Citation
Influence of Southeastern Fox Squirrel (Sciurus niger) Pelage Coloration on Ectoparasite Richness and Abundance

Material Information

Title:
Influence of Southeastern Fox Squirrel (Sciurus niger) Pelage Coloration on Ectoparasite Richness and Abundance
Creator:
Bäck, Rowan Emily
Publication Date:
Language:
English

Subjects

Subjects / Keywords:
Colors ( jstor )
Ectoparasites ( jstor )
Fleas ( jstor )
Lice ( jstor )
Mammals ( jstor )
Modeling ( jstor )
Parametric models ( jstor )
Species ( jstor )
Squirrels ( jstor )
Ticks ( jstor )
Genre:
Undergraduate Honors Thesis

Notes

Abstract:
Within the southeastern United States, the fox squirrel (Sciurus niger) exhibits great variation in its pelage coloration. Current research shows that there is a relationship between ectoparasites and pelage coloration in some mammals, but it is not known if this occurs in southeastern fox squirrels, which are north America’s most variably colored mammal. To assess the effect of pelage coloration on the ectoparasite richness and abundance of fox squirrels in Florida, I collected and identified ectoparasites from 25 fox squirrels and I quantified pelage coloration values of hue, saturation, and lightness on the dorsal, ventral, and tail surfaces. Louse were found to have the greatest relationship with pelage coloration, largely because they comprised the majority of the ectoparasites. Dorsal Coloration was the top model for explaining the number of lice on fox squirrels. This model included Dorsal Coloration Hue (b: -0.008, SE: 0.001, 95% CI: -0.011–-0.007), Dorsal Coloration Saturation (b: 0.028, SE: 0.008, 95% CI: 0.011–0.045), and Dorsal Coloration Lightness (b: -0.044, SE: 0.006, 95% CI: -0.057–-0.032). These results concur with the higher prevalence of louse on the dorsum of fox squirrels. Red, yellow, orange, and purple attracted lice, while green and blue repelled them. Since this study yielded low variability in ectoparasite species, it is recommended that live, trap-caught fox squirrels be used to ensure ectoparasite diversity. ( en )
General Note:
Awarded Bachelor of Science, magna cum laude, on May 2, 2017. Major(s): Zoology
General Note:
College or School: College of Liberal Arts and Sciences
General Note:
Advisor: Robert Alan McCleery. Advisor Deptarment or School: Wildlife Ecology and Conservation

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Rowan Emily Bäck. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

(For Office Use Only) Honors Thesis Submission Form Major: ______ Designation : ____ Graduation Term: __ Name:Rowan Back UFID:3318-3480 Additional Authors: Email:rowane @ ufl.edu Major:Zoology Advisor Name:Dr. Robert A McCleery Advisor Email:ramcc\eery @ ufl.edu Advisor Department:Wildlife Ecology and Conservation Thesis Title: Influence of Southeastern Fox Squirrel (Sciurus niger) Pelage Coloration on Ectoparasite Richness and Abundance Abstract (200 words max): Within the southeastern United States the fox squirrel (Sciurus niger) exhibits great variation in its p e la ge colo ra tion. Current research shows that there is a relationship between ectoparasite s and pel a ge co l o ration in some mammals, but it is not known if this occurs in southeastern fox squirrels whic h a re north America's most variably colored mammal. To assess the effect of pelage c oloration on the ectoparasite richness and abundance of fox squirrels in Florida I collected and identified ectoparasites from 25 fox squirrels and I quantified pelage coloration values of hue, saturation, and lightness on the dorsal, ventral, and tail surfaces. Louse were found to have the greatest relationship with pelage coloration, largely because they comprised the majority of the ectoparasites. Dorsal Coloration was the top model for explaining the number of lice on fox squirrels. This model included Dorsal Coloration Hue ([3: -0.008, SE: 0.001, 95% CI : -0 011--0.007), Dorsal Coloration Saturation ([3: 0.028, SE: 0.008, 95% CI: om 1-0.045), and Dorsal Coloration Lightness ([3: -0.044, SE: 0.006, 95% CI: 0.057-0.032). These results concur with the higher prevalence oflouse on the dorsum of fox squirrels Red, yellow, orange, and purple attracted lice, while green and blue repelled them. Since this study yielded low variability in ectoparasite species, it is recommended that live, trapcaught fox squirrels be used to ensure ectoparasite diversity.

PAGE 2

Student Signature/Date fp"J; ticIp'lJ.lQLlJ. aD!'/Thesis Advisor Signature/Date (-Z.: __ ') -2.. t:J Departmental Honors Coordinator Signature __ __ __ Pleale Indicate your preference for public acce to ),our Ihe.l. b)' InlUallnllhe approprlale .tatemenl below: I grant permission to the University oJ Florida to lI.ttthe title and ahJtract oJ thlJ the.dJ In a publicly accessible database __ I do not grant permission to the University oJ Florida to list the title and abJtract oJ thlJ thesiJ publicly If you wish to make the entire theals publicly available, you mUll also complete the Interact DIstribution Permissions Form, available at http://digital.uflib ufl.edu/procedures/copvrightlGrantofPermissions.doc If you do not Indude this form, your thesis wiU be archived but wiU not be viewable onUne.



PAGE 1

Influence of Southeastern Fox Squirrel (Sciurus niger) Pelage Coloration on Ectoparasite Richness and Abundance Rowan Emily Bck rowane@ufl.edu University of Florida Department of Biology Faculty Advisor: Dr. Robert A. McCleery CLAS Latin Honors

PAGE 2

ABSTRACT Within the southeastern United States, the fox squirrel ( Sciurus niger) exhibits great variation in its pelage coloration. Current research shows that there is a relationship between ectoparasites and pelage coloration in some mammals, but it is not known if this occurs in southeastern fox squirrels, which are north Americas most variably colored mammal. To assess the effect of pelage coloration on the ectoparasite richness and abundance of fox squirre ls in Florida, I collected and identified ectoparasites from 25 fox squirrels and I quantified pelage coloration values of hue, saturation, and lightness on the dorsal, ventral, and tail surfaces. Louse were found to have the greatest relationship with pel age coloration, largely because they comprised the majority of the ectoparasites. Dorsal Coloration was the top model for explaining the number of lice on fox squirrels. This model included Dorsal Coloration Hue (!: -0.008, SE: 0.001, 95% CI: -0.011-0.007), Dorsal Coloration Saturation (!: 0.028, SE: 0.008, 95% CI: 0.011!0.045), and Dorsal Coloration Lightness (!: -0.044, SE: 0.006, 95% CI: -0.057-0.032). These results concur with the higher prevalence of louse on the dorsum of fox squirrels. Red, yellow, orange, and purple attracted lice, while green and blue repelled them. Since this study yielded low variability in ectoparasite species, it is recommended that live, trap-caught fox squirrels be used to ensure ectoparasite diversity. INTRODUCTION The fox squirrel (Sciurus niger) is a large tree squirrel that occurs naturally east of the Rocky Mountains (Hall, 1981; Koprowski, 1994). In the southeastern United States, 6 subspecies of fox squirrel (S. n. vulpinus, S. n. niger, S. n. shermani, S. n. bachmani, S. n. avicennia, and S. n. cinereus) inhabit the coastal plain into southern Virginia and west into Alabama They are

PAGE 3

collectively known as southeastern fox squirrels and are well-known for having a variety of pelage colorations (Baumgartner, 1943; Moore, 1956; Loeb and Moncrief, 1993). Four subspecies of southeastern fox squirrels are listed as occurring in FloridaS. n. niger, S. n. bachmani, S. n. shermani, and S. n. avicennia (Moore 1956). Currently, Florida considers the Shermans fox squirrel (S. n. shermani) to be a species of special concern and the Big Cypress fox squirrel (S. n. avicennia) to be threatened (Florida Fish and Wildlife Conservation Commission [FWC] 2016). It is generally recognized that south eastern fox squirrels have three prominent color phasesred, gray, or blackhowever, additional coloration variations have also been observed, such as white, tan, and agouti (Loeb and Moncrief, 1993; Moore, 1956) (Figure 1). Shermans fox squirrel typically has a white nose and ears with a black dorsal portion of the head and variable coloration on the remainder of the pelage, usually falling within all dark, all tan, or dark/tan mixed (Humphrey, 1992; Kiltie, 1992). Big Cypress fox squirrels most common color phase is buff with a white nose and eartips, a black crown, and an agouti or black -agouti dorsum, but they can vary between black and tan (Ashton and Humphrey, 1992; Moore, 1956). Southern fox squirrels (S. n. niger) commonly occur in the gray-white color phase as well as the black color phase (Moore 1956). Certain coat colorations have been shown to attract ectoparasites and animals can develop certain coloration polymorphisms to deter ectoparasites (Hubbard, 2010; Jacquin et al, 2013; Reissmann and Ludwig, 2013; Payne, 2017). For example, black horses from Cameroon attract more Amblyomma ticks than those that were brown, white, or gray (Payne, 2017). In birds, dark colored, feral pigeons had more ectoparasites when inhabiting moderately urbanized environments than their paler counterparts (Jacquin et al, 2013). The clear link between an

