Hubert Wiktor Askanas

DOES MATING SYSTEM AFFECT THE PHYSIOLOGICAL STRESS OF PROVISIONING BIRDS? A COMPARATIVE STUDY OF POLYGYNANDROUS BICKNELL S THRUSH (CATHARUS BICKNELLI) AND SOCIALLY MONOGAMOUS SWAINSON S THRUSH (CATHARUS USTULATUS) BREEDING SYMPATRICALLY. Hubert Wiktor Askanas B.Sc. Honours, University of New Brunswick, 2008 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTERS OF SCIENCE IN BIOLOGY IN THE GRADUATE ACADEMIC UNIT OF THE DEPARTMENT OF BIOLOGY IN THE FACULTY OF SCIENCE Supervisor: Antony W. Diamond, Ph.D., Department of Biology/Ornithology Examining Board: Les Cwynar, Ph. D., Department of Biology/Botany, Chair Andy Didyk, Ph. D., Department of Biology/Nursing External Examiner: Graham Forbes, Ph. D., Department of Forestry/ Environmental Management, U.N.B. THIS THESIS IS ACCEPTED BY THE DEAN OF GRADUATE STUDIES THE UNIVERSITY OF NEW BRUNSWICK (September), 2012 Hubert Wiktor Askanas, (2012)

ii Abstract Bicknell s Thrush is one of two migratory bird species in North America with a polygynandrous mating system. Having multiple male feeders at the nest, and a skewed sex ratio, suggests that Bicknell s Thrush adults partition the feeding effort. Since nestling provisioning is physiologically stressful, I predicted that Bicknell s Thrush would be less stressed than a monogamous counterpart during the nestling provisioning period. I compared Bicknell s Thrush with Swainson s Thrush, a close relative and concurrent monogamous breeder in the study area. Heterophil to lymphocyte ratio (H:L) was used as a surrogate for corticosterone in the analysis of physiological stress. To test the sensitivity of H:L to physiological stress, I compared breeding and overwintering birds. Overwintering birds were less physiologically stressed than breeding birds, and had no circulating hematozoa. Both species of Thrush were found to be parasitized during the breeding season, but hematozoa infections and infestation with trombiculid mites had no effect on body condition or H:L. Most Bicknell s Thrush females provided at rates similar to either sex of Swainson s Thrush. Most Bicknell s Thrush males provided less than 27% of the feeds at a nest, but total rates for males feeding two broods were similar to female Bicknell s Thrush. I hypothesized that male Bicknell s Thrush would have the lowest H:L, but could not conclude that they were the least stressed during the provisioning period. The results were most likely confounded by Swainson s Thrush whose nests had failed. I did see the potential of H:L as a tool for deducing nest success in demographic studies where catching adults is easier than finding nests: provisioning birds, of either species, had significantly higher H:L ratios than those sampled after their nest had failed.

iii Dedication In memory of my mother, Teresa, who lost her battle with cancer before the completion of this project, you are missed. To my father Wiktor, I would never have realized my full potential without you. I would also like to thank my wife Chrissie, son Caelum, and daughter Aniela, for their support and patience.

iv Table of contents Abstract...ii Dedication...iii Table of contents...iv List of tables...vii List of figures...ix Chapter 1: Introduction...1 - General introduction and objectives...1 - General methods - Study sites...4 - Introduction to chapters...4 - Literature cited...6 Chapter 2: Susceptibility to hematozoan infection is influenced by high physiological stress during the breeding season, and is correlated with negative effects on condition...9 - Abstract...10 - Introduction...11 - Methods...12 - Results - Samples...15 - Hematozoa and leukocyte profile...15 - Hematozoa and condition...17 - Discussion - Effects of hematozoa on leukocyte profile...17 - Effects of hematozoa on male condition...21 - Acknowledgments...22 - Literature cited...22

v Chapter 3: The effect of trombiculid larvae infestation on condition and relative leukocyte profiles of two Thrushes...32 - Abstract...33 - Introduction...34 - Methods...35 - Results...36 - Discussion...37 - Acknowledgments...37 - Literature cited...38 Chapter 4: Seasonal effects on relative leukocyte profiles and prevalence of hematozoa in Bicknell's Thrush, and evidence of local transmission in New Brunswick, Canada...43 - Abstract...44 - Introduction...45 - Methods...46 - Results - Samples...48 - Hatch-year birds with hematozoan infections...48 - Relative leukocyte profile...48 - Discussion - Hatch-year birds with hematozoan infection...49 - Seasonal effects on relative leukocyte profile...49 - Seasonal effect on physiological stress and circulating hematozoa...50 -Acknowledgments...51 - Literature cited...52 Chapter 5: Total feeding rate of a male Bicknell s Thrush concurrently feeding two broods...57 - Abstract...58 - Introduction...59

vi - Field site description...60 - Methods...60 - Results...60 - Discussion...61 - Acknowledgments...62 - Literature cited...62 Chapter 6: Comparing feeding rates and physiological stress of two concurrently breeding song-birds with different mating systems; do many beaks make light work?..67 - Abstract...68 - Introduction...69 - Methods...71 - Results...73 - Discussion...74 - Acknowledgments...77 - Literature cited...77 Chapter 7: Conclusions...86 - Literature cited...88 Legend for appendices...91 Appendix A: Adult banding records...92 Appendix B: Chick and hatch year banding records...99 Appendix C: Proportion of feeds...102 Appendix D: Total feeding rates...103 Appendix E: Feeding rate at nest MG/YLG...104 Curriculum Vitae

vii List of Tables Table 2.1: The infection rate of hematozoa in adult Bicknell s and Swainson s Thrush, in the Christmas Mountains, NB, Canada. Birds were sampled between 26 May and 3 August, 2008 to 2010...26 Table 2.2: Analysis of white blood cell profiles of Bicknell s and Swainson s Thrush in NB, Canada. We used Type III ANOVA with year, species and age as factors. Significant results are in bold. The columns are cell types. Eos = Eosinophil, Mon = Monocyte, Bas = Basophil, Het = Heterophil, Lym = Lymphocyte, H:L = Heterophil to Lymphocyte ratio...27 Table 2.3: Table 2.3. Analysis of white blood cell profile of Bicknell s and Swainson s Thrush in NB, Canada, after pooling data by year, species, and age. We used type III ANOVA with week, sex, and parasite species as factors. Significant results are in bold. Due to the interaction between week and sex, for basophils, the significant effect from week should be ignored. The columns are cell types and ratios. Eos = Eosinophil, Mon = Monocyte, Bas = Basophil, Het = Heterophil, Lym = Lymphocyte, H:L = Heterophil to Lymphocyte ratio...28 Table 2.4: Analysis of the condition of male Bicknell s and Swainson s Thrush in NB, Canada. Condition was calculated from the scaled mass index. a) Type III ANOVA shows no effect of year, age, or species. b) After pooling data by year, age and species, Type III ANOVA showed no effect of week or parasite infection on condition. Of the 108 sampled birds, 56 were infected by Leucocytozoon, 7 by Trypanosoma, 13 by

viii Leucocytozoon and Trypanosoma, and 11 by Leucocytozoon and Haemaproteus. 21 birds had no parasites...29 Table 3.1: Analysis of leukocyte profiles in Bicknell s and Swainson s Thrush, sampled between 23 July and 2 August of 2009 and 2010, in NB, Canada. Type III ANOVA showed an effect of trombiculid infestation on relative basophil counts. No other cell line was affected by trombiculid larvae infestation...39 Table 5.1: Feeding rates of adult Bicknell s Thrush, at two broods which shared a male provider, during summer 2010, in the Christmas Mountains, NB, Canada. Feeding rate is given in feeds per hour. Individuals are labelled with their leg-band colours followed by their sex (Ex: female banded Light Green/Mauve over Medium Blue = LGM/MB F). The unbanded male may be more than one bird. The sole chick from BN1 was depredated during the evening of 4 July... 64 Table 6.1: Proportion of feeds at nests of Bicknell s (n=5) and Swainson s Thrush (n=4), in New Brunswick, Canada, during the breeding seasons 2008 to 2010. Bicknell s Thrush females, at nests with 2 or more males, provided 54% of feeds (n=4, St Dev= 14%)...80 Table 6.2: Type III ANOVA showed no effect of age or year on H:L of Bicknell s and Swainson s Thrush, during the provisioning period (between 16 June and 23 July), so data were pooled...81 Table 6.3: Type III ANOVA showed no effect of species or sex on H:L during the provisioning period (between 16 June and 23 July)...82

ix List of Figures Figure 1.1: Two sites were selected for this study, due to their proximity to one another and co-occurrence of breeding Bicknell s and Swainson s Thrush. North Pole Mountain and Mount Mitchell are found in north-central New Brunswick, south of Mount Carleton Provincial Park. They are approximately 12.45 km apart, and share similar elevations (approximately 635 metres)...8 Figure 2.1: Intra-seasonal changes to white blood cell profile and hematozoan infection of Bicknell s and Swainson s Thrush in NB, Canada, during the summers of 2008 to 2010. Week one began on 26 May. a) Heterophils were significantly higher during weeks one, three, and eight (p= 0.036, 0.045, 0.023, respectively). Lymphocytes were significantly lower during weeks one and three (p= 0.004, 0.004, respectively). There was a significant negative correlation between heterophils and lymphocytes (r = -0.775, p < 0.0001) b) Heterophil to Lymphocyte ratio (H:L) was significantly higher during weeks one, three, and eight (p= 0.006, 0.016, 0.038, respectively) c) Types of hematozoa were not taken into account; only the mean number of species infecting individual birds was plotted. Week three was significantly high (p= 0.005). d) Condition, measured as standardized residuals, did not change significantly during the breeding season. There was a significant negative correlation between the mean number of hematozoa per individual and condition, during weeks two through nine (r=-0.745, p= 0.034). Week 1 was omitted from the analysis in order to minimize the interference from

x migration and lack of insect vectors, and week 10 was omitted due to low sample size...30 Figure 3.1: This Swainson s Thrush, with trombiculid larvae in its auditory canal and a tick above the right eye, was caught on 27 July, 2010, in NB, Canada...40 Figure 3.2: Trombiculid mite larva, still clinging to some flesh. Photo was taken through a compound microscope at approximately 300X magnification (10X objective, 10X ocular, with 3X zoom digital camera)...41 Figure 3.3: Body condition of Swainson Thrush males, either infested or not, caught in NB, Canada, between 23 July and 2 August of 2009 and 2010. Condition was not affected by larval mite infestation (P=0.247). Means are indicated by +, median by a horizontal line, upper and lower quartiles by the top and bottom of the box, and maximum and minimum by the whiskers...42 Figure 4.1: Figure 4.1: Relative leukocyte profile of Bicknell s Thrush sampled during the breeding season in New Brunswick (NB; n=23), Canada, during 2008 to 2010, and while overwintering in the Dominican Republic (DR; n=22) in 2010. Means are indicated by +, median by a horizontal line, upper and lower quartiles by the top and bottom of the box, and maximum and minimum by the whiskers. Monocytes were not included in the figure as they were not suitable for the scale, mean counts for New Brunswick birds were 0.9 and in Dominican Republic 0.0. Birds from the Dominican Republic had significantly fewer monocytes (0.9 cell less, Tukey s HSD= 0.045), fewer heterophils (8.2 cells less, Tukey s HSD = 0.003), but more basophils (18.3 cells more, Tukey s HSD < 0.0001) than New Brunswick birds. Lymphocyte counts (Tukey s HSD

xi = 0.407), and eosinophil counts (Tukey s HSD = 0.077) were not significantly different...55 Figure 4.2: Heterophil to lymphocyte ratio (H:L) of Bicknell s Thrush sampled during the breeding season in New Brunswick (NB; n=23), Canada, during 2008 to 2010, and while overwintering in the Dominican Republic (DR; n=24) in 2010. Means are indicated by +, median by a horizontal line, upper and lower quartiles by the top and bottom of the box, and maximum and minimum by the whiskers. H:L was significantly higher (Tukey s HSD = 0.014) for New Brunswick birds (mean 0.290), compared to those in the Dominican Republic (0.134)...56 Figure 5.1: This male Bicknell s Thrush, banded medium-green over yellow/light-green (MG/YLG) was first captured at Mount Mitchell, NB, in 2008. He is at least 4 years old. The nest on the left is BN1, and contains one chick. The nest on the right is BN2, and contains four chicks...65 Figure 5.2: Feeding rate of a male Bicknell s Thrush feeding two broods in NB, Canada, during the summer of 2010. Individual feeds for the first nest (BN1) are shown as diamonds and those from the second nest (BN2) as squares. The start and end of the videos are labelled in black. The sole chick from BN1 was depredated during the evening of 4 July...66 Figure 6.1: The total feeding rates at nests of Bicknell s and Swainson s Thrush, in NB, Canada. Data from the 2008 to 2010 breeding seasons were pooled. The number of nestlings multiplied by the age of nestlings (days) was better correlated to feeding rate (R 2 =0.71), than age of nestlings (R 2 =0.43) or number of nestlings alone (R 2 =0.59).

