0.80 L a y i n g Date*Brood S i z e 3 0.0003 0.0001 0.18 0.91 Asymptotic Mass 1 0.01 0.01 21.66 0.0001 E r r o r • 97 0.05 123 0.005 T o t a l 121 0.09 Table 1.9: ANOVA source of v a r i a t i o n table for LGR^ j (day"1) on the Fraser River 1994: Sources of Variation df SS MS F value P Brood S i z e 3 0.002 0.0007 0.69 0 .58 P o s i t i o n ( B r o o d Size) 6 0.0007 0.0001 0.12 0 . 99 L a y i n g Date 1 0.001 0.001 1.01 0 .34 L a y i n g Date*Brood S i z e 3 0.002 0.007 0.70 0 .57 Asymptotic Mass 1 0.0003 0.0003 0.30 0 .59 E r r o r 11 0.01 0.001 T o t a l 25 0.03 Table I.10: ANOVA source of v a r i a t i o n table for the time taken to grow from 10 - 90% ( t 1 0 . 9 0 Mass) at a l l colonies: Sources of Variation df SS MS F value P Colony 4 59.66 14.91 4. 12 0.003 Brood S i z e 3 9.34 3.11 0. 86 0.46 P o s i t i o n ( B r o o d Size) 6 15.69 2.62 0. 72 0. 63 L a y i n g Date 1 13.47 13.47 3. 72 0.06 Colony*Brood S i z e 12 14.08 1.17 0. 32 0.98 L a y i n g Date*Colony 4 50. 63 12.66 3. 49 0. 009 L a y i n g Date*Brood S i z e 3 9.14 3.05 0. 84 0.47 Asymptotic Mass 1 25.00 25.00 6. 90 0.009 LGR M 1 477.99 477.99 131 .92 0.0001 E r r o r 162 586.99 3.62 T o t a l 197 2338.52 Table I.11: Comparison of adjusted l e a s t squared mean f o r t (days) f o r colony differences: FRSR 1994 FRSR 1993 FFI 1994 MAND 1993 FFI 1993 26. 62 a 25. 63 a 24 .56 a 23. 16 a 17 .50 a 124 Table 1.12: Results of p a r t i a l F and Scheffe's multiple comparison tests for the i n t e r a c t i o n between colony and laying date at a l l colonies: P a r t i a l F Test Parameter Value SSR (with colony) 1751.52 SSR (without colony) 1586.43 no. c o e f f . Res. 8 MSR (with colony) 50.04 Scheffe's Multiple Comparison Test Colony Colony Slope Intercept 1 2 S2 S2 FFI 93 FFI 94 5.06 4.98 FFI 93 MAND 93 0.02 0.27 FFI 93 FRSR 93 0.18 0.0004 FFI 93 FRSR 94 0.84 1.99 FFI 94 MAND 93 9.86 a 11.86 a FFI 94 FRSR 93 2.81 2.46 FFI 94 FRSR 94 11.27 a 11.91 a MAND 93 FRSR 93 0.41 0.15 MAND 93 FRSR 94 1.05 1.35 FRSR 93 FRSR 94 1.73 1.31 S c r i t i c a l _ 9.76 a d i f f e r e n c e s s i g n i f i c a n t between c o l o n y 1 and 2, p> 0.05 P a r t i a l F = 0.41 ^ c r i t i c a l = 2.01 125 Table 1.13: ANOVA source of v a r i a t i o n table for t 1 0 . 9 0 M a s s at a l l colonies except Five Finger Island 1994: Sources of Variation df SS MS F value P Colony 3 8.58 2.86 1.01 0.39 Brood S i z e 3 5.54 1.85 0. 65 0.58 P o s i t i o n ( B r o o d Size) 6 11.05 1.84 0. 65 0. 69 La y i n g Date 1 8.11 8.11 2.85 0.09 Colony*Brood S i z e 9 11.53 1.28 0.45 0.91 L a y i n g Date*Colony 3 7.64 2 .55 0.90 0.44 L a y i n g Date*Brood S i z e 3 5. 65 1.88 0. 66 0.57 Asymptotic Mass 1 9.86 9.86 3.47 0.06 LGR M 1 365.75 365.75 128.57 0.0001 E r r o r 132 375.50 2.84 T o t a l 162 1604.15 Table 1.14: ANOVA Finger source Island of vari a t i o n 1994: table for ^10-90 Mass ° n Five Sources of Variation df SS MS F value P Brood S i z e 3 1.97 0.66 0.40 0.75 P o s i t i o n ( B r o o d Size) 6 49.47 8.24 5.05 0.005 L a y i n g Date 1 0.87 0.87 0.54 0.48 L a y i n g Date*Brood S i z e 3 1.98 0.66 0.41 0.75 Asymptotic Mass 1 3.76 3.76 2.30 0.15 LGR M 1 62.19 62.19 38.12 0.0001 E r r o r 15 24.47 1.63 T o t a l 30 221.93 126 Table 1.15: Comparison of adjusted least square means of fcio-9o Mass (days) for ne s t l i n g p o s i t i o n nested i n brood s i z e 1 : 2/4 1/3 3/3 2/2 1/1 4/4 1/2 1/4 3/4 2/3 23 .2a 22 .3a 22 . l a 22 . l a 22 . l a 22 . 