Host-Odor Attractants of Philornis Downsi [Diptera: Muscidae], an Invasive Bird Parasite in the Galapagos Islands

SUNY College of Environmental Science and Forestry Digital Commons @ ESF

Dissertations and Theses

Spring 5-14-2019

HOST-ODOR ATTRACTANTS OF PHILORNIS DOWNSI [DIPTERA: MUSCIDAE], AN INVASIVE BIRD PARASITE IN THE GALAPAGOS ISLANDS

Follow this and additional works at: https://digitalcommons.esf.edu/etds

Part of the Environmental Health and Protection Commons, and the Environmental Monitoring Commons

Recommended Citation Doherty, Kristin, "HOST-ODOR ATTRACTANTS OF PHILORNIS DOWNSI [DIPTERA: MUSCIDAE], AN INVASIVE BIRD PARASITE IN THE GALAPAGOS ISLANDS" (2019). Dissertations and Theses. 71. https://digitalcommons.esf.edu/etds/71

This Open Access Thesis is brought to you for free and open access by Digital Commons @ ESF. It has been accepted for inclusion in Dissertations and Theses by an authorized administrator of Digital Commons @ ESF. For more information, please contact [email protected], [email protected]. HOST-ODOR ATTRACTANTS OF PHILORNIS DOWNSI [DIPTERA: MUSCIDAE],

AN INVASIVE BIRD PARASITE IN THE GALAPAGOS ISLANDS

Kristin M. Doherty

A thesis submitted in partial fulfillment of the requirements for the Master of Science Degree State University of New York College of Environmental Science and Forestry Syracuse, New York April 2019

Department of Environmental and Forest Biology

Approved by: Stephen A. Teale, Major Professor John Wagner, Chair, Examining Committee Melissa K. Fierke, Department Chair S. Scott Shannon, Dean, The Graduate School

Acknowledgements

I would like to thank Stephen A. Teale for cultivating my interest in entomology as an undergraduate and especially for the years of guidance that followed. I could not have asked for a more creative leader to help and inspire. Thank you to my committee members, Margaret Voss and

Katalin Boroczky, both of whom took my ideas and helped to turn them into answerable research questions. Their knowledge in ornithology, metabolism and analytical chemistry enabled me to recognize the most important findings of this project. Additionally, thank you to Dong Cha who offered guidance both in the area of electrophysiological testing and as a reader.

I also wish to express my gratitude to all those associated with the Charles Darwin Research

Station in the Galapagos Islands. Charlotte Causton introduced our lab to the Philornis downsi team with a workshop in 2012 and has been absolutely indispensable since. Thank you to Alejo Mieles,

Arno Cimadom, Sarah Knutie, Piedad Lincango, and all those with the Charles Darwin Foundation for assisting me with trapping experiments and sample acquisition when I was unable to personally travel to Santa Cruz. Thank you to James Gibbs and George Heimpel for transporting P. downsi specimens to the United States, with permits to import the specimens granted by USDA-APHIS. This study was funded by the Galapagos Conservancy and the International Community Foundation with a grant awarded by The Leona M. and Harry B. Helmsley Charitable Trust to Stephen Teale, Birgit

Fessl and Charlotte Causton.

Thank you to my lab-mates Alejo Mieles, Hajar Faal, Laura Hansen, and Tian Xu for all of your assistance and thank you to Max Collignon for recognizing my potential role in the P. downsi project. Finally, thank you to all of my friends and family who generously supported me the last several years.

iii

Table of Contents

List of Tables……………………………………………………………………………………………………………………………...vi

List of Figures..……………………………………………………………………………………………………………..…………...vii

List of Appendices…………………………………………………………………………………………..……………………..…viii

Abstract…………………………………………………………………………………………………………...……………………...…ix

Introduction…………………………………………………………………………………………………………..……………………1

PHILORNIS DOWNSI, AN AVIAN ECTOPARASITE……………………………….………………………………1

IMPACTS ON HOSTS IN THE GALAPAGOS………………………………………………...……………………….2

PHILORNIS DOWNSI MANAGEMENT...…………………………………………………....…………………………4

HOST-ODOR BASED ATTRACTION IN DIPTERA………………………………...………………………………7

AVIAN SOURCES OF OLFACTORY ATTRACTION OF PHILORNIS DOWNSI……………………………9

Methods……………………………………………………………………………………………………………………………………11

LIVE ANIMALS………………………………………………………………………………….……………………………11

Birds………………………………………………………………………………….………………………………11

Insects………………………………………………………………………..………………………………………12

VOLATILE COLLECTION AND ANALYSIS…………………………………………………………………………12

Zebra finch nest headspace………………………………………….………………………………………12

Wild bird nest headspace……………………………………………………….……………………………14

Fecal samples…………………………………………………………………………………….……….………15

Uropygial gland secretions…………………………………………………………………………..………16

Egg headspace……………………………………………………………………………………………………17

ELECTROPHYSIOLOGICAL AND BEHAVIORAL ASSESSMENT……..……………………………………17

GC-EAD/EAG………………………………………………………………………………………………………17

Laboratory bioassays…………………………………………………………………………………….……19

Field bioassays……………………………………………………………………………………………………20

Results…………………………………………………………………………………………………………………………………...…23

VOLATILE COLLECTION AND ANALYSIS…………………………………………………………………………23

Laboratory bird nest headspace…………………………………………………………………………...23

Wild bird nest headspace………………………………………………………………………………….…23

Fecal samples……………………………………………………………………………………………………..23

Uropygial gland secretions…………………………………………………………..………………………24

Egg headspace……………………………………………………………………………………………………24

ELECTROPHYSIOLOGICAL AND BEHAVIORAL ASSESSMENT…………………..………………………24

GC-EAD/EAG………………………………………………………………………………………………………24

Laboratory bioassays……………………………………………………………………………………….…26

Field bioassays……………………………………………………………………………………………………26

Discussion…………………………………………………………………………………………………………………………………31

WHOLE-NEST VOLATILES…………………………………………………………………………………………...…31

INDIVIDUAL SOURCES OF HOST VOLATILES……………………………………………………………..……35

Feces………………………………………………………………………………………………………….………35

Uropygial glands……………………………………………………………………………………………...…36

Eggs………………………………………………………………………………………………………………..…37

LABORATORY BIOASSAYS………………………………………………………………………………………...……39

Conclusion…………………………………………………………………………………………………………………………...……41

References…………………………………………………………………………………………………………………………...……42

Appendices……………………………………………………………………………………………………………………………….51

Resume……………………………………………………………………………………………………………………………………..61

List of Tables

Table 1. Collection and analysis methods for zebra finch nest dynamic headspace collection………..13

Table 2. Collection and analysis methods for central New York wild bird nest headspace collection………………………………………………………………………………………………………………………….……….15

Table 3. Collection and analysis methods for Galapagos wild bird nest headspace collection…..…….15

Table 4. Occurrence of bird-related volatiles in the literature………………………………………………………33

List of Figures

Figure 1. GC-EAD response to acetone………………………………………………………………………………………..25

Figure 2. GC-EAD response to uropygial gland secretion……………………………………………………..………25

Figure 3. Mean (±SE) number of flies per trap captured for moderate release-rate acetone experiment……………………………………………………………………………………………………………………………..…27

Figure 4. Mean (±SE) number of flies per trap captured for high/low release-rate acetone experiment……………………………………………………………………………………………………………..…………………28

Figure 5. Mean (±SE) number of flies per trap captured for hexanal and food-source experiment………………………………………………………………………………………………………………………………..29

Figure 6. Mean (±SE) number of flies per trap captured for hexanal and food-compound experiment at El Barranco (A) and Los Gemelos (B) ……………………………..……………………………………30

vii

List of Appendices

Appendix I. Schematics of headspace collection methods……………………………………………………………..51

Appendix II. Antennal preparation for Philornis downsi GC-EAD/EAG recording……………..…………….52

Appendix III. Methodology and results of GC-EAD/EAG assays………………………………….….………………53

Appendix IV Schematics of olfactometers used for behavioral response assays…………….………………54

Appendix V. Chemical compounds identified from GC-MS analysis of zebra finch dynamic headspace collection………….……………..…………………..…………………………………………………..……………….55

Appendix VI. Chemical compounds identified from GC-MS analysis of zebra finch static headspace collection……………………………………………………………..……………………………………………………………………56

Appendix VII. Chemical compounds identified from GC-MS analysis of New York wild bird dynamic headspace collection……………………………………………………………..…………………………………...……...………57

Appendix VIII. Chemical compounds identified from GC-MS analysis of zebra finch uropygial gland secretions……………………………………………………………..……………..……………………………………………………58

viii

Abstract

K.M. Doherty. Host-odor Attractants of Philornis downsi [Diptera: Muscidae], an Invasive Bird Parasite in the Galapagos Islands, 71 pages, 4 tables, 6 figures, 2019.

Since its discovery in 1997, Philornis downsi has been shown to reduce fledgling success and fitness in many bird species within the Galapagos, including the endemic and critically endangered Darwin’s finches. Despite its impact, there are currently no practical methods to control the nest parasite. This study investigated potential kairomones that adult P. downsi use to locate bird nests, with the goal of developing an efficient olfaction-based method to monitor and control the insect. Volatile organic compounds were collected from the feces, uropygial glands, and eggs of laboratory- reared zebra finches, as well as whole-nest headspace from zebra finches, central New York birds, and finches in the Galapagos. Selected compounds were then tested for electrophysiological and behavioral response using GC-EAD/EAG and laboratory/field bioassays with adult P. downsi. Acetone, collected from eggs, and uropygial gland extracts was shown to illicit electrophysiological responses in the adult fly, but no compounds tested demonstrated attractiveness.

Key words: Acetone, Bioassay, Egg metabolism, EAG, GC-EAD, GC-MS, Kairomone, Nest parasite, Uropygial gland, Zebra finch

K.M. Doherty Candidate for the degree of Master of Science, April 2019 Stephen A. Teale, Ph.D. Department of Environmental and Forest Biology State University of New York College of Environmental Science and Forestry Syracuse, New York

Introduction

PHILORNIS DOWNSI, AN AVIAN ECTOPARASITE

Discovered in the Galapagos Islands in 1997, Philornis downsi Dodge & Aitken, 1968,

(Diptera: Muscidae) has since become considered a “highly invasive” species and a threat to the islands’ native, endemic, and critically endangered avifauna (Fessl and Tebbich 2002; Causton et al.

2006). Philornis downsi adult females oviposit in the nest material (O’Connor et al. 2010a) of at least

18 species of Galapagos birds (Fessl et al. 2006b). Once hatched, the first instar larvae migrate to the nares of nesting chicks where they inconspicuously feed on blood. Second and third instar larvae then feed at night on the blood of the nesting chicks externally, occasionally becoming semi- subcutaneous and feeding on internal tissue as well (Fessl et al. 2006b). Philornis downsi may also exhibit saprophagous behavior on chicks which have died due to heavy parasitism (O’ Connor et al.

2010a). Antibody immunity studies suggest larvae are also able to feed on brooding adult female birds (Huber et al. 2010), and this has elicited some controversy among researchers as to when P. downsi females oviposit in the nest. After feeding for 4-6 days (or less if their host has prematurely died from parasitism or fledged) (Dudaniec and Kleindorfer 2006), larvae embed themselves in the bottom of the nest material, secreting a milky substance to cement the debris around them (Fessl et al. 2006b). In this cemented cocoon they pupate for approximately two weeks (Dudaniec and

Kleindorfer 2006).

Post-eclosion behavior has not been studied, as adults are only known from reared immature specimens collected from bird nests or adults captured in insect traps. Video evidence has recorded adult females ovipositing in nests but they only remained for a maximum of 10 minutes while walking on inner nest surfaces (O’ Connor et al. 2010a). It is generally accepted that the adults are non-parasitic and feed on decaying organic matter, which is supported by high survival of reared adults on a papaya/protein-based diet developed by Lincango and Causton

(2008).

IMPACTS ON HOSTS IN THE GALAPAGOS

Due to the parasite’s generalist tendencies, a wide variety of terrestrial avian species have been observed as hosts including several endemic and endangered passerines. In several species studied, including the small ground finch (Geospiza fuliginosa), medium ground finch (Geospiza fortis), small tree finch (Camarhynchus parvulus), large tree finch (Camarhynchus psittacula), warbler finch (Certhidea olivacea), woodpecker finch (Cactospiza pallida) and medium tree finch

(Camarhynchus pauper), 100% of nests inspected were parasitized by P. downsi (Fessl et al. 2006a;

Dudaniec et al. 2007; O’Connor et al. 2010b). Fessl and Tebbich (2002) found a 97% P. downsi prevalence rate across 12 species of birds, including eight endemic, three native, and one introduced. To date, the largest number of P. downsi parasites found in a single nest is 182, belonging to the endemic Galapagos mockingbird (Nesomimus parvulus) (Fessl and Tebbich 2002).

The extensively studied small ground finch was found to have the least intensely parasitized nests of the five finch species examined with 33 ± 3 (mean ± SE) parasites per nest (Kleindorfer and

Dudaniec 2009). However, in a single year of another small ground finch study the mean number of parasites per nests was found to be 55.3, equating to 22.4 larvae per nestling (Fessl and Tebbich

2002). Other Darwin’s finches, including the woodpecker finch, small tree finch, warbler finch, medium ground finch, and large tree finch, were found to have 31.6, 28.8, 21.2, 18.8, and 18.5 P. downsi larvae per chick, respectively (Fessl and Tebbich 2002; Dudaniec et al. 2007). Across all studied host species, the average number of larvae per nest has been calculated to have increased

46% from 2000 to 2016, indicating that the severity of nest parasitism is intensifying (Kleindorfer and Dudaniec 2016).

Such intense nest infestation by P. downsi decreases the fitness of fledglings and increases nestling mortality in host species. Galligan and Kleindorfer (2009) studied physical deformation and found beak malformation due to P. downsi early instar activity in the nares of small ground finch fledglings, evidenced by larger nares, as well as shorter and shallower beaks. Although

2 affected birds were not observed to have lower foraging rates, non-hereditary beak changes due to

P. downsi could affect the archetypal natural selection in finch populations. In addition to physical deformation, Fessl et al. (2006a) found that blood loss among chicks can be substantial in cases of P. downsi nest parasitism. In small and medium ground finches, nestlings in parasite-reduced nests

(via 1% permethrin application) as opposed to untreated nests gained body mass faster, had higher haemoglobin concentrations, and had over a two-fold increase in fledgling success (from 33.93% to

86.58%). In another study of the small ground finch, there was a direct correlation between higher haemoglobin level, higher fledgling success rate and lower parasite intensity (Dudaniec et al. 2006).