xii Bicknell s Thrush feeding rates were not significantly different than those of Swainson s Thrush (Tukey s HSD=0.088)...83 Figure 6.2: Heterophil to lymphocyte ratio (H:L) of male and female Bicknell s Thrush and Swainson s Thrush, sampled between June 16 and 23 July in NB, Canada. Samples from 2008 to 2010 were pooled. Means are represented by + (labelled to the right of the box plot), median by a horizontal line, upper and lower quartiles by the top and bottom of the box, and maximum and minimum by the whiskers...84 Figure 6.3: Figure 6.3: Heterophil to lymphocyte ratio (H:L) of Bicknell s and Swainson s Thrush, provisioning or not, superimposed on H:L data collected from all birds of both species in NB, Canada. Data from 2008 to 2010 were pooled. Nestling provisioning occurred during weeks four through eight of sampling, between 16 June and 23 July. Nestling provisioning at individual nests was usually completed within two weeks, starting mid June, but failed nests and subsequent re-nesting attempts prolonged the period during which birds might have been provisioning. During this period, H:L was significantly higher for birds who were feeding nestlings, compared to those whose nest had failed (Tukey s HSD =0.007)...85

1 Chapter 1: INTRODUCTION General introduction and objectives We often think of birds as being monogamous, and rightly so with 90% of bird species having some degree of pair bonding with just one partner (Lack 1968), yet there are some species whose mating system is based entirely on pairing with multiple partners during one breeding season. Bicknell s Thrush (Catharus bicknelli) is just such a species. It is one of the rarest song-birds in Atlantic Canada, and one of only 2 species in Canada that exhibit a polygynandrous mating system (Briskie 1998, Rimmer et al. 2001). Bicknell s Thrush males attempt to pair with more than 1 female and females with more than 1 male, resulting in up to four males feeding chicks at one nest (Strong et al. 2004). Are breeding birds less physiologically stressed while providing for nestlings, if there is more than one male provider? The extent of a bird s parental effort and, consequently its physiological stress, may be dictated by its mating system. Bicknell s Thrush has a biased sex ratio, with approximately 2 males for every female and an average of 2 male providers per nest (Strong et al. 2004), while Swainson s Thrush (Catharus ustulatus) is socially monogamous, with only one male feeder at the nest (Mack and Yong 2000). The difference in the number of male providers at the nests of these closely-related species implies different levels of parental effort in the two species, creating a unique opportunity to study the effects of mating systems on physiological stress. High physiological stress is caused by intense physical activity, starvation, and dehydration, so theoretically, the more an adult provides for chicks the greater its physiological

2 stress. Having multiple male feeders should lower the feeding rate of individual providers, unless all of the adults provide at a similarly high rate, in which case it would increase the total number of feeds for nestlings. In New Brunswick, Swainson s Thrush is the most numerous thrush that cooccurs with Bicknell s Thrush (Nixon et al. 2001, McKinnon 2009). Because both species of Thrush nest in the same area, their sympatry creates a natural experiment to compare their mating systems without having to account for differences in elevation, weather, and resource availability. It has been suggested that Bicknell s Thrush engages in polygynandrous mating in order to overcome the harsh environment and low resource availability of their breeding grounds (Strong et al. 2004). Since polygynandrous mating systems are rare in birds (Goetz et al. 2003), any knowledge we can gain about such a system may lead to a better understanding of the evolution of mating systems. The breeding range for Bicknell s Thrush extends from the north-eastern United States, east to Cape Breton Island, Nova Scotia, and consists of several isolated highelevation populations (Bredin and Whittam 2009). Unfortunately, this species was disappearing from the high elevation forests of New Brunswick, Canada, at an annual decline of 17% during the years 2002-2009 (Bird Studies Canada unpubl. data, from IBTCG 2010). In November 2009, the species was up listed to Threatened status by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC 2009).There are many possible anthropogenic causes, such as habitat degradation at wintering and breeding grounds (IBTCG 2010), and incidental take during tree harvesting and precommercial thinning (Bredin and Whittam 2009). Pre-commercial thinning occurs when a stand is 10 to 15 years of age, the same age range for stands in which Bicknell s

3 Thrush prefers to breed (Chisholm and Leonard 2008, Bredin and Whittam 2009). Considering 91% of Canadian Bicknell s Thrushes are found in lands managed for logging (COSEWIC 1999), this is of particular concern for the Canadian population of Bicknell s Thrush. This is reflected in one of the International Bicknell s Thrush Conservation Group s action plans, to partner with timber companies and management agencies to develop and implement Best Management Practices for Bicknell s Thrush habitat (IBTCG 2010). This project was one in a series whose goals were to better understand Bicknell s Thrush breeding behaviour and habitat use in New Brunswick (Fraser unpl., Nixon et al. 2001, Bredin and Whittam 2009, McKinnon 2009). This thesis is presented in articles format. Antony Diamond contributed in the editing of all seven chapters of this dissertation. Kevin Fraser and Emily Fraser contributed to data collection in 2008, and helped in the editing of chapters 2, 4, and 6. Lynda Leppert was responsible for white blood cell counts, hematozoon identification, and the editing chapters 2, 3, 4, and 6. Rebecca Holberton contributed financially for blood smear analysis and helped in the editing of chapters 2, 3, 4, and 6. Samples of mite larvae were sent to Manigandan Lejeune-Virapin (University of Calgary; Faculty of Veterinary Medicine) and James W. Mertins (United States Department of Agriculture; Animal and Plant Health Inspection Service) for identification (chapter 3). Hubert Askanas was responsible for; i) the design of the research proposal, ii) field work for 2009 and 2010 (field assistant in 2007 and 2008) including all aspects of logistics, finances, and data collection, iii) all data analyses, and iv) manuscript preparation.

4 General methods Study sites.- I selected 2 sites in the Christmas Mountains, north-central New Brunswick, due to their relatively high densities of Bicknell s Thrush and co-occurrence with Swainson s Thrush (Nixon et al. 2001, McKinnon 2009). The two sites, at Mount Mitchell and North Pole Mountain, are at similar elevations (635 metres) and approximately 12.45 km apart (Fig 1.1). The stands were composed of thick earlyregeneration Balsam Fir (Abies balsamea), in which Bicknell s Thrush tend to nest (Askanas pers. obs., Nixon et al. 2001, McKinnon 2009, Aubry et al. 2011). Starting in late May, these sites were monitored for arriving Bicknell s and Swainson s Thrushes. Sampling of thrushes occurred through to early August (from 2008-10). Detailed accounts of methods can be found in the individual chapters. Introduction to Chapters.- Upon analyzing blood smears in 2008, I realized that the target species were parasitized by hematozoa (also known as blood parasites). In 2009, I also found larval trombiculid mites infesting our birds. Since this study involved relative leukocyte profiles, using heterophil to lymphocyte ratio as a surrogate for corticosterone in the analysis of physiological stress, I tested for any effects of hematozoa (Chapter 2) and trombiculid mites (Chapter 3) on relative leukocyte profiles. For the purpose of this study, physiological stress was considered to be an increase in physical activity and/or decrease in energy stores caused by stressors such as provisioning, defending territory, courting, incubating, and the decrease in food intake due to the aforementioned stressors. During high physiological stress, birds release corticosterone into their bloodstream (Remage-Healey and Romero 2001).

5 Corticosterone mobilizes energy stores, in the form of fatty acids and glucose, giving the birds extra energy (Dhabhar et al. 1995). Corticosterone also affects relative leukocyte profiles, as it increases circulating heterophil numbers and sequesters lymphocytes (Davis et al. 2008). Heterophils belong to the innate immune system (primary phagocytic cell; Campbell 1995, Davis 2008), which also includes basophils (initiate inflammation and have a role in allergic reaction; Campbell 1995), eosinophils (antiparasitic; Rupley 1997), and monocytes (phagocytic; Rupley 1997, Davis 2008). Lymphocytes belong to the adaptive immune system, and are responsible for antigen production (Campbell 1995). As Bicknell s Thrush is a threatened species, I also tested for effects of parasites on condition. I calculated body condition by using morphometrics. Birds in good condition were those with high mass, given their body size (Peig and Green 2010). If two birds had the same mass but one had a larger body size, the smaller bird would be in better condition. In 2010, after the International Bicknell s Thrush Conservation Group meeting in the Dominican Republic, I went into the field and was able to sample overwintering Bicknell s Thrushes. We tested for differences in physiological stress, relative leukocyte profiles and hematozoa in overwintering birds (Chapter 4). Also in 2010, I was able to concurrently record video of two Bicknell s Thrush broods sharing a male provider (Chapter 5). I used these data from feeding-rate analysis to help explain the differences in physiological stress during the nestling provisioning stage of the breeding season, at which point I tested for effects of mating systems on physiological stress (Chapter 6).

6 Literature cited Aubry, Y., A. Desroches, and G. Seutin. 2011. Response of Bicknell s Thrush (Catharus bicknelli) to boreal siviculture and forest stand edges: a radio-tracking study. Journal of Canadian Zoology 89:474-481. Bredin, K., and B. Whittam. 2009. Conserving the Bicknell s Thrush. Bird Studies Canada 25pp. [Online] available at www.bsc-eoc.org/download/acbithrpt.pdf Briskie, J. V., R. Montgomerie, T. Poldmaa, and P. T. Boag. 1998. Paternity and paternal care in the polygynandrous Smith's Longspur. Behavioral Ecology and Sociobiology 43(3):181-190. Campbell, T.W. 1995. Avian Hematology and Cytology. Iowa State University Press, Ames, Iowa. Chisholm, S. E., and M. L. Leonard. 2008. Effect of forest management on a rare forest specialist, the Bicknell s Thrush (Catharus bicknelli). Canadian Journal of Zoology 86:217-223. COSEWIC. 1999. Status report of Bicknell s Thrush Catharus bicknelli in Canada. Committee on the Status of Endangered Wildlife in Canada, Ottawa, Ontario. v+43 pp. COSEWIC. 2009. COSEWIC assessment and status report on the Bicknell's Thrush Catharus bicknelli in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa, Ontario. vii + 44 pp. Davis, A. K., D. L. Maney, and J. C. Maerz. 2008. The use of leukocyte profile to measure stress in vertebrates: a review for ecologists. Functional Ecology 22:760-772. Dhabhar, F. S., A. H. Miller, B. S. McEwen, and R. L. Spencer. 1995. Effects of stress on immune cell distribution dynamics and hormonal mechanisms. Journal of Immunology 154:5511-5527. Goetz, J. E., K. P. McFarland, and C. C. Rimmer. 2003. Multiple paternity and multiple male feeders in Bicknell's thrush (Catharus bicknelli). Auk 120(4):1044-1053. International Bicknell s Thrush Conservation Group. 2010. A Conservation Action Plan for Bicknell s Thrush (Catharus bicknelli). J. A. Hart, C. C. Rimmer, R. Dettmers, R. M. Whittam, E. A. McKinnon, and K. P. McFarland, Eds. International Bicknell s Thrush Conservation Group. [Online] available at www.bicknellsthrush.org Lack, D. 1968. Ecological Adaptations for Breeding in Birds. Methuen, London. 409 pp.

