Ecological separation in a polymorphic terrestrial salamander by Cari-Ann Hickerson | Papers by Cari-Ann

Journal of Animal Ecology 2008, 77, 646–653 Blackwell Publishing Ltd doi: 10.1111/j.1365-2656.2008.01398.x Ecological separation in a polymorphic terrestrial salamander Carl D. Anthony1*, Matthew D. Venesky2 and Cari-Ann M. Hickerson3 Department of Biology, John Carroll University, University Heights, OH 44118, USA; 2Department of Biology, John Carroll University, University Heights, OH 44118, USA; and 3Department of Biological, Geological and Environmental Sciences, Cleveland State University, Cleveland, OH 44115, USA 1 Summary 1. When studying speciation, researchers commonly examine reproductive isolation in recently diverged populations. Polymorphic species provide an opportunity to examine the role of reproductive isolation in populations that may be in the process of divergence. 2. We examined a polymorphic population of Plethodon cinereus (red-backed salamanders) for evidence of sympatric ecological separation by colour morphology. Recent studies have correlated temperature and climate with colour morphology in this species, but no studies have looked at differences in diet or mate choice between colour morphs. We used artificial cover objects to assess salamander diet, mating preference and surface activity over a 2-year period at a field site in northeastern Ohio. 3. We detected differences in diet between two colour morphs, striped and unstriped. The diets of striped individuals were significantly more diverse and were made up of more profitable prey than the diets of unstriped salamanders. 4. Opposite sex pairs were made up of individuals of the same colour morph and striped males were found more often with larger females than were unstriped males. 5. We corroborate findings of earlier studies suggesting that the unstriped form is adapted to warmer conditions. Unstriped individuals were the first to withdraw from the forest floor as temperatures fell in the late fall. We found no evidence that the colour morphs responded differently to abiotic factors such as soil moisture and relative humidity, and responses to surface temperatures were also equivocal. 6. We conclude that the two colour morphs exhibit some degree of ecological separation and tend to mate assortatively, but are unlikely to be undergoing divergence given the observed frequency of intermorph pairings. Key-words: climate change, colour polymorphism, mate choice, Plethodontidae, sympatric speciation Introduction Understanding how new species form is a fundamental goal of evolutionary biology, and sympatric speciation (the divergence of species without geographical isolation) has recently received considerable attention, both theoretical and empirical (Maynard Smith 1966; Kondrashov & Mina 1986; Rice & Salt 1990; Dieckmann & Doebeli 1999). Population divergence that occurs in sympatry is thought to be maintained, in part, by non-random mating (see Bagnoli & Guardiani 2005). For example, the ecological speciation hypothesis predicts that divergent selection on traits can lead to the evolution of reproductive isolation (Schluter 2001) and studies have *Correspondence author. E-mail: canthony@jcu.edu provided evidence that reproductive isolation can evolve in the presence of gene flow (Maynard Smith 1966; Robinson 2000; Hollander, Lindegarth & Johannesson 2005). Reproductive isolation among individuals in sympatry has the potential to occur as populations specialize on a particular resource, such as food (Skúlason, Snorrason & Noakes 1993) or habitat (Danley et al. 2000; Takahashi 2004). Adaptive trade-offs between different morphologies within a single population may further drive this divergence (Robinson 2000). Reproductive isolation can also occur if different forms vary temporally in their breeding activity (Whiteman & Semlitsch 2005). In addition, differential predation pressure may lead to reproductive isolation by selecting for correlated characters, such as colour pattern and preference for predatorfree microhabitat (Nosil 2004). © 2008 The Authors. Journal compilation © 2008 British Ecological Society Ecological separation in a salamander 647 Polymorphic species provide a good model system for studying speciation because different phenotypes can be expressed in a single population, each facing different selection pressures (West-Eberhard 1986). Thus, polymorphic species can provide a way to study speciation as reproductive isolation emerges within a population (Whiteman & Semlitsch 2005). Plethodon cinereus Green 1818 is a common terrestrial salamander found throughout eastern North America and Canada. Individuals of P. cinereus exhibit colour polymorphism and have two common colour phases: the striped phase and unstriped phase (Pfingsten & Walker 1978). The striped phase (the typically more abundant of the two