Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Foraging behaviour of captive black-tailed deer (Odocoileus hemionus columbianus) Gillingham, Michael Patrick 1985

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

Item Metadata

Download

Media
831-UBC_1985_A1 G54_4.pdf [ 8.35MB ]
Metadata
JSON: 831-1.0096617.json
JSON-LD: 831-1.0096617-ld.json
RDF/XML (Pretty): 831-1.0096617-rdf.xml
RDF/JSON: 831-1.0096617-rdf.json
Turtle: 831-1.0096617-turtle.txt
N-Triples: 831-1.0096617-rdf-ntriples.txt
Original Record: 831-1.0096617-source.json
Full Text
831-1.0096617-fulltext.txt
Citation
831-1.0096617.ris

Full Text

FORAGING BEHAVIOUR OF CAPTIVE BLACK-TAILED DEER (Odocoileus hemionus columbianus) by MICHAEL PATRICK GILLINGHAM .Sc., Macdonald College of McGill University, 1978 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Forestry) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September 1985 © Michael Patrick Gillingham, 1985 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of FORESTRY  The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date October 11, 1985 i i ABSTRACT A review of the l i t e r a t u r e on b l a c k - t a i l e d deer (Odocoileus hemionus columbianus Richardson) feeding habits reveals considerable v a r i a t i o n among animals, locations, and seasons. Processes a f f e c t i n g food selection, however, are poorly understood. Optimal foraging theory was explored as a means of predicting deer foraging behaviour and diet breadth. Because of complex constraints and objectives involved in predicting diet selection, food preference was determined under ad libitum conditions. Feeding behaviour of two deer was studied in a 0.5-ha enclosure to examine the e f f e c t s of density and d i s t r i b u t i o n of their preferred foods on diet selection. When deer had to search for food, diet selection remained the same as that under ad libitum conditions as long as preferred food was abundant. Deer nearly exhausted their highly preferred food before switching to lower ranked foods. This switch was gradual, as deer continued to search for preferred food. The amount of preferred food already eaten during a t r i a l was p o s i t i v e l y correlated with the time that animals would continue searching before switching to lower-ranked foods. Switching was related to the amount and type of food encountered and not to the amount of food in the the pen. Dispersion of the preferred food (clumped versus unclumped) had no s i g n i f i c a n t e f f e c t on the amount of food eaten, but did s i g n i f i c a n t l y influence the types of food encountered by one of the two animals. Both animals became more e f f i c i e n t (intake per distance travelled) at finding preferred foods with increasing experience with a s p e c i f i c d i s t r i b u t i o n of food. Animals increased their e f f i c i e n c y of finding apples by repeating searching patterns which had been e f f e c t i v e during previous t r i a l s . Performance was poor, however, when d i s t r i b u t i o n s were changed. When preferred food was abundant, platforms containing preferred food were not always completely cleared of food the f i r s t time a platform was v i s i t e d . Intake rates of non-preferred foods tended to increase with declining abundance of preferred food. This increase was not caused by changes in the amount of non-preferred food eaten at feeding stations, but rather by the rate at which non-preferred feeding stations were v i s i t e d . The influence of i n t r a s p e c i f i c plant v a r i a t i o n on food habit studies and the u t i l i t y of preference indices are discussed. I conclude that foraging bouts are highly dynamic and that some foraging questions may not be adequately answered i f t h i s internal variation i s ignored. iv TABLE OF CONTENTS ABSTRACT i i LIST OF TABLES v i i LIST OF FIGURES x ACKNOWLEDGEMENTS x i i Chapter 1 - General Introduction 1 Terminology 8 Thesis Structure 10 Chapter 2 - The effects of learning of food d i s t r i b u t i o n s on food selection and search paths 12 Introduction 12 Hypotheses 14 Methods 16 S t a l l t r i a l s 18. Pen t r i a l s 20 Results 28 S t a l l t r i a l s 28 Food Preference: Hypotheses 1 and 2 28 Factors influencing selection 32 Pen t r i a l s 35 Effec t of searching on selecti o n : Hypothesis 3 .. 38 Effec t of food abundance on food selection: Hypothesis 4 42 Effec t of memory on food selection: Hypotheses 5 and 6 44 Search Paths 50 Detection distance 53 Other effects on selection, 54 Discussion 55 S t a l l T r i a l s 55 Pen T r i a l s 57 Conclusions 66 Chapter 3 - The ef f e c t s of food a v a i l a b i l i t y and d i s t r i b u t i o n on food selection 68 Introduction 68 Hypotheses 7'0 Methods 72 S t a l l t r i a l s 73 Pen t r i a l s 74 Results 87 S t a l l T r i a l s 87 Food Preference: Hypotheses 1 and 2 87 Pen t r i a l s 96 Variation in intake rates 97 Food encountered versus food available 101 Clearance of apple platforms 101 Eff e c t s of searching on selection: Hypothesis 3 .103 Eff e c t s of density and dispersion 110 Preference Ratios 115 Consumption of non-preferred foods: Hypothesis 8 116 Food switching: Hypothesis 9 117 Turning frequency of searching: Hypothesis- 10 ...12-2 Discussion 122 S t a l l T r i a l s 122 v i Pen T r i a l s 126 Conclusions 138 Chapter 4: Implications 141 Literature Cited 154 v i i LIST OF TABLES Table 2.1. Food selection of two b l a c k - t a i l e d deer based on proportion of bites taken and time spent eating foods 30 Table 2.2. Proportions of weight consumed of pelleted ration, apples, and pelleted a l f a l f a between two bla c k - t a i l e d deer under ad libitum conditions 31 Table 2.3'. Preference of two bl a c k - t a i l e d deer for pelleted dairy ration, apples, and pelleted a l f a l f a based on time b i t i n g each food to 5, 10, and 60 min ... 33 Table 2.4. Comparison of weight, weight per b i t e , and weight per s b i t i n g between two b l a c k - t a i l e d deer during ad libitum s t a l l t r i a l s 34 Table 2.5. Summary of sample sizes for pen t r i a l s conducted in 1981 by subject and d i s t r i b u t i o n 37 Table 2.6. Food selection of the female during 1981 pen t r i a l s based on time eating each food 39 Table 2.7. Food selection of the male during 1981 pen t r i a l s based on time eating each food 41 Table 2.8. Comparison of the proportion of apples consumed, by d i s t r i b u t i o n , for both deer 43 Table 2.9. Influence of consecutive exposures to the same d i s t r i b u t i o n on the e f f i c i e n c y of locating apples 47 Table 2.10. Comparison of. search paths before and after changes in food d i s t r i b u t i o n 51 Table 2.11. Comparison by contingency table analyses of v i i i the search paths used during d i s t r i b u t i o n s 52 Table 3.1. Summary by subject and food d i s t r i b u t i o n of 1983 pen t r i a l s " 80 Table 3.2. Preference of both b l a c k - t a i l e d deer based on handling time for 1983 ad libitum s t a l l t r i a l s 89 Table 3.3. Examination of consistency of selection for both deer during 1983 ad libitum s t a l l t r i a l s 91 Table 3.4. Comparison of weight consumed and weight available during 1983 s t a l l t r i a l s 92 Table 3.5. Estimation of i n i t i a l weight of apples consumed by each animal during 1983 s t a l l t r i a l s 94 Table 3.6. Average intake rate based on t o t a l weight consumed for both animals during 1983 s t a l l t r i a l s .... 95 Table 3.7. Comparison of consumption rates of pelleted dairy ration and pell e t e d a l f a l f a by the male 98 Table 3.8. Comparison of time required to consume apples during pen t r i a l s among time intervals 100 Table 3.9. Comparison of available platforms and types of platforms seen during 1983 pen t r i a l s 102 Table 3.10. Comparison of number of apple platforms uncleared of apples, during male t r i a l s 104 Table 3.11. Selection by the female during 1983 pen t r i a l s based on cumulative time handling each food 105 Table 3.12. Selection by the male during 1983 pen t r i a l s based on cumulative time handling each food 106 Table 3.13. Comparison of female; f.ood selection among 0-5, 5-10, and 10-60 min intervals during 1983 pen t r i a l s .. 108 Table 3.14. Comparison of male food selection among 0-5, ix 5-10, and 10-60 min intervals during s t a l l t r i a l s 109 Table 3.15. The eff e c t s of apple density, dispersion, and their interaction, on the foraging of the male and female b l a c k - t a i l e d deer 111 Table 3.16. Comparison of strata means for which s i g n i f i c a n t density effects were present during pen t r i a l s 114 Table 3.17. Comparison of consumption rates of apples within and among pen t r i a l s with the female 118 Table 3.18. Comparison of consumption rates of apples within and among pen t r i a l s with the male 119 Table 3.19. Comparison of consumption rates of non-apple foods for the male during pen t r i a l s 120 X LIST OF FIGURES Figure 2.1. Photograph showing ad libitum food presentation during s t a l l t r i a l s used to esta b l i s h preference 19 Figure 2.2. Photograph of a pen t r i a l showing food platforms, observation tower, and grid system for animal location 22 Figure 2.3. Schematic presentation of d i s t r i b u t i o n s 1 (a) and 2 (b) used during 1981 pen t r i a l s 24 Figure 2.4. Method of comparing search paths between t r i a l s and d i s t r i b u t i o n s 27 Figure 2.5. Comparison of f i r s t bite during 1981 s t a l l t r i a l s r e l a t i v e to food type and location 36 Figure 2.6. The effect of experience within a d i s t r i b u t i o n on the e f f i c i e n c y of finding apples 45 Figure 2.7. Minimum distance required to locate a l l apple platforms for d i s t r i b u t i o n s 1 and 2 49 Figure 3.1. Distributions of food types used during 1983 pen t r i a l s 76 Figure 3.2. Reconstruction of a sample search path for 1983 pen t r i a l s i l l u s t r a t i n g 5-m detection distance ... 81 Figure 3.3. Example of the technique used to compare intake rates of foods within 1983 pen t r i a l s 86 Figure 3.4. Examination, of turning r a d i i as- a. function, of food consumption 88 Figure 3.5. Graphical representation of the interaction x i e f f e c t of apple density and dispersion during 1983 pen t r i a l s 112 Figure 4.1. A model of the influence of i n t r a s p e c i f i c v a r i a t i o n in quality on diet selection 145 ACKNOWLEDGEMENTS I would l i k e to acknowledge my supervisor, Dr. Fred L. Bunnell, who, in addition to his encouragement, stimulated some of the ideas that prompted this investigation. Dr. C. Lee Gass contributed much time and e f f o r t , improving the quality of thi s thesis, and providing me with the opportunity to improve as a researcher. Discussions with Dr. David M. Shackleton provided both in s p i r a t i o n and opportunities to int e r r e l a t e d i s c i p l i n e s . The advice and comments of other members of my committee, Drs. James P. Kimmins and Carl J. Walters, were appreciated. I wish to thank Dr. Chris C. Shank, who provided much helpful advice both with t h i s study and science in general. My thanks also to fellow graduate students for both discussions and confrontations which helped to better formulate my ideas. More than one c r i s i s was abated through the assistance of Peter W. Sanders and members of the sta f f at the U.B.C. Research Forest. I give them my thanks. Diane G. Car-t.wright, provided f i e l d assistance during the f i r s t year of thi s study. This research was supported by N.S.E.R.C. through a grant to Fred L. Bunnell and a scholarship to the author. I also received a fellowship from Canadian Forest Products. 1 CHAPTER 1 - GENERAL INTRODUCTION Food habits of North American Cervidae have received much attention, and Columbian b l a c k - t a i l e d deer (Odocoileus hemionus  columbianus) have been no exception (Cowan 1945, Klein 1965, see reviews by Crouch 1979 and 1981). Knowledge of the types and quantities of foods consumed can be revealing, and i s one step to understanding an animal's natural history, but l i s t i n g of these diet components t e l l s us l i t t l e about food selection processes (Robbins 1983). For example, the fact that deer consume fireweed (Epilobium) i s of l i t t l e s ignificance i f we have no knowledge of t h i s plant's n u t r i t i o n a l value, or of how deer obtain i t (see Mautz 1974). Several factors influence diet composition and selection of ruminants: d i g e s t i b i l i t y (Smith 1952, Radwan and Crouch 1974); p a l a t a b i l i t y (Heady 1964, Longhurst et a l . 1968, Tucker et a l . 1976) including odour, taste (Arnold et a l . 1980), and chemical composition (Arnold and H i l l 1972:, Krueger et a l . 1974, Radwan and Crouch 1974, Connolly et a l . 1980, L e s l i e et a l . 1984); and a v a i l a b i l i t y (Dublin 1980, LaGory et a l . 1985). Results of these d i f f e r e n t studies vary greatly (see Crouch 1979). When Spalinger (1980) compared food habit studies of Nevada mule deer (0. h. hemionus), he found that selection varied s i g n i f i c a n t l y among study s i t e s , seasons, and years. Food habits of herbivores are usually examined and compared among habitats or species using broad groupings of 2 plants; forage i s described in terms of species, families, or even graze, or browse (e.g.,, Dublin 1 980, Hanley 1980, Rochelle 1980, Hanley and McKendrick 1983 and 1985, L e s l i e et a l . 1984). A comprehensive theory to explain diet selection of large herbivores i s lacking. The effects of d i s t r i b u t i o n and abundance of food items on food selection remain largely unknown for native ungulates. This thesis describes research I conducted to determine the effects of d i s t r i b u t i o n and abundance of preferred foods on diet selection by black-tailed deer. The role of experience with a food d i s t r i b u t i o n on food selection was also examined. In order to conduct this research I needed a method of predicting food selection under conditions of equal food a v a i l a b i l i t y . The ecological l i t e r a t u r e contains many examples of pri n c i p l e s of optimization applied to theories of natural selection (see Krebs 1978). Animals make decisions about where, when, and what to eat (Bunnell and Gillingham 1985), and optimality approaches have been used to examine these decisions. Aspects of foraging behaviour that have been considered by optimal foraging theory have included breadth of diet (e.g., Emlen 1966, MacArthur and Pianka 1966, Schoener 1971, Charnov 1976a), movement strategies (Beukema 1968, Cody 1971, Smith 1974a and b), and the optimal use of patchy environments (MacArthur and Pianka 1966, Krebs et a l . 1974, Charnov 1976b). Reviews include those by Schoener (19*71 ) , Pyke et a l . (1977), Krebs (1978), Kamil and Sargent (1981), Emlen (1984), and Krebs and McCleery (1984). 3 Optimal foraging models usually assume that animals maximize energy intake while foraging. Schoener (1971) grouped animals into two broad classes: (1) 'energy maximizers' whose potential reproductive success increases as a function of net energy gain and (2) 'time minimizers', whose reproductive success does not increase beyond a net energy gain threshold. Large herbivores may spend almost a l l of their time feeding with l i t t l e time remaining for other a c t i v i t i e s . Like predators, they should, omit hard-to-handle items from their diets i f they can ignore the time l o s t between items eaten, and concentrate on energy gain. On the other hand, a mouse that does not know when a foraging bout w i l l end with a dash for cover, should concentrate on food items than can be dealt with fastest (see Emlen 1984). C l a s s i f y i n g foragers as time minimizers and nutrient or energy maximizers i s n o n t r i v i a l (Hixon 1982). A researcher could pose both objectives for the same animal at d i f f e r e n t points in i t s l i f e - c y c l e , within a sea son<a 1 fiF-amewor k Optimal foraging theory has been used to predict foraging behaviour of large herbivores. Belovsky (1978, 1981) formulated a linear programming model for moose (Alces alces) based on sodium l i m i t a t i o n (Jordan et a l . 1973). Hanley (1980) pointed out a number of weaknesses with Belovsky's model, and concluded the model was not capable of Belovsky's expectations. Spalinger (1980) had only l i m i t e d success in predicting foraging behaviour of mule deer using linear programming techniques,, although Belovsky (1984) suggested that the f a i l u r e 4 of the predictions was due to problems with the model and not the technique., Belovsky (1984) reviewed and compared three types of models that have been applied to herbivore foraging: contingency models (Owen-Smith and Novellie 1982), nutrient content weighted by food abundance (Stenseth and Hansson 1979), and linear programming (Belovsky 1978). He noted that the multiple constraint foraging systems of herbivores best lent themselves to linear programming techniques, but that a very s p e c i f i c objective function, such as sodium maximization, was required. Construction of any optimal foraging model involves postulating what i s being optimized, ranking food items r e l a t i v e to this currency, and hypothesizing constraints facing the animal. Food 'value' or 'currency' i s not necessarily a parameter that can be scaled in a single dimension (Emlen 1984). None of these components can be e a s i l y stated for b l a c k - t a i l e d deer. Ruminant t a c t i c s are based on a high e f f i c i e n c y of nutrient extraction and u t i l i z a t i o n (Bell 1971). Compared with hind-gut fermenters, ruminants cannot s i g n i f i c a n t l y increase passage rates of digesta as d i g e s t i b i l i t y decreases (Janis 1976). D i g e s t i b i l i t y might be considered a currency, given nutrient (including protein) constraints. Protein could be as equally important an objective for deer (Ullrey et a l . 1967a, Nagy et a l . 1969). Many authors (including Westoby 1974, 1978, Freeland and Janzen 1 974, Ldndroth 1979) have argued that an optimal mix of foods and nutrients in the diet or the avoidance of plant secondary compounds (see E l l i s et a l . 1976) are also 5 important objectives for herbivores. Wild deer have been poisoned by single species diets (Quinton 1985), but because they are generalist herbivores, t h i s problem i s l i k e l y rare (Freeland and Janzen 1974). Black - t a i l e d deer have been shown to strongly reject certain poisonous plants although they have high tolerance to other s p e c i f i c toxins (Dean and Winward 1974). For a variety of herbivores including hare (Lepus  americanus) and moose, Bryant and Kuropat (1980) found that preferred foods were not ranked on energy or nutrient content but that there was a strong negative c o r r e l a t i o n between selected foods and the extractable resin content of plant tissue (terpenes and phenols). This selection often resulted in reduced ingestion of c a l o r i e s or protein. For large herbivores, energy a c q u i s i t i o n may be only a seasonal problem (see Nudds 1980). Objectives may s h i f t from energy maximization in seasons of high a v a i l a b i l i t y of forage, to minimization of energy expenditure during periods of high energy d e f i c i t . Most changes in plant n u t r i t i v e status are caused by numerous inter r e l a t e d factors whose ef f e c t s often cannot be distinguished from one another (Nelson and Leege 1982). Possible objectives for the forager may also covary (see L e s l i e et a l . 1984). For example, protein content of forage may vary with the d i g e s t i b i l i t y of the plant. Interaction between mixed diets' and d i g e s t i b i l i t y may be pos i t i v e or negative. Some mixed species diets have been shown to be more di g e s t i b l e than the sum of diet components digested separately using 6 bla c k - t a i l e d deer digesta in v i t r o (Rochelle 1980). Much l i t e r a t u r e documents the i n h i b i t o r y e f fects of plant compounds on digestion (Oh et a l . 1967 and 1968, Jung 1977, Freeland and Janzen 1974). Dietz et a l . (cited in Klein 1969) found that p a l a t a b i l i t y of big sagebrush (Artemisia t r i d e n t a t a ) , in southwestern Colorado, depended on i t s proximity to other species. White et a l . (1982) indicated that mastication of big sage was reponsible for lowering monoterpenoid content of ingesta to 23% of unmasticated lev e l s for pygmy rabbits (Brachylagus idahoensis). Time scale must be considered when using optimal foraging theory to predict foraging behaviour of herbivores. Insects (Charnov 1976b) and hummingbirds (Gass and Montgomerie 1981) may need to meet requirements on a meal to meal basis, but ruminants may meet seasonal requirements without approaching optimality during individual foraging bouts. Though often unstated, a basic assumption of the objective function postulated by the reseacher i s : the animal i s capable of ranking food along the same gradient as the model. Deer probably select for correlates of d i g e s t i b i l i t y or protein, not the quantities themselves as i t i s unlikely an animal recognizes nitrogen, crude f i b r e , energy, or ash because they do not exist separately as such within the plant (Arnold and Dudzinski 1978). Unless deer are able to assay q u a l i t y without ingestion, some types of food may be consumed simply to sample available food quality (Freeland and Janzen 1974, Westoby 1978). In addition, the observer must be able to measure the value of the consumed food item in the same currency as the 7 animal. Without a currency potential food items cannot be ranked. Searching and handling a c t i v i t i e s for ruminants are not ea s i l y d i f f e r e n t i a t e d . The forager continues to search while handling (chewing and swallowing) a food item and defers some of the handling time u n t i l l a t e r rumination periods. Bite size may vary from plant to plant and species to species (C o l l i n s et a l . 1978, Trudell and White 1981, Wickstrom et a l . 1984); d i g e s t i b i l i t y and rumen turnover may depend not only on forage quality, but also on synergistic interactions among d i f f e r e n t ingesta in the rumen (Rochelle 1980). Because of the complexity of the problem, I decided that I could not treat the foraging behaviour of b l a c k - t a i l e d deer as an optimization problem. Given the present knowledge of constraints and objectives, the optimal foraging model would have been as much a test of the model's assumptions as i t would have been of deer diet s e l e c t i o n . Accurate a p r i o r i predictions of food ranking could not be made without an appropriate currency. However, consistent selection by deer for food types under ad libitum conditions could be used to estimate animals' ranking of available foods. I used consistency of selection during ad libitum t r i a l s as a c r i t e r i o n for establishing preference ranking. Whether animals could survive on the preferred food alone was not examined. The n u t r i t i o n a l basis for food selection was beyond the- scope of thi s study. Once a clear ad libitum ranking of food preference was established, the effects of d i s t r i b u t i o n and abundance of the 8 preferred food, and the role of experience with food d i s t r i b u t i o n were examined by introducing the dee.r to various s p a t i a l food patterns in a r e l a t i v e l y simple environment. This approach enabled me to examine the foraging behaviour of black - t a i l e d deer under r e l a t i v e l y controlled conditions. I was able to examine the eff e c t s of d i s t r i b u t i o n and abundance of preferred foods on diet selection, determine the effect of an animal's f a m i l i a r i t y with a food d i s t r i b u t i o n on foraging behaviour, and determine whether or not food selection could be predicted by simply overlaying ad libitum preference and food d i s t r i b u t i o n . Terminology Much of the l i t e r a t u r e pertaining to feeding habits of ungulates suffers from ambiguous terminology. To c l a r i f y my use of terms, a series of d e f i n i t i o n s follows. Preference has a variety of meanings. Some, including Vangilder et a l . (1982>), have used i.t- to mean- a p r i n c i p a l food. More commonly, a preferred food i s defined as one that i s taken in higher proportion than i t s r e l a t i v e a v a i l a b i l i t y (Petrides 1975, Crawley 1983), and a variety of methods exist for i t s ca l c u l a t i o n : the index of e l e c t i v i t y (Ivlev 1961), contingency tables (Hanson and Libisky 1964, Buchler 1976), chi-square and Bonferroni z s t a t i s t i c s (Neu et a l . 1974), and difference rankings (Johnson 1980). The l i t e r a t u r e concerning animal preference has been recently reviewed by Skiles (1984). The ra t i o method (Van Dyne and Heady 1965, Jacobs 1974, Chesson 9 1978) i s perhaps the most widely used. Johnson (1980) pointed out that for a given data set the presence of, other items will, influence the rating for a given item. Krueger (1972) compared four r e l a t i v e preference indices and found that ranking of preference values was d i f f e r e n t for every index that he tested. Whether preference should change with abundance depends on whether i t i s defined under ad libitum conditions or from f i e l d r e s u l t s . Preference should be a more s t a t i c quantity, defined under conditions of equal a v a i l a b i l i t y ( E l l i s et a l . 1976, Johnson 1980, Crawley 1983). Therefore, I defined preference under ad libitum conditions. I considered a food preferred i f i t was consumed s i g n i f i c a n t l y more than others. Abundance of a food item was the amount of the component present, but not necessarily available, or known, to the animal. A v a i l a b i l i t y of food implied a c c e s s i b i l i t y to the animal. The difference between abundance and a v a i l a b i l i t y was not merely dependent on the physical reach of the animal. An animal foraging through the enclosure might be able to detect food types for only a fixed width on either side of i t s search path. Different forage ratios would be determined based on what was considered available (see Johnson 1980). Selection referred to the foods that were eaten. The animal's preference, modified in part by a v a i l a b i l i t y and d i s t r i b u t i o n , resulted in selection of food items. Under ad libitum conditions, preference- and selection were synonymous. P a l a t a b i l i t y and unpalatability were used to describe foods that were acceptable to the animal. 10 A bite was considered to la s t as long as the animal's mouth remained in contact with the food substrate. The animal's ingestion of p e l l e t s d i f f e r e d from the normal process of b i t i n g and chewing. Handling time was considered to be the combination of biting-and chewing food material for 1983 data. Actual ingestion (a bite) was not included in handling time during 1981. Rumination was excluded in both cases. Unless otherwise stated s t a t i s t i c a l significance was based on a c r i t i c a l value (a) of 0.05. Because Scheffe's test i s conservative (Jones 1984), an a of 0.10 was used when these were employed. Thesis Structure Data presented in t h i s thesis were co l l e c t e d over a three-year period. Two sets of intensive f i e l d t r i a l s were conducted; due to changes in experimental design and in preferences by the deer over this time, they are presented separately at the expense of some overlap. Hypotheses addressed during the research are presented in the introductions to Chapters 2 and 3. Chapter 2 examines the effects of d i s t r i b u t i o n and experience on food selection by b l a c k - t a i l e d deer. Animals were repeatedly exposed to the same food d i s t r i b u t i o n and the effects of experience were examined r e l a t i v e to food s e l e c t i o n , e f f i c i e n c y of finding food, and searching behaviour. Ef f e c t s of food a v a i l a b i l i t y and d i s t r i b u t i o n on se l e c t i o n 11 were more rigorously addressed through the modified design used in 1983. Chapter 3 examines these tests, as well as discussing the influence of assumptions concerning a v a i l a b i l i t y on predictions of food selection and preference. Chapter 4 discusses the implications of these controlled tests under more variable conditions. Interactions of i n t r a s p e c i f i c plant v a r i a t i o n with animal preferences are discussed. The results of my experiments are incorporated into t h i s discussion and the u t i l i t y of preference r a t i o s are examined. 12 CHAPTER 2 - THE EFFECTS OF LEARNING OF FOOD DISTRIBUTIONS ON FOOD SELECTION AND SEARCH PATHS Introduction For many years, animal psychologists have studied the e f f e c t s of experience on animal behaviour (see Kamil and Yoerg 1982); rats have been studied in mazes, pigeons in Skinner boxes. Investigators of foraging behaviour have recently used sim i l a r tests and devices to examine the wisdom of foraging, decisions and the a b i l i t y of foragers to learn to exploit their environment (see Shettleworth 1984). For example, Beukema (1968) found that hunger and experience affected foraging e f f i c i e n c y in three-spined sticklebacks (Gasterosteus  aculeatus). Learned food aversion has also been studied in a number of species. Zahoric and Houpt (1977) examined the a p p l i c a b i l i t y of learned aversion experiments ( n u t r i t i o n a l wisdom) to large herbivores. When deleterious substances were-presented- in combination with single-item d i e t s , c a t t l e learned to avoid the novel food provided the poison was fast acting. Effects of either a long delay in deleterious action, or a mixed-species d i e t , remain untested. Herbivores are known to avoid toxic plants (see Crouch 1981), or to select foods that are high in nutrients that are lacking in their diets (Arnold and Dudzinski 1978). There are examples, however, of deer eating food that i s not n u t r i t i o n a l l y good for them (Ullrey et a l . 1967b and 1972). Fraser and Reardon (1980) conducted a feeding 13 experiment at a mineral l i c k and showed that moose and white-tailed deer (Odocoileus virginianus) consistently selected aqueous solutions of NaCl, NaHC03, and Na2SOi, and they concluded that the animals were selecting sodium. Gluesing and Balph (1980) found that sheep that previously had a l f a l f a a vailable in their pastures, were more l i k e l y to consume a l f a l f a in a new pasture. Matthews and Kilgour (1980) discussed the e f f e c t of prior experience or exposure on subsequent preference and intake for a variety of grazers. Although there i s c o n f l i c t i n g evidence on the extent that herbivores learn and are n u t r i t i o n a l l y wise, present foraging models make many assumptions about an animal's f a m i l i a r i t y with the environment and i t s a b i l i t y to learn. The marginal value theorem (Charnov 1976b) assumes that animals recognize food patches of varying q u a l i t y . Westoby (1974), Freeland and Janzen (1974), and Lindroth (1979) have suggested that animals sample various foods. Knowledge of the d i s t r i b u t i o n of the available foods should enable foragers to better exploit t h e i r environments; an expectation of what might be encountered could influence the types and quantities of food eaten. Optimal use of food resources by a forager might also be based on expectation of intake. Optimal foraging models predict that when an intake expectation i s not met or exceeded, diet selection should be altered, re s u l t i n g in either increased or decreased diet breadth (see Krebs 1978). Not only should thi.S' expectation change while foraging (Charnov 1976b), but i n i t i a l food selection might also be based on previous foraging experience. 