Fessl et al. (2006b) used figures for conversion efficiency for blood to parasite biomass, nestling mass to blood ratio, larval and nestling mass per nest as well as other factors to estimate percent blood loss per nestling for four species of Darwin’s finches. Calculations indicated a range of 32-

55% of nestling blood was lost to P. downsi larvae, with an average of 9-13.5 larvae per nestling. In

2005, parasitism accounted for 95.2% of the 90.0% fledgling mortality in small ground finch, medium ground finch, and cactus finch (Geospiza scandens) nests. Other years in the same study,

2000 and 2004, found 32.4% and 60.0% mortality due to parasitism for 61.5% and 71.4% fledgling mortality rates, respectively. In all three years, 72.7-100% of nestlings had damaged nasal cavities and in 2004 and 2005, 26.7- 50.0% of nestlings had infected auditory canals, 22.0-33.0% had wounds or contusions and 22.0-33.0 % had larval-induced openings under the wings, legs and on their backs (Fessl et al. 2006b). Since 2000, overall in-nest mortality due to P. downsi for Darwin’s finches has been calculated to be 55±6.2% (mean ± SE) across studies (Kleindorfer and Dudaniec

2016).

Other avian species may be even more vulnerable to P. downsi infestation. On Floreana

Island, the endemic warbler finch has not been found for nearly half a century (Grant et al. 2005).

While P. downsi was only discovered in 1997, museum records indicate the species was in the

Galapagos since at least 1964 (Causton et al. 2006). It is speculated that P. downsi may have

3 contributed to the apparent extirpation of the warbler finch on Floreana. While the warbler finch exists on other islands in the archipelago, each island population of the species, and other endemic birds, is genetically distinct. Such loss of island populations and genetic diversity leaves a diminished ecosystem and lost scientific opportunities (Grant et al. 2005).

Population reduction of the medium tree finch on Floreana Island due to habitat destruction and predator introduction has resulted in its listing as a critically endangered species on the 2009

International Union for Conservation of Nature Red List. In addition to these factors, O’Connor et al.

(2010b) determined that P. downsi is the primary cause of mortality among nestlings in this species.

Based on body size studies from other Darwin’s finches, the medium tree finch experiences a higher than expected parasite intensity. With a 100% nest infestation rate, 41% of fledgling failure can be attributed to parasitism, contributing to 75% overall mortality (O’Connor et al. 2010b).

The mangrove finch (Camarhynchus heliobates), however, is a species that could arguably be most affected by P. downsi. Critically endangered and with only 80-100 individuals estimated to be remaining, threats such as rat predation and habitat destruction have wreaked havoc upon breeding success (Lawson et al. 2017). With rat control, fledgling success in the mangrove finch’s restricted habitat can increase by 28% but P. downsi infestation has been shown to decrease fledgling success by at least an additional 14% (Fessl et al. 2010). Genetic diversity of the last remaining mangrove finch population is historically low, with allelic diversity approaching levels seen in the Fernandina population prior to its extinction in the 1970s (Lawson et al. 2016).

Population viability analyses suggest increasing breeding success of the mangrove finch to a rate that can protect the species will require intensified rat control and also the initiation of a P. downsi control program (Fessl et al. 2010).

PHILORNIS DOWNSI MANAGEMENT

Thus far, an effective management strategy for P. downsi is lacking because research since its discovery in 1997 has focused primarily on impacts. Some limited methods for immediate

4 control have been studied, such as non-pheromonal trapping (Lancango and Causton 2009; Muth

2007), physical barriers to parasitism (Koop et al. 2011), and manual application of pesticides to nests (Fessl et al. 2006a). None of these methods are practical for a large-scale management program but have been used in the most urgent cases such as temporarily mitigating impacts on the mangrove finch. Moreover, these attempted control methods display a limited ability to reduce P. downsi populations.

Other, more long-term, solutions are being investigated, but the lack of knowledge of P. downsi biology has prevented these potential solutions from coming to fruition. For example, sterile-insect technique (SIT), where sterilized males are repeatedly introduced to a population to reduce pest reproduction, would be a viable option for P. downsi control because the island populations are finite and the flies would be released as adults, which are harmless. Additionally, because the Galapagos is a unique and fragile island ecosystem, SIT would be acceptable due to its benign environmental nature. It has no non-target effects because it is species-specific and is non- polluting, chemically or genetically (Alphey 2002).

Biological control, in which a predator or parasitoid of the pest is introduced, is another promising method of P. downsi control. It is a long-term solution that, if executed correctly, is the only management program that is species-specific, self-sustaining and able to work on a very large scale. Biocontrol has already been successfully implemented on the Galapagos using a predatory coccinellid beetle against a pest scale insect, Icerya pushasi (Causton et al. 2006), therefore many infrastructural and administrational details are already in existence. However, choosing a biocontrol agent that will achieve the desired, risk-free result is a long process that requires extensive biological and ecological knowledge of both the pest and the potential control organism.

One of the greatest obstacles to implementing a successful SIT or biological control program is this lack of knowledge of P. downsi behavioral ecology as well as P. downsi population dynamics.

Sterile insect technique programs often fail without explanation despite a high sterile male to wild

5 male ratio. These programs usually lack prior estimation of wild female populations or evaluation of mating competitiveness of the sterile males (Ito and Yamamura 2005). Presently, the only way to exhaustively assess population levels of P. downsi is through examination of bird nests for parasite load which is a laborious task and limited to a small spatial scale (Dudaniec et al. 2007). Both SIT and biocontrol require captive breeding of the pest organism on a sufficiently large scale to allow for a steady source of sterilized males or an experimental laboratory population to ensure host specificity (Hoddle 2003). To date, it has been difficult to rear the insect as adult mating and first instar larva survival remain unexplored or challenging. Recently, diets of chicken-blood, protein powder and milk powder/brewer’s yeast have increased the egg-to-adult rearing success rate from

0.6% to ~10% with eggs oviposited by wild-caught adult females (Lahuatte et al. 2016). This success rate, as well as in-captivity reproduction, must be improved and studied so that captive breeding populations can be established on the mass scale required of these programs.

With more research, the semiochemistry of P. downsi could be a source of effective, long- term, environmentally-safe methods of control or population monitoring. As reviewed by Witzgall et al. (2010), chemical ecology works on the principal that olfactory cues, such as pheromones, host-derived odors or food-related volatiles, can elicit a behavioral response in a particular insect species. Management techniques based on the chemical ecology of a pest species can be used in combination with other techniques or on their own to reduce pest populations. Several widely used management practices utilize olfactory cues in order to lower pest populations: attract-and-kill, where insects are attracted to a baited trap and exterminated; mass trapping, where insects are trapped in large enough numbers to lower the population; or mating disruption, where the wide- spread release of pheromones prevents the sexes from finding each other and mating. If using semiochemicals to lower insect populations is unnecessary due to the presence of another control mechanism, these compounds can be used to assess effectiveness of the control. Spread delineation with traps, where a pest’s movement and establishment is outlined in order to detect incipient

6 populations, is a method often used to slow the spread of a species or to identify a new population that can be eradicated if quickly detected. Population assessment via standardized trapping systems offers baseline information in a variety of pest management scenarios (Witzgall et al.

2010).

Philornis downsi population monitoring is challenging because both adults and larvae are difficult to locate and quantify. Furthermore, indicators of impact, such as fledgling success and beak deformation rates, are highly variable among host species, locations and years (Fessl and

Tebbich 2002; Dudaniec et al. 2007; Galligan and Kleindorfer 2009). Semiochemically based trapping is potentially the most efficient means of monitoring P. downsi populations, which is important in the short-term for assessing the threat to high-risk bird populations and critical in the long-term for assessing the progress of other control strategies such as SIT or biocontrol (Witzgall et al. 2010).

The severity of P. downsi parasitism makes it critical to quickly develop a control technique that is efficient in impact, scale, and longevity. Developing an appropriate lure requires minimal knowledge of insect population dynamics and ecology and can often be developed in less time than multi-phase programs. Methods of control or monitoring that utilize semiochemicals can be instituted without extensive assessment of non-target effects, and because lures can usually be inexpensively manufactured once developed, they can be cost-effective as well (Witzgall et al.

2010).

HOST-ODOR BASED ATTRACTION IN DIPTERA

While lures are often based on sex or aggregation pheromones, the addition of other odorants derived from food, oviposition substrates, and hosts can synergize attraction. These odors occasionally function as the only attractant in a lure if a pheromone has not been identified

(Witzgall et al. 2010). Many investigations of host volatiles have related to insect pests of agricultural or forestry concern. Plant-based chemicals, such as the triterpenes released by squash

7 blossoms (Cucurbitaceae) or the sesquiterpenes released by stressed green ash trees (Fraxinus pennsylvanica) can elicit behavioral responses in pests such as the western corn root worm

(Chrysomelidae) and the emerald ash borer (Buprestidae), respectively (Metcalf et al. 1980; Crook et al. 2008).

Additionally, several parasitic Diptera have been tested for attraction to host volatiles. Most notably, disease-vectoring mosquitoes in the genera Aedes, Anopheles, and Culex, have been found to be attracted to carbon dioxide, lactic-acid and other skin, breath and urine-derived odors released by hosts (Takken 1991). These compounds are often used in combination with light as trap lures in high-priority areas of mosquito control (Burkett et al., 2002). Virgin females of the hematophagous leishmaniasis vector, Lutzomyia longipalpis Lutz and Neiva (Diptera: Psychodidae), show a strong increase in attraction and a sharp decrease in the proportion of non-responders to male sex pheromone when host odors from a hamster are added (Bray and Hamilton 2007). Fannia conspicua Malloch, (Muscidae), is an emerging pest of humans, deer, and cattle in California and is attracted to the host odors carbon dioxide, ammonia, and the combination of the two (Mohr et al.

2010).

Host-derived odor studies of other Diptera are compatible with the suggestion that P. downsi could be attracted to bird or nest odors. Several species of Culex, including C. quinquefasciatus Say, which prefers avian to human hosts, showed strong attraction to the odor of bird feathers, especially in combination with carbon dioxide (Allan et al. 2006). In a separate study,

C. pipiens L. and C. restuans Theobald exhibited attraction to crow uropygial gland secretions in field trapping experiments (Russel and Hunter 2005). Finally, the non-parasitic Caribbean fruit fly,

Anastrepha suspensa Loew (Tephritidae) is more attracted to crude avian feces than the release of ammonia from feces alone should indicate, likely due to the food-related odor of protein decomposition. The authors argue that other chemicals in the feces that remain to be identified are responsible for the increase in attraction (Epsky et al. 1997). While the fruit fly would be attracted

8 to avian feces as a food odor and not a host odor as in P. downsi, this study indicates that odors of bird feces contain chemical compounds that are attractive to other species of Diptera.

AVIAN SOURCES OF OLFACTORY ATTRACTION OF PHILORNIS DOWNSI

Volatiles released by brooding female birds, assisting males, nestlings or developing embryos may attract P. downsi to bird nests. The uropygial gland secretes preening materials in most orders of birds and may produce both nonvolatile and volatile compounds (Campagna et al.

2012). In a large study, including 11 species in four families of Passeriformes, researchers analyzed uropygial gland secretions and found a variety of monoesters that were generally similar across species, though their identity was not confirmed (Haribal et al. 2005).

Uropygial gland secretions vary by season, age, and breeding status. In the Upupidae, an odorous secretion containing several classes of volatile compounds is only produced by breeding females and nestlings (Martin-Vivaldi et al. 2009). In the dark-eyed junco (Passeriformes), a change in the volatile fraction of uropygial secretions was observed in both males and females when they entered the breeding season (Soini et al. 2007). Volatile extracts containing sex-specific proportions of straight-chain alkanols were found to be similar among four Passeriform species, suggesting that these compounds are of common phylogenetic origin (Zhang et al. 2009). In the grey catbird,

Dumetella carolinensis (Mimidae), seasonal changes in uropygial gland volatiles have been found in males (Whelan et al. 2010). Furthermore, a change in ester composition has been observed in several bird species during the breeding season. This change is found only in the brooding sex in species with uniparental incubation, but in both sexes in those with biparental incubation

(Reneerkens et al. 2007). These studies show that birds inhabiting nests vary the composition of uropygial gland secretions according to sex and reproductive stage which suggests that they have the potential to be targeted attractants for P. downsi.

In addition to brooding- or nesting-associated volatiles originating from the uropygial gland, the concentration of bird odors created by constant nest occupancy alone could result in a

9 chemical profile unique to nests. Accumulation of fecal material is a promising source of attraction in P. downsi as evidenced by the coprophagus habits of other Philornis spp. (Dudaniec and

Kleindorfer 2006) as well as reported attraction to volatile products of protein decomposition such as ammonia and putrescine (Lincango and Causton 2009). In one study, ground bird nest predation rates were higher in nests experimentally closer to fecal material (Petit et al. 1989), suggesting that feces emit an odor that mammalian predators recognize. Bird eggs may be another source of attractive odors, as nitrogenous waste typically accumulates in the allantoic fluid (ten Busch et al.

1997) and volatile components could either be released through the shell during development, or released in high concentrations at the time of hatching.

The passerine hosts of P. downsi may relay location information to their parasite though the release of volatiles from their nests. Specifically, the sources may include adult bird uropygial glands, fecal accumulation within the nests, or developing bird eggs. The objective of this study is to collect and identify the volatiles from nests, as well as more specific host volatile sources, and assess them for attraction with adult P. downsi. Attractive compounds could then be used as lures to trap the adult P. downsi for either population monitoring or control.

Methods

LIVE ANIMALS

Birds

Nest odors were sampled from three groups of birds: (1) captive zebra finches, (2) wild birds in Central New York, and (3) wild finches in the Galapagos.

Zebra finches (Taeniopygia guttata) were maintained in accordance with the Institutional

Animal Care and Use Committee of SUNY-ESF (Protocol #120301). The colony was started with two pairs in 2012. They were fed a seed-based diet for finches (Kaytee® Supreme®, Kaytee Products,

Inc 521 Clay Street, Chilton Wisconsin 53014), with occasional additions of millet sprigs, boiled egg shells/whites, fresh fruit, root vegetables and leafy greens. The colony was kept under photoperiod regimes between 14:10 and 12:12 (L:D), with day length increasing or decreasing to stimulate or discourage breeding as needed. The light source was a timer-controlled, double-tube fluorescent lamp (1.2 m). Nesting boxes were constructed of galvanized steel electrical boxes and lined with sterile cotton in order to minimize contaminating odors, though the birds occasionally built nests in other areas of the cage. Birds that were no longer needed for laboratory experiments were distributed into the local pet trade.

