Responses of Eastern Gray Squirrels (Sciurus carolinensis) to Predator Calls and Their Modulation by Coat Color.
Eastern gray squirrels (Sciurus carolinensis) are found in two main color morphs: a gray morph and a darker melanistic morph (Koprowski, 1994). Historically both color morphs shared much of the same range. By 1930 the melanistic morph was extirpated from Ohio (Schorger, 1949; Creed and Sharp, 1958). Then, in 1961, fewer than two-dozen melanistic squirrels were imported from London, Ontario, Canada, to the Kent State University campus in Northeastern Ohio (Office of the University Registrar, Kent State University). Since then, the melanistic morph has expanded throughout much of northeastern Ohio. Growing evidence indicates there are strong pleiotropic interactions between coloration in animals and other aspects of their behavior and physiology. Indeed, darker coloration has been linked to increased aggression, resistance to stressors, and increased cardiovascular tone (Pritchard and White, 2007; Ducrest et al., 2008).
Additionally, much work recently has focused on the impacts of urbanized habitats on natural behaviors and ecology (Gilbert, 2012). For example, eastern gray squirrels showed lower "giving up density" when foraging near human-occupied structures (Bowers and Breland, 1996). Similar reduction of antipredator defenses were documented among fox squirrels (Sciurus niger) living in urbanized areas (McCleery, 1990). One previous report on eastern gray squirrels living in an urbanized environment found no difference between color morphs in their response to approach by human and canid predator threats (Gustafson and VanDruff, 1990), but the possibility of differential responses among color morphs to avian predator threats and urban noise stimuli, potential analogs of predator stimuli (Frid and Dill, 2002), have not been reported.
Herein, we report our findings on the association between coat color and antipredator behavior in the eastern gray squirrel. We hypothesized antipredator behavior would be reduced (that is, tolerance to possible threats would be higher) in the melanistic morph relative to the gray morph. Our results expand the range of species and behaviors investigated in association with coat color and contribute to an understanding of the relative success of the melanistic morph in urbanized areas.
All observations conducted in this study took place between July and August 2012 and June and October 2013. This study was conducted with the approval of the Hiram College Institutional Animal Care and Use Committee.
The study took place on Hiram College campus (41[degrees]18'46.8"N, -81[degrees]8'24"W) and surrounding residential village. The area contains many green spaces and multiple tree species. The dominant tree species in this area are: sugar maple (Acer saccharum, 16.8%), white pine (Pinus strobus, 12.7%), pin oak (Quercus palustris, 9.6%), serviceberry (Amelanchie rcanadensis, 8.9%), and Norway maple (Acer platanoides, 7.19%). Eastern gray squirrels (Sciurus carolinensis) in this area are well habituated to human presence.
The main predator known to take squirrels on campus and in the surrounding area is the red-tailed hawk (Buteo jamaicensis), although other predators known to be in the area include the raccoons (Procyon lotor), coyotes (Canis latrans), red foxes (Vulpes vulpes), and domestic cats (Felis catus). As with other native populations in an urbanized habitat (e.g., Gustafson and VanDruff, 1990), the majority (~75%) of eastern gray squirrels found on our campus and in the residential area of Hiram Village have melanistic pelage. This proportion has been stable since censuses were first taken in 2007 (T. Koehnle, pers. Obs.).
The campus and surrounding village were divided in half. Each half of this area had only one of each type of sound played on any given day, to minimize habituation among squirrels not being observed. To further minimize habituation, only two trials were conducted on any given day, one on one side of the study site and the other at the opposite side. Selection of squirrels for observation and placement of the playback equipment was ad lib.
A squirrel was located visually, and the sound system (Bose SoundDock Series II Digital Music System) and iPod (Apple iPod Classic) was then set up with a clear line of sight to the squirrel. The sound system was set up 6 m to 15 m from the squirrel, distances previously determined empirically to provide the closest possible approach with the lowest levels of apparent disturbance to the focal animal (T. Koehnle, pers. obs.). To ensure equal representation of each morph, the color observed in each trial was alternated between trials.