14 The s e l e c t i v i t y of a foraging animal may depend on how the animal's perception of the available food interacts with i t s preference and how long the animal searches. Whether or not f a m i l i a r i t y with a d i s t r i b u t i o n and an i n i t i a l expectation of a v a i l a b i l i t y w i l l a l t e r t h i s preference i s of interest. In general, I believe that researchers studying ungulate feeding behaviour do not pay enough attention to the problem of estimating forage a v a i l a b i l i t y . In addition, i f an animal's f a m i l i a r i t y with a d i s t r i b u t i o n of food influences i t s diet selection then the present concept of available forage used in estimating preference r a t i o s (see Johnson 1980, Skiles 1984) may need to be reexamined. Using a c o n t r o l l e d food d i s t r i b u t i o n , and fixed food items, I examined the e f f e c t of food d i s t r i b u t i o n on food selec t i o n . By comparing food preference under ad libitum conditions with s e l e c t i o n under varying food d i s t r i b u t i o n s and f a m i l i a r i t y with the d i s t r i b u t i o n s , I was able to test the effe c t s of s p a t i a l v a r i a t i o n and learning, on diet s e l e c t i o n . Hypotheses Because of the absence of an a p r i o r i preference for the test foods, ad li b i t u m c a f e t e r i a - s t y l e feeding t r i a l s ( s t a l l t r i a l s ) were used to test the following hypotheses: (1) Deer would exhi b i t a consistent preference for food types, when repeatedly given the same choice of foods; 15 (2) These preferences would not d i f f e r between animals. In conducting c a f e t e r i a - s t y l e feeding t r i a l s to test these two hypotheses, I also tested the time spent eating foods, duration of food deprivation, and the effect of i n i t i a l feeding preference on consumption throughout the t r i a l . The e f f e c t of making animals search for food on food consumption was examined after a stable preference was established in s t a l l t r i a l s . By repeatedly presenting an animal with the same food d i s t r i b u t i o n , and then a l t e r i n g i t , I tested the following hypotheses: (3) The addition of a s p a t i a l component to food abundance would not -alter food selection by deer when food remained unlimited r e l a t i v e to consumption; (4) Food s e l e c t i v i t y would be altered (and thus d i f f e r from ad libitum preference) by low abundance of preferred foods at the start of t r i a l s and an increase in the use of lower-ranked food items would be observed; (5) Memory plays a role in the; selection of food, and therefore-, deer would exploit a food d i s t r i b u t i o n more e f f i c i e n t l y after repeated exposure to the same food d i s t r i b u t i o n ; (6) If animals are using previous experience with a d i s t r i b u t i o n to find preferred food, the e f f i c i e n c y with which preferred foods are located should be lower in t r i a l s in which a d i f f e r e n t s p a t i a l d i s t r i b u t i o n of food i s encountered. In addition to the above hypotheses I examined the intern a l dynamics of foraging bouts. Data from 1981 pen t r i a l s 1 6 were also used to: (a) examine s h i f t s in the rate at which platforms were cleared of apples; (b) evaluate s i m i l a r i t i e s in search paths used in d i f f e r e n t d i s t r i b u t i o n s ; (c) examine distances at which the contents of platforms appeared to be detected by the deer; and (d) examine i f wind speed and dire c t i o n influenced the distance that the contents of feeding platforms were detected by the animals. Methods I n i t i a l ad libitum preference was established by cafe t e r i a t r i a l s (hereafter termed " s t a l l t r i a l s " ) . Inclusion of s p a t i a l v a r i a t i o n was accomplished by repeating similar experiments in a 0.5-ha enclosure (hereafter termed "pen t r i a l s " ) . A l l work was carried out at the the University of B r i t i s h Columbia Research Forest, Maple Ridge, B.C. Results presented in t h i s chapter are based on data I c o l l e c t e d from July 1st to August 31st, 1981. Two yearling b l a c k - t a i l e d deer-(one male and: one, female) were used in the study. These animals were selected based on their t r a c t a b i l i t y from six captive-born, mother-raised deer. The other four animals were too wild to tolerate the degree of handling required for my experiments. Data c o l l e c t e d from experiments using small numbers of animals cannot be accurately extrapolated to population parameters. However, there are numerous examples of studies conducted with only a few animals (e.g., Krueger 1972, Wallmo et a l . 1973, Trudell and White 1981, Wickstrom 1983, Cooper 1 7 1985, Gillingham and Bunnell 1985, Holechek and Valdez 1985). The aim. of these studies was either to evaluate existing techniques used for studying ungulates or to test for the presence or absence of s p e c i f i c behaviours during ruminant foraging. In my study, larger numbers of animals would have presented a better sample of foraging behaviour. T r i a l s conducted in the 0.5-ha enclosure, however, required large amounts of setup and c o l l e c t i o n time, permitting no more than two t r i a l s per day within the normal peak foraging times of the captive deer (during daylight hours); 0500-0900" h and 1600-1900 h (pers. observ.). It was necessary to frequently expose deer to food d i s t r i b u t i o n s to assess the effect of learning. The use of two animals on a d a i l y basis was considered preferable to infrequent use of a larger sample. Data were recorded using a MORE behaviour recorder (Observational Systems, Redmond, WA): a real-time data logger enabling continuous recording of occurrence and duration of behaviours. The device was modified to permit two observers to simultaneously record information. One observer recorded the animal's behaviour while the second logged grid locations where subjects changed di r e c t i o n or ingested food. Standard meteorological data, including temperature, wind speed and d i r e c t i o n , and cloud cover, were recorded prior to each t r i a l . T r i a l s were conducted only during rain-free periods because moisture r a d i c a l l y altered the p a l a t a b i l i t y of the- pelleted rations. 18 S t a l l t r i a l s C a f eteria-style s t a l l t r i a l s were conducted in triangular-shaped enclosures (Fig. 2.1) similar to those normally used to house deer at the Research Forest. A door at the front of the pen allowed the observer to add food and leave the pen without missing observations of feeding behaviours. Three types of food were selected based on the feeding habits of captive deer at the Research Forest and t h e i r l i k e l i h o o d of re s u l t i n g in constant preference rankings: apples, which were considered p o t e n t i a l l y highly preferred food; pelleted 16% protein dairy ration (Buckerfields, Abbotsford, B.C.); and pe l l e t e d a l f a l f a (Buckerfields). A l l three foods were placed in 60 x 60 x 30-cm feeding buckets in the pen under ad libitum conditions (Fig. 2.1). Animals were fasted for 5-10 h prior to s t a l l t r i a l s . Weight of each food type was recorded before and after each t r i a l . A t r i a l began when food was placed in the pen. Locations of feeding and other behaviours were recorded throughout the 60-min t r i a l . Recorded behaviours included: b i t i n g - the actual intake of food; handling - masticating food while not ingesting; walking; r e s t i n g / l y i n g ; grooming; drinking; standing a l e r t - interruption of another behaviour and staring, s n i f f i n g , or l i s t e n i n g in a fixed d i r e c t i o n ; and voiding urine and faeces. Because the duration of a bite (that time the animal remained in contact with the feeding substrate) could be highly variable, I did not consider the number of bites to be the most 19 F i g u r e 2.1. Photograph showing ad l i b i t u m food p r e s e n t a t i o n during s t a l l t r i a l s . Animals were allowed access to apple p i e c e s , p e l l e t e d d a i r y r a t i o n , and p e l l e t e d a l f a l f a f o r 60 min. 20 appropriate measure for use in comparing food sele c t i o n . Therefore, time (in s) eating, each food type was used to compare food preference. The number of bi t e s , t o t a l duration of time b i t i n g , and weight eaten were recorded for each food type and expressed as proportions of the consumption of a l l foods. Preference for food types was based on rankings of test foods from highest to lowest. I also used one-way analyses of variance (Sokal and Rohlf 1981) to compare differences within animals among t r i a l s (hypothesis 1) and between animals (hypothesis 2). Comparisons were made using the proportion of weight, bit e s , or time b i t i n g , as well as average bite size and a c t i v i t y patterns. Thirteen s t a l l t r i a l s (seven with the male and six with the female) were completed, but problems with the data acqu i s i t i o n system resulted in four t r i a l s with the male and six with the female being analysed. Because weight consumed during a t r i a l was obtained by difference before and after the observation period, rain during t r i a l 2 wet the foods and made accurate estimation of weight consumed impossible. As a res u l t , preference estimates and discussions for the female are based on six t r i a l s for time measures and f i v e for weight. S t a l l t r i a l s were interspersed with pen t r i a l s to enable examination of temporal s h i f t s in preference. Pen t r i a l s A vegetation-free enclosure was needed to allow control of food d i s t r i b u t i o n and abundance. The 0.5-ha pen was marked off 21 into 5-m intervals enabling two observers, positioned in a tower -8 m above the ground (see F i g . 2.2), to estimate the animal's location to the nearest square-metre. An animal was fasted in i t s pens and moved into a 5 x 10-m holding area adjacent to the pen -1 h before a t r i a l . At other times each deer was isolated with ad libitum access to water, pelleted dairy ration, and forage. Naturally growing vegetation was removed from the enclosure by chemical d e f o l i a t i o n . Daily hand removal of weeds could not keep up with regrowth of vegetation and a reapplication of herbicides between tests with d i s t r i b u t i o n s 2 and 3 was required. No t r i a l s were conducted for =2 wk after chemical applications because of potential problems with chemical residues in the pen. Dead standing vegetation was cut and removed, and hand removal of sprouting weeds was car r i e d out d a i l y . These combined measures were not s u f f i c i e n t to eliminate a l l vegetation. Small amounts of weeds provided an alternate forage source, precluding comparison to s t a l l t r i a l s e lection. I a r b i t r a r i l y chose 10% of bites on weeds as an upper l i m i t and t r i a l s were excluded i f t h i s threshold was exceeded. Weed growth also eliminated several planned t r i a l s . A t o t a l of 1 44 30 x 30-cm platforms, supported =-30 cm above the ground, were positioned at 5-m intervals throughout the pen (Fig. 2.2). Food types placed on these platforms varied with the test d i s t r i b u t i o n . Platforms were divided into 36 groups of four to create 'patches' of food. Leaving one-fourth of the platforms without food provided areas with no available food. Nine 'patches' of apples, pelleted dairy 22 Figure 2.2. Photograph of pen t r i a l showing food platforms as seen from the observation tower. Platforms were spaced at 5-m in t e r v a l s and were used as a g r i d system for animal l o c a t i o n . 23 ration, pelleted a l f a l f a , or no food were randomly assigned locations in the pen. Two randomizations produced d i s t r i b u t i o n s 1 and 2 (Fig. 2.3). D i s t r i b u t i o n 3 was s p a t i a l l y i d e n t i c a l to d i s t r i b u t i o n 2 but contained only one piece of apple per .platform instead of four pieces. I chose 1/16 apple pieces (=*9 g) so that uniform apple size could be e a s i l y maintained. Food was set out immediately before each t r i a l , requiring = 1 h. F i f t y ml of pelleted dairy ration (=31 g) or pelleted a l f a l f a (=27 g), or four pieces of apple (each -9" g) were placed at each of the stations designated for food. T r i a l s were run at =0700 and =1700 h. Because feeding behaviour might have varied with time of day, I chose to use individuals at the same time each day. Male t r i a l s were conducted in the morning and female t r i a l s in the afternoon to eliminate variation between feeding periods within animals.. As a r e s u l t , any differences in behaviour between animals cannot be separated from diurnal e f f e c t s because of r e s t r i c t i o n s on sample s i z e . Food deprivation prior to t r i a l s varied from =2-10 h, but was usually >8 h. A rope from the holding area door to the tower was used to allow an animal to enter the enclosure. The door was closed once the animal had entered the experimental area, i n i t i a t i n g a t r i a l . Eight behaviours were recorded including walking, running, b i t i n g , chewing, alertness, grooming, rest i n g , and voiding. Walking, included a l l locomotory movements except running. T r i a l s were run for =2 h or terminated sooner for one of 24 a HOLDING PEN X OBSERVATION TOWER O o o o • o • D • • • • o O O o o o • D • • • • • • o o • • o o D D o O • • • • • • o o • O o O • • • • O o • • • • D • • • o o • • • • D o o o D D • • • D • • o o • • • • D D o o D D o o • • • • o o D • o o D D • • b X o o o o • • O O D • a D • • o o o o • • O o o • D • • • o o • • • • o o • • • • O O o o D D • • o o • • • • o O • • O O o o o o o o D a • • • • O O o o D • o o • o D o o • • • D • D o • • • • a • • • a D D o • • • • F i g u r e 2.3 Schematic p r e s e n t a t i o n of d i s t r i b u t i o n s 1 (a) and 2 (b) used d u r i n g 1981 pen t r i a l s . Apples are rep r e s e n t e d by •, p e l l e t e d d a i r y r a t i o n by O, and a l f a l f a p e l l e t s by • . Gaps i n the d i s t r i b u t i o n r e p r e s e n t f e e d i n g p l a t f o r m s that r e c e i v e d no food. D i s t r i b u t i o n 3 was s p a t i a l l y i d e n t i c a l to d i s t r i b u t i o n 2, but the amount of food at each apple p l a t f o r m was reduced by 75%. Platforms were 5 m a p a r t . 25 two reasons. If the animal lay down and rested, the foraging bout was considered completed. Twice the data recorder's memory f i l l e d , ending the t r i a l . By recording locations wherever an animal changed di r e c t i o n the complete search path and distance t r a v e l l e d could be determined (± =1 m). When an animal ate at a platform I recorded the location and number of b i t e s . Entry of both b i t i n g and chewing durations enabled c a l c u l a t i o n of time spent consuming each food type. I used time spent ingesting apples, pelleted dairy ration, and pelleted a l f a l f a to compare food selection. Observations were grouped to 5, 10, 20, and 90 min, to examine selection on a cumulative basis. Selection among t r i a l s and d i s t r i b u t i o n s was tested by analysis of variance on arcsin square root transformed proportions (Sokal and Rohlf 1981) of time spent eating a food divided by the t o t a l time eating for the above time i n t e r v a l s . The e f f e c t s of repeated exposure to a d i s t r i b u t i o n , as well as the frequency and duration of a l e r t behaviours on food selection were examined using linear regression and correlation techniques (MIDAS; Fox and Guire 1976). Variation in apple intake rates within t r i a l s was tested by examining the time spent clearing platforms of a l l apple pieces. Platforms that were cleared during 0-5, 5-10, and 10-90 min i n t e r v a l s were used. Possible variation within and among t r i a l s was tested (ANOVA) for both animals. For each animal, mean time required to clear platforms- of apples during 0-5, 5-10, and 10-90 min intervals was regressed on the number of times an animal had been exposed to a d i s t r i b u t i o n , to test 26 i f duration of bites varied with experience in a -distribution. Time eating apples per distance t r a v e l l e d (s/m) to 5, 10, 15, 20, and 90 min was calculated and regressed on experience within a d i s t r i b u t i o n to test hypothesis 5, that animals would become more e f f i c i e n t at obtaining a preferred food with experience in a d i s t r i b u t i o n . One-tailed t-tests (Sokal and Rohlf 1981) were used to determine i f the slopes were >0. Slopes represented the change in t o t a l time (s) consuming apples per m t r a v e l l e d with each exposure to a d i s t r i b u t i o n . Running bouts (play), during which animals ran around the pen, were excluded from distance t r a v e l l e d . Comparisons of apple consumption per distance t r a v e l l e d between the 0-5 and 0-90 min periods within a d i s t r i b u t i o n were made by analysis of covariance (SLTEST; Le 1984) for d i s t r i b u t i o n s in which both slopes were s i g n i f i c a n t l y >0. Because animals may have been using the same area of the pen or a l t e r i n g the area i n i t i a l l y searched based on previous experience, a technique was required to compare search paths among t r i a l s . The pen was divided into four s t r i p s , each containing two rows of platforms (Fig. 2.4), and the use of each s t r i p for the f i r s t 100 and 200 m t r a v e l l e d during the t r i a l was measured. Reconstructed search paths (Fig. 2.4) were subdivided into consecutive 5-m segments and the numbers of these segments that ended in each of the four s t r i p s were accumulated for the f i r s t 100 and 200 m of the search path. I used 5-m segments because of the distance between platforms. If no platforms were v i s i t e d , an animal crossing one of the four s t r i p s (Fig. 2.4) would count two segments when the s t r i p 27 F i g u r e 2 . 4 . Method of comparing s e a r c h p a t h s between t r i a l s and d i s t r i b u t i o n s . Apple p l a t f o r m s a r e r e p r e s e n t e d by •, p e l l e t e d d a i r y r a t i o n by O, and p e l l e t e d a l f a l f a by • . To s p a t i a l l y c h a r a c t e r i z e i n i t i a l s e a r c h i n g , the p o r t i o n of the pen c o n t a i n i n g food was p a r t i t i o n e d i n t o 4 s t r i p s , each of 2 i d e n t i c a l columns. The t r a n -s i t i o n from the s o l i d t o d o t t e d l i n e i n d i c a t e s the end of the f i r s t 100-m of the s e a r c h p a t h . D i v i d i n g the f i r s t 100 and 200 m of the p a t h i n t o the number of 5-m segments e n d i n g i n each s t r i p r e s u l t e d i n the c u m u l a t i v e f r e q u e n c y below the d i s t r i b u t i o n . The 200-m c a l c u l a t i o n i s not shown. 28 was crossed at right angles. Contingency table analyses (Sokal and Rohlf 1981) were used to examine, s h i f t s in patterns of use within and between search paths of d i s t r i b u t i o n s 1 and 2. The difference between d i s t r i b u t i o n 2 and 3 was in quantity of apple per platform, rather than s p a t i a l and so t h i s t r a n s i t i o n was not s i m i l a r l y examined. Influence of wind (direction and speed) and pen shape on the distance at which food types might be detected was examined by breaking the approaches to platforms (by food type) into N, S, E, and W components. Attempts were then made to relate s i g n i f i c a n t differences between the distance components as detected by t-tests to wind conditions and pen structure. Results S t a l l t r i a l s Food Preference: Hypotheses 1 and 2 Neither hypothesis 1 (consistent preference ranking) nor hypothesis 2 (consistency between animals) were rejected. Considering a l l t r i a l s , consumption of apples consistently ranked highest for both animals. Although both deer preferred apples under ad libitum conditions, extent of the preference varied between animals from t r i a l to t r i a l , and in some cases within t r i a l s . Table 2.1 compares the food preference for pelleted ration, apples, and pell e t e d a l f a l f a between animals, 29 based on e n t i r e 60-min t r i a l s . No s i g n i f i c a n t differences (P < 0.05) were observed between animals for the proportion of time eating or the proportion of t o t a l bites allocated to any food type (Table 2.1). Analyses of variance for the number of bites and the length of time eating each food type for each individual were a l l s i g n i f i c a n t l y d i f f e r e n t , indicating unequal use of food types. Using either duration or number of b i t e s , ranking of means showed a preference of: apples (highest), pelleted dairy ration, and a l f a l f a (lowest) for both animals. The male exhibited a s i g n i f i c a n t preference for apples over pelleted ration for duration (P = 0.049) and number of bites (P = 0.054) by Scheffe's test (Scheffe 1959). There was also a s i g n i f i c a n t difference (ANOVA) between a l l foods for the number of bites (P = 0.002) and the time spent eating apples and pelleted dairy ration (P = 0.063) for the female. Preference was also examined on a weight basis (Table 2.2) because differences in 'food size' and hardness of food or both might have biased calculations of intake from time b i t i n g . There were no differences (P > 0.05) when the proportion of each food eaten was compared between animals. Comparisons of weight consumed among foods showed highly s i g n i f i c a n t differences in selection (P < 0.001) for both animals. Based on Scheffe's t e s t , both animals preferred apples over pelleted dairy r a t i o n (P < 0.001); second ranked dairy ration was not consumed s i g n i f i c a n t l y more than a l f a l f a (P > 0.10). In addition to the o v e r a l l preference for apples described above, both animals showed a consistent preference for foods in Table 2.1. Comparison of food s e l e c t i o n between one male and female b l a c k - t a i l e d deer (Odocoileus hemionus coIumbianus) based on p r o p o r t i o n of b i t e s taken and time spent e a t i n g p e l l e t e d r a t i o n , a p p l e s , and p e l l e t e d a l f a l f a . S t a l l t r i a l s were conducted under ad l i b i t u m c o n d i t i o n s . The p r o p o r t i o n of each food type s e l e c t e d ( b i t e s or time) d i d not d i f f e r s i g n i f i c a n t l y between animals when the sin _',/x transformed p r o p o r t i o n s were compared. Subject T r i a l P r o p o r t i o n of b i t e s P r o p o r t i o n of time Rat i on App1es A l f a l f a Rat ion Apples A l f a l f a Male Female 1 0. . 324 0 . 459 0 .2 16 0. . 320 0 . 389 • 0 . 291 2 0. . 206 0 .651 0 . 143 0. . 344 0 .559 0 .097 3 0. . 333 0 . 542 0 . 125 0. .362- 0 .507 0 .131 4 0. . 143 0 .857 0 .000 0 .063 0 .938 0 .000 X 0. . 252 0 .627 0 .121 0 272 0 .598 0 . 130 (SD) (0. 093) (0. .172) (0. .090) (0. 141 ) (0. .237) (0 . 121 ) 1 0. 059 0. .941 0. .000 0. .031 0. ,969 0. .000 2 b. 426 0. .574 0 .000 0. 573 0. ,427 0. .000 3 0. 456 0. .491 0. .053 0. 374 0. ,591 0. .034 4 0. 425 0. . 575 0. .000 0. 300 0. . 700 0. .000 5 0. 348 0. 609 0. .043 0. 221 0. . 731 0. ,047 6 0. 239 0. 674 0. .087 0. 154 0. , 793 0. ,053 X 0. 326 0. 644 0. 031 0. 276 0. 702 0. 022 (SD) (0. 152) (0. 157) (0. 036) (0. 188) (0. 183) (0. 025) P 3 p > 0 . 5 0 n S p > 0 . 7 5 n S p > o . i o n s p > 0 . 7 5 n S p > 0 . 5 0 n S p > 0.05 P r o b a b i l i t y t h a t both animals a l l o c a t e the same p r o p o r t i o n of time or b i t e s to each food type as t e s t e d by a n a l y s i s of v a r i a n c e . 31 T a b l e 2.2. Comparison of proportions of weight consumed of pelleted ration, apples, and pelleted a l f a l f a between one male and female black-t a i l e d deer {Odocoi I eus hemi onus col umbi anus) during ad libitum s t a l l t r i a l s conducted to establish preference. Subject T r i a l Proportion of Weight Consumed Ration Apples A l f a l f a Male Female 1 0.320 0.531 0. 1 50 2 0.244 0.672 0.084 3 0.309 0.636 0.055 4 0. 104 0.896 0.000 X 0.244 0.684 0.072 (SD) (0.099) (0. 154) (0.062) 1 0.017 0.983 0.000 2 * * 3 0.352 0.640 0.006 4 0.217 0.783 0.000 5 0.036 0.954 0.010 6 0.063 0.923 0.015 X 0. 1 37 0.857 0.006 (SD) (0.144) (0.143) (0.006) Rain during t r i a l 2 caused an underestimation of. weight of pelleted food consumed. Ignoring rain, pelleted dairy ration s t i l l represented 87% of the t o t a l weight of food consumed. 32 each t r i a l (Table 2.3): the male spent more time eating apples throughout a l l t r i a l s ; apples ranked f i r s t in a l l but one t r i a l with the female. The male ingested s i g n i f i c a n t l y higher amounts of pelleted ration and pelleted a l f a l f a per bite and per s b i t i n g than did the female (from Table 2.4; P < 0.05). Apple intake rate (g/s) was similar for both animals (dictated largely by the uniform size of pieces). Intake rate by the female on pelle t e d dairy ration was smaller than on apples (P < 0.001). The male's intake rate was not affected by food type (P > 0.10). The female's smaller bite size when eating p e l l e t s explains in part the apparent preference for p e l l e t s based on time b i t i n g in t r i a l 2 and the start of t r i a l 3 when the o v e r a l l weight of apples consumed was greater. Within animals, food type did not a f f e c t bite length (P > 0.10). Apple bites were longer for the female than for the male (P < 0.05). Although apples were not the most consumed food during every minute of every t r i a l , departures from apple consumption ranking highest were s u f f i c i e n t l y rare that I did not reject either hypothesis 1 or 2. Factors influencing selection The proportion of time feeding or the amount of food consumed by either animal was not related (P > 0.10) to the number of s t a l l t r i a l s conducted. Food deprivation was not related to feeding intensity during any portion of t r i a l s with either animal (P > 0.10). Table 2.3. P r e f e r e n c e of two b l a c k - t a i l e d deer (Odocoi1 eus hemionus columbianus) f o r p e l l e t e d r a t i o n , a p p l e s , and p e l l e t e d a l f a l f a compared w i t h i n ad l i b i t u m s t a l l t r i a l s based on time spent b i t i n g each f o o d type to 5, 10, and 60 min. Time s i n c e P r o p o r t i o n of time b i t i n g S u b j e c t T r i a l t r i a l began (min) P e l l e t e d r a t i o n Apples P e l l e t e d A l f a l f a 1 5 ± 0.362 0. 594 0.043 10* 0. 320 0.389 0. 291 2 5 0. 286 0.653 0.061 10 0.317 0. 599 0.085 60 0.344 0. 559 0.097 3 5 • 0.074 0.889 0.037 10 0. 199 0.631 0.171 60 0.362 0. 507 0.131 4 5. 0.063 0.938 0.000 10* 0.063 0.938 0.000 1 5 0.041 0.959 0.000 10 0.036 0. 964 0.000 60 0.031 0.969 0.000 2 5 0.032 0.968 0.000 10 0.402 0. 598 o.ooo 60 0.573 0.427 0.000 3 5 0.746 0. 254 0.000 10 0.656 0. 344 0.000 60 0.374 0. 591 0.034 4 5 0.000 1 .000 0.000 10 0. 223 0. 777 0.000 60 0. 300 0. 700 0.000 5 5 0.067 0.933 0.000 10 0.117 0. 883 0.000 60 0. 221 0. 731 0.047 6 5 0. 122 0. 842 0.036 10 0.062 0.920 ,-0.018 60 0. 154 0. 793 '0.053 animal o n l y f e d f o r f i r s t 10 minutes of t r i a l . Co co Table 2.4. Comparison of weight, w e i g h t / b i t e , and weight/s b i t i n g between one male and one female b l a c k - t a i l e d deer (0. h. columbianus) d u r i n g ad l i b i t u m s t a l l t r i a l s . The d u r a t i o n of a b i t e was the l e n g t h of time an animal i n g e s t e d w h i l e i n c o n t a c t w i t h a p a r t i c u l a r f o o d type. Subj e c t T r i a l Weight (g) consumed Weight/bi t e (g) Weight/second b i t i n g ( g / s ) Rat i on Apples A l f a l f a Rat i on Apples A l f a l f a Rat i on Apples A l f a l f a Male 1 129.7 215 .5 60. 7 10.8 12 .7 7.6 1 . 3 1 .8 0.7 2 155 . 7 428 .7 53.6 12.0 10 . 5 6.0 2.4 4 . 1 3.0 3 198 . 7 408 .2 35 . 3 12.4 15 . 7 5.9 1 . 5 2 .2 0.7 4 34 .9 299 .6 0.0 17.5 25 .0 0.0 2 . 7 1 .5 0.0 X 129.8 338 .0 37.4 13.2 16 .0 4.9 2.0 2 . 4 1 . 1 (SD) (69.3) (99 .4) (27. 1) (3.0) (6 .4) (3.4) (0.7) (1. 2) (1.3) Female 1 7.5 433 . 7 0.0 3.8 13 .6 0.0 0.8 1 . ,4 0.0 2 385. . 7 * 12 . 4 * 0. 9 * 3 254 . 7 463 .6 4.6 9.8 16. .6 1 .5 0.8 1 . 0 0.2 4 142.8 514 . 4 0.0 8.4 22 . 4 0.0 1 . 2 1 . 9 0.0 5 24 .6 650. . 5 6.6 1 .0 15, . 5 2 . 2 0.3 2 . 0 0.3 6 42 .0 619. ,4 10.0 3.8 20. 0 2.5 0.9 2 , 4 0.6 X 94 .3 511. .2 4 . 2 5.4 16 , .8 1 . 2 0.8 1 . 6 0.2 (SD) (103.9) ( 105. 0) (4.3) (3.6) (3. 8) (1.2) (0.3) (0. 6) (0.2) A b s o r p t i o n of m o i s t u r e d u r i n g t r i a l p r e c l u d e d measurement of weight consumed d u r i n g t r i a l 35 Preference for apples was generally higher at the st a r t of the t r i a l s , but apples were not always eaten f i r s t (Fig. 2.5). Although samples were small and the design was unbalanced, the data were informative. The female's i n i t i a l food selection was determined by the position of feeding buckets and not by the type of food. Five of the six bites were taken from the.middle bucket (Fig. 2.1), the sixth occurring the only time a l f a l f a p e l l e t s were in the middle. A l f a l f a was avoided, but otherwise the middle food was taken. Both non-centre, f i r s t bites by the male also occurred when a l f a l f a was in the middle. P e l l e t s were favoured over apples for the f i r s t b i t e , perhaps because deer were maintained on pelleted dairy ration between t r i a l s . There was no difference (P > 0.10; male: n = 4; female: n = 6) between the proportion of apples taken during the entire t r i a l when f i r s t bites of apples and p e l l e t s were compared. Data from the s t a l l t r i a l s supported a consistent preference ranking of apples > pelleted dairy ration > pelleted a l f a l f a for comparison with the pen t r i a l experiments. Pen t r i a l s Thirty-nine pen t r i a l s were complete; 22 with the male and 17 with the female. Consumption of weeds and problems with the behaviour recorder resulted in a number of t r i a l s not being analysed (Table 2.5). 36 F i g u r e 2.5. Comparison of f i r s t b i t e d u r i n g 1981 s t a l l t r i a l s r e l a t i v e to food type and l o c a t i o n f o r two b l a c k - t a i l e d deer. For both animals, a l l non-centre b i t e s o c c u r r e d when a l f a l f a p e l l e t s were p l a c e d i n the c e n t r e . P o s i t i o n , and not p r e f e r r e d food type, appeared to i n f l u e n c e f i r s t b i t e c h o i c e . 37 Table 2.5. Summary of sample sizes for pen t r i a l s in 1981, by subject and d i s t r i b u t i o n . Completed t r i a l s were not usable i f either problems with the data recorder precluded data r e t r i e v a l (n = 1) or i f more than 10% of bites were of other than experimental foods. Dis t r i b u t i o n Subject Male Female # Completed # Usable # Completed # Usable 1 12 11 a 6 5 a' b 2 7 7 8 6 a 3 3 3 3 2 C Total T r i a l s 22 21 17 13 a T r i a l s discarded due to consumption of weeds; >10% of bites on non-experimental foods. k For comparing i n i t i a l search areas between d i s t r i b u t i o n s 1 and 2, the 6th t r i a l was included. c T r i a l lost due to f a i l u r e of behavioural recorder. 