phases) has a red dorsal stripe, whereas the unstriped phase (also called the ‘lead phase’) lacks a red dorsal stripe and has a completely black dorsum. Highton (1959) determined that the colour polymorphism in P. cinereus has a genetic basis and Highton (1975) argued that multiple pairs of genes probably interact to produce the different colour morphs. Striped/unstriped dimorphism occurs in at least eight species of Plethodon (Petranka 1998), yet we know little with regard to the function of the dimorphism. A number of studies have correlated warmer and dryer climates with higher frequencies of the unstriped morph of P. cinereus (Burger 1935; Test 1952; Williams, Highton & Cooper 1968; Lotter & Scott 1977), and in a recent study (Gibbs & Karraker 2006) global climate change and forest disturbance were implicated in an observed increase in the frequency of the unstriped form throughout the species range. In addition to providing strong evidence of the effects of climate change on the evolution of polymorphism, Gibbs & Karraker (2006) outlined three lines of evidence that suggest that colour polymorphism responds to temperature-related selection: (1) increased mortality in colder climates for the unstriped morph (Lotter & Scott 1977; Moreno 1989); (2) early retreat by unstriped morphs to underground refugia as the temperature cools in the autumn (Lotter & Scott 1977; Moreno 1989); and (3) possession by unstriped morphs of a standard metabolic rate that allows activity in warmer and drier conditions (Moreno 1989; Petruzzi, Niewiarowski & Moore 2006). Although there is growing evidence suggesting that climate and temperature play a role in the maintenance of colour polymorphism in P. cinereus, other selective forces have not been ruled out. For example, Moreno (1989) reported that unstriped individuals of P. cinereus had a significantly higher incidence of tail breakage (a common measure of predation pressure) compared to striped individuals. Venesky & Anthony (2007) reported similar differences in tail breakage between colour morphs and also found differences in behaviour between morphs when encountering a snake predator in laboratory arenas. Trophic polymorphism has been reported in adjacent populations of this species (Adams & Rohlf 2000; Maerz, Myers & Adams 2006), but the effect that colour morphology has on diet is unknown. In the current study, we sought to use observations taken over a 2-year period at a single collection site to evaluate whether there was any evidence for ecological separation of the two colour morphs. To determine if the colour morphs exhibit potential for reproductive isolation, we collected data on the surface activity patterns (date of activity and temperature at which the salamander was found) of 1095 individuals of P. cinereus. If the colour morphs are differentiated seasonally, we expected that they would be active at different times of the year and that they would be active at different temperatures. If activity is dependent upon temperature and differs between colour morphs, we predicted that the two morphs might forage in different microhabitats or at different times of day and thus have the opportunity to select different prey types. We tested this hypothesis by obtaining the gut contents of 81 individuals of P. cinereus from one date in our field survey. Finally, we examined 94 male–female pairs for evidence of assortative mating by colour morphology in an effort to detect reproductive isolation between the two colour morphs. Materials and methods On 12 and 13 April 2004, we placed 288 artificial cover objects (ACOs) on the forest floor in the Cuyahoga Valley National Park (CVNP), Ohio (41°13′ 46·62′ N, 81°31′ 7·77′ W). The field site lies on a north-facing slope (elevational range 260–271 m) and is dominated by Acer saccharum (sugar maple), Fagus grandifolia (American beech), Liriodendron tulipifera (tulip poplar) and Quercus rubra (red oak). We used white ceramic floor tiles measuring 30·5 × 30·5 cm as ACOs. ACOs were arranged in 32 arrays of nine tiles each; and each tile was separated by a distance of 1 metre. The arrays were placed as part of a larger study examining the regulatory role of P. cinereus in the forest floor food web. The design of this larger study entails the removal and permanent displacement of red-backed salamanders from half (n = 144) of the ACOs. To assess seasonal activity and response to abiotic factors of the two colour forms, we used only data collected from the 144 removal ACOs. When collecting data on mate preference, we utilized all 288 ACOs, but we photo-checked salamanders found repeatedly within the same arrays. This ensured that each salamander was counted only once. Beginning on 30 September 2004, and on each subsequent sampling date (to 29 September 2006), we turned each ACO by hand and recorded the colour morphology (striped (red) or unstriped (lead)), sex, mass, snout–vent length (SVL to nearest 0·01 mm) and number of P. cinereus under each object. We also recorded the ground surface temperature, soil temperature and air temperature at the central ACO in each nine-tile array (32 points spaced evenly throughout the site). We used an infrared temperature sensor (Cole-Parmer Instrument Company, IL, USA) to measure surface temperature of the ground under the ACO. Soil moisture was taken at 10 cm below the soil surface with a Kelway soil moisture meter (94302, Forestry Suppliers, Inc, MS, USA). Data collection usually started between 0800 and 0900 h and sometimes lasted until 1600–1700 h. We visited the arrays in random order to remove any temporal bias in sampling. We sampled the site every 2 weeks in 2004 and 2005 and every week in 2006. Sampling artificial cover at weekly frequencies does not appear to affect cover quality negatively (Marsh & Goicochea 2003). During 2006, abiotic measurements were taken every other week. We assumed that when an adult male and female red-backed salamander shared an ACO, they were a mated pair because previous studies (Gillette, Jaeger & Peterson 2000; Peterson 2000; Jaeger, Gillette & Cooper 2002) have used a maximum intrapair distance of 30 cm to define mated pairs of this species. Our cover objects were © 2008 The Authors. Journal compilation © 2008 British Ecological Society, Journal of Animal Ecology, 77, 646–653 648 C. D. Anthony et al. only 30·5 cm2, thus it is unlikely that two adults cohabitating beneath an ACO were not aware of each other’s presence. Additionally, adults of this species are territorial (Jaeger 1984; Mathis et al. 1995), including those from north-eastern Ohio (Gall, Anthony & Wicknick 2003; Hickerson, Anthony & Wicknick 2004; this study) and do not otherwise share cover objects as small as the ones used in this study (Mathis 1990; but see Quinn & Graves 1999; Maerz & Madison 2000). For example, of 518 adult salamanders observed under the 144 removal ACOs at our field site, 438 occurred alone and 74 were found paired with another adult. Of the 74 paired individuals, only six were found in same-sex pairs (three female–female pairs). We found no instances of males co-occurring and only two cases where three adults co-occurred. To determine sex, we examined closely the shape of the snout for differences in shape and size. Adult males of P. cinereus exhibit an enlarged snout when in reproductive condition; the snout of females appears blunt relative to a reproductively active male. We counted only pairs that included a male in obvious reproductive condition. We failed to determine reproductive status of females. Under every ACO that yielded a presumptive mated pair, we recorded the colour morphology and sex of each adult salamander. For analysis, we excluded the three cases where more than two adults were found together. We used a G-test of independence to test the hypothesis that individuals of P. cinereus do not mate assortatively based upon colour morphology. We used residuals from an ordinary least squares regression as an index of body mass for a given size (SVL) for males and females (Schulte-Hostedde et al. 2005). The residuals were then used in a mixed-model analysis of variance (ANOVA) with male and female size as paired dependent variables and colour pair category (1, striped male and female; 2, striped male–unstriped female; 3, unstriped male–striped female; and 4, unstriped male and female) as the independent variable to examine whether differences in sizes between sexes differ among the four colour pair categories. We predicted that successful males would be paired with females that were significantly larger than themselves because fecundity is correlated positively with female size in P. cinereus (Nagel 1977; Lotter 1978). We used  for Windows, version 13 for the statistical analyses. To examine differences in prey consumed by the two colour morphs, we collected gut contents from salamanders observed under the ACOs on a single sampling day (7 October 2005). Gut contents were extracted by flushing the gut with water (Fraser 1976; Mitchell, Wicknick & Anthony 1996) and prey items were extracted as salamanders were observed between 0930 and 1630 h. We transferred gut contents to individually marked vials and identified prey to order or some cases family (e.g. Formicidae). Importance values (Ix) (Powell et al. 1990) were calculated for each prey taxon taken by striped and unstriped salamanders using the equation: Fig. 1. Body size comparisons among different colour pairs. Body size indices of all possible colour/gender combinations as indicated by the residuals from ordinary least squares regression of body mass and snout–vent length. White salamander symbols: red-striped salamanders; black symbols: unstriped salamanders. used the Shannon diversity