Following this setup, the general color of the squirrel, date, time, location, and weather were recorded. Five minutes were then allowed to pass for habituation of the squirrel to the observer's presence and the recording of some pelage characteristics (vide infra). After these 5 min, a timer was started. During the first 2 min, the number of behaviors of the focal squirrel, consisting of vigilance, freezing, escaping, and tail flagging were recorded. At the 2-min mark one of three sound stimuli--predator call, bird call, or artificial noise--was played for 10 sec outlined below. For 2 min after the sound stimulus, the same antipredator behaviors were recorded. At the end of the observation period, detailed pelage characteristics of the squirrel were recorded so squirrels could be individually identified. In addition, the distance the squirrel was away from the nearest tree at the time the sound was played was measured. Canopy cover at the initial location of the squirrel was recorded using a spherical crown densiometer (Model C, Robert E. Lemmon, Forest Densiometers).
DESCRIPTION OF BEHAVIORS
Several antipredator behaviors were observed and recorded. These included vigilance, freezing, fleeing, and tail flagging. Vigilance was defined as a squirrel exhibiting a bipedal stance while not handling or consuming food. Freezing was defined as a total cessation of movement for 5 sec or more. Fleeing was defined as the squirrel running in any direction, either up a tree or across the ground. Tail flagging was defined as a burst of vertical flips of the tail.
Three different types of sounds were played to the observed squirrels. One type was a nonalarm call of native birds found in this area. The territory call of the American robin (Turdus migratarius) and territory call of the black capped chickadee (Poecile atricapillus) were used (Stokes Field Guide to Bird Songs, Eastern Region and The Birds of North America). The second type of sound was urban noise with no inherent predatory valence. These sounds, a car alarm (SoundJay.com) or a loud buzzer (SoundBible.com), were chosen because actual predator calls used in the other trials might have elicited a ceiling effect, masking real differences between morphs in terms of their boldness. The sounds are representative of typical urban noise stimuli, may be analogous to predator stimuli (Frid and Dill, 2002), and are less novel in this environment than other possible control sounds, such as white noise. The third sound category comprised two exemplars of the red-tailed hawk (Buteo jamaicensis) scream (Stokes Field Guide to Bird Songs, Eastern Region and The Birds of North America). All sounds were played for 10 sec in duration and were of similar sound pressure level, measured 1 m from the front of the speaker (73 [+ or -] 0.8 dB).
Trapping and tagging were not permitted by the campus Institutional Animal Care and Use Committee. Therefore, squirrels were individually identified by pelage characteristics following an adaptation of the coat color typology offered by Creed and Sharp (1958). Using binoculars (8 X 42) and long pre- and post-observation periods permitted detailed descriptions and creation of shaded diagrams using colored pencils. Exhaustive observations were made from a distance between 6 and 15 m, and observation times ran from 5 min before testing until the description was completed after the test, up to 30 min after the end of the testing.
Squirrels were initially categorized as gray or melanistic based on the principal color of the dorsum. Where the dominant color was gray, the squirrel was categorized as a gray morph. All other colors (shades of brown or black) were categorized as melanistic. After categorizing the dorsum, the ventral color was recorded. Then, the dominant color of the dorsal and ventral surface of the tail and dorsal striping of the tail were noted, along with the color of the flanks and presence of striping of the flanks along the dorsal/ventral boundary. Then, the color of the cheek patches and the presence of any stripes or flashes between the cheeks and the blending with the dorsal color were noted. Color patches around the eyes and foreand hindlimb color and color gradients along the medial and lateral surface were noted next. Finally, individual color anomalies (Creed and Sharp, 1958) were noted, including the color of the tip of the nose, the color of the foot pads on the forelimbs, the presence of spots, flashes, flecking, dappling, other unique color markings, loss of all or part of an ear or tail, and altered gait or hair loss indicative of injury or disease.
Color pencil drawings of the pelage and written descriptions were collated and scanned for similarity among squirrels. Individuals were sorted by dorsal, ventral, flank, tail, facial, limb, and anomalous color pattern characteristics. If two squirrels were similar overall with no unique color anomalies or injuries to differentiate them, the observation made first in time was retained in the data set and the other, similar observation (s) were excluded. If two squirrels appeared similar except for an injury (loss of ear or tail), and the first observation was of an uninjured squirrel, the second, injured observation was dropped. However, if there were two similar squirrels and the first observation included an injury, but the second observation appeared unharmed, then both observations were retained in the final data set. For the results reported herein, only data from the first playback instance for any given squirrel were included in the analysis. Therefore, all observations are statistically independent.