38 Effect of searching on selection: Hypothesis 3 Although there were differences in proportions of the food types selected between s t a l l and pen t r i a l s , the ranking of apples (highest), pelleted dairy ration, and pelleted a l f a l f a (lowest) persisted when apples were abundant in the enclosure. Hypothesis 3, regarding s p a t i a l effects on selection (page 15), was not rejected. For d i s t r i b u t i o n s 1 and 2, food selection by the male and female d i f f e r e d s i g n i f i c a n t l y to 5, 10, 15, 20, and 90 min into the t r i a l (ANOVA; for a l l P < 0.01). The animals were therefore considered separately. The female selected apples more than other foods during a l l time inter v a l s within d i s t r i b u t i o n s 1 and 2 (Table 2.6). While she was exposed to d i s t r i b u t i o n 1, she ate exclusively apples for the f i r s t 20 min of each t r i a l . For the two t r i a l s with d i s t r i b u t i o n 3 (which represented a 75% reduction in apple pieces per platform), apples were i n i t i a l l y selected more than other foods and were, consumed for more of the. time; (not always-s i g n i f i c a n t l y so), during the f i r s t 20 min (Table 2.6). D i s t r i b u t i o n did not affect the time the female spent eating apples (P = 0.259) during the f i r s t 5 min of t r i a l s . By 10 min, and through a l l subsequent intervals (0-15, 0-20, and 0-90 min), Scheffe's test showed time spent eating apples in d i s t r i b u t i o n 3 to be s i g n i f i c a n t l y lower (for a l l periods P < 0.001). No difference existed between d i s t r i b u t i o n s 1 and 2 (P > 0.15 for a l l contrasts). The male was considerably less consistent in food T a b l e 2.6. Food selection by the female black-tailed deer during 1981 pen t r i a l s based on time spent eating food types. Only use of dairy ration and apples is presented since a l f a l f a p e l l e t s were rarely consumed. Time Period (min) 0-5 0-10 0-15 0-20 0-90 D i s t r i b u t i o n Experience Time Bi t i n g (s) Time Biting (s) Time Bi t i n g (s) Time B i t i n g (s) Time B i t i n g (s) Apples Ration Apples Ration Apples Ration Apples Ration Apples Ration 1 1 15 0 32 0 32 0 37 0 221 89 2 50 0 122 0 157 0 157 0 339 4 3 56 0 60 0 78 0 120 0 278 18 4 93 0 140 0 183 0 190 0 282 12 5 58 0 115 0 179 0 216 0 273 32 2 1 49 0 102 0 131 0 163 0 271 63 2 , 50 17 103 17 129 17 139 17 281 65 3 70 15 120 15 149 15 176 22 181 50 4 61 0 86 0 152 0 159 0 203 1 5 79 0 121 0 138 0 138 0 208 1 6 112 0 316 0 410 0 424 0 564 8 3 2 a 29 2 29 25 45 31 45 40 59 99 3 65 2 85 11 85 26 85 58 89 344 a t r i a l 1 was completed with d i s t r i b u t i o n 3 but data were i r r e t r i e v a b l e . 40 selection than was the female (Table 2.7). Although he spent more time eating apples in d i s t r i b u t i o n s 1 and 2, he tended to feed on pelleted dairy ration during the f i r s t 5 min of two t r i a l s with d i s t r i b u t i o n 1. With three exceptions ( t r i a l s 2, 4, and 10 with d i s t r i b u t i o n 1), time spent eating apples was greatest by 15 min, with a tendency to increase the proportion of time spent eating pelleted dairy ration after 20 min in some t r i a l s . He shifted to pelleted dairy ration during the 15-20 min i n t e r v a l with a l l t r i a l s of d i s t r i b u t i o n 3 (see Table 2.7). The male's selection of apples to 5 min was not affected by d i s t r i b u t i o n ( t r i a l 1 with d i s t r i b u t i o n 1 was excluded because feeding did not occur u n t i l =*20 min into the t r i a l ) . Unlike the female, there were no differences between the time eating apples in d i s t r i b u t i o n s 1 and 3 (P > 0.10), possibly due to the male adapting more slowly to the pen s i t u a t i o n at the start of the experiments. To 10 and 15 min, Scheffe's tests indicated that his selection of apples was greater (P < 0.10) during d i s t r i b u t i o n 2 than for 1 and 3 with no difference between the l a t t e r two d i s t r i b u t i o n s . There was no difference between the selection of apples in d i s t r i b u t i o n s 1 and 2 by 20 min. Selection of apples, however, was s i g n i f i c a n t l y lower with d i s t r i b u t i o n 3 (P < 0.001). For both animals, food selection throughout t r i a l s with d i s t r i b u t i o n s 1 and 2 did not d i f f e r from ad libitum preference when apples were abundant in the pen. Exceptions were restricted, to: t r i a l s in. which food, consumption was low or towards the end of t r i a l s (with the male) when apple abundance was reduced. I did not consider these data s u f f i c i e n t to Table 2.7. S e l e c t i o n of food types by the male b l a c k - t a i l e d deer d u r i n g 1981 pen t r i a l s based on time e a t i n g . Only use of d a i r y r a t i o n and app l e s i s p r e s e n t e d s i n c e a l f a l f a p e l l e t s were r a r e l y consumed. Time P e r i o d (min) 0-5 0-10 0-15 0-20 0-90 D i s t r i b u t i o n E x p e r i e n c e Time B i t i n g ( s ) Time B i t i n g ( s ) Time B i t i n g ( s ) Time B i t i n g ( s ) Time B i t i n g ( s ) Apples R a t i o n Apples R a t i o n Apples R a t i o n Apples R a t i o n Apples R a t i o n 1 0 0 0 0 0 0 0 18 87 76 2 a 20 32 24 32 29 32 32 32 89 87 4 3 22 9 26 9 26 1 1 27 1 1 54 99 5 33 19 75 19 77 21 78 21 151 123 6 22 10 22 10 64 1 1 64 12 205 303 7 37 5 86 5 86 5 98 7 180 30 8 35 4 47 4 65 9 78 9 99 72 9 5 2 25 3 51 7 64 7 153 71 10 31 0 31 27 58 50 82 51 184 123 1 1 32 47 88 51 136 51 136 51 308 143 12 55 8 101 12 103 40 103 40 193 122 1 2 2 b 6 37 8 66 9 75 9 179 236 2 30 14 90 14 108 19 108 65 155 92 3 35 0 64 1 84 3 84 3 182 42 4 54 6 1 19 6 130 6 143 1 1 162 38 5 34 6 73 6 88 6 1 12 6 156 158 6 45 0 55 1 55 1 55 1 159 22 7 35 1 54 1 81 1 92 9 184 175 1 26 12 38 14 69 100 71 132 91 354 2 43 21 59 21 70 21 80 164 102 507 3 22 16 44 77 64 77 73 1 1 1 79 283 t r i a l 3 was completed w i t h d i s t r i b u t i o n 1 but data were i r r e t r i e v a b l e , a l f a l f a e a t e n f o r 28 s. 42 reject hypothesis 3. Selection approximated preference when food abundance was unlimited. Eff e c t of food abundance on food s e l e c t i o n : Hypothesis 4 Hypothesis 4 was rejected: food selection was not altered by low i n i t i a l abundance of preferred foods over the range tested. Increased consumption of lower-ranked foods did not occur u n t i l apples were no longer av a i l a b l e . The number of apples eaten by both animals as a proportion of the t o t a l number in the pen (144 pieces in d i s t r i b u t i o n s 1 and 2 and 36 pieces in d i s t r i b u t i o n 3) was always greatest in d i s t r i b u t i o n 3 (Table 2.8; a l l P < 0.05) during a l l time i n t e r v a l s . Both animals v i s i t e d more apple platforms during the f i r s t 15 min of t r i a l s with d i s t r i b u t i o n 3 than with d i s t r i b u t i o n s 1 and 2. For example, during the f i r s t 5 min of t r i a l s , the mean proportion of apples consumed (Table 2.8) indicate that, on average, ^7, ^10, ^22 platforms were v i s i t e d by the female and =*4,, =*6, and =*12 by the male for d i s t r i b u t i o n s 1, 2 r and: 3, respectively. The data did not suggest that the lower selection of apples after 15 min in d i s t r i b u t i o n 3 was due to a switching from the preferred s t a l l food, but rather to a depletion of apples in the available food supply. From 15 min on, the proportion of apples consumed was lower during d i s t r i b u t i o n 1 than d i s t r i b u t i o n 2 for both animals (Table 2.8). This result may have been caused by low consumption of apples at the st a r t of the experiments. When fewer apples were encountered on platforms, the deer Table 2.8. Comparison of the mean proportions of available apples consumed per number of apple pieces in the pen by d i s t r i b u t i o n for both animals. Means are presented as the measured proportions; ANOVAs were carr i e d out after sin"Vx transformations on the data. Underlined means were not s i g n i f i c a n t l y d i f f e r e n t by Scheffe's test with an a = 0.10. Dis t r i b u t i o n Subject Time Interval (min) 1 2 3 remale 0- -5 0 . ,20 0 . ,27 0 . ,60 0- -10 0 . ,33 0 . ,47 0 . , 69 0- -15 0 . ,42 0. ,60 0. .81 0- -20 0 . ,47 0 . ,66 0 . , 83 o- -90 0 . , 85 0 . .94 0. . 96 Male o- -5 0 . . 12 0, . 17 0, . 3 2 o- -10 0, . 19 0, . 3 5 0, .51 o- -15 0. . 23 0, . 43 0, . 6 5 o- -20 0, . 26 0, . 47 0, . 7 6 o- -90 0, . 48 0, . 7 6 0, . 9 3 44 did not respond by eating more lower-ranked food. Instead, they continued searching u n t i l apples were no longer located, and then switched to less preferred foods. Their selection was altered by apple a v a i l a b i l i t y and not by i n i t i a l apple density. Consequently, I rejected hypothesis 4. The number of times the male was exposed to d i s t r i b u t i o n 1 affected the proportion of apples in the pen that were eaten ( a l l P < 0.05) during the 0-15, 0-20, and 0-90-min i n t e r v a l s . This was determined by regressing the mean proportion of apples eaten (Table 2.8) on the t r i a l r e p licate number ( i . e . experience within a d i s t r i b u t i o n ) . There was no effect of experience on the proportion of apples eaten during any t r i a l s with the female, d i s t r i b u t i o n s 2 and 3 with the male, or to 10 min during t r i a l s with d i s t r i b u t i o n 2 with the male ( a l l P > 0.10). Effe c t of memory on food selection: Hypotheses 5 and 6 As noted, experience within-: a d i s t r i b u t i o n was not s i g n i f i c a n t l y related to the proportion of available apples consumed for a l l cases with the female and. most with the male. Improvements in searching e f f i c i e n c y , however, could be masked in that analysis by variations in foraging i n t e n s i t y . Using time spent eating apples per m t r a v e l l e d (s/m) as a measure of ef f i c i e n c y of exploiting a d i s t r i b u t i o n , I did not reject hypothesis 5 (the influence of experience on e f f i c i e n c y , page 15). Figure 2.6 presents changes in the e f f i c i e n c y of locating 45 10-.E IA D LU -J LU > O u -o z H < LU LU 0.0 o O 0 O DISTRIBUTION 1 DISTRIBUTION 2 DISTRIBUTION 3 EXPERIENCE WITHIN A DISTRIBUTION F i g u r e 2 . 6 . The e f f e c t o f e x p e r i e n c e w i t h i n a d i s t r i b u t i o n on t h e e f f i c i e n c y o f f i n d i n g a p p l e s e a r l y i n t r i a l s . E a c h d i s t r i b u t i o n i s r e p r e s e n t e d by a d i f f e r e n t p o i n t t y p e . The d a t a shown a r e f o r t h e f e m a l e t o 5 min i n t o e a c h t r i a l . D a t a were not a v a i l a b l e f o r t r i a l 6 w i t h d i s t r i b u t i o n 1, t r i a l s 7 and 8 w i t h d i s t r i b u t i o n 2, and t r i a l 1 w i t h d i s t r i b u t i o n 3. 46 apples with repeated exposure to the same d i s t r i b u t i o n f o r the female to 5 min i n t o t r i a l s . The female consumed more apples per d i s t a n c e t r a v e l l e d with the number of times she was exposed to a d i s t r i b u t i o n . When the f i r s t 5 min of t r i a l s were compared, both animals showed s i g n i f i c a n t i n c r e a s e s i n t h e i r e f f i c i e n c y of f i n d i n g apples with repeated exposure to the same d i s t r i b u t i o n (Table 2.9). The r a t e of i n c r e a s e i n t h i s e f f i c i e n c y with experience in a d i s t r i b u t i o n tended to decrease when longer p e r i o d s of t r i a l s were e v a l u a t e d (Table 2.9)'. Animals improved the r a t e at which apples were l o c a t e d s i g n i f i c a n t l y f a s t e r d u r i n g the f i r s t 5 min of t r i a l s than d u r i n g e n t i r e t r i a l s . During three of the d i s t r i b u t i o n - s u b j e c t combinations (Table 2.9: both with the female and one with the male), t h e r e was no i n c r e a s e i n the e f f i c i e n c y of f i n d i n g apples w i t h i n d i s t r i b u t i o n s by 90 min. During the f i r s t exposure of the male to d i s t r i b u t i o n 1, i n which there was i n c r e a s e d e f f i c i e n c y of f i n d i n g apples by the end of the t r i a l s , the in c r e a s e i n e f f i c i e n c y was s i g n i f i c a n t l y l e s s by a n a l y s i s of co v a r i a n c e (P = 0.021). A d d i t i o n a l l y , the female i n c r e a s e d her e f f i c i e n c y w i t h repeated exposure f a s t e r than d i d the male (Table 2.9). The l e n g t h of time r e q u i r e d t o c l e a r p l a t f o r m s of a l l four apple p i e c e s was examined to t e s t f o r d i f f e r e n c e s i n consumption r a t e s between t r i a l s . C o n s i d e r i n g only p l a t f o r m s that were c l e a r e d from 0-5, 5-10, or 10-90 min, ANOVA i n d i c a t e d that there were no, d i f f e r e n c e s i n the time to c l e a r a p l a t f o r m of apples by time p e r i o d w i t h i n any t r i a l f o r e i t h e r animal. P o o l i n g t r i a l s by s u b j e c t , there were no d i f f e r e n c e s between Table 2.9. The i n f l u e n c e of c o n s e c u t i v e exposures to the same d i s t r i b u t i o n f o r one male and one female b l a c k - t a i l e d deer (0. h. columbianus). R e g r e s s i o n s l o p e s i n d i c a t e the improvement w i t h each t r i a l (over n t r i a l s ) i n the s of apple consumption per m t r a v e l l e d over the s p e c i f i e d time i n t e r v a l . The number of times an animal had been exposed to a d i s t r i b u t i o n was used as the independent v a r i a b l e . Data were not a v a i l a b l e f o r t r i a l 4 ( d i s t r i b u t i o n 1) f o r the male, and t r i a l s G ( d i s t r i b u t i o n 1), 7 and 8 ( d i s t r i b u t i o n 2) f o r the female. Animal Time P e r i o d (min) s 1 ope D i s t r i b u t i o n 1 p(P>or n s 1 ope D i s t r i but i on 2 P(p>0) n Male 0-5 0-10 0-15 0-20 0-90 0.027 0.637 0.029 0.644 0.026 0.584 0.023 0.746 0.008 0.453 0.003** 0.003** 0.006** 0.001 * * 0.023* 0.043 0.025 0.022 0. 003 0.007 0.566 0. 349 0.473 0.018 0.315 0.051* 0. 162 0.088* 0. 773 0. 190 Female 0-5 0-10 0-15 0-20 0-90 0. 145 0.736 0.117 0.323 0.117 0.561 0.090 0.410 0.002 0.006 0.063* 0. 258 0. 145 0. 245 0.898 0.098 0. 103 0. 100 0.047 0.022 0. 994 0. 731 0. 775 0. 595 0. 501 0.000** 0.030* 0.021* 0.072* 0.116 t e s t i n g p f o r a p o s i t i v e s l o p e ( o n e - t a i l e d t e s t ) * P < 0.100 ** P < 0.020 48 time periods for the female (P = 0.497) or the male (P = 0.064). Overall mean times to c l e a r apple platforms were 9.22 ± 7.23 s (± SD) and 8.66 ± 2.57 s for female and male, respectively. There were no relationships ( a l l P > 0.10) when mean clearance time of apples to 5, 10, and 90 min was regressed on experience within a d i s t r i b u t i o n for either animal. Consequently, increases in the e f f i c i e n c y of locating apples could not be explained by variation in the rate at which apples were consumed at platforms. Variation in rates of clearing platforms were also unrelated to differences in the duration of food deprivation (P > 0.10). Early e f f i c i e n c y was a product of finding more apple platforms. Searching e f f i c i e n c y , rather than feeding intensity, explained changes in the rate at which apple platforms were located. To evaluate how e f f i c i e n t the animals became, minimum distance required to locate a l l of the apple platforms was computed for both d i s t r i b u t i o n s 1 and 2 (see F i g . 2.7). Taking mean clearance times of 9.22 s and 8.66 s per platform for the female and male and multiplying by 36 platforms, the minimum distances y i e l d 1.39 and 1.31 s of apples consumed per m t r a v e l l e d for the female and male, respectively, in d i s t r i b u t i o n 1 and 1.37 and 1.29 s apples per m t r a v e l l e d for d i s t r i b u t i o n 2. These numbers are maximum intake rates that could be obtained by an omniscient forager throughout the distribution.. There was not only a s i g n i f i c a n t increase in e f f i c i e n c y of eating apples within a d i s t r i b u t i o n (see F i g . 2.6), but also a 49 F i g u r e 2.7. Minimum d i s t a n c e r e q u i r e d to l o c a t e a l l apple p l a t f o r m s f o r d i s t r i b u t i o n s 1 and 2. Apples are r e p r e s e n t e d by p e l l e t e d d a i r y r a t i o n by o, and p e l l e t e d a l f a l f a by • . 50 reduction in e f f i c i e n c y when the d i s t r i b u t i o n was changed from 1 to 2. As a re s u l t , I did not reject hypothesis 6: because animals were using previous experience with a d i s t r i b u t i o n to find preferred food, exploitation e f f i c i e n c y declined in t r i a l s in which a d i f f e r e n t d i s t r i b u t i o n was encountered. Search Paths The search paths used by the animals for the f i r s t 100 and 200 m of t r i a l s were influenced by previous experience with food d i s t r i b u t i o n . The results of applying the technique outlined in Figure 2.4 to the three t r i a l s immediately before and a f t e r the change from d i s t r i b u t i o n 1 to 2 are presented in Table 2.10. T r i a l s used to i l l u s t r a t e the frequency technique in Figure 2.4 were reconstructions of one of the l a s t t r i a l s with d i s t r i b u t i o n 1 ( t r i a l 16) and the f i r s t t r i a l with d i s t r i b u t i o n 2 ( t r i a l 20). Although d i s t r i b u t i o n of apples changed, the animal followed b a s i c a l l y the same search path, missing several clumps, of apples by 5 m. Table- 2.11 presents., the results of contingency table analyses on the frequency data in Table 2.10. The search path of the f i r s t female t r i a l with d i s t r i b u t i o n 2 more closely resembled searching in d i s t r i b u t i o n 1 both to 100 and 200 m than subsequent t r i a l s with d i s t r i b u t i o n 2. Although the male's search path did not d i f f e r s i g n i f i c a n t l y between t r a n s i t i o n a l t r i a l s (Table 2.11), i t varied much more at the start of t r i a l s . As a re s u l t , to 100 m, the male's search path used immediately after the t r a n s i t i o n Table 2.10. Frequency of end p o i n t s of 5-m segments p r e s e n t e d by d i s t r i b u t i o n and animal to 200 m i n t o the t r i a l . T e s t s of s i g n i f i c a n c e a r e p r e s e n t e d i n T a b l e 2.13 and c a l c u l a t i o n of f r e q u e n c i e s i n F i g u r e 2.4. Fema1e Male D i s t r i b u t i o n T r i a l # Occurrence in columns T r i a l H . Occurrence i n columns 1-2 3-4 5-6 7-8 1-2 3-4 5-6 7-8 1 4 7 6 18 5 10 0 9 16 10 5 '4 8 19 3 1 1 5 10 15 5 6 3 4 8 19 8 12 2 7 16 12 2 1 3 10 18 4 1 2 2 16 14 2 12 15 8 4 2 2 12 1 1 8 3 16 1 1 3 0 3 . 10 13 9 8 a l t h o u g h t h i s t r i a l c o n t a i n e d >10% use of weeds, these d a t a were i n c l u d e d because of the importance of the t r a n s i t i o n a l t r i a l . Table 2.11. Comparison of the s e a r c h paths used d u r i n g two d i s t r i b u t i o n s f o r the male and female b l a c k - t a i l e d deer (0. h. columbianus). Data p r e s e n t e d are based on the number of 5-m segments t h a t ended i n p a r t i c u l a r columns of the d i s t r i b u t i o n to 100 and 200 m t r a v e l l e d ( F i g . 2.4). D i s t r i b u t i o n 1 D i s t r i b u t i o n 2 Female T r i a l number 4 5 G 1 2 3 H o : A l l segments equal (t o 100m / to 200m) P=0.000** / P=0.000** 1 i a H o : Equal w i t h i n d i s t r i b u t i o n I 1 I P=0.519 / P=0.808 P=0.001** / P=0.000** H o : T r a n s i t i o n a l t r i a l s equal L P=0.060 / P=0.107 H o : F i r s t t r i a l w i t h d i s t r i b u t i o n 2 equal to d i s t r i b u t i o n 1 P=0.600 / P=0.885 P=0.292 / P=0.729 Male T r i a l number 10 1 1 1 2 1 2 3 H o : A l l segments equal I I ( t o 100m / to 200m) P=0.007** / P=0.014** H o : Equal w i t h i n d i s t r i b u t i o n H o : T r a n s i t i o n a l t r i a l s equal H o : F i r s t t r i a l w i t h d i s t r i b u t i o n j I I P=0.007** / P=0.297 P=0.992 / P=0.004** I I P=0.164 / P=0.496 2 equal to d i s t r i b u t i o n 1 P=0.004** / P=0.126 P=0.668 / P=0.204 3 P of Ho 53 from d i s t r i b u t i o n 1 to d i s t r i b u t i o n 2 did not d i f f e r s i g n i f i c a n t l y f.rom subsequent t r i a l s . To 200 m, a similar r e l a t i o n existed for the female. It i s interesting to compare the search path used by the female at the end of d i s t r i b u t i o n 1 t r i a l s (Fig. 2.4, t r i a l 16) and the 'optimal' search path (Fig. 2.7, d i s t r i b u t i o n 1). The animals had learned to exploit the closest part of this d i s t r i b u t i o n to the release point. Use of a highly similar search path at the start of d i s t r i b u t i o n 2 resulted in lower apple consumption as shown in Figure 2.4. Detection distance From how far away did deer detect food? Average straight l i n e approach distances to bites was =5 m for a l l foods; largely an a r t i f a c t of the inter-platform distance ( i . e . the high frequency of 5 m approaches). There was, however, no way of knowing i f the animal 'knew' the contents of a platform when i t f,irst started towards i t or i f i t determined the contents while approaching. Distances at which platforms were rejected or missed, provided more accurate information. During early portions of t r i a l s , when deer were c l e a r l y searching for apples, distances at which apples were not 'detected' (when a similar distance in another d i r e c t i o n did not y i e l d a piece of apple) and distances at which platforms were refused were examined. Several examples can be found in Figure 2.4. At the start of t r i a l 16, only apple pieces were eaten. However, after v i s i t i n g the second pell e t e d dairy 54 ration platform (Fig. 2.4), the female turned l e f t to another p e l l e t e d dairy ration platform, missing an apple -5 m ahead. After proceeding a l l the way to the pelleted dairy ration platform, a right turn yielded an apple but the animal turned l e f t missing an apple 5 m ahead and not rejecting the pelleted dairy ration platform u n t i l = 1 m away. There were other t r i a l s in which animals changed di r e c t i o n towards apple platforms at =5 m. Apples were not always detected at 5 m and the distance may have been shorter. I concluded that 5 m was a conservative estimate of the detection distance under these conditions. Other effects on selection There were no s i g n i f i c a n t c o r r e l a t i o n s between either the duration or frequency of al e r t behaviours (an indication of dis t r a c t i o n s ) and selection of apples for both animals within d i s t r i b u t i o n s or when a l l t r i a l s were pooled by animal (for a l l P > 0.10; male: n = 21; female: n = 13). Although wind di r e c t i o n and speed,, possibly in combination with the pen configuration, may have had some effect on searching by each subject, none was detected. S t r a i g h t - l i n e approach distances to bites of a l l foods were broken into N, S, E, and W components. Sig n i f i c a n t differences between N-S and E-W components or both occurred during 15 out of 21 t r i a l s with the male, and 10 out of 13 t r i a l s with the female. These differences were not related to wind d i r e c t i o n or wind speed and persisted even without wind ( a l l P > 0.10). Although these biases were clear within t r i a l s , neither animal exhibited a 55 consistent d i r e c t i o n a l bias among t r i a l s . Discussion S t a l l T r i a l s Analysis of the time b i t i n g and weight of food consumed indicated preferences for apples (highest), p e l l e t e d dairy ration, and pelleted a l f a l f a (lowest). This ranking was consistent between animals and did not result in rejection of hypotheses 1 (consistent selection under ad libitum conditions) or 2 (consistency between animals). To examine e f f e c t s of sp a t i a l components on selection, a constant preference for a single food would have been desirable. Although deer c l e a r l y preferred apples in s t a l l t r i a l s , t h i s was not the case in every t r i a l . V ariation in selection of apples among t r i a l s was partly due to the measurements that I used as surrogates of se l e c t i o n . Use, of. in.s-tantaneous weights: of f.ood- consumed might- have reduced t h i s v a r i a t i o n , but weight of foods consumed were only obtainable for entire t r i a l s . Time b i t i n g foods provided a more accurate estimate of food selection than did number of b i t e s . The proportion of time b i t i n g each food type provided a good prediction of preference for apples (Table 2.3) in a l l t r i a l s in which apples were consumed greatest by weight (Table 2.2). Time b i t i n g also predicted the female's preference for pelle t e d dairy ration during t r i a l 2. Although accurate weights of food consumed 5 6 were not available for t h i s t r i a l (rain caused an increase of 5 % in unused a l f a l f a ) , an underestimate of the g of pelleted dairy ration consumed indicated a preference for ration on a weight basis. This result was not indicated when the proportion of bites were considered. Duration of bites was tremendously variable. Bite size, in the more t r a d i t i o n a l sense, can vary greatly (Trudell and White 1981, Wickstrom et a l . 1984). In this study differences in the available size of food 'pieces' (^ 9 g for apples and =*0.2 g for p e l l e t s ) , combined with differences in the hardness of foods, underestimated intake of the larger food items. Larger 'bites' require longer chewing time. A combination of bi t i n g and chewing time would better estimate intake and would be required i f differences in intake within t r i a l s and among foods were to be examined (Chapter 3). Differences in preference for apples among t r i a l s may have resulted from variation in apple q u a l i t y . For these t r i a l s , c u l l apples were obtained from l o c a l produce r e t a i l e r s . Although spoiled apples were never used, there was variation in the firmness and moisture content among apples, and as a res u l t , between some t r i a l s . This variation may have affected apple p a l a t a b i l i t y among and within t r i a l s . Results of s t a l l t r i a l s indicate that apples > pelleted dairy ration > pelleted a l f a l f a was the preference ranking of experimental foods (hypothesis 1), for both animals (hypothesis 2). Although there was considerable variation in the degree of th i s preference from t r i a l to t r i a l , and even though f i r s t bites were not exclusively apples (Fig. 2 . 5 ) , i n i t i a l selection 57 of apples was high. Apples were predicted to be strongly selected during, the i n i t i a l portion of pen tr i a l s , . Unless the s p a t i a l component of the pen t r i a l s (or the fact that they were not ad libitum) altered selection, o v e r a l l ranking of food selection in pen t r i a l s should have been apples > pelleted dairy ration > pelleted a l f a l f a . Pen T r i a l s Selection of apples in pen t r i a l s was altered from ad libitum preference for both animals. The female consumed more apples during pen t r i a l s although apple consumption by the male declined. When apples were abundant in the pen, there was no difference between preference rankings during s t a l l t r i a l s and selection during pen t r i a l s . Increased selection of apples by the female during pen t r i a l s could not be explained by temporal s h i f t s in preference because s t a l l t r i a l s were interspersed with pen t r i a l s . Satiation with apples in s t a l l t r i a l s might;* account: for greater, consumption of apples during pen t r i a l s . More than 1 kg of apples was available in s t a l l t r i a l s with no di s t r a c t i o n s or movements required between bites, a situ a t i o n not replicated in pen t r i a l s . Pen t r i a l s lasted longer than s t a l l t r i a l s but thi s difference did not explain differences in consumption of apples. Considering only the f i r s t 10 min of pen t r i a l s , =610 and =855 g of apples were consumed on average in d i s t r i b u t i o n s 1 and 2 respectively (based on average proportions of availa b l e 58 apples eaten [Table 2.8] and =^ 1300 g of apples in the pen). The maximum weight of apples consumed during s t a l l t r i a l s , was ^650 g (Table 2.4). In contrast, the male's selection of apples was lower during i n i t i a l pen t r i a l s than in s t a l l t r i a l s . Although he preferred apples in a l l s t a l l t r i a l s (Table 2.3), the male spent more time eating pelleted dairy ration than apples during the second t r i a l with d i s t r i b u t i o n 1 (Table 2.7). Consumption of pelleted dairy ration at the f i r s t one or two platforms encountered, during these pen t r i a l s , resulted in proportionately lower use of apples. Addition of a s p a t i a l component to food abundance did not greatly a l t e r food selection by the deer when the preferred food was abundant; hypothesis 3 (page 15) was not rejected. Preference ranking (based on time biting) for the female was not affected by searching when apples were abundant and apples ranked highest throughout pen t r i a l s with d i s t r i b u t i o n s 1 and 2. Food selection by the male was more variable, but the same conclusion was reached concerning hypothesis 3. The male demonstrated strong selection for apples during the f i r s t 20 min of t r i a l s during d i s t r i b u t i o n 2. As a result of t h i s selection, apple a v a i l a b i l i t y was reduced, res u l t i n g in high use of pelleted dairy ration l a t e r in t r i a l s . When apples were available, selection strongly resembled ad libitum s t a l l preference. Apples- were' not' as strongly selected by the male- in d i s t r i b u t i o n 1. During the f i r s t few t r i a l s with d i s t r i b u t i o n 1, the male spent time exploring the pen, in ef f e c t learning to 59 'play the game'. Selection of apples during these t r i a l s was reduced. However, there was only one other t r i a l (Table 2.7: t r i a l 11) in which apples did not rank f i r s t when they were abundant. As a result of combined data from d i s t r i b u t i o n s 1 and 2, I did not reject hypothesis 3 for either animal. Optimal foraging theory predicts that the animal's preference ranking corresponds to food p r o f i t a b i l i t y (MacArthur and Pianka 1966, Pyke et a l . 1977), generally measured in terms of energy intake while foraging. Because preference was based on ad libitum feeding and not a p r i o r i predictions of food p r o f i t a b i l i t y , a threshold for including lower-ranked foods in the accepted diet could not be made. My data showed that food selection did not d i f f e r from ad libitum preference u n t i l apples were almost unavailable; hypothesis 4 was rejected. Selection of lower-ranked foods did increase in d i s t r i b u t i o n 3 when apple supply was low, and later in d i s t r i b u t i o n 1 and 2 t r i a l s when apple supply was depleted, but not u n t i l apples were rarely located. It could be argued that the animals eventually switched food types and thus hypothesis 4 should not be rejected. If the deer preferred apples so much they considered i t 'profitable' (sensu optimal foraging theory; Emlen 1966, MacArthur and Pianka 1966) to continue to search for apples u n t i l they were almost exhausted, then r e j e c t i o n of hypothesis 4 may be misleading because i n i t i a l food abundance should not be expected to influence s e l e c t i o n . I.t is also possible that because deer, were under no pressure to forage e f f i c i e n t l y during t r i a l s , they may have been prepared simply to search for a l l of the highly preferred 60 food f i r s t , and then f i l l up on lower-ranked foods that were also of high q u a l i t y . Hypothesis 4, however, examined the response of animals to i n i t i a l food abundance. The animals switched to less preferred foods once apple a v a i l a b i l i t y was low. The higher proportion of apple platforms cleared of apples per unit time in d i s t r i b u t i o n 3 than in d i s t r i b u t i o n 2 was probably the result of apple density. D i s t r i b u t i o n 3 d i f f e r e d from d i s t r i b u t i o n 2 only in the number of apples pieces present on a platform. Because less time was required to clear platforms of apples in d i s t r i b u t i o n 3 (75% less apple to eat), we should expect more platforms to have been v i s i t e d in d i s t r i b u t i o n 3 than in d i s t r i b u t i o n s 1 and 2 over a fixed time period. High s e l e c t i v i t y at the start of foraging bouts i s contrary to conventional wisdom. Increases in the degree of selection exhibited later in foraging bouts after i n i t i a l unselective foraging, has been reported for c a t t l e (Bos taurus; Hafez and Schein 1962) and mountain goats (Oreamnos americanus; Geist 1971). Early consumption of pelleted dairy ration in several t r i a l s by the male might be explained by early lack of s e l e c t i v i t y for apples ( i . e . f i l l i n g an empty stomach). The fact that high quality food was available may have caused the deer to be highly selective even at low apple a v a i l a b i l i t y and then to f i l l up on other foods when apples were no longer located; a result that might not be expected in the wild. Memory has been shown to play a role in the foraging behaviour of a number of organisms. Gass and Sutherland (1985) 61 found rufous hummingbirds (Selasphorus rufus) began foraging each day using a strategy that had worked the previous day. Although the d i s t r i b u t i o n had changed overnight, the action that worked most e f f e c t i v e l y for the birds the l a s t foraging day was repeated. A number of species of birds, including t i t s (Parus spp.; Cowie et a l . 