To collect nest volatiles from wild birds in central New York, 40 nesting boxes were made according to the New York State Bluebird Society (http://www.nysb.org/) specifications and hung on trees or fence posts around an active hay field or an old field in Pompey, NY (42°52'53.7"N

76°01'13.5"W) in March of 2012. No successful nesting occurred in the boxes in 2012. In early

March of 2013 and 2014, the boxes were cleaned small mammal or old bird nests and checked periodically for nesting until the nesting season ended in late July. Birds that utilized the nesting boxes included eastern bluebirds (Sialia sialis), tree swallows (Tachycineta bicolor), and house wrens (Troglodytes aedon).

Wild bird nests were also odor-sampled in the moist upland area of Los Gemelos

(0°37'42.7"S 90°23'23.6"W) and the arid lowland area of El Barranco (0°44'14.5"S 90°18'05.1"W) on Santa Cruz Island, Galapagos Province, Ecuador. Nests belonging to the green warbler-finch

(Certhidea olivacea) and small tree-finch (Camarhynchus parvulus) were located, assessed for nesting status, and subsequently sampled.

Insects

Philornis downsi specimens used for electrophysiological and behavioral experiments were obtained from the Charles Darwin Research Station in Puerto Ayora, Ecuador. Puparia were collected by collaborators (see Acknowledgements) from wild bird nests that had fledged or were abandoned due to brood mortality. Puparia were kept separate until eclosion, then transferred to jars containing sex and age cohorts. Wild-caught adults were collected from ball traps baited with blended papaya and the sexes were kept separate unless otherwise noted. Adults were fed a diet of blended papaya, brown sugar, protein powder, and milk powder in separate petri dishes. Flies were kept at 23.5°C with 70% relative humidity on a 14:10 light:dark schedule.

VOLATILE COLLECTION AND ANALYSIS

Zebra finch nest headspace

Five different methods were used to collect volatile compounds from zebra finch nest headspace. All methods involved air collection at 100-200ml/min through a Porapak Q (80-100 mesh; Sigma-Aldrich Co., St. Louis, MO, USA) adsorbent trap (Appendix I-A). Sampling time ranged from two to twenty-four hours. Nest headspace was sampled from nesting boxes with nesting activity, nests with eggs and nests with chicks. Control samples were air from nesting boxes without evidence of nesting activity, unfiltered laboratory air and a cotton-lined nesting box without previous contact with nesting birds. Prior to first use, the adsorbent was cleaned with ~15 ml elution solvent or Soxhlet extraction with chloroform for 8 hours then rinsed with elution solvent.

Volatile compounds were eluted from the traps with analytical grade solvent (pentane or an equal parts mixture of dichloromethane and hexane) and stored at -60°C or -4°C until analysis or behavioral assay experiments. Extracts were then analyzed using coupled gas chromatography- mass spectrometry (GC-MS; GC 7890A – MS 5975C VL MSD, Agilent Technologies, Santa Clara, CA,

USA) equipped with either an HP-5MS (30 m X 0.25 mm, 0.25 µm film thickness; Agilent

Technologies) or DB-WAX (30 m X 0.25 mm, 0.25 µm film thickness; J&W Scientific, Folsom, CA,

USA) column. Sample aliquots (1µl) were injected in splitless mode with a purge time of one minute and a solvent delay of four minutes. For analysis using the HP-5MS column, the injector temperature was set to 280°C, and the oven was programmed for an initial temperature of 40°C for one minute, then increased 3°C per minute to 300°C and held for 20 minutes. The transfer line was set to 280°C. For analyses using the DB-WAX column, the injector temperature was set to 250°C and the oven was programmed for an initial temperature of 40°C for one minute, then increased 5°C per minute to 240°C and held for 20 minutes. The transfer line was set at 250oC. The electron ionization

(70 eV) mass spectrometer scan range was m/z 40 to 500 u for all analyses. Aspects of the sampling and analysis methods were varied as detailed in Table 1.

Table 1. Collection and analysis methods for zebra finch nest dynamic headspace collection. Sampling Collection Analysis Sample method Date Controls Sampling Extraction Column size Inactive a) HP-5MS 16 Mar-Jun 24hr @200 ml/min 1 nest/Lab 1 ml pentane 2013 50 mg Porapak Q b) DB-WAX 21 air Inactive Jun-Oct 24hr @200 ml/min 2 nest/Lab 500 µl pentane DB-WAX 8 2013 50 mg Porapak Q air Inactive 250 µl Jul 4hr @100 ml/min 3 nest/Box dichloromethane: DB-WAX 4 2014 50 mg Porapak Q with cotton hexane (1:1) Inactive 250 µl Jul 8hr @ 100 ml/min 4 nest/Box dichloromethane: DB-WAX 5 2014 50 mg Porapak Q with cotton hexane (1:1) 2hr @ 100 ml/min Inactive 250 µl Jul 50 mg Porapak Q 5 nest/Box dichloromethane: DB-WAX 10 2014 Nest entrance with cotton hexane (1:1) covered with jar lid

In addition to dynamic headspace collection, volatile compounds originating from zebra finch nests were collected using solid-phase microextraction (SPME) in August 2014 (Appendix I-

B). Carboxen/polydimethylsiloxane (75 µm) coated SPME field collector fibers (Sigma-Aldrich Co.

LLC, St. Louis, MO) were exposed to zebra finch nests in various stages of nesting, including nest building, nests with eggs, or nests with chicks. An electrical box filled with sterile cotton that had not had direct contact with the birds was used as a control. All nesting box and the control box entrances were covered with a solvent-washed aluminum-lined jar lid to reduce outside air contamination and did not contain brooding adult birds during sampling. Fibers were exposed to boxes for 30 minutes and then immediately placed in the GC-MS injection port. The injector was set to splitless mode at 250°C. The carrier gas was helium and the column was DB-WAX (30 m X 0.25 mm, 0.25 µm film thickness; J&W Scientific, Folsom, CA, USA). The oven temperature program was

45°C for 1 minute, a ramp of 10°C/min to 240°C, then 20 minutes isothermal. The transfer line temperature was set to 250oC. The mass spectrometer scan range was m/z 33–450 u. There were eight samples, including three controls.

Wild bird nest headspace

Three collection methods were used to sample nest headspace from wild birds in central

New York. In all three methods, nest air was collected for six hours on Porapak Q adsorbent traps

(Appendix I-C). Bird boxes were sampled if they contained nests with eggs or chicks and the species of occupant was noted. Controls included empty nest boxes or boxes with unused nests when available. Analysis of compounds was done with GC-MS equipped with a DB-WAX column with parameters identical to laboratory nest headspace samples. Details of the three methods are in

Table 2.

Table 2. Collection and analysis methods for central New York wild bird headspace collection. Sampling Collection Analysis Sample method Date Controls Sampling Extraction Column size May-Jun 6hr@500 ml/min 1 ml 1 Empty box DB-WAX 3 2013 500 mg Porapak Q dichloromethane 6hr@500 ml/min 2 times with 500 µl 2 Jun 2013 Empty box DB-WAX 25 500 mg Porapak Q dichloromethane 250 µl Jun-July Empty box/ 6hr@200 ml/min 3 dichloromethane: DB-WAX 15 2014 unused nest 100 mg Porapak Q hexane (1:1)

Two methods (Table 3) were used to sample nest volatiles from nesting finches in the

Galapagos. For all methods, headspace from within a wild bird nest was pulled via vacuum pump through a Porapak Q adsorbent trap, or through a Porapak Q adsorbent trap and an activated carbon (50-200 mesh) adsorbent trap in series. Species included warbler finch, small tree finch, and medium ground finch and nest conditions included nests with eggs, nests with chicks, or nests with both. After collection, traps were eluted with solvent and subsequent samples were transported to

Syracuse, NY for analysis. Gas chromatography-mass spectrometry analysis was performed using a

DB-WAX column as outlined for zebra finch nest headspace analysis.

Table 3. Collection and analysis methods used for Galapagos wild bird nest headspace collection. Sampling Collection Analysis Sample method Date Controls Sampling Extraction Column Size Feb Open air ≥5 m 24hr @ 500 ml/min 1 ml 1 DB-WAX 8 2013 away from nest 5 mg Porapak Q dichloromethane 6hr @ 200 ml/min 250 µl Mar Open air ≥5 m 2 100 mg Porapak Q; dichloromethane: DB-WAX 5 2014 away from nest 100 mg activated carbon hexane (1:1)

Fecal samples

Feces were collected from the zebra finch colony for headspace collection and direct extraction. For headspace collection, fresh (≤10 minutes old) feces were collected on Teflon sheeting that lined the bottom of the bird cages. A small piece (4 cm2) of Teflon sheeting

15 surrounding each fecal sample was excised. Each fecal sample was weighed before being placed in a glass vial (2 ml) topped with a Teflon-lined septum. Charcoal filtered air was drawn into the vial containing feces and collected on Porapak Q (6.5 mg). The flow rate was 40 mL/min and headspace was collected for 30 minutes. To extract the samples, the adsorbent traps were rinsed with two separately collected aliquots (100 µl) of hexane:dichloromethane solution (1:1). The control was a vial containing a piece of Teflon (4 cm2) from the cage bottom but no with feces. Four samples were taken including one control and eight aliquots were analyzed.

Direct extracts of feces were made by scraping fresh fecal samples off of the newspaper- lined cage bottom with a spatula; controls were scraped newspaper from the cage without feces.

Feces or a clean spatula were placed in a glass vial containing sodium sulfate (as a desiccant) and hexane:dichloromethane solution (1:1, 300 µl), briefly agitated by hand, and left to sit at ambient temperature for 30 minutes. Glass filter paper was used to press the insoluble materials to the bottom of the vial before a solvent aliquot (1 µl) was removed from the top and injected into the

GC-MS. Thirteen samples, including two controls, were taken and analyzed.

For both the headspace extraction and direct extraction of feces, GC-MS analysis took place using a DB-WAX column using the same parameters as was used for nest headspace analysis.

Uropygial gland secretions

Uropygial gland secretions were collected from July to November 2014 from male and female laboratory-reared zebra finches in various breeding stages, including non-breeding, nest building, or brooding. The secretions were collected by gently squeezing the glands with clean forceps and applying a capillary pipette (1 µl) tip to the exuded droplet of secretion. The capillary pipette was then placed in a vial with 250 µl of dichloromethane, hexane:dichloromethane solution

(1:1), or anhydrous methyl alcohol and vortexed for 30 seconds. A total of 37 samples were taken and were kept at -4°C until GC-MS analysis or biological assessment with P. downsi. Prior to agitation, sample vials were weighed to estimate uropygial gland secretion volume.

Analysis of gland secretions was done on a DB-WAX column using the previously stated methods. For secretion samples that were dissolved in anhydrous methyl alcohol, volatile components were collected on carboxen/polydimethylsiloxane solid-phase microextraction fibers by piercing the vial septum and exposing the fiber for 30 minutes.

Egg headspace

Mature, viable zebra finch eggs were collected from zebra finch nests within three days of estimated hatch start (pipping). Viability and pipping time was estimated by observing the developmental stage of the embryo and the size of the air space by candling the eggs. For headspace volatile collection, each egg was placed under a heat lamp that heated the work area to 37°C. A small

(~1 mm) hole was made in the shell and shell membrane using a sterile pin. A capillary pipette was inserted into the hole and acted as a protective sleeve for a solid-phase microextraction fiber

(carboxen/polydimethylsiloxane, 75 µm). The tip of the fiber was placed 1-2 mm into the air space for 30 minutes (Appendix I-D). After sampling, the egg was sealed with a thin layer of non-toxic glue

(Elmer’s Products, Inc., Westerville, OH) and placed into the nest to hatch. Viability was confirmed by observed hatching within three days. The controls for the egg headspace experiment were SPME fibers (1) not exposed to egg headspace, and (2) exposed to non-viable eggs when available. Fibers were placed in the GC-MS injector port set to 250°C and analysis parameters were the same as those used for zebra finch nest SPME samples.

ELECTROPHYSIOLOGICAL AND BEHAVIORAL ASSESSMENT

GC-EAD/EAG

Acetone and uropygial gland secretion samples were tested for electrophysiological response in adult P. downsi using coupled gas chromatography-electroantennographic detection

(GC-EAD) and electroantennography (EAG). Samples were prepared by collecting 5-10 µl of air with a gas-tight syringe from the sample headspace (EAG only), by taking 2 µl of sample extract, or by

17 collection of sample headspace via SPME fiber (carboxen/polydimethylsiloxane, 75 µm). Gas chromatography-electroantennographic detection was performed using an HP 5890 Series II GC

(Hewlett-Packard, Sunnyvale, CA, USA) equipped with an HP-5MS capillary column (30 m X 0.25 mm ID, 0.25 µm film thickness; Agilent Technologies, Wilmington, DE) and helium as the carrier gas at a flow rate of 1 ml/min. The SPME fibers or 2 µl aliquots were injected in splitless mode with the injector temperature set to 220°C. The oven was programmed for an initial temperature of 50°C for one minute, then increased 10°C per minute to 280°C and held for one minute. Nitrogen make-up gas was added at 8 ml/min to the column flow using a “Y” glass splitter (Supelco, Bellefonte, PA) and the column effluent was split 1:1 in the oven via an additional “Y” glass splitter. One arm of the second splitter led to the flame ionization detector (FID) (280°C) and the other to the heated EAD port (280°C) introduced into a glass cooling-jacket with a humidified air stream (300 ml/min) directed toward the antennal mount preparation.

Electroantennography was performed by manually introducing (“puffing”) the headspace volatiles of the samples into a humidified air stream (300 ml/min in a cooling jacket) directed 2 cm upstream of the antennal mount preparation. At the conclusion of each EAG trial, a control sample of equivalent volume of headspace from an empty vial was introduced.

Antennal preparations of adult P. downsi were made by mounting the excised head and prothorax of individual flies on one end of a 300 µl polypropylene microinsert (Thermo Scientific,

Rockwood, TN) filled with either conductive saline solution (7.5 g NaCl, 0.21 g CaCl2, 0.35 g KCl,

0.20 g NaHCO3, 1 l of deionized water) or conductive gel (EcG electrosurgery-TENS). A gold wire electrode immersed in a saline-filled capillary tube (1 mm dia.) was inserted at 45° into the third antennal segment (recording electrode) and another gold wire electrode inserted into the base of the head of the adult fly dorsally (indifferent electrode). This preparation was mounted on a custom acrylic holder (Appendix II). The output signal from the antennae was amplified (10x) and recorded on a computer using the Syntech IDAC-2 system and dedicated software (v1.2.4; Syntech).

Sample acquisition and injection for each trial is detailed in Appendix III.

Laboratory bioassays

Two olfactometer designs were used to test P. downsi behavioral response to positive controls or nest odor treatments. The first olfactometer was constructed using a plexiglass arena

(51 cm x 51 cm x 123 cm) with two doors on the front (long) wall. Two ball traps (ISCA

Technologies, Riverside, CA) were placed directly over two circular openings (11.5 cm diameter) at the top of the olfactometer. All ball traps used for laboratory or field bioassays consisted of two sections: a clear plastic top containing a small treatment compartment conventionally used for lure placement, and a yellow plastic bottom section with a central hole allowing insect entry but discouraging insect exit. In this bottom section, a donut-shaped moat served as the reservoir for any papaya-based mixtures. A double fluorescent lamp (1.2 m) illuminated the arena from above the ball traps. Vacuum was applied to a hole in the middle of the arena front wall and drew air out of the arena at 500 ml/min; air entered through the ball traps and into the arena (Appendix IV-A).

To assess the efficacy of the olfactometer with P. downsi adults, the ball traps received either the positive control (15 ml papaya) or negative control (15 ml deionized water) in the trap moat. The positive control was papaya juice made by blending one whole papaya without seeds, ~250 ml water and ~200 g sugar. Treatment placement was randomized for each trial. To test nest odor attractiveness, extracts from sampling method 2 of zebra finch nest headspace collections (Table 1) were used instead of papaya positive control. In this case, filter paper was placed in the lure compartment of the ball traps and 250 µl of either nest odor extract or pentane control was placed on the filter paper. For both olfactometer efficacy and nest odor attraction experiments, groups of five wild-caught adult female flies were released in the center of the arena for each trial. After four hours, the ball traps were examined to determine the number of flies captured in each.

The second olfactometer design involved a small glass y-tube (1 cm diameter, 6 cm arm length, 7 cm base length) placed at a 35° incline with a full-spectrum light suspended above. The y-

19 tube was placed in an open-top white cardboard box to eliminate external visual cues. Each arm received charcoal-filtered humidified laboratory-supplied air at 40 ml/min which passed over a filter paper before entering the arms of the y-tube (Appendix IV-B). To test the efficacy of the olfactometer design, the filter papers were treated with either aged blended papaya juice (250 µl; positive control) or deionized water (250 µl; negative control) randomly assigned to the arms. The blended papaya juice was aged 15 days in a ball trap at ambient temperature (~70oC). Laboratory- reared female flies were released individually at the base of the y-tube and observed for 30 minutes.

Field bioassays

Acetone was evaluated for attractiveness to P. downsi in two trapping experiments using ball traps. In the first experiment, the trapping efficacy of a moderate acetone release rate was tested. For this experiment, traps were placed at El Barranco for seven days, (20 to 26 Oct, 2014) and hung 3–5 m above the ground from various tree species. Traps were spaced ~10 m apart along

5 transects and were checked and rotated every three days. Treatments were randomly assigned and consisted of (1) blended papaya and water-filled polyethylene pipette, (2) blended papaya and acetone-filled pipette, and (3) water and an acetone-filled pipette. Blended papaya was the positive control and was made from one whole papaya without seeds, ~250 ml water and ~200 g sugar. The papaya juice or water was placed in the bottom of the ball trap, and the pipettes were suspended from the top of the ball trap and hung bulb-down inside the trap cavity. Pipettes were made from disposable polyethylene pipettes (2 ml), filled with 2 ml of industrial-grade acetone or water and heat-sealed. The release rate of acetone was gravimetrically shown to be 25.16 ± 1.10 (mean ± SD) mg/day at ambient temperature in laboratory experiments.

In the second experiment, attractiveness of acetone to P. downsi was evaluated at low and high release rates. Traps for this experiment were hung at El Barranco for nine days (27 Oct to 04

Nov, 2014) and were hung 3–5 m above the ground ~10 m apart. Every 2–4 days the traps were

20 checked and rotated. Treatments were randomly assigned along three transects and included (1) papaya juice, (2) papaya juice and low release rate acetone, (3) water and low release rate acetone,

(4) papaya juice and high release rate acetone, and (5) water and high release rate acetone. Papaya juice was made as above, but acetone lures were made using glass screw cap vials (2 ml). The low release rate treatment was made by using a polyethylene septum in the vial cap cut from the pipettes used in the moderate release rate experiment. The high release rate treatment was made by using a parafilm septum in the vial cap. Laboratory gravimetric measurements indicated the low and high acetone release rates to be 3.83 ± 6.52 mg/day and 64.30 ± 0.99 mg/day (mean ± SD), respectively. Blended papaya or water was placed at the bottom of the ball traps for this experiment, while the acetone lures were suspended in the trap cavity from the top of the ball trap.

In addition to acetone, hexanal was evaluated for attractiveness via field bioassay in two experiments during March–April 2015. These experiments combined potential host-related attractants and food-related attractants in order to reduce the number of individual trapping experiments during the field season. For the first experiment, a total of 42 ball traps were hung 10 m apart and 5–6 m high along seven transects in El Barranco. Traps were placed in the field for four days and checked and rotated every other day. Treatments were (1) papaya juice, (2) papaya juice and hexanal, (3) blackberry juice, (4) blackberry juice and hexanal, (5) yeast, and (6) yeast and hexanal. To make the positive controls containing only food attractants, 500 ml of papaya juice or blackberry juice were combined with 100 g of sugar and 1.5 l of water. Fifteen grams of yeast was added to 1.5 l of warmed (30–35oC) water and 75 g sugar mixture. Each trap bottom was filled with

150 ml of a food attractant. Hexanal was added to traps by placing a 2 cm2 piece of absorbent cloth with 1 ml hexanal inside a 2 mm thick polypropylene bag. These bags were taped to the top interior of the ball traps.

A second trapping experiment was performed at both El Barranco and Los Gemelos. Ball traps were hung 10 m apart and 5–6 m high along eight transects. Traps were checked and rotated

21 every other day during the six-day experiment. Treatments were (1) blackberry juice alone, (2) hexanal and blackberry juice, (3) hexanal, acetic acid, and blackberry juice, (4) methyl butanol, butanediol, linalool and blackberry juice, and (5) a mixture of all five compounds and blackberry juice. Twenty microliters of each compound was added to 150 ml blackberry juice/sugar/water mixture in the bottom of the ball trap.

For each of the four trapping experiments, the number and sex of adult P. downsi captured were recorded. Significant differences among means of the treatments were evaluated using SAS® software version 9.4 (SAS System software, SAS Institute Inc. 2012) and was assessed for each sex separately. All data were evaluated for adherence to analysis of variance (ANOVA) assumptions using Shapiro-Wilks and Levene’s tests. Pairwise comparisons after variance analysis were performed using post-hoc Tukey’s HSD tests at α=0.05. If data sets were not found to meet ANOVA assumptions, they were analyzed by fitting Poisson or negative binomial regression models (via

PROC GENMOD), with model choice based on standard model fit parameters such as Akaike information criterion (AIC) and Bayesian information criterion (BIC). Treatment differences were found using LSMEANS with a Tukey-Kramer adjustment for pairwise comparisons and an alpha of

0.10.

Results

VOLATILE COLLECTION AND ANALYSIS

Laboratory bird nest headspace

Dynamic headspace sampling of zebra finch nests resulted in few volatile compounds being consistently detected by GC-MS. Among the most consistently isolated compounds were straight- chain aldehydes (hexanal, heptanal, nonanal) and alcohols (1-octanol, 1-octen-3-ol, 1-hexanol, (E)-

2-nonenal). All compounds that were not found in controls are listed in Appendix V.

In analyzing SPME-collected nest samples, some trace compounds were detected. Acetone, ethanol, hexanal, and acetic acid were found in all non-control samples which included zebra finch nests being built, containing eggs and containing chicks. Several compounds, including 1-pentanol,

3-hydroxy 2-butanone, and 1-heptene, were found only in nests that contained eggs or chicks.

Acetone, hexanal, nonanal, and 1-octanol were verified using reference compounds.

Wild bird nest headspace

All sampling methods used to collect wild bird nest odors, from both New York bird nests and Galapagos bird nests resulted in significant nesting material contamination. The green leaf volatiles present in these samples, as well as other contaminants, compromised these samples extensively and no compounds of bird origin were identified. Despite this, four compounds

(heptanal, nonanal, 1-octanol, and (E)-2-octenal) identified from zebra finch nest samples were also found in central New York wild bird nest samples.

Fecal samples

Dynamic headspace collection of zebra finch feces resulted in samples that contained no compounds except for methyl butyrate found in one sample. Direct fecal extracts contained six trace components, three of which were too low in abundance to identify. Of the remaining three

23 compounds, two were found in more than one sample. These were identified as bis (2-ethylhexyl) ester hexanedioic acid and squalene.

Uropygial gland secretions

Uropygial gland secretions taken from laboratory-reared zebra finches yielded complex chemical samples with high inter-sample variation. Compounds eluted between 18 and 57 minutes and were mostly esters. Because of the chemical classes and late eluting times, many of these compounds are likely non-volatile at ambient conditions. A comprehensive list of compounds found in uropygial gland samples is in Appendix VIII.

Egg headspace

Samples collected from developing egg air spaces were found to contain only acetone, verified by external standard. Two non-viable eggs did not contain acetone, but instead contained methyl-branched butanoic acids and ketones.

ELECTROPHYSIOLOGICAL AND BEHAVIORAL ASSESSMENT

GC-EAD/EAG

Acetone was evaluated for antennal response with adult Philornis downsi, due to its discovery in egg headspace. Antennal responses were elicited four times from a wild-caught female

P. downsi during an EAG recording. Two out of seven GC-EAD trials with acetone resulted in an electroantennographic response (Figure 1). Results for individual trials are found in Appendix III.

Uropygial gland secretion samples were tested for antennal response with adult female P. downsi. Out of eight GC-EAD trials, seven resulted in an electroantennographic response to uropygial gland compounds. Retention indices for antenally active compounds ranged from 707 to

2823 and indicate the presence of possibly 40 different antenally active compounds (Figure 2).

Figure 1. Selected GC-EAD trial of acetone exposure to Philornis downsi antennae, with flame ionization detector peak (bottom) and simultaneous electroantennographic response (top).

Figure 2. Selected GC-EAD trial of uropygial gland secretion exposure to Philornis downsi antennae, with flame ionization detector peaks (bottom) and simultaneous electroantennographic responses (top).

Laboratory bioassays

The large plexiglass arena olfactometer was evaluated for efficacy and used to determine attraction of P. downsi to various nest odor samples. Using blended papaya as a positive control and deionized water as a negative control, four trials were performed. Of the four trials, a choice of the positive control was made by one out of five flies during the first trial. No flies in any of the trials chose the negative control. During three trials to assess the attraction of P. downsi adults to nest odor extracts, no flies made a choice of either the nest odor treatment or the negative control by the end of the four hour observation window.

The Y-tube olfactometer was evaluated for efficacy only. Out of four trials, one trial resulted in the fly choosing the negative control. No flies in any trial chose the positive control during the 30- minute observation window.

Field bioassays

Significant differences were found in trap catch between treatments for both male and female P. downsi in the first experiment using acetone as a trap lure (F2,12 = 4.75, p=0.0302; F2,12-

=4.78, p=0.0027; respectively), as assessed by ANOVA and subsequent post-hoc Tukey’s HSD multiple comparisons testing. McPhail traps baited with water and moderate release-rate acetone were found to capture significantly fewer adult male and female P. downsi compared to traps baited with water and the positive-control papaya, though for both sexes, papaya and acetone baited trap catches did not differ significantly from either treatment. The water and acetone treatment did not result in any adult P. downsi being captured for any of the five replicates (Figure 3). Male catch rate was overall very low, with only a total of eight (8) males caught in all 15 traps.

Figure 3. Observed mean number of adult Philornis downsi captured in traps baited with papaya and water (C), papaya and moderate release-rate acetone (PA), and water and moderate release-rate acetone (WA). Bars with corresponding letters that differ have significantly different means (Tukey’s HSD, α=0.05), with capitalized male and lowercase female letters.

The second trapping experiment involving acetone included treatments with low release- rate and high release-rate acetone. The data for this experiment did not conform to ANOVA assumptions of normality or homogeneity of variance for either sex so these were fit to regression models, with the male and female data sets fitting a scale-corrected Poisson model and a negative binomial model the best, respectively. However, there were no adult P. downsi captured in any trap containing only water and acetone, regardless of release-rate which resulted in aliased parameters and inappropriate least-square means testing. The water and acetone treatment data were removed from the analyses, and the resulting data sets were then suitable for ANOVA. For both male and female P. downsi trap catches, there was no observed treatment effect and there were no significant differences among treatment catch means (male: F2,6 = 0.50, p=0.6297; female: F2,6=0.26,

27 p=0.7817) (Figure 4). As with the first acetone field bioassay, male trap rates were extremely low, with only two (2) male P. downsi captured in all 15 McPhail traps.

Figure 4. Observed mean number of adult Philornis downsi captured in traps baited with papaya (C), papaya and low release-rate acetone (PL), water and low release-rate acetone (WL), papaya and high release-rate acetone (PH), and water and high release-rate acetone (WH). Bars with corresponding letters that differ have significantly different means, with capitalized male and lowercase female letters. No significant differences were found among treatments (ANOVA F-test, α=0.05).

For the trapping experiment testing hexanal and food odors, male and female trap catches conformed to assumptions of normality and homogeneity of variance so were analyzed using

ANOVA. For both sexes, there was no evidence of treatment effect on trap catch (male: F5,36 = 1.74, p=0.1494; female: F5,36=1.71, p=0.1566) (Figure 5).

Figure 5. Observed mean number of adult Philornis downsi captured in traps baited with papaya juice (P), papaya juice + hexanal (PH), blackberry juice (B), blackberry juice + hexanal (BH), yeast (Y), and yeast + hexanal (YH). Bars with corresponding letters that differ have significantly different means, with capitalized male and lowercase female letters. No significant differences were found among treatments (ANOVA F-test, α=0.05).

The final trapping experiment was performed at two locations with treatments including hexanal, blackberry juice, and several food odor compounds. Data from the two sites were analyzed separately. Trap catch data from traps placed in El Barranco were found to meet ANOVA assumptions. For both male and female P. downsi there were no differences in trap catch between the treatments (male: F4,35 = 0.90, p=0.4748; female: F4,35=1.97, p=0.1206) (Figure 6A). Los Gemelos trap catch data for male and female P. downsi severely violated assumptions of homogeneity of variance and female trap catches were not normally distributed. Regression analysis using the negative binomial distribution was used for both data sets. Treatment was not found to influence trap catch, and no statistically significant treatment differences were found (Figure 6B).

Figure 6. Observed mean number of adult Philornis downsi captured in traps baited with blackberry juice (C), hexanal and blackberry juice (T1), hexanal, acetic acid and blackberry juice (T2), methyl butanol, butanediol, linalool and blackberry juice (T3) and all five compounds and blackberry juice (T4) from both El Barranco (A) and Los Gemelos (B). No significant treatment differences were found (A: ANOVA F-test, α=0.05; B: Tukey-Kramer, α=0.10), as denoted by shared capitalized letters for male means and lowercase letters for female means.

Discussion

WHOLE-NEST VOLATILES

Because of the variety of methods utilized and fluid nature of nesting status among individual zebra finches, it is difficult to draw conclusions from the dynamic headspace sampling of finch nests. The method that collected the most compounds was a 24 hr sampling on 50 mg Porapak

Q extracted with 500 µl pentane (Method 2, Table 1), but this method did not detect four (heptanal,

1-hexanol, decanal, or hexanoic acid) of the 12 compounds consistently identified from all methods combined. This suggests that none of the methods used for dynamic headspace sampling was suitable for collecting all nest volatiles. Hexanal and nonanal were found in all methods but also in all stages of nesting status from nest building to chick-rearing. They were found in 78% and 80%, respectively, of all non-control samples taken. Compounds that were found only in specific nesting statuses were among the least reproducible (Appendix V). The methods used to collect nest odors via dynamic headspace sampling may not have been the most suitable way of collecting chemical compounds that are emitted from zebra finch nests at various nesting stages.

Static headspace sampling of zebra finch nests yielded greater reproducibility of compounds detected. Reduced sampling time needed for static headspace sampling could have reduced the number of opportunities for contamination, amplified by the ability to exclude the adult bird from the nest during sampling. Additionally, the solventless method of solid-phase microextraction allowed GC-MS analysis to include earlier elution times leading to the identification of hexanal, acetone, and ethanol peaks. Prior to this method, hexanal was obscured by the solvent peak during dynamic headspace sampling and acetone and ethanol were unknown components of nest headspace. All non-control samples included acetone, ethanol, hexanal, 1-pentanol and acetic acid. Three compounds (1-pentanol, 3-hydroxy-2-butanone, and 1-heptene) were only found in nests that actively contained eggs or chicks. However, more nest odor collections using this method

31 should be performed to ensure reliability of the results due to the small sample size of the eight nests sampled.

In addition to zebra finch nests in captivity, wild bird nest headspace was sampled. Wild birds in central New York experience exposure to natural diets and have seasonal breeding patterns, like those of P. downsi hosts. However, the wild New York birds were not kept in controlled laboratory conditions and headspace samples were heavily contaminated with green- leaf volatiles from nesting materials and surrounding flora. Dynamic headspace sampling using current methods was not effective and no patterns of volatile composition were discernable. It cannot be concluded that the four compounds (heptanal, nonanal, 1-octanol, and (E)-2-octenal) identified from wild bird nests and zebra finch nests originated from birds directly, as these compounds are common green leaf, fruit and fungal volatiles (Hui et al. 2010 ; Thakeow et al. 2008) that could have originated from the environment.

Nests of known hosts were sampled on the field in Galapagos to eliminate both laboratory- based factors and questionable host-status present in zebra finch and central New York bird nest samples. Because of the remote location of the nests in the Galapagos, samples were heavily contaminated upon receipt and results were inconclusive. These host-nest headspace samples would contain the true odor utilized by P. downsi to chemically locate the nests, so methods should be altered and transportation issues addressed in order to further pursue this area of research.

Despite issues involving inappropriate methodology and small sample size, the compounds that were isolated from headspace sampling of zebra finch nests may be compounds that comprise the odor profile of reproductive zebra finches. Many of the aldehydes, as well as other compounds, found in nest samples have been found in other bird-related volatile studies (Table 4). Most of these studies involve seabirds and ground-nesting birds which are not typical hosts of P. downsi, so these compounds may not be potential host-specific P. downsi attractants. However, in January of 2016 whole-nest headspace collections for zebra finch nests with 10-day old chicks were collected and

analyzed by Kohlwey et al. (2016). Several of the 15 major constituents of these samples were also

found in zebra finch headspace samples in this study, including (E)-2-heptenal, 1-octen-3-ol,

octanal, nonanal and hexanal.

Table 4. Sources of bird-related volatiles found in literature and zebra finch nest headspace. Source Junco Catbird Zebra Compound Chicken Auklet Prion Auklet Chicken Chicken uropygial uropygial finch feathers1 feathers2 feathers3 headspace4 feces5 skin/feet1 glands6 glands7 nests8 1-hexanol * * 1-octanol * * 1-octen-3-ol * (E)-2-heptenal * (E)-2-octenal * Decanal * * * * * * Heptanal * * * Hexanal * * * * Nonanal * * * * * * Octanal * * * * * Acetic acid * * * Hexanoic acid * * * * 1Gallus gallus domesticus (Bernier et al. 2008); 2Aethis cristatella (Hagelin et al. 2003, Douglas et al. 2001); 3Pachyptila desolata (Bonadonna et al. 2007); 4A. cristatella (Douglas et al. 2001); 5G. g. domesticus (Garner et al. 2008, Cooperband et al. 2008); 6Junco hyemalis (Soini et al. 2007); 7Dumetella carolinensis (Shaw et al. 2011);8Taeniopygia guttata (Kohlwey et al. 2016).

Some compounds found in zebra finch nest headspace samples have been shown to play a

role in host-location in other host-parasite interactions. Sand flies (Lutzomyia longipalps Lutz and

Neiva [Diptera:Psychodidae]) have shown antennal activity in response to 1-pentanol, 3-hydroxy-2-

butanone, and hexanal (Dougherty et al. 1999), volatile components of fox (Vulpes vulpes) glands.

Host odor compounds, including 1-pentanol, 1-hexanol, 1-octanol, 1-octen-3-ol, heptanal, octanal,

nonanal, decanal, (E)-2-heptenal, (E)-2-nonenal, and hexanoic acid, have elicited antennal

responses from Culex and Anopheles mosquito species (Du and Millar 1999; Cooperband 2008; Syed

and Leal 2009; Carey et al. 2010). Several species of tsetse fly (Glossina spp.) have shown antennal

33 response to (E)-2-heptenal, (E)-2-nonenal, octanal, and nonanal (Gikonyo et al. 2002) and exhibited attraction to 1-octen-3-ol and acetone (Vale 1980; Vale and Hall 1985). Hexanal, a major volatile component of chicken and rabbit feces, elicited oviposition from L. longipalpis (Dougherty et al.

1995). Krčmar, Mikuška and Radolić (2009) attracted several tabanid species with traps baited with 1-octen-3-ol and acetone, components of livestock odor, while Anderson (2001) observed deer botflies (Cephenemyia spp.) ovipositing on models baited with 1-octen-3-ol. While these compounds are mostly isolated from mammalian hosts and elicit electrophysiological or behavioral response in insects not closely related to P. downsi, they have the capability of being semiochemicals that dipteran parasites use to locate their hosts, as illustrated by the many examples in the literature.

Out of the compounds isolated from whole-nest odor headspace collection, hexanal was chosen for field attraction study using two trapping experiments in the Galapagos. Hexanal was chosen due its repeated identification in headspace collections as well as its presence in the literature as a bird-associated volatile with antennal detection in insects. However, neither trapping experiment indicated any statistically significant positive or negative influence of hexanal on trap catches compared to controls. In all three analyses, the addition of hexanal to blackberry juice did not result in a significant difference in the number of P. downsi caught, indicating that hexanal is not attractive to P. downsi adults. Alternatively, because hexanal is a volatile that has been found in the odor of milk, fish, leaves and fruit (including papaya) (Burdock 2001), it is likely commonly found in the environment throughout the range of P. downsi. It is possible that hexanal is an attractive olfactory component of bird nests, but only when in the presence of other volatile indicators, or synergists. Inagaki et al. (2014) found that hexanal was a constituent of the odor released specifically by stressed rats. It had been previously demonstrated that this odor, when exposed to other rats, induced stress. Hexanal, when combined with another component from the odor, 4- methylpentanal, produced a stress response in other rats but this pheromone combination only worked when the two compounds were exposed together. If an omnipresent volatile such as

34 hexanal is a host-odor cue for P. downsi, a synergist is likely necessary. It also must be noted that neither the experimental lure release rate for hexanal nor the biological release rate of hexanal from bird nests was quantified before the field bioassays were performed. While the hexanal trapping results may indicate that hexanal alone is not attractive to P. downsi, this can only be concluded for the release-rates tested in this study.

Apart from hexanal, several other compounds identified from whole-nest headspace collections should be tested for attractiveness to P. downsi. As previously mentioned, 1-octen-3-ol is elicits oviposition behavior in deer botflies and was also found by Kohlwey et al. (2016) in zebra finch nests. Three aldehydes, octanal, nonanal, and decanal, are found most commonly in the literature as bird-produced volatile compounds. These compounds, because of their presence in the literature, should be investigated first, but all compounds found in zebra finch headspace may function as attractants to P. downsi either on their own or in mixtures. In 2016, acetic acid and ethanol, two compounds found in the headspace collection of yeast but also zebra finch nests, were tested as lures for P. downsi in field experiments. These two compounds, while investigated as a food-source olfactory cue, were found to attract an equivalent number of flies as the yeast-sugar positive control (Cha et al. 2016). The success of these two compounds in eliciting attraction may also be due to their presence in host nests as well as food sources, and should be considered as potential synergists with hexanal and other nest-derived compounds.

INDIVIDUAL SOURCES OF HOST VOLATILES

Feces

Identification of potential olfactory cues for P. downsi from whole-nest headspace samples proved complicated, so specific sources of nest odor were examined individually. Solvent extraction for fecal odor sampling was inappropriate, as evidenced by the lack of compounds found in direct fecal extract. Dynamic headspace collection of fecal headspace only yielded one compound, methyl butyrate, which has been previously recorded as a volatile component of human feces (Ahmed et al.

2013). It is likely there are other volatile components of zebra finch fecal material, though they were not detected using these methods.

Uropygial glands

Unlike the investigation of fecal odors, chemical analysis of zebra finch uropygial gland secretions and egg headspace yielded intriguing results. Zebra finch uropygial gland secretions, when solvent extracted, were shown to be chemically complex. Retention indices calculated for peaks indicated the compounds found in zebra finch uropygial glands are similar, but not identical, to compounds found in the uropygial gland secretions of several other songbirds (Soini et al. 2013).

Because of the sample complexity, zebra finch uropygial gland samples were exposed to P. downsi adult antennae during GC-EAD analysis in order to identify only the compounds that were antenally active.

The electrophysiological responses elicited from adult P. downsi when exposed to uropygial gland samples suggest that the components of gland secretions are detected by the insect. The complicated structure of these compounds and the number of them that elicited response supports the hypothesis that they may play a role in host identification or host location. More trials should be done with the goal of repeatability by exposing several flies to a single gland sample and exposing several gland samples to a single fly.

The chemical profiles of zebra finch uropygial gland secretions would be useful knowledge for many purposes. Zebra finches, as a representative avian study species, should have the components of their uropygial gland secretion cataloged to allow further study of uropygial gland function. For example, detailed studies involving the influence of stress exposure, breeding status, diet changes and relatedness on uropygial gland secretions could be done using zebra finches in the laboratory to increase the understanding of these factors’ interaction with secondary chemical compound production in vertebrates.

The sample acquisition method for zebra finch uropygial gland secretions proved effective and non-destructive to the birds and was approved for use on natural hosts of P. downsi. It is important to continue this area of research both to find olfactory cues that could elicit attraction in

P. downsi and for other studies involving the venerated Galapagos finches. Variation in secretion composition between species could be correlated with rates of nest parasitism by P. downsi to understand host-parasite interactions or with phylogenetic relatedness to enhance the understanding intra-and inter-island evolution and speciation.

Eggs

The air space in zebra finch eggs consistently contained acetone only when the eggs were viable indicating that acetone is produced by all developing zebra finch embryos. Acetone is a by- product of lipid metabolism, though most of the relevant literature involves diabetic humans as subjects. In the case of a diabetic human, improper insulin production leads to the inability for cells to use glucose as an energy source. The body begins to use lipids in the absence of carbohydrates, ultimately forming ketone bodies in the liver that can then be recirculated in the blood as an energy source for other cells. This metabolic state is called ketosis. The ketone bodies formed in the liver, acetoacetate and β-hydroxybutyrate, are converted in non-liver cells to acetyl coenzyme A for energy, but spontaneous decarboxylation of acetoacetate produces acetone (Norris 1997). In the condition hyperketonemia, high levels of ketone bodies in the blood of diabetics results in the accumulation of acetone, which can volatilize and be measured in the breath and skin headspace of diabetics (Crofford et al. 1977; Naitoh et al. 2002). It is likely that this is a major metabolic pathway in developing zebra finch embryos, resulting in the acetone collected in egg airspace. The primary source of energy for developing chicken embryos, the yolk, is composed of only 1% carbohydrates while almost 60% of the dry mass of the yolk is lipids (Bies 1985; Speake et al. 1998). Up to 90% of the total energy requirement of embryos is obtained from the oxidation of fatty acids derived from yolk lipids (Noble and Cocchi 1990; Sato et al. 2006).

The production of acetone by developing birds may function as a highly specific cue for P. downsi adult females. Studies have shown that the pathways for lipid metabolism vary in their usage over development in chicken embryos. Moran (2006) found that lipid metabolism is low early in incubation due to the lack of oxygen able to freely diffuse to the embryo. Beis (1985) found that ketone body-utilizing enzyme levels increase from days 13–17, then decrease until hatching on day 20. After hatching, the enzyme levels in muscles increase dramatically. Similar patterns were found in turkey embryos when gene expression for the production of these enzymes was studied

(Oliviera et al. 2012). This indicates that shifts in lipid metabolism, and therefore acetone production, may create a specific pattern that relays information about egg hatching time to P. downsi adults. Additionally, there may be a difference between the acetone production in altricial

(birds that are blind, mostly immobile, and unable to acquire nutrition themselves after hatching) and precocial birds. This is notable because known P. downsi hosts are exclusively altricial birds.

Studies examining three altricial and two precocial species of bird embryos found differences in patterns of oxygen consumption, which is required for many metabolic processes including ketone body formation. In altricial birds, oxygen consumption increased throughout development, while precocial oxygen consumption plateaued in the final quarter of incubation (Vleck et al. 1979).

In addition to an electrophysiological response using GC-EAD/EAG, a potential behavioral response to acetone was explored when traps baited with acetone at a variety of release rates were placed in the Galapagos. Overall catch rates were extremely low for males, even in positive controls, indicating there were a low number of males in the population during the trapping experiments.

Female P. downsi were more responsive to traps, but the conclusions that can be drawn are limited.

Results indicate for all three release rates across the two experiments, acetone paired only with water was not attractive, as all such baited traps caught zero flies of either sex. The addition of acetone to the positive-control papaya did not alter trap catches significantly. However, when low- release rate acetone was added to papaya it nearly doubled the mean number of female flies

38 captured (Figure 4), going from 6.0 flies per trap to 11.33 flies per trap, but because of the large standard error of the low release-rate treatment, the effect was not significant. The large standard error could be due to low replication (n=3) leading to a non-representative sample, but may also be due to the wide release-rate range of the delivery mechanism calculated in the laboratory (3.83 ±

6.52 mg/day, mean ± SD). Generally, it is likely the rate at which acetone is released from developing bird embryos would be low, and it is possible one or more of the traps baited with papaya and low release-rate acetone had a biologically comparable release-rate. Preliminary experiments not described in this paper indicate that acetone can diffuse through zebra finch egg shells and the finding of acetone in this study’s static headspace collection show this acetone is detectable at the whole-nest scale. From this information, it must be determined the rate at which acetone is released from eggs, particularly how it changes throughout incubation. This would not only aid in creating a more effective lure for P. downsi, but it would provide a non-destructive way to measure the metabolic processes throughout avian embryo development.

LABORATORY BIOASSAYS

Neither in the plexiglass arena nor in the y-tube olfactometers were positive controls able to elicit responses from adult P. downsi. It was observed in both olfactometers that flies remained mostly inactive for the entire duration of the trials. The development of an effective olfactometer is critical to continue investigating chemical attractants of P. downsi whether they are host-emitted kairomones or parasite-emitted pheromones. The reliance upon remote field bioassays for behavioral data is laborious and time-consuming, and it is imperative that attraction be quantified in laboratory conditions. Manipulation of light sources may capitalize on phototaxis to induce movement in the flies and alterations in odor delivery (via air flow rates or the addition of turbulence) may elicit searching behavior. Recently, in experiments not reported in this paper responses have been successfully elicited to positive control samples with adult P. downsi using a vertical y-tube olfactometer with an overhead light source. However, a novel olfactometer design

39 may provide additional advantages. Because adult behavior, including flight orientation habits, is under-studied in the fly’s natural environment, olfactometer design efforts should mimic natural conditions. An olfactometer developed by Eisemann (1988) placed gravid sheep botflies (Lucillia cuprina Wiedemann) in a 1 m3 cage downwind or upwind of visible sheep hosts, finding that flies successfully oriented themselves to hosts only when the source was placed upwind. A larger container, such as the one used by Eisemann, would allow quantification of attraction in P. downsi but also observation of kinesis or taxis patterns in P. downsi generally. Information about how P. downsi orients itself to stimulus is an important aspect of developing an efficient olfactometer, and could lead to enhanced trap design as well.

Conclusion

This study investigated the potential of host-odor cues to act as attractants for the invasive bird parasite, Philornis downsi, with the ultimate goal of providing a novel lure that can be used for population monitoring and control. The chemical composition of bird nest odors was analyzed both from whole nests and from individual sources of volatiles including bird feces, uropygial glands and eggs. While the collection and analysis of these chemical compounds presented challenges, it was ultimately the inability collect reproducible electrophysiological and behavioral responses of the flies to the host odors that prevented the development of a successful lure. This study provides a preliminary library of host-odor compounds that can be tested upon the development of an appropriate olfactometer and precise GC-EAD methodology. Of these compounds, the discovery of acetone in the headspace of developing bird embryos is most interesting as it provides not only a potential P. downsi attractant but is thus far undocumented in the literature as a by-product of embrionic avian metabolism.

References

Ahmed I, Greenwood R, Costello B, Ratcliff NM, Probert CS. 2013. An investigation of fecal volatile

organic metabolites in irritable bowel syndrome. PLoS ONE 8(3):e58204.

Allan SA, Bernier UR, Kline DL. 2006. Laboratory evaluation of avian odors for mosquito (Diptera:

Culicidae) attraction. J. Med. Entomol. 43(2): 225–231.

Alphey L. 2002. Re–engineering the sterile insect technique. Insect Biochem Mol Bio. 32(10): 1242–

1247.

Anderson JR. 2001. Larviposition by nasopharyngeal bot fly parasites of Columbian black-tailed

deer: a correction. Med Vet Entomol. 15(4):438–442.

Barclay HJ. 2005. Mathematical models for the use of sterile insects. In: Dyck VA, Hendrichs J,

Robinson, AS, editors. Principles and Practice in Area-Wide Integrated Pest Management.

Dordrecht (Netherlands): Springer. p. 147–174.

Beis A. 1985. Activities of the enzymes of ketone body metabolism in the developing chick. Comp

Biochem Physiol B. 81(3): 671–675.

Bernier UR, Allan SA, Quinn BP, Kline DL, Barnard DR, Clark GG. 2008. Volatile compounds from the

integument of white leghorn chickens (Gallus gallus domesticus L.): Candidate attractants of

ornithophilic mosquito species. J Sep Sci. 31: 1092–1099.

Bonadonna F, Miguel E, Grosbois V, Jouventin P, Bessiere J. 2007. Individual odor recognition in

birds: an endogenous olfactory signature of petrels’ feathers?. J Chem Ecol. 33(9): 1819–

1829.

Bray DP, Hamilton JGC. 2007. Host odor synergizes attraction of virgin female Lutzomyia longipalpis

(Diptera: Psychodidae). J Med Entomol. 44(5): 779–787.

Burdock GA. 2001. Fenaroli’s handbook of flavor ingredients. Fourth edition. Boca Raton (FL):CRC

Press. p. 741–742.

Burkett DA, Lee WJ, Lee KW, Kim HC, Lee HI, Lee JS, Shin EH, Wirtz RS, Cho HW, Claborn DM, et al.

2002. Late season commercial mosquito trap and host seeking activity evaluation against

mosquitoes in a malarious area of the Republic of Korea. Korean J Parasit. 40(1): 45–54.

Campagna S, Mardon J, Celerier A, Bonadonna, F. 2012. Potential semiochemical molecules from

birds: A practical and comprehensive compilation of the last 20 years studies. Chem. Senses.

37(1): 3–25.

Carey AF, Wang G, Su CY, Zwiebel LJ, Carlson JR. (2010). Odorant reception in the malaria mosquito

Anopheles gambiae. Nature. 464(7285): 66–71.

Causton CE, Peck SB, Sinclair BJ, Roque-Albedo L, Hodgson CJ, Landry B. 2006. Alien insects: threats

and implication for the conservation of the Galapagos Islands. Ann Entomol Soc Am. 99(1):

121–143.

Cha DH, Mieles AE, Lahuatte PF, Cahuana A, Lincango MP, Causton CE, Tebbich S, Cimadom A, Teale

SA. 2016. Identification and optimization of microbial attractants for Philornis downsi, an

invasive fly parasite on Galapagos birds. J Chem Ecol. 42: 1101–1111.

Cooperband MF, McElfresh JS, Millar JG, Carde RT. 2008. Attraction of female Culex quinquefasciatus

Say (Diptera: Culicidae) to odors from chicken feces. J Insect Physiol. 54(7): 1184–1192.

Crofford OB, Mallard RE, Winton RE, Rogers NL, Jackson JC, Keller U. 1977. Acetone in breath and

blood. Trans Am Clin Climatol Assoc. 88: 128–139.

Crook DJ, Khrimian A, Francese JA, Fraser I, Poland TM, Sawyer AJ, Mastro VC. 2008. Development of

a host-based semiochemical lure for trapping emerald ash borer, Agrilus planipennis

(Coleoptera: Busprestidae). Environ Entomol. 37(2): 356–365.

De Oliviera J, Druyan S, Uni Z, Ashwell C, Ferket P. 2012. Metabolic profiling of late-term turkey

embryos by microarrays. Poultry Sci. 92(4): 1011–1028.

Dougherty MJ, Guerin PM, Ward RD, Hamilton JGC. 1999. Behavioural and electrophysiological

responses of the phlebotomine sandfly Lutzomyia longipalpis (Diptera: Psychodidae) when

exposed to canid host odour kairomones. Physiol Entomol. 24(3): 251–262.

Dougherty MJ, Guerin PM, Ward RD. 1995. Identification of oviposition attractants for the sandfly

Lutzomyia longipalpis (Diptera: Psychodidae) in volatiles of faeces from vertebrates. Physiol

Entomol. 20(1): 23–32.

Douglas HD, Co JE, Jones TH, Conner WE. 2001. Heteropteran chemical repellents identiﬁed in the

citrus odor of a seabird (crested auklet: Aethia cristatella): evolutionary convergence in

chemical ecology. Naturwissenschaften. 88(8): 330–332.

Du YJ, Millar JG. 1999. Electroantennogram and oviposition bioassay responses of Culex

quinquefasciatus and Culex tarsalis (Diptera: Culicidae) to chemicals in odors from bermuda

grass infusions. J Med Entomol. 36(2): 158–166.

Dudaniec RY, Fessl B, Kleindorfer S. 2007. Interannual and interspecific variation in intensity of the

parasitic fly, Philornis downsi, in Darwin’s finches. Biol Conserv. 139(3): 325–332.

Dudaniec RY, Gardner MG, Donnellan S, Kleindorfer S. 2008. Genetic variation in the invasive avian

parasite, Philornis downsi (Diptera, Muscidae) on the Galápagos archipelago. BMC Ecol. 8(1):

13–25.

Dudaniec RY, Gardner MG, Kleindorfer S. 2010. Offspring genetic structure reveals mating and nest

infestation behavior of an invasive parasitic fly (Philornis downsi) of Galápagos birds. Biol

Invasions. 12(3):581–592.

Dudaniec RY, Kleindorfer S, Fessl B. 2006. Effects of the introduced ectoparasite Philornis downsi on

haemoglobin level and nestling survival in Darwin’s small ground finch (Geospiza

fuliginosa). Austral Ecol. 31(1): 88–94.

Dudaniec RY, Kleindorfer S. 2006. The effects of the parasitic flies Philornis (Diptera: Muscidae) on

birds. Emu. 106(1): 13–20.

Eisemann, C.H. 1988. Upwind flight by gravid Australian sheep blowfies, Lucilia cuprina

(Wiedemann)(Diptera:Calliphoridae), in response to stimuli from sheep. Bull Entomol Res.

77(2): 287–294.

El-Sayed AM. 2018. The Pherobase: Database of Pheromones and Semiochemicals [Internet]. [cited

2019 March 26]. Available from: http://www.pherobase.com.

Epsky ND, Dueben BD, Heath RR, Lauzon CR, Prokopy RJ. 1997. Attraction of Anastrepha suspense

(Diptera:Tephritidae) to volatiles from avian fecal material. Florida Entomol. 80(2): 270-277.

Fessl B, Kleindorfer S, Tebbich S. 2006a. An experimental study on the effects of an introduced

parasite in Darwin’s finches. Biol Conserv. 127(1): 55–61.

Fessl B, Sinclair BJ, Kleindorfer S. 2006b. The life-cycle of Philornis downsi (Diptera: Muscidae)

parasitizing Darwin’s finches and its impacts on nestling survival. Parasit. 133(6): 739–747.

Fessl B, Tebbich S. 2002. Philornis downsi- a recently discovered parasite on the Galapagos

archipelago- a threat for Darwin’s finches? Ibis. 144(3): 445–451.

Fessl B, Young GH, Young RP, Rodgríguez-Matamoros J, Dvorak M, Tebbich S, Fa JE. 2010. How to

save the rarest Darwin’s finch from extinction: the mangrove finch on Isabela Island. Phil

Trans R Soc B. 365(1543): 1019–1030.

Galligan TH, Kleindorfer S. 2009. Naris and beak malformation caused by the parasitic fly, Philornis

downsi (Diptera: Muscidae), in Darwin’s small ground finch, Geospiza fuliginosa

(Passeriformes: Emberizidae). Biol J Linn Soc. 98(3): 577–585.

Garner CE, Smith S, Elviss NC, Humphrey TJ, White P, Ratcliffe NM, Probert CS. 2008. Identiﬁcation

of Campylobacter infection in chickens from volatile faecal emissions. Biomarkers. 13(4):

413–421.

Gikonyo NK, Hassanali A, Njagi PGN, Gitu PM, Midiwo JP. 2002. Odor composition of preferred

(buffalo and ox) and nonpreferred (waterbuck) hosts of some savanna tsetse flies. J Chem

Ecol. 28(5): 969–981.

Grant PR, Grant BR, Petren K, Keller LF. 2005. Extinction behind our backs: the possible fate of one

of the Darwin’s finch species in Isla Floreana, Galapagos. Biol Conserv. 122(3): 499–503.

Hagelin JC, Jones IJ, Rasmussen LEL. 2003. A tangerine-scented social odour in a monogamous

seabird. Proc R Soc Lond B Biol Sci. 270(1522): 1323–1329.

Haribal M, Dhondt AA, Rosane D, Rodriguez E. 2005. Chemistry of preen gland secretions of

passerines: different pathways to same goal? Why?. Chemoecol. 15: 251–260.