To avoid inadvertent inclusion of seasonal changes in pelage in individual descriptions, the squirrels were sampled between June and early October. In the eastern gray squirrel, the spring molt is completed by late May or early June, and the autumn molt begins in early to mid-October (Sharp, 1958). Juvenile gray morphs often have a series of dark stripes or bands along the length of the ventral surface of the tail, which are obscured by the growth of other hairs as the squirrels age (Sharp, 1958). Therefore, all morphs included in the analysis could be differentiated on characteristics other than striping of the ventral surface of the tail.
As the behaviors observed occurred at relatively low frequency during the 2 min before and 2 min after playback, all data were converted into simple presence/absence comparisons. Behavioral data from individual squirrels were entered into a model comparing behaviors before and after playback. The proportion of individuals showing each behavior was compared pre- and post-playback using McNemar's test for correlated proportions. Behavioral data were compared across different types of playback, or between color morphs using the z-ratio to find the significance of the difference between two independent proportions. Canopy cover and distance of the squirrel at initial sighting from the nearest tree were compared using the t-test.
In total, 158 playback sessions were conducted. After application of exclusion criteria, 98 individual observations remained in the data set. There were 35 squirrels (18 gray) in the bird call playbacks, 35 squirrels (18 gray) in the noise playbacks, and 28 squirrels (17 gray) in the predator call playbacks. The latter set is smaller because several dark brown or black melanistic individuals had to be excluded from the set owing to overall visual similarity.
Melanistic morphs fell roughly into the color typology defined by Creed and Sharp (1958). Many squirrels (86%) in the sample fell into the Group 1 category, having rusty, chocolate, or dark brown dorsal or ventral surfaces. Unique brown striping on the dorsum or flanks, along with white, red, or orange patches were common in the sample. Facial anomalies comprised brown patches under the nose, around the eyes, gray or brown stripes along the cheeks, and various ear injuries. Tails ranged from rusty red to chocolate to dark brown or black in these individuals, with light, black, red, or brown fringes, striping, and flecking.
None of the melanistic squirrels in the sample fit exactly within Creed and Sharp's (1958) Group 2: jet black with silver/white flecks on the dorsum or tail. While such squirrels have been seen in the study area, closer inspection reveals that they also often have red or gold flecking or brown patches (4% of the sample population). Indeed, one individual was jet black except for a tail fringed in bright orange. Among the melanistic morphs only 11% could be described as pure jet black, or Group 3 defined by Creed and Sharp (1958). One jet black individual had pale forelimb pads and another had a stubby tail (in the first observation), permitting their identification. One jet black individual's data with no injuries or color anomalies was kept in the first season, and similar squirrels were subsequently excluded from the data set.
Coat color among the gray squirrels was highly variable in terms of the depth of the gray coloration. Many gray morphs had brown flanks, tail, or brown ventral coloration; only 10% had the canonically described pure gray dorsum and pure white ventral surface. Among even these, all individuals had unique color anomalies in the head (white or brown patches around the eyes or ears, different nose colors), tail (brown stripes on the dorsal surface), or injuries which permitted their identification.
Four melanistic morphs had tail injuries. One was jet black but was the first such individual observed. One had an ear injury as well, when first observed. One individual had a brown face, uniquely identifying it. The last one had facial markings that differentiated it. Two melanistic morphs had loss of hair on the tail likely related to mite infection. One had a white spot on the back of its head and the other had a gray patch on its dorsum. As these were unique to these individuals, subsequent regrowth of the tail hair would not have led to their mistaken inclusion in the data set. None of the gray morphs observed had hair loss or tail injuries.
In aggregate, squirrels observed showed increasing tendency to initiate flight movements upon playback of the test sounds as the level of implied threat increased (Fig. 1). Escape behavior was increased in all groups by the initiation of sound playback to some degree. Bird calls caused the lowest proportion of flights (31% of all squirrels, compared to 14% at baseline before the playback), whereas artificial noise caused an intermediate level of flight responses (57% of all squirrels, compared to 14% at baseline) and predator calls induced the highest levels of flight initiation (76% of all squirrels, compared to 4% at baseline). The rate of flight initiation was not different between treatments at baseline (P = 0.25). The proportion of flight responses seen after noise playback was statistically significantly higher than after bird call playback (P = 0.03). Similarly, the proportion of flight responses observed after predator call playback was significantly higher than bird call playback (P = 0.001), although it did not reach significance with respect to the noise playbacks (P = 0.07).