0a 22 . 0a 21. 7a 20 . 6a 15. 8b 1 #/# represents nes t l i n g position/brood size Table 1.16: ANOVA source of v a r i a t i o n table for the asymptotic culmen length at a l l colonies: Sources of Variation df SS MS F value P Colony 4 724.80 181.20 7.28 0.001 Brood Size 3 56.19 18 .73 0.75 0.52 Position(Brood Size) 6 87.76 14.63 0.59 0.74 Laying Date 1 26.08 26.08 1.05 0.31 Colony*Brood Size 12 232.81 19.40 0.778 0.67 Laying Date*Colony 4 746.55 186.64 7.49 0.001 Laying Date*Brood Size 3 56.79 18 . 93 0.76 0.52 Error 143 3561.01 24.90 Total 176 5489.03 Table I.17: Comparisons of adjusted least square means of culmen length (mm) for colony: FRSR 1994 FFI 1994 MAND 1993 FFI 1993 FRSR 1993 81. 91 a 65 .53 a 63. 97 a 58 . 68 a - 27.06 b 127 Table 1.18: Results of p a r t i a l F and Scheffe's multiple comparison tests for the i n t e r a c t i o n between colony and laying date at a l l colonies: P a r t i a l F Test Parameter Value SSR (with colony) 1928.02 P a r t i a l F = 3.10 SSR (without colony) 479.72 F c r i t i c a l = 2.01 no. c o e f f . Res. 8 MSR (with colony) 58.42 Scheffe's Multiple Comparison Test Colony Colony Slope Intercept 1 2 S2 S2 FFI 93 FFI 94 0.19 0.18 FFI 93 MAND 93 0.22 0.37 FFI 93 FRSR 93 18 .23 a 23.01 a FFI 93 FRSR 94 1.26 1.95 FFI 94 MAND 93 1.36 1.74 FFI 94 FRSR 93 19.11 a 23.90 a FFI 94 FRSR 94 2 . 63 3.08 MAND 93 FRSR 93 33.15 a 32.91 a MAND 93 FRSR 94 0.87 1.12 FRSR 93 FRSR 94 25.11 a 26.14 a S c r i t i c a l = 9-76 a d i f f e r e n c e s s i g n i f i c a n t between c o l o n y 1 and 2 Table 1.19: ANOVA source of v a r i a t i o n table for culmen length at a l l colonies except Fraser River 1993 Sources of Variation df SS MS F value p Colony 3 16.50 5.50 0.27 0.84 Brood S i z e 3 45.97 15.32 0.76 0.52 128 P o s i t i o n ( B r o o d Size) 6 45.24 7. 54 0 37 0 89 L a y i n g Date 1 32 . 64 32 .64 1 62 0 21 Colony*Brood S i z e 9 87 .59 9. 73 0 48 0 88 L a y i n g Date*Colony 3 12.01 4 . 00 0 20 0 90 L a y i n g Date*Brood S i z e 3 46. 63 15 .54 0 77 0 51 E r r o r 119 2395.26 20 .13 T o t a l 147 2935.10 Table 1.20: ANOVA source of v a r i a t i o n table for culmen length at the Fraser River 1993: Sources of Variation df SS MS F value P Brood S i z e 2 3.18 1.59 0.009 0 . 91 P o s i t i o n ( B r o o d S i z e ) 4 134.75 33. 69 1.93 0 .15 L a y i n g Date 1 167.23 167.23 9.59 0. 007 L a y i n g Date*Brood S i z e 1 0.004 0.004 0.0002 0 . 99 E r r o r 17 296.39 17.43 T o t a l 27 1291.91 Table 1.21: ANOVA source growth rate of v a r i a t i o n (LGRC) at a l l table for colonies: the l o g i s t i c Source of Variation df SS MS F Value P Colony 4 0.01 0.003 6. 60 0.0001 Brood S i z e 3 0.005 0.002 3. 82 0.01 P o s i t i o n ( B r o o d Size) 6 0.003 0.0005 1. 04 0.40 L a y i n g Date 1 0.00007 0.00007 0. 16 0.69 Colony*Brood S i z e 12 0.001 0.0008 1. 73 0.07 L a y i n g Date*Colony 4 0.01 0.003 6. 57 0. 0001 La y i n g Date*Brood S i z e 3 0.005 0.002 3. 