7 Mack, D. E. and W. Yong. 2000. Swainson's Thrush (Catharus ustulatus), The Birds of North America Online. (A. Poole, Ed.). Ithaca: Cornell Lab of Ornithology. McKinnon, E. A. 2009. Bicknell s Thrush (Catharus bicknelli) in managed forests: nestsite election, diet, and co-occurrence with Swainson s Thrush (C. ustulatus). M.Sc. Thesis, University of New Brunswick, Fredericton, New Brunswick. 127 pp. Nixon, E. A., S. B. Holmes and A. W. Diamond. 2001. Bicknell s Thrushes (Catharus bicknelli) in New Brunswick clear-cuts: their habitat associations and co-occurrence with Swainson s Thrushes (Catharus ustulatus). Wilson Bulletin 113:33-40. Peig, J., and A. J. Green, 2010. The paradigm of body condition: a critical reappraisal of current methods based on mass length. Functional Ecology 24:1323-1332. Remage-Healey, L., and M. Romero. 2001. Corticosterone and insulin interact to regulate glucose and triglyceride levels during stress in a bird. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology 281:994-1003. Rimmer, C. C., K. P. McFarland, W. G. Ellison, and J. E. Goetz. 2001. Bicknell s Thrush (Catharus bicknelli). In The Birds of North America, No. 592 (A. Poole and F. Gill, eds.). The Birds of North America, Inc., Philadelphia, PA. Rupley, A. E. 1997. Manual of Avian Practices. 556 pp. W.B. Saunders Company, Philadelphia PA. Strong, A. M., C. C. Rimmer, and K. P. McFarland. 2004. Effects of prey biomass on reproductive success and mating strategy of Bicknell s Thrush (Catharus bicknelli), a polygynandrous songbird. Auk 121(2):446-451.

8 Figures Figure 1.1: Two sites were selected for this study, due to their proximity to one another and co-occurrence of breeding Bicknell s and Swainson s Thrush. North Pole Mountain and Mount Mitchell are found in north-central New Brunswick, south of Mount Carleton Provincial Park. They are approximately 12.45 km apart, and share similar elevations (approximately 635 metres).

9 Chapter 2: SUSCEPTIBILITY TO HEMATOZOAN INFECTION IS INFLUENCED BY HIGH PHYSIOLOGICAL STRESS DURING THE BREEDING SEASON, AND IS CORRELATED WITH NEGATIVE EFFECTS ON CONDITION Hubert Askanas 1, Antony Diamond 2, Kevin C. Fraser 3, Emily A. Mckinnon 4, Lynda Leppert 5, Rebecca Holberton 6 1,2,3,4 Department of Biology, University of New Brunswick, 10 Bailey Drive, Fredericton, NB, Canada, E3B 5A3. 5 Georgia Aquarium, Veterinary Services, 225 Baker St NW, Atlanta, GA 30313 6 School of Biology & Ecology, University of Maine, 221 Murray Hall, Orono, ME 04469 k439e@unb.ca 1, diamond@unb.ca 2, fraserkev@gmail.com 3, emilymck@yorku.ca 4, uuncia@gmail.com 5, rebecca.holberton@maine.edu 6 Corresponding author: Hubert Askanas, 110 Taurus Drive, Hanwell, NB, Canada, E3C 1N1, k439e@unb.ca 3 Emily A. Mckinnon is now at the Department of Biology, York University, Toronto, ON, M3J 1P3

10 Abstract Bicknell s Thrush is declining rapidly in New Brunswick, Canada. Since infection with hematozoa is known to be detrimental to a bird s fitness, we were interested in gauging parasitic effects on bird condition and determining whether this species is at higher risk of infection during the most physiologically stressful periods of the breeding season. We included Swainson s Thrush in the study, as we expected hematozoa were similarly affecting both species. Of 111 Swainson s Thrush and 59 Bicknell s Thrush sampled between 2008 and 2010, 69% were parasitized by Leucocytozoon debrueili, 24% by Haemoproteus neseri and/or Haemaproteus minutus, 8% by Trypanosoma avium, and 5% by unknown Microfilaria. We found no parasites in 39% of Bicknell s Thrush and 10% of Swainson s Thrush. There was no effect of hematozoa on the relative leukocyte profiles of Bicknell s and Swainson s Thrush. Most of the variation in relative leukocyte profile was attributed to the physiological stress experienced during the breeding season (high heterophil to lymphocyte ratio during arrival, courting/defending territory, and feeding nestlings). As relative leukocyte counts do not reflect the potential immune function of birds, and seem not to change in response to the presence/absence of hematozoa, they should be reserved for the analysis of physiological stress. Significant peaks in physiological stress occurred concurrently with one significant and one non-significant peak in circulating hematozoa, and increases in hematozoa were correlated with poor body condition. Despite the correlation, there were no statistically significant effects of Leucocytozoon, Trypanosoma, Leucocytozoon + Trypanosoma, or Leucocytozoon + Haemoproteus infection on condition.

11 Introduction Bicknell s Thrush is declining in New Brunswick, Canada, at a rate of 17% annually (Bird Studies Canada unpubl. data, from IBTCG 2010), and has recently been up-listed to Threatened status by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC 2009). During the 2008 field season we found Bicknell s Thrush (Catharus bicknelli) to be heavily parasitized both internally and externally, so we became concerned about the effects of parasites on their health. Swainson s Thrushes (Catharus ustulatus) were included in this study, as they are closely related to, breed sympatrically with, and are more numerous than Bicknell s Thrush in our study area. Since many studies have demonstrated the negative effects of blood parasites (also known as hematozoa) on many bird hosts (Dufva 1996, Ots and Hõrak 1998, Merino et al. 2000, Schrader et al. 2003), we expected that hematozoa were similarly affecting Bicknell s and Swainson s Thrush. Hematozoa included Leucocytozoon debrueili, Trypanosoma avium, Haemoproteus neseri, and Haemaproteus minutes. We also found an unknown Microfilaria. Leucocytozoon species have been shown to reduce body condition in Cirl Buntings (Emberiza cirlus) (Figuerola et al. 1999), Trypanosoma to reduce reproductive success in Aquatic Warblers (Acrocephalus paludicola) (Dyrcz et al. 2005), Haemoproteus to reduce reproductive success and condition in Blue Tits (Parus caeruleus) (Merino et al. 2000), and Microfilaria to be fatal (Simpson et al. 1996), By sampling throughout the breeding season, we were able to monitor hematozoa prevalence during the time in the birds yearly cycle when they were expected to be most susceptible to blood parasites. Proceeding with the hypothesis that

12 wild populations of birds have persistent hematozoan infections at levels too low to be pathogenic (Atkinson and Van Riper 1991), and that susceptibility to parasitic infection increases during high physiological stress (Weatherhead and Bennett 1991, Nordling et al. 1998), we predicted that high physiological stress during the breeding season would be positively correlated with prevalence of hematozoa, and negatively correlated with condition. In addition, we predicted that birds infected by hematozoa would have lower condition. We also wanted to monitor changes to relative leukocyte profiles, and since eosinophils are hypothesized to be the major anti-parasitic cell line (Rupley 1997), we predicted that birds infected by hematozoa would have higher numbers of circulating eosinophils. Methods Our study sites were located in the Christmas Mountains, north-central New Brunswick, Canada. We captured birds five days a week at dusk and dawn, in arrays of mist-nets (weather permitting). The 2008 to 2010 field seasons were two and a half months each, with data presented by week (1-10) starting 26 May. We took several morphometric measurements including mass, tarsus length, tail length, wing chord length, and bill length. Condition indices were chosen by correlation analysis, and condition was calculated using scaled-mass index (Peig and Green 2009). The effects of haematozoa on condition and leukocyte profiles were tested using type III ANOVA, unless otherwise stated. Principal component analysis was performed on white blood cell profiles to determine which cell lines were responsible for the greatest source of variation. All statistical tests were performed using XLSTAT 2011, using α=0.05.

13 Blood smears were made from the peripheral blood of Bicknell s and Swainson s Thrush. Blood was sampled by puncturing a brachial vein with a sterile 30 gauge needle, and a total of 80 µl of blood was collected in two heparinised microcapillary tubes. Blood smears were prepared immediately after collection. The smears were stained with Volu-Sol Wright-Giemsa (Utah) stain, and 100 white blood cells were counted under a compound microscope. The smears were analyzed for endoparasite presence/absence and intensity. Mean number of parasite species per individual was calculated as the sum of parasites per individual in any given week, divided by the number of individuals. If half of the birds caught had one parasite species and the other half had two, the mean number of parasite species per individual would be 1.5. The number of Haemoproteus sp. counted within 10,000 red blood cells is considered the blood parasite intensity level. Leucocytozoon sp. and Trypanosoma sp. were not included in intensity levels because they reside outside the red blood cells. Note that relative white blood cell counts do not represent a bird s ability to fight infection, rather they show the proportion of cell types circulating in blood (Dhabhar 2002). Relative white blood cell counts also do not account for leukocytes which are sequestered into other tissues, nor the ability to mobilize those leukocytes for an immune response. We were testing whether hematozoa changed the proportions of circulating leukocytes, not the bird s ability to mount an immune response. Heterophil to Lymphocyte ratio (H:L) was used as the indicator of physiological stress. Some researchers believe that it is equivalent to, if not better than, sampling glucocorticoid levels as an indicator of physiological stress (Davis et al. 2008). Heterophils are phagocytic cells which increase in circulation in response to stress,

14 inflammation and infection (Harmon 1998). Lymphocytes are responsible for immunoglobulin production, otherwise known as humoral immunity (Campbell 1996). These two blood cell types are sensitive to the stress caused by increased physical activity. During high physiological stress, birds release corticosterone into their bloodstream (Remage-Healey and Romero 2001). Corticosterone has an effect on relative leukocyte profiles, as it decreases circulating lymphocytes and increases circulating heterophils (Davis et al. 2008). It is more energetically costly to maintain and deploy lymphocytes than heterophils (Apanius 1998), so during high physiological stress, lymphocytes are sequestered into other tissues and heterophils are released into circulation. The negative correlation between heterophils and lymphocytes, during high physiological stress, gave rise to the method of using the ratio of the two cell types (H:L) as an indicator of physiological stress (Gross and Siegal 1983, Ots et al. 1998, Ilmonen et al. 2003, Owen and Moore 2006). Corticosterone also increases and maintains the amount of circulating glucose and lipids (Remage-Healey and Romero 2001), giving the birds extra energy (Dhabhar et al. 1995). Theoretically, the release of stored body reserves should lower a bird s mass, thereby decreasing its body condition. We also tested the effects of parasitism on the three other white blood cells; monocytes, basophils, and eosinophils. Monocytes are thought to be phagocytic cells (Rupley 1997, Davis 2008); basophils to be responsible for allergic reactions and the initiation of inflammation (Campbell 1995); and eosinophils to be anti-parasitic (Rupley 1997).

15 We used scaled mass index using the following equation, as in Peig and Green (2009):, where and are the measurements (Mass and some type of body Length) from individual i, and is the mean length from the sampled population. We calculated bsma for each species by calculating the slope of an ordinary least squares regression of ln(mass) on ln(right wing), as they were the most closely correlated measurements for both species (R²=0.137 p=0.031 in Bicknell s, R²=0.083 p=0.009 in Swainson s), and by dividing the slope by its Pearson s correlation coefficient r. Outliers were removed, and scaled mass was calculated for each species separately. Condition was measured as the standardized residual of scaled mass versus date. Results Samples.- During the 2008 to 2010 breeding seasons we collected 170 blood samples, of which 130 were useable for white blood cell counts. Of those 130 samples, 25 were from Bicknell s Thrush and 105 from Swainson s Thrush. A complete overview of the parasites infecting adult birds, in the original 170 samples, can be found in Table 2.1. Hematozoa and leukocyte profile.- We removed outliers and any parasite species which had fewer than five individuals. The remaining 23 Bicknell s Thrush (8 female and 15 male) and 85 Swainson s Thrush (33 female, 50 male, 2 unknown sex) samples were used to test if we could pool data by year, species, and age. Values for R 2, F, and P can be found in Table 2.2. Of those 108 birds, 56 were infected by Leucocytozoon, 7 by

16 Trypanosoma, 12 by Leucocytozoon and Trypanosoma, 11 by Leucocytozoon and Haemaproteus, and 21 with no parasites. Haemaproteus parasite loads for the 11 infected birds are as follows; 1, 2, 6, 6, 10, 13, 20, 32, 56, 71, and 94 parasites per 10, 000 red blood cells. Type III ANOVA showed no effect of age or species on white blood cell profiles, but year did have a significant effect on monocytes: in 2008, the average monocyte count was 3.487, dropping to 2.801 in 2009 and 0.071 in 2010. We used week, sex, and parasite species as factors to test for effects of Leucocytozoon, Trypanosoma, Leucocytozoon + Trypanosoma, and Leucocytozoon + Haemoproteus on condition and relative leukocyte profile. The first week began with the first data point, on 26 May. There was a significant interaction between sex and week on basophils, with females having higher counts during the first week (p= 0.013). There was a significant effect of week on heterophils and lymphocytes (Fig 2.1a), and heterophil to lymphocyte ratio (H:L) (Fig 2.1b). For H:L, weeks one, three, and eight were significantly higher than other weeks (Fig 2.1b). Values for R 2, F, and P can be found in Table 2.3. There was a significant negative correlation between heterophils and lymphocytes (r = -0.775, p < 0.0001; Fig 2.1a). Mean number of parasite species, infecting individual birds, can be seen in Figure 2.1c. There were significantly more parasite species found in blood smears collected in week three (Fig 2.1c). Principal component analysis revealed that heterophils and lymphocytes were responsible for 72% of variation on factor 1, which accounted for 44% of the total variation in relative leukocyte profiles. Eosinophils were responsible for 63% of variation on factor 2, which accounted for 24% of total variation. There was no correlation between H:L and condition (r=-0.468, p=0.246).