Disaggregating by color (Fig. 2a), it is possible to see how coat color was correlated with flight initiation tendency. In response to bird calls, 35% of melanistic and 27% of gray morphs initiated flight behavior. McNemar's test revealed that neither group's behavior was a significant increase above its own baseline before sound playback (melanistic, P = 1.0; gray, P = 0.625), and the two morphs did not differ from each other (P = 0.632). Noise playbacks initiated a higher overall level of flight behavior (c.f. Fig. 1). In this instance 41% of melanistic and 72% of gray morphs initiated flight behavior. However, neither the melanistic nor gray morph reached significance compared to its own baseline (melanistic, P = 0.125; gray, P = 0.343). Although the gray morphs were more likely to initiate flight, this trend between morphs did not quite reach statistical significance (P = 0.065). Predator calls, however, showed a clear difference between behavior in the two color morphs. Just 45% of melanistic morphs initiated flight in response to a predator call, similar to the noise call (vide supra), whereas 100% of gray morphs did so (P < 0.001). Compared to the preplayback baseline, melanistic morphs did not initiate flight behavior at a higher level (P = 0.5), whereas the gray morphs did so (P < 0.001).
Other trends emerged in freezing behavior (Fig. 2b). In response to bird calls, 6% of melanistic and 0% of gray morphs initiated escape behavior. McNemar's test revealed neither group's behavior was a significant increase above its own baseline before sound playback (melanistic, P = 1.0; gray, P = 1.0), and the two morphs did not differ from each other (P = 0.296). Noise playbacks generated a higher overall level of freezing behavior, with 53% of melanistic and 39% of gray morphs freezing after playback. This was a significant increase in freezing behavior compared to baseline before playback for both morphs (melanistic, P = 0.039; gray, P = 0.031). Although both morphs significantly increased their freezing behavior in response to the call, they did not differ from each other (P = 0.404). In response to predator calls, 36% of melanistic morphs and 41% of gray morphs froze. For the melanistic morphs, there was a slight trend towards increase but it did not reach significance compared to baseline (P = 0.125), whereas for the gray morphs, there was a statistically significant increase in freezing (P = 0.015). Although both groups increased their freezing after playback, the two groups did not differ significantly from each other in response to predator calls (P = 0.799).
Unlike flight initiation and freezing behavior, vigilance was not significantly affected by treatment (Fig. 2c). After playback of bird calls, 46% of squirrels exhibited vigilant postures. Analyzing the data in separate groups divided by color morph, 29% of melanistic morphs exhibited vigilance compared to 61% of grays. This difference between groups was not statistically significant (P = 0.059), and in neither group, was it a significant change from the preplayback baseline (melanistic P = 1; gray P = 0.179). In response to the noise playback, 49% of all squirrels exhibited vigilance (59% of melanistic, 39% of grays). This difference between color morphs was not significant (P = 0.238), and in neither group, was it a significant change from baseline (melanistic P = 0.507; gray P= 1). In response to predator calls, 46% of all squirrels exhibited vigilance after playback (45% of melanistic squirrels, 47% of gray squirrels). These rates of vigilance were not significantly elevated above preplayback baseline in either group (melanistic P = 0.453, gray P = 1), nor compared to each other (P = 0.933).
As with vigilance behavior, tail flagging was not appreciably influenced by call playback (Fig. 2d). In response to bird calls, 37% of all squirrels initiated tail flagging (29% of melanistic and 44% of gray morphs). This small difference between groups did not reach statistical significance (P = 0.357), and neither group was significantly different from its preplayback baseline (melanistic P = 1; gray P = 0.25). Noise playbacks caused 45% of all morphs to tail flag (35% of melanistic and 55% of gray morphs). Though this trend towards higher levels of antipredator behavior in the gray morphs was visible in the data, the two groups did not differ from each other (P = 0.229), and neither group differed from its preplayback baseline (melanistic P = 0.507; gray P = 0.343). Playback of predator calls generated tail flagging in 46% of all morphs (45% of melanistic, 47% of gray). This increase was not different between groups (P = 0.933), nor compared to baseline (melanistic P = 0.062, gray P = 0.453).