83 0.01 Asymptotic Culmen Length 1 0.007 0.007 14 .47 0.0002 129 E r r o r T o t a l 142 0.07 0.0005 176 0.14 Table 1.22: Comparisons of adjusted least square means of the LGRC (day"1) at a l l colonies: FFI 1993 FRSR 1993 FRSR 1994 MAND 1993 FFI 1994 0 . 35a 0 . 33 a b 0 . 19b 0 . 16b 0.11b Table 1.23: Comparisons of adjusted least square means of the LGRC (day"1) for Brood Size at a l l colonies: Brood of 2 Brood of 4 Brood of 1 Brood of 3 0 .23a 0.23a 0.22a 0.22a Table 1.24: Results of the p a r t i a l F and Scheffe's multiple comparison tests for the interaction between colony and laying date at a l l colonies: P a r t i a l F Test Parameter Value SSE (with colony) 0.07 P a r t i a l F = 1.98 SSE (without colony) 0.04 ^critical = 2.01 no. c o e f f . Res. 8 MSE (with colony) 0.0004 Scheffe's Multiple Comparison Test: Colony Colony Slope Intercept 1 2 S* S2 FFI 93 FFI 94 14.32 a 14.21 a 130 FFI 93 MAND 93 17.53 3 16. 52 a FFI 93 FRSR 93 20.75 a 16. 95 a FFI 93 FRSR 94 1.76 0. 41 FFI 94 MAND 93 0.00 0. 04 FFI 94 FRSR 93 2.01 2 . 96 FFI 94 FRSR 94 3.66 3. 84 MAND . 93 FRSR 93 2 .52 2 . 69 MAND 93 FRSR 94 4.24 4. 42 FRSR 93 FRSR 94 8.20 8. 17 ^ critical ~ 9 . 7 6 a d i f f e r e n c e s s i g n i f i c a n t between c o l o n y 1 and 2, p> 0.05 Table 1.25: Results of the p a r t i a l F and Scheffe's multiple comparison tests for the interaction between brood siz e and laying date at a l l colonies: P a r t i a l F Test Parameter Value SSE (with brood) 0.07 P a r t i a l F = 3.33 SSE (without brood) 0.06 F c r i t i C a i = 2.01 no. c o e f f . Res. 6 MSE (with brood) 0.0005 Scheffe's Multiple Comparison Test Brood Brood Slope Intercept 1 2 S f S2 1 2 0.01 0. .02 1 3 1.38 2. .67 1 4 0.0001 0. .21 2 3 . 3.07 3. .97 2 4 0.004 0. .64 3 4 0.03 7 . 62 S critical = 9 - 7 6 a d i f f e r e n c e s s i g n i f i c a n t between b r o o d 1 and 2, p> 0.05 131 Table 1.26: ANOVA source of variation table for 1 o g i s t i c growth rate f o r culmen (LGRj.) at a l l colonies except Five Finger Island 1993: Source of Variation df SS MS F Value P Colony- 3 0.0002 0.00006 0 63 0.60 Brood S i z e 3 0.0005 0.0002 1 91 0.13 P o s i t i o n ( B r o o d Size) 6 0.0003 0.00004 0 47 0.83 Laying Date 1 0.0001 0.0001 1 .24 0.27 Colony*Brood Siz e 9 0.002 0.0002 2 .17 0.03 Laying Date*Colony 3 0.0002 0.00007 0 .71 0.55 Laying Date*Brood Size 3 0.005 0.0002 1 . 90 0.13 Asymptotic Culmen Length 1 0.005 0.005 58 .43 0.0001 E r r o r 110 0.01 0.00009 T o t a l 139 0.05 Table 1.27: Comparisons of adjusted least square between colony and brood size (day"1) except Five Finger Island 1993: means for LGRC at a l l colonies 4/2 4/4 4/1 3/3 5/2 3/2 3/4 4/3 0.21 a 0.18 a 0.16 a 0 .15 a 0.15 a 0.14 a 0.14 a 0 . 14 a 3/1 5/4 2/3 /2/1 5/3 5/1 2/4 2/2 0.14 a 0.13 a 0.13 a 0.11 a 0.10 b 0.09 b 0.07 b 0.04 b where: #/# = c o l o n y / b r o o d s i z e ; 2= F i v e F i n g e r I s l a n d 1994, 3= Mandarte I s l a n d 1993, 4= F r a s e r R i v e r 1993 and 5= F r a s e r R i v e r 1994 132 Table 1.28: ANOVA source of v a r i a t i o n table f o r the l o g i s t i c growth rate f o r Culmen (LGR,) on Five Finger Island 1993: Source of Variation df SS MS F Value P Brood S i z e 2 0.001 0. 0006 1 42 0 .27 P o s i t i o n ( B r o o d Size) 5 0.002 0. 0004 1 12 0 .38 L a y i n g Date 1 0.0003 0. 0003 0 74 0 .40 L a y i n g Date*Brood S i z e 2 0.001 0. 