17 Hematozoa and condition.- Initial analysis of condition was confounded by gravid females, so male condition was used to determine any effects from parasitic infection. We used 49 Bicknell s Thrush and 89 Swainson s Thrush to calculate scaled mass index (Fig 2.1d). To test for effects of parasites on condition, we used 54 birds infected by Leucocytozoon, 8 by Trypanosoma, 8 by Leucocytozoon and Trypanosoma, 9 by Leucocytozoon + Haemaproteus, and 19 without any hematozoa detected. Type III ANOVA showed no effect of year, species, or age (Table 2.4a), so data were pooled. Type III ANOVA showed no significant effect of the aforementioned parasite species, or combination of parasite species, on condition (Table 2.4b). There was, however, a significant negative correlation between mean number of hematozoon per individual and mean condition (r=-0.745, p= 0.034). Discussion Effects of hematozoa on leukocyte profiles.- Taking week and sex into account, hematozoa had no effect on white blood cell profiles (Table 2.3). The majority of the variation in white blood cell profiles of Bicknell s and Swainson s Thrush, during the breeding season, can be explained by changes in heterophil and lymphocyte counts (the sum of variation, of heterophils and lymphocytes, represents 72% of the variation on factor 1, in a principal component analysis). High physical activity resulted in high H:L, which can be seen in Figures 2.1a and 2.1b. There was a significant negative correlation between heterophils and lymphocytes (r = -0.775, p < 0.0001) which corresponded with

18 high energy events. Year, species and age had no effect on heterophils, lymphocytes or H:L (Table 2.2). Heterophils, lymphocytes, and H:L were all significantly affected by week, a surrogate for different stages of the breeding season (Table 2.3). The H:L ratios for weeks one, three and eight were all significantly higher than the seasonal mean (Fig 2.1b). During the first to fourth week of sampling, the birds were recovering from migration, claiming territory, courting, mating, and nest building. During the fourth to eighth week, most birds were feeding nestlings. The chronic physical activity during this stage, which increased as chicks grew, caused H:L to peak during the eighth week of sampling. As mentioned previously, it has been hypothesized that the increase in physiological stress during migration and breeding may compromise a bird s immune system (breeding: Ots and Hõrak 1996, Nordling et al.1998, Al-Murrani et al. 2002, Verhulst et al. 2005; migration: Owen and Moore 2006). High physiological stress could allow new infections to occur, and latent infections to flare up (Weatherhead and Bennett 1991). The blood smears revealed hematozoa infections throughout the breeding season. Unfortunately we cannot determine parasite load of Leucocytozoon or Trypanosoma using blood smears, so we could not determine if parasite loads increase during the most stressful periods of the breeding season. The intensity of Haemaproteus infection (N = 11, range 1 to 94 parasites per 10,000 red blood cells) had no apparent effect on leukocytes. We saw a relationship between high H:L values and an increase in the mean number of hematozoa species infecting individuals (except for the first week of

19 sampling, Figs 2.1b and 2.1c). A connection between physiological stress and hematozoa has been observed in other studies, where birds with high H:L ratios (Al- Murrani et al. 2002), or breeding birds (Nordling et al. 1998), were more susceptible to infection. During our first week of sampling, H:L was high but the number of parasite species per individual was low, most likely due to the low numbers of vectors (biting insects) while the birds were arriving in spring. Otherwise, peaks in the mean number of parasite species infecting individuals occurred during the same weeks in which H:L peaked, reinforcing the conclusions made by Al-Murrani et al. (2002) and Nordling et al. (1998); that high physiological stress does make a bird more susceptible to parasitic infection. Despite the correlation, H:L was not a good predictor of infection, nor was it expected to be, as high physiological stress does not guarantee that a bird will be bitten by an infected insect, and birds infected by many parasite species are not necessarily physiologically stressed at the time of sampling. The mean number of parasite species found in individual hosts was significantly higher in the third week of sampling, which corresponds to a significant spike in H:L during the same week. A similar increase in parasites and H:L was apparent in week eight, but it was not significant for parasites. Aside from heterophils and lymphocytes, eosinophils accounted for a large portion of the variation in the principal component analysis (63% of variation on factor 2, which accounted for 24% of total variation). We were unable to determine the cause of variation in eosinophil counts from our samples. Eosinophils are thought to be involved in fighting parasitic infection as well as promoting inflammation (Rupley 1997). Parasitized birds had slightly elevated counts (2.0 cells) but the difference was not significant. Unexpectedly, some birds with no hematozoa had high eosinophil counts

20 (range between 0 and 30 cells), and some parasitized birds had low counts (Leucocytozoon = 0 to 36, Leucocytozoon and Haemaproteus = 1 to 31, Leucocytozoon and Trypanosoma = 0 to 33, Trypanosoma = 0 to 30). It is also hypothesized that high physiological stress will lower the number of circulating eosinophils (Cardinet 1964). Active birds had slightly lower levels (2.5 cells), but again the difference was not significant. Monocyte counts for New Brunswick birds differed significantly by year, though the cause is unknown (Table 2.2). Monocytes are thought to phagocytise foreign particles and defend against bacteria (Rupley 1997, Davis 2008), but there was no significant difference in monocyte counts of parasitized versus non-parasitized birds overall. There was a significant effect of the interaction of sex and week on basophils, with females having significantly higher counts during the first week of sampling, and males showing a brief non-significant increase in week three. Very little is known about the function of basophils in birds (Rupley 1997), though it is thought they are responsible for allergic reactions and the initiation of inflammation (Campbell 1995). The high basophil counts could be explained by the developing brood patch in females and cloacal protuberance in males, though this is speculative. Recent work on mice has shown that basophils have roles in parasite defence (Min 2008, Shen et al. 2008, Min et al. 2012), aiding in TH-2 immunity, but we found no effect of, or correlation with, parasite infection. As basophils aid lymphocytes in humoral immunity, and lymphocytes are sequestered during the breeding season, it is possible that basophils are being sequestered as well.

21 The effects of hematozoa on male condition.- After pooling by year, species and age (Table 2.4a) we found no significant effects of hematozoa on the condition of male Bicknell s and Swainson s Thrush (Table 2.4b). Our findings are similar to other studies (Baker 1976, Fallis and Desser 1977), suggesting that hematozoa do not cause pathological infections. It has been hypothesized that migration acts as a filter for heavily parasitized birds (Booth and Elliot 2002), and therefore those which survive the journey to the breeding grounds have low level infections. In support of this, Garvin et al. (2006) concluded that several migrant species arriving on the northern coast of the Gulf of Mexico, during spring migration, were in better condition if they were not infected by hematozoa. In addition, our birds had the lowest number of hematozoan species during the first week of sampling (Fig 2.1c, mean 0.667). As mentioned before, we cannot determine parasite loads for Leucocytozoon, or Trypanosoma. In Haemaproteus, there was no correlation between parasite intensity and condition (N = 11, range 1 to 94 parasites per 10,000 red blood cells). Having no data on the level of parasitism that would negatively affect our birds (or parasite loads for Leucocytozoon and Trypanosoma), we can neither refute nor prove that the parasitized birds arrived in New Brunswick with low-level infections, and that heavily parasitized birds would have shown pathological symptoms. In support of the theory that hematozoa maintain chronic infections in the wild (Atkinson and Van Riper 1991) and at levels low enough to make it difficult to show pathogenicity (Atkinson and Van Riper 1991, Weatherhead and Bennett 1991), we did find a significant negative correlation between the mean number of hematozoa per individual and mean condition (Figs 2.1c, 2.1d), although ANOVA showed no effect by

22 single or paired parasite species (Table 2.4). It is difficult to separate the effects of physiological stress and parasitism on condition, as high physiological stress increases susceptibility to parasitic infection. Both H:L and mean number of parasite species per individual increase as condition decreases. Merino et al. (2000) suggest a medication experiment would be better at determining the negative impacts of hematozoa on condition. Acknowledgments We thank the New Brunswick Wildlife Trust Fund and Environment Canada for providing major funding for this project, Vermont Institute of Natural Sciences, Canadian Wildlife Service, and Dr. Jim Goltz for technical assistance (Department of Agriculture, Aquaculture and Fisheries; Veterinary Laboratory Services). This project was also supported by the Atlantic Laboratory for Avian Research (formerly Atlantic Cooperative Wildlife Ecology Research Network), the Department of Natural Resources, Bird Studies Canada, and Science Horizons (Environment Canada). And a big thanks to all the field assistants, Julia Hughs, Katy Leger, Graham Dixon- MacCallum, Greg Jongsma, Patrick Blake, Marie-Paule Godin, Richard Aracil, and Carl-Adam Wegenschimmel. Literature cited Apanius, V. 1998. Stress and immune defence. Advances in the Study of Behavior 27:133-153. Atkinson, C. T., and C. Van Riper III. 1991. Pathogenicity and epizootiology of avian hematozoa: Plasmodium, Leucocytozoon and Haemoproteus. In Bird-Parasite Interactions: Ecology, Evolution, and Behavior (J. Loye, and M. Zuk, Eds.), pp. 19-48. New York: Oxford University Press.

23 Al-Murrani, W.K., I. K. Al-Rawi, and N.M. Raof. 2002. Genetic resistance to Salmonella typhimurium in two lines of chickens selected as resistant and sensitive on the basis of heterophil/lymphocyte ratio. British Poultry Science 43:501-507. Baker, J. R. 1976. Biology of the trypanosomes of birds. In Biology of the kinetoplastida (ed.w. H. R. Lumsden), pp. 131-174. London: Academic Press. Booth, C. E., and P. F. Elliot. 2002. Hematological responses to hematozoa in North American and neotropical songbirds. Comparative Biochemistry and Physiology, Part A: Molecular and Integrative Physiology 133(3):451-467. Campbell, T.W. 1995. Avian Hematology and Cytology. Iowa State University Press, Ames, Iowa. Campbell, T.W. 1996. Clinical pathology. Reptile Medicine and Surgery (D.R. Mader, Ed.), pp. 248-257. Philadelphia: W.B. Saunders Company. Cardinet, G. H., J. F. Littrell, and O.W. Schalm. 1964. Effects of sustained muscular activity upon blood morphology of the horse. The California Veterinarian 18:31-35. COSEWIC. 2009. COSEWIC assessment and status report on the Bicknell's Thrush Catharus bicknelli in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa, Ontario. vii + 44 pp. Davis, A. K., D. L. Maney, and J. C. Maerz. 2008. The use of leukocyte profile to measure stress in vertebrates: a review for ecologists. Functional Ecology 22:760-772. Dhabhar, F. S., A. H. Miller, B. S. McEwen, and R. L. Spencer. 1995. Effects of stress on immune cell distribution dynamics and hormonal mechanisms. Journal of Immunology 154:5511-5527. Dhabhar, F. S. 2002. A hassle a day may keep the doctor away: stress and the augmentation of immune function. Integrative and Comparative Biology 42:556-564. Dufva, R. 1996. Blood parasites, health, reproductive success, and egg volume in female Great Tits Parus major. Journal of Avian Biology 27:83-87. Dyrcz, A., M. Wink, A. Kruszewicz, and B. Leisler. 2005. Reproductive success is correlated with blood parasite levels and body condition in the promiscuous aquatic warbler (Acrocephalus paludicola). Auk 122(2):558-565. Fallis, A. M., and S. S. Desser. 1977. On species of Leucocytozoon, Haemoproteus, and Hepatocystis. In Parasitic protozoa (J. P. Kreier, Ed.), pp. 239-266. NewYork: Academic Press.