PAGE 4

animals coloration and ectoparasites suggests there may be associations with an immunological response, that in turn, can influence the birds fitness (Jacquin et al, 2013; Reissmann and Ludwig, 2013). Thus, pelage coloration may have evolved over time as a result of less fit morphs succumbing to the effects of having a high burden of ectoparasites (Jacquin et al, 2013). Several species of ticks, mites, fleas, and sucking lice have been documented as occurring on fox squirrels in Florida (Forrester, 1992; Coyner et al, 1996). Six species of ticks were identified: I. texanus, I. scapularis, A. americanum, A. maculatum, A. tuberculatum, and D. variabilis (Rogers, 1953; Moore, 1957; Forrester, 1992; Coyner et al, 1996). The two species of fleas found include O. howardii and H. glacialis affinis (Fuller, 1943; Moore, 1957; Layne, 1971; Forrester, 1992; Coyner et al, 1996). The three species of sucking lice identified were H. sciuricola, N. sciurinus, and Enderlinus longiceps (Moore, 1957; Forrester, 1992; Coyner et al, 1996). Also found were six species of mites (E. alfreddugesi, A. megaventralis, Listrophorus sp., A. casalis, N. whartoni, and E. diversa) and one species of botfly larvae (Cuterebra sp.) (Moore, 1957; Forrester, 1992; Coyner et al, 1996). Presently, there are no studies that explore if pelage coloration variation in fox squirrels is related to ectoparasite load. This study can be useful in predicting which colors attract ectoparasites, and thus, the risk of an animal to certain arthropod -vectored diseases as well as the fitness of an animal. Since fox squirrels vary significantly in their pelage colorations, they are a good model to use in this study (Moore, 1956). Therefore, my objective was to determine if the ectoparasite burden of southeastern fox squirrel subspecies of Florida was related to their pelage coloration. Specifically, I quantified the coloration of pelages on the dorsal, ventral, and tail surfaces of fox squirrels and related them to the measure of ectoparasites, including the total number of ectoparasites and the number of ectoparasite species.

PAGE 5

METHODS I received road-killed fox squirrels collected throughout Florida from 20122015. I stored squirrels were in a freezer and then transferred them to a refrigerator before necropsy. Fox squirrels came with locational coordinates and I plotted them spatially using ArcGIS 10.4 (ESRI, Redlands, California). Ectoparasite Collection and Identification I used a flea comb and a toothbrush to run through the fur to collect any ectoparasites hanging loosely on the hairsthis was done dorsally, ventrally, through the tail, behind the ears, and around the anal region (Aspinall, 2014). I conducted this process over white legal-sized paper to collect any ectoparasites that may have fallen off the squirrel or were not in the teeth of the comb or toothbrush. The comb-through procedure was repeated several times to ensure majority of the ectoparasites had been collected. I then viewed squirrels under a dissecting microscope to find any ectoparasites that remained. I used forceps to pluck them from the skin, if any were found. Each type of ectoparasites (fleas, lice, ticks) was capped at 50 individuals that could be collected. I stored the collected ectoparasites in plastic vials filled with 70% isopropanol until mounting. I utilized Hoyers medium to slide-mount representative lice and fleas from each squirrel, while ticks and the rest of the lice and fleas remained in 70% isopropanol (Rondon and Corp, N.D.). I slide-mounted lice on their dorsum with their appendages spread apart, while I mounted fleas laterally with their legs separated (Palma, 1978; Brigham Young University, N. D.). I viewed lice and fleas under a light microscope, using the 4X and 10X objective lenses. I utilized dichotomous keys from the CDC to identify the ectoparasites (Stojanovich and Pratt,

PAGE 6

1961; Pratt, 1966; Pratt and Stark, 1973). To identify lice, I relied on the length of the claws and legs, number of antennal segments, presence/absence of sclerotized plates on the abdomen, abdominal plate articulations, and spine-like seta at the posterior-apical angle on the first antennal segment (Stojanovich and Pratt, 1961). Flea identification depended on the presence/absence of the genal, pronotal, and abdominal combs, length of labial palps, and number of plantar bristles (Pratt and Stark, 1973). The ticks were first categorized on the basis of being hard or soft. The color (or ornateness) was then examined along with other characteristics such as palp length, terminal spurs, and presence/absence of eyes (Pratt, 1966). I identified each ectoparasite to species level. Pelt Collection and Photographing I flensed pelts from the squirrel carcasses and sent them to the Florida Museum of Natural History to be vouchered for the collections. Borax was used to cure the pelts after the remaining fat was stripped from the skin and then pelts hung on a drying rack for 1-2 weeks to further preserve the skin. I photographed pelts using a Nikon D5100 DSLR camera in a 30 x 30 x 30 light box with two 100W LED lights placed on either side to illuminate the inside. A white background was used to ensure the pelts were distinct in the image. Pelage Analysis Using Digital Color Meter (Apple Inc., Cupertino, California), I measured RGB (red, green, blue) values of the dorsal, ventral, and tail surfaces in each squirrels photograph. By measuring RGB values, I could obtain the exact numerical color composition of a squirrels pelage in terms of the colors red, green, and blue (Lupton and Phillips, 2014). I isolate d the specific amount of each of respective color to quantitatively determine ectoparasites preferences. I selected three points to read RGBs on each surface that were equally spaced apart

PAGE 7

and spanned the entire section. For example, on the ventral surface, the three points selected included posterior to the axillary region, anterior to the groin, and one in th e center of the two previous points. The three individual RGB values obtained were averaged together to create an overall RGB value for that section. RGB values only exist when evaluating pixels on a computer monitor (Lupton and Phillips, 2014), so I converted the RGB values to Hue, Saturation, Lightness (HSL) values (hereafter, pelage color covariates). Hue is the actual color and has a range of 0 to 360 (or degrees on the color wheel) (Figure 2). Zero or 360 is red, 120 is green, and 240 is blue. Saturation denotes the amount of gray in a color as a percentage value; a value of 0% indicates mostly gray while 100% is white (Figure 3, top). Lightness is the amount of white or black that is mixed in the color; 0% is dark and 100% is light (Figure 3, bottom). Statistical Analysis I examined the relationships between my ectoparasite metrics (the total number of fleas, ticks, louse, ectoparasite species, and ectoparasite individuals). Then, I assessed the influence of color covariates (HSL values as continuous predictors) on ectoparasite metrics using generalized linear models and a Poisson distribution in R (R Core Team, 2014). I conducted a separate analysis for each ectoparasite response variable. I evaluated single and multivariable models using the color of each body segment (e.g., Dorsal Hue + Dorsal Saturation + Dorsal Lightness) or for all body segments (e.g., Dorsal Hue + Ventral Hue + Tail Hue) as predictors. I evaluated my candidate models using Akaike Information Criterion corrected for small sample size (AICc), where the level of importance was assessed by model weights (Burnham and Anderson, 2002). I determined the covariates predictive importance by inspecting the conditional beta coefficient (!) estimates and their 95% confidence intervals (CI), wit h significance defined as CIs for a variable that did not overlap zero.

PAGE 8

RESULTS I processed 25 fox squirrels (16 males, 9 females) to quantify and identify ectoparasite metrics (Table 1). These individuals represented S. n. niger and S. n. shermani subspecies in the Florida panhandle and the central part of the peninsula (Figure 3). I dropped the number of ectoparasites from my analysis because of a high Pearson's correlation coefficient with lice (|0.997|), a result of most ectoparasites in my stud y being lice and overall few ticks and fleas recovered from fox squirrel carcasses My top model for ectoparasite richness was the null model, indicating pelage color ation variables were not influential (Table 2). My top model for assessing the number of ticks on fox squirrels had one variable, Ventral Coloration Saturation (!: 0.180, SE: 0.079, 95% CI: 0.028!0.360) (Table 3); however, the significance of this variable, and its ranking above the null model is may be spurious, because only 2 of the 25 individuals had ticks (1 tick and 2 ticks, respectively). All other models had beta coefficients with 95% confidence intervals that included zero. The squirrel with 1 tick had a Ventral Coloration Saturation of 13.2% (more gray), while the squirrel with 2 ticks had a value of 36.4% (more pigmented in color, but still in the gray range). For fleas, the best model included the combination of Ventral Coloration Hue + Saturation + Lightness; however, the top model had more parameters than could be estim ated from the data making this ranking misleading. I dropped the model and reran the analysis, which resulted in the models of Ventral Coloration Hue, Coloration Saturation, and Lightness all as the top three models, respectively (Table 4). Ventral Coloration Hue (!: -0.079, SE: 0.016, 95% CI: -0.115"0.048), Coloration Saturation (!: -0.346, SE: 0.101, 95% CI: -0.587-0.178), and Lightness (!: -0.197, SE: 0.051, 95% CI: -0.317-0.107) all negatively influenced the number of fleas on fox squirrels. The squirrel with the highest number of fleas (7) had the lowest Ventral

PAGE 9

Coloration Hue, Coloration Saturation, and Coloration Lightness (26, 9.1, and 50.4, respectively) compared to the other four squirrels with fleas, meaning that its coloration was mo re red with gray undertones and was mid-level in brightness. For the other four squirrels with only 1 flea each, their Ventral Coloration Hue ranged from 62 -69, Ventral Coloration Saturation was between 12.1-18.2%, and Ventral Coloration Lightness ranged from 58-65.5%. These squirrels were more yellow-green with less gray undertones, and brighter in color. My top model for assessing the number of lice on fox squirrels was the Tail Coloration (Tail Hue + Tail Saturation + Tail Brightness). Although the Tail had 100% support over all other models, no lice were actually found on the tail during ectoparasite collection. Therefore, I reran the analysis after dropping the Tail model. The top model for the number of lice was Dorsal Coloration (a combination of Hue, Coloration Saturation, and Lightness) (Table 5). Dorsal Coloration Hue (!: -0.008, SE: 0.001, 95% CI: -0.011-0.007), and Dorsal Coloration Lightness (!: -0.044, SE: 0.006, 95% CI: -0.057-0.032) negatively influenced the number of lice on fox squirrels, whereas Dorsal Coloration Saturation (!: 0.028, SE: 0.008, 95% CI: 0.011!0.045) was a positive predictor. I found that lice parasitized squirrels within the Dorsal Coloration Hue ranges of 4-75 and 270-300, meaning that they preferred squirrels with more red, orange, yellow, and purple coloration. In relation to Dorsal Coloration Saturation, it appeared that lice generally favored squirrel pelages with more gray undertones, or within the 2 -7% range on the saturation scale. For Dorsal Coloration Lightness, I noticed there was a large cluster of lice in the range of 38-56%, indicating that they preferred pelages with mid -level brightness of coloration. Hue was also a competing model as it was only 1.5 AIC values from Dorsal Coloration, Dorsal Coloration Hue had a relatively small effect ( !: -0.004, SE; 0.001, 95% CI:

PAGE 10

0.007-0.002) and Ventral Coloration Hue was not significant ( !: 0.010, SE: 0.009, 95% CI: 0.0070.028). DISCUSSION Overall, my findings suggest that color covariates did not influence ectoparasite metrics equally across different ectoparasite types. This is likely due to the ectoparasites collected primarily consisting of louse species of the 25 squirrels processed, 21 had lice which accounted for 96.2% of all ectoparasites collected. The lack of diversity in the dataset was evident with the overall lack of variability in ectoparasite richness between fox squirrel individuals (Table 1); however, this was likely influenced by the small sample size. Ticks and fleas comprised only 3.8% of all ectoparasites collected in this study. I found that Ventral Coloration Saturation was the top model explaining tick prevalence, which concurs with what we know about tick behavior. Specifically, ticks typically engage in a behavior known as questing, where they climb onto weeds, grasses, and bushes to wait for hosts to latch onto while raising their first pair of legs (Mullen and Durden, 2009; CDC, 2015). Therefore, the ventral surface is the most likely surface to be seen by ticks while questing. Some species of ticks are able to see shadows and use this to locate hosts (Centers for Disease Control and Prevention [CDC], 2015). With the Ventral Coloration Saturation being more gray, t he ticks may have viewed the squirrels ventral surface as a large shadow and used this for host detection (Mullen and Durden, 2009; Centers for Disease Control and Prevention [CDC], 2015). Additionally, ticks, especially Argasid ticks, are intermittent feeders and, although they are considered obligate parasites, they spend 90% of their lifecycle off the host (Mullen and Durden, 2009; Mathison and Pritt, 2014); this likely explains the low abundance of ticks found on the fox squirrel carcasses. Perhaps a greater influence of the low tick abundance on the fox squirrels is a

PAGE 11

result of their attraction to a heat source (Mehlhorn, 2008; Mullen and Durden, 2009). The ticks likely vacated the road-killed squirrel carcass due to its loss in body heat (Mullen and D urden, 2009). The reliability in the top models with flea abundance was also uncertain, as fleas would have similarly abandoned the host upon its carcass dropping in temperature, since they are also notorious for being attracted to a hosts body warmth (Marquardt, 2004; Mullen and Durden, 2009). Fleas also have a unique adaptation called the sensilium, which detects air currents, thus, aiding in finding a new host by its movements (Mullen and Durden, 2009; Morand et al, 2015). Anatomically, fleas are one of the ectoparasites most likely to leave a carcass early since their legs have evolved into being quite long and spring -like to help jump long distances. This adaptation could have contributed to them quickly finding a host and dispersing from the carcass prior to it being collected. The robustness of the lice dataset is likely a result of them being permanent, obligate ectoparasites that cannot survive off the host for more than a few hours (Monello and Gompper, 2008). Additionally, lice require direct con tact for transfer, which is unlikely to occur with a road-killed squirrel (Mullen and Durden, 2009). With majority of lice being found at mid -level Dorsal Coloration Lightness values, it can be inferred that lice may be less common parasites of fox squirrels as their pelage coloration becomes more black or white. As surfaces become darker, they reflect less solar radiation, becoming much hotter in temperature than lighter surfaces (Jessen, 2012). Since lice easily desiccate in high temperatures, they most likely avoid these darker areas of the squirrels pelage (Mullen and Durden, 2009; Jessen, 2012). Lice also have some seasonality, peaking in the winter months, perhaps further supporting that dark fur exceeds their thermal tolerance (Mullen and Durden, 20 09).

PAGE 12

The relationship between pelage coloration and ectoparasite load in my study appeared to differ across various types of ectoparasites likely as a function of the anatomy and ecology of the ectoparasites (Mullen and Durden, 2009; Payne, 2017). Ticks are most likely attracted to more gray Ventral Coloration Saturation due in part to their eyesight since they recognize shadows, the deeper the gray of the ventral surface, the more likely they are to be attracted to the squirrel (Mullen and Durden, 2009; Centers for Disease Control and Prevention [CDC], 2015). Additionally, the pelage coloration serves to create a small microclimate within the squirrels fur that significantly influences lice ( Mullen and Durden, 2009; Jessen, 2012). Since lice cannot survive in high temperatures, they choose areas of the pelt that are lighter in color to avoid the heat from fur that absorbs more radiation (and are thus, darker in color) (Jessen, 2012). I found that pelages with red coloration attracted fleas and lice. As me ntioned previously, specific colors (red, orange, yellow, purple) and mid-level lightness attracted more lice. This further supports the idea that the squirrels pelts have microclimates since these colors generally reflect more radiation than absorb, making the temperatures more hospitable for lice (Jessen, 2012). Overall, based on the findings of this study, fox squirrels with primarily red-colored dorsal and ventral surfaces that is mid-level in lightness (i.e. not too black or white) are most vulnerable to ectoparasite infestations. The link between ectoparasites and pelage coloration is important for predicting the fitness of an animal as well as its risk for certain diseases. The pelage coloration can allow for the prediction of ectoparasite species on the fox squirrels and, therefore, the diseases they are at risk of due to the arthropod vectors. Ectoparasites have been found to negatively impact animals fitness levels by reducing their body condition, thus making them more susceptible to predation, disease, or even death (Marquardt, 2004; Mehlhorn, 2008; Mullen and Durden, 2009; Jacquin et

PAGE 13

al, 2013; Reissmann and Ludwig, 2013). To combat the possibility of reduced fitness, many animals have evolved pelage colorations that deter ectoparasites (Reissm ann and Ludwig, 2013). Zebras are thought to have evolved their stripes to prevent tsetse and tabanid fly bites, since the flies find striped objects difficult to land on (Caro et al, 2014). The same logic may apply to fox squirrels; their vast array in pelage coloration and patterns might be an evolutionary arms race between the squirrels and ectoparasites, with their pelage colorations evolving often to have even more varieties that ectoparasites are not familiar with or attracted to. ACKNOWLEDGEMENTS Id like to thank Dr. Robert McCleery and Dr. Daniel Greene for supporting this project and providing their invaluable guidance. I am also grateful for those that spent time assisting me with collecting and processing the squirrels, especially Forest Hayes, Courtney Tye, Joshua Ringer, Sara Stiehler, and Verity Mathis. I especially thank Bambi Clemons at the Florida Fish and Wildlife Conservation Commission for allowing me to utilize the labs facilities and supplies as well as for training me on proper necropsy technique. I would also like to thank Matthew Back for graciously loaning his photography equipment to me as well as providing his expertise.

PAGE 14

Literature Cited Agarwal, A. 2011. A Simple Way to Understand Hue, Saturation, and Luminosity. Digital Inspiration. URL: https://www.labnol.org/home/hue-saturation-luminosity/20104/. Aspinall, Victoria. 2014. Diagnostic Laboratory Techniques. Clinical Procedures in Veterinary Nursing. Butterworth-Heinemann, Oxford, UK. Baumgartner, L. L. 1943. Pelage Studies of Fox Squirrels (Sciurus niger rufiventer). American Midland Naturalist. 29(3):588-590. Brigham Young University. N. D. Flea Mounting Procedures. URL: https://fleasoftheworld.byu.edu/Systematics/MountingTechniques Burnham and Anderson. 2002. Model Selection and Multimodel Interference: A Practical Information-Theoretic Approach. Springer, New York, USA. Caro, T., Izzo, A., Reiner Jr., R. C., Walker, H., and T. Stankowich. 2014. The function of zebra stripes. Nature Communications. 5:3535 Centers for Disease Control and Prevention (CDC). 2015. Ticks: Lice Cycle of Hard Ticks That Spread Disease. Atlanta, Georgia, USA. Coyner, D. F, J. B. Wooding and D. F. Forrester. 1996. Comparison of Parasitic Helminths and Arthropods From Two Subspecies of Fox Squirrels (Sciurus niger) in Florida. Journal of Wildlife Diseases. 32(3):492-497. Florida Fish and Wildlife Conservation Commission (FWC). 2016. Floridas Imperiled Species. Management Plan. Tallahassee, Florida, USA. Forrester, D. J. 1992. Squirrels. Parasites and Diseases of Wild Mammals in Florida. University Press of Florida, USA. Fuller, H. S. 1943. Studies on Siphonaptera of eastern North America. Bulletin of the Brooklyn

PAGE 15

Entomology Society. 38:18-23. Hall E. R. 1981. The mammals of North America. 2nd edition. New York: John Wiley and Sons. Humphrey, S. R. 1992. Big Cypress Fox Squirrel (Sciurus niger avicennia). Rare and endangered biota of Florida. Volume 1. Mammals. University Press of Florida, USA. Humphrey, S. R. 1992. Shermans Fox Squirrel (Sciurus niger shermani). Rare and endangered biota of Florida. Volume 1. Mammals. University Press of Florida, USA. Hubbard, J. K., Uy, J. A. C, Hauber, M. E., Hoekstra, H. E., and R. J. Safran. 2010. Vertebrate pigmentation: from underlying genes to adaptive function. Trends in Genetics. 26(5): 231-239. Jacquin, L., Rcapet, C., Prvot-Julliar, A. C., Leboucher, G., Lenouvel, P., Erin, N., Corbel, H., Frantz, A., and J. Gasparini. 2013. A potential role for parasites in the maintenance of color polymorphism in urban birds. Oecologia. 173(3):1089-1099. Jessen, C. 2012. Temperature Regulation in Humans and Other Mammals. Springer, New York, USA Kantola, A.T., and S.R. Humphrey. 1990. Habitat use by Shermans fox squirrel (Sciurus niger shermani) in Florida. Journal of Mammalogy 71(3):411-419. Kiltie, R. A. 1989. Wildfire and the Evolution of Dorsal Melanism in Fox Squirrels, Sciurus niger. Journal of Mammalogy. 70(4): 726-739. Kiltie, R. A. 1992. Comparisons among Fox Squirrels from the Mississippi River Delta. Journal of Mammalogy. 73(4): 906-913. Koprowski J. L. 1994. Sciurus niger. Mammalian species 479:19. Layne, J.N. 1968. Fleas (Siphonaptera) of Florida. Florida Entomology. 54:35 -51. Loeb, S. C., and N. D. Moncrief. 1993. The biology of fox squirrels (Sciurus niger) in the

PAGE 16

Southeast: a review. Pages 1-9 in N. D. Lupton, E., and J. C. Phillips. 2014. Graphic Design: The New Basics. Adams Media, Avon, Massachusetts, USA. Mathison, B. A., and B. S. Pritt. 2014. Laboratory Identification of Arthropod Ectoparasites. Clinical Microbiology Reviews. 27(1): 48-67. Moncrief, J. W. Edwards, and P. A. Tappe, eds. Proceedings of the Second Symposium on Southeastern Fox Squirrels, Sciurus niger. Virginia Museum of Natural History, Martinsville, USA. Marquardt, W. 2004. Fleas, the Siphonaptera. Biology of Disease Vectors. Elsevier Academic Press. Burlington, Massachusetts, USA. Mehlhorn, H. 2008. Ticks. Encyclopedia of Parasitology: A -M. Springer Media. Monello, R. J., and M. E. Gompper. 2009. Relative Importance of Demographics, Locale, a nd Seasonality Underlying Louse and Flea Parasitism of Raccoons ( Procyon lotor). Journal of Parasitology. 95(1): 56-62. Moore, J. C. 1956. Variation in the Fox Squirrel in Florida. American Midland Naturalist. 55(1): 41-65. Moore, J. C. 1957. The natural history of the fox squirrel, Sciurus niger shermani. Bulletin of the American Museum of Natural History. 113:1-71. Palma, R. L. 1978. Slide-mounting of Lice: a Detailed Description of the Canada Balsam technique. The New Zealand Entomologist. 6(4): 432 -436. Payne, V. K., Mbafor, F. L., Pone, J. W., and J. Tchoumboue. 2017. Preliminary Study of Ectoparasites of Horses in the Western Highlands of Cameroon. Veterinary Medicine and Science.

PAGE 17

Pratt, H. D., and H. E. Stark. 1973. Fleas: Pictorial Key to Some Common Species in the United States. Department of Health, Education, and Welfa re, Public Health Service, Communicable Disease Center (CDC). Atlanta, Georgia, USA Pratt, H. D., and C. J. Stojanovich. 1966. Acarina: Illustrated Key to Some Common Adult Female Mites and Adult Ticks. Pictorial Keys to Arthro pods, Reptiles, Birds, and Mammals of Public Health Importance. U. S. Dept. of Health, Education, and Welfare, Public Healh Service, Communicable Disease Center (CDC). Atlanta, Georgia, USA. R Core Team, 2014. R: A Language and Environment for Statistical Computing. Austria, Vienna. Reissmann, M., and A. Ludwig. 2013. Pleiotropic effects of coast colour -associated mutations in humans, mice, and other mammals. Seminars in Cell and Developmental Biology 24: 576586 Rogers, A. J. 1953. A study of the ixodid ticks of northern Florida, including the biology and life history of Ixodes scapularis Say (Ixodidae: Acrina). PhD dissertation. University of Maryland, College park. 191 pp. Rondon, S., and M. Corp. N. D. Preserving Insects and Related Arthropods. Oregon State University. URL: http://extension.oregonstate.edu/umatilla/sites/default/files/PRESERVING__INSECTS.p df Stojanovich, C. J., and H. D. Pratt. 1961. Anoplura: Pictorial Key to Some Common Genera of Sucking Lice.. Lice of Public Health Importance and Th eir Control. Department of Health, Education, and Welfare, Public Health Service, Communicable Disease Center. Atlanta, Georgia, USA

PAGE 18

University of Rhode Island. Frequently Asked Questions: Seasonal Information. TickEncounter Resource Center. URL: http://www.tickencounter.org/faq/seasonal_information Weigl, P. D., M. A. Steele, L.J. Sherman, J. C. Ha, and T.S. Sharpe. 1989. The ecology of the fox squirrel (Sciurus niger) in North Carolina: implications for survival in the Southeast. Bulletin of Tall Timbers Research Station 24:1 -94. Wooding, J. B. 1990. Status, life history, and management of fox squirrels in Florida. Final Performance Report, Florida Game and Fresh Water Fish Commission, Tallahassee, Florida, USA.

PAGE 19

TABLES AND FIGURES Fig. 1. The range of pelage colorations seen in fox squirrels. The first squirrel on the left exhibits complete melanism, while the last squirrel is primarily white.

PAGE 20

Fig. 2. The color wheel in which each degree represents a specific hue of color in HSL values. Source: http://dba.med.sc.edu/price/irf/Adobe_tg/models/hsb.html Fig. 3. Saturation and lightness scales using the red hue. Source: https://www.techfry.com/css-tutorial/css3-colors-rgba-hsl-and-hsla

PAGE 21

Fig. 4. The distribution of the 25 fox squirrels processed in this study across Florida, USA, 20122015.

PAGE 22

Fig. 5. Top: Sample pelt from the first cluster of pelages with a Dorsal Hue Coloration of 4 -75. Bottom: Sample pelt from the second cluster of pelages with a Dorsal Hue Coloration of 270 300. Table 1. Arthropod ectoparasites found in this study. Arthropod Species Number Present Fleas Orchopeas howardii 11 Lice Neohaematopinus sciurinus Hoplopleura sciuricola 79 277 Ticks Ixodes scapularis Amblyomma maculatum 1 2

PAGE 23

Table 2. Candidate models with color covariates, Akaikes Information Criterion (AICc), difference in AICc ("AICc) between a model and the model with the lowest AICc, Akaike weight (Wi) and number of parameters (k) for the candidate models used to estimate the relationship between pelage coloration and ectoparasite richness. Model AICc "AICc Wi k Null 57.111 0.000 0.252 1 Tail Lightness 59.250 2.139 0.086 2 Dorsal Hue 59.363 2.252 0.082 2 Tail Hue 59.374 2.264 0.081 2 Dorsal Saturation 59.398 2.287 0.080 2 Ventral Saturation 59.457 2.347 0.078 2 Ventral Hue 59.466 2.355 0.077 2 Ventral Lightness 59.472 2.361 0.077 2 Dorsal Lightness 59.482 2.371 0.077 2 Tail Saturation 59.482 2.371 0.077 2 Hue 64.655 7.545 0.006 4 Tail (Hue + Saturation + Lightness) 64.658 7.548 0.006 4 Lightness 64.685 7.575 0.006 4 Dorsal (Hue + Saturation + Lightness) 64.752 7.642 0.006 4 Saturation 64.835 7.724 0.005 4 Ventral (Hue + Saturation + Lightness) 64.903 7.792 0.005 4 Table 3. Candidate models with color covariates, Akaikes Information Criterion (AICc), difference in AICc ("AICc) between a model and the model with the lowest AICc, Akaike weight (Wi) and number of parameters (k) for the candidate models used to estimate the relationship between pelage coloration and tick burden. Model AICc "AICc Wi k Ventral Saturation 19.340 0.000 0.275 2 Lightness 20.852 1.512 0.129 4 Ventral Lightness 21.088 1.748 0.115 2 Tail Saturation 21.668 2.328 0.086 2 Tail Hue 21.871 2.531 0.078 2 Null 22.282 2.941 0.063 1 Tail Lightness 22.320 2.979 0.062 2 Saturation 23.646 4.305 0.032 4 Ventral (Hue + Saturation + Lightness) 23.712 4.371 0.031 4 Dorsal Hue 23.856 4.516 0.029 2 Dorsal Lightness 24.020 4.680 0.026 2 Ventral Hue 24.094 4.754 0.026 2 Dorsal Saturation 24.426 5.086 0.022 2 Tail (Hue + Saturation + Lightness) 25.399 6.059 0.013 4 Hue 25.687 6.347 0.012 4 Dorsal (Hue + Saturation + Lightness) 28.885 9.544 0.002 4

PAGE 24

Table 4. Candidate models with color covariates, Akaikes Information Criterion (AICc), difference in AICc ("AICc) between a model and the model with the lowest AICc, Akaike weight (Wi) and number of parameters (k) for the candidate models used to estimate the relationship between pelage coloration and flea burden. Model AICc "AICc Wi k Ventral Hue 38.543 0.000 0.289 2 Ventral Saturation 38.913 0.369 0.240 2 Ventral Lightness 39.842 1.299 0.151 2 Hue 40.131 1.588 0.131 4 Lightness 41.140 2.597 0.079 4 Dorsal Lightness 42.650 4.107 0.037 2 Tail (Hue + Saturation + Lightness) 43.380 4.837 0.026 4 Saturation 43.621 5.077 0.023 4 Tail Hue 45.062 6.519 0.011 2 Dorsal (Hue + Saturation + Lightness) 45.503 6.959 0.009 4 Dorsal Hue 46.798 8.255 0.005 2 Null 59.286 20.742 0.000 1 Tail Lightness 60.622 22.078 0.000 2 Dorsal Saturation 60.849 22.305 0.000 2 Tail Saturation 61.523 22.979 0.000 2 Table 5. Candidate models with color covariates, Akaikes Information Criterion (AICc), difference in AICc ("AICc) between a model and the model with the lowest AICc, Akaike weight (Wi) and number of parameters (k) for the candidate models used to estimate the relationship between pelage coloration and louse burden. Model AICc "AICc Wi k Dorsal (Hue + Saturation + Lightness) 507.866 0.000 0.684 4 Hue 509.412 1.545 0.316 4 Tail Hue 535.787 27.920 0.000 2 Saturation 544.174 36.308 0.000 4 Tail Saturation 546.828 38.962 0.000 2 Dorsal Hue 557.175 49.309 0.000 2 Ventral (Hue + Saturation + Lightness) 563.540 55.674 0.000 4 Ventral Hue 576.096 68.229 0.000 2 Tail Lightness 585.308 77.441 0.000 2 Lightness 586.135 78.268 0.000 4 Dorsal Lightness 590.252 82.385 0.000 2 Ventral Saturation 591.893 84.026 0.000 2 Null 595.654 87.788 0.000 1 Dorsal Saturation 597.175 89.308 0.000 2 Ventral Lightness 598.025 90.159 0.000 2





PAGE 1

/ Honors Thesis GRANT OF PERMISSIONS: Internet Distribution Agreement In reference to the following Undergraduate Honors Thesis, including any supplementary file(s) : Rowan Back Influence of Southeastern Fox Squirrel (Sciurus niger) Pelage Coloration on Ectoparasite Richness and Abundance University Of Florida: Gainesville,FL. 29 April 2017 I, 8oUJa...rJ f3dclL as copyright holder or licensee with the authority to grant copyright permissions for the aforementioned title(s), hereby authorize the University of Florida, acting on behalf of the Board of Trustees of the University of Florida, to digitize, distribute, and archive the title{s) for nonprofit, educational purposes via the Internet or successive technologies This is a non-exclusive grant of permissions for on-line and off-line use for an indefinite term Off-line uses shall be consistent either, for educational uses, with the terms of U S. copyright legislation's "fair use" provisions or, by the University of Florida, with the maintenance and preservation of an archival copy Digitization allows the University of Florida to generate imageand text-based versions as appropriate and to provide and enhance access using search software. This grant of permissions prohibits use of the digitized versions for commercial use or profit. Signature of Copyri llouJan t5d d Printed or Typed Name of Copyright Holder 3i30/ OLO 1'::;DfueoiSignature Attention: IR Manager Digital Production Services; Smathers Libraries University of Florida P O Box 117003 Gainesville, FL32611-7003 P: 352 273.2831 IRManager@uflib ufl edu Form online : http://digital.uflib ufl edu/procedures/copyrlghtiGrantofPermissions.doc



PAGE 1

Influence of Southeastern Fox Squirrel ( Sciurus niger ) Pelage Coloration on Ectoparasite Richness and Abundance Rowan Emily BŠck UFID: 3318 3480 rowane@ufl.edu University of Florida Department of Biology Faculty Advisor: Dr. Robert A. McCleery CLAS Latin Honors

PAGE 2

ABSTRACT Within the southeastern United States, the fox squirrel ( Sciurus niger ) exhibits great var iation in its pelage coloration Current research shows that there is a relationship between ectoparasites and pelage coloration in some mammals, but it is not known if this occurs in southeastern fox squirrels, which are north America 's most variably colored mammal. To assess the effect of pelage coloration on the ectoparasite richness and abundance of fox squirre ls in Florida, I collected and identified ectoparasites from 25 fox squirrels and I quantified pelage coloration values of hue, saturation, and lightness on the dorsal, ventral, and tail surfaces. Louse were found to have the greatest relationship with pel age coloration, largely because they comprised the majority of the ectoparasites. Dorsal Coloration was the top model for explaining the number of lice on fox squirrels. This model included Dorsal Coloration Hue ( : 0.008, SE: 0.001, 95% CI: 0.011 0.007 ) Dorsal Coloration Saturation ( : 0.028, SE: 0.008, 95% CI: 0.011 0.045) and Dorsal Coloration Lightness ( : 0.044, SE: 0.006, 95% CI: 0.057 0.032) These results concur with the higher prevalence of louse on the dorsum of fox squirrels. Red, yellow, orange, and purple attracted lice, while green and blue repelled them. Since this study yield ed low variability in ectoparasite species, it is recommended that live, trap caught fox squirrels be used to ensure ectoparasite diversity. INTRODUCTION The fox squirrel ( Sciurus niger ) is a large tree squirrel that occurs naturally east of the Rocky Mountains ( Hall, 1981; Koprowski, 1994 ). In the southeastern U nited States, 6 subspecies of fox squirrel ( S n. vulpinus, S. n. niger, S. n. shermani, S. n. bachmani, S. n. avicennia and S. n. cinereus ) inhabit the coastal plain into southern Virginia and west into Alabama They are

PAGE 3

collectively known as south eastern fox squirrels and are well known for having a variety of pelage colorations (Baumgartner, 1943; Moore, 1956 ; Loeb and Moncrief, 1993) Four subspecies of southeastern fox squirrel s are listed as occurring in Florida S. n. niger S. n. bachmani S n. shermani and S. n. avicennia (Moore 1956). Currently, Florida considers the Sherman's fox squirrel ( S n. shermani ) to be a species of special concern and the Big Cypress fox squirrel ( S. n. avicennia ) to be threatened (Florida Fish and Wildlife Conse rvation Commission [FWC] 2016). It is generally recognized that south eastern fox squirrels have three prominent color phases red, gray, or black however, additional coloration variations have also been observed, such as white, tan, and agouti (Loeb and Moncrief, 1993; Moore, 1956) (Figure 1). Sherman's fox squirrel typically has a white nose and ears with a black dorsal portion of the head and variable coloration o n the remainder of the pelage, usually falling within all dark, all tan, or dark/tan mixed (Humphrey, 1992; Kiltie, 1992). Big Cypress fox squirrels' most common color ph ase is buff with a white nose and eartips, a black crown, and an agouti or black agouti dorsum, but they can vary between black and tan (Ashton and Humphrey, 1992; Moore, 1956). Southern fox squirrels ( S. n. niger ) commonly occur in the gray white color ph ase as well as the black color phase (Moore 1956). Certain coat colorations have been shown to attract ectoparasites and animals can develop certain coloration polymorphisms to deter ectoparasites (Hubbard, 2010; Jacquin et al 2013; Reissmann and Ludwig, 2013; Payne, 2017). For example, black horses from Cameroon attract more Amblyomma ticks than those that were brown, white, or gray (Payne, 2017). In birds, dark colored, feral pigeons had more ectoparasites when inhabiting moderately urbanized environment s than their pale r counterparts (Jacquin et al 2013). The clear link between an

PAGE 4

animal's coloration and ectoparasites suggests there may be associations with an immunological response, that in turn, can influence the birds' fitness (Jacquin et al 2013; Re issmann and Ludwig, 2013). Thus, pelage coloration may have evolved over time as a result of less fit morphs succumbing to the effects of having a high burden of ectoparasites (Jacquin et al 2013). Several species of ticks, mites, fleas, and sucking lice have been documented as occurring on fox squirrels in Florida (Forrester, 1992; Coyner et al, 1996). Six species of ticks were identified: I texanus I scapularis A americanum A maculatum A t uberculatum and D variabilis (Rogers, 1953; Moore, 1957; Forrester, 1992; Coyner et al 1996). The two species of fleas found include O. howardii and H. glacialis affinis (Fuller, 1943; Moore, 1957; Layne, 1971; Forrester, 1992; Coyner et al 1996). The three species of sucking lice identified were H. sciuricola N. sciurinus and Enderlinus longiceps (Moore, 1957; Forrester, 1992; Coyner et al 1996). Also found were six species of mites ( E. alfreddugesi A. megaventralis Listrophorus sp., A. casalis N whartoni and E. diversa ) and one species of botfly larvae ( Cuterebra sp.) (Moore, 1957; Forrester, 1992; Coyner et al 1996). Presently, there are no studies that explore if pelage coloration variation in fox squirrels is related to ectoparasite load. This study can be useful in predicting which colors attract ectoparasites, and thus, the risk of an animal to certain arthropod vectored diseases as well as the fitness of an animal. Since fox squirrels vary significantly in their pelage colorations, they are a good model to use in this study (Moore, 1956). Therefore, my objective was to determine if the ectoparasite burden of southeastern fox squirrel subspecies of Florida was related to their pelage coloration. Specifically, I quantified the coloration of pelages on the dorsal, ventral, and tail surfaces of fox squirrels and related them to the measure of ectoparasites, including the total number of ectoparasites and the number of ectoparasite species.