index (H′) and compared the indices with a two-tailed t-test. Univariate regression analysis was used to examine the relationship between numbers of salamanders observed and each of the following variables: collection date, surface temperature, relative humidity and soil moisture. We used t-tests to compare mean values at which the different colour morphs were collected. We employed two-tailed tests except when examining temperature because previous studies have established that the unstriped morph prefers warmer temperatures (Test 1952; Lotter & Scott 1977; Moreno 1989). Results MORPH-SPECIFIC SALAMANDER PAIRINGS Ix = [(nx /N ) + (vx /V ) + ( fx /F )]/3 where nx, vx and fx represent the number, volume and frequency (number of guts containing that prey) of each prey type and N, V and F represent the sums of those values across all guts. Prey volumes were estimated with the equation for an ellipsoid using length and width measurements made on each prey item. The index, Ix, ranges from 0 to 1 and provides a measure of the importance of each prey category in an organism’s diet (Anderson & Mathis 1999). We compared numbers of the six most important prey taxa using a χ2 test and we used t-tests (two-tailed) to compare mean numbers of ants, collembolans and mites consumed by the two colour morphs. Alpha was reduced to 0·017 to account for multiple comparisons. To estimate the prey niche breadth of the two colour morphs, we Over the five field seasons, we observed 94 reproductive pairs of adult P. cinereus. Of the 94 pairs, 68 pairs were same-colour pairs and 26 non-matching colour-pairs were observed. Of the 68 same-colour pairs, eight were unstriped pairs. Of the 26 non-matching pairs, approximately equal numbers were composed of unstriped males paired with striped females (n = 12) and striped males paired with unstriped females (n = 14). Males were more likely to be found with females of the same colour morphology (G statistic = 6·74; P = 0·058). Although striped and unstriped males did not differ in size, the females that they were paired with did differ (Fig. 1). We detected significant size differences between males and females in different colour combinations (i.e. significant sex × colour morphology interaction; F = 8·25, d.f. = 3, P < 0·001; Fig. 1). The largest females were found paired with striped males and the smallest females were found with unstriped males (Fig. 1). Of the unstriped females, only the largest were found with striped males. The smallest females were unstriped and found only with males of the same colour morphology (Fig. 1). SALAMANDER GUT CONTENTS On 7 October 2005, we collected the gut contents of 81 P. cinereus (striped n = 67; unstriped n = 14). The gut contents contained © 2008 The Authors. Journal compilation © 2008 British Ecological Society, Journal of Animal Ecology, 77, 646–653 Ecological separation in a salamander 649 Fig. 2. Diet differences between striped (n = 67) and unstriped (n = 14) salamanders collected 7 October 2005. Striped salamanders ate significantly more Collembola and had a significantly more diverse diet consisting of more profitable prey. The peak in larval Diptera consumption resulted from one unstriped salamander consuming 60% of all fly larvae. Table 1. Importance values (Ix) for dominant prey categories of striped and unstriped colour morphs of the red-backed salamander, Plethodon cinereus. Importance values (bold type) were calculated from (in parentheses) the total number of each prey type (nx), the total volume of each prey type (vx) and the frequency ( fx) of each prey type Striped morph (n = 67) 0·325 (124, 12·83, 43) 0·319 (95, 67·22, 41) 0·272 (83, 104·93, 30) 0·098 (19, 41·59, 12) 0·199 (37, 102·86, 23) 0·046 (7, 8·73, 7) Unstriped morph (n = 14) 0·278 (12, 1·82, 8) 0·086 (6, 0·61, 2) 0·451 (15, 21·06, 10) 0·184 (6, 9·10, 4) 0·129 (3, 6·56, 3) 0·124 (13, 0·77, 2) Prey taxon Acari Collembola Formicidae Isopoda Diptera (adults) Diptera (larvae) Fig. 3. Surface activity of striped (open circles, dashed line) and unstriped (closed circles, solid line) colour morphs of red-backed salamanders as it relates to (a) relative humidity, and (b) percentage soil moisture. Relative humidity – striped: R2 = 0·775, F1,15 = 46·123, P < 0·0001; unstriped: R2 = 0·734, F1,15 = 29·774, P < 0·0001. Percentage soil moisture – striped: R2 = 0·651, F2,7 = 6·543, P < 0·025; unstriped: R2 = 0·747, F2,7 = 10·334, P < 0·008. Note that the scales of the right and left axes differ. two colour forms. Striped salamanders had significantly more diverse diets compared to unstriped salamanders (striped H′ = 2·70, unstriped H′ = 2·39, d.f. = 25, t = 15·5, P < 0·001). a total of 489 prey items representing 14 categories (Fig. 2). The diet of P. cinereus was made up primarily of Acari (28%), Collembola (21%), Hymenoptera (ants; 20%), Diptera (7·1%) and Isopoda (5·5%). There was no significant difference in total volume of prey between the two colour morphs (mean prey volume gut−1: striped = 9·80 mm2, unstriped = 4·36 mm2; Z = 1·58, P = 0·114, two-tailed), but striped individuals tended to have more prey in their guts than did unstriped salamanders (mean number prey gut−1: striped = 6·36, unstriped = 4·50; Z = 1·84, P = 0·066, two tailed). Importance values indicated that the dominant prey categories differed between colour morphs (d.f. = 5, χ2 = 58·2, P < 0·001; Table 1). The diet of striped morphs was dominated by mites, Collembola and ants (with I-values of 0·325, 0·319 and 0·272, respectively). However, for unstriped morphs, ants were the dominant category (I = 0·451) followed by mites (I = 0·279) and isopods (I = 0·184). Collembola were not an important prey category for unstriped morphs (I = 0·086) and these salamanders consumed significantly fewer Collembola than did striped morphs (P = 0·010, two-tailed). Diet differences were also reflected in estimates of niche diet breadth for the ABIOTIC AND SEASONAL EFFECTS ON SURFACE ACTIVITY The percentage of unstriped individuals active at the surface ranged from 10% to 35·3% at our field site. Striped and unstriped salamanders were active on the forest floor under similar relative humidity and percentage soil moisture conditions (Fig. 3). There were no significant differences in mean relative humidity (striped = 71·5%; unstriped = 69·2%; t = 1·28, P = 0·22, two-tailed) or mean soil moisture (striped = 65·9%; unstriped = 66·9%; t = 1·03, P = 0·30, two-tailed) at which the two forms were collected. Over all seasons, the two morphs did not differ significantly in their surface activity relative to soil temperature (mean for striped = 13·6 °C; unstriped = 13·7 °C; t = 0·31, P = 0·76, two-tailed), but unstriped individuals withdrew from the forest floor first as the autumn season progressed (Fig. 4). Comparisons made within season returned equivocal results. In spring and early summer we found no significant difference in surface temperature at which the two forms were active (Julian dates 92–210, © 2008 The Authors. Journal compilation © 2008 British Ecological Society, Journal of Animal Ecology, 77, 646–653 650 C. D. Anthony et al. Fig. 4. Surface activity of striped (open circles, dashed line) and unstriped (closed circles, solid line) colour morphs of red-backed salamanders by season (as expressed by Julian date). Striped: R2 = 0·455, F3,42 = 11·700, P < 0·0001; unstriped: R2 = 0·517, F3,42 = 14·990, P < 0·0001. As the autumn season progressed, unstriped (lead) phase individuals withdrew from the forest floor at a faster rate than did striped (red) phase individuals. Note that the scales of the right and left axes differ. striped = 14·01 °C; unstriped = 13·51 °C; t = 1·09, P = 0·14, one-tailed, Fig. 5), nor was there a difference in activity in the late summer and autumn (Julian dates 214–336, striped = 13·22 °C; unstriped = 13·78 °C; t = 1·32, P = 0·094, one-tailed; Fig. 5). Discussion Reproductive isolation is the driving force behind speciation and ultimately generates biological diversity. Ecological speciation theory predicts that reproductive isolation can occur between groups when natural selection favours alternate phenotypes (Schluter 2001). These phenotypes may exist in a single population where resource polymorphisms (Smith & Skúlason 1996) arise as a result of divergent selection. We monitored a polymorphic population of red-backed salamanders over a 2-year period for evidence of ecological separation along several niche dimensions. The differences in temperature-dependent seasonal activity, diet and mating preferences that we detected between the two colour forms suggest that there is potential for environmental and genetic isolation between the two colour morphs of P. cinereus. Our data corroborate previous studies illustrating that the colour morphs of P. cinereus differ in their response to climate in their seasonal activity patterns (Greer 1973; Lotter & Scott 1977; Moreno 1989; Gibbs & Karraker 2006). Salamanders have an ecological thermal range that influences their activity (Zweifel 1957) and they can regulate the temperature to which they are exposed by moving vertically within the soil horizon. Animals move deeper into the soil during the winter to stay warm and during the summer to stay cool (Vernberg 1953). Thus, relative numbers of the different colour morphs active at the surface should reflect temperature preference of those morphs. As predicted, we found that unstriped salamanders retreated from the surface earlier than did their striped counter- Fig. 5. Surface activity (cubic regression) of striped (open circles, dashed line) and unstriped (closed circles, solid line) colour morphs of red-backed salamanders as it relates to surface temperature taken under artificial cover