24 Figuerola, J., R. G. Muñoz, and D. Ferrer. 1999. Blood parasites, leucocytes and plumage brightness in the Cirl Bunting, Emberiza cirlus. Ecology 13:594-601. Garvin, M. C., C. C. Szell, and F. R. Moore. 2006. Blood Parasites of nearcticneotropical migrant passerine birds during spring trans-gulf migration: Impact on host body condition. Journal of Parasitology 92(5):990-996. Gross, W. B., and H. S. Siegal. 1983. Evaluation of the heterophil/lymphocyte ratio as a measure of stress in chickens. Avian Diseases 27:972-979. Harmon, B. G. 1998. Avian heterophils in inflammation and disease resistance. Poultry Science 77:972 977. Ilmonen, P., D. Hasselquist, A. Langefors, and J. Wiehn. 2003. Stress, immunocompetence and leukocyte profiles of pied flycatchers in relation to brood size manipulation. Oecologia 136:148-154. International Bicknell s Thrush Conservation Group. 2010. A Conservation Action Plan for Bicknell s Thrush (Catharus bicknelli). J. A. Hart, C. C. Rimmer, R. Dettmers, R. M. Whittam, E. A. McKinnon, and K. P. McFarland, Eds. International Bicknell s Thrush Conservation Group. [Online] available at www.bicknellsthrush.org Merino, S., J. Moreno, J. J. Sanz, and E. Arriero. 2000. Are avian blood parasites pathogenic in the wild? A medication experiment in blue tits (Parus caeruleus). Proceedings of the Royal Society B 267:2507-2510. Min, B. 2008. Basophils: what they can do versus what they actually do. Nature Immunology 9:1333-1339. Min, B., M. A. Brown, and G. Legros. 2012. Understanding the roles of basophils: breaking dawn. 135(3):192-197. Nordling, D., M. Andersson, S. Zohari, and L. Gustafsson. 1998. Reproductive effort reduces specific immune response and parasite resistance. Proceedings of the Royal Society B 265:1291-1298. Ots, I., and P. Hõrak. 1996. Great Tits Parus major trade health for reproduction. Proceedings of the Royal Society of London B 263:1443-1447. Ots, I., and P. Hõrak. 1998. Health impact of blood parasites in breeding Great Tits. Oecologia 116:441-448.

25 Ots, I., A. Murumägi, and P. Hõrak. 1998. Hematological health state indices of reproducing Great Tits: a response to brood size manipulation. Functional Ecology 12:750-756. Owen, J. C., and F. R. Moore. 2006. Seasonal differences in immunological condition of three species of thrushes. Condor 108:389-398. Peig, J., and A. J. Green. 2009. New perspective for estimating body condition from mass/length data: the scaled mass index as an alternative method. Oikos 118:1883-1891. Remage-Healey, L., and M. Romero. 2001. Corticosterone and insulin interact to regulate glucose and triglyceride levels during stress in a bird. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology 281:994-1003. Rupley, A. E. 1997. Manual of Avian Practices. 556 pp. W.B. Saunders Company, Philadelphia PA. Schrader, M. S., E. L. Walters, F. C. James, and E. C. Greiner. 2003. Seasonal prevalence of a haematozoan parasite of Red-bellied Woodpeckers (Melanerpes carolinus) and its association with host condition and overwinter survival. Auk 120:130-137. Shen, T., S. Kim, J. Do, L. Wang, C. Lantz, J. F. Urban, G. LeGros, and B. Min. 2008. T cell-derived IL-3 plays key role in parasite infection-induced basophil production but is dispensable for in vivo basophil survival. Immunology 20(9):1201-1209. Simpson, V. R., G. Mackenzie, and E. A. Harris. 1996. Fatal microfilarial infection in red-billed blue magpies (Urocissa erythrorhynchus). Veterinary Record 138:522-523. Verhulst, S., B. Riedstra, and P. Wiersma. 2005. Brood size and immunity costs in zebra finches Taeniopygia guttata. Journal of Avian Biology 36:22-30. Weatherhead, P. J., and G. F. Bennett. 1991. Ecology of red-winged blackbird parasitism by haematozoa. Canadian Journal of Zoology 69:2352-2359.

26 Tables Table 2.1. The infection rate of hematozoa in adult Bicknell s and Swainson s Thrush, in the Christmas Mountains, NB, Canada. Birds were sampled between 26 May and 3 August, 2008 to 2010. Host Hematozoan 2008 2009 2010 Total Species Species N=22 N=14 N=23 N=59 Bicknell's Leucocytozoon debreuli 0.59 0.57 0.35 0.49 Thrush Haemaproteus neseri 0.23 0.36 0.26 0.27 Trypanosoma avium 0.05 0.07 0.04 0.05 Microfilaria (nematode) 0.05 0.00 0.00 0.02 No Parasites 0.32 0.29 0.52 0.39 N=41 N=41 N=29 N=111 Swainson's Leucocytozoon debreuli 0.85 0.78 0.72 0.79 Thrush Haemaproteus neseri 0.29 0.20 0.14 0.22 Trypanosoma avium 0.05 0.07 0.17 0.09 Microfilaria (nematode) 0.00 0.15 0.07 0.07 No Parasites 0.05 0.15 0.10 0.10

27 Table 2.2. Analysis of white blood cell profiles of Bicknell s and Swainson s Thrush in NB, Canada. We used Type III ANOVA with year, species and age as factors. Significant results are in bold. The columns are cell types. Eos = Eosinophil, Mon = Monocyte, Bas = Basophil, Het = Heterophil, Lym = Lymphocyte, H:L = Heterophil to Lymphocyte ratio. Year Species Age Year*Species Year*Age Species*Age Eos Mon Bas Het Lym H:L R² 0.124 0.315 0.076 0.091 0.085 0.111 F 1.538 5.011 0.889 1.093 1.008 1.362 P 0.145 < 0.0001 0.538 0.375 0.439 0.216 F 2.347 4.014 0.695 0.199 0.608 0.699 P 0.101 0.021 0.502 0.820 0.547 0.500 F 1.043 1.175 0.781 0.589 2.018 0.277 P 0.310 0.281 0.379 0.445 0.159 0.600 F 0.109 1.591 2.211 3.347 2.923 3.699 P 0.742 0.210 0.140 0.070 0.090 0.057 F 0.089 0.388 1.467 0.140 0.544 0.217 P 0.914 0.680 0.236 0.869 0.582 0.806 F 0.636 0.940 0.363 0.923 1.117 1.008 P 0.531 0.394 0.696 0.401 0.331 0.369 F 0.592 0.012 1.470 0.671 0.220 0.589 P 0.443 0.915 0.228 0.415 0.640 0.445

28 Table 2.3. Analysis of white blood cell profile of Bicknell s and Swainson s Thrush in NB, Canada, after pooling data by year, species, and age. We used type III ANOVA with week, sex, and parasite species as factors. Significant results are in bold. Due to the interaction between week and sex, for basophils, the significant effect from week should be ignored. The columns are cell types and ratios. Eos = Eosinophil, Mon = Monocyte, Bas = Basophil, Het = Heterophil, Lym = Lymphocyte, H:L = Heterophil to Lymphocyte ratio. Week Sex Parasite Species Week*Sex Week* Parasite Species Sex* Parasite Species Week*Sex* Parasite Species Eos Mon Bas Het Lym H:L R² 0.497 0.422 0.718 0.533 0.499 0.502 F 1.047 0.774 2.701 1.210 1.056 1.067 P 0.433 0.820 0.000 0.245 0.421 0.406 F 0.321 1.416 5.382 2.494 2.582 2.809 P 0.965 0.205 < 0.0001 0.018 0.015 0.009 F 1.176 0.074 0.481 0.299 0.049 0.322 P 0.283 0.787 0.491 0.587 0.825 0.573 F 0.813 0.698 1.001 1.249 0.985 0.770 P 0.523 0.597 0.415 0.301 0.423 0.549 F 1.523 0.061 2.355 0.477 0.329 0.283 P 0.179 1.000 0.036 0.847 0.938 0.958 F 1.302 0.625 0.791 1.089 0.700 0.762 P 0.213 0.887 0.722 0.386 0.820 0.754 F 0.275 0.143 0.575 0.590 0.220 0.524 P 0.893 0.965 0.682 0.671 0.926 0.718 F 0.882 0.147 0.763 0.206 0.076 0.068 P 0.456 0.931 0.520 0.892 0.973 0.976

29 Table 2.4. Analysis of the condition of male Bicknell s and Swainson s Thrush in NB, Canada. Condition was calculated from the scaled mass index. a) Type III ANOVA shows no effect of year, age, or species. b) After pooling data by year, age and species, Type III ANOVA showed no effect of week or parasite infection on condition. Of the 108 sampled birds, 56 were infected by Leucocytozoon, 7 by Trypanosoma, 13 by Leucocytozoon and Trypanosoma, and 11 by Leucocytozoon and Haemaproteus. 21 birds had no parasites. a) R²=0.128 F P b) R²=0.279 F P Age 1.527 0.219 Week 0.604 0.789 Species 0.03 0.862 Parasite Species 1.468 0.222 Year 0.582 0.560 Week* Parasite Species 0.610 0.879 Age*Species 0.477 0.491 Age*Year 2.018 0.137 Species*Year 0.398 0.673 Age*Species*Year 0.11 0.896

30 Figures week 1 week 2 week 3 week 4 week 5 week 6 week 7 week 8 week 9 week 10 week 1 week 2 week 3 week 4 week 5 week 6 week 7 week 8 week 9 week 10 week 1 week 2 week 3 week 4 week 5 week 6 week 7 week 8 week 9 week 10 week 1 week 2 week 3 week 4 week 5 week 6 week 7 week 8 week 9 week 10 Figure 2.1: Intra-seasonal changes to white blood cell profile and hematozoan infection of Bicknell s and Swainson s Thrush in NB, Canada, during the summers of 2008 to 2010. Week one began on 26 May. a) Heterophils were significantly higher during weeks one, three, and eight (p= 0.036, 0.045, 0.023, respectively). Lymphocytes were significantly lower during weeks one and three (p= 0.004, 0.004, respectively). There was a significant negative correlation between heterophils and lymphocytes (r = -0.775,

31 p < 0.0001) b) Heterophil to Lymphocyte ratio (H:L) was significantly higher during weeks one, three, and eight (p= 0.006, 0.016, 0.038, respectively) c) Types of hematozoa were not taken into account; only the mean number of species infecting individual birds was plotted. Week three was significantly high (p= 0.005). d) Condition, measured as standardized residuals, did not change significantly during the breeding season. There was a significant negative correlation between the mean number of hematozoa per individual and condition, during weeks two through nine (r=-0.745, p= 0.034). Week 1 was omitted from the analysis in order to minimize the interference from migration and lack of insect vectors, and week 10 was omitted due to low sample size.

32 Chapter 3: THE EFFECT OF TROMBICULID LARVAE INFESTATION ON CONDITION AND RELATIVE LEUKOCYTE PROFILES OF TWO THRUSHES Hubert Askanas 1, Antony Diamond 2, Lynda Leppert 3, Rebecca Holberton 4, Manigandan Lejeune-Virapin 5, 1,2 Department of Biology, University of New Brunswick, 10 Bailey Drive, Fredericton, New Brunswick, Canada, E3B 5A3. 3 Georgia Aquarium, Veterinary Services, 225 Baker St NW, Atlanta, GA 30313 4 School of Biology & Ecology, University of Maine, 221 Murray Hall, Orono, ME 04469 5 Department of Ecosystem and Public Health, Faculty of Veterinary Medicine, University of Calgary, 3280 Hospital Drive, Calgary, Alberta, T2N 4Z6 k439e@unb.ca 1, diamond@unb.ca 2, uuncia@gmail.com 3, rebecca.holberton@maine.edu 4, mlejeune@ucalgary.ca 5, Corresponding author: Hubert Askanas, 110 Taurus Drive, Hanwell, NB, Canada, E3C 1N1, k439e@unb.ca

33 Abstract We discovered trombiculid mite larvae infesting several species of passerines, in New Brunswick, Canada. They were found near the cloaca and in the auditory canal of Bicknell s Thrush (Catharus bicknelli), Swainson s Thrush (Catharus ustulatus), Whitethroated Sparrow (Zonotrichia albicollis), Fox Sparrow (Passerella iliaca) and Darkeyed Junco (Junco hyemalis). The mites were identified as belonging to the genus Neotrombicula, and most closely resembled the species microti or harperi. Most infestations (18/22, for Bicknell s and Swainson s Thrush combined) occurred at the end of the breeding season, after 23 July, indicating that these mites were being transmitted locally. There was no effect of mites on the condition of male Bicknell s and Swainson s Thrush. Parasitized birds did show a statistically significant increase in mean basophil counts (2.8 cells more). There were no effects of trombiculid infestation on lymphocytes, heterophils, monocytes, or eosinopils. The samples were obtained during the onset of infestation, so effects of chronic infestation are unknown.