0006 1 39 0 .27 Asymptotic Culmen Length 1 0.002 0 .002 6 23 0 .02 E r r o r 19 0.008 0. 0004 T o t a l 31 0.02 Table 1.29: ANOVA source of v a r i a t i o n table for the time taken to grow from 10 to 90 % ( t 1 0 . 9 0 ^ 1 at a l l colonies: Sources of Variation df SS MS F value P Colony 4 32.23 8.07 1.44 0.23 Brood S i z e 3 4.44 1.48 0.27 0.85 P o s i t i o n ( B r o o d Size) 6 28.90 4.82 0.86 0.53 L a y i n g Date 1 23.43 23.43 4.20 0.04 Colony*Brood S i z e 11 37.89 3.44 0.62 0.81 L a y i n g Date*Colony 4 33.18 8.29 1.49 0.21 La y i n g Date*Brood S i z e 3 4.39 1.46 0.26 0.85 Asymptotic Culmen Length 1 10.52 10.52 1.88 0.17 LGR C 1 676.95 676.95 121.22 0.0001 E r r o r 106 591.96 5.58 T o t a l 140 3356.46 133 Table 1.30: ANOVA source of v a r i a t i o n table for asymptotic t a r s a l length at a l l colonies: Sources of Variation df SS MS F value P Colony 4 109.25 27.31 3.46 0 .01 Brood S i z e 3 8.76 2.92 0.37 0 .77 P o s i t i o n ( B r o o d Size) 6 13.84 2.31 0.29 0 .94 L a y i n g Date 1 0.002 0.002 0.00 0 .99 Colony*Brood S i z e 12 129.52 10.79 1.37 0 .19 L a y i n g Date*Colony 4 111.92 27.98 3.55 0. 009 L a y i n g Date*Brood S i z e 3 8.61 2.87 0.36 0 .78 E r r o r 136 1072.16 7.88 T o t a l 169 1407.40 Table 1.31: Comparisons length (mm) of adjusted least for colony at a l l square means of colonies: t a r s a l FRSR 1994 MAND 1993 FFI 1993 FFI 1994 FRSR 1993 96.29 a 86. 22 a 84.43 a 80 . 53 a 54.84 b Table 1.32: Results of partial F and Scheffe's multiple comparison tests for the interaction between colony and la y i n g date at a l l colonies: P a r t i a l F Test Parameter Value SSR (with colony) 335.16 P a r t i a l F = 3.45 SSR (without colony) 54.71 ^critical = 2.01 no. c o e f f . r e s . 8 MSR (with colony) 10.16 134 Scheffe's Multiple Comparison Test Colony Colony Slope Intercept 1 2 S* S2 FFI 93 FFI 94 0.73 0.85 FFI 93 MAND 93 0.59 0.44 FFI 93 FRSR 93 9.35 11.20 a FFI 93 FRSR 94 0.02 0.04 FFI 94 MAND 93 0.05 0.18 FFI 94 FRSR 93 8.70 10.09 a FFI 94 FRSR 94 0.91 1.58 MAND 93 FRSR 93 12.09 a 12.22 a MAND 93 FRSR 94 0.76 0.97 FRSR 93 FRSR 94 13.41 a 13.68 a S critical — 9.76 a d i f f e r e n c e s s i g n i f i c a n t between c o l o n y 1 and 2, p> 0.05 Table 1.33: ANOVA source of va r i a t i o n table f o r asymptotic t a r s a l length at a l l colonies except Fraser River 1993: Sources of Variation df SS MS F value Colony 3 14.43 4 .81 0. 63 0 59 Brood S i z e 3 9.60 3 .20 0. 42 0 74 P o s i t i o n ( B r o o d Size) 6 17.54 2 . 92 0. 39 0 89 L a y i n g Date 1 . 18.64 18 .64 2. 46 0 12 Colony*Brood S i z e 9 88.85 9 .87 1. 30 0 24 L a y i n g Date*Colony 3 12.86 4 .28 0. 56 0 64 L a y i n g Date*Brood S i z e 3 9.02 3 .01 0. 40 0 76 E r r o r 113 857.47 7 .59 T o t a l 141 1089.00 135 Table 1.34: ANOVA source of v a r i a t i o n table for asymptotic t a r s a l length on the Fraser River 1993: Sources of Variation df SS MS F value P Brood S i z e 2 9.33 4.67 0.43 0.66 P o s i t i o n ( B r o o d Size) 4 16.09 4.02 0.37 0.83 L a y i n g Date 1 35.28 35.28 3.22 0.09 L a y i n g Date*Brood S i z e 1 4.55 4 .55 0.42 0.53 E r r o r 17 186.35 10.96 T o t a l 27 316.35 Table 1.35: ANOVA source of v a r i a t i o n table f o r the l o g i s t i c growth rate f o r Tarsus (LGR,,) at