1981, Shettleworth and Krebs 1982) and corvids (Bossema 1979, Vander Wall and Balda 1981), store large numbers of seeds in scattered locations, recovering them days or even months l a t e r . Under experimental conditions, marsh t i t s (Parus p a l u s t r i s ) performed more e f f i c i e n t l y when finding seeds they had cached themselves as compared to control seeds hidden by observers (Shettleworth 1983). I found that memory played a similar role in food selection of deer. Deer learned to better exploit food d i s t r i b u t i o n s with repeated exposure to them. As a result, hypothesis 5 was not rejected. Figure 2.6 and Table 2.9 i l l u s t r a t e that there was learning of some form during successive t r i a l s with each d i s t r i b u t i o n . For example, time eating apples per distance t r a v e l l e d (s/m) during the f i r s t 5 min of t r i a l s increased s i g n i f i c a n t l y with repeated exposure to the same d i s t r i b u t i o n . The rate of improvement decreased as the t r i a l s progressed and slowed to 0 by 90 min. Increases in foraging e f f i c i e n c y (time eating apples per distance travelled) with repeated exposure to the same d i s t r i b u t i o n , were not accounted for by the animals clearing a higher proportion of apple platforms with repeated exposure to the same d i s t r i b u t i o n . With the exception of the 15-90 min i n t e r v a l for the male in d i s t r i b u t i o n 1 , there was no 62 s i g n i f i c a n t r e l a t i o n between the proportion of cleared platforms and the number of consecutive exposures to a d i s t r i b u t i o n . The exception with the male resulted from him spending less time foraging during the f i r s t few t r i a l s . Further evidence of the role of memory in the use of search path or food location was provided by declines in apple location e f f i c i e n c y when d i s t r i b u t i o n s were changed. These data led to acceptance of hypothesis 6, that because animals were using previous experience with a d i s t r i b u t i o n to find preferred food, e f f i c i e n c y of exploitation should decline in t r i a l s in which a new d i s t r i b u t i o n was encountered. Lower e f f i c i e n c y on d i s t r i b u t i o n 2 than on d i s t r i b u t i o n 1 suggested that a strategy learned on d i s t r i b u t i o n 1 had to be unlearned on d i s t r i b u t i o n 2. This result was again consistent with the findings of Gass and Sutherland (1985) for hummingbirds. Comparison of search paths before and after d i s t r i b u t i o n changes (Fig. 2.4) and the simple comparisons made in Table 2.11 both suggested that the area of the pen i n i t i a l l y searched accounted for changes in searching e f f i c i e n c y within and between d i s t r i b u t i o n s . For example, in t r i a l 20 (Fig. 2.4), the female unsuccessfully pursued a search path similar to that she used in d i s t r i b u t i o n 1 and did not detect clumps of apples 5 m away. What role does memory play in the foraging behaviour of deer encountering food under less controlled conditions? Deer probably re l y on. a number of strategies* in addition to memory to locate food resources. Seed-caching birds often use strategies l i k e preferences for certa i n types of s i t e s 63 (Shettleworth 1983) to f a c i l i t a t e their recovery of caches under natural conditions.. However, the food resources of deer are temporally dynamic, unlike seed caches or platforms containing apples. Once browsed, food resources are not replenished overnight as they may be for nectar feeding animals. Although deer can learn more e f f i c i e n t ways of searching for food in an individual patch, they were ea s i l y fooled. Deer did not recognize when the d i s t r i b u t i o n had changed at the outset of a t r i a l . However, they had no way of knowing before sta r t i n g to search that they were dealing with a di f f e r e n t d i s t r i b u t i o n . Optimal foraging theory is a formulation of information animals may need to 'know', and ways they may need to use i t , to maximize rates of intake while foraging. These requirements may include d e t a i l s of food a v a i l a b i l i t y in the individual's environment. Spatial memory at a patch l e v e l would be as b e n e f i c i a l to a foraging deer as i t i s to birds (Shettleworth and Krebs 1982, Gass and Sutherland 1985). These data present evidence of s p a t i a l memory for only one patch and because changes in d i s t r i b u t i o n were not immediately recognized by the animal, foraging e f f i c i e n c y decreased (Fig. 2.6) when di s t r i b u t i o n s changed. Whether deer can retain information about several patches simultaneously remains untested. Reduction in e f f i c i e n c y of locating apples between d i s t r i b u t i o n s 2 and 3 (Fig. 2.6) was not due to changes in apple d i s t r i b u t i o n . Because the t r a n s i t i o n from d i s t r i b u t i o n 2 to 3 represented no change in s p a t i a l pattern, declines in the time handling apples per distance t r a v e l l e d were l i k e l y due to 64 the number of apple pieces available per platform v i s i t e d . However,, a delay of two weeks between t r i a l s with d i s t r i b u t i o n s 2 and 3 (due to weed problems), may also have contributed to differences in the r e s u l t s . Even with the lower apple a v a i l a b i l i t y , e f f i c i e n c y of finding apples continued to increase within d i s t r i b u t i o n 3 (Fig. 2.6). Repeated exposure of animals to the same d i s t r i b u t i o n did not a f f e c t apple consumption per distance t r a v e l l e d for three of the four animal-distribution combinations, when entire t r i a l s were considered-(Table 2.9). The o n l y d i s t r i b u t i o n with s i g n i f i c a n t improvement (male with d i s t r i b u t i o n 1), was l i k e l y caused by low foraging intensity at the start of the experiments. Over the t o t a l duration of t r i a l s , distance t r a v e l l e d by the animals would be covariate with foraging i n t e n s i t y . Considerable walking towards the end of t r i a l s would mask e f f e c t s of i n i t i a l high e f f i c i e n c y . A l t e r n a t i v e l y , animals may only have been remembering areas of the pen where intake was high from previous t r i a l s and headed there at the start of t r i a l s . Searching might have followed a more random pattern l a t e r in t r i a l s , although I did not test t h i s . The distance at which food i s detected by large herbivores has not been well studied. Owen-Smith and Novellie (1982) constructed a foraging model for kudu (Tragelaphus  strepsiceros ), in which the animal was assumed to encounter food within 0.5 m of i t s search path. Potvin and Huot (1983) assumed a 1-m browse s t r i p for white-tailed deer in a carrying capacity model. Data were not presented to substantiate either a 0.5 or 1.0-m detection distance and measurement of search 65 path width under natural conditions seems very d i f f i c u l t . Q u a l i t a t i v e assessment of my data suggested that the deer did not detect food more than 5 m away from themselves (see Fig. 2.4). Apples were rarely missed by <5 m. I believe this distance was largely an a r t i f a c t of the 5-m spacing of platforms. If platforms had been spaced at 7, 10, or 12 m, the data may have indicated a higher detection distance. The estimate i s important, however, in determining i f the contents of platforms 1, 2, or 3 rows (5, 10, or 15 m) away were 'known' when a foraging decision was made. Given a choice of <5, =5, and >5 m, I concluded 5 m was a good estimate of search path width. The much shorter distances assumed by Owen-Smith and Nbvellie (1982) and by Potvin and Huot (1983) may be a re s u l t of an interaction of forage abundance and detection distance; the denser the vegetation, the smaller the detection window for the forager. Neither duration of food deprivation nor amount of a l e r t behaviour (frequency and duration) explained s i g n i f i c a n t amounts of v a r i a t i o n in selection or foraging intensity during s t a l l or pen t r i a l s . Both might be expected to cause the animal to spend less time looking for preferred food (see for example Beukema 1976). Physiological status of an animal has been shown to influence food intake rate (Arnold and Dudzinski 1978). Fat sheep tend to feed for less time and less intensely than do sheep in poorer condition (Arnold and B i r r e d ! 1977). However, my captive deer were in excellent condition and they were not p h y s i o l o g i c a l l y stressed. Absence of a hunger effect in my 66 study might have resulted from animals not being stressed enough, either in terms of. the length of food deprivation, or an absence of need to forage e f f i c i e n t l y within the pen t r i a l s . Conclusions Ad libitum s t a l l t r i a l s produced a consistent preference ranking of apples > pelleted dairy ration > pel l e t e d a l f a l f a for both animals. This resulted in acceptance of hypotheses 1 and 2 (page 14) and a food ranking for tests of s p a t i a l e f f e c t s of food d i s t r i b u t i o n on s e l e c t i o n . Variation in selection among t r i a l s may have been in part att r i b u t a b l e to v a r i a t i o n in the quality of apples used and the measure used as a surrogate of selection. While time b i t i n g provided a useful index of food intake, intake rate could be estimated more accurately by combining time b i t i n g and chewing into handling time. This l a t t e r measure would be more appropriate for estimating intake within t r i a l s and among foods. Selection o>f; food in pen t r i a l s ^ resembled' ad libitum' preference ranking as long as apples were abundant. Making animals search for food did not in i t s e l f cause food, selection to d i f f e r from preference. Apples continued to be eaten u n t i l the animal f a i l e d to locate t h i s preferred food at which time an increase in the selection of the second-ranked food occurred. Deer exploited food d i s t r i b u t i o n s better with repeated exposure to them. Improvement in the time b i t i n g apples per distance t r a v e l l e d during early portions of t r i a l s resulted 67 from animals learning more e f f e c t i v e search paths with experience in a d i s t r i b u t i o n . Animals did not immediately recognize new di s t r i b u t i o n s and by using search paths that they had used in previous d i s t r i b u t i o n s , the deer greatly reduced their intake rates at the start of t r i a l s on new d i s t r i b u t i o n s . Deer would have to recognize d i f f e r e n t patches i f they were to gain from knowledge of food d i s t r i b u t i o n within a patch. It remains to be tested whether or not deer learn e f f i c i e n t ways of foraging in a number of patches simultaneously, or i f such knowledge would be ben e f i c i a l in a changing environment. 68 CHAPTER 3 - THE EFFECTS OF FOOD AVAILABILITY AND DISTRIBUTION ON FOOD SELECTION Introduct ion Foraging animals encounter food and decide, based in part on experience (Gluesing and Balph 1980), p a l a t a b i l i t y (Radwan and Crouch 1974), and requirements (see Arnold and Dudzinski 1978), whether to eat a food item or continue without b i t i n g . This decision may involve the animal's perception of what is avail a b l e . Rates of encountering food types, based on a random walk model, can be highly dependent on the underlying d i s t r i b u t i o n of foods (Gillingham unpubl. data). The higher the degree of contagion, the larger the sample needed by the 'forager' (or observer) to accurately describe a d i s t r i b u t i o n . Plants are rarely randomly dispersed, and are often highly aggregated (Goodall 1974). Because of these problems, food preference and selection cannot be e a s i l y separated from density and distribution' of food items under natural conditions. Researchers use measures of preference to describe the choices a foraging animal makes. However, unless the animal i s omniscient with respect to forage a v a i l a b i l i t y , the choices i t makes cannot accurately be compared to the r e l a t i v e quantities of forage a v a i l a b l e . The implication i s not that the animal need be omniscient, but rather that the observer makes t h i s assumption i m p l i c i t l y when a l l foods are described as equally a v a i l a b l e . Differences in selection, such as those described 69 by Spalinger (1980), may be due not only to changes in the r e l a t i v e abundance of forage types, but also to temporal and/or s p a t i a l changes in forage q u a l i t y . Temporal and s p a t i a l factors have been shown to influence forage q u a l i t y . Percent crude protein content of deer browse was related to season and s i t e exposure, elevation, and l i g h t i n t e n s i t y (Dealy 1959). Concentrations of minerals within twigs of Vacciniurn parvifolium s h i f t on a diurnal basis ( E l l i s 1985). McCann (1985) found v a r i a b i l i t y in crude protein content of one-year-old s a l a l (Gaultheria shallon) leaves within twigs, among plants, and among s i t e s . I n t r a s p e c i f i c v a r i a b i l i t y in food quality i s l i k e l y to be a factor in forage selec t i o n . Even given conditions of constant food q u a l i t y and known food a v a i l a b i l i t y , optimal foraging theory (sensu Charnov 1976b) predicts that the d i s t r i b u t i o n of preferred food influences diet s e l e c t i o n . The abundance of high qu a l i t y foods should determine what foods are eaten and/or when to forage. To determine the effects of food d i s t r i b u t i o n on diet selection by deer, food consumption was compared between ad libitum feeding t r i a l s and d i s t r i b u t i o n s of food in which food abundance and dispersion of preferred food varied. As in 1981, I used uniform foods to control for quality v a r i a t i o n s within foods. Data from the 1981 t r i a l s (Chapter 2) suggested that deer were not omniscient with respect to food d i s t r i b u t i o n . In most cases they did not detect preferred food beyond =5 m of their p o s i t i o n . Therefore, I expected food selection to be based on 70 i n i t i a l preference modified by the types of food encountered, and not by t o t a l food abundance (in the entire enclosure). I assumed that preference was exhibited under ad libitum conditions and that animals retained t h i s preference when encountering food with varying a v a i l a b i l i t y . By using this preference for test foods, I could test the e f f e c t s of preferred food density and dispersion on food sele c t i o n . Based on the 1981 t r i a l s , I assumed the contents of platforms were not detected by animals unless the animal passed within 5 m of the platform. Using t h i s estimate of the types of foods encountered by the animals, forage a v a i l a b i l i t y and food selection were examined. The effects of food dispersion on diet selection were tested by presenting the deer with d i s t r i b u t i o n s equal in density but d i f f e r i n g in the dispersion of the preferred food. Hypotheses To e s t a b l i s h preference for test foods during 1983 experiments, ad libitum s t a l l t r i a l s again required testing of hypotheses 1 and 2: (1) Deer would exhibit a consistent preference for apples, pelle t e d dairy ration, and pelleted a l f a l f a ; (2) These preferences would not d i f f e r between animals. Given a prediction of preference for use during pen t r i a l s , I tested hypothesis 3*, this time over a larger range' of a v a i l a b i l i t y of preferred foods: 71 (3) The addition of a s p a t i a l component to food abundance w i l l not a l t e r food s e l e c t i v i t y by deer when each food i s unlimited r e l a t i v e to the deer's consumption; The range of densities and dispersions of apples presented to the animals enabled testing of the following hypotheses re l a t i n g to food a v a i l a b i l i t y : (7) Clumped d i s t r i b u t i o n s of preferred food w i l l allow the animal to exploit food patches rather than individual platforms, thus increasing the consumption of preferred foods for a given food density; (8) Deer w i l l respond to a lower abundance of preferred foods by eating more of lower-ranked foods at each feeding location; (9) When deer stop eating only preferred food and begin eating foods of lower rank, the t r a n s i t i o n w i l l not be abrupt. Deer w i l l continue to search for apples, increasing the rate at which non-apples are eaten as less preferred food is encountered; (10) When preferred foods are encountered, the animal w i l l continue to search for food in that area, concentrating the search path after preferred foods have been located. Data from 1983 pen t r i a l s were also used to address other aspects of deer foraging behaviour and their r e l a t i o n to forage a v a i l a b i l i t y : (a) using an estimate of search path width, foods encountered by the deer were compared with r e l a t i v e abundance of food in the enclosure; (b) relationships between the proportion of platforms cleared of apples when f i r s t 72 encountered by the animals and apple density and d i s t r i b u t i o n were examined; and (c) u s e : a v a i l a b i l i t y ratios (food used/food available) were used to examine e f f e c t s of density and d i s t r i b u t i o n of preferred food on c l a s s i c a l preference estimates. Methods Data were c o l l e c t e d from June 27 to September 23, 1983 using methods sim i l a r to 1981 experiments (page 16). The same animals, one male and one female, both three years old, were used throughout the t r i a l s . Between 1981 and 1983 t r i a l s , the two test animals were kept with the rest of the captive herd at the Research Forest. A l l animals had ad libitum access to pelleted dairy ration and a l f a l f a . Their access to apples over t h i s period was infrequent. S t a l l and pen t r i a l s were run between =*0600 and ^ 0900 h; t r i a l s with the male always preceded tests with the female. Test foods included apples,, pelle t e d dairy ration, and pelleted a l f a l f a . During the 1981 t r i a l s , apples had been obtained according to market a v a i l a b i l i t y ; variety and qual i t y varied. For the data presented in t h i s chapter, red del i c i o u s apples were purchased by the case and cold stored at the U.B.C. Research Forest, eliminating most of the apple to apple var iat ion. Data were recorded using a MORE behavioural recorder; only one observer was used throughout a l l experiments. Behaviours recorded did not d i f f e r from 1981: walking, running, b i t i n g , 73 chewing, alertness, grooming, resting, and voiding. Based on conclusions reached from the 1981 experiment., handling time was based on combined b i t i n g and chewing durations. Although large and small bites might each require the same duration, more chewing'would be required to process the larger b i t e , given a r e l a t i v e l y constant food-size threshold for swallowing. I used t h i s handling time measure to estimate consumption rates throughout t r i a l s , and to compare estimated weight of foods eaten among foods and d i s t r i b u t i o n s . S t a l l t r i a l s The animals' preference for apples, pelleted dairy ration, and p e l l e t e d a l f a l f a was established in c a f e t e r i a - s t y l e s t a l l t r i a l s s i milar to 1981 tests (see F i g . 2.1 and description on page 18). T r i a l s lasted 40 min. Duration of food deprivation varied from 8.3-10.9 h for the male and 9.2-12.3 h for the female. Location of foods, placed in 60 x 60 x 30-cm feeding buckets, were; assigned randomly- for each t r i a l - . Eighteen ad libitum s t a l l t r i a l s were conducted, nine with each animal. Changes in preference over the summer were tested by conducting six t r i a l s with each deer before the f i r s t pen t r i a l and three each after completion of the pen t r i a l s . 'Ad libitum' was i n i t i a l l y considered to be =1200 g each of apples, p e l l e t e d dairy ration, and pelleted a l f a l f a (from 1981 data). Because both animals increased the amount of apples consumed during the t r i a l s , the weight of apples available during s t a l l t r i a l s was increased to =2000 and =1500 74 g for the male and female, respectively. Because deer ate a l l the available apples during some t r i a l s , I partitioned these t r i a l s by the time at which apples were no longer available. I used time spent handling each food type (biting and chewing) from the start of the t r i a l to 5, 10, and 40 min to estimate preference under ad libitum conditions. Preference for foods was determined by comparing the observed handling times of a l l three foods. The time of the f i r s t switch from apples and the weight of apples consumed to that point were estimated by assuming a constant intake per handling time rel a t i o n s h i p throughout a t r i a l . An average food-specific intake rate was calculated by d i v i d i n g the t o t a l weight of a food consumed by the t o t a l time handling that food during a t r i a l . Average intake rates were calculated for each food and t r i a l , and compared between animals and among food types, using analysis of variance. Location and type of f i r s t bite were also examined and compared between animals. Pen t r i a l s The vegetation-free enclosure used in 1983 t r i a l s was unmodified from 1981. Test animals were fasted in the main pens and moved into the holding area adjacent to the 0.5-ha pen -1 h before t r i a l s . Fasting varied from 7.5-11.7 h for the male and 8.9-12.6 h for the female. Between t r i a l s , each animal was isolated with ad libitum access to water, pelle t e d dairy ration, and a l f a l f a hay. Male t r i a l s began at =0600 h 75 and female t r i a l s at =0730 h. Forty-one 1-h pen t r i a l s were conducted. One-hour t r i a l s were long enough to gather the relevant data, they permitted a single observer to conduct t r i a l s with both animals each morning (=4 h time period), and they did not overload the storage c a p a b i l i t i e s of the behavioural recorder. By completing both male and female t r i a l s within a 4-h period, inclement weather affected both t r i a l s ; both animals could be subjected to the same d i s t r i b u t i o n each day, simplifying experimental protocol. A t r i a l commenced when the deer entered the feeding enclosure from the holding pen. Despite d a i l y mechanical removal of weeds, t r i a l s were interrupted for =4 weeks from July 26 to August 31 to reapply herbicides. Following the break, the female continued to search for weeds and test foods were almost ignored. Three t r i a l s were attempted and discarded; subsequent t r i a l s used only the male. Eight d i s t r i b u t i o n s were used during the experiments (Fig. 3.1). Four densities of apples were presented in clumped and unclumped pai r s . Clumped d i s t r i b u t i o n s consisted of groups of nine 30 x 30-cm platforms (in a 3 x 3 configuration) randomly placed throughout the pen (Fig. 3.1). Pelleted dairy ration and pelleted a l f a l f a were randomly assigned to the remainder of the 144 platforms. In non-clumped (dispersed) apple d i s t r i b u t i o n s , a food type was randomly assigned to each platform. Four pieces of apple- ( t o t a l weight =36 g), 50 ml of pelle t e d dairy ration (=32 g), or 50 ml of pelleted a l f a l f a (=27 g) were placed on each platform. No platforms were l e f t Figure 3.1. D i s t r i b u t i o n s used during 1983 pen t r i a l s . Apples are represented by •, pelleted dairy ration by O, and a l f a l f a p e l l e t s by •. Every platform contained either =36 g of apples in four equal pieces, or =50 ml of p e l l e t e d ration or a l f a l f a . Platforms were 5 m apart. 77 O D O O O O O O O O O O O O O O O O a o * « « a o o o o o o o o o « * « O Q » » « O O O D O D O D O O « « » 0 O » » » O D D O D O D O 0 O » » « O Q O O O O O O O O O O O O O O D O O D O D D O » » « O D O O O D O O O O O O D O O » » « O O O O D O D O O O O O O O D « « » D O O O O O O O O DISTRIBUTION 5 CLUMPED MEDIUM-LOW O D O O O O O O » D O O O D O » O D D O D O » D O D O » D O D » O O Q « • • • 0 0 0 0 « O D O » 0 0 » D « 0 D O O D O D O o a o D D O D O O D O O D O » 0 0 » D O O O » 0 0 » Q » 0 0 » O O O D O O » O D O O D O O O O O O O O O O O O D O « 0 0 » D O O « • O D 0 » 0 » 0 » 0 O 0 O 0 0 » 0 O DISTRIBUTION 6 DISPERSED MEDIUM-LOW DISTRIBUTION 7 CLUMPED-LOW O O O D O O O D » 0 » D O D O « O D 0 O O 0 O » O O O O O O O » 0 O 0 O O O O O Q O O O O O D O O O O D O O O D « 0 0 0 » O D » O O O O D O O D o o o « a o o a o o o a * o D * o o o a o o o o a o a o o o o o o o o o O » O D 0 « D D » O O » 0 0 O » O 0 D D O n o D O O O D D O O « O D » 0 DISTRIBUTION 8 DISPERSED-LOW 78 D 0 0 D 0 » » » O D D O O D O » » » • • • 0 0 » » » 0 D O » » » D » » » • • • • O O 0 O 0 0 O » » » O 0 0 0 • O D O O O O » » » O O O D D » » » D D O » » » D » « » 0 « » » D « » » o o o » « « o » » » o » » » o » » » 0 D D » « » 0 D D O D « » » 0 D D O D I S T R I B U T I O N 1 CLUMPED-HIGH O 0 O » O » D O D 0 » » 0 » » » • • • • O » D 0 D » D « O O D » D D » 0 » D O « 0 » » O D D O » • O D » 0 » « 0 » 0 » » O O D » D I S T R I B U T I O N 2 D1SPERSED-H1GH D O O D O D O O » » « O D D O D O D 0 D O D 0 0 0 D * * 9 0 0 0 0 D 0 Q D O D D O D O O » « « O O C O O D O • • • O O O D O D O O O O O » » » O • • • • • • • • D O O O O D « « « D 0 0 0 0 « « » 0 0 » « » O O O O O D O O D O D O O D O « « * O O D O D O D I S T R I B U T I O N 3 CLUMPED MEDIUM-HIGH D a o o « o « D « D o o a * o * o » O D » O D O O » 0 « 0 » » O D O O » • 0 0 0 0 » O D O D D O O O « D « D 0 0 » 0 » O D » 0 « 0 « O » O O O O • 0 0 » O O O O O O O O O O O D « D 0 0 » 0 0 0 » 0 » D O » O D O » D » • O O O Q « O D O O « 0 0 » 0 0 * D D I S T R I B U T I O N 4 DISPERSED MEDIUM-HIGH 79 empty. Relative densities were chosen to r e f l e c t a broad range of, apple a v a i l a b i l i t i e s . I used test d i s t r i b u t i o n s with ratios (apples:pelleted dairy r a t i o n r p e l l e t e d a l f a l f a ) of: (1) 2:1:1 (high apple density); (2) =1:1:1 (medium-high apple density); (3) =1:2:2 (medium-low apple density); and (4) =1:3.5:3.5 (low apple density). The number of t r i a l s , by animal and d i s t r i b u t i o n , used in the analysis are presented in Table 3.1. Dis t r i b u t i o n s were altered d a i l y to avoid e f f e c t s of animals learning p a r t i c u l a r d i s t r i b u t i o n s . The sequence of d i s t r i b u t i o n presentation progressed from the clumped-high to the dispersed-low d i s t r i b u t i o n (Fig. 3.1), returning to d i s t r i b u t i o n 1 after a l l eight had been used. The animal's location in the pen was recorded (nearest m) whenever the deer changed d i r e c t i o n . From these data the entire search path was reconstructed and the distance the animal t r a v e l l e d as a function of time was calculated. I used t o t a l time handling each food (biting and chewing) as a measure of the consumption of each food type, and to compare the selection of the animals among foods. A 5-m distance on either side of the search path (Fig. 3.2) was used to estimate food encountered (hereafter termed "seen"). Although the term "seen" suggests the sense of v i s i o n , I w i l l use i t to refer to an estimate of the platforms that the forager encountered or was exposed to. For each t r i a l , platforms within 5. m of the animal's search path (shaded area of F i g . 3.2) were recorded along with the time. These data were used as an estimate of the platforms detected by the Table 3.1. Summary by subject and d i s t r i b u t i o n of 1983 pen t r i a l s . Numbers in the la s t two columns are numbers of t r i a l s of each d i s t r i b u t i o n run with each i n d i v i d u a l . This table does not include the la s t three t r i a l s with the female when she was dropped from the experiment. Di s t r i b u t i o n Subject Apple Density Dispersion Male Female . High (72) a Clumped 3 2 Di spersed 3 2 Medium-High (45) Clumped 3 2 Dispersed 3 2 Medium-Low (27) Clumped 4 2 Dispersed 3 1 Low (18) Clumped 3 1 Dispersed 3 1 a Number of platforms (out of 144) which contained apples. F i g u r e 3.2. A h y p o t h e t i c a l example of a s e a r c h p a t h f o r 1983 pen t r i a l s i l l u s t r a t i n g 5-m d e t e c t i o n d i s t a n c e . The d i s p e r s e d - l o w a p p l e d i s t r i b u t i o n i s d e p i c t e d . The dark c e n t r a l l i n e r e p r e s e n t s the a n i m a l ' s p a t h and the c r o s s - h a t c h e d s t r i p a 5-m i n t e r v a l on e i t h e r s i d e . P l a t f o r m s , and t h e i r a s s o c i a t e d f o o d s , were assumed to be 'seen' i f they f e l l w i t h i n the 10-m s t r i p and were r e c o r d e d a l o n g w i t h the time at which they were passed. 