34 Introduction In August 2009, during a study on Bicknell s and Swainson s Thrush, we noticed orange coloured polyps around the cloaca of 2 Swainson s Thrushes. During the subsequent field season (2010), we found the parasite on 15 Swainson s Thrushes and five Bicknell s Thrushes, as well as several White-throated Sparrows (Zonotrichia albicollis), Fox Sparrows (Passerella iliaca) and a Dark-eyed Junco (Junco hyemalis). These birds had been infested by larval trombiculid mites (Fig 3.1). Upon review of the specimens, the mites most closely resembled Neotrombicul microti (Fig 3.2- mite identification performed by Manigandan Lejeune-Virapin and James W. Mertins), but the identification is not yet conclusive. The 6-legged larval stage of trombiculids is parasitic, and feeds upon the flesh of vertebrates including birds but more commonly rodents (Wrenn 1974, Whitaker and French 1982). This is the first account of Bicknell s Thrush being infested by larval trombiculid mites. The infestation of Bicknell s Thrush was of particular concern as it is a threatened species (COSEWIC 2009) and is declining rapidly in New Brunswick (Bird Studies Canada unpubl. data, from International Bicknell s Thrush Conservation Group (IBTCG 2010)). We compared leukocyte profiles and condition of parasitized versus nonparasitized Bicknell s and Swainson s Thrush. Although leukocyte profile is not reflective of potential immune response, it does represent the mobilization of cells in and out of circulation. We hypothesized that trombiculid larvae would negatively affect condition and cause changes in infested birds relative leukocyte profiles. We predicted that infested birds would be in poorer condition and, since eosinophils are thought to be

35 the major anti-parasitic cell line (Rupley 1997), would have more circulating eosinophils than non-infested birds. Methods Our study area was in the Christmas Mountains, north-central New Brunswick, Canada. We captured birds five days a week, from 25 May to 5 Aug, during the 2009 and 2010 breeding seasons. We recorded mass, wing chord, tarsus, tail, and bill length/width/depth. Birds were searched for ectoparasites during processing. Mites were scraped or cut off the infested area, and preserved in 70% alcohol. Blood samples were obtained by puncturing the brachial vein with a sterile 30 gauge needle. Blood was collected into two heparinised micro-capillary tubes, and smears were made immediately. Blood smears were fixed in methanol, and later stained with Volu-Sol Wright-Giemsa (Utah) stain. 100 white blood cells were counted under a compound microscope, for relative leukocyte profiles. Scaled mass was used as a condition indicator, using the following equation, as in Peig and Green (2009):, where and are measurements (Mass and Length) from individual i. is the mean length from the sampled population. We calculated bsma for Swainson s Thrush by calculating the slope of an ordinary least squares regression of ln(mass) on ln(right wing) and dividing the slope by its Pearson s correlation coefficient r. Outliers were removed, and scaled mass was calculated for each species separately. Condition was measured as the standardized residual of scaled mass versus date. We limited the analysis of body

36 condition to males, as the mass of female thrush was too variable during the breeding season. We used Bicknell s and Swainson s Thrush samples, taken between 23 July and 2 August, to test for differences in condition and relative leukocyte profiles. For condition, we used seven infested males and six un-infested males. For leukocyte profiles, we used 10 parasitized and 13 non-parasitized birds. In 2009, between 23 July and 2 August, there was only 1 infested bird. In 2010, all Thrushes caught between 23 July and 2 August were parasitized. Since the analysis of parasitized versus non-parasitized birds could also be viewed as the effect of year (2010 versus 2009 respectively), our conclusions were limited. We could not make any conclusions on the effects of trombiculids on leukocytes or condition, if there was a year effect. Monocytes were affected by year (P=0.021, from Table 2.2 in this thesis chapter 2), so they were removed from the analysis. None of the other cell lines were affected by year, so data were pooled. Results Basophils were significantly higher (2.55 cells) in parasitized versus nonparasitized birds (Tukey s HSD = 0.017). There were no other effects of trombiculids on relative leukocyte profiles (Table 3.1). There was no effect of trombiculids on the condition of male Swainson s Thrush (Fig 3.2 ; P=0.247).

37 Discussion We found no evidence that trombiculid larvae were affecting the condition of Bicknell s and Swainson s Thrush. A small, yet significant, increase in basophil counts was observed. There has been some recent work, in mice, which suggests that basophils are antigen presenting cells, which promote TH-2 immunity (humoral immunity) and are connected with parasite defence (Min 2008, Shen et al. 2008, Min et al. 2012). Until avian models are used to determine the function of basophils in birds, we are hesitant to make any conclusions about their increase in infested birds. Unfortunately, we were ending our field season just as the mite infestations were beginning, and may have missed any effects of chronic infestation on these birds. Acknowledgments We thank the New Brunswick Wildlife Trust Fund and Environment Canada for providing major funding for this project, Vermont Institute of Natural Sciences, Canadian Wildlife Service, Dr. Jim Goltz (Department of Agriculture, Aquaculture and Fisheries; Veterinary Laboratory Services) for technical assistance, and James W. Mertins (United States Department of Agriculture; Animal and Plant Health Inspection Service) for assistance in identifying the mite samples. This project was also supported by the Atlantic Laboratory for Avian Research (formerly Atlantic Cooperative Wildlife Ecology Research Network), the Department of Natural Resources, Bird Studies Canada, and Science Horizons (Environment Canada). And a big thanks to all the field assistants, Julia Hughs, Katy Leger, Graham Dixon-MacCallum, Greg Jongsma, Patrick Blake, and Marie-Paule Godin.

38 Literature cited COSEWIC. 2009. COSEWIC assessment and status report on the Bicknell's Thrush Catharus bicknelli in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa, Ontario. vii + 44 pp. International Bicknell s Thrush Conservation Group 2010. A Conservation Action Plan for Bicknell s Thrush (Catharus bicknelli). J. A. Hart, C. C. Rimmer, R. Dettmers, R. M. Whittam, E. A. McKinnon, and K. P. McFarland, Eds. International Bicknell s Thrush Conservation Group. Available at www.bicknellsthrush.org Min, B. 2008. Basophils: what they can do versus what they actually do. Nature Immunology 9:1333-1339. Min, B., M. A. Brown, and G. Legros. 2012. Understanding the roles of basophils: breaking dawn. 135(3):192-197. Peig, J., and A. J. Green. 2009. New perspective for estimating body condition from mass/length data: the scaled mass index as an alternative method. Oikos 118:1883-1891. Rupley, A.E. 1997. Manual of Avian Practices. 556 pp. W.B. Saunders Company, Philadelphia PA. Shen, T., S. Kim, J. Do, L. Wang, C. Lantz, J. F. Urban, G. LeGros, and B. Min. 2008. T cell-derived IL-3 plays key role in parasite infestation-induced basophil production but is dispensable for in vivo basophil survival. Immunology 20(9):1201-1209. Whitaker, J. O., and T. W. French. 1982. Ectoparasites and other associates of some insectivores and rodents from New Brunswick. Canadian Journal of Zoology 60:2787-2797. Wrenn, W. J. 1974. Notes on the ecology of chiggers (Acarina:Trombiculidae) from northern Michigan and the description of a new species of Euschoengastia. Journal of the Kansas Entomological Society 47(2):227-238.

39 Tables Table 3.1: Analysis of leukocyte profiles in Bicknell s and Swainson s Thrush, sampled between 23 July and 2 August of 2009 and 2010, in NB, Canada. Type III ANOVA showed an effect of trombiculid infestation on relative basophil counts. No other cell line was affected by trombiculid larvae infestation. Het Lymph Eo Baso H:L R² 0.010 0.004 0.000 0.241 0.016 Trombiculid F 0.215 0.083 0.007 6.668 0.346 P 0.647 0.775 0.936 0.017 0.563

40 Figures Figure 3.1: This Swainson s Thrush, with trombiculid larvae in its auditory canal and a tick above the right eye, was caught on 27 July, 2010, in New Brunswick, Canada.

41 Figure 3.2: Trombiculid mite larva, still clinging to some flesh. This photo was taken through a compound microscope at approximately 300X magnification (10X objective, 10X ocular, with 3X zoom digital camera).

42 Figure 3.3: Body condition of Swainson Thrush males, either infested or not, caught in New Brunswick Canada, between 23 July and 2 August of 2009 and 2010. Means are indicated by +, median by a horizontal line, upper and lower quartiles by the top and bottom of the box, and maximum and minimum by the whiskers. Condition was not affected by larval mite infestation (P=0.247).

43 Chapter 4: SEASONAL EFFECTS ON RELATIVE LEUKOCYTE PROFILES AND PREVALENCE OF HEMATOZOA IN BICKNELL'S THRUSH, AND EVIDENCE OF LOCAL TRANSMISSION IN NEW BRUNSWICK, CANADA Hubert Askanas 1, Antony Diamond 2, Kevin C. Fraser 3, Emily A. Mckinnon, Lynda Leppert 5, Rebecca Holberton 6 1,2,3,4 Department of Biology, University of New Brunswick, 10 Bailey Drive, Fredericton, NB, Canada, E3B 5A3. 5 Georgia Aquarium, Veterinary Services, 225 Baker St NW, Atlanta, GA 30313 6 School of Biology & Ecology, University of Maine, 221 Murray Hall, Orono, ME 04469 k439e@unb.ca 1, diamond@unb.ca 2, fraserkev@gmail.com 3, emilymck@yorku.ca 4, uuncia@gmail.com 5, rebecca.holberton@maine.edu 6 Corresponding author: Hubert Askanas, 110 Taurus Drive, Hanwell, NB, Canada, E3C 1N1, k439e@unb.ca 3 Emily A. Mckinnon is now at the Department of Biology, York University, Toronto, ON, M3J 1P3

44 Abstract It is well documented that neotropical migrants have fewer circulating hematozoa while overwintering than while breeding. Since high physiological stress is known to cause the relapse of infections and increase susceptibility to new infections, we explored the hypothesis that low physiological stress while overwintering allows migrant birds to mount a more effective immune response. We used heterophil to lymphocyte ratio (H:L) as a surrogate for corticosterone in the analysis of physiological stress. As predicted, overwintering Bicknell s Thrush had significantly lower H:L than those on the breeding grounds, and infection rates were significantly higher in breeding birds. All 24 samples from the Dominican Republic were parasite-free, whereas 61% of 59 New Brunswick birds were infected by hematozoa. There was a major shift in the relative leukocyte profile between overwintering and breeding birds. The most notable, and statistically significant, changes were an increase in mean basophil count (18.3 cells more) and a decrease in mean heterophil count (8.2 cells less). Basophils have recently been recognized as having a role in parasitic defence. With no data on suitability and prevalence of vectors in the Dominican Republic, we were unable to conclude whether low H:L and high basophil counts were major factors in the absence of hematozoa in overwintering birds. The infection of chicks and recently fledged birds, by Leucocytozoon debrueili and Trypanosoma avium, revealed that these parasites were being transmitted in New Brunswick.

45 Introduction Inter-seasonal effects on the prevalence of hematozoa in neotropical migrants have been observed in several species (Bennett et al. 1980, Greiner 1975, Bennett and Cameron 1974, Garvin et al. 2004, Valkiuñas et al. 2004, Fallon et al. 2005). Birds have many hematozoa circulating in their blood during their breeding season in North America (Greiner et al. 1975), and hematozoa infections decrease significantly while overwintering in the neotropics (Bennett et al. 1980, Garvin et al. 2004). Blood parasites spend part of their life cycle inside insects, such as black flies (simulids), mosquitoes (culicids), horse flies (tabanids), louse flies (hippoboscids), mites (trombiculids), and midges (ceratopogonids) (Bennett and Cameron 1974, Bennett and Coombs 1975, Merino and Potti 1995, Garvin et al. 2004, Sehgal et al. 2005). Using Leucocytozoon as an example, birds become infected after being bitten by a black fly which carries Leucocytozoon sporozoites (Desser 1967). The sporozoites migrate to the birds liver, and mature into merozoites. The merozoites then infect erythrocytes and leukocytes to produce gametocytes which, when ingested by a black fly feeding on blood or flesh, mature in their gut. A macrogametocyte fuses with a microgametocyte to mature into an oocyst, which produces sporozoites. The sporozoites migrate to the salivary glands of the black fly, ready to infect another host. The high numbers of vectors in breeding grounds, and the low prevalence and lack of suitable vectors on the wintering grounds, are thought to be the major factors in the prevalence of hematozoa infection, though little is known about neotropical vectors (Garvin et al. 2004, Valkiuñas et al. 2004, Fallon et al. 2005).