a l l colonies: Source of Variation df SS MS F Value P Colony 4 0.002 0.0005 0.39 0.81 Brood S i z e 3 0.001 0.0004 0.30 0.82 P o s i t i o n ( B r o o d Size) 6 0.004 0.0007 0.58 0.75 L a y i n g Date 1 0.00005 0.00005 0.04 0.84 Colony*Brood S i z e 12 0.01 0.0009 0.66 0.79 L a y i n g Date*Colony 4 0.002 0.0005 0.37 0.83 L a y i n g Date*Brood S i z e 3 0.001 0.0004 0.29 0.83 Asymptotic T a r s a l Length 1 0.04 0.04 30.75 0.0001 E r r o r 132 0.17 0.001 T o t a l 166 0.36 136 Table 1.36: ANOVA source of v a r i a t i o n table for the time taken to grow from 10 - 90 % ( t 1 0 . 9 0 T a r s u s) at a l l colonies: Sources of Variation df SS MS F value P Colony- 4 57.40 14.35 0. 67 0 . 62 Brood S i z e 3 42.96 14 .32 0.67 0 .57 P o s i t i o n ( B r o o d Size) 6 138.97 23.16 1.08 0 .38 L a y i n g Date 1 33.08 33.08 1.54 0 .22 Colony*Brood S i z e 10 108.46 10.85 0.51 0 .88 L a y i n g Date*Colony 4 51.39 12.85 0.60 0 . 66 L a y i n g Date*Brood S i z e 3 41.82 13.94 0. 65 0 .59 Asymptotic T a r s a l Length 1 11.56 11.56 0.54 0 .46 LGR-rp 1 57.14 57.14 2. 66 0 .11 E r r o r 97 2082.90 21.47 T o t a l 130 2886.26 Table 1.37: ANOVA source of v a r i a t i o n table f o r the asymptotic s t r u c t u r a l size at a l l colonies: Sources of Variation df SS MS F value P Colony 4 826.17 206.54 6.91 0.0001 Brood S i z e 3 51.71 17 .23 0.58 0.63 P o s i t i o n ( B r o o d Size) 6 42.01 7.00 0.23 0.96 L a y i n g Date 1 8.78 7.78 0.29 0.59 Colony*Brood S i z e 12 330.98 27.58 0.92 0.52 L a y i n g Date*Colony 4 853.70 213.42 7.14 0.0001 L a y i n g Date*Brood S i z e 3 51.75 17.25 0.58 0.63 E r r o r 124 3705.25 29.88 T o t a l 157 5901.07 137 Table 1.38: Comparisons of adjusted l e a s t square means of tio-go Tarsus (days) at a l l colonies: FRSR 1994 MAND 1993 FFI 1994 FFI 1993 FRSR 1993 146 . 66a 127 . 85a 120.81a 113. 66a 25. 62b Table 1.39: Results of partial F and Scheffe's multiple comparison tests for the interaction between colony and laying date at a l l colonies: P a r t i a l F Test Parameter Value SSE (with colony) 2195.82 P a r t i a l F = 6.85 SSE (without colony) 558.67 F c r i t i c a l = 2.01 no. c o e f f . Res. 8 MSE (with colony) 29.88 Scheffe's Multiple Comparison Test Colony Colony Slope Intercept 1 2 S2 S2 FFI 93 FFI 94 0.38 0.40 FFI 93 MAND 93 0.03 0.004 FFI 93 FRSR 93 . 17.11 a 23.19 a FFI 93 FRSR 94 0.18 0.45 FFI 94 MAND 93 0.55 0.81 FFI 94 FRSR 93 22.45 a 27 .74 a FFI 94 FRSR 94 1.26 1.59 MAND 93 FRSR 93 34.73 a 35.03 a MAND 93 FRSR 94 0.53 0.69 FRSR 93 FRSR 94 20.80 a 22.30 3 ^ critical = 9-76 a d i f f e r e n c e s s i g n i f i c a n t between c o l o n y 1 and 2 138 Table 1.40: ANOVA source of v a r i a t i o n f o r the asymptotic s t r u c t u r a l s i z e at a l l colonies except Fraser River 1993: Sources of Variation df SS MS F value P Colony 3 17.11 5.71 0.19 0 .90 Brood S i z e 3 59.05 19. 68 0. 65 0 .58 P o s i t i o n ( B r o o d Size) 6 91.29 15.21 0.50 0 .80 L a y i n g Date 1 63.47 63.47 2.10 0 . 15 Colony*Brood S i z e 9 146.54 16.28 0.54 0 .84 L a y i n g Date*Colony 3 9.30 3.10 0.10 0 . 96 L a y i n g Date*Brood S i z e 3 58.63 19.54 0.65 0 .59 E r r o r 101 3047.57 30.17 T o t a l 129 3723.83 Table 1.41: ANOVA source of v a r i a t i o n table f o r the asymptotic s t r u c t u r a l s i z e on the Fraser River 1993: Sources of Variation df SS MS F value P Brood S i z e 2 15.86 7.93 0.27 0 .77 P o s i t i o n ( B r o o d Size) 4 98 .53 24 . 63 0.84 0 .52 L a y i n g Date 1 274.54 274.54 9.37 0. 007 L a y i n g Date*Brood S i z e 1 3.72 3.72 0.13 0 .73 E r r o r 17 497.92 29.29 T o t a l 27 1775.04 Table 1.42: ANOVA at a l l source of colonies v a r i a t i o n table for condition index Sources of Variation df SS MS F value P Colony 4 14 .5 3.6 1.35 0 .26 Brood S i z e 3 P o s i t i o n ( B r o o d Size) 6 L a y i n g Date 1 Colony*Brood S i z e 9 L a y i n g Date*Colony 4 L a y i n g Date*Brood S i z e 3 E r r o r 110 T o t a l 140 11.1 9.8 21.0 13. 9 14.2 11.5 295.8 712.1 3.7 1.6 21.0 1.5 3.6 3.8 2.7 1.38 0. 61 7.80 0.57 1.32 1.43 0.25 0.72 0.006 0.82 0.27 0.24 140 CD si -U a -H si •H cn CD •H o rH 0 CJ CD CD H Si J-> 4-> rcj CQ a rrj in O B u o u T J CD 4-> CQ CD M • u r o i ON. CD CTl , si U co a> H — • H — • M | CD 0) „ (fl Cn M — —. co CO (fl (O XI - H C h -it dp H • -— 01 O H Z <D -u — — (fl o\° TJ (0 — -C co (fl (fl 2 2 r Q O Q a H rrj iH CJ) •H in rH O -U CD co O CD a U H o 4H O -U •H 4-J rrj CD H -H 4J Q CO -H TJ a CD 01 di • H O d) a w ro rH LD o IT) • -=r r— co co • o ro i LO CD • CD rH -—- -—- .—. . . r— CO ----- - — rH LO , ro CXC m ro ro rH o rH CO CN ro LD LO ON • ON CM LO r -CN r- o rH rH • • 1 CN LO rH rH rH ON m CNJ rH IT) o o — — ro 1—1 1 '— — CM — CNI ON CO ^ rH ro r - CT CD Cd — ro . CD CO LO 00 LO r - LO . . o ro ro r- vD 00 CO CTl CD rH o • CN o ro ro : ON 00 O • CN 00 o CD • oo CD CTv CM CNJ IT) o CNJ ro o ro • ON rH o CN rsi LO rH r- . O rH CM ro CN1 o O rH -—• CN — ro rH ro CN rH CM — '— LD rH CNJ LD CM ro o -— rH • LO cn —. ON • ro ON • CM o ro LO • o co r- • ro LO LO ON • co CM uo LO — ro • co ro • LO LD • o ro C\J rH CO ro ro ~ -r- ON • r-co • CM CM o CM csl — ro LO • LD co o yo • r— CM • O VD CO — ro o co • ro co CD I D • ro ON ro CM — CM LD O • CD ON O —. CM OS • CO CN • o r-j— _ CD ro yo CM ro co . LO o ro o CD ro — o ro csi • r - r-rH CM o ro • CD ON • r- r-co •— CM ON — . rH O • o CM • rH CO r- -— r-CM rH • r- rH ro ON [-- rH —v — , O to (fl (fl M 3 +J X! •—, u -u UJ CU CO rfl (fl bi rH 3 Si CU cu M - v TJ CU >H U (fl • H — V '—V rH 3 bi - H • U i> (0 ^4 O SH 4-) (fl cn 3 O (0 CM • H rH H a bi G to - H tn CD t<H • H q o j j CO (U bi C u cu g to CO 3 si CU C to u M rH 3 CU O -u (fl CO CO (fl rfl -q CU -LJ U to a o -q a .q crj O C to CU Cfl CO to • H O CM bi 3 4-i -q O <V O -q CM M 0 O q a u • H CU • H >H CO CU u -u - H cu • H ' H UH 0 . in O CU - U cu <X3 14H c^ O TS • H 3 • H O CU Co c - H O O M O q U (fl • H U 3 C Q, rd O rfl O Xi CJ X ! CJ rfl M CU Oi "~ Cu EH CO Oi CM ^ -rH CU rH CU C (fl p +J a -w U q 3 CU (fl CD (0 . V SH o o 4-) (0 C to X) « CU -H CU *H O M rH - H C0 0 T J 0 cu -q T J -q M DH (fl CL, u s - CO ^ CU a c D T J 0) CO • H MH n3 •H CD -P T! a - H d) M T J 0 • H -q co rH Ct cu co c c cu 3 (0 CD TJ •H rH -H (0 0 -p -q 0 EH — CO 3 >H CU cu -u O Q, C tfl (fl X rH CU T J -q a rfl co CO 0) 4 J O >, • H TJ UH O • H E= U E (fl CM ro ro o ID • VD LD • CM CO ro — CD ON . . CO CNJ LD ON ro o VD ON CD CC ro LD CM iD ro rH rH •CT o CD co rH CM 1 o co • ro ^p O LD LD ro ro o LD rH • • rO rH rH rH — — ro CM o