Habitat use by the two color morphs did not differ in this study. Melanistic morphs, when initially sighted by observers, were situated in areas with an average of 91.6 [+ or -] 1.9% canopy cover, whereas gray morphs had an average of 87.8 [+ or -] 2.1% canopy cover. These values do not differ statistically (P = 0.182). Similarly, melanistic morphs, when initially sighted for playback, averaged, 2.8 [+ or -] 0.34 m from the nearest tree, whereas gray morphs were 3.38 [+ or -] 0.28 m from the nearest tree, a difference that also is not statistically significant (P = 0.191).
Based on well-documented pleiotropic relationships between pigmentation and boldness (vide infra), we expected to find differences in behavior in response to predator threats between melanistic and gray morphs in our study. In particular, we hypothesized melanistic morphs would show increased boldness in response to predator call playbacks. The study population broadly reflected the color typology set out by Creed and Sharp (1958). They found that 81% of melanistic squirrels fit into their Category 1, dark brown, and only 9% were Category 3, or jet black. We found that 86% fit into Category 1 in our sample, while 11% fit into Category 3. We did not observe any squirrels in Category 2 (10% of Creed and Sharp's sample ), jet black with only white flecking.
This study found urban noise and predator calls caused eastern gray squirrels to initiate flight behavior, with higher levels of escape stemming from perceived predator threat (Fig. 1). Gray morphs were much more likely to escape after predator call playback than melanistic morphs (100% vs. 45%, respectively). Playback of urban noises increased freezing behavior in both morphs. Tail flagging behavior was increased more in response to bird calls and urban noises in gray squirrels than melanistic morphs but not significantly so. McRae and Green (2014) found tail flagging in eastern gray squirrels mostly was directed at terrestrial threats, so the lack of findings here with potential aerial predators may not be surprising. Vigilance behavior did not seem to be modified by playback, at least over the short times observed after playback in this study.
Frid and Dill (2002) theorized urban stimuli are directly analogous to predator stimuli, and the results of this study provide some support for this approach. Human-generated noise playbacks induced antipredator behavior in the sampled population, although not at the same high rate observed for actual predator call playbacks. The higher proportion of melanistic squirrels seen in this and other study sample populations therefore may indicate not only increased boldness, but increased tolerance for such urban stimuli. The difference in propensity to initiate flight in response to predator stimuli may not relate directly to effects of pigmentation on boldness per se, but might instead stem from differential habitat use of the two morphs. Therefore, we found no differences between the morphs in terms of habitat use. Further analysis of only the trials involving squirrels receiving hawk calls revealed a similar homogeneity (melanistic distance from tree 2.7 [+ or -] 0.5 m vs. gray distance 3.1 [+ or -] 0.5 m; P = 0.546). Another interpretation would be that gray morphs are not less bold but simplyless vigilant than melanistic morphs, so are more apt to be surprised by novel or predatory stimuli. However, this can be tested in part from within the existing data set. By combining counts of vigilance across all trials during the preplayback period, we found 31% of melanistic and 34% of gray morphs showed vigilant postures during the 2-min baseline. Therefore, it seems that both habitat use and baseline vigilance are similar overall between gray and melanistic morphs in this sample. On this basis we interpreted the differences in behavior seen in this study as being related to the inherent boldness of the animals and not to other color-related aspects of behavior.
Linkage between boldness and melanism in the eastern gray squirrel previously was explored in the context of simulated terrestrial predatory attack (Gustafson and VanDruff, 1990). In that study, also conducted in a campus environment, no differences were found between the two color morphs in response to simulated attacks by humans, humans with dogs, or dogs alone. These authors discounted the possible relationship between pigmentation and boldness in this species. However, a predator approaching its prey is a clear and present danger, whereas the sound stimuli used in this study merely suggested the possibility of attack. It is possible, therefore, that Gustafson and VanDruff (1990) demonstrated a ceiling effect, in which no differences could be observed owing to the severity or perceived immanence of the threat.