46 Since physiological stress has also been shown to influence susceptibility to hematozoa (Nordling et al. 1998, this thesis Chapter 2), we were interested in comparing the physiological stress of breeding and overwintering birds. Breeding is an energetically expensive event (Ots et al. 1998, Askanas et al. this thesis Chapter 2), making birds more susceptible to the acquisition of new infections and the return of latent infections (Weatherhead and Bennett 1991). Presumably, overwintering birds are less physiologically stressed, as they are solely preoccupied with territorial defence and self maintenance, and can consequently afford to allocate more resources to immune function. We hypothesized that Bicknell s Thrush would have fewer circulating hematozoa while on the wintering grounds because they are less physiologically stressed, and predicted that overwintering birds would have lower H:L than those on the breeding grounds. We sampled nestlings and hatch-year (recently fledged) birds on the breeding grounds to determine if hematozoa were being transmitted locally. Hematozoa in adult blood smears could simply be the return of latent infections, or the acquisition of new ones during migration. Given that blood parasites are transmitted horizontally through a vector, newly hatched chicks are parasite free. The presence of hematozoa in nestling and hatch-year birds therefore confirms local transmission. Methods Our breeding-ground study sites were located in the Christmas Mountains, northcentral New Brunswick, Canada. We captured Bicknell s Thrush five days a week at dusk and dawn. The 2008 to 2010 field seasons were two and a half months each (late

47 May to early August). Bicknell s Thrushes from the Dominican Republic were sampled in Sierra de Bahoruco National Park, during 9-12 and in Guaconeho 24-28 November 2010. All statistical tests were performed using XLSTAT 2011. Tukey s HSD was used to determine differences between blood cell counts and H:L of breeding versus overwintering Bicknell s Thrush. Blood samples were collected into heparinised micro-capillary tubes, by puncturing the brachial vein of captured Bicknell s Thrushes with a sterile 30 gauge needle. Blood smears were made immediately after sampling, fixed with methanol, and later stained with with Volu-Sol Wright-Giemsa (Utah) stain. 100 blood cells were counted, under a compound microscope. The smears were analyzed for presence/absence of hematozoa. H:L was used as a surrogate for corticosterone in the analysis of physiological stress. It is considered as good as, if not better than, corticosterone for the analysis of physiological stress (Davis et al. 2008). Corticosterone levels can increase within minutes of handling a bird, while H:L remains stable during processing (Davis 2005). During high physiological stress, corticosterone is released into circulation (Remage- Healey and Romero 2001). This releases stored glucose and lipids to boost energy (Dhabhar et al. 1995), but it also has effects on relative leukocyte profiles, causing an increase of circulating heterophils and the sequestering of lymphocytes (Davis et al 2008). Lymphocytes are more costly to maintain and deploy, so during high stress they are stored in other tissues (Apanius 1998).

48 Results Samples.- Of 25 useable blood smears sampled from Bicknell s Thrush in New Brunswick, and after removing outliers, 23 were used for relative leukocyte profile analysis. Hematozoa included Leucocytozoon debreuli, Haemaproteus neseri, Trypanosoma avium, and Microfilaria. Possible vectors included simulids (Leucocytozoon, Trypanosoma, Microfilaria), ceratopogonids (Haemaproteus, Microfilaria), hippoboscids (Haemaproteus, Trypanosoma), culicids (Haemaproteus, Trypanosoma, Microfilaria), trombiculids (Trypanosoma), and tabanids (Microfilaria). We found no parasites in 23 of 59 birds (39%) sampled in New Brunswick (this thesis chapter 2). We also sampled 24 Bicknell s Thrush in the Dominican Republic. 22 samples were useable for leukocyte profile analysis, and no hematozoa were found in their blood smears. Hatch-year birds with hematozoan infections.- Out of 15 hatch-year Bicknell s Thrush (chicks and fledglings), six were found to be infected by Leucocytozoon, and one by Trypanosoma. Relative leukocyte profile.- After outliers for H:L analysis, the leukocyte counts of Bicknell s Thrush from the Dominican Republic were compared to those of New Brunswick (Fig 4.1). Relative leukocyte profiles showed that Bicknell s Thrush in the Dominican Republic had significantly more circulating basophils (18.3 more, Tukey s HSD < 0.0001), fewer monocytes (0.9 cell less Tukey s HSD =0.045), and fewer heterophils than the New Brunswick birds (8.2 cells less, Tukey s HSD = 0.003). Mean

49 lymphocyte counts were not significantly different (mean counts from Dominican Republic were 4.0 cells less, Tukey s HSD 0.407), much like eosinophils (4.5 cells less, Tukey s HSD = 0.077). H:L was significantly higher in New Brunswick birds (Fig 4.2), compared to those in the Dominican Republic (0.156 higher, Tukey s HSD = 0.014). Mean eosinophil counts were not significantly different in parasitized versus nonparasitized birds (5.1 cells higher in parasitized birds, Tukey s HSD = 0.075). Discussion Hatch-year birds with hematozoan infections.- The presence of hematozoa in hatchyear birds indicates that Leucocytozoon and Trypanosoma are completing their life-cycle in New Brunswick. Sometime during the breeding season, an insect vector must have become infected by feeding on a parasitized bird, and subsequently infected a hatch-year bird. Seasonal effects on relative leukocyte profiles.- The blood smears of Bicknell s Thrush in the Dominican Republic were unique in their lack of hematozoa, as well as their monocyte, basophil, and heterophil counts (Fig 4.1). Mean monocyte counts in Dominican Republic birds (0.0) were only one cell less than New Brunswick birds (0.9), so monocytes were discarded as a source of variation in the leukocyte profile. Basophil counts were significantly higher in overwintering birds, with a large difference in the counts (18.3 cells). Besides heterophils, which are negatively correlated with physiological stress, we were unable to explain why these cell lines differ.

50 Basophils are thought to be inflammatory in nature (Campbell 1995), though recent studies on mice have reported that basophils are antigen-presenting cells with a role in parasitic defence through TH-2 immunity (antigen-specific humoral immunity; Min 2008, Shen et al. 2008, Min et al. 2012). This could explain the absence of hematozoa in the birds from the Dominican Republic, but until avian models are used to test the functions of basophils, this is speculative. Similarly with monocytes, whose role is to fight infection (Rupley 1997), we do not know why they were absent from the Dominican Republic samples. We expected eosinophils to be most affected by hematozoa, as they are thought to be anti-parasitic (Rupley 1997), but the difference of 5.1 cells in parasitized versus non-parasitized birds was not statistically significant. Seasonal effects on physiological stress and circulating hematozoa.- Several studies have shown the relationship between physiological stress and susceptibility to infection (Weatherhead and Bennett 1991, Nordling et al. 1998). In Bicknell s and Swainson s Thrush, we found that susceptibility to hematozoa infection during the breeding season increased with high physiological stress, as increases in the number of circulating hematozoa per individual occurred concurrently with peaks in H:L (this thesis chapter 2). Similarly, we found some evidence to support the theory that low physiological stress in overwintering birds reduces circulating hematozoa. Bicknell s Thrush sampled in the Dominican Republic had low mean H:L (mean 0.134; Fig 4.2) and no parasites, while birds from New Brunswick had significantly higher mean H:L (mean over the breeding season 0.290; Fig 4.2) and 36 of 59 birds (61%) were parasitized (this thesis chapter 2). For comparison, Askanas et al. (this thesis Chapter 2) reported that mean

51 H:L in breeding Bicknell s and Swainson s Thrush during peak provisioning was 0.599, and Owen and Moore (2006) reported mean H:L in migrating Swainson s Thrush to be 1.04, Veery (Catharus fuscescens) 1.37, and Wood Thrush (Hylocichla mustelina) 1.19. Experiencing less physiological stress on the wintering grounds may have allowed the 24 overwintering birds to mount a more effective immune response. As it may also be possible that hematozoa have no suitable neotropical vectors (Garvin et al 2004), and that prevalence of vectors in the neotropics is low during the winter, a study on the presence/absence and prevalence of hematozoa in vectors of the Dominican Republic is needed. In Costa Rica, Valkiuñas et al. (2004) reported four species of Neotropical migrants being infected by local species of hematozoa, so it seems probable that suitable vectors are present in the Dominican Republic as well (such as mosquitoes, Askanas per obs). If hematozoa are found in suitable vectors of the Dominican Republic, during the time Bicknell s Thrush are overwintering, it would support the theory that high physiological stress compromises the immune system during the breeding season and that low physiological stress will return immune function while overwintering. Acknowledgments We thank the New Brunswick Wildlife Trust Fund and Environment Canada for providing major funding for this project, Chris Rimmer, the Vermont Institute of Natural Sciences, Canadian Wildlife Service, and Museo Nacional de Historia Natural (Santo Domingo) for their help in the Dominican Republic, and Dr. Jim Goltz for technical assistance (Department of Agriculture, Aquaculture and Fisheries; Veterinary Laboratory Services). This project was also supported by the Atlantic Laboratory for

52 Avian Research (formerly Atlantic Cooperative Wildlife Ecology Research Network), the Department of Natural Resources, Bird Studies Canada, and Science Horizons (Environment Canada). And a big thanks to all the field assistants, Julia Hughs, Katy Leger, Graham Dixon-MacCallum, Greg Jongsma, Patrick Blake, Marie-Paule Godin, Richard Aracil, Carl-Adam Wegenschimmel, as well as Juan Klavins and the rest of the Dominican Republic crew. Literature cited Apanius, V. 1998. Stress and immune defence. Advances in the Study of Behavior 27:133-153. Askanas, H., A. Diamond, K. Fraser, E. Fraser, L. Leppert, and R. Holberton. In progress. Susceptibility to hematozoon infection is influenced by high physiological stress during the breeding season, and is correlated with negative effects on condition. Master s thesis chapter 2. Bennett, G. F., and M. Cameron. 1974. Seasonal prevalence of avian hematozoa in passeriform birds of Atlantic Canada. Canadian Journal of Zoology 52(10):1259-1264. Bennett, G. F., and R. F. Coombs. 1975. Ornithphilic vectors of avian hematozoa in insular Newfoundland. Canadian Journal of Zoology 53(9):1241-1246. Bennett, G. F., H. Witt, and E. M. White.1980. Blood parasites of some Jamaican birds. Journal of Wildlife Diseases 16(1):29-38. Campbell, T.W. 1995. Avian Hematology and Cytology. 108 pp. Iowa State University Press, Ames, Iowa. Davis, A. K. 2005. Effect of handling time and repeated sampling on white blood cell counts. Journal of Field Ornithology 76(4):334-338. Davis, A. K., D. L. Maney, and J. C. Maerz. 2008. The use of leukocyte profile to measure stress in vertebrates: a review for ecologists. Functional Ecology 22:760-772. Desser, S. S. 1967. Schizogony and gametogony of Leucocytozoon somondi and associated reactions in the avian host. Journal of Protozoology 14:244-254.

53 Dhabhar, F. S., A. H. Miller, B. S. McEwen, and R. L. Spencer. 1995. Effects of stress on immune cell distribution dynamics and hormonal mechanisms. Journal of Immunology 154:5511-5527. Fallon, S. M., E. Bermingham, and R. E. Ricklefs. 2005. Host specialization and geographic localization of avian malaria parasites: a regional analysis in the Lesser Antilles. The American Naturalist 165:466-480. Garvin, M. C., P. P. Marra, and S. K. Crain. 2004. Prevalence of hematozoa in overwintering American Redstarts (Setophaga ruticilla): No evidence for local transmission. Journal of Wildlife Diseases 40(1):115-118. Greiner, E. C., F. Gordon, E. M. Bennett, and R. F. Coombs. 1975. Distribution of the avian hematozoa of North America. Canadian Journal of Zoology 53(12):1762-1787. Merino, S., and Potti, J. 1995. High prevalence of hematozoa in nestlings of a passerine species, the Pied Flycatrcher (Ficedula hypoleuca). Auk 112(4):1041-1043. Min, B. 2008. Basophils: what they can do versus what they actually do. Nature Immunology 9:1333-1339. Min, B., M. A. Brown, and G. Legros. 2012. Understanding the roles of basophils: breaking dawn. 135(3):192-197. Nordling, D., M. Andersson, S. Zohari, and L. Gustafsson. 1998. Reproductive effort reduces specific immune response and parasite resistance. Proceedings of the Royal Society B 265:1291-1298. Ots, I., A. Murumägi, and P. Hõrak. 1998. Hematological health state indices of reproducing Great Tits: a response to brood size manipulation. Functional Ecology 12:750-756. Owen, J. C., and F. R. Moore. 2006. Seasonal differences in immunological condition of three species of thrushes. Condor 108: 389-398. Sehgal, R., H. I. Jones, and T. B. Smith. 2005. Molecular evidence of host specificity of parasitic nematode microfilariae in some African rainforest birds. Molecular Ecology 14:3977-3988. Shen, T., S. Kim, J. Do, L. Wang, C. Lantz, J. F. Urban, G. LeGros, and B. Min. 2008. T cell-derived IL-3 plays key role in parasite infection-induced basophil production but is dispensable for in vivo basophil survival. Immunology 20(9):1201-1209.