PAGE 5

METHODS I received road killed fox squirrels colle cted throughout Florida from 2012 2015. I stored squirrels were in a freezer and then transferred them to a refrigerator before necropsy. Fox squirrels came with locational coordinates and I plotted the m spatially using ArcGIS 10.4 (ESRI, Redlands, C alifornia ). Ectoparasite Collection and Identification I used a flea comb and a toothbrush to run through the fur to collect any ectoparasites hanging loosely on the hairs this was done dorsally, ventrally, through the tail, behind the ears, and around the anal regio n (Aspinall, 2014). I conducted this process over white legal sized paper to collect any ectoparasites that may have fallen off the squirrel or were not in the teeth of the comb or toothbrush. The comb through procedure was repeated several times to ensure majority of the ectoparasites had been collected. I then viewed squirrels under a dissecting microscope to find any ectoparasites that remained. I used forceps to pluck them from the skin, if any were found. Each type of ectoparasites (fleas, lice, ticks) w as capped at 50 individuals that could be collected. I stored the collected ectoparasites in plastic vials filled with 70% isopropanol until mounting. I utilized Hoyer's medium to slide mount representative lice and fleas from each squirrel, while ticks and the rest of the lice and fleas remained in 70% isopropanol (Rondon and Corp, N.D.). I slide mounted lice on their dorsum with their appendages spread apart, while I mounted fleas laterally with their legs separated (Palma, 1978; Brigham Young Universi ty, N. D.). I viewed lice and fleas under a light microscope, using the 4X and 10X objective lenses. I utilized dichotomous keys from the CDC to identify the ectoparasites (Stojanovich and Pratt,

PAGE 6

1961; Pratt, 1966; Pratt and Stark, 1973). To i dentify lic e, I relied on the length of the claws and legs, number of antennal segments, presence/absence of sclerotized plates on the abdomen, abdominal plate articulations, and spine like seta at the posterior apical angle on the first antennal segment (Stojanovich and Pratt, 1961). Flea identification depended on the presence/absence of the genal, pronotal, and abdominal combs, length of labial palps, and number of plantar bristles (Pratt and Stark, 1973). The ticks were first categorized on the basis of being "har d" or "soft." The color (or "ornateness") was then examined along with other characteristics such as palp length, terminal spurs, and presence/absence of eyes (Pratt, 1966). I identified each ectoparasite to species level. Pelt Collection and Photographi ng I flensed pelts from the squirrel carcasses and sent them to the Florida Museum of Natural History to be vouchered for the collections. Borax was used to cure the pelts after the remaining fat was stripped from the skin and then pelts hung on a drying rack for 1 2 weeks to further preserve the skin. I photographed pelts using a Nikon D5100 DSLR camera in a 30" x 30" x 30" light box with two 100W LED lights placed on either side to illuminate the inside. A white background was used to ensure the pelts we re distinct in the image. Pelage Analysis Using Digital Color Meter (Apple Inc., Cupertino, California), I measured RGB (red, green, blue) values of the dorsal, ventral, and tail surfaces in each squirrel's photograph. By measuring RGB values, I could ob tain the exact numerical color composition of a squirrel 's pelage in terms of the colors red, green, and blue (Lupton and Phillips, 2014). I isolate d the specific amount of each of respective color to quantitatively determine ectoparasites' preferences. I selected three points to read RGBs on each surface that were equally spaced apart

PAGE 7

and spanned the entire section. For example, on the ventral surface, the three points selected included posterior to the axillary region, anterior to the groin, and one in th e center of the two previous points. The three individual RGB values obtained were averaged together to create an overall RGB value for that section. RGB values only exist when evaluating pixels on a computer monitor (Lupton and Phillips, 2014) so I conver ted the RGB values to Hue, Saturation, Lightness (HSL) values (hereafter, pelage color covariates). Hue is the actual color and has a range of 0 to 360 (or degrees on the color wheel) (Figure 2). Zero or 360 is red, 120 is green, and 240 is blue. Saturatio n denotes the amount of gray in a color as a percentage value; a value of 0% indicates mostly gray while 100% is white (Figure 3, top). Lightness is the amount of white or black that is mixed in the color; 0% is dark and 100% is light (Figure 3, bottom). S tatistical Analysis I examined the relationships between my ectoparasite metrics (the total number of fleas, ticks, louse, ectoparasite species, and ectoparasite individuals). Then, I assessed the influence of color covariates (HSL values as continuous pre dictors) on ectoparasite metrics using generalized linear models and a Poisson distribution in R (R Core Team, 2014). I conducted a separate analysis for each ectoparasite response variable. I evaluated single and multivariable models using the color of each body segment (e.g., Dorsal Hue + Dorsal Saturation + Dorsal Lightness) or for all body segments (e.g., Dorsal Hue + Ventral Hue + Tail Hue) as predictors. I evaluated my candidate models using Akaike Information Criterion corrected for small sample si ze (AICc), where the level of importance was assessed by model weights (Burnham and Anderson, 2002). I determined the covariates predictive importance by inspecting the conditional beta coefficient ( ) estimates and their 95% confidence intervals (CI), wit h significance defined as CIs for a variable that did not overlap zero.

PAGE 8

RESULTS I processed 25 fox squirrels (16 males, 9 females) to quantify and identify ectoparasite metrics ( Table 1) These individuals represented S n. niger and S. n. shermani subspecies in the Florida panhandle and the central part of the p eninsula (Figure 3). I dropped the number of ectoparasites from my analysis because of a high Pearson's correlation coefficien t with lice (|0.997|), a result of most ectoparasites in my stud y being lice and overall few ticks and fleas recovered from fox squirrel carcasses M y top model for ectoparasite richness was the null model, indicating pelage color ation variables were not influential (Table 2) My top model for assessing the number of t icks on fox squ i rrels had one variable, Ventral Color ation Saturation ( : 0.180, SE: 0.079, 95% CI: 0.028 0.360 ) (Table 3) ; however, the significance of this variable, and it's ranking above the null model is may be spurious because only 2 of the 25 individuals had ticks (1 tick and 2 ticks, respectively). All other models had beta coefficients with 95% confidence intervals that included zero The squirrel with 1 tick had a Ventral Coloration Saturation of 13.2% (more gray), while the squirrel with 2 ticks had a value of 36.4% (more pigmented in color, but still in the gray range). For fleas, the best model included the combination of Ventral Coloration Hue + Saturation + Lightness; however, the top model had more parameters than could be estim ated from the data making this ranking misleading. I dropped the model and reran the analysis, which resulted in the models of Ventral Coloration Hue, Coloration Saturation, and Lightness all as the top three models, respectively (Table 4 ). Ventral Colorat ion Hue ( : 0.079, SE: 0.016, 95% CI: 0.115 0.048), Coloration Saturation ( : 0.346, SE: 0.101, 95% CI: 0.587 0.178), and Lightness ( : 0.197, SE: 0.051, 95% CI: 0.317 0.107) all negatively influenced the number of fleas on fox squirrels The squirrel with the highest number of fleas (7) had the lowest Ventral

PAGE 9

Coloration Hue, Coloration Saturation, and Coloration Lightness (26, 9.1, and 50.4, respectively) compared to the other four squirrels with fleas, meaning that its coloration was mo re red with gray undertones and was mid level in brightness. For the other four squirrels with only 1 flea each, their Ventral Coloration Hue ranged from 62 69, Ventral Coloration Saturation was between 12.1 18.2%, and Ventral Coloration Lightness ranged f rom 58 65.5%. These squirrels were more yellow green with less gray undertones, and brighter in color. My top model for assessing the number of lice on fox squirrels was the Tail Coloration (Tail Hue + Tail Saturation + Tail Brightness). Although the Tai l had 100% support over all other models, no lice were actually found on the tail during ectoparasite collection. Therefore, I reran the analysis after dropping the Tail model. The top model for the number of lice was Dorsal Coloration (a combination of Hu e, Color ation Saturation, and Lightness) (Table 5 ). Dorsal Coloration Hue ( : 0.008, SE: 0.001, 95% CI: 0.011 0.007), and Dorsal Coloration Lightness ( : 0.044, SE: 0.006, 95% CI: 0.057 0.032) negatively influenced the number of lice on fox squirrels, whereas Dorsal Color ation Saturation ( : 0.028, SE: 0.008, 95% CI: 0.011 0.045) was a positive predictor I found that lice parasitized squirrels within the Dorsal Coloration Hue ranges of 4 75 and 270 300, meaning that the y preferred squirrels with more red, orange, yellow, and purple coloration. In relation to Dorsal Coloration Saturation, it appeared that lice generally favored squirrel pelages with more gray undertones, or within the 2 7% range on the saturation scale. F or Dorsal Coloration Lightness, I noticed there was a large cluster of lice in the range of 38 56%, indicating that they preferred pelages with mid level brightness of coloration. Hue was also a competing model as it was only 1.5 AIC values from Dorsal C ol oration, Dorsal Coloration H ue had a relatively small effect ( : 0.004, SE; 0.001, 95 % CI:

PAGE 10

0.007 0.002) and Ventral Coloration Hue was not significant ( : 0.010, SE: 0.009, 95% CI: 0.007 0.028) DISCUSSION Overall, my findings suggest that color covariates did not influence ectoparasite metrics equally across different ectoparasite types. This is likely due to the ectoparasites collected primarily consisting of louse species of the 25 squirrels processed, 2 1 had lice which accounted for 96.2% of all ectoparasites collected. The lack of diversity in the dataset was evident with the overall lack of variability in ectoparasite richness between fox squirrel individuals (Table 1); however, this was likely influen ced by the small sample size. Ticks and fleas comprised only 3.8% of all ectoparasites collected in this study. I found that Ventral Coloration Saturation was the top model explaining tick prevalence, which concurs with what we know about tick behavior. Specifically, ticks typically engage in a behavior known as questing, where they climb onto weeds, grasses, and bushes to wait for hosts to latch onto while raising their first pair of legs (Mullen and Durden, 2009; CDC, 2015). Therefore, the ventral surfa ce is the most likely surface to be seen by ticks while questing. Some species of ticks are able to see shadows and use this to locate hosts (Centers for Disease Control and Prevention [CDC], 2015). With the Ventral Coloration Saturation being more gray, t he ticks may have viewed the squirrel's ventral surface as a large shadow and used this for host detection (Mullen and Durden, 2009; Centers for Disease Control and Prevention [CDC], 2015). Additionally, ticks, especially Argasid ticks, are intermittent fe eders and, although they are considered obligate parasites, they spend 90% of their lifecycle off the host (Mullen and Durden, 2009; Mathison and Pritt, 2014); this likely explains the low abundance of ticks found on the fox squirrel carcasses. Perhaps a g reater influence of the low tick abundance on the fox squirrels is a