Pigmentation in vertebrates has a very strong genetic component and is genetically linked to expression of other genes and behaviors through the proopiomelanocortin (POMC) system (Ducrest et al., 2008; Roulin and Ducrest, 2011). Pigmentation was linked to a wide variety of other traits, including modulation of inflammatory activity and the HPA/stress response (e.g., Leone et al., 2015; Dores et al., 2014). Pigmentation also was related to energy metabolism and aggression (e.g., Roulin and Ducrest, 2011; Dores et al, 2014; Morgan and Cone, 2006; West and Packer 2002).
Against this background many studies linking pigmentation and boldness have been conducted. In a study of the eastern Hermann's tortoise (Eurotestudo boettgeri), Maflia et al. (2011) found darker individuals had higher levels of aggression and were bolder in the presence of a human. Similarly, Mateos-Gonzales and Senar (2012) found in siskins (Carduelis spinus) with larger and darker bib coloration, latency to approach a novel object was reduced. Among quadruped mammals, darker coat colors have been correlated with increased aggressiveness or boldness in the Korean native jindo dog (Canis lupus familiarisr, Kim et al, 2010), rodents such as deer mice (Peromyscus maniculatus-, Hayssen, 1997), and rats (Rattus norvegiaiy, Cottle and Price, 1987).
Genetic controls of pigmentation in the eastern gray squirrel only recently have been described in detail (Microbie et al, 2009). As of yet, few studies of the possible impacts of different pigmentation genes on behavior in this species have been conducted. Gustafson and VanDruff (1990) commented on the abundance of the melanistic morph in urbanized areas compared to their relative absence in areas with less disturbance. It is possible their relative boldness makes them better adapted to such areas, where natural predators are fewer in number overall due to human habitation. Untangling these relationships will require more intense study.
Acknowledgments.--Many thanks to J. Kastee and V. Hansbro for their assistance in developing early pilot versions of this project. Thanks also to P. Harness for commentary on early drafts. This work was supported in part through the generosity of The Paul and Maxine Frohring Foundation.
Bowers, M. A. and B. Breland. 1996. Foraging of gray squirrels on an urban-rural gradient: use of the GUD to assess anthropogenic impact. Ecol. Applications, 6:1135-1142.
Cottle, C. A. and E. O. Price. 1987. Effects of the nonagouti pelage-color allele on the behavior of captive wild Norway rats (Rattus noivegirus). J. Comp. Psyc., 101:390-394.
Creed, W. A. and W. M. Sharp. 1958. Melanistic gray squirrels in Cameron County, Pennsylvania. J. Mammal., 39:532-537.
Dores, R. M., R. L. Londraville, J. Prokop, P. Dams, N. Dewey, and N. Lesinski. 2004. Molecular evolution of GPCRs: melanocortin/melanocortin receptors. J. Mol. Endocrinol., 52:29-42.
Ducrest, A., L. Keller, and A. Roulin. 2008. Pleiotropv in the melanocortin system, coloration and behavioural syndromes. Trends. Ecol. Evolut., 23:502-510.
Frid, A. and L. Dill. 2002. Human-caused disturbance stimuli as a form of predation risk. Corner. Ecol., 6:11-27.
Gilbert, O. 2012. The ecology of urban habitats. Springer Science Be Business Media.
Gustafson, E.J. and L. W. VanDruff. 1990. Behavior of black and gray morphs of Sciurus carolinensis in an urban environment. Am. Midi. Nat., 123:18(5-192.
Hayssen, V. 1997. Effects of the nonagouti coat-color allele on behavior of deer mice (Peromyscus maniculatus): a comparison with Norway rats (Rattus norvegicus). J. Comp. Psyc., 111:419-423.
Kim, Y. K, S. S. Lee, S. 1. On, J. S. Kim, E. H. Suh, and K. A. Houpt. 2010. Behavioural reactivity of the
Korean native Jindo dog varies with coat colour. Behav. Process, 84:568-572. Koprowski, J.L. 1994. Sciurus Carolinensis. Mam?n. Species., 480:1-9.
Leone, S., G. Noera, and A. Bertolini. 2015. Developments and new vistas in the field of melanocortins. Biomol. Concepts, 6:361-382.