54 Remage-Healey, L., and M. Romero. 2001. Corticosterone and insulin interact to regulate glucose and triglyceride levels during stress in a bird. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology 281:994-1003. Rupley, A. E. 1997. Manual of Avian Practices. W.B. Saunders Company, Philadelphia PA. Valkiuñas, G., T. A. Iezhova, D. R. Brooks, S. V. Brant, M. E. Sutherlin, and D. Causey. 2004. Additional observations on blood parasites of birds in Costa Rica. Journal of Wildlife Disease 40(3):555-561. Weatherhead, P. J., and G. F. Bennett. 1991. Ecology of red-winged blackbird parasitism by haematozoa. Canadian Journal of Zoology 69:2352-2359.

55 Figures Figure 4.1: Relative leukocyte profile of Bicknell s Thrush sampled during the breeding season in New Brunswick (NB; n=23), Canada, during 2008 to 2010, and while overwintering in the Dominican Republic (DR; n=22) in 2010. Means are indicated by +, median by a horizontal line, upper and lower quartiles by the top and bottom of the box, and maximum and minimum by the whiskers. Monocytes were not included in the figure as they were not suitable for the scale, mean counts for New Brunswick birds were 0.9 and in Dominican Republic 0.0. Birds from the Dominican Republic had significantly fewer monocytes (0.9 cell less, Tukey s HSD= 0.045), fewer heterophils (8.2 cells less, Tukey s HSD = 0.003), but more basophils (18.3 cells more, Tukey s HSD < 0.0001) than New Brunswick birds. Lymphocyte counts (Tukey s HSD = 0.407), and eosinophil counts (Tukey s HSD = 0.077) were not significantly different.

56 Figure 4.2: Heterophil to lymphocyte ratio (H:L) of Bicknell s Thrush sampled during the breeding season in New Brunswick (NB; n=23), Canada, during 2008 to 2010, and while overwintering in the Dominican Republic (DR; n=24) in 2010. Means are indicated by +, median by a horizontal line, upper and lower quartiles by the top and bottom of the box, and maximum and minimum by the whiskers. H:L was significantly higher (Tukey s HSD = 0.014) for New Brunswick birds (mean 0.290), compared to those in the Dominican Republic (0.134).

57 Chapter 5: TOTAL FEEDING RATE OF A MALE BICKNELL S THRUSH CONCURRENTLY FEEDING TWO BROODS Hubert Askanas 1, Antony Diamond 2 1,2 Department of Biology, University of New Brunswick, 10 Bailey Drive, Fredericton, NB, k439e@unb.ca 1, diamond@unb.ca 2, Corresponding author: Hubert Askanas, 110 Taurus Drive, Hanwell, NB, Canada, E3C 1N1, k439e@unb.ca

58 Abstract Catharus bicknelli (Bicknell s Thrush) is polygynandrous, with both sexes attempting to pair-bond with more than one partner during the breeding season, resulting in multiple male feeders at the nest. The total feeding rate in females is easily determined from nest video, but total feeding rate in males is complicated by those feeding two broods. While studying Bicknell s Thrush in the province of New Brunswick, we observed two broods that shared a male feeder. Due to the inherently challenging nature of Bicknell s Thrush research, accounts of males concurrently feeding two broods are very scarce; to our knowledge, this is the first study to use concurrent video in calculating the total feeding rate of a male Bicknell s Thrush feeding two broods. We recorded video of both nests concurrently for two days before one of the nests was depredated. The male provided at a rate of 1.2 feeds per hour at one nest, and 2.0 feeds per hour at the other. The total feeding rate for this male (3.2 feeds per hour) was similar to the females (2.9 and 3.3 feeds per hour).

59 Introduction Since 2007, we have been studying Bicknell s Thrush in the Christmas Mountains of New Brunswick, Canada. This area is considered low elevation habitat (600 metres) for a species normally nesting at higher elevations (>900 metres, Able and Noon 1976). It prefers to breed in thick, stunted (or early regeneration) stands of Balsam Fir (Abies balsamea) (Nixon et al. 2001, Aubry et al. 2011), which makes finding nests particularly difficult. Bicknell s Thrush is one of two species in Canada known to show polygynandry (Strong et al. 2004), the other being Smith s Longspur (Briskie 1998). Bicknell s Thrush males will attempt to mate with several females and females with several males. Males will provide food for nestlings in nests built by females with which they have mated (Goetz et al. 2003). This typically results in two males (occasionally up to four) providing at a nest. Although it is well known that Bicknell s Thrush is polygynandrous, we still know very little about the consequences of this mating system for their breeding biology. During summer 2010, we found two Bicknell s Thrush nests, approximately 84 metres apart, which shared a male provider. The first nest (hereafter BN1) had one chick with only one male feeder. The chick was 1-2 days old on 29 June. The second nest (BN2), found on 3 July, had four chicks and at least one other unbanded male feeder. The four chicks in BN2 were 1-2 days older than the chick in BN1. We videorecorded both nests for two consecutive days before the chick in BN1 was depredated. This is the first account, for New Brunswick, of a male providing concurrently at two nests. There are three such accounts from Vermont, and three from Quebec (Rimmer et al. 2001).

60 Field site description Our field site was located on Mount Mitchell, North-central New Brunswick, Canada. The study area was 635 metres in elevation, and the forest stand was composed of thick early-regeneration Balsam Fir (Abies balsamea). Methods Thrushes were caught in mist nets, occasionally luring them into the nets by playing their song (MP3 player and speakers), and colour banded for identification on video. Females were equipped with a radio-transmitter in order to find their nests. We installed semi-permanent video camera mounts as far from the nest as possible (3-5 metres), while still being able to see the rim of the nest. This allowed us to set up and remove the video cameras quickly, limiting disturbance of the nest. The camera mount was waterproof and camouflaged. We used Sony DCR-SR45 hard-drive-disk camcorders, with extended batteries, to record the nests continuously for six to seven hours. Results Only one male, banded medium-green over yellow/light-green (hereafter MG/YLG), fed chicks at both nests (Figure 5.1). The initial feeding rates of this male, from 29 June to 1 July at BN1, were relatively low (Table 5.1). Even though he was the sole male feeder, he did not feed the chick in this nest after 4:15 pm, between 29 June and 1 July. In addition to MG/YLG not providing at BN1 in the evening, his feeding rate

61 was also lower at BN1 during the day (1.2 feeds per hour, compared to 2.0 feeds per hour; Fig 5.2). During the two days of concurrent video, the total feeding rate of MG/YLG averaged 3.2 feeds per hour, which is comparable to the females; 2.9 at BN2, and 3.3 at BN1 (derived from Table 5.1). The male provided 27% of all feeds at BN1, and 20% at BN2. After the depredation event at BN1, the feeding rate of MG/YLG at BN2 remained consistent, and the feeds by the unbanded male(s) decreased from 4.6 feeds per hour to 1.3 (Table 5.1). Discussion The total feeding rate for the male (MG/YLG) providing at two nests (3.2 feeds per hour) was similar to the rates recorded for his female partners (2.9 and 3.3 feeds per hour). The division of effort between nests for the New Brunswick Bicknell s Thrush male (1.2 feeds per hour for one chick, and 2.0 for four chicks) was similar to a male in Vermont, with sequential broods, with 1.0 visit per hour at a nest with two male feeders and 1.9 at a nest where he was the sole male feeder (Goetz et al. 2003). The male from Vermont stopped feeding the chicks at the nest with a second male feeder, soon after they fledged, and doubled his feeding effort for his newly-hatched brood in which he was the sole male feeder (Goetz et al. 2003). Males apparently prioritize feeding efforts depending on the number and stage of nestlings, as well as the presence of other male feeders.

62 Acknowledgments We would like to thank the New Brunswick Wildlife Trust Fund and Environment Canada for providing major funding for this project, Vermont Institute of Natural Sciences, Canadian Wildlife Service, and Dr. Jim Goltz for technical assistance (Department of Agriculture, Aquaculture and Fisheries; Veterinary Laboratory Services). This project was also supported by the Atlantic Laboratory for Avian Research (formerly Atlantic Cooperative Wildlife Ecology Research Network), the Department of Natural Resources, Bird Studies Canada, and Science Horizons (Environment Canada). Thanks to all the field assistants, Julia Hughs, Patrick Blake, and Marie-Paule Godin. And thanks to Kevin Fraser for providing comments. Literature cited Able, K. P., and B. R. Noon. 1976. Avian community structure along elevational gradients in the Northern United States. Oecologia 26:275-294. Aubry, Y., A. Desroches, and G. Seutin. 2011. Response of Bicknell s Thrush (Catharus bicknelli) to boreal siviculture and forest stand edges: a radio-tracking study. Canadian Journal of Zoology 89:474-482. Briskie, J. V., R. Montgomerie, T. Poldmaa, and P. T. Boag. 1998. Paternity and paternal care in the polygynandrous Smith's longspur. Behavioral Ecology and Sociobiology 43(3):181-190. Goetz, J. E., K. P. McFarland, and C. C. Rimmer. 2003. Multiple paternity and multiple male feeders in Bicknell's thrush (Catharus bicknelli). Auk 120(4):1044-1053. Nixon, E.A., S. B. Holmes, and A. W. Diamond. 2001. Bicknell s Thrushes (Catharus bicknelli) in New Brunswick clear cuts: their habitat associations and co-occurrence with Swainson s Thrushes (Catharus ustulatus). Wilson Bulletin 113(1):33-40.

63 Rimmer, C. C., K. P. McFarland, W. G. Ellison, and J. E. Goetz. 2001. Bicknell's Thrush (Catharus bicknelli), In The Birds of North America, No. 592 (A. Poole and F. Gill, Eds.). The Birds of North America, Inc., Philadelphia, PA. Strong, A. M., C. C. Rimmer, and K. P. McFarland. 2004. Effects of prey biomass on reproductive success and mating strategy of Bicknell s Thrush (Catharus bicknelli), a polygynandrous songbird. Auk 121(2):446-451.

64 Tables Table 5.1: Feeding rates of adult Bicknell s Thrush, at two broods which shared a male provider, during summer 2010, in the Christmas Mountains, New Brunswick, Canada. Feeding rate is given in feeds per hour. Individuals are labelled with their leg-band colours followed by their sex (Ex: female banded Light Green/Mauve over Medium Blue = LGM/MB F). The unbanded male may be more than one bird. The sole chick from BN1 was depredated during the evening of 4 July. Nest BN1 29-Jun 30-Jun 01-Jul 02-Jul 03-Jul 04-Jul RW/DB F 1.29 2.02 1.32 no video 3.48 3.15 MG/YLG M 0.71 0.29 0.50 no video 1.45 1.00 # of chicks 1 1 1 1 1 1 Age of chick 1 2 3 4 5 6 Nest BN2 03-Jul 04-Jul 05-Jul 06-Jul 07-Jul 08-Jul LGM/MB F 2.00 3.74 2.01 3.12 2.01 0.00 MG/YLG M 1.80 2.16 2.20 2.14 1.50 1.16 Unbanded M 4.61 4.60 1.28 2.53 1.79 0.87 # of chicks 4 4 4 4 3 2 Age of chicks 6 7 8 9 10 11

65 Figures Figure 5.1: This male Bicknell s Thrush, banded medium-green over yellow/light-green (MG/YLG) was first captured at Mount Mitchell, NB, in 2008. He is at least 4 years old. The nest on the left is BN1, and contains one chick. The nest on the right is BN2, and contains four chicks.