82 forager throughout the t r i a l . The frequencies of the three food types seen were compared to the frequencies of food platforms in the pen by contingency table analysis (0-5, 0-10, 0-15, 0-60 min i n t e r v a l s ) , to examine apple density and / dispersion e f f e c t s on the types of food encountered. Measurements of food, before and after t r i a l s , were based on volume for l o g i s t i c a l reasons. Sixty samples, of both pelleted dairy ration and a l f a l f a , were measured to estimate weight per volume relationships. Mean values of 0.625 ± 0.011 (x ± SD) for pelleted dairy ration arid 0.548 ± 0.016 (g/ml)' for a l f a l f a were obtained and used in conversions of volume to weight. Oven dry weights were obtained by drying eight samples each of apples, pelleted dairy ration, and pell e t e d a l f a l f a at =90°C ,in drying ovens for =24 h, u n t i l no difference was found between consecutive weighings. Dry weight as a proportion of wet weight (g) was found to be 0.141 ± 0.006 (x ± SD), 0.866 ± 0.002, and 0.872 ± 0.003 for apples, pelleted dairy ration, and pelleted a l f a l f a , respectively. Mean values were used to relate consumed and available foods on a dry weight basis. Although selection was based on time handling foods, an estimate of the weight consumed was needed to examine variation in t o t a l intake among t r i a l s . Following each t r i a l , samples of apple, pell e t e d dairy ration, and p e l l e t e d a l f a l f a platforms v i s i t e d during the t r i a l were measured. Only platforms that were v i s i t e d only once during the t r i a l were used to avoid confounding estimates of intake- rates. Intake rates of apples, pelleted dairy ration, and pelleted a l f a l f a were estimated from platforms v i s i t e d only once during a t r i a l . Estimates of 83 intake rate (ml/s and g/s based on the above conversions) were based on weight consumed and duration of handling time corresponding to a feeding platform. Analysis of variance and Scheffe's tests were used to examine differences in estimated intake rates for 0-5, 5-10, 10-20, and 20 to 60-min interv a l s as a function of apple density and dispersion. Mean consumption rates (ml/s) of apples, pelleted dairy ration, and pelleted a l f a l f a , for each of the four time i n t e r v a l s , were used to estimate weight consumed as a function of time into the t r i a l . Estimation of c l a s s i c a l preference ( u s e : a v a i l a b i l i t y r a t i o s ; sensu Van Dyne and Heady 1965) was examined in a number of ways. The amount of food available was determined on both a wet and dry weight basis, for the entire enclosure and for the 10-m-wide, seen corridor. Ratios of the amount of food consumed divided by the amount available were examined to 5, 10, 15, and 60 min and compared among apple densities and d i s t r i b u t i o n s , using contingency tables. Deer did not always clear apple platforms when they f i r s t v i s i t e d them. A platform was considered to be uncleared i f any pieces of apples remained after the deer l e f t the platform for the f i r s t time. Relationships between the number of platforms cleared and the time and distance t r a v e l l e d to the f i r s t uncleared platform, and apple abundance and d i s t r i b u t i o n were examined through c o r r e l a t i o n analysis. Attempts were also made to relate the number of, uncleared apple platforms to the number of apple platforms seen, and the t o t a l number of platforms v i s i t e d to 5, 10, 15, and 60 min, as well as duration of food 84 deprivation. Effects of apple, density and patch dispersion on food consumption were evaluated using two-way analysis of variance (GENLIN; Greig and Bjerring 1980). The following model was used: Yijk = u + ai + 0j + (aP)ij_ + eijk Yijk - response variable tested; u - parametric mean value; ai - the fixed e f f e c t of the j^th density; 02 - the fixed e f f e c t of the j t h dispersion; (aP)ij^ - the interaction of the _ith density and j t h dispersion; eijk - the error associated with the j t h density, j t h dispersion, and kth experimental u n i t . Response variables (Yijk) tested included: t o t a l time handling apples pe^ r distance travelled.; t o t a l time handling and estimated weight consumed of a l l foods per distance t r a v e l l e d ; t o t a l distance t r a v e l l e d ; t o t a l time handling apples to 5, 10, 60, 5-10, and 10-60 min; weight of apples and pelleted foods consumed; proportion of platforms seen to 5, 10, and 60 min that contained apples; and u s e : a v a i l a b i l i t y r a t i o s for apples on a dry weight basis (dry weight of apples consumed divided by the dry weight of apples a v a i l a b l e ) . For t r i a l s with the male, ANOVAs were run using a l l four apple d e n s i t i e s and both dispersions (clumped and dispersed). Because of the smaller 85 number of female t r i a l s successfully conducted, density was pooled to high (high, and medium-high d i s t r i b u t i o n s ) and low (medium-low and low d i s t r i b u t i o n s ) l e v e l s ; both dispersion l e v e l s were tested. Consecutive bites on the same food types were examined to determine i f : (1) apple density affected the animal's switching to lower-ranked foods; and (2) intake rates of non-apples increased later in a t r i a l when apple a v a i l a b i l i t y decreased (related to hypothesis 9, page 71). Only apples and p e l l e t s were considered; pelleted dairy ration and pelleted a l f a l f a were grouped together. A minimum of three consecutive bites on apples or p e l l e t s were needed to estimate cumulative intake rates (Fig. 3.3). Consecutive bites on apples or p e l l e t s were grouped together (Fig. 3.3) and analysis of covariance (Le 1984) used to: (1) estimate the intake rate for each group of consecutive b i t e s ; and (2) examine intake rates for trends within t r i a l s . Consecutive bites on the same food numbering two or less could not be analysed. Because handling time was only correlated with intake, and did not permit dir e c t comparisons of apples and p e l l e t s , cumulative intake rates were not compared between apples and p e l l e t s . A number of factors were examined (correlation analysis) for t h e i r r e l a t i o n to the f i r s t b i t e of a p e l l e t and the f i r s t switch to p e l l e t s (three or more consecutive bites) within each t r i a l . Factors examined included time into the t r i a l , time since the l a s t apple bite, proportion of platforms seen that contained apples, the number of platforms passed since the l a s t apple b i t e , and the number of apple platforms in the pen. 8.6 TIME SINCE START OF TRIAL F i g u r e 3.3. Example of the technique used to compare intake r a t e s of foods w i t h i n 1983 pen t r i a l s . Intake r a t e s were determined f o r c o n s e c u t i v e b i t e s on apples or non-apples numbering three or more. Apples are represented by •, p e l l e t e d d a i r y r a t i o n by Or and p e l l e t e d a l f a l f a by •. Within food type, inake r a t e s of c o n s e c u t i v e b i t e s were compared f o r tre n d s with time i n t o t r i a l u s i ng a n a l y s i s of c o v a r i a n c e . 87 Examination of the turning frequency of the foraging animal, employing a technique similar to that of Smith (1974a) was used to examine changes in turning frequencies with food encountered (hypothesis 10, page 71). The three lin e segments (two turns) before and after an apple or pelleted dairy ration bite were considered (see F i g . 3.4). If another bite occurred during t h i s i n t e r v a l , the data were discarded. Paired-t^ tests were used to compare the s t r a i g h t - l i n e distance over these i n t e r v a l s . Results S t a l l T r i a l s A l l 18 ad libitum s t a l l t r i a l s yielded complete data sets and documented a strong preference for apples over the two pellet e d foods. Food Preference: Hypotheses 1 and 2 Neither hypothesis 1 or 2 (consistent preferences within and between animals) were rejected. Comparison of observed handling times indicated, throughout a l l t r i a l s for both animals, that apples were selected more than any other food type (Table 3.2). Based on cumulative data, pelleted dairy ration and pelle t e d a l f a l f a were consumed far less than apples, but the r e l a t i v e ranking between them was ambiguous. Considering the entire 40-min t r i a l , however, pelleted dairy BITE OF APPLE F i g u r e 3.4. E x a m i n a t i o n of t u r n i n g r a d i i as a f u n c t i o n of food consumption. Three l i n e segments, d u r i n g which no b i t i n g was o bserved, b e f o r e and a f t e r a s u c c e s s f u l b i t e were compared. P a i r e d - ! t e s t s were used t o t e s t d i f f e r e n c e s i n t u r n i n g , D, v e r s u s D 2 b e f o r e and a f t e r b i t e s on a p p l e s and p e l l e t e d d a i r y r a t i o n . T a b l e 3.2. Preference of two black-tailed deer (Odocoi leus hemionus columbianus) for pelleted ration, apples, and pelleted a l f a l f a compared within ad libitum s t a l l t r i a l s . Handling time was compared at 5, 10, and 40 min into t r i a l s . Time handling foods (s) 0 to 5 min 0 to 10 min 0 to 40 min Subject T r i a l • Ration Apples A l f a l f a Ration Apples A l f a l f a Ration Apples A l f a l f a Male Fema1e 1 40 232 0 56 385 47 150 498 50 2 0 297 0 63 468 18 63 479 18 3 0 300 0 38 539 0 184 722 47 4 0 284 0 0 491 0 84 710 29 5 0 290 0 0 526 0 17 915 108 6 0 287 0 0 58 1 0 189 952 165 7 0 272 0 0 501 0 0 795 19 8 0 276 0 0 562 0 15 742 84 9 0 292 0 0 581 0 0 969 0 1 5 291 0 5 484 0 213 939 0 2 0 300 0 0 600 0 165 695 0 3 0 296 0 0 596 0 167 944 5 4 0 300 0 44 560 0 95 699 45 5 14 202 14 21 385 25 53 852 25 6 0 287 0 0 583 0 24 945 0 7 0 207 0 0 31 1 0 0 396 0 8 0 300 0 0 492 0 0 1056 0 9 0 261 0 0 473 0 0 1041 0 90 ration consistently ranked above pelleted a l f a l f a for the female, but not for the male. Apples made up a very high proportion of the t o t a l handling time. Except for the f i r s t 40 s of the f i r s t s t a l l t r i a l , the male spent the f i r s t 5 min of a l l t r i a l s eating apples. During six of the t r i a l s , he ate apples exclusively for the f i r s t 10 min. The female exhibited s i m i l a r , strong selection of apples (see Table 3.2). Apples were also strongly preferred throughout t r i a l s (Table 3.3). With one exception ( t r i a l 2 with the female) apples ranked highest during a l l time inte r v a l s for both animals. During t r i a l 2 with the female, preference for pelleted dairy ration after 10 min of observation resulted from exhaustion of the apple supply (see below). Ad libitum conditions were not maintained throughout a l l t r i a l s because the deer increased their consumption of apples during the experimental period (Table 3.4). The male exhibited a much greater range in amount of apples consumed than the female, eating 2.3 times more apples during the l a s t t r i a l than the f i r s t t r i a l . For the seven t r i a l s in which the apple supply was exhausted, time at which the l a s t apple was eaten and the proportion of the t o t a l handling time spent eating apples to that point in the t r i a l , were calculated (Table 3.4). Exhaustion of the apple supply explained a l l but one occurrence of p e l l e t preference during the 10-60 min i n t e r v a l (from Tables 3.3 and 3.4). Even during- t h i s one. instance, apples s t i l l ranked highest in terms of consumption time. T r i a l s 7 through 9, for both animals, (Table 3.4) were Table 3.3. Examination of c o n s i s t e n c y of p r e f e r e n c e of two b l a c k - t a i l e d deer f o r p e l l e t e d r a t i o n , a p p l e s , and p e l l e t e d a l f a l f a compared w i t h i n ad l i b i t u m s t a l l t r i a l s based on time spent h a n d l i n g each food type from 0-5. 5-10, and 10-40 min. Time h a n d l i n g foods ( s ) 0 to 5 min 5 to 10 min 10 to 60 min S u b j e c t T r i a l R a t i o n Apples A l f a l f a R a t i o n Apples A l f a l f a R a t i o n Apples A l f a l f a Male 1 40 232 0 16 153 47 94 113 3 2 0 297 0 63 171 18 0 11 0 3 0 300 0 38 239 0 146 183 47 4 0 284 0 0 207 0 84 219 29 5 0 290 0 0 236 0 17 389 108 6 0 287 0 0 294 0 189 371 165 7 0 272 0 0 229 0 0 294 19 8 0 276 0 0 286 0 15 180 84 9 0 292 0 0 289 0 0 388 0 Female 1 5 291 0 0 193 0 208 455 0 2 0 300 0 0 300 0 165 95 0 3 0 296 0 0 300 0 167 348 5 4 0 300 0 44 260 0 51 139 45 5 14 202 14 7 183 11 32 467 0 6 0 287 0 v 0 296 0 24 362 0 7 0 207 0 0 104 0 0 85 0 8 0 300 0 0 192 0 0 564 O 9 0 261 0 0 212 0 0 568 0 Table 3.4. Compar ison of weight consumed v e r s u s we ight a v a i l a b l e d u r i n g ad l i b i t u m s t a l l t r i a l s f o r two b l a c k - t a i l e d d e e r . I n c r e a s e s i n the weight of a p p l e s consumed d u r i n g the e x p e r i m e n t a l p e r i o d r e s u l t e d i n the need to i n c r e a s e weight of a v a i l a b l e a p p l e s . Subj e c t T r i a l We i ght of a p p l e s (g) Time ( t ) of t r i a l P r o p o r t i o n o f when a p p l e s u p p l y h a n d l i ng t i me was e x h a u s t e d on a p p l e s to Ava i1ab1e Taken (s ) , t ime t / Mal e 1 1 177 . 3 869. .9 a 2 1 156 . 2 905, ,8 - -3 1 160 . 1 1 160. . 1 1244 0. .912 4 1517 .6 1 195. .3 - -5 1533 .4 1401 . , 5 - -6 1436 . 4 1436. .4 1801 0. .778 7 1484 . 2 1484 , . 2 1035 1 . .000 8 1596 .6 1596 . .6 1118 1 .000 9 2037 .0 2008 . 3 - -Female 1 1 106. . 1 1 106. 1 1930 0. . 878 2 1 149. . 1 1 149 . 1 757 0. .97 1 3 1521 .9 1235 . 1 - -4 1393. .6 1393. 6 757 0. 928 5 1607 . 8 895 . 2 - -6 1376. 5 1308 . 4 - -7 1506. 1 152 . 2 - -8 1488. 7 1033. 4 - -9 1545 . 3 1065. 0 a hyphen i n d i c a t e s t h a t not a l l a v a i l a b l e a p p l e s were consumed d u r i n g the t r i a l . to 93 conducted after the completion of pen t r i a l s . Apple consumption (t o t a l weight) by the male appeared to increase during the pen t r i a l interlude while consumption by the female decreased. Unlike the 1981 t r i a l s , type of food and not location of the feeding buckets determined the food eaten f i r s t . Although apples were presented three times in each feeding location for both animals, the female always selected apples for the f i r s t b i t e . With the exception of the f i r s t t r i a l with the male, in which p e l l e t e d dairy ration located in the central feeding p o s i t i o n was eaten f i r s t , apples made up the f i r s t bite of the remaining t r i a l s (17 of 18 for both deer). Time handling apples before the f i r s t bite of p e l l e t s was taken to estimate the i n i t i a l weight of apples eaten (Table 3.5). The i n i t i a l consumption of apples by the male ranged as high as 2 kg; =600 g was the minimum amount consumed. The female ate between =150 g and =1.1 kg. For comparative purposes, the t o t a l weight of apples available during the pen t r i a l s ranged from =2.6 kg (high d i s t r i b u t i o n ) down to =648 g (low d i s t r i b u t i o n ) . S t a l l intake rates (g/s) were calculated by subject, food, and t r i a l (Table 3.6), and compared by analysis of variance. Intake rate d i f f e r e d s i g n i f i c a n t l y between subjects for apples and p e l l e t e d dairy ration. No s i g n i f i c a n t difference existed for a l f a l f a , but this food was consumed during only two t r i a l s with the female' y i e l d i n g a very small sample f o r comparisons, (Table 3.6). In addition to the greater weight of apples consumed during the la s t three t r i a l s with the male, apple Table 3.5. E s t i m a t i o n of weight (g) of a p p l e s consumed by two b l a c k - t a i l e d deer u n t i l f i r s t s w i t c h from a p p l e s to p e l l e t e d foods d u r i n g ad l i b i t u m s t a l l t r i a l s . E s t i m a t e s assume a c o n s t a n t i n t a k e r a t e (g/s) throughout the 40-min t r i a l s . Weight consumed to s w i t c h i s based on h a n d l i n g time to the f i r s t s w i t c h as a p r o p o r t i o n of the t o t a l h a n d l i n g time. S u b j e c t T r i a l Time of 1st s w i t c h H a n d l i ng t i me to s w i t c h ( s ) Ha n d l i n g t ime f o r t r i a l ( s ) T o t a l weight consumed (g) E s t i m a t e d weight consumed to s w i t c h (g) Male 378 361 380 1327 600 737 1035 1035 2400 347 342 372 668 419 702 795 742 969 498 479 722 710 915 952 795 742 969 869.9 905.8 1160.1 1 195 . 3 1401.5 1436.4 1484.2 1596.6 2008.3 606 . 1 646 . 7 597 . 7 1124.6 641 .8 1059.2 1484.2:: 1596.6 2008.3 C Female 217 630 861 390 270 650 2400 2400 2400 217 600 816 380 192 612 396 1056 104 1 939 695 944 699 852 945 396 1056 104 1 1106. 1 1 149. 1 1235 1393 895 1308 152 1033 1065 261 . 5 992 .0 1067.6 757.6 201 . 7 847.3 152.2 C 1033.4 C 1065.0° a b c P e l l e t s consumed f o r 40 s b e f o r e a p p l e s were Apple s u p p l y exhausted p r i o r to s w i t c h i n g to Only a p p l e s eaten d u r i n g t r i a l . f i r s t e a t e n i n t h i s t r i a l o t h e r f o o d s . Table 3.6. Average intake rates based on total weight consumed/total time handling each food type for two bl a c k - t a i l e d deer during 40-min ad libitum s t a l l t r i a l s . P r o b a b i l i t i e s shown at the bottom of the table are based on analysis of variance between subjects, for each food type. Subject T r i a l Average intake rate (total weight/time handling (g/s)) Dairy Ration Apples Pelleted Al f a l fa Male 51 32 24 84 10 65 0.80 1 .75 1 .89 1 .60 1 .68 1 .53 1.51 1 .87 2.15 2 .07 36 36 61 36 0. 38 0. 36 37 35 X + SD 0.9210.48 1.78±0.23 0.39±0.09 Female .37 25 32 53 23 14 1.18 1 .63 1.31 1 . 98 1 .05 1 . 38 0. 38 0.92 0.08 0.80 1 .02 X ± SD 0.31+0.13 P(no animal difference) O.OII 1.21+0.45 0.004 0.44±0.51 0.770 n S P < 0.05 P < 0.01 P > 0.05 96 intake rate in s t a l l t r i a l s was higher (P < 0.01) after the pen t r i a l s were completed, although no differences existed for pel l e t e d dairy ration or pelleted a l f a l f a . Apple intake rate of the female in the s t a l l t r i a l s was lower (P = 0.04) for the three post-pen t r i a l s . For the female, intake rate was p o s i t i v e l y correlated (r_ = 0.87, P = 0.002, n = 9) with the weight of apples consumed in a t r i a l . This was not the case (r = 0.42, P = 0.257, n = 9) for the male. Intake rate was not related to the duration of food deprivation for either animal (both P > 0.10). Food deprivation also did not explain differences in weight of apples consumed for either animal (male: £ = 0.22, P = 0.564; female: r = 0.39, P = 0.302, n = 9 for both). Inclusion of the weight of a l l foods consumed did not improve t h i s relationship. S t a l l t r i a l s suggested that both animals strongly preferred apples to the other food types a v a i l a b l e ; a preference that could be compared to pen t r i a l s . There was l i t t l e difference in the use of pelleted dairy ration and pellet e d a l f a l f a , although, p a r t i c u l a r l y for the female, pellet e d a l f a l f a appears to rank lower in terms of preferred foods under ad libitum conditions. i Pen t r i a l s Thirty-eight pen t r i a l s were completed and analysed: 13 t r i a l s with the female and 25 with the male. Three t r i a l s with the female were deleted, and she was not used following herbicide re-application. 97 Variation in intake rates within and between t r i a l s , differences between proportions, of food seen and food available in the enclosure, and changes in the amount of apples eaten per platform with changing apple density, a l l relate d i r e c t l y or i n d i r e c t l y to a number of hypotheses. For t h i s reason they w i l l be treated before o v e r a l l density and dispersion effects and individual hypotheses are addressed. Variation in intake rates Intake rates varied with time into the t r i a l for both animals. To enable comparisons of consumption between food types and differences in the rates of consumption of foods throughout t r i a l s , intake rates were estimated for 0-20 and 20-60 min intervals for pelleted dairy ration and pelleted a l f a l f a . Intake rates (g/s) for the male (Table 3.7) were found to be s i g n i f i c a n t l y higher during the 0-20 min i n t e r v a l than later in the t r i a l , for both pelleted dairy ration and p e l l e t e d a l f a l f a . Only one* value with: a l f a l f a , was- obtained with the female: 0.91 g/s. No difference (P > 0.10, n = 7) existed in the female's intake rate of pelleted dairy ration before and aft e r 20 min of observation and data were therefore pooled: 1.25 ± 0.50 (x ± SD, n = 7). These two values for the female, and the mean values presented in Table 3.7, were used to estimate the weight of food consumed from the observed handling time for a l l t r i a l s . Substantially greater use of apples during pen t r i a l s enabled apple intake rates to be compared among shorter time 98 Table 3 . 7 . Comparison of consumption rates of pelleted dairy ration and pellet e d a l f a l f a between 0-20 and 20-60 min interv a l s for t r i a l s with the male. Presented means are based on measured consumption (g) at platforms v i s i t e d only once during time intervals (s). P r o b a b i l i t i e s at right of table indicate results of analysis of variance between time i n t e r v a l s . Food Time Interval (min) n X SD P (g/s) Dairy ration 0-20 * 8 1 .58 0.67 0.016 20-60 1 6 0.93 0.52 A l f a l f a 0-20 4 0.96 0.31 0.038 20-60 10 0.60 0.25 99 int e r v a l s . The time required by the deer to eat four pieces (=36 g tota l ) of apple varied, s i g n i f i c a n t l y with time into the t r i a l ( a l l P < 0.001; Table 3.8). Variances were not homogeneous ( a l l P < 0.001) between intervals; the variance increased with the mean. Sig n i f i c a n t increases in the handling time per apple eaten later in the t r i a l (Table 3.8) indicated, not only an increased mean handling time, but also that animals might have foraged less intensely. For both animals, time required to handle =36 g of apples was s i g n i f i c a n t l y shorter during the f i r s t 5 min than during the remainder of a t r i a l (Table 3.8). Handling time was biased towards p e l l e t s when used to compare apple and p e l l e t consumption. Intake rates of apples of 2.22 and 1.88 g/s for the male and female, respectively (from Table 3.8 based on 36 g per platform) were higher than average values for pelleted foods (Table 3.7). This means handling time overestimated consumption of p e l l e t s when compared d i r e c t l y to apples. Estimates of weights of each food consumed were therefore used to make ove r a l l comparisons among foods. Although intake rates were higher at the start of t r i a l s , i n i t i a l intake rates were not related to the duration of food deprivation ( a l l P > 0.10). Intake rates (g/s handling) were similar between s t a l l and pen t r i a l s . There were no differences for either animal for pelleted a l f a l f a (P > 0.05, n = 22 [male] and 4 [female]). Male intake rate- was- also not di f f e r e n t for pelleted dairy ration (P > 0.05, n = 31), but intake rates of the female were greater (P < 0.01, n = 13) in pen than in s t a l l t r i a l s . Table 3.8. Comparison of mean time r e q u i r e d to consume =36 g of a p p l e s (one p l a t f o r m ) d u r i n g 0-5-10, 10-20, and 20-60 min i n t e r v a l s f o r male and female b 1 a c k - t a i 1 e d deer. Means s h a r i n g a common under s c o r e were not s i g n i f i c a n t l y d i f f e r e n t by S c h e f f e ' s t e s t (a = 0.10). Mean v a l u e s a r e p r e s e n t e d i n seconds. Time p e r i o d (min) Subject 0-5 5-10 10-20 20-60 X ± S D n X ± S D n X ± S D n X + SD n Male 16.2 5.4 210 18.7 6.7 125 21.3 7.8 161 19.7 6.4 104 a Female 19.1 5.8 71 24.0 7.1 61 25.8 10.5 53 28.9 9.3 87 a mean v a l u e f o r 10-20 min i n t e r v a l s i g n i f i c a n t l y g r e a t e r than 5-10 and 20-60 min i n t e r v a l s 101 Average intake rates for apples (g/s), during pen t r i a l s (based on t o t a l weight/total handling, time), were not d i f f e r e n t than s t a l l intake rates (P > 0.05), for either animal. Food encountered versus food available S i g n i f i c a n t differences between the frequency of food types encountered' (seen) and food in the enclosure were found for only the male and then only for clumped apple d i s t r i b u t i o n s with medium-low and low apple densities (Table 3.9). Frequencies of foods encountered did not d i f f e r from t o t a l a v a i l a b i l i t y for any t r i a l with the female, dispersed t r i a l s with the male, or clumped apple d i s t r i b u t i o n t r i a l s with medium-high or high apple density with the male (contingency table analysis; a l l P > 0.10). For these t r i a l s , the female's search path accurately sampled the food d i s t r i b u t i o n . Clearance of apple platforms Deer did not always eat a l l pieces of apples when platforms containing apples were f i r s t encountered, even though deer continued eating apples at other locations. Subsequent v i s i t s to the i n i t i a l platform removed the remaining apple pieces. The male ate a l l apples during a l l f i r s t v i s i t s to platforms during only seven t r i a l s : two with dispersed-low d i s t r i b u t i o n , one each with a l l others excluding the high-density d i s t r i b u t i o n . The female removed a l l apple pieces during her only encounters with dispersed-medium-low and dispersed-low d i s t r i b u t i o n s . Table 3 . 9 . R e s u l t s o f c o m p a r i s o n s o f f r e q u e n c i e s o f a l l a v a i l a b l e p l a t f o r m s a n d c u m u l a t i v e f r e q u e n c y o f a p p l e s w i t h i n 5 m o f t h e m a l e ' s s e a r c h p a t h f o r c l u m p e d low a n d c l u m p e d m e d i u m - l o w d i s t r i b u t i o n s . P r o b a b i l i t i e s a r e b a s e d o n c o n t i n g e n c y t a b l e a n a l y s i s . No d i f f e r e n c e s w e r e f o u n d f o r a n y t r i a l s w i t h t h e f e m a l e . T r i a l n u mber Number o f A p p l e P 1 a t f o r m s D i s t r i b u t i o n P o f no d i f f e r e n c e b e t w e e n ' s e e n ' a n d a v a i 1 a b 1 e t o 5 m i n 10 m i n 15 m i n 20 m i n 6 0 m i n 9 13 25 34 36 38 4 0 27 18 27 27 18 27 18 CML C LC CML CML CL CML CL ns 0.005 * * ns n s 0 . 0 0 0 * * 0 . 0 0 0 * * 0 . 0 1 7 * n s 0 . 0 0 3 * * n s n s 0 . 0 0 0 * * 0 . 0 0 3 * * n s n s ns 0 . 0 2 5 * n s 0 . 0 0 0 * * 0 . 0 0 5 * * 0 . 0 1 5 * ns 0 . 0 0 5 * * 0 . 0 0 6 * * n s 0 . 0 0 2 * * 0 . 0 0 3 * * ns n s 0 . 0 1 4 * n s n s 0 . 0 1 6 * 0 . 0 0 8 * * n s f C1umped-mediurn-1ow a p p l e d i s t r i b u t i o n . ° P > 0.05 C l u m p e d - l o w a p p l e d i s t r i b u t i o n . * * P < 0.01 * P < 0.05 1 03 The male l e f t more apple platforms uncleared per number of apple platforms v i s i t e d and per number of apple platforms seen when apples were abundant (Table 3.10). Due to limited sample size (n = 25 in a l l ) , t r i a l s with the female were pooled to high (72 and 45 apple platforms) and low (27 and 18 apple platforms) d e n s i t i e s . The female also l e f t more platforms uncleared per number v i s i t e d (P = 0.037, n = 25) and per number seen (P = 0.019, n = 25) at high density. As might be expected, the number of apple platforms seen was p o s i t i v e l y correlated with the number of apple platforms v i s i t e d for the male (r = 0.989, P < 0.01, n = 25) and female (r = 0.949, P < 0.01, n = 13). The fact that the correlations were not perfect indicated either that not a l l apples were seen within the 5-m detection distance, or that they were seen and not selected. For both animals, duration of food deprivation was not related to the clearance of apple platforms (for a l l P > 0.15). E f f e c t s of searching on se l e c t i o n : Hypothesis 3 Hypothesis 3 was not rejected: when apples were abundant, pen t r i a l s were e s s e n t i a l l y ad libitum and selection did not d i f f e r from preference. To 5 and 10 min into almost a l l t r i a l s , apples were the most consumed food by both animals (Tables 3.11 and 3.12). Exceptions included: the f i r s t 5 min of the f i r s t female t r i a l in which a l f a l f a was consumed longer than apples', and to 10 min' during the female'& only encounter with the clumped low d i s t r i b u t i o n . The femaie selected apples more than other food types (Table 3.11) based on t o t a l handling 104 ' Table 3.10 Comparison of the p r o p o r t i o n of apple p l a t f o r m s v i s i t e d but not c l e a r e d of apples by the male. T r i a l s were pooled by number of a v a i l a b l e apples, i g n o r i n g d i s p e r s i o n . U n d e r l i n e d means were not s i g n i f i c a n t l y d i f f e r e n t by S c h e f f e ' s t e s t . Mean values are presented. Measure Number of apples a v a i l a b l e P of no d i f f e r e n c e s 72 48 27 18 # not c l e a r e d # v i s i t e d 0. 185 0.033 0.039 0.032 0.038 # not c l e a r e d # seen 0.208 0.036 0.043 0.034 0.023 a # v i s i t e d r e f e r s to apple p l a t f o r m s o n l y . k # seen re p r e s e n t s number of apple p l a t f o r m s passed w i t h i n 5 m. Table 3.11. S e l e c t i o n o f f o o d t y p e s b y a f e m a l e b l a c k - t a i l e d d e e r d u r i n g 1 9 8 3 p e n t r i a l s b a s e d o n c o m p a r i s o n s o f c u m u l a t i v e t i m e h a n d l i n g e a c h f o o d t y p e t o 5 , 1 0 , a n d 6 0 m i n . S i g n i f i c a n c e o f t e s t s o f e q u a l i t y f o r e a c h p e r i o d a r e s h o w n b e s i d e m o s t c o n s u m e d f o o d i t e m s . C o n s u m p t i o n o f a l f a l f a w a s r a r e a n d i s n o t p r e s e n t e d i n t h i s t a b l e a l t h o u g h t h i s f o o d w a s i n c l u d e d i n s t a t i s t i c a l c o m p a r i s o n s . D i s t r i b u t i o n s a r e a b b r e v i a t e d a s f o l l o w s : C H - c l u m p e d h i g h ; C M H - c l u m p e d m e d i u m h i g h ; C M L - c l u m p e d m e d i u m l o w ; a n d C L - c l u m p e d l o w . A b b r e v i a t i o n s b e g i n n i n g w i t h a D a r e d i s p e r s e d i n s t e a d o f c l u m p e d . C u m u l a t i v e T i m e H a n d l i n g ( s ) f r o m s t a r t o f t r i a l N u m b e r o f T r i a l # D i s t r i b u t i o n A p p l e 0 - 5 m i n 0 - 1 0 m i n 0 - 6 0 m i n P l a t f o r m s A p p l e s D a i r y R a t i o n A p p l e s D a i r y R a t i o n A p p l e s D a i r y R a t i o n 2 C H 7 2 4 7 a 5 1 4 0 * * 2 6 9 1 3 * * 8 7 4 D H 7 2 1 8 4 * * 0 3 4 2 * * 1 5 8 2 3 * * 5 9 6 C M H 4 5 1 2 0 * * 2 0 2 0 9 * * 2 0 5 8 1 * * 9 4 8 D M H 4 5 1 0 5 * * 4 4 2 1 1 * * 6 6 9 7 9 * * 2 5 3 1 0 C M L 2 7 1 5 7 * * 1 8 2 8 7 * * 2 4 5 3 9 * * 7 0 1 2 D M L 2 7 1 2 8 * * 3 7 3 0 1 * * 4 5 4 7 0 * * 4 5 1 4 C L 1 8 9 8 * * 0 2 9 2 * * 0 4 6 8 * * 4 8 1 6 D L 1 8 7 -j * * 2 7 7 1 1 5 6 * * 3 8 8 * * 1 7 3 1 8 C H 7 2 2 6 9 * * 0 4 6 6 * * 0 1 2 2 2 * * • 0 2 0 D H o 7 2 1 6 3 * * 0 2 7 9 * * 0 7 4 5 * * 1 0 2 2 C M H 4 5 1 1 6 * * 1 5 2 7 9 * * 2 3 6 2 5 * * 2 3 2 4 D M H 4 5 1 6 5 * * 0 3 2 7 * * 0 8 7 2 * * 0 2 6 C M L 2 7 1 5 2 * * 0 2 3 0 * * 0 4 9 5 * * 0 a A l f a l f a w a s e a t e n f o r 6 6 s w h i c h w a s n o t s i g n i f i c a n t l y g r e a t e r t h a n a p p l e s d u r i n g t h i s p e r i o d . * P < 0 . 0 1 Table 3.12. S e l e c t i o n of f o o d types by a male b l a c k - t a i l e d deer d u r i n g 1983 pen t r i a l s based on comparisons of c u m u l a t i v e time h a n d l i n g each food type to 5, 10, and SO min. S i g n i f i c a n c e of t e s t s of e q u a l i t y f o r each p e r i o d are shown b e s i d e most consumed food items. Consumption of a l f a l f a was r a r e and i s not p r e s e n t e d a l t h o u g h i t was i n c l u d e d i n s t a t i s t i c a l comparisons. See T a b l e 3.11 f o r d i s t r i b u t i o n a b b r e v i a t i o n s . ' Cumulative Time H a n d l i n g ( s ) from s t a r t of t r i a l Number of T r i a l tt D i s t r i b u t i o n Apple 0-5 min 0-10 min 0-60 min P l a t f o r m s Apples D a i r y R a t i o n Apples D a i r y R a t i o n Apples D a i r y R a t i o n 1 CH 72 126** 0 126** 0 375** 143 3 DH 72 186** 0 244** 0 457** 0 5 CMH 45 68** 0 68** 0 191** 0 7 DMH 45 122** 0 262** 6 314** 6 9 ' CML 27 129** 24 \ 72* * 24 556** 83 1 1 DML 27 147** 0 220** 40 436** 196 13 CL 18 183** 7 222** 53 327 433 15 DL 18 87** 40 159** 61 277ns 263 17 CH 72 235** 0 447** 0 1174** 0 19 DH 72 230** 0 4 \ 1 * * 0 840** 0 21 CMH 45 174** 0 323** 0 688** 0 23 DMH 45 218** 0 363** 0 777** 138 25 CML 27 158** 0 248** 12 436** 225 27 CH 72 220** 6 306** 6 557** 6 29 DH 72 195** 9 385** 9 600** 9 31 CMH 45 100** 0 262** 0 676** 0 33 DMH 45 179** 0 304** 0 662** 0 34 CML 27 152** 0 322** 0 4 -j 2** 159 35 DML 27 88** 3 2 10** 25 420** 25 36 CL 18 223** 0 302** 15 319** 35 37 DL 18 102** 6 226** 12 359** 12 38 CML 27 178** 0 365** 0 570** 139 39 DML 27 1 16** 0 245** 0 507** . 