PAGE 11

result of their attraction to a heat source (Mehlhorn, 2008; Mullen and Durden, 2009). The ticks likely vacated the road killed squirrel carcass due to its loss in body heat (Mullen and D urden, 2009). The reliability in the top models with flea abundance was also uncertain, as fleas would have similarly abandoned the host upon its carcass dropping in temperature, since they are also notorious for being attracted to a host's body warmth (Ma rquardt, 2004; Mullen and Durden, 2009). Fleas also have a unique adaptation called the sensilium, which detects air currents, thus, aiding in finding a new host by its movements (Mullen and Durden, 2009; Morand et al 2015). Anatomically, fleas are one of the ectoparasites most likely to leave a carcass early since their legs have evolved into being quite long and spring like to help jump long distances. This adaptation could have contributed to them quickly finding a host and dispersing from the carcass p rior to it being collected. The robustness of the lice dataset is likely a result of them being permanent, obligate ectoparasites that cannot survive off the host for more than a few hours (Monello and Gompper, 2008). Additionally, lice require direct con tact for transfer, which is unlikely to occur with a road killed squirrel (Mullen and Durden, 2009). With majority of lice being found at mid level Dorsal Coloration Lightness values, it can be inferred that lice may be less common parasites of fox squirre ls as their pelage coloration becomes more black or white. As surfaces become darker, they refl e ct less solar radiation, becoming much hotter in temperature than lighter surfaces (Jessen, 2012). Since lice easily desiccate in high temperatures, they most likely avoid these darker areas of the squirrel's pelage (Mullen and Durden, 2009; Jessen, 2012). Lice also have some seasonality, peaking in the winter months, perhaps further supporting that dark fur exceeds their thermal tolerance (Mullen and Durden, 20 09).

PAGE 12

The relationship between pelage coloration and ectoparasite load in my study appeared to differ across various types of ectoparasites likely as a function of the anatomy and ecology of the ectoparasites (Mullen and Durden, 2009; Payne, 2017 ) Ticks are most likely attracted to more gray Ventral Coloration Saturation due in part to their eyesight since they recognize shadows, the deeper the gray of the ventral surface, the more likely they are to be attracted to the squirrel (Mullen and Durden, 2009; Centers for Disease Control and Prevention [CDC], 2015). Additionally, the pelage coloration serves to create a small microclimate within the squirrel's fur that significantly influences lice ( Mullen and Durden, 2009; Jessen, 2012 ). Since lice cannot survi ve in high temperatures, they choose areas of the pelt that are lighter in color to avoid the heat from fur that absorbs more radiation (and are thus, darker in color) (Jessen, 2012). I found that pelages with red coloration attracted fleas and lice. As me ntioned previously, specific colors (red, orange, yellow, purple) and mid level lightness attracted more lice. This further supports the idea that the squirrels' pelts have microclimates since these colors generally reflect more radiation than absorb, maki ng the temperatures more hospitable for lice (Jessen, 2012). Overall, based on the findings of this study, fox squirrels with primarily red colored dorsal and ventral surfaces that is mid level in lightness (i.e. not too black or white) are most vulnerable to ectoparasite infestations. The link between ectoparasites and pelage coloration is important for predicting the fitness of an animal as well as its risk for certain diseases. The pelage coloration can allow for the prediction of ectoparasite species on the fox squirrels and, therefore, the diseases they are at risk of due to the arthropod vectors. Ectoparasites have been found to negatively impact animals' fitness levels by reducing their body condition, thus making them more susceptible to predation, disease, or even death (Marquardt, 2004; Mehlhorn, 2008; Mullen and Durden, 2009; Jacquin et

PAGE 13

al 2013; Reissmann and Ludwig, 2013). To combat the possibility of reduced fitness, many animals have evolved pelage colorations that deter ectoparasites (Reissm ann and Ludwig, 2013). Zebras are thought to have evolved their stripes to prevent tsetse and tabanid fly bites, since the flies find striped objects difficult to land on (Caro et al 2014). The same logic may apply to fox squirrels; their vast array in pe lage coloration and patterns might be an evolutionary arms race between the squirrels and ectoparasites, with their pelage colorations evolving often to have even more varieties that ectoparasites are not familiar with or attracted to. ACKNOWLEDGEMENTS I 'd like to thank Dr. Robert McCleery and Dr. Daniel Greene for supporting this project and providing their invaluable guidance. I am also grateful for those that spent time assisting me with collecting and processing the squirrels, especially Forest Hayes, Courtney Tye, Joshua Ringer, Sara Stiehler, and Verity Mathis. I especially thank Bambi Clemons at the Florida Fish and Wildlife Conservation Commission for allowing me to utilize the lab's facilities and supplies as well as for training me on proper necr opsy technique. I would also like to thank Matthew Back for graciously loaning his photography equipment to me as well as providing his expertise.

PAGE 14

Literature Cited Agarwal A. 2011. A Simple Way to Understand Hue, Saturation, and Luminosity. Digital Inspiration. URL: https://www.labnol.org/home/hue saturation luminosi ty/20104/ Aspinall, Victoria. 2014. Diagnostic Laboratory Techniques. Clinical Procedures in Veterinary Nursing. Butterworth Heinemann, Oxford, UK. Baumgartner, L. L. 1943. Pelage Studies of Fox Squirrels ( Sciurus niger rufiven ter ). American Midland Naturalist. 29(3):588 590. Brigham Young University. N. D. Flea Mounting Procedures. URL: https://fleasoftheworld.byu.edu/Systematics/MountingTechniques Burnham and Anderson. 2002. Model Selection and Multimodel Interference: A Practical Information Theoretic Approach. Springer, New York, USA. Caro, T., Izzo, A., Reiner Jr., R. C., Walker, H., and T. Stankowich. 2014. The function of zebra stripes. Nature Communications. 5:3535 Centers for Disease Control and Prevention (CDC). 2015. Ticks: Lice Cycle of Hard Ticks That Spread Disease. Atlanta, Georgia, USA. Coyner, D. F, J. B. Wooding and D. F. Forrester. 1996. Comparison of Parasitic Helminths and Arthropods From Two Subspecies of Fox Squirrels ( Sciurus niger ) in Florida. Journal of Wildlife Diseases. 32(3):492 497. Florida Fish and Wildlife Conservation Commission (FWC). 2016. Florida's Imperiled Species. Management Plan. Tallahassee, Florida, USA. Forrester, D. J. 1992. Squirrels. Parasites and Diseases of Wild Mammals in Florida. University Press of Florida, USA. Fuller, H. S. 1943. Studies on Siphonaptera of eastern North America. Bulletin of the Brooklyn

PAGE 15

Entomology Society. 38:18 23. Hall E R. 1981. The mammals of North America. 2 nd edition. New York: John Wiley and Sons. Humphrey, S. R. 1992. Big Cypress Fox Squirrel ( Sciurus niger avicennia ). Rare and endangered biota of Florida. Volume 1. Mammals. University Press of Florida, USA. Humphrey, S. R. 1992. Sherman's Fox Squirrel ( Sciurus niger shermani ). Rare and endangered biota of Florida. Volume 1. Mammals. University Press of Fl orida, USA. Hubbard, J. K., Uy, J. A. C, Hauber, M. E., Hoekstra, H. E., and R. J. Safran. 2010. Vertebrate pigmentation: from underlying genes to adaptive function. Trends in Genetics. 26(5): 231 239. Jacquin, L., RÂŽcapet, C., P rÂŽvot Julliar, A. C., Leboucher, G., Lenouvel, P., Erin, N., Corbel, H., Frantz, A., and J. Gasparini. 2013. A potential role for parasites in the maintenance of color polymorphism in urban birds. Oecologia. 173(3):1089 1099. Jessen, C. 2012. Temperature Regulation in Humans and Other Mammals. Springer, New York, USA Kantola, A.T., and S.R. Humphrey. 1990. Habitat use by Sherman's fox squirrel ( Sciurus niger shermani ) in Florida. Journa l of Mammalogy 71(3):411 419. Kiltie, R. A. 1989. Wildfire and the Evolution of Dorsal Melanism in Fox Squirrels, Sciurus niger Journal of Mammalogy. 70(4): 726 739. Kiltie, R. A. 1992. Comparisons among Fox Squirrels from the Mississippi River Delta. Journal o f Mammalogy. 73(4): 906 913. Koprowski J. L. 1994. Sciurus niger Mammalian species 479:1 9. Layne, J.N. 1968. Fleas (Siphonaptera) of Florida. Florida Entomology. 54:35 51. Loeb, S. C., and N. D. Moncrief. 1993. The biology of fox squirrels ( Sciurus niger ) in the

PAGE 16

Southeast: a review. Pages 1 9 in N. D. Lupton, E., and J. C. Phillips. 2014. Graphic Design: The New Basics. Adams Media, Avon, Massachusetts, USA. Mathison, B. A., and B. S. Pritt. 2014. Laboratory Identification of Arthropod Ectoparasites. Clinical Microbiology Reviews. 27(1): 48 67. Moncrief, J. W. Edwards, and P. A. Tappe, eds. Proceedings of the Second Symposium on Southeastern Fox Squirrels, Sciurus niger Virginia Museum of Natural History, Martinsville, USA. Marquardt, W. 2004. Fleas, the Siphonaptera. Biology of Disease Vectors. Elsevier Academic Press. Burlington, Massachusetts, USA. Mehlhorn, H. 2008. Ticks. Encyclopedia of Parasitology: A M. Springer Media. Monello, R. J., and M. E. Gompper. 2009. Relative Importance of Demographics, Locale, a nd Seasonality Underlying Louse and Flea Parasitism of Raccoons ( Procyon lotor) Journal of Parasitology. 95(1): 56 62. Moore, J. C. 1956. Variation in the Fox Squirrel in Florida. American Midland Naturalist. 55(1): 41 65. Moore, J. C. 1957. The natural history of the fox squirrel, Sciurus niger shermani Bulletin of the American Museum of Natural History. 113:1 71. Palma, R. L. 1978. Slide mounting of Lice: a Detai led Description of the Canada Balsam technique. The New Zealand Entomologist. 6(4): 432 436. Payne, V. K., Mbafor, F. L., Pone, J. W., and J. Tchoumboue. 2017. Preliminary Study of Ectopar asites of Horses in the Western Highlands of Cameroon. Veterinary Medicine and Science.