Hellman W, Fierke M. 2009. 2009 Philornis downsi trapping and rearing. Puerto Ayora

(Galapagos):Charles Darwin Foundation.

Hoddle MS. 2003. Classical biological control of arthropods in the 21st century. In: Van Driesche RG,

editors. Proceedings of the International Symposium on Biological Control of Arthropods;

2002 Jan 14–18; Honolulu, Hawaii. Morgantown (WV): USDA Forest Service. p. 3–16.

Huber SK, Owen JP, Koop JAH, King MO, Grant PR, Grant BR, Clayton DH. 2010. Ecoimmunity in

Darwin’s finches: invasive parasites trigger acquired immunity in the medium ground finch

(Geospiza fortis). PLoS ONE [Internet]. [cited 5 Feb 2015]; 5(1). Available from:

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0008605#pone-

0008605-g003 ; doi: 10.1371/journal.pone.0008605

Hui YH, Chen F, Nollet LML, editors. 2010. Handbook of Fruit and Vegetable Flavors. Hoboken

(USA): John Wiley & Sons.

Inagaki H, Kiyakawa Y, Tamogami S, Watanabi H, Takeuchi Y, Mori Y. 2014. Identification of a

pheromone that increases anxiety in rats. Proc Natl Acad Sci USA. 111(52):18751–18756.

Ito Y, Yamamura K. 2005. Role of population and behavioral ecology in the sterile insect technique.

In: Dyck VA, Hendrichs J, Robinson, AS, editors. Principles and Practice in Area-Wide

Integrated Pest Management. Dordrecht (Netherlands): Springer. p. 177–205.

journal.pone.0019706 ; doi: 10.1371/journal.pone.0019706

Kleindorfer S, Dudaniec RY. 2009. Love thy neighbour? Social nesting pattern, host mass and nest

size affect ectoparasite intensity in Darwin’s tree finches. Behav Ecol Sociobol. 63(5): 731–

739.

Kleindorfer S, Dudaniec RY. 2016. Host-parasite ecology, behavior and genetics: a review of the

introduced fly parasite Philornis downsi and its Darwin’s finch hosts. BMC Zoo. 1(1). doi:

10.1186/s40850-016-0003-9

Kohlwey S, Krause ET, Baier MC, Müller C, Caspers BA. 2019. Chemical analyses reveal family-

specific nest odor profiles in zebra finches (Taeniopygia guttata): a pilot study. In: Schulte

BA, Goodwin TE, Ferkin MH, editors. Chemical Signals in Vertebrates 13. Springer Science &

Business Media B.V. p. 167–175

Koop JAH, Huber SK, Laverty SM, Clayton DH. 2011. Experimental demonstration of the fitness

consequences of an introduced parasite of Darwin’s finches. PLoS ONE [Internet]. [cited 5

Feb 2015]; 6(5). Available from: http://journals.plos.org/plosone/article?id=10.1371/

Krčmar S, Mikuška A, Radolić V. 2009. Comparison of sampling tabanids (Diptera: Tabanidae) by

four different potential attractants. J Appl Entomol. 134(7): 608–613.

Lahuatte PF, Lincango MP, Heimpel GE, Causton CE. 2016. Rearing larvae of the avian nest parasite,

Philornis downsi (Diptera:Muscidae), on chicken blood-based diets. Journal Insect Sci. 16(1):

84. 1–7.

Lawson LP, Fessl B, Vargas FH, Farrington HL, Cunninghame HF, Mueller JC, Nemeth E, Sevilla PC,

Petren K. 2017. Slow motion extinction: inbreeding, introression, and loss in the critically

endangered mangrove finch (Camarhynchus heliobates). Conserv Genet. 18(1): 159–170.

Lincango P, Causton C. (2009). Ensayos de atrayentes para la captura de la mosca parásito, Philornis

downsi (Diptera: Muscidae) en las Islas Galápagos. Puerto Ayora (Galapagos): Fundación

Charles Darwin.

Lincango P, Causton C. 2008. Crianza en cautiverio de Philornis downsi, en las islas Galapagos.

Puerto Ayora (Galapagos): Fundación Charles Darwin.

Linstrom PJ, Mallard WG, Editors. 2019. NIST Chemistry WebBook, NIST Standard Reference

Database Number 69. Gaithersburg (MD): National Institute of Standards and Technology

https://doi.org/10.18434/T4D303.

Martin-Vivaldi M, Ruiz-Rodrigues M, Soler JJ, Peralta-Sanchez JM, Mendez M, Valdivia E, Martin-

Platero AM, Martinez-Bueno M. 2009. Seasonal, sexual, and developmental differences in

hoopoe Upupa epops preen gland morphology and secretions: evidence for a role of

bacteria. J Avian Biol. 40(2): 191–205.

Metcalf RL, Metcalf RA, Rhodes AM. 1980. Cucurbitacins as kairomones for diabroticite beetles. Proc

Natl Acad Sci USA. 77(7), 3769–3772.

Mohr RM, Mullens BA, Gerry AC. 2010. Evaluation of ammonia, human sweat, and bovine blood as

attractants for the female canyon fly, Fannia conspicua (Diptera:Muscidae), in southern

California. J Vector Ecol. 36(1): 55–58.

Moran E. 2006. Nutrition of the developing embryo and hatchling. Poultry Sci. 86(5): 1043–1049.

Muth A. 2007. Control of Philornis downsi, bird parasite. Puerto Ayora (Galapagos): Charles Darwin

Foundation.

Naitoh K, Tsuda T, Nose K, Kondo T, Takasu A, Hirabayashi T. 2002. New measurement of hydrogen

gas and acetone vapor in gases emanating from human skin. Instrum Sci Technol. 30(3):

267–280.

Noble RC, Cocchi M. 1990. Lipid metabolism and the neonatal chicken. Prog Lipid Res. 29(2) 107–

140.

Norris DO. 1997. Vertebrate endocrinology. San Diego (USA): Academic Press. p. 568–569.

O’ Connor JA, Robertson J, Kleindorfer S. 2010a. Video analysis of host-parasite interactions in

nests of Darwin’s finches. Oryx. 44(4): 588–584.

O’ Connor JA, Sulloway FJ, Robertson J, Kleindorfer S. 2010b. Philornis downsi parasitism is the

primary cause of nestling mortality in the critically endangered Darwin’s medium tree finch

(Camarhynchus pauper). Biodivers Conserv. 19(3): 853–856.

Petit KE, Petit LJ, Petit DR. 1989. Fecal sac removal: Do the pattern and distance of dispersal effect

the chance of nest predation? Condor. 91(2): 479–482.

Reneerkens J, Almeida JB, Lank DB, Jukema J, Lanctot RB, Morrison RIG, Rijpstra WIC, Schamel D,

Schekkerman H, Damste JSS, et al. 2007. Parental role division predicts avian preen wax

cycles. Ibis. 149(4): 721–729.

Russell CB, Hunter FF. 2005. Attraction of Culex pipiens/restuans (Diptera: Culicidae) mosquitoes to

bird uropygial gland odors at two elevations in the Niagara region of Ontario. J Med

Entomol. 42(3): 301–305.

Sato M, Tachibana T, Furuse M. 2006. Heat production and lipid metabolism in broiler and layer

chickens during embryonic development. Comp Biochem and Physiol A. 143(3): 382–388.

Shaw CL, Rutter JE, Austin AL, Garvin MC, Whelan RJ. 2011. Volatile and semivolatile compounds in

gray catbird uropygial secretions vary with age and between breeding and wintering

grounds. J Chem Ecol. 37(4): 329–339.

Soini HA, Schrock SE, Bruce KE, Wiesler D, Ketterson ED, Novotny MV. 2007. Seasonal variation in

volatile compound proﬁles of preen gland secretions of the dark-eyed junco (Junco

hyemalis). J Chem Ecol. 33(1): 183–198.

Soini HA, Schrock SE, Bruce KE, Wiesler D, Ketterson ED, Novotny MV. 2007. Seasonal variation in

volatile compound profiles of preen gland secretions of the dark-eyed junco (Junco

hyemalis). J Chem Ecol. 33(1): 183–198.

Soini HA, Whittaker DJ, Wiesler D, Ketterson ED, Novotny MV. 2013. Chemosignaling diversity in

songbirds: Chromatographic profiling of preen oil volatiles in different species. J

Chromatogr A. 1317(22): 186–192.

Speake BK, Murray AMB, Noble RC. 1998. Transport and tranformations of yolk lipids during

development of the avian embryo. Prog Lipid Res. 37(1): 1–32.

Syed Z, Leal WS. 2009. Acute olfactory response of Culex mosquitoes to a human- and bird-derived

attractant. P Natl Acad Sci USA. 106(44): 18803-18808.

Takken W. 1991. The role of olfaction in host-seeking of mosquitoes: A review. Int J Trop Insect Sci.

12(1-2-3): 287–295.

Ten Busch MT, Milakofsky L, Hare T, Nibbio B, Epple A. 1997. Regulation of substances in allantoic

and amniotic fluid of the chicken embryo. Comp Biochem Physiol. 116A(2): 131–136.

Thakeow P, Angeli S, Weißbecker B, Schütz S. 2008. Antennal and behavioral responses of Cis boleti

to fungal odor of Trametes gibbosa. Chem Senses 33(4): 379–387.

Vale GA, Hall DR. 1985. The use of 1-octen-3-ol, acetone and carbon dioxide to improve baits for

tsetse flies, Glossina spp. (Diptera: Glossinidae). B Entomol Res. 75(2): 219–232.

Vale GA. 1980. Field studies of the responses of tsetse flies (Glossinidae) and other Diptera to

carbon dioxide, acetone and other chemicals. B Entomol Res. 70(4): 563–570.

Vleck C, Hoyt D, Vleck D. 1979. Metabolism of avian embryos: patterns in altricial and prococial

birds. Physiol Zool. 53(3): 363–377.

Whelan RJ, Levin TC, Owen JC, Garvin MC. 2010. Short chain carboxylic acids from gray catbird

(Dumetella carolinesis) uropygial secretions vary with testosterone levels and photoperiod.

Comp Biochem Physiol. B: Biochem Mol Biol. 156(3): 183–188.

Witzgall P, Kirsch P, Cork A. 2010. Sex pheromones and their impact on pest management. J Chem

Ecol. 36(1): 80–100.

Zhang JX, Zuo MX, Sun L. 2009. The volatile composition of uropygial glands contains information

about sex, individual, and species in Bengalese finches, Lonchura striata. Curr Zool. 55(5):

357–365.

APPENDIX I— Schematics of headspace collection methods

A) Dynamic headspace collection of zebra finch nests

B) Static headspace collection of zebra finch nests

C) Dynamic headspace collection of wild New York bird nests

D) Static headspace collection of zebra finch egg air-pocket headspace

APPENDIX II— Antennal preparation of Philornis downsi for GC-EAD/EAG recording

APPENDIX III— Methodology and results of GC-EAD/EAG assays

*Specimen A- wild-caught female, 37 days old; Specimen B- wild-caught female, 38 days old; Specimen C- wild-caught female, 38 day old; Specimen D- wild-caught male, 42 days old; Specimen E- wild-caught male, 42 days old; Specimen F- laboratory-reared female, 72 days old; Specimen G- laboratory-reared female, 71 days old; Specimen H- laboratory-reared female, 80 days old; Specimen I- laboratory-reared female, 80 days old; Specimen J- laboratory-reared female, 80 days old; Specimen K- laboratory-reared female, 80 days old; Specimen L- wild caught female, 25 days old; Specimen M- wild caught female, 25 days old; Specimen N- wild-caught female, 29 day old; Specimen O- wild-caught female, 29 days old. †Antennal response. ‡No antennal response. •Error.

Number P. downsi Sample Method Sample preparation of trials Specimen* 5µl headspace from 2ml vial containing 15µl 2†† EAG A acetone on glass paper 1† EAG 5µl headspace from stock bottle of acetone A 10µl headspace from 2ml vial containing 1† EAG A 15µl acetone on glass paper Carboxen/PDMS SPME fiber exposed for 5 1‡ GC-EAD B seconds in 2ml vial containing 1µl acetone Acetone Carboxen/PDMS SPME fiber exposed for 10 3‡‡‡ GC-EAD C, D, E seconds in 2mL vial containing 1µL acetone Carboxen/PDMS SPME fiber exposed for 10 1‡ GC-EAD F seconds in 2ml vial containing 1µl acetone Carboxen/PDMS SPME fiber exposed for 15 1†• GC-EAD G seconds in 4ml vial containing 1ml acetone Carboxen/PDMS SPME fiber exposed for 10 1†• GC-EAD H seconds in 4ml vial containing 1 µl acetone 2 µl of DCM:Hexane (1:1) extract from 2†† GC-EAD I nesting male zebra finch #1 2 µl of DCM:Hexane (1:1) extract from non- 1† GC-EAD J, K nesting male zebra finch #2 2 µl of DCM:Hexane (1:1) extract from 1† GC-EAD I nesting female zebra finch #3 Uropygial 2 µl of methanol extract from non-nesting 1†• GC-EAD L glands female zebra finch #4 2 µl of DCM:Hexane (1:1) extract from non- 1‡• GC-EAD M nesting male zebra finch #5 2 µl of methanol extract from nesting female 1†• GC-EAD N zebra finch #6 2 µl of DCM:Hexane (1:1) extract from 1†• GC-EAD O nesting female zebra finch #7

APPENDIX IV— Schematics of olfactometers used for behavioral assays

A) Large plexiglass arena olfactometer

B) Y-tube olfactometer view from above (left) and side (right)

APPENDIX V— Chemical compounds identified from GC-MS analysis of zebra finch nest dynamic headspace collection

Method descriptions (1–5) are in Table 1. Compound ID was obtained through ion match to NIST 08 mass spectral library and with the calculated retention indices reported here. a- Compound ID verified by injection of corresponding reference standard. b- Compound ID is supported by match of retention index to data in the Pherobase chemical database or NIST Chemistry WebBook. Analyses for these compounds were performed on a DB-WAX column.