68 40 CL 18 189** 0 189** 42 313** 123 41 DL 18 1 14** 0 157** 4 215** 58 * * P < 0.01 107 time for the t r i a l . Pelleted dairy ration was selected more during the male's f i r s t exposure to the clumped-low d i s t r i b u t i o n (Table 3.12; 0-60 min). Apples were not selected more than pelleted dairy ration during the following dispersed-low d i s t r i b u t i o n t r i a l with the male. However, apples were not abundant in the low apple d i s t r i b u t i o n s , therefore hypothesis 3 was not rejected. The female selected apples to the same extent whether selection was considered on a cumulative (Table 3.11) or non-cumulative (Table 3.13) basis. Non-cumulative selection was examined by p a r t i t i o n i n g t r i a l s into 0-5, 5-10, and 0-60 min i n t e r v a l s . The s i m i l a r i t y between cumulative and non-cumulative selection indicated apples were highly selected during nearly a l l intervals and when they were not (5-10 min of t r i a l 16; Table 3.13), s h i f t s to other foods were dramatic. Food selection by the male varied much more within t r i a l s than selection by the female. The male consumed more pellet e d food during the 5-10 min i n t e r v a l (Table 3.14) during two of the three times he was presented with the clumped-low d i s t r i b u t i o n . On high and medium-high densities, apples were the food most consumed by the male from 10-60 min. Only on clumped medium-low, and clumped and dispersed low apple d i s t r i b u t i o n s did the male's handling time for apples not exceed that for other foods. However, the presence of one of these three l a t t e r d i s t r i b u t i o n s did not necessarily mean that apples would not be the-most selected food type (Table 3.14). Table 3.13. Comparison of s e l e c t i o n of food types by a female b l a c k - t a i l e d deer between 0-5, 5-10, and 10-60 min I n t e r v a l s d u r i n g 1983 pen t r i a l s based on comparisons of time h a n d l i n g each f o o d type. S i g n i f i c a n c e of t e s t s of e q u a l i t y of s e l e c t i o n are shown next to most consumed f o o d item. P r o b a b i l i t y at r i g h t of t a b l e i n d i c a t e s l i k e l i h o o d of equal s e l e c t i o n between t r i a l p e r i o d s . See T a b l e 3.11 f o r d i s t r i b u t i o n a b b r e v i a t i o n s . T r i a l tt D i s t r i b u t i o n Number of Appl e P l a t f o r m s H a n d l i ng ( s ) from s t a r t of t r i a l p 0-5 min 5-10 min 10-60 m i n Apples Rat ion A l f a l f a Apples Rat i on A l f a l f a Apples Rat i on A l f a l f a 2 CH 72 47 5 6 6 n S 93** 21 1 773** 61 1 0 .00 4 DH 72 184** 0 0 158** 15 0 481** 44 0 0 .00 6 CMH 45 120** 20 0 89** 0 O 372** 74 14 0 .00 8 DMH 45 105** 44 0 106** 22 7 452** 77 27 0 .00 10 CML 27 157** 18 0 130** 6 0 252** 46 38 0 .00 12 DML 27 128** 37 0 173** 8 0 169** 0 0 0. .00 14 CL 18 98** 0 5 194** 0 0 176** 48 0 0. .00 16 DL 18 71** 27 0 0 129** 3 317** 17 10 0, ,00 18 CH 72 269** 0 0 197** 0 O 756** 0 0 0. 81 20 DH 72 163** 0 0 116** 0 0 466** 10 0 0. 55 22 CMH 45 1 16** 15 0 163** 8 0 346** 0 0 0. 00 24 DMH 45 165** 0 0 162** 0 0 545** 21 9 0. 03 26 CML 27 152** 0 0 78** 0 0 265** 0 0 0. 90 P < 0.01 Table 3.14. Comparison of s e l e c t i o n of food types by a male b l a c k - t a i l e d deer between 0-5, 5-10, and 10-60 min i n t e r v a l s d u r i n g 1983 pen t r i a l s based on comparisons of time h a n d l i n g each f o o d type. S i g n i f i c a n c e of t e s t s of e q u a l i t y of s e l e c t i o n a re shown next to most consumed f o o d item. P r o b a b i l i t y at r i g h t of t a b l e i n d i c a t e s l i k e l i h o o d of equal s e l e c t i o n between t r a i l p e r i o d s . See T a b l e 3.11 f o r d i s t r i b u t i o n a b b r e v i a t i o n s . H a n d l i n g ( s ) from s t a r t of t r i a l Number of T r i a l D i s t r i b u t i o n Apple 0-5 min 5-10 min 10-60 min H P l a t f o r m s Apples R a t i o n A l f a l f a Apples R a t i o n A l f a l f a Apples R a t i o n A l f a l f a 1 CH 72 126** 0 0 0 0 0 249** 143 19 0 .00 3 DH 72 186** 0 0 58** 0 0 213** 0 0 0 .86 5 CMH 45 68** 0 0 0 0 0 123** 0 0 0 .43 7 DMH 45 122** 0 0 140** 6 0 52* * 0 0 0 . 52 9 CML 27 129** 24 10 43** 0 0 384** 59 6 0 .00 1 1 • DML 27 147** 0 16 73** 40 0 266** 156 1 1 0 .00 13 CL 18 183** 7 0 39 46ns 24 105 380** 38 0 .00 15 DL 18 87** 40 27 72** 21 7 1 18 202** 94 0 .00 17 CH 72 235** 0 0 2 12* * 0 0 727** 0 0 0 .82 19 DH 72 230** 0 0 181** 0 0 429** 0 0 0, .96 21 CMH 45 174** 0 0 149** 0 0 365** 14 25 0 .00 23 DMH 45 218** 0 0 145** 0 0 414** 138 228 0. ,00 25 CML 27 158** 0 3 90** 12 15 188 213ns 152 0, ,00 27 CH 72 220** 6 0 86** 0 13 251** 0 0 0. .00 29 DH 72 195** 9 7 190** 0 9 215** 0 0 0, ,00 31 CMH 45 100** 0 4 162** 0 0 414** 0 17 0. , 20 33 DMH 45 179** 0 4 125** 0 0 358** 0 10 0. .70 34 CML 27 152** 0 0 170** 0 0 90 159 176ns 0. 00 35 DML 27 88** 3 0 1 2 2 * * 22 0 210** 0 5 0. 00 36 CL 18 223** 0 0 79** 15 0 17 20 39ns 0. 00 37 DL 18 102** 6 0 124** 6 6 133** 0 61 0. 00 38 CML 27 178** 0 0 187** 0 0 205ns 139 192 0. 00 39 DML 27 1 16** 0 0 129** 0 0 262** 68 25 0. 00 40 CL 18 189** 0 0 0 42 63ns 124** 81 14 0. oo 41 DL 18 114** 0 16 43** 4 29 58 54 63ns 0. 00 "° P > 0.05 * * P < 0.01 1 10 Effects of density and dispersion Hypothesis 7 was rejected: consumption of apples was not greater on clumped than on unclumped d i s t r i b u t i o n s . Results of two-way analyses of variance involving the e f f e c t s of apple density and dispersion, and the i r interaction are presented in Table 3.15. For four of the 16 ANOVAs conducted ( a l l t r i a l s with the male), density and/or dispersion effects were dependent on each other (Fig. 3.5). In addition to s i g n i f i c a n t interactions between density and dispersion for t o t a l time handling apples and the proportion of platforms seen that contained apples (Table 3.15), t o t a l time handling apples per m tr a v e l l e d (P = 0.062) and estimated weight of a l l food per m t r a v e l l e d (P = 0.051) were marginally non-significant. When apple platforms were abundant (72 out of 144), more apples were found per distance t r a v e l l e d (and more t o t a l weight of food consumed) than when apples were dispersed (Fig. 3.5a and b). For both variabiles.,- high- apple-density increased, intake per distance t r a v e l l e d (Scheffe's test) when apples were dispersed (no difference existed with clumped d i s t r i b u t i o n s ) . For a l l other densities, t r a n s i t i o n from clumped to dispersed d i s t r i b u t i o n s resulted in a an i n s i g n i f i c a n t reduction in mean intake (Fig. 3.5a and b). The e f f e c t s of density on time handling apples, and the proportion of apples seen to 5 min into the t r i a l , varied with dispersion, and dispersion e f f e c t s varied with density (Fig. 3.5c and d). E f f e c t s of density and dispersion on the Table 3.15. S i g n i f i c a n c e of two-way a n a l y s i s of v a r i a n c e t e s t i n g the e f f e c t s of a p p l e d e n s i t y , d i s p e r s i o n , and t h e i r i n t e r a c t i o n , on a number of response v a r i a b l e s . W i t h i n the t a b l e , TTH i n d i c a t e s t o t a l time h a n d l i n g and EWC i s the e s t i m a t e d weight consumed. Male Female Response V a r i a b l e S i g n i f i c a n c e of E f f e c t of S i g n i f i c a n c e of E f f e c t of D e n s i t y D i s p e r s i o n I n t e r a c t i o n 3 D e n s i t y D i s p e r s i o n I n t e r a c t i o n TTH apples/m t r a v e l l e d ( 6 0 min) 0 . 0 0 1 * * 0 . 3 0 7 0 , 0 6 2 0 . 0 1 5 * 0 . 2 6 3 0 . 2 0 2 TTH apples/m t r a v e l l e d ( 1 0 min) 0 . 0 6 0 0 . 7 8 8 0 129 0 . 124 0 . 7 6 4 O, . 4 1 8 TTH a l l foods/m t r a v e l l e d ( 6 0 min) 0 . 170 0 . 6 7 8 0 . , 0 7 6 0 . 100 0 . 334 0 , . 8 9 4 EWC a l l foods/m t r a v e l l e d ( 6 0 min) 0 . 0 0 2 * * 0 . 3 9 5 0 , , 051 0 . 0 2 3 * 0 . 2 2 3 0 , . 2 0 3 T o t a l d i s t a n c e (m) 0 . 774 0 . 6 8 3 0 . 6 6 3 0 . 9 7 7 0 . , 3 6 9 O. 3 0 9 TTH a p p l e s ( 0 - 5 min) 0 . 0 3 5 * 0 . . 261 0 . 0 1 2 * 0 . 4 8 8 0 . . 9 2 3 0 . , 4 7 6 ( 0 - 1 0 min) 0 .251 0 . , 8 4 7 0 . 394 O . 4 4 8 0 . . 7 2 5 0 . 4 2 9 ( 0 - 6 0 min) 0 . 0 4 8 * 0 . 8 5 2 0 . 9 5 0 0 . 0 1 0 * * 0 . , 5 4 3 0 . 9 5 5 ( 5 - 1 0 min) 0 . 3 0 7 0 . , 3 6 7 0 . 8 3 5 0 . 5 3 5 0 . 6 1 0 0 . 5 1 2 ( 1 0 - 6 0 min) 0 . 0 4 4 * 0 . 701 0 . 7 8 7 0 , , 0 0 6 * * 0 . 6 0 6 0 . 6 0 9 Weight of Apples consumed 0 . 0 4 9 * 0 . 851 0 . 9 3 0 0 , . 4 0 4 0 . 9 5 8 0 . 6 8 0 EWC of a l l p e l l e t e d foods o . . 0 3 4 * 0 . 3 7 2 0 . 5 2 6 0 . . 4 5 6 0 . 3 5 9 0 . 4 7 0 % Apple p l a t f o r m s seen ( 0 - 5 min) 0 . . 0 0 0 * * 0 . 0 0 5 * * 0 . 0 1 8 * 0 . , 0 0 2 * * 0 . , 3 3 4 0 . 8 2 7 ( 0 - 1 0 min) 0 . 0 0 0 * * 0 . 0 0 2 * * 0 . 314 0 . , 0 0 1 * * 0 . 481 0 . 611 ( 0 - 6 0 min) 0 . . 0 0 0 * * 0 . 0 1 8 * 0 . 9 5 2 0 , , 0 0 0 * * 0 . 411 0 . 6 9 8 P r e f e r e n c e f o r a p p l e s (dry weight) 0 . 0 0 1 * * 0 . 6 8 2 0 . 6 1 8 0 . 0 0 0 * * 0 . 3 2 5 0 . 2 2 3 a I n t e r a c t i o n e f f e c t of d e n s i t y and d i s p e r s i o n of a p p l e s . D e n s i t y p o o l e d to two c l a s s e s : h i g h and low (see t e x t ) . c P r o p o r t i o n of dry weight consumed which were a p p l e s d i v i d e d by the p r o p o r t i o n of d r y weight of a l l foods i n the pen which were a p p l e s . / 1 1 2 TIME HANDLING APPLES TO 5 MIN. PROPORTION OF APPLE PLATFORMS * SEEN' TO 5 MIN. CLUMPED DISPERSED CLUMPED DISPERSED F i g u r e 3.5. G r a p h i c a l r e p r e s e n t a t i o n of the i n t e r a c t i o n e f f e c t s of ap p l e d e n s i t y and d i s p e r s i o n f o r the male. The c a p t i o n above each p l o t r e p r e s e n t s the response v a r i a b l e t e s t e d . Apple d e n s i t y i s r e p r e s e n t e d by HD ( h i g h d e n s i t y ) , MHD (medium-high d e n s i t y ) , MLD (medium-low d e n s i t y ) , , and LD (low d e n s i t y ) . 'Seen* i n d i c a t e s the p l a t f o r m was w i t h i n 5 m of the a n i m a l ' s s e a r c h p a t h . 1 13 four variables in Figure 3.5 interacted. Table 3.16 presents mean values for variables that showed s i g n i f i c a n t density e f f e c t s (excluding results with interactions) during male and female t r i a l s . The female ate more of a l l foods per distance t r a v e l l e d at higher density. Although t o t a l time handling apples for the 60-min t r i a l was s i g n i f i c a n t l y higher with increasing apple density, for both animals, the e f f e c t of apple density was more apparent l a t e r in a t r i a l . As a r e s u l t , s i g n i f i c a n t effects of density on the t o t a l time handling apples from 10-60 min of the t r i a l for' male and female subjects contributed to the former e f f e c t . Density affected the proportion of apple platforms seen by both animals. Although apple dispersion did not a f f e c t the proportion of apples seen for the female, i t did for the male. Si g n i f i c a n t interaction between density and dispersion to 5 min (Fig. 3.5) precluded further examination, but to 10 and 60 min of male t r i a l s , both density and dispersion affected what was seen. In both these cases, clumping increased the mean proportion of platforms seen that contained apples: 0.35 ± 0.01 (x ± SD) versus 0.28 ± 0.02 to 10 min, and 0.31 ± 0.01 compared to 0.27 ± 0.01 to 60 min for clumped and dispersed d i s t r i b u t i o n s . Density had no ef f e c t on the t o t a l time handling a l l food per distance t r a v e l l e d or the t o t a l distance t r a v e l l e d for either animal (Table 3.15). There was also no e f f e c t of density or dispersion on the- estimated t o t a l weight of, a l l foods consumed by either animal (P > 0.05). Density did affect both the weight of apples and combined weight of other foods Table 3.16. Comparison of s t r a t a means f o r v a r i a b l e s i n which two-way a n a l y s i s of v a r i a n c e found d e n s i t y to be a s i g n i f i c a n t e f f e c t . Note t h a t f o r female t r i a l s h i g h and medium-high, and low and medium-low d e n s i t i e s were poole d . Mean v a l u e s s h a r i n g the same un d e r s c o r e were not s i g n i f i c a n t l y d i f f e r e n t by S c h e f f e ' s t e s t . A b b r e v i a t i o n s and s i g n i f i c a n c e s a r e g i v e n i n T a b l e 3.15. Subj e c t Var i a b l e High Apple D e n s i t y Med i um-h i gh Med i urn-1ow Low Male TTH a p p l e s (0-60 min) ( s ) (10-60 min) ( s ) TTH ap p l e s / m a (g/m) EWC a l l foods/m a (g/m) Weight of a p p l e s consumed (g) Weight of p e l l e t s ' 3 consumed (g) 667 . 2 347 . 3 1 .07 2 .06 1250.4 23 .9 551 . 3 287.7 0. 73 1 . 37 1024.4 36 . 5 483.9 229 . 3 0. 56 1 . 20 906.8 128.3 301 . 7 92 . 5 0.41 0.96 592.9 154 . 3 Female P r o p o r t i o n of p l a t f o r m s (0-10 min) seen (0-60 min) P r e f e r e n c e f o r a p p l e s 0. 52 0. 52 8.3 0.31 0.31 16.9 0. 26 0.20 22 . 7 TTH apples/m (g/m) 1 .1 C 0. .6 EWC a l l foods/m (g/m) 1 .6 1 . 0 TTH a p p l e s (0-60 min) (s) 805 .5 472 . 0 (10-60 min) (s) 523 .9 235 . 8 P r o p o r t i o n of p l a t f o r m s (0-5) 0. .42 O. . 18 seen (0-10 min) 0. .43 0. . 19 (0-60 min) 0. 43 0. 17 P r e f e r e n c e f o r a p p l e s 10. 9 31 . 1 0. 20 0.15 2 8 . 1 i n t e r a c t i o n p r e s e n t e d i n F i g u r e 3.5. e s t i m a t e d combined weight of p e l l e t e d d a i r y r a t i o n and p e l l e t e d a l f a l f a , d e n s i t i e s p o o l e d f o r doe t r i a l s due to sample s i z e s . 115 eaten by the male. As the density of apples declined, there was a decrease in the weight of apples eaten and an increase in use of the other foods. Although mean weight of apples eaten by the female declined from 892.0 ± 461.0 g (x ± SD, n = 11) to 694.5 ± 101.8 g (n = 5) and combined mean weight of p e l l e t s increased from 39.3 ± 44.4 (n = 11) to 57.7 ± 54.4 g (n = 5), between pooled high and low apple density, the large variance precluded s t a t i s t i c a l significance ( a l l P > 0.30). Preference Ratios For both animals decreases in the density of apples in the enclosure increased the preference r a t i o for apples (on a dry-weight ba s i s ) . Apples represented 10.9% of the available dry weight (DW) at high apple density, and only 1.7% at low densities. Consequently, apples were never avoided ( i . e . preference r a t i o always > 1.0). It could be argued that s i g n i f i c a n t differences in the r a t i o above 1.0, do not make much difference'. However, the r e l a t i v e ranking- of- foods must be considered. Pelleted dairy ration made up 52.3% of the available dry weight (DW) of food at low density (48.2% based on wet weight). On a wet weight basis, both animals avoided pelleted dairy ration during a l l t r i a l s (weight used/weight available < 1.0). Considering dry weight u s e r a v a i l a b i l i t y r a t i o s , pelleted dairy ration was selected during t r i a l s 13 (ra t i o 1.27) and 15 (1.05) with the male, and t r i a l 16 (1.13) with the female. Both animals avoided pelleted dairy ration during other t r i a l s . For 1 16 the same t r i a l s , the preference ratios for apples on a dry weight basis were 15.8, 16.8, and 22.4, respectively. These data indicated that apples were preferred much more than pellet e d dairy ration. However, during t r i a l 13 (clumped-low apple d i s t r i b u t i o n with the male) t o t a l time handling foods favoured pell e t e d dairy ration (Table 3.8). Based on consumption rates for pelleted dairy ration (Table 3.11), 318 g of ration were eaten (=275 g DW) compared to =524 g (=74 g DW) of apples during t r i a l 13. Preference ra t i o s for wet weights during the same t r i a l were 7.0 for apples and 0.54 for pelleted dairy ration. Consumption of non-preferred foods: Hypothesis 8 Deer ate no more non-preferred food per station v i s i t e d at lower apple density so hypothesis 8 was rejected. T r i a l s were partitioned into 0-20 and 20-60 min intervals and estimated weight of p e l l e t s (pelleted dairy ration and pelleted a l f a l f a treated- separately) per platform v i s i t e d computed. A three-way analysis of variance using density, dispersion, time i n t e r v a l , and their interactions was used to examine food consumption by platform v i s i t . There were no effects of density, dispersion, or time into a t r i a l on the weight of pelleted dairy ration consumed per platform for either animal ( a l l P > 0.10). The male ate more pelleted a l f a l f a per platform later in t r i a l s (20-60 min) although data were highly variable. His a l f a l f a consumption increased from 4.7 ± 4.5 (x ± SD) to 55.7 ± 46.9 g after 20 min into the t r i a l (after log transformation P < 1 17 0.001 ) . Food switching: Hypothesis 9 Hypothesis 9 (page 71) was not rejected: there was a tendency for p e l l e t intake rates to increase as apple abundance declined. Comparing cumulative apple intake rates between sets of consecutive apple bites within t r i a l s (Tables 3.17 and 3.18), no consistent trends were apparent. For both animals, although there was a tendency for intake rates at the start of t r i a l s to be higher, t r i a l s occurred with: (1) no s i g n i f i c a n t differences between cumulative rates; (2) s i g n i f i c a n t increases; and (3) decreases in the estimated intake rate when a l l slopes within a t r i a l are compared. Only one t r i a l with the female and ten with the male contained more than one switch to pelleted foods (combined pelleted dairy ration and pelleted a l f a l f a ) . For the male, comparison of estimated consumption rates of non-apple rates throughout t r i a l s (Table 3.19) suggested; that intake rates of p e l l e t s (pelleted dairy ration and pelleted a l f a l f a combined) tended to be higher l a t e r in t r i a l s . T r i a l 25 (Table 3.19) was the only case in which the l a s t series of bites on pelleted foods was the greatest (not always s i g n i f i c a n t ) . For the male, the more apples seen, the longer he was l i k e l y to wait before eating p e l l e t s (r = 0.657, P < 0.01, n = 20), and also the greater number of platforms without apples that would be passed (r = 0.674, P < 0.01, n = 20). Not s u r p r i s i n g l y , the elapsed time between last apple b i t e and Table 3.17. Comparison of consumption r a t e s of a p p l e s w i t h i n and among female pen t r i a l s . A n a l y s i s of c o v a r i a n c e was used to examine d i f f e r e n c e s i n s l o p e s (time h a n d l i n g a p p l e s / e l a p s e d t r i a l time) between groups of s u c c e s s i v e a p p l e b i t e s . U n d e r s c o r e d s l o p e s d i d not d i f f e r s i g n i f i c a n t l y by a n a l y s i s of c o v a r i a n c e . See T a b l e 3.11 f o r d i s t r i b u t i o n a b b r e v i a t i o n s . T r i a l Apple P l a t f o r m s Slope of l i n e s f i t t e d to c o n s e c u t i v e b i t e s of a p p l e s .. . D i s t r i b u t i o n „ . .st „nd _ r d .th _ t h a Number Prese n t 1 2 3 4 5 2 CH 72 0. .46 0 .78 0 .80 0 .57 0. 95 4 DH 72 0. .74 0. . 73 0 .51 6 CMH 45 0. .51 0 . 56 8 DMH 45 0. 66 0 .58 0. .60 0. .42 10 CML 27 0. .60 0 .87 0. .65 0, .44 0. 83 12 DML 27 0. .53 0 .46 0 .44 14 CL 18 0. 50 0 .90 0. .39 b 16 DL 18 0. 38 0. .36 0. 15 18 CH 72 0. 83 0. .77 0. 44 20 DH 72 0. 61 0. .92 0. 25 22 CMH 45 0. 86 0. 57 o. 53 0. 53 24 DMH 45 0. 65 0. .76 0. 40 26 CML 27 0. 61 0. .34 a 1 s t column r e p r e s e n t f i r s t s t r i n g of apple b i t e s , 2 n d the second, and so on. b s t r d 1 and 3 s l o p e s not s i g n i f i c a n t l y d i f f e r e n t . Table 3.18. Comparison of consumption r a t e s of appl e s w i t h i n and among male pen t r i a l s . A n a l y s i s of c o v a r i a n c e was used to examine d i f f e r e n c e s i n s l o p e s (time h a n d l i n g a p p l e s / e l a p s e d t r i a l time) between groups of s u c c e s s i v e a p p l e b i t e s . U n d e r s c o r e d s l o p e s d i d not d i f f e r s i g n i f i c a n t l y by a n a l y s i s of c o v a r i a n c e . See Ta b l e 3.11 f o r d i s t r i b u t i o n a b b r e v i a t i o n s . T r i a l Apple P l a t f o r m s Slope of l i n e s f i t t e d to c o n s e c u t i v e b i t e s of a p p l e s .. . D i s t r i b u t i o n _, . .st _nd _ r d .th ,-th a Number Present 1 2 3 4 5 1 CH 72 0 .71 O . 93 0 .72 b 0 . 52 3 DH 72 0 . 76 0 . 79 0 . 75 5 CMH 45 0 .63 0 . 84 7 DMH 45 0 .58 0 . 46 9 CML 27 0. .63 0. .62 0 . 96 0 .70 0. 72 1 1 DML 27 0, , 62 0. . 36 0 .87 0. .51 0.51 13 CL 18 0. . 76 0. . 56 15 DL 18 0. . 55 0. . 76 0. .40 17 CH 72 0. .71 0. . 39 0. .40 0. .41 19 DH 72 0. .66 0. .43 21 CMH 45 0. . 52 0. .43 0. .90 23 DMH 45 0. .62 0. . 29 0. . 42 60 b 25 CML 27 0. 59 0. 67 0. 91 0. 27 CH. 72 0. 76 0. 69 0. 74 0. 46 29 DH 72 0. 69 0. 72 0. 83 0. 41 31 CMH 45 0. 72 0. 52 0. 53 33 DMH 45 0. 60 0. 53 0. 35 35 DML 27 0. 51 0. 33 0. 44 3G CL 18 0. 84 0. 82 3 0 b 37 DL 18 0. 37 O. 72 0. 39 DML 27 0. 36 0. 58 0. 39 b 40 CL 18 0. 61 0. 83 41 DL 18 0. 52 0. 52 a 1 s t column r e p r e s e n t f i r s t s t r i n g of apple b i t e s , 2 n c' the second, and so on. b a l l u n d e r s c o r e d s l o p e s were not s i g n i f i c a n t l y d i f f e r e n t . Table 3.19. Comparison of consumption r a t e s of combined non-apple foods f o r the male. U n d e r l i n e d s l o p e s (time h a n d l i n g p e l l e t s / e l a p s e d time) were not s i g n i f i c a n t l y d i f f e r e n t by a n a l y s i s of c o v a r i a n c e . Groups of b i t e s were d e l i m i t e d by a p p l e consumption (see t e x t ) . D i s t r i b u t i o n a b b r e v i a t i o n s a r e g i v e n i n T a b l e 3.11. T r i a l Apple P l a t f o r m s Slope of l i n e s f i t t e d to c o n s e c u t i v e non-apple b i t e s ., . D i s t r i b u t i o n „ . .st _nd _ r d .th _ t h Number Prese n t 1 2 3 4 5 11 DML 27 0 .23 0 .21 0 . 78 13 CL 18 0, , 30 0, , 24 0. . 26 0.79 15 DL 18 0. .50 0. , 36 0. .21 0.31 23 DMH 45 0. . 17 0. . 35 0. .21 25 CML 27 0. 65 C 0. 21 0. 23 0.29 34 CML 27 0. 41 0. .49 0. 80 38 CML 27 0. 21 0. .22 0. . 53 0.81 39 DML 27 0. 08 0. 42 40 CL 18 0. 31 0. 12 0. 83 41 DL 18 0. 31 0. 60 b 1 column r e p r e s e n t f i r s t s t r i n g of a p p l e b i t e s , 2 the second, and so on. f i r s t and f i f t h s l o p e not s i g n i f i c a n t l y d i f f e r e n t . f i r s t non-apple s t r i n g of b i t e s d i d n ' t occur u n t i l 17 min i n t o t r i a l . 121 f i r s t non-apple bite, and the number of a l l platforms passed were highly correlated (r = 0.875, P < 0.01, n = 20). The longer the male had been feeding on apples, the longer he was l i k e l y to search before eating p e l l e t s (r = 0.672, P < 0.01, n = 20). Again, time into the t r i a l and number of apples seen were highly related (r = 0.928, P < 0.01, n = 20). There were no correlations between the f i r s t non-apple bite and the number of platforms i n i t i a l l y containing apples in the pen (P > 0.30, n = 20). For the female, elapsed time from the l a s t apple bite increased with the number of apples seen (r_ = 0.688, P < 0.05, n = 11). However, elapsed time and number of platforms passed were also related (r_ = 0.720, P < 0.05, n = 11). Because the female rarely switched to p e l l e t s , factors r e l a t i n g to the f i r s t switch to p e l l e t s could only be examined for the male. There were t r i a l s in a l l but the clumped-low apple d i s t r i b u t i o n in which switching did not occur for the male. Elapsed time between the l a s t apple bite and f i r s t switch to pe l l e t s was p o s i t i v e l y correlated with the proportion of apples seen (r = 0.702, P < 0.01, n = 12), and with the number of apples seen (r = 0.686, P < 0.05, n = 12), but not the number of apples in the pen (r = 0.473, P > 0.05, n = 12). The time of the switch was related to apple abundance (r = 0.653, P < 0.05, n = 12), to the number of apples seen (r = 0.789, P <0.01, n=12), and to the proportion of apples seen (r = 0.622, P <, 0.05, n = 12). However, the number of platforms passed between the l a s t apple bite and f i r s t non-apple switch were not related to density, time into the t r i a l , apples seen, or even 1 22 elapsed time. Turning frequency of searching: Hypothesis 10 Hypothesis 10 was rejected. S t r a i g h t - l i n e distances t r a v e l l e d a f t e r apple bites did not decrease, but instead tended to be shorter before reaching the platforms (see Fi g . 3.4). Mean differences before and after apple bites (Ax) were not s i g n i f i c a n t by paired-t t e s t s : Ax = 2.16 m for the male (P = 0.063, n = 32) and 3.00 m for the female (P = 0.154, n = 22). Considering only cases when no bites occurred for three turns on either side of a p e l l e t e d dairy ration b i t e , there was no difference in the distance t r a v e l l e d following the bi t e : Ax = -1.69 m (P = 0.772, n = 13) for the male and Ax = -4.57 m (P = 0.176, n = 6) for the female. Discussion S t a l l T r i a l s Results of 1983 s t a l l t r i a l s demonstrated that both animals strongly preferred apples. Although their r e l a t i v e preference for pelleted dairy ration and pelleted a l f a l f a was ambiguous under the conditions I imposed, both animals selected apples much more than either kind of pelleted food. Ambiguity between the preference ranking of pelle t e d dairy ration and pelleted a l f a l f a , ; for the male, resulted f:rom low use of these foods. Neither hypotheses 1 nor 2 were rejected, as I concluded that preference rankings were consistent within and 1 23 between animals. Selection of apples was much stronger during 1983 s t a l l t r i a l s than in 1981. This increased selection may have resulted from more uniform apple qu a l i t y in 1983; v a r i a t i o n in preference for apples was reduced during t r i a l s . Weight of apples consumed also increased s u b s t a n t i a l l y from 1981 to 1983 s t a l l t r i a l s . Because deer increased their consumption of apples during the experimental period, ad libitum conditions disappeared during some of the s t a l l t r i a l s , r esulting in s h i f t s in selection of apples within these t r i a l s . Age (and thus size differences), in addition to apple q u a l i t y , may have explained increases in apple weight consumed between years: maximum of =650 g in 1981 (Table 2.4) and =2000 g (Table 3.4) in 1983. Within 1983 t r i a l s , the male increased consumption from =870 to =2008 g per t r i a l ; the amount eaten increased with each t r i a l . This increase was only p a r t l y due to increased apple a v a i l a b i l i t y (=1180 to =2040 g), because they sometimes l e f t apples uneaten. The female's consumption of apples was not related to experience in s t a l l t r i a l s . She ate considerably fewer apples following her exclusion from pen t r i a l s ( t r i a l s 7-9; Table 3.4) than she did during the f i r s t six t r i a l s . During the two weeks between her l a s t pen t r i a l and her next s t a l l t r i a l , apples were not available to her. Unexplained low apple consumption during t r i a l 5 precluded s t a t i s t i c a l significance between consumption before and after' pen t r i a l s . Preference among foods can be strongly influenced by experience (Arnold and Mailer 1977). Young deer raised in 124 c a p t i v i t y , on formulated diets, frequently prefer food items d i f f e r e n t from those consumed readily by wild animals (Verme and U l l r e y 1984). Consumption of apples by the male may have increased from t r i a l to t r i a l as a result of his rumen being microbially better adapted to handle apples. Although apples were highly d i g e s t i b l e , and digestion would have begun almost immediately, i t was unlikely increases in apple consumption of as much as =500 g per day can be explained by faster digestion during a 40-min feeding t r i a l . Differences in intake rates between apples and' pelleted food (both animals; Table 3.6) resulted from differences in available food size; while apple pieces weighed =9 g, individual p e l l e t s weighed =0.1-0.3 g. Differences in intake rates between animals may have been due to either size differences between animals, res u l t i n g from my use of di f f e r e n t sexes of deer, or simply due to differences in feeding behaviour. The captive deer exhibited high intake rates under ad libitum conditions. Wickstrom et a l . (1984), working with mule deer, found asymptotic grass consumption rates of 2.22 g dry matter per min (g DM/min) and s i g n i f i c a n t l y higher averages of 3.76 ± 1.18 g DM/min (± SD) for grasses, forbs, and browse in coniferous understory. Average intake rates in s t a l l t r i a l s (based on Table 3.6) were =48 and =15 g DM/min for the male on pelleted dairy ration and apples, and =16 and =10 g DM/min, respectively, for the female1. S t a l l t r i a l rates were based only on time handling and indicate that searching may have been a factor in the mule deer data (Wickstrom et a l . 1984). 125 However, in my study the minimum bite sizes available to the deer were =0.1 g DM for apples and =0.2 g DM for p e l l e t s . These are much larger than food sizes encountered by wild ungulates (Dunham 1980, Trudell and White 1981, Wickstrom et a l . 1984). Most n u t r i t i o n a l l i t e r a t u r e for ungulates i s based on dry weight of forage (g DM). In t h i s study, I used time handling and in some cases wet weight to estimate food sele c t i o n . A d ditionally, apples contained more than 6.2 times as much water than both pelleted foods. Within the 40-min s t a l l t r i a l s (or 1-h pen t r i a l s ) , processing of the food items was limited to b i t i n g and chewing; digestion was not a factor. Dry matter intake may be important when researchers compare n u t r i t i o n a l quality and forage intake on a daily basis. However, I believe that time handling foods or an estimate of weight of food consumed better describe how foraging behaviour is partitioned within feeding bouts. The p o s i t i v e c o r r e l a t i o n between the t o t a l weight of apples eaten and apple intake rate of the female, may have resulted from my use of handling time as a measure of intake rate. When weight consumed for the entire t r i a l was high, much of the t r i a l was spent eating apples. Intake of new apple pieces from the feeding buckets commenced before the l a s t bite was completely handled, thus reducing handling time. Low consumption of apples, and pausing and/or walking between bi t e s , would also decrease the 'intake- rate-' as measured. With such a small sample for the female (n = 9), her consumption of only =152 g of apples during t r i a l 7 (Table 3.5) may have 126 resulted in the s i g n i f i c a n t c o r r e l a t i o n ; no such trend was apparent for the male,. I expected duration of food deprivation to influence at least i n i t i a l intake rates, i f not t o t a l weight of food consumed, during the t r i a l . However, I found no such relationships. I could calculate intake rates only for the entire t r i a l and t h i s measure may have been insens i t i v e to i n i t i a l food deprivation responses. However, neither weight consumed to the f i r s t switch from apples, time of f i r s t pause from b i t i n g , nor duration of the t r i a l spent b i t i n g , was correlated with food deprivation. More l i k e l y , the range of food deprivation used in t h i s study was i n s u f f i c i e n t to influence deer behaviour. For fasting metabolic rates, food deprivation of =48 h has been used (Regelin et a l . 1985) and fasting durations of =7 h were used by Milchunas et a l . (1978) when they examined food digestion and rates of passage. Wickstrom et a l . (1984) considered =12 h of fasting s u f f i c i e n t to ensure 'high grazing i n t e r e s t ' . My use of the animals every day precluded fasting for more than 12-14 h unless animals were allowed access to food only during experimental periods. The l a t t e r situation might have resulted in deterioration in animal condition and would not have allowed the assumption of apple preference determined in s t a l l t r i a l s . Pen T r i a l s I could compare preference obtained from s t a l l t r i a l s to pen t r i a l selection only when food conditions were similar 127 between the two types of t r i a l s . The presence of weeds as a large component of selected foods precluded using some t r i a l s . Although t o t a l elimination of weeds during a l l t r i a l s was impossible, consumption of non-test foods during most t r i a l s was minor. During the three t r i a l s that I discarded, the female spent a large amount of time searching for weeds that were scarce and had a very small return compared to the large amounts of apples and p e l l e t s a v a i l a b l e . Whether weeds were important n u t r i t i o n a l l y or were simply novel foods, i s unknown. Both animals had access to a balanced pelleted d i e t , as well as green forage to ensure adequate rumination, so the l a t t e r explanation seems more l i k e l y . Selection of weeds may also have been analogous to sampling for changes in the environment (sensu Freeland and Janzen 1974). Also, animals usually select green herbage over dry food (Arnold and Dudzinski 1978), and the female may have preferred the weeds because 'green forage' i s normally more n u t r i t i o u s . Food s e l e c t i o n during pen t r i a l s did not d i f f e r from ad libitum preference when apples were abundant. As a result, hypothesis 3 was not rejected. With the exception of the f i r s t t r i a l run with the female, both animals strongly selected apples throughout the f i r s t 5 min of a l l t r i a l s . On a cumulative basis, apples were highly preferred during a l l t r i a l s (60 min; Table 3.11) by the female, and with one exception (the f i r s t exposure to the clumped-low d i s t r i b u t i o n ; Table 3.12), by the male-. Apples were- a 1 so selected, more than, other foods throughout a l l t r i a l s (0-5, 5-10, and 10-60 min intervals) with high and medium-high apple d e n s i t i e s . These 128 data were consistent with 1981 findings. Animals behaved d i f f e r e n t l y in response to lower apple a v a i l a b i l i t y . Apples were highly selected during a l l but one time i n t e r v a l by the female; the one exception (Table 3.13: 5-10 min) occurred with the dispersed-low apple d i s t r i b u t i o n . Apples ranked f i r s t during the subsequent 10-60 min i n t e r v a l , presumably because she again encountered apples. Although she spent less time handling apples at low density (Table 3.15), the female's selection of pelleted foods did not increase. Ranking remained the same because she spent less time eating apples when they were less abundant (Table 3.16). The male showed an increase in the use of p e l l e t s as apple density declined (Table 3.16) and as a result, he consumed more pellete d dairy ration than apples when apple a v a i l a b i l i t y was low (Table 3.14). An obstacle to comparing data on food selection with predictions of optimal foraging theory, and s p e c i f i c a l l y c o r o l l a r i e s of the marginal value theorem, was the absence of a common currency, which i s required. Predicted s h i f t s in diet selection are often based on intake rates of energy or other nutrient (MacArthur and Pianka 1966). Belovsky (1978) was able to rank foods for moose only because of a clear n u t r i t i o n a l requirement for sodium. In t h i s study, lack of a clear n u t r i t i o n a l requirement resulted in predicting preference from ad libitum conditions. Comparisons of food selection also may have been influenced by my use of handling time as a surrogate for the actual weight consumed and for intake rates. Weight of apples 1 29 consumed could be accurately estimated (based on the number of pieces consumed) from the observation tower and estimates of intake rate obtained for every platform that was v i s i t e d only once during t r i a l s . Pelleted foods were much more d i f f i c u l t to assess. As a result of time constraints, multiple v i s i t s to platforms, and limited use of some foods (especially for the female), small numbers of samples were available to estimate weight consumed from handling time. If I had used only time handling foods to estimate consumption of food among time intervals and food types, these estimates would have been biased because consumption rates d i f f e r e d among foods and t r i a l i n t e r v a l s . S i m i l a r l y , my estimates of consumption rates may have been biased because of the small samples on which they were based. Although handling time did indicate what portion of the t r i a l was spend feeding, instantaneous weight consumed would have been more appropriate, but the l a t t e r was unobtainable. Bite rates and/or intake rates are often reported in re l a t i o n to forage a v a i l a b i l i t y (Arnold and Dudzinski 1978, Trudell and White 1981, Wickstrom et a l . 1984). However changes in intake rate within foraging bouts are not well studied. My data indicated that the weight of food consumed per time handling food was higher at the start of t r i a l s . This was true for a l l foods and i t demonstrated that feeding bouts were dynamic, not only in terms of food selection (Chapter 2), but also in terms1 of intake rate. Although the relationship between food use and food a v a i l a b i l i t y has been addressed (Van Dyne and Heady 1965, 1 30 Johnson 1980), the e f f e c t of an animal depleting i t s resources on food selection has received only limited consideration (Nelson 1984). My results suggest that a number of factors, including the frequency of finding preferred food and the intake rates of non-preferred foods, were related to food selec t i o n . There were s i g n i f i c a n t relationships between intake, the amount of food already eaten, and time spent searching for preferred food without eating. The tendency for apple intake rates to decline within t r i a l s (Table 3.17 and 3.18), probably r e f l e c t e d declining apple a v a i l a b i l i t y . The duration of food deprivation before a t r i a l was not related to the degree of selection shown by either animal. There was an increase in intake rate for p e l l e t s (consecutive bites; hypothesis 9 rejected) l a t e r in t r i a l s when apples were not located. Hypothesis 9 could only be evaluated for the male. Increases in p e l l e t intake rate were not related to apple density or dispersion. This increase resulted from the male eating more pelleted a l f a l f a per platform v i s i t e d as well as an increase in the rate at which a l l pelleted platforms were v i s i t e d l ater in t r i a l s . There was a s i g n i f i c a n t relationship between the number of apple platforms encountered by both animals, and the length of time they continued searching before starting to eat p e l l e t s . Although there were no s i g n i f i c a n t relationships between elapsed time, or distance t r a v e l l e d , from the last apple bite to the f i r s t non-apple; bi t e , the• mor.e apples seen by both animals, the longer (time and distance) they were l i k e l y to continue searching for apples before taking a p e l l e t b i t e . 131 Similar relationships existed for the f i r s t food switch (three or more consecutive bites of p e l l e t s ) for the male. It could be argued that the more apples seen, the more eaten, and thus the longer an animal was prepared to search. However, the number of apples in the pen could not be shown to influence switching to p e l l e t s . Deer treated their preferred food in a similar manner to the marginal value theorem's prediction for leaving patches (Krebs et a l . 1974, McNair 1982). Both animals seemed to use a po s t e r i o r i rules, such as the time Or distance t r a v e l l e d from the l a s t bite of preferred food, to determine when to start eating less preferred foods. A l l of these results (for intake rates, switches in food selection, and forage encountered) indicated that food selection cannot be e a s i l y predicted by simple comparisons of use and a v a i l a b i l i t y (e.g., Skiles 1984). Clumping of apples did not s i g n i f i c a n t l y increase the consumption of apples or the rate at which apples were located (Table 3.15). Therefore, I rejected hypothesis 7. In fact, clumping decreased the consumption of apples by the male at high density (Fig. 3.5). At high apple density, half (72 of 144) of the platforms contained apples. Weight of apples eaten per m t r a v e l l e d was s i g n i f i c a n t l y higher at high apple densities than other densities (Fig. 3.5) when the apples were randomly dispersed. For a l l other densities, mean time handling apples per distance' t r a v e l l e d was lower for dispersed than for clumped apple d i s t r i b u t i o n s . Although density of apples s i g n i f i c a n t l y affected the 132 proportion of apple platforms seen by both animals throughout-t r i a l s (Table 3.15), apple dispersion did not a f f e c t food seen by the female. Similar differences in dispersion produced s i g n i f i c a n t differences between the proportion of platforms seen and the frequency of available foods for the male (Table 3.9). Why the female's searching accurately sampled the types of food in the pen, while the male's search path did not, was unclear. Differences in the effect of food dispersion between subjects might have been explained i f the female had t r a v e l l e d faster; in e f f e c t sampling more area per unit time. Examining t r i a l s in which no difference existed between seen and available apples, the female t r a v e l l e d farther (to 5 m) than did the male. Although the mean distance t r a v e l l e d by the female was larger, the difference was not s i g n i f i c a n t . If these non-significant differences in distance t r a v e l l e d were i n s u f f i c i e n t to explain the d i f f e r i n g e f f e c t s of dispersion, the female must have searched in a d i f f e r e n t manner, resu l t i n g in a more representative sample of food a v a i l a b i l i t y . Researchers who employ standard u s e : a v a i l a b i l i t y r a t i o s , make i m p l i c i t assumptions regarding the forager's knowledge of food d i s t r i b u t i o n . If we assume an animal has perfect knowledge of i t s environment, a r e l a t i v e l y long-lived animal may 'know' what habitats are available within i t s home-range. The investigator might then assume that the animal i s found where, i t prefers to be (McLellan 1985). Social interactions, and energetic costs of moving to the preferred s i t e , may even then influence habitat s e l e c t i o n . 1 33 At the scale of food patches or individual plants, omniscience seems unl i k e l y . Results for the male showed that food d i s t r i b u t i o n affected differences between the types of food seen by the animal and actual food abundance. Although data for the female did not support the same conclusion, the marked s h i f t to pelleted dairy ration from 5-10 min of t r i a l 16 (Table 3.13), and then back to apples from 10-60 min, probably ref l e c t e d encounter rates of apples. So, although experience with a d i s t r i b u t i o n can improve the rate of location of preferred food, the agreement between food encountered and t o t a l a v a i l a b i l i t y may depend on the degree of dispersion of food. The more clumped and/or rare s p e c i f i c foods are, the more familiar an animal may need to be with that d i s t r i b u t i o n for i t s sample of food types to approximate food abundance. Although u s e : a v a i l a b i l i t y ratios predicted preference for apples for both animals throughout t r i a l s (pen and s t a l l ) , they were s i g n i f i c a n t l y affected by food abundance in my experiments (Table 3.15). It could be argued, that to the range manager the: important aspect of a preference r a t i o i s i t s value r e l a t i v e to 1.0 (<, =, or >). However, other investigators may be interested in ranking foods r e l a t i v e to other foods (Neu et a l . 1974, Johnson 1980). The data I have presented demonstrate problems with estimating r e l a t i v e ranking, u t i l i z i n g u s e : a v a i l a b i l i t y r a t i o s . The most important food in a diet may not be the most preferred. When apples were abundant, consumption was high but the preference ratio was r e l a t i v e l y low (although s t i l l > 1.0). 134 Declining apple a v a i l a b i l i t y decreased use of apples and yet preference for apples increased. However, a simple preference r a t i o did predict apples to be the most preferred food during a l l pen t r i a l s , an outcome that was also predicted by s t a l l t r i a l s . Continuous sampling has been suggested as a way for foragers to track a changing environment (Westoby 1974, Freeland and Janzen 1974). This concept has evolved and more recently has been termed " r i s k - s e n s i t i v e foraging" (Caraco et a l . 1980, Caraco 1981 and 1982). Whether or not sampling was the explanation for my captive deer tending to leave more apple platforms uncleared when apples were abundant i s unknown. When deer are foraging they can choose to continue foraging on the same plant, move to a d i f f e r e n t plant of the same species, or move to a new forage species. Deer tend to move from one shrub to another, often of the same species, when foraging even when they could have continued eating at the same plant (pers. observ.). Mixing of d i e t s , an underlying objective of continually moving while foraging, or perhaps the avoidance of secondary plant compounds (Freeland and Janzen 1974, Lindroth 1979) are among possible explanations for t h i s behaviour. However, in my study captive deer eating highly palatable foods l e f t food on a platform, proceeded to eat the same food from another platform and then l a t e r returned to consume the remaining food from the i n i t i a l platform. This behaviour was highly related to- the abundance- of• the preferred, food and seldom occurred when apples were rare. The deer's expectation of finding a "better" food might 135 explain t h i s behaviour. Foraging by expectation of food intake has been proposed as a means of animals e f f i c i e n t l y foraging in patches (Krebs et a l . 1974, Charnov 1976b). Deer were able to return l a t e r to consume uneaten food i f no 'better' food was found. When apples were abundant, the deer could e a s i l y f i n d them, and the animals may have simply decided to see what else was available in the pen. With these data I was unable to discern between any of the above hypotheses, or the p o s s i b i l i t y that animals were simply maximizing their short-term apple intake rate. When apples were abundant, and thus the distance to the next apple platform was short, the deer could take two or three pieces of apple in one 'bite' and chew them while moving on to the next platform. When the distance between apple platforms was greater (low density), higher intake rates would be obtained by eating a l l apple pieces when a platform was detected. Although there was a tendency for handling time of apples per distance t r a v e l l e d during the i n i t i a l 10 min of t r i a l s to be greater at higher density, t h i s difference was not s i g n i f i c a n t (Table 3.15, P = 0.06). Furthermore, even a s i g n i f i c a n t difference would not prove the above speculation, because the deer may have been responding in a way consistent with one of the above hypotheses. In most cases, uneaten apples were eaten l a t e r when the deer returned to the same platform. Because t r i a l s were run in order of decreasing apple a v a i l a b i l i t y through the eight d i s t r i b u t i o n s , I could not detect whether deer were responding to apple a v a i l a b i l i t y encountered during the last t r i a l , or 136 were deciding to eat a l l apples on a platform on the basis of i n i t i a l encounter rates during a t r i a l . Conditioning to low apple a v a i l a b i l i t y might explain the increased use of apples between the f i r s t and second exposure of both animals to d i s t r i b u t i o n 1 (see Tables 3.11 and 3.12). At the start of the second set of t r i a l s , the deer may have expected apples to be rare. Crawley (1983) discussed the idea that most habitat selection operates by animals moving more slowly and turning more frequently in good habitats, than in bad. Predators often change their searching pattern in response to encounters with prey items (see Curio 1976, Begon and Mortimer 1981). They often increase their rates of turning immediately following an intake of food, which leads to predators remaining in the v i c i n i t y of their l a s t food item (Banks 1957). Zach and F a l l s (1976, 1977) found ovenbirds turned more and made shorter moves after capturing prey. However, I found no s i g n i f i c a n t differences in the turning frequency before and after successful apple b i t e s . This result may be largely an a r t i f a c t of the way that I analysed the data. Because apples were often in patches, turns between successive apple bites could not be attributed to inherent turning as opposed to the detection of another apple, and were excluded from the analysis. The fact that the animals did not turn more under dispersed conditions may simply mean that they already knew the contents of adjacent platforms. However, apple platforms were often missed by- -5 m.. It should be noted that Smith (1974a and b) and Zach and F a l l s (1976, 1977), whose studies showed that predators increased the 1 37 frequency of turning after successful capture of prey, dealt with foragers preying on food that were out of sight. If an animal's food resource i s clumped, the forager should respond by feeding more in areas where the prey i s located and less time where prey i s absent. The marginal value theorem (Charnov 1976b) states that after a l l energetic and time expenses have been accounted for, foragers should concentrate on any patch that y i e l d s a higher return than another. Increases in the frequency of turns after successful prey capture have been interpreted as behavioural responses to hunting for clumped prey. My data did not show t h i s r e s u l t , and a possible explanation i s that I excluded sequences of searching in which consecutive apples platforms were encountered. Did the forager turn more often because the presence of other prey were detected, or did i t turn in response to an expectation of finding more food? When describing searching behaviour, a researcher should c a r e f u l l y d i f f e r e n t i a t e between movements in response to expectation of prey location, and movements in response to prey that have already been detected. The s p a t i a l scale of my experiments may also have been important. Nine apple platforms spaced uniformly at 5-m intervals may not have represented a 'patch' to the deer. Instead, the problem of finding apples may be approached by the deer, as one of checking the contents of 144 platforms, and not of locating 2, 3, 4, or 6 apple 'patches'. Alternation of dispersed with clumped apple d i s t r i b u t i o n s may have reinforced the l a t t e r hypothetical behaviour. 1 38 Conclusions In the 1981 and 1983 s t a l l t r i a l s combined, both animals exhibited a consistent preference for apples. The stronger, less variable selection of apples in 1983 may have resulted from the high, consistent apple q u a l i t y . Ambiguous ranking of p e l l e t e d dairy ration and pelleted a l f a l f a during some 1983 s t a l l t r i a l s resulted from the strong selection of apples and decreased consumption of pelleted foods. High intake rates exhibited by the deer under ad libitum conditions were caused by the sizes of pieces of foods that I used, and were not comparable to other studies of deer food intake. When apples were abundant, the addition of a s p a t i a l component to the experiments did not a l t e r selection from the preference exhibited in s t a l l t r i a l s . At high apple density, apples were highly selected (hypothesis 3). Possibly as a resul t of the a r t i f i c i a l nature of the t r i a l s , and of the animals not being under pressure to forage e f f i c i e n t l y , apples continued to be strongly selected' as apple, a v a i l a b i l i t y declined. Non-rejection of hypothesis 9 indicated that the s h i f t in food selection was gradual; length of time for complete switching to p e l l e t s related to the amount of food already eaten. Even when apples were rare, the deer gradually switched to p e l l e t s , possibly indicating that they expected to find more of the preferred food. At high apple density the deer did not consume a l l avai l a b l e apples when they were f i r s t encountered. Although I could not esta b l i s h the reason for t h i s behaviour, I believe 139 that i t may have indicated an expectation of intake on the part of the deer. If the deer knew apple platforms were p l e n t i f u l and that they could be returned to, i t may have enabled the deer to search in hope of finding some 'better' food. Clumping of apple platforms (hypothesis 7) did not s i g n i f i c a n t l y a f f e c t intake of preferred foods at any food density. Possibly because of undetected differences in the searching patterns used by both animals, the degree of clumping (in combination with density) affected the proportion of foods seen and differences between food seen and t o t a l apple a v a i l a b i l i t y for the male, but not the female. These equivocal results did not indicate that the female was aware of a l l apple locations but rather that the frequency of foods she encountered did not d i f f e r s i g n i f i c a n t l y from the r e l a t i v e frequency of a l l foods in the enclosure. The 1983 data did not support hypothesis 8 unequivocally. There was no s i g n i f i c a n t e f f e c t of density on the weight of p e l l e t s eaten per platform v i s i t e d . The male did eat more pel l e t e d a l f a l f a per platform v i s i t e d l a t e r in the t r i a l when apple abundance was lower. The one animal that could be tested had higher p e l l e t intake rates (per time searching) towards the end of the t r i a l s . This behaviour was associated with f a i l u r e to locate preferred food. There were no detectable differences in either the rate or pattern of searching by either animal after locating apple platforms. I concluded that- deer did not a l t e r their searching behaviour as a result of locating preferred foods. However, t h i s conclusion might not hold i f a more natural food d i s t r i b u t i o n was encountered by foraging deer. 141 CHAPTER 4: IMPLICATIONS What do the results of this research mean when they are applied to a highly variable and dynamic system in nature? What roles do the types of food encountered, the a v a i l a b i l i t y of highly preferred food items, the animal's experience with a pa r t i c u l a r food d i s t r i b u t i o n , the amount of food already eaten, and variation in plant quality play in food selection? Implications of my research and i t s relation to studies of food selection are discussed in thi s chapter. A major concern facing the ungulate ecologist i s that of forage a v a i l a b i l i t y . The problem i s not just one of how to measure a v a i l a b i l i t y , but measuring i t in a way that is relevant to the animal. In my experiments, the searching behaviour of captive animals was more closely related to what they had encountered within =5 m of the search path than to t o t a l food a v a i l a b i l i t y in the enclosure. Commonly employed use r a v a i l a b i l i t y ratios wo.uld include.', a l l platforms in the enclosure in a ca l c u l a t i o n of preference whereas forage encountered by one of the deer at lower apple densities d i f f e r e d s i g n i f i c a n t l y from overal l a v a i l a b i l i t y . True preference for a food, as I have used the term, can only be measured when a l l foods are equally available (Crawley 1983). Even under my "highly controlled" ad libitum conditions in the s t a l l t r i a l s , i t could be argued that size differences between pieces of apples, pelleted dairy ration, and pelleted a l f a l f a resulted in discernible a v a i l a b i l i t y differences. 142 Because of the d i f f i c u l t y in finding conditions when foods are equally available,,, f i e l d experiments that evaluate true preference are rare. Horton (1964) reported on deer acc i d e n t a l l y entering a fenced exclosure that contained equal amounts of young Pinus banksiana, P. strobus, P. resinosa, and Picea glauca. He found that P. banksiana was consumed more than any of the other seedlings. However, preference for P. banksiana may have been pa r t l y due to these p a r t i c u l a r jack pine trees being t a l l e r and thus more available. In most f i e l d experiments, preference and a v a i l a b i l i t y are probably inseparable. In addition, the amount of food eaten at each 'plant' v i s i t may i t s e l f be a function of the a v a i l a b i l i t y of highly preferred food items (see Chapter 3). Coupled with problems in estimating what forage i s available to the animal, food intake i s not e a s i l y documented. Fractions of d i f f e r e n t foods in the diet are usually determined from d i r e c t observations of forage intake (Wallmo et a l . 1973), from analyses of rumen contents, by the u t i l i z a t i o n method (Nelson and Leege 1982), or by microhistological analyses of herbivore faeces. With the l a t t e r technique, accuracy can be good afte r allowing for differences in mastication and digestion in f i s t u l a t e d animals (Golley and Buechner 1968). Given differences in retention time, however, the results represent a composite sample of intake over several hours or days, with food a v a i l a b i l i t y at the time of consumption unknown and unknowable by these methods. Even given the investigator's perfect knowledge of what was available and consumed, i t i s unclear that the c r i t e r i a of equal a v a i l a b i l i t y for preference 143 determinations would be met. Studies of food selection are also confronted with variations in forage q u a l i t y . Even i f we assume that the only factor determining the s p a t i a l d i s t r i b u t i o n of animals is food and that animals have perfect knowledge of the food in their environment, we would expect animals to aggregate in patches of habitat where t h e i r nutrient intake would be highest (Crawley 1983). Even in t h i s simple system the aggregation of other animals making similar choices would soon reduce the patch q u a l i t y . Individuals may also exchange information on a changing food resource (Krebs 1974). Furthermore, one foraging species can influence the food q u a l i t y of another species. Willms et a l . (1981) showed that grazing by c a t t l e in autumn affected the s p a t i a l d i s t r i b u t i o n of deer on Agropyron spicatum ranges in spring. Deer selected grazed instead of ungrazed ranges, once shoots exceeded stubble height. Constraints, therefore, become very complicated, even i f we assume that animals know where a l l potential food i s located a l l of the time, as discussed in Chapter 1. The o v e r a l l abundance of food ( i . e . i t s dry weight per unit area of habitat) may change r e l a t i v e l y slowly within a season, but i t s qu a l i t y may change more rapidly. Factors that determine good qua l i t y forage may also change between seasons. In winter a good q u a l i t y forage may be one that has a high energy content (Nudds 1980). For females during late gestation or l a c t a t i o n , q u a l i t y may represent high protein content (Ullrey et a l . 1967a, Nagy et a l . 1969). An animal tracking a changing q u a l i t y of forage may just sample d i f f e r e n t foods 1 44 (Westoby 1974) depending on whether or not ingestion or tasting is required to determine forage q u a l i t y (see Arnold and Dudzinski 1978). Wild deer do not often encounter foods of uniform quality as are pellet e d dairy ration and pelleted a l f a l f a . Janzen (1979:337) stated that "... a l i s t of Latin binomials feeding on other Latin binomials c a r r i e s almost no information when i t is remembered that the secondary-compound chemistry of two different plant parts on the same plant is much more l i k e l y to be different than the same." Variation in chemical composition has been documented between plant parts (Janzen et a l . 1976), among forage s i t e s (McCann 1985), and on a diurnal basis within twigs ( E l l i s 1985). In addition, herbivores are highly selective of the parts of a plant (see Crawley 1983). In part due to sampling problems, forage quality is rarely examined at the bite l e v e l ; minimum samples for some laboratory analyses require larger sized samples than the average bite of a deer. The above evidence would suggest, however, that v a r i a b i l i t y between bites on the same plant species does e x i s t . What role does this v a r i a b i l i t y play in determining diet selection and what are the implications of ignoring i t when studying food selection? Food p a l a t a b i l i t y generally increases with food q u a l i t y . Let us assume that we are dealing with a n u t r i t i o n a l l y wise animal, consuming f i v e species of browse (A through E), and that there i s a 1:1 relationship between forage q u a l i t y and p a l a t a b i l i t y to the forager. Figure 4.1a i l l u s t r a t e s t h i s hypothetical relationship, given the standard composite 1 4 5 >-H • < D O B A < D O B PALAT ABILITY PALATABILITY (-< o B PALATABILITY < D O B D • — • PALATABILITY Fi g u r e 4.1. A model of the i n f l u e n c e of i n t r a s p e c i f i c v a r i a t i o n i n n u t r i e n t q u a l i t y on d i e t s e l e c t i o n . See d e s c r i p t i o n of model i n t e x t . 146 sampling approach of c l i p p i n g a species, drying i t , combining a l l samples, and then sub-sampling from a homogenous mixture of a l l samples, we would obtain a single mean value for each quality measure per species. By introducing v a r i a t i o n in nutrient quality around the mean (e.g., protein content) for each forage 'species' and retaining the 1:1 r e l a t i o n s h i p between quality and p a l a t a b i l i t y , Figure 4.1b r e s u l t s . After giving our n u t r i t i o n a l l y wise forager a p a l a t a b i l i t y threshold (Fig. 4.1c), we do not find the discrete inclusion or exclusion of whole species, as the a p p l i c a t i o n of a prey-oriertted optimal foraging theory might predict. Rather a mixed diet occurs, including species E, D, and nearly half of species C. A lowering of the p a l a t a b i l i t y threshold (Fig. 4.1d) might result from lower a v a i l a b i l i t y of the more highly preferred species (D and E), and/or an increased intake requirement. In t h i s case C, D, and E should always be eaten, and only a small proportion of species B w i l l be rejected. The actual proportions in the diet w i l l s t i l l depend on r e l a t i v e a v a i l a b i l i t y and thus on encounter rates. The amount of preferred food already encountered w i l l influence the rate at which the acceptance threshold i s altered (from 1983 data, Chapter 3). This model i s oversimplified, but two conclusions can s t i l l be made. F i r s t , the model i l l u s t r a t e s how a simple set of rules, employed by a forager in a complicated environment, can result in a mixed-species d i e t . I am not proposing that sampling and tracking, the environment (sensu Westoby 1974) are not occurring. However, the mixed diets predicted in Figures 4.1c and d resulted from simple thresholds; sampling c r i t e r i a 147 were not invoked. The model also indicates that predicted preferences may not follow a simple, ranking. In the situation i l l u s t r a t e d in Figure 4.1b, v a r i a b i l i t y in plant quality r e s u l t s in some of species D and C being preferred to some members of species E (overlap of error bars p a r a l l e l to p a l a t a b i l i t y a x i s ) . Ignoring v a r i a t i o n , I would not have predicted almost half of species C to be as palatable as the range of species D. A prediction of this type of model i s that |the quality of plants of the same species selected by an animal (measured, for example, by esophageal f i s t u l a ) should exceed that of a randomly hand-picked sample of plants picked from the same clump of food. Data for the evaluation of this model are not available, largely because v a r i a b i l i t y i s ignored when food habits are studied. If large herbivores do not select food at the taxon l e v e l , then we must determine on what scale animals rank foods and how t i g h t l y they adhere to thi s ranking when food a v a i l a b i l i t y i s not a factor. Foraging strategies are a d d i t i o n a l l y influenced by animal behaviour in a variable environment. Selection is not r e s t r i c t e d to food type or species and plant part. Techniques of determining u s e : a v a i l a b i l i t y ratios have also been used to analyse selection of habitats (Neu et a l . 