PAGE 17

Pratt, H. D., and H. E. Stark. 1973. Fleas: Pictorial Key to Some Common Species in the United States. Department of Health, Education, and Welfa re, Public Health Service, Communicable Disease Center (CDC). Atlanta, Georgia, USA Pratt, H. D., and C. J. Stojanovich. 1966. Acarina: Illustrated Key to Some Common Adult Female Mites and Adult Ticks. Pictorial Keys to Arthro pods, Reptiles, Birds, and Mammals of Public Health Importance. U. S. Dept. of Health, Education, and Welfare, Public Healh Service, Communicable Disease Center (CDC). Atlanta, Georgia, USA. R Core Team, 2014. R: A Language and Environment for Statistical Computing. Austria, Vienna. Reissmann, M., and A. Ludwig. 2013. Pleiotropic effects of coast colour associated mutations in humans, mice, and other mammals. Seminars in Cell and Developmental Biology 24: 576 586 Rogers, A. J. 1953. A study of the ixodid ticks of northern Florida, including the biology and life history of Ixodes scapularis Say (Ixodidae: Acrina). PhD dissertation. University of Maryland, College park. 191 pp. Ro ndon, S., and M. Corp. N. D. Preserving Insects and Related Arthropods. Oregon State University. URL: http://extension.oregonstate.edu/umatilla/sites/default/files/PRESERVING__INSECTS.p df Stojanovich, C. J., and H. D. Pratt. 1961. Anoplura: Pictorial Key to Some Common Genera of Sucking Lice.. Lice of Public Health Importance and Th eir Control. Department of Health, Education, and Welfare, Public Health Service, Communicable Disease Center. Atlanta, Georgia, USA

PAGE 18

University of Rhode Island. Frequently Asked Questions: Seasonal Information. TickEncounter Resourc e Center. URL: http://www.tickencounter.org/faq/seasonal_information Weigl, P. D., M. A. Steele, L.J. Sherman, J. C. Ha, and T.S. Sharpe. 1989. The ecology of the fox squirrel (Sciurus niger) in North Carolina: implications for survival in the Southeast. Bulletin of Tall Timbers Research Station 24:1 94. Wooding, J. B. 1990. Status, life history, and management of fox squirrels in Florida. Final Performance Report, Florida Game and Fresh Water Fish Commission, Tallahassee, Florida, USA.

PAGE 19

T ABLES AND FIGURES Fig. 1. The range of pelage colorations seen in fox squirrels. The first squirrel on the left exhibits complete melan ism, while the last squirrel is primarily white.

PAGE 20

Fig. 2. The color wheel in which each degree represents a specific hue of color in HSL values. Source: http://dba.med.sc.edu/price/irf/Adobe_tg/models/hsb.html Fig. 3. Saturation and lightness scales using the red hue. Source: https://www.techfry.com/css tutorial/css3 colors rgba hsl and hsla

PAGE 21

Fig. 4. The distribution of the 25 fox squirrels processed in this study across Florida, USA, 2012 2015.

PAGE 22

Fig. 5. Top: S ample pelt from the first cluster of pelages with a Dorsal Hue Coloration of 4 75. Bottom: Sample pelt from the second cluster of pelages with a Dorsal Hue Coloration of 270 300. Table 1. Arthropod ectoparasites found in this study. Arthropod Species Number Present Fleas Orchopeas howardii 11 Lice Neohaematopinus sciurinus Hoplopleura sciuricola 79 277 Ticks Ixodes scapularis Amblyomma maculatum 1 2

PAGE 23

Table 2. Candidate models with color covariates, Akaike's Information Criterion (AICc), difference in AICc ( AICc) between a model and the model with the lowest AICc, Akaike weight (W i ) and number of parameters (k) for the candidate models used to estimate the relationship between pelage coloration and ectoparasite richness. Model AICc AICc W i k Null 57.111 0 .000 0.252 1 Tail Lightness 59.25 0 2.139 0.086 2 Dorsal Hue 59.363 2.252 0.082 2 Tail Hue 59.374 2.264 0.081 2 Dorsal Saturation 59.398 2.287 0.08 0 2 Ventral Saturation 59.457 2.347 0.078 2 Ventral Hue 59.466 2.355 0.077 2 Ventral Lightness 59.472 2.361 0.077 2 Dorsal Lightness 59.482 2.371 0.077 2 Tail Saturation 59.482 2.371 0.077 2 Hue 64.655 7.545 0.006 4 Tail (Hue + Saturation + Lightness) 64.658 7.548 0.006 4 Lightness 64.685 7.575 0.006 4 Dorsal (Hue + Saturation + Lightness) 64.752 7.642 0.006 4 Saturation 64.835 7.724 0.005 4 Ventral (Hue + Saturation + Lightness) 64.903 7.792 0.005 4 Table 3. Candidate models with color covariates, Akaike's Information Criterion (AICc), difference in AICc ( AICc) between a model and the model with the lowest AICc, Akaike weight (W i ) and number of parameters (k) for the candidate models used to estimate the relationship between pelage coloration and tick burden. Model AICc AICc W i k V entral S aturation 19. 34 0 0 .000 0.275 2 Lightness 20.852 1.512 0.129 4 Ventral L ightness 21.088 1.748 0.115 2 Tail S aturation 21.668 2.328 0.086 2 Tail H ue 21.871 2.531 0.078 2 Null 22.282 2.941 0.063 1 Tail L ightness 22.32 0 2.979 0.062 2 Saturation 23.646 4.305 0.032 4 Vent ral (Hue + Saturation + Lightness) 23.712 4.371 0.031 4 Dorsal H ue 23.856 4.516 0.029 2 Dorsal L ightness 24.02 0 4.68 0 0.026 2 Ventral H ue 24.094 4.754 0.026 2 Dorsal S aturation 24.426 5.086 0.022 2 Tail (Hue + Saturation + Lightness) 25.399 6.059 0.013 4 Hue 25.687 6.347 0.012 4 Dors al (Hue + Saturation + Lightness) 28.885 9.544 0.002 4

PAGE 24

Table 4. Candidate models with color covariates, Akaike's Information Criterion (AICc), difference in AICc ( AICc) between a model and the model with the lowest AICc, Akaike weight (W i ) and number of parameters (k) for the candidate models used to estimate the relationship between pelage coloration and flea burden. Model AICc AICc W i k Ventral H ue 38.543 0.000 0.289 2 V entral S aturation 38.913 0.369 0.240 2 Ventral L ightness 39.842 1.299 0.151 2 Hue 40.131 1.588 0.131 4 Lightness 41.140 2.597 0.079 4 Dorsal L ightness 42.650 4.107 0.037 2 Tail (Hue + Saturation + Lightness) 43.380 4.837 0.026 4 Saturation 43.621 5.077 0.023 4 Tail H ue 45.062 6.519 0.011 2 Dors al (Hue + Saturation + Lightness) 45.503 6.959 0.009 4 Dorsal H ue 46.798 8.255 0.005 2 Null 59.286 20.742 0.000 1 Tail L ightness 60.622 22.078 0.000 2 Dorsal S aturation 60.849 22.305 0.000 2 Tail S aturation 61.523 22.979 0.000 2 Table 5. Candidate models with color covariates, Akaike's Information Criterion (AICc), difference in AICc ( AICc) between a model and the model with the lowest AICc, Akaike weight (W i ) and number of parameters (k) for the candidate models used to estimate the relationship between pelage coloration and louse burden. Model AICc AICc W i k Dors al (Hue + Saturation + Lightness) 507.866 0.000 0.684 4 Hue 509.412 1.545 0.316 4 Tail H ue 535.787 27.920 0.000 2 Saturation 544.174 36.308 0.000 4 Tail S aturation 546.828 38.962 0.000 2 Dorsal H ue 557.175 49.309 0.000 2 Vent ral (Hue + Saturation + Lightness) 563.540 55.674 0.000 4 Ventral H ue 576.096 68.229 0.000 2 Tail L ightness 585.308 77.441 0.000 2 Lightness 586.135 78.268 0.000 4 Dorsal L ightness 590.252 82.385 0.000 2 V entral S aturation 591.893 84.026 0.000 2 Null 595.654 87.788 0.000 1 Dorsal S aturation 597.175 89.308 0.000 2 Ventral L ightness 598.025 90.159 0.000 2