Maflia, A., K. Wakamatsub, and A. Roulina. 2011. Melanin-based coloration predicts aggressiveness and boldness in captive eastern Hermann's tortoises. Anim. Behav., 81:859-863.
Mateos-Gonzalez, F. and J. C. Senar. 2012. Melanin-based trait predicts individual exploratory behaviour in siskins, Carduetis spinus. Anim. Behav., 83:229-232.
Mccleery, R. A. 2009. Changes in fox squirrel anti-predator behaviors across the urban-rural gradient. Landscape Ecology, 24:483-493.
McRae, T. R. and S. M. Green. 2014. Joint tail and vocal alarm signals of gray squirrels (Sciurus carolinensis). Behaviour, 151:1433-1452.
Microbie, H., A. Thomas, and J.O. Kelly. 2009. The genetic basis of melanism in the gray squirrel (Sciurus carolinensis). J. Hered., 100:709-714.
Morgan, C. and R. D. Cone. 2006. Melanocortin-5 receptor deficiency in mice blocks a novel pathway influencing pheromone-induced aggression. Behav. Genet., 36:291-300.
Office of the University Registrvr - Kent State University. A brief history of the black squirrel at Kent State University. Kent State University, [cited 8 June 20101. Available: http://www.registrars. kent.edu/home/site_navs./history.htm.
Pritchard, L. E. and A. White. 2007. Neuropeptide processing and its impact on melanocortin pathways. Endocrinology, 148:4201 -4207.
Roulin, A. and A. L. Ducrest. 2011. Association between melanism, physiology and behaviour: a role for the melanocortin system. Eur. J. Pharm., 660:226-233.
Schorger, A. W. 1949. Squirrels in early Wisconsin. Trans. Wis. Acad. Sri. Arts Lett., 39:195-247.
Sharp, W. M. (1958). Aging gray squirrels by use of tail-pelage characteristics. J. Wildlife Management, 22:29-34.
West, P. M. and C. Packer. 2002. Sexual selection, temperature, and the lion's mane. Science, 297:1330-1343.
Submitted: 7 February 2017
Accepted: 30 June 2017
Department of Entomology and Nematology, 1 Shields Avenue, University of California, Davis 95616
THOMAS J KOEHNLE (1)
Neuroscience Program, Department of Biology, 11715 Carfteld lid, Hiram College, Hiram, Ohio 44234
(1) Corresponding author: e-mail: email@example.com
Caption: Fig. 1.--A higher proportion of squirrels of all color morphs initialed flight behaviors after playback of noise or predator calls than bird calls (*, P < 0.05 compared to bird call). Though more squirrels tended to initiate flight after predator calls than noise, this trend did not reach statistical significance (P = 0.07)
Caption: Fig. 2.--Threatening playback calls (a) caused differential escape responses in the two color morphs of the eastern gray squirrel, Sciurus carolinensis, in Hiram, Ohio. Predator calls caused gray morphs (gray bars) to initiate flight behavior more often than melanistic morphs (black bars, *; P < 0.001) and more often than their own baseline behavior before the playback (see text). A similar trend was evident for noise playback, though it did not reach statistical significance (P = 0.065). Freezing behavior (b) was increased significantly above baseline in both morphs following noise playback (*; melanistic, P = 0.039; gray, P = 0.031). Although both morphs significantly increased their freezing behavior in response to the call, they did not differ from eacli other statistically (P = 0.404). In response to predator calls, only the gray morphs showed a statistically significant increase in freezing from baseline (*; P = 0.015). Playback of the various sound stimuli (c) did not alter the proportion of squirrels adopting vigilant postures in either color morph. Playback of the various sound stimuli (d) did not alter the proportion of squirrels engaged in tail flagging in either color morph
|Printer friendly Cite/link Email Feedback|
|Author:||Bohls, Patricia; Koehnle, Thomas J.|
|Publication:||The American Midland Naturalist|
|Date:||Oct 1, 2017|
|Previous Article:||Understory Vegetation Structure and Soil Characteristics of Geomys pinetis (Southeastern Pocket Gopher) Habitat in Southwestern Georgia.|
|Next Article:||Evidence for Migratory Spawning Behavior by Morphologically Distinct Cisco (Coregonus artedi) from a Small Inland Lake.|