Linear Retention Methods % of Identification Samples Sample types Index detected samples 1076 Hexanal ab 1a, 1b, 2, 3, 4, 5 36 out of 46 78% Nest building, nesting, incubating, chicks 1180 Heptanal b 1a, 3, 4 15 out of 18 83% Nest building, nesting, incubating, chicks 1283 Octanal b 1b, 2 11 out of 21 52% Nesting, incubating, chicks 1322 E -2-heptenal b 2 5 out of 6 83% Nest building, incubating, chicks 1351 1-hexanol b 3, 5 3 out of 9 33% Incubating 1384 Nonanal ab 1a, 1b, 2, 3, 4, 5 37 out of 46 80% Nest building, nesting, incubating, chicks 1561 1-octanol ab 1b, 2, 5 17 out of 28 61% Nest building, nesting, incubating, chicks 1428 E -2-octenal b 1b, 2 10 out of 21 48% Nest building, nesting, incubating, chicks 1448 1-octen-3-ol b 2,3,5 10 out of 15 67% Nest building, incubating, chicks 1494 Decanal b 1a, 1b, 4 18 out of 31 58% Nesting, incubating, chicks 1529 E -2-nonenal b 1b, 2 16 out of 21 76% Nest building, nesting, incubating, chicks 1998 Hexanoic acid 5 4 out of 6 67% Incubating, chicks

APPENDIX VI— Chemical compounds identified from GC-MS analysis of zebra finch nest static headspace collection

Compound ID was obtained through ion match to NIST 08 mass spectral library. a- Compound ID was verified with the corresponding reference standard. Analyses for these compounds were performed on a DB-WAX column.

Retention Compound Compound ID % of samples Sample types Time (mins) A 1.43 Acetone a 100% Nesting, Incubating, Chicks B 2.32 Ethanol 100% Nesting, Incubating, Chicks C 4.15 Hexanal a 100% Nesting, Incubating, Chicks D 6.60 1-pentanol 100% Incubating, Chicks E 7.11 3-hydroxy 2-butanone 67% Incubating, Chicks F 8.77 1-heptene 67% Incubating, Chicks G 11.55 Acetic acid 100% Nesting, Incubating, Chicks

APPENDIX VII— Chemical compounds identified from GC-MS analysis of New York wild bird nest headspace collection Method descriptions are in Table 2. Compound ID was obtained through ion match to NIST 08 mass spectral library, and with calculated retention indices reported. a- Compound ID was verified with the corresponding reference standard. b- Compound ID supported by match of retention index with data in the Pherobase chemical database or NIST Chemistry WebBook. *Retention index not available in literature. Analyses for all compounds were performed on a DB-WAX column. Compound Retention Time (mins) Linear Retention Index Identification Methods detected A 4.52 1008 Tricycloheptane * 3 B 4.62 1005 Decane b 2 C 4.71 1010 Unknown 2 D 4.77 1021 α-pinene ab 1,2,3 E 5.55 1062 Camphene b 1,3 F 6.45 1104 Undecane b 2,3 G 6.50 1105 β-pinene b 2 H 6.70 1118 β-phellandrene 2,3 I 6.86 1124 Unknown 3 J 7.39 1144 1-undecanol * 2 K 7.47 1150 3-carene b 2,3 L 7.71 1161 β-myrcene b 2,3 M 7.85 1167 4-methyl-3-heptanone * 2 N 8.35 1185 Heptanal b 2, 3 O 8.52 1192 Limonene b 1,2,3 P 8.75 1204 β-phellandrene b 3 Q 8.79 1206 Eucalyptol * 3 R 8.91 1210 Unknown 2,3 S 10.05 1254 Styrene b 1 T 10.30 1265 Benzene * 3 U 10.55 1273 Unknown 1 V 10.60 1276 (+)-4-carene 3 W 10.80 1283 Cyclohexene * 2 X 10.94 1289 Unknown 1 Y 11.62 1316 4-methylene-5-hexenal * 3 Z 12.17 1337 5-hepten-2-one * 1,2 AA 12.66 1356 1-hexanol b 2 BB 13.10 1373 Dimethyl trisulfide b 2 CC 13.50 1388 Nonanal ab 2,3 DD 13.55 1390 Bicycloheptan-2-one * 1 EE 14.21 1416 Hexyl ester butanoic acid 2,3 FF 14.50 1428 E -2-octenal b 2 GG 14.67 1435 Unknown 1 HH 14.91 1444 Unknown 2 II 15.44 1464 Dodecadien-2-ol * 2 JJ 15.45 1466 2-heptanone 1,2 KK 15.81 1480 3-cyclohepten-1-one * 2 LL 15.99 1488 Fumaric acid * 1,2 MM 16.10 1492 2-ethyl-1-hexanol b 1,2 NN 16.21 1496 Spiro-octan-5-one * 2 OO 16.25 1498 Dicyclobutylidene oxide * 2 PP 16.29 1500 Spiro-octan-5-one * 1 QQ 16.50 1508 Camphor * 3 RR 16.51 1509 Bicycloheptan-2-one * 1 SS 16.74 1518 Benzaldehyde b 2 TT 17.15 1535 Unknown 3 UU 17.61 1554 1-octanol ab 1,3 VV 18.35 1586 Bicycloheptan-2-ol * 1,2 WW 18.65 1598 Unknown 1,2 XX 20.96 1697 3-cyclohexene-1-methanol * 1,2 YY 21.09 1703 Bicycloheptan-2-ol * 1 ZZ 21.29 1712 Cyclopenten-1-ol * 2 AAA 23.39 1809 Unknown 2 BBB 23.39 1808 9-decen-1-yl ester hexanoic acid * 3 CCC 24.21 1847 Benzenemethanol b 1 DDD 25.10 1888 3-decen-5-one * 3 EEE 25.26 1896 Pyrazole * 2 FFF 26.17 1941 Unknown 3 GGG 28.10 2039 Dodectrien-3-ol * 2 HHH 31.90 2243 Dimethoxybenzaldehyde * 2 III 32.58 2282 Diethyltoluamide * 2

APPENDIX VIII— Chemical compounds identified from GC-MS analysis of zebra finch uropygial gland secretions

Sample types refer to gender and breeding status of birds that yielded a given compound, where F = nonbreeding female, M=nonbreeding male, FB= breeding female, MB = breeding male. Calculated retention indices are reported. Analysis for these compounds was performed on a DB-WAX column. Retention Time Linear Retention Methods % of Compound Samples Sample types (mins) Index detected samples A 18.166 1577 2 3 out of 10 30% F, M, MB B 18.734 1600 2 3 out of 9 33% F, FB, MB C 19.981 1651 2 2 out of 5 40% F, FB D 20.274 1667 2 2 out of 5 40% F, FB E 22.476 1766 2 8 out of 14 57% F, M, FB, MB F 23.779 1826 2 2 out of 5 40% M, FB G 24.513 1860 2 9 out of 14 64% F, M, FB, MB H 24.749 1871 2 8 out of 14 57% F, M, FB, MB I 26.609 1964 2 5 out of 13 38% M, FB, MB J 26.878 1977 2 3 out of 9 33% F, FB, MB K 26.940 1980 2 1 out of 5 20% M L 27.127 1990 2 5 out of 13 38% M, FB, MB M 28.345 2052 2 1 out of 5 20% M N 29.822 2121 2 3 out of 9 33% F, FB, MB O 29.823 2130 1 1 out of 1 100% M P 30.930 2188 2 1 out of 4 25% MB Q 32.554 2282 2 5 out of 14 36% F, FB, M, MB R 32.950 2303 2 1 out of 5 20% M S 33.892 2359 1 1 out of 1 100% M T 34.180 2374 2 9 out of 13 69% M, FB, MB U 34.218 2378 2 1 out of 5 20% M V 35.134 2433 2 1 out of 5 20% M W 35.570 2458 2 13 out of 15 87% F, M, FB, MB X 35.590 2461 1 1 out of 1 100% M Y 37.025 2547 2 11 out of 14 79% F, M, FB, MB Z 37.032 2550 1 1 out of 1 100% M AA 37.087 2554 2 1 out of 5 20% M BB 37.245 2561 2 14 out of 14 100% F, M, FB, MB CC 37.246 2564 1 2 out of 2 40% M , MB DD 37.580 2584 1 1 out of 1 100% M EE 37.590 2582 2 14 out of 14 100% F, M, FB, MB FF 38.056 2611 2 1 out of 4 25% FB GG 38.322 2629 2 11 out of 14 79% F, M, FB, MB HH 38.331 2632 2 1 out of 5 20% M

II 38.345 2633 1 1 out of 1 100% M JJ 38.639 2650 2 12 out of 14 86% F, M, FB, MB KK 38.640 2652 1 2 out of 2 100% M , MB LL 38.642 2652 2 2 out of 5 40% M MM 38.819 2664 2 2 out of 5 40% M NN 38.835 39 1 2 out of 2 100% M, MB OO 38.876 2665 2 12 out of 14 86% F, M, FB, MB PP 38.965 2671 2 6 out of 13 46% F, M, FB, MB QQ 39.153 2684 2 12 out of 14 86% F, M, FB, MB RR 39.180 2687 2 2 out of 5 40% M SS 39.773 2725 2 11 out of 14 79% F, M, FB, MB TT 39.782 2727 1 1 out of 1 100% M UU 39.835 2731 2 1 out of 5 20% M VV 40.190 2754 2 14 out of 14 100% F, M, FB, MB WW 40.190 2755 1 2 out of 2 100% M , MB XX 40.365 2767 2 2 out of 5 40% M YY 40.370 2767 1 2 out of 2 100% M , MB ZZ 40.390 2768 2 12 out of 14 86% F, M, FB, MB AAA 40.708 2790 2 11 out of 14 79% F, M, FB, MB BBB 40.730 2791 2 2 out of 5 40% M CCC 40.957 - 2 2 out of 5 40% M DDD 40.995 - 2 5 out of 8 63% FB, MB EEE 41.027 - 2 11 out of 14 79% F, M, FB, MB FFF 41.037 - 1 1 out of 1 100% M GGG 41.037 - 2 2 out of 5 40% M HHH 41.215 - 2 11 out of 14 79% F, M, FB, MB III 41.301 - 2 2 out of 5 40% M JJJ 41.413 - 2 1 out of 5 20% M KKK 41.700 - 2 2 out of 5 40% M LLL 41.715 - 1 2 out of 2 100% M , MB MMM 41.721 - 2 12 out of 14 86% F, M, FB, MB NNN 41.903 - 2 1 out of 5 20% M OOO 42.082 - 2 1 out of 5 20% M PPP 42.106 - 2 4 out of 10 40% F, M, FB, MB QQQ 42.319 - 2 1 out of 5 20% M RRR 42.335 - 2 1 out of 4 25% MB SSS 42.626 - 2 1 out of 5 20% M TTT 42.636 - 2 10 out of 14 71% F, M, FB, MB UUU 42.949 - 2 8 out of 13 62% M, FB, MB VVV 42.953 - 2 1 out of 5 20% M WWW 43.473 - 2 6 out of 8 75% M, FB, MB XXX 43.474 - 2 1 out of 5 20% M YYY 43.696 - 2 1 out of 5 20% M

ZZZ 44.586 - 2 10 out of 14 71% F, M, FB, MB AAAA 44.598 - 2 1 out of 5 20% M BBBB 46.824 - 2 1 out of 5 20% M CCCC 50.903 - 1 1 out of 1 100% M DDDD 51.998 - 1 1 out of 1 100% M EEEE 55.987 - 1 1 out of 1 100% M FFFF 56.086 - 1 1 out of 1 100% M GGGG 56.716 - 1 1 out of 1 100% M HHHH 57.502 - 1 1 out of 1 100% M IIII 45.301-56.838 - 2 1 out of 5 20% M JJJJ 45.332-56.774 - 2 12 out of 14 86% F, M, FB, MB

Kristin M. Doherty [email protected] 347-266-6220 930 Hollywood Ave Bronx, NY 10465 Education State University of New York College of Environmental Science and Forestry, Syracuse NY Bachelor of Science with Honors, Environmental Biology Graduated May 2012, Magna Cum Laude Cumulative GPA of 3.727 Dean’s List: Spring 2009, Fall 2009 President’s List: Spring 2010, Summer 2010, Fall 2010, Spring 2011, Fall 2011, Spring 2012 Alpha Xi Sigma Honor Society: Spring 2010, Fall 2010, Spring 2011, Fall 2011, Spring 2012

State University of New York College of Environmental Science and Forestry, Syracuse NY Master of Science, Entomology Expected Graduation May 2019 Cumulative GPA of 4.000

Previous Work Experience NYC Department of Environmental Protection, Queens NY; May-August 2009 College Intern: Processed investigator field reports for air and noise pollution violations. Sent corrective correspondence to those found in violation.

SUNY Environmental Science and Forestry, Syracuse NY; May 2011-August 2013 Research Assistant: Assisted PhD, post-doctorate, and major professors with various research projects. Primarily responsible for setting up and executing field and laboratory experiments, as well as creating presentations for conferences and workshops. Also included general lab maintenance.

SUNY Environmental Science and Forestry, Syracuse NY; August 2013-May 2015 Teaching Assistant: Assisted professors with planning course curricula. Independently taught laboratory portions for courses, including leading experiments and field trips.

Nick Bell Lawn and Landscape, Brockport NY; Spring and Summers of 2012-2015 Landscaper: Performed standard duties such as weeding, raking, and mulching as well as laying patio pavers and decorative walls. Primarily assisted in garden design and installation.

Blue Moon Mexican Café, Bronxville NY; August 2015-August 2016 Hostess: Responsible for seating the dining room, taking pick-up/delivery orders, and corresponding with customers on miscellaneous issues.

Nicosia’s Bronx Bread Company, Bronx NY; May 2016-Ongoing Office Manager: Managed customer accounts including incoming wholesale bread orders, accounts receivable, and accounts payable. Also managed daytime production, miscellaneous supply inventory, and employee records.

Extracurricular Experience Undergraduate assistant at SUNY ESF; August 2010-May 2011 Alpha Xi Sigma Honor Society Vice President; September 2010-December 2011 Thornden Park Cleanup; August 2008- December 2011 Rescue Missions ‘Thrifty Shopper’ Thrift Store; November 2008- May 2012

Skills Experienced in Microsoft Office (Word, Excel, Publisher, and Powerpoint), NoWait restaurant application, Flexibake management software, Statistical Analysis Software, ArcGIS software and Global Positioning Systems equipment. In possession of a valid NYS driver’s license. 61

Kristin M. Doherty [email protected] 347-266-6220 930 Hollywood Ave Bronx, NY 10465

References Anthony Costa Blue Moon Mexican Café, Manager [email protected] (917) 795-7219

Alex Toddman Nicosia’s Bronx Bread Company, Manager (646) 410-7750

Stephen Teale SUNY Environmental Science and Forestry, Major Professor [email protected] (315) 470-6758