O rH CM CD ro O rH rH — — — — — - — rH ~—' CM — ON o CO rH ON . O , 0O CD LD CD CM rH LD LD rH CM rH LD CM ro ^< LO o o rH [- ON rH CD C ro '.') • ON r-CD 1 CO • CM 1 • rH • rH r - r— CO • CD r - O ON • LD ^p ON rH ON OO ON ro o CD rH ro CM LD ro o O rH ON rH CD cn CD rH CD rH •—- CO LD rH CM — — - -— r- — r - rH CO CN •— c -H — , OH CO rH —, 3 0 CO •U 0 3 rt) CO -u -H C in 0) M M Cl J-) — O (C! ai co x: T J CL, Cn to C! to to rfl 3 o -q to CO rH -U s: <D 0 fa -q to 4-1 o o 0 •H to •H 0 >,-H UH (a UH -U M -u -V Xi •H a ci ca o cu o <u rd 1 1 o CO rfl LH 4-) ro, cd — CM ~ - co ^ -CN ro cn — r— •ry • O CO • O to to 3 & (0 o CM (0 £ -t! • H CL M CD x! trj co I I. ro ro ro VD VO o o o rH rH CO CO • r l 0 (1) a CO o ro o . "J 1 in ON oo ro CM 00 rH ro CM rH ro VD VD r-cn CM ro co ro ro CM ON o ON ON uO VD ro r -rH O VD rH rH ON CO VD ro O o i n rH O o r -ro LO i n ro rH CM uO rH rH O 00 CM CM 00 CM o o m o CD -3" CM 00 m 00 r- in o 00 rH CO ro m CNl rH o o o ro co ro ro co co CM o cn i n ro —. -—^ O a CO ca rd • H — X . Y 3 4-1 Si —, a co o -u ca CD co rH - V • cd <a tn rH 3 3 CO 4-1 XJ QJ QJ M —s M O 3 ca CD "~H H 0 fa CD CO 4-1 —i —X rH 3 tn • H 4-1 Q) -U ca " H CQ 0 tn U 4 J O Q, a t~ CD Cn 3 O ca CD CM • H C <0 SH >H SH 4-1 C tn c "> • H tn rd X .—^ o ca 0) CO • H q 0 4 J CO H aj tn rH QJ OH si u cu 6 w CO 3 cu ca •d - q Cn co a to SH M . H 3 QJ o 4-1 cd co co C CO cd 3 ^ ^i QJ "a co - q QJ 4-1 M to a cd Co 4-1 4-1 O - q X -C3 0 a « QJ c/3 co aj CO QJ CO CO 4-1 rH 4-1 • H 0 CM tn 3 ca 4-) si vu o t l ^ O CC o -c; 0 O q TJ t> £ i CO 4-1 O 0 o • H 0) • H M CO QJ U 4-1 • H QJ rH ' H • H T3 •H CO • H O VH a VM 0 QJ 4-1 QJ <TS VM Q< QJ ^ VM O VH ca VM 4 J SM 4-) -H 3 • H 0 QJ CO C E; • H i f c o • H 5 * ^1 . -H Q , SH CO O -H O q • H N^ 0 3 a -a O E 0 QJ O CD (B rH m o CD O si e> .C u cd >H 0 CM cd o to cd >-H 4-1 CM a, ~— CM S - EH CO CM v O — CM " ~ Pi — CM V - CO Appendix IV: Diet of nestling Double-crested Cormorants at Mandarte Island i n 1971 (Robertson 1971) and 1993. Species Mandarte Island 1971 Mass (g) (%) No. F i s h (%) Mandarte Island 1993 Mass (g) (%) No. F i s h (%) P a c i f i c H e r r i n g (Clupea harengus) Northern Anchovy (Engraulis morax) P a c i f i c Salmon (Oncorhynchus spp) Threespine S t i c k l e b a c k (Gasterosteus aculeatus) S h i n e r Perch (Cymatogaster aggregata) S t r i p e d Seaperch (Embiotoca l a t e r a l i s ) P a c i f i c Snake P r i c k l e b a c k {Lumpenus saggitta) Penpoint Gunnel (Apodichthys flavidus) Crescent Gunnel (Pholis laeta) Saddleback Gunnel (Pholis oranta) U n i d e n t i f i a b l e Gunnels (Pholidae spp) T o t a l Gunnel spp (Pholidae spp) P a c i f i c Sandlance (Ammodytes hexapterus) R o c k f i s h (Sebastes spp) P a c i f i c Staghorn S c u l p i n (Leptocottus armatus) S t a r r y Flounder ( P l a t i c h t h y s s t e l l a t u s ) Shrimp (Heptacarpus spp) 109.9 (2.7) 7 (1.3) 4.6 (0.1) 1 (0.2) 66.4 (0.9) 1 (0.2) 5.1 (0.1) 2 (0.4) 1459.4 (20.5) 85 (15.5) 248.2 (3.5) 7 (1.3) 726.8 (10.2) 63 (11.5) 2542.6 (35.7) 130 (23.8) 1130.5 (15.9) 124 (22.8) 3673.1 (51.6) 254 (46.6) 331.1 (4.6) 112 (20.5) 419.4 (5.9) 15 (2.7) 48.1 (0.5) 30.0 (0.3) 4.3 (0.1) (0.3) (0.2) (0.2) 524.6 (5.9) 37 (5.9) 1205.6 (13.5) - 83 (13.2) 2028.7 (22.8) 78 (12.4) 602.0 (6.8) 5 (8.7) 298.6 (3.4) 38 (6.0) 702.2 (7.9) 60 (9.5) 3631.6 (40.8) 231 (36.7) 679.0 (7.6) 172 (27.3) 712.2 (8.0) 8 (1.3) 1734.1 (19.5) 71 (11.3) 325.0 (3.7) 21 (3.3) 3.8 (0.4) 2 (0.3) 143 CO TS SH -H XI Cn a •H 4-> rrj CU XI CQ -H MH o 4-) TS 1) XI CQ • -H 4H >i CU U MH O H a o •H 4 J -H CQ O o u 4-1 CU •H SH 4-> XI -H TS CU % C D ^ . . . G « 0 0 G CD M CD MH-OS rH rH LO (Tl to c#> • • < o m CM rH X I oo 00 CM C D • H ^ rH (Tl 00 CM 00 C D ftdP • H ~ — • • o (Tl LO LO C D tA rH CM rH CM CM oo c • H <D • P 0 ( H C M dp W C D • H O C D ft W >1 (U SH CM co cu • H O CD a, rrj rd . TS 4-1 a CU rd • H XI TS O c U rd CD O ,—. X LO CU 4-> 00 4-1 00 (0 (Tl C en CU rH O rH s 00 CM rH C\] 00 CD CM r~ (Ti rH CM CM rH CO CO o DM TS C rd SH O rH CU X ! ^ U oo 4-i cn (0 rH PQ — o CM o oo LO CM rH CD r> 00 rH rH rH o LO CM LO rH o CM CN CN CM CM CM CM rH CD O CM CO CM • O T i TS TS TS T i • • O • c c c c a LO CTl CM LO T3 T i G T J G TS TS a xs G CM CD 00 CM L O rH rH CTi CM 00 rH CTi CTl L O CTl CTl LO LO CTl CD CM 00 00 LO 00 CM 00 L O o CD LO I 1 CD L O CD LO CD r~ r -CM CTi 4 J trj - H M CD U O CO •H CO q CU a. rrj U CO to T S O - H SH q >i 3 nj "H > ctj X o SH U SH X tn rH rrj tj q • H co c tq OH — < — CO 3 SH 3 rrj CO o co cu SH cu X ) > E= S H O 3 o rrj C O co — co • H . to q CD rH DH CU rrj SH O CU M U rd S to •H CO q cu rrj U to 3 CO 3 X -H. SH to O 3 - H O C M 3 CU U M - H CO rrj O SH H U O <D O ^ 4-1 § X — co — 4-1 — CU CO CnTS c CU Crj OH 3 U CO CO CO "H to q ^ cu O Ci, (0 C/f CO rrj a co CO G trj SH X ! CU X rrj 4-) - H SH 3 O C O CO •H CO 4-i q CU CU CM trj U C a rO CJ trj CU M ft 3 rd Co CJ — CQ CU 4J rrj e •H 4-1 CQ <U U CU 4-> 4-1 rrj e >i u T5 C! 0 T5 (U CQ rrj X) TS CU 4-) a cu CQ CU SH a TS cd TS cu SH cu > a o o CO 4-> CO TS tO 4-1 rO T i 0 a T) CD a o • H - U rd •H > CU H XI XI m 4-1 o M rrj CO CQ O r H o X -H CU 4-> • H C >i rO T J Cn rO T J rrj T J >i rrj T J (0 T J o\° o\° >1 rO T J CU 4-> O 4J to (0 o o\° o CTl (0 TJ >i rO X I >i ro T J 4-1 o CU c CO £ r H U cu CU C . g CO CO - H B CU r H CO SH CJ o SH (0 ro • H SH SH CJ £ 4-> 4H <4H X m X CO CU CJ SH SH SH CU cu x : C CO CO CO T J o O o O o T J 4-1 CU SH rO C o 4H MH 4H CJ C 3 £ CO CO cu B - H - H o r H CO SH Xj QJ 4H cu CU CU CU >1 SH rO ro E 4-1 CO 4-J O 4-1 4-1 4-) 4-1 Cn CU Cn CJ B 4-1 CU CJ (0 (0 (0 rO (0 SH CJ C 3 C SH >i >, SH SH SH SH CU c SH SH SH SH (0 CP u c rO O O o O r H r H 4-> X! H c . C Xi £ xi cu 4-1 4H 4H 4H 4H rO rO H 4-> CU cu 4-1 4-> 4-> 4-> SH 4J 4-J O c s 3 3 3 cu O r H r H r H r H O O o CU t r O O o o r H (X ro rO to (0 4-1 4-J B SH CU SH SH SH SH XI B > > > > • H Cn CU SH Cn Cn Cn Cn rO • H SH SH SH SH 4H 4H r H m N CU CU cu CU O O CU CU X! CJ CJ O O • H CU 4-1 4-1 4-> 4-> 4-> 4-) • H 4-1 • H - H • H • H r H > c C c c 4-1 4-J 4-> c 4-1 4-1 4-1 4-1 O - H • H - H • H • H C C r H r H CO cu CO CO CO CO X! 4-1 CU CD CU O O cu u • H • H • H - H (0 rO CU CU CU CD CJ CJ B to CO Cn SH Cn Cn Cn Cn 4-1 r H B 6 £5 B SH SH rO X! • i H CU O O O o cu QJ • H • H • H - H CU CD 2 < < Q D-i h—1 HH HH S ai E H EH EH EH CM OH I—i ai H rj Q O ^ of cT CJ U 0 0 (5 U ^ HH l-H HH S 04 4-J 2 EH EH 

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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088450/manifest

Comment

Related Items