1974, Johnson 1980). The decisions of animals in the f i e l d , though, involve even more than simple choices of plants, patches, and habitats. Rather, animals face a large number of problems that vary in s p a t i a l and temporal scale simultaneously (Gass and Montgomerie 1981). Gass (1985) discussed a h i e r a r c h i c a l way in which animals may make decisions, including those of foraging. 1 48 Therefore, i t may be incorrect to interpret p a r t i c u l a r observed actions as resulting, simply from decisions of foraging or habitat selection. In a variable environment, we should also expect to find a d a p t a b i l i t y in the forager's behaviour. Adaptable behaviours enable the animal to respond to a variety of changes in i t s environment (see Gass 1985). As a ruminant, however, a deer i s somewhat ph y s i o l o g i c a l l y buffered against an environment that i s variable in terms of forage c h a r a c t e r i s t i c s . Buffering also argues against selection in s e n s i t i v i t y to small changes in the animal's environment. These arguments are not meant to imply that standard preference ratios w i l l not help to i d e n t i f y highly preferred or avoided foods. By the r a t i o method, deer preferred apples during a l l 1983 pen t r i a l s . There are many more elaborate predictors of preference than simple ra t i o s used in range and w i l d l i f e management (see Nelson 1984, Skiles 1984). The problem i s not in designing elaborate predictions, however, but in describing what was actually available to the forager. In reviewing measures that have been to predict forage preferences, Skiles (1984:173) described c a f e t e r i a feeding experiments as unrigorous and involving l i t t l e manipulation of the data. Although c a f e t e r i a t r i a l s may not be elegant, they do permit measurement of food preference in the absence of confounding by a v a i l a b i l i t y . I have shown that ad libitum preference cannot be; inferred accurately from f i e l d observations. Attempts to weight estimates for varying a v a i l a b i l i t y f a i l because actual 1 49 a v a i l a b i l i t y as encountered by the forager i s always unknown to the observer. Furthermore, arguing that the animal i s omniscient on a seasonal time scale does not take into account short-term changes that occur in forage quality and a v a i l a b i l i t y . What then do these results t e l l us when they are applied to a highly dynamic and variable natural system as i s the normal environment inhabited by b l a c k - t a i l e d deer? My study used a r t i f i c i a l foods and examined deer foraging behaviour under unnatural conditions. The study did not produce a simple formula for predicting the effects of forage abundance and d i s t r i b u t i o n on food selection. It did show, however, that foraging bouts are dynamic and that even given a constant preference for foods, diet selection varied within observation periods with the amount of food encountered, the amount of food already eaten, and experience during previous t r i a l s with a similar food d i s t r i b u t i o n . In the pen, both animals ate most of the highly preferred food f i r s t . Even at low i n i t i a l apple density, animals searched for apples u n t i l , depending in part on the amount of food already eaten, they gradually increased their consumption of non-preferred foods. It seems unlikely that wild deer would normally forage in this manner, but i t does raise the question of whether deer are more selective at the start or the end of foraging bouts. These data demonstrate that captive deer are more selective at t'he start of foraging bouts. Forage abundance i s in part a function of past consumption and the time of measurement may influence a v a i l a b i l i t y 150 conclusions. My data indicate that similar dynamics occur within foraging bouts. However, i f preferred foods are unlimited, the forager may not be able to s i g n i f i c a n t l y lower the food abundance. Data from 1983 pen t r i a l s suggested that food dispersion may influence the types of food encountered. For example, over the range of dispersions tested, clumping of apples resulted in one captive animal encountering d i f f e r e n t proportions of food than were a c t u a l l y present in the enclosure. In a meadow or forest, with many types of forage "available" and variable plant dispersion, i t is l i k e l y that foragers select items based on a sample of ov e r a l l abundance (as does the inve s t i g a t o r ) . To c l e a r l y understand the choices an animal makes, the researcher must know what foods have been available to the forager. The model presented above proposes that even within the foods encountered there may not be a 1:1 correspondence between forage species and forage q u a l i t y . By invoking u s e : a v a i l a b i l i t y r a t i o s the investigator i s attempting to describe forage selection, and yet, what was act u a l l y 'available' at the time the feeding decision was made i s always unknown to him with currently used methods. In t h i s study the animals' experience within a food d i s t r i b u t i o n increased the rate at which they located preferred foods. Although natural foraging conditions may not replenish themselves on a da i l y basis (as apples on platforms d i d ) , the area foraged on a given day may be a function of past experience (Gass and Sutherland 1985). As such, forage a v a i l a b i l i t y may not be randomly sampled from day to day by 151 animal or investigator. When preferred food was abundant o v e r a l l , captive deer l e f t some food behind at feeding s i t e s . Similar behaviour observed under natural conditions has.been explained,by suggesting that deer are sampling (Westoby 1978) and/or that they are avoiding toxic secondary compounds (Freeland and Janzen 1974). I was unable to discriminate between possible explanations for this behaviour. However, the fact that there was an e f f e c t of food abundance on the amount of food l e f t indicates that abundance of plants in the f i e l d ' should be examined r e l a t i v e to sampling and toxin aversion hypotheses. Changes in intake rates within foraging bouts, coupled with changes in selection discussed above, indicate that foraging bouts are dynamic. Therefore, i t may be incorrect for researchers to assume that behaviour in foraging bouts i s invariant and that they can randomly sample (e.g., scan sampling) for intake rates or forage s e l e c t i o n . If deer in the wild do behave d i f f e r e n t l y within foraging bouts, then the length of time an animal has been foraging would be an important measure when observing behaviour. What role does i n t r a s p e c i f i c plant v a r i a t i o n play in food selection and what are the consequences of ignoring t h i s possible source of variation? Experiments described in t h i s thesis were conducted using r e l a t i v e l y uniform quality foods. Even then, lower variation in apple quality may account for more consistent food selection during 1983 t r i a l s compared with those of 1981. The model presented in Figure 4.1 implies that some individuals of d i f f e r e n t plant species may be more similar 1 52 in quality than individuals of the same species. Different preference rankings for di f f e r e n t assays of quality ( i . e . protein, digestible energy, secondary compounds), would result in di f f e r e n t predicted diets although the animal's decisions may have been based on a number of factors. "The animal's n u t r i t i o n a l environment i s a vast array of chemical compounds conveniently ordered by w i l d l i f e s c i e n t i s t s into aggregations from species to communities" (Robbins 1983:234). When reseachers approach food selection at the species l e v e l , or lump forage to broader groupings of taxa, description may be f a c i l i t a t e d , but understanding an animal's foraging strategies does not necessarily increase. Researchers must question the worth of c o l l e c t i n g large amounts of data that may not enhance understanding of what i s r e a l l y occurring. If i n t r a s p e c i f i c variation i s greater than i n t e r s p e c i f i c v a r i a t i o n , then the plant species may not be the appropriate unit to describe food selection. Such variation might help explain some of the discrepancies documented between numerous food habits studies such as those described by Spalinger (1980). Hence, d i s t r i b u t i o n and abundance of forage act in combination with the searching of the forager to affect diet s e l e c t i o n . It i s unlikely that these can ever be accurately described for the wild situation, at least for so complex an animal or diet as that of deer. The approach to be taken in studying ungulate foraging should be dictated: by the quest ions: the researcher- is: asking? Simple or complex preference indices may indicate the degree of diet selection exhibited by the forager and potential 153 competition between species (Nelson 1984). Similarly, i f the objective of a study i s to document food habits, then a rough measure of how rare foods are may indicate the degree of sel e c t i o n . However, th i s t e l l s us l i t t l e about the foraging behaviour of the animal and how the forager i s using the habitat. The primary value of a preference index is to enable the researcher to rank various plants under a s p e c i f i c set of circumstances (Krueger 1972). As such, they are largely d e s c r i p t i v e . Such indices are not very useful in studying foraging decisions or even in predicting diet consumption for a general case largely because of assumptions researchers must make about what was available to the animal when a foraging decision was made. Without a better understanding of the v a r i a b i l i t y of plant quality and how i t relates to taxonomic c l a s s i f i c a t i o n of foods, and the mechanisms used in selection, we w i l l not move closer to being able to accurately predict diet selection of free-ranging animals. 154 LITERATURE CITED Arnold, G.W., E.S. DeBoer, and C.A.P. Boundy. 1980. The influence of odour, and taste on the food preference and food intake of sheep. Aust. J. Anim. S c i . 31:571-587. •, and H.A. B i r r e l l . 1977. Food intake and grazing behaviour of sheep varying in body condition. Anim. Prod. 24:343-353. -, and M.L. Dudzinski. 1978. Development in Animal and  Veterinary Sciences Vol 2: Ethology of free-ranging  domestic animals. Elsevier S c i . Publ. Co., New York. 198 pp. •, and J.L. H i l l . 1972. Chemical factors a f f e c t i n g selection of food plants by ruminants, pp. 71-101 i_n J.B. Harborne (ed.). Phytochemical ecology. Academic Press, London. •, and R.A. Mailer. 1977. Effects of n u t r i t i o n a l experience in early and adult l i f e on the performance and dietary habits of sheep. Appl. Anim. Ethol 3:5-26. Banks, C.J. 1957. The behaviour of individual c o c c i n e l l i d larvae on plants. Anim. Behav. 5:12-24. Begon,- M-. and M.. Mortimer. 1981 . Population ecology: a, una f l e d  study of animals and plants. Sinaeur Assoc. Inc. Sunderland, Mass. 200 pp. B e l l , R.V. 1971. A grazing ecosystem in the Serengeti. S c i . Amer. 225:86-93. Belovsky, G.E. 1978. Diet optimization in a generalist herbivore: the moose. Theor. Pop. B i o l . 14:105-134. . 1981. Food plant selection by a generalist herbivore: the moose. Ecology 62:1020-1030. •. 1984. Herbivore optimal foraging: a comparative test of three models. Amer. Natur. 124:97-115. 155 Beukema, J . J . 1968. Predation by the three-spined stickleback Gasterosteus aculeatus L., the influence of hunger and experience. Behaviour 31 ; 1 -1 2.6. Bossema, I. 1979. Jays and oaks: an ecoethological study of a symbiosis. Behaviour. 70:1-117. Bryant, J.P. and P.J. Kuropat. 1980. Selection of winter forage by subarctic browsing vertebrates: the role of plant chemistry. Ann. Rev. Ecol. Syst. JJ_: 261 -285. Buchler, E.R. 1976. Prey selection by Myotis lucifugus. (Chiroptera: V e s p e r t i l i o n i d a e ) . Amer. Natur. 110:619-628• Bunnell, F.L. and M.P. Gillingham. [1985]. Foraging behavior: the dynamics of dining out. i_n R.J. Hudson and R.G. White (eds.). Bioenergetics of WiId Herbivores. CRC Press, Roca Baton, F l a . (in pre¥s). Caraco, T. 198. R i s k - s e n s i t i v i t y and foraging groups. Ecology 62:527-531. •. 1982. Aspects of risk-aversion in foraging white-crowned sparrows. Anim. Behav. 30:719-727. •, S. Martindale, and T.S. Whitham. 1980. An empirical demonstration of r i s k - s e n s i t i v e foraging preferences. Anim. Behav. 28:820-830. Charnov, E.L. 1976a. Optimal foraging: attack strategy of a mantid. Amer. Natur. 110:141-151. •. 1976b. Optimal foraging: the marginal value theorem. Theor. Popul. B i o l . , 9:129-136. Chesson, J . 1978. Measuring preference in selec t i v e predation. Ecology 59:211-215. Cody M.L. 1971. Finch flocks in the Mohave Desert. Theor. Pop. B i o l . 2:142-148. C o l l i n s , W.B., P.J. Urness, and D.D. Austin. 1978. Elk diets and a c t i v i t i e s on d i f f e r e n t lodgepole pine habitat 1 56 segments. J. Wildl. Manage. 42:799-810. Connolly, G.E., B.O. E l l i s o n , J.W. Flemming, S. Geng, R.E. Kepner, W.M. Longhurst, J.H. Oh, and G.F. Russell. 1980. Deer browsing of Douglas-fir trees in r e l a t i o n to v o l a t i l e terpene composition and iR v i t r o fermentability. For. S c i . 26:179-193. Cooper, S.M. 1985. Factors influencing the u t i l i z a t i o n of woody plants and forbs by ungulates. PhD. Thesis, Univ. of Witwatersrand, Johannesburg, South A f r i c a . Cowan, I. McT. 1945. The ecological relationships of the food of Columbian black-tailed deer, Odocoileus hemionus  columbianus (Richardson), in the coast forest region of southern Vancouver Island, B r i t i s h Columbia. Ecol. Monogr. jj>: 1 1 0-1 39. Cowie, R.J., J.R. Krebs, and D.F. Sherry. 1981. Food storing by marsh t i t s . Anim. Behav. 29:1252-1259. Crawley, M.J. 1983. Herbivory: the dynamics of animal-plant interactions. Studies in ecology Vol. 10. Univ. of C a l i f . Press. Berkeley, C a l i f . 437 pp. Crouch, G.L. 1979. Food habits of b l a c k - t a i l e d deer on forested habitats in the P a c i f i c Northwest, pp 53-59 rn O.C. Wallmo and J.W. Schoen (eds.). Sitka Black-tailed deer: Proceedings of a conference in Juneau Alaska. USDA For. Serv. Ser. No-., R1-0-48-. . 1981. Coniferous forest habitats. Part I. Food habits and n u t r i t i o n , pp 423-433 ijn O.C. Wallmo, (ed.). Mule and  Black-tailed deer of North America. Univ. of Nebraska Press. Lincoln, Nebr. Curio, E. 1976. The ethology of predation. Springer-Verlag. B e r l i n . 250 pp. Dealy, H.T. 1959. The influences of logging practices on Columbian bla c k - t a i l e d deer in the Blue River area of Oregon.. M.Sc. Thesis, Oregon State College:, C o r v a l l i s , Oreg. 65 pp. 157 Dean, R.E. and A.H. Winward. 1974. An investigation into the p o s s i b i l i t y of tansy ragwort poisoning of bla c k - t a i l e d deer. J. Wi l d l . Dis. 10:166-169. Dublin, H.T. 1980. Relating deer diets to forage quality and quantity: the Columbian white-tailed deer (Odocoileus  virginianus leucurus). M.Sc. Thesis, Univ. of Wash. Seattle, Wash. 135 pp. Dunham, K.M. 1980. The feeding behaviour of a tame impala (Aepyceros melampus). Afr. J. Ecol. 18:253-257. E l l i s , J.E., J.A. Weins, C.F. Rodell, and J.C. Anway. 1976. A conceptual model of diet selection as an ecosystem process. J . Theor. B i o l . 60:93-108. E l l i s , R.M. 1985. Concentration of minerals in Vaccinium  parvifolium, a key browse species for bl a c k - t a i l e d deer. pp. 397-402 in Meehan, W.R., T.R. Me r r e l l , J r . , and T.A. Hanley (eds.). Fish and W i l d l i f e relationships in  old-growth f o r e s t s : Proceedings of a symposium (Juneau, AK., 12-14 A p r i l 1982). Bookmasters, Ashland, Ohio. Emlen, J.M. 1966. The role of time and energy in food preference. Amer. Natur. 100:611-617." . 1984. Population biology: the coevolution of population  dynamics and behavior. MacMillan Publ. Co. New York. 547 pp. Fox, D.J. and K.E. Guire. 1976. Midas. S t a t i s t i c a l Research Laboratory, Univ. of Michigan, Mich. 203 pp. Fraser, D. and E. Reardon. 1980. Attraction of wild ungulates to mineral-rich springs in central Canada. Holarct. Ecol. 3:36-40. Freeland, W.J. and D.H. Janzen. 1974. Strategies in herbivory by mammals: the role of plant secondary compounds. Amer. Natur. 108:269-289. Gass, C L . 1985. Behavioral foundations of adaptation, pp. 63-107 i_n P.P.G. Bateson and P.H. Klopfer (eds.). Perspectives in ethology. Vol 6. Plenum Press, New York. 1 58 Gass, C.L. and R.D. Montgomerie. 1981. Hummingbird foraging behavior: decision-making and energy regulation, pp. 159-194 i_n A.C. Kami! and T.D. Sargent (eds.). Foraging  behavior: ecological, ethological, and psychological approaches. Garland STPM Press, New York. •, and G.D. Sutherland. [1985], Specialization by t e r r i t o r i a l hummingbirds on experimentally enriched patches of flowers: energetic p r o f i t a b i l i t y and learning. Can. J . Zool. 6J3: (i_n press). Geist, V. 1971 . Mountain sheep: a study in behavior and evolution. Univ. of Chicago Press. Chicago. 383 pp. Gillingham, M.P. and F.L. Bunnell. 1985. R e l i a b i l i t y of motion-sensitive radio c o l l a r s for estimating a c t i v i t y of b l a c k - t a i l e d deer. J. W i l d l . Manage. 49:951-958. Gluesing, E.A. and D.F. Balph. 1980. An aspect of feeding behavior and i t s importance to grazing systems. J . Range Manage. 33:426-427. Golley, F.B. and H.K. Buechner. 1968. A p r a c t i c a l guide to the  study of the productivity of large herbivores. I.B.P. Handbook 7. Blackwell S c i . Publ. Oxford. Goodall, D.W. 1974. A new method for the analysis of s p a t i a l pattern by random pairing of quadrats. Vegetatio. 29: 135-146. Greig, M. and J . Bjerring. 1980. GENLIN: a general least squares analysis of variance package. University of B.C. Computing Centre Publ. Vancouver, B.C. 55 pp. Hafez, E.S.E. and M.W. Schein. 1962. The behaviour of c a t t l e , pp. 247-296 i_n E.S.E. Hafez (ed.). The behaviour of  domestic animals. Williams and Wilkins Co. Baltimore, Md. Hanley, T.A. 1980. N u t r i t i o n a l constraints on food and habitat se l e c t i o n by sympatric ungulates. PhD. Thesis, Univ. of Washington, Seattle, Wash. 176 pp. •, and J.D. McKendrick. 1983. Seasonal changes in chemical composition and n u t r i t i v e value of native forages in a spruce-hemlock forest, Southeastern Alaska. USDA For. 159 Serv. Res. Pap PNW-312. 41 pp. Hanley, T.A. and J.D. McKendrick. 1985. Potential n u t r i t i o n a l l i m i t a t i o n s for blac k - t a i l e d deer in a spruce-hemlock forest, Southeastern Alaska. J. Wildl. Manage. 49: 103-114. Hanson, W.R. and R.F. Libisky. 1964. Association of pheasants with vegetative types in east-central I l l i n o i s . Trans. N, Amer. Wi l d l . Natur. Res. Conf. 29:295-306. Heady, H.F. 1964. P a l a t a b i l i t y of herbage and animal preference. J. Range Manage. 17:76-82. Hixon, M.A. 1982. Energy maximizers and time minimizers: theory and r e a l i t y . Amer. Natur. 119:596-599. Holechek, J.L. and R. Valdez. 1985. Evaluation of in v i t r o digestion for improving botanical estimates of mule deer fecal samples. J. Mamm. 66:574-577. Horton, K. 1964. Deer prefer jack pine. J. For. 62:497-499. Ivlev, V.S. 1961. Experimental ecology of the feeding of  fishes . Yale Univ. Press. New Haven, Conn. 302 pp. Jacobs, J. 1974. Quantitative measurement of food selection. Oecologia. (Berlin) 14:413-41 7. Janis, C. 1976. The evolutionary strategy of the Equidae and the o r i g i n s of rumen and cecal digestion. Evolution. 30: 757-774. Janzen, D.H. 1979. New horizons in the biology of plant defences, pp. 331-350 iri G.A. Rosenthal and D.H. Janzen (eds.). Herbivores: their interaction with secondary  plant metabolites. Academic Press. New York. •, H.B. Juster, and I.E. Liener. 1976. I n s e c t i c i d a l action of the- phytohemagglutinin in. black beans- on a Bruchid. beetle. Science 192:795-796. 160 Johnson, D.H. 1980. The comparison of usage and a v a i l a b i l i t y measurements for evaluating resource preference. Ecology 61:65-71. Jones, D. 1984. Use, misuse, and role of multiple-comparison procedures in ecological and a g r i c u l t u r a l entomology. Environ. Entomol. 13:635-649. Jordan, P.A., D.B. Botkin, A.S. Dominski, H.S. Lowendorf, and G.E. Belovsky. 1973. Sodium as a c r i t i c a l nutrient for the moose of Isle Royale. Proc. N. Amer. Moose Conf. Workshop. 9:13-42. Jung, H.G. 1977. Responses of mammalian herbivores to secondary plant compounds. The B i o l o g i s t 59:123-136. Kamil, A.C. and T.D. Sargent (eds.). 1981. Foraging behavior: ecological, ethological, and psychological approaches. Garland STPM Press, New York. 534 pp. -, and S.I. Yoerg. 1982. Learning and foraging behavior, pp.325-364 i_n P.P.G. Bateson and P.H. Klopfer (eds.). Perspectives in ethology. Vol 5_: Ontogeny. Plenum Press, New York. Klein, D.R. 1965. Ecology of deer range in Alaska. Ecol. Monogr. 3_5:259-284. -. 1969. Food selection by North- American, deer and their response to o v e r - u t i l i z a t i o n of preferred plant species, pp. 25-46. iin A. Watson (ed.). Animal  populations in r e l a t i o n to their food resources. B r i t . Ecol. Soc. Symp. #10. Krebs, J.R. 1974. Colonial nesting and s o c i a l feeding as strategies for exploiting food resources in the Great Blue Heron (Ardea herodias). Behaviour 51:99-131. -. 1978. Optimal foraging: decision rules for predators, pp. 23-63 in J.R. Krebs and N.B. Davies (eds.). Behavioral ecology: an evolutionary approach. Blackwell S c i . Pubs,., Oxford. , and R.H. McCleery. 1984. Optimization in behavioural ecology, pp. 91-121 in J.R. Krebs and N.B. Davies (eds.). 161 Behavioural ecology: an evolutionary approach. (2nd e d i t i o n ) . Blackwell S c i . Publ. Oxford. Krebs, J.R., J. Ryan, and E.L. Charnov. 1974. Hunting by expectation or optimal foraging? A study of patch use by chickadees. Anim. Behav. 22:953-964. Krueger, W.C. 1972. Evaluating animal forage preference. J. Range. Manage. 25:471-475. •, W.A. Laycock, and D.A. Price. 1974. Relationships of taste, smell, sight, and touch to forage s e l e c t i o n . J. Range. Manage. 27:258-262. LaGory, M.K., K.E. LaGory, and D.H. Taylor. 1985. Winter browse a v a i l a b i l i t y and use by white-tailed deer in Southeastern Indiana. J . W i l d l . Manage. 49:120-124. Le, C. 1984. UBC SLTEST: Equality of slope t e s t . Univ. of B.C. Computing Centre Publ. Vancouver, B.C. 10 pp. L e s l i e , D.M., E.E. Starkey, and M. Vavra. 1984. Elk and deer diets in old-growth forests in western Washington. J. Wil d l . Manage. 48:762-775. Lindroth, R.L. 1979. Diet optimization by generalist mammalian herbivores. Theor. B i o l . 61:41-58. Longhurst, W.M., H.K. Oh, M.B. Jones, and R.E. Kepner. 1968. A basis for the p a l a t a b i l i t y of deer forage plants. Trans. N. Amer. W i l d l . Natur. Res. Conf. 33:181-192. MacArthur, R.H., and E.R. Pianka. 1966. On optimal use of a patchy environment. Amer. Natur. 100:603-609. Matthews, L.R. and R. Kilgour. 1980. Learning and associated factors in ruminant feeding behaviour, pp. 123-144 i_n Y. Ruckebusche and P. Thivend ( e d S e ) . Digestive physiology  and metabolism in ruminants. AVI Publ. Co., Westport, Conn. Mautz, W.W. 1974. Deer n u t r i t i o n a l and physiological research so what? Trans. Northeast Sec. Wildl. Soc. 31:113-122. 162 McCann, R.K. 1985. Variation in crude protein of s a l a l (Gaultheria shallon), a key winter browse species of coastal b l a c k - t a i l e d deer. B.Sc. Thesis. Univ. of B.C. Vancouver, B.C. 51 pp. McLellan, B. [1985]. Use a v a i l a b i l i t y analysis and selection of timber by g r i z z l y bears. Proceedings of the Gr i z z l y Bear Habitat Symposium ( A p r i l 30-May 2, 1985), Missoula, Mont. (in press). McNair, J.M. 1982. Optimal giving up times and the marginal value theorem. Amer. Natur. 19:511-529. Milchunas, D.G., M.I. Dyer, O.C. Wallmo, and D.E. Johnson. 1978. In- v i v o / i n - v i t r o relationships of Colorado mule deer forages. Colorado Div. of Wi l d l . Spec. Rep. No. 43. 44 pp. Nagy, J.G., T. Hakonson, and K.L. Knox. 1969. Eff e c t of quality on food intake in deer. Trans. N. Amer. Wi l d l . Nat. Res. Conf. 34:146-154. Nelson, J.R. and T.A. Leege. 1982. N u t r i t i o n a l requirements and food habits, pp. 323-367 in J.W. Thomas and D.E. Toweill (eds.). Elk of North America: ecology and management. Stackpole Books. Harrisburg, Pa. •. 1984. A modeling approach to large herbivore competition, pp. 491-524 in Developing strategies for  rangeland management. Westview Press, London., Neu, C.W., CR. Byers, and J.M. Peek. 1974. A technique for analysis of u t i l i z a t i o n a v a i l a b i l i t y data. J. Wi l d l . Manage. 38:541-545. Nudds, T.D. 1980. Forage 'preference': t h e o r e t i c a l considerations of diet selection by deer. J. Wildl. Manage. 44:735-740. Oh, H.K., M.B. Jones, and W.M. Longhurst. 1968. Comparison of rumen microbial i n h i b i t i o n r e s u l t i n g from various essential oils: isolated, from r e l a t i v e l y unpalatable plant species. Appl. Microbiol. 16:39-44. 1 63 Oh, H.K., T. Saki, M.B. Jones, and W.M. Longhurst. 1967. Effects of various es s e n t i a l o i l s isolated from Douglas-fi.r needles upon sheep and deer rumen microbial a c t i v i t y . Appl. Microbiol. 15:777-784. Owen-Smith, N. and P. Novellie. 1982. What should a clever ungulate eat? Amer. Natur. 119:151-178. Petrides, G.A. 1975. P r i n c i p a l foods versus preferred foods and their relations to stocking rates and range condition. B i o l . Conserv. 7:161-169. Potvin, F. and J. Huot. 1983. Estimating carrying capacity of a white-tailed deer wintering area in Quebec. J. Wildl. Manage. 47:463-475. Pyke, G.H., H.R. Pulliam, and E.L. Charnov. 1977. Optimal foraging: a selective review of theory and te s t s . Q. Rev. B i o l . 52:137-154. Quinton, D.A. 1985. Saskatoon serviceberry toxic to deer. J . Wildl. Manage. 4_9:362-364. Radwan, M.A. and G.L. Crouch. 1974. Plant c h a r a c t e r i s t i c s related to feeding preference by blac k - t a i l e d deer. J . Wildl. Manage. 38:32-41. Regelin, W.L., C.C. Schwartz, and A.W. Franzmann. 1985. Seasonal metabolism of: adult moose. J . Wildl. Manage. 49: 388-393. Robbins, C.T. 1983. W i l d l i f e feeding and n u t r i t i o n . Academic Press. New York. 343 pp. Rochelle, J.A. 1980. Mature forests, l i t t e r f a l l and patterns of forage quality as factors in the n u t r i t i o n of bla c k - t a i l e d deer on northern Vancouver Island. PhD. Thesis, Univ. of B.C., Vancouver, B.C. 295 pp. Scheffe H. 1959. The analysis of variance. John Wiley and Sons, Inc. New York. Schoener, T.W. 1971. Theory of feeding strategies. Ann. Rev. Ecol. Syst. 2:369-404. 164 -. 1984. Learning and Behavioural Ecology, pp. 1 70-194 iji J.R. Krebs and N.B. Davies (eds.). Behavioural Ecology: an evolutionary approach (2nd e d i t i o n ) . Blackwell S c i . Publ. Oxford. Shettleworth, S.J. 1983. Memory in food-hoarding birds. Scient. Amer. 284:102-110. . and J.R. Krebs. 1982. How marsh t i t s f i n d their hoards: the roles of si t e preference and s p a t i a l memory. J. Exp. Psychol.: Anim. Behav. Processes. 8:354-375. Skiles, J.W. 1984. A review of animal preference, pp. 153-213 in Developing strategies for rangeland management. Westview Press, London. Smith, A.D. 1952. D i g e s t i b i l i t y of some native forages for mule deer. J . Wildl. Manage. 16:309-312. Smith, J.N.M. 1974a. The food searching behaviour of two European thrushes I . Descriptions and analysis of search paths. Behaviour 48:276-302. -. 1974b. The food searching behaviour of two European thrushes. I I . The adaptiveness of search patterns. Behaviour 49:1-61. Sokal, R.P. and F.J. Rohlf. 1981. Biometry. W.H. Freeman and Company. San Francisco. 8.59, pp. Spalinger, D.E. 1980. Mule deer habitat evaluation based upon n u t r i t i o n a l modelling. M.A. Thesis. Univ. of Nevada. Reno, Nev. 190 pp. Stenseth, N.C. and L. Hansson. 1979. Optimal food sele c t i o n : a graphic model. Amer. Natur. 113:373-389. Trudell, J . and R.G. White. 1981. The effect of forage structure and a v a i l a b i l i t y on food intake, b i t i n g rate, bites s i z e , and d a i l y eating time of reindeer. J . Appl. Ecol. 18:6.3-8 I.-Tucker, R.E., W. Majak, P.D. Parkinson, and A. McLean. 1976. P a l a t a b i l i t y of Douglas f i r - f o l i a g e to mule deer in 165 re l a t i o n to chemical and s p a t i a l factors. J. Range. Manage. 29:486-489. Ullrey , D.E., W.G. Youatt, H.E. Johnson, L.D. Fay, and B.L. Bradley. 1967a. Protein requirements of white-tailed deer fawns. J . Wildl. Manage. 3J_:679-685. U l l r e y , D.E., W.G. Youatt, H.E. Johnson, L.D. Fay, and B.E. Brent. 1967b. D i g e s t i b i l i t y of cedar and jack pine browse for the white-tailed deer. J. Wi l d l . Manage. 31. 448-454:. -, , , A.B. Cowan, R.L. Covert, and W.T. Magee. 1972. D i g e s t i b i l i t y and estimated metabolizability of aspen browse for while-tailed deer. J. Wildl. Manage. 36:885-891. Van Dyne, G.M. and H.F. Heady. 1965. Botanical composition of sheep and c a t t l e diets on a mature annual range. Hil g a r d i a . 34:465-492. Vander Wall, S.B. and R.P. Balda. 1981. Ecology and evolution of food-storage behavior in conifer-seed-caching corvids. Z. Tierpsychol. 56:217-242. Vangilder, L.D.., 0. Torgerson, and W.R. Porath. 1982. Factors influencing diet selection by white-tailed deer. J. Wil d l . Manage. 46:711-718. Verme, L.,J. and D.E.. Ul.lrey. 1984. Physiology and n u t r i t i o n , pp. 91-118 ir\ L.K. Halls (ed.). White-tailed deer: ecology and management. Stackpole Books. Harrisburg, Pa. Wallmo, O.C, R.B. G i l l , L.H. Carpenter, and D.W. Reichert. 1973. Accuracy of f i e l d estimates of deer food habits. J . Wil d l . Manage. 37:556-562. Westoby, M. 1974. An analysis of diet selection by large generalist herbivores. Amer. Natur. 108:290-304. •. 1978. What are the b i o l o g i c a l bases of varied diets? Amer. Natur. 112:627-631. 166 White, S.M., B.L. Welch, and J.T. Flinders. 1982. Monoterpenoid content of pygmy rabbit stomach ingesta. J. Range Manage. 35:107-109. Wickstrom, M.L. 1983. Food intake and foraging energetics of elk (Cervus elaphus nelsoni) and mule deer (Qdocoileus  hemionus hemionus). MSc. Thesis, Wash. State Univ., Pullman, Wash. 64 pp. Wickstrom, M.L., C.T. Robbins, T.A. Hanley, D.E. Spalinger, and S.M. Parish. 1984. Food intake and foraging energetics of elk and mule deer. J . W i l d l . Manage. 48:1285-1301. Willms, W., A.W. Bailey, A. McLean, and R. Tucker. 1981. The e f f e c t s of f a l l d e f o l i a t i o n on the u t i l i z a t i o n of blue bunchgrass and i t s influence on the d i s t r i b u t i o n of deer in spring. J. Range Manage. 34:16-18. Zach, R. and J.B. F a l l s . 1976. Ovenbird (AVES: Parulidae) hunting behaviour in a patchy environment: an experimental study. Can. J . Zool. 54:1863-1879. •, and . 1977. Influence of capturing a prey on subsequent search in the ovenbird (AVES: Parulidae). Can. J. Zool. 55:1958-1969. Zahoric, D.M. and K.A. Houpt. 1977. The concept of n u t r i t i o n a l wisdom: a p p l i c a b i l i t y of laboratory learning models to large herbivores, pp 45-67 i_n L.M. Barker, M. Best, and M.. Domjan (eds.). Learning Mechanisms in Food; Selection. Baylor Univ. Press. Waco, Tex. 

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.831.1-0096617/manifest

Comment

Related Items