objects. Spring and summer (Julian dates 92– 209) – striped: R2 = 0·454, F3,31 = 8·583, P < 0·0001; unstriped: R2 = 0·355, F3,31 = 8·408, P < 0·003. Summer and autumn (Julian dates 210–336) – red phase: R2 = 0·394, F3,33 = 7·148, P < 0·001; unstriped: R2 = 0·338, F3,33 = 5·619, P < 0·003. Note that the scales of the right and left axes differ. parts, but the temperatures at which they were active did not differ. These results mirror those reported by previous workers (Lotter & Scott 1977; Moreno 1989). However, in a recent study of two populations near to our study site, Petruzzi et al. (2006) reported a higher substrate temperature for the striped morph (at one locality) and a decrease in the surface activity of striped morphs relative to unstriped forms in the autumn season. They argued that the relationship between temperature and colour morph frequency is more complex than thought previously as it differed between seasons and localities in their study. We agree with this assessment. Despite substantial sample sizes (n = 1095, this study; n = 993, Petruzzi et al. 2006) and populations in close proximity (4·0–9·3 km apart), two research teams report different responses of polymorphic P. cinereus to temperature. These inconsistencies may result from the complex interactions between salamander thermal and hydric requirements in different microhabitats. For example, numerous laboratory studies have illustrated that plethodontids respond to temperature gradients (e.g. Spotila 1972), but detecting differences in the field may be difficult because lungless salamanders are restricted to moist environments (Feder & Pough 1975) that probably lack thermal diversity (Feder 1982). © 2008 The Authors. Journal compilation © 2008 British Ecological Society, Journal of Animal Ecology, 77, 646–653 Ecological separation in a salamander 651 An alternative explanation for differences among morphs may lie in selective differences between localities. For example, the presence of different predators in different seasons and at different localities could influence the surface activity of the colour morphs. A number of studies have suggested that, in P. cinereus, differential predation of colour morphs occurs. In an experiment examining survivorship between two colour morphs of P. cinereus (all-red (erythristic) and striped), striped morphs were eaten by birds significantly more often than all-red morphs (Brodie & Brodie 1980). This is due presumably to Batesian mimicry of toxic red-spotted newt efts (Notophthalmus viridescens) by all-red morphs of P. cinereus (Tilley, Lundrigan & Brower 1982; Cassell & Jones 2005). Additionally, Lotter & Scott (1977) and Moreno (1989) both discuss the possibility of disproportional predation rates between the striped and unstriped colour morphs of P. cinereus based on differences in tail breakage between the two forms and Venesky & Anthony (2007) found that striped and unstriped individuals of P. cinereus respond differently to snake predators in experimental arenas. It is likely that the presence of visually orientated (bird) predators or chemically orientated (snake) predators at different sites has the potential to effect surface activity and possibly local adaptation by populations at different sites. Our estimates of diet composition in P. cinereus are similar to those reported in previous studies. Mites (28%), springtails (21%), ants (20%), flies (7·1%) and isopods (5·5%) made up the bulk of the diet at our site. Jaeger (1990) listed mites (50%), flies (21%), springtails (13%) and ants (5·5%) to be the most common prey by number in the guts of 519 P. cinereus in Virginia. Maglia (1996) reported mites (37%), ants (21%), unidentified larvae (16%) and springtails (10%) as the most common prey in 192 guts of P. cinereus from Tennessee. Adams & Rohlf (2000) reported mites, ants and beetles as the most common prey taken by P. cinereus in Pennsylvania and Maerz et al. (2006) listed mites, spiders, springtails and beetles among 12 common prey types taken by P. cinereus in New York and Pennsylvania. Salamanders at our site in northeastern Ohio have similar diets, but we also found evidence that the colour morphs have diverged in their prey base with striped salamanders consuming a higher diversity of prey that appear to be of higher quality. Admittedly, our survey was not extensive and represents only a snapshot of the gut contents of one field season. However, analysing diet samples collected on a single day (Arif, Adams & Wicknick 2007) removes the confounding effects of rainfall and season on invertebrate abundance and may provide a more accurate picture of prey preferences when making comparisons between groups within a single population. At our field site, the diets of unstriped salamanders were dominated by ants and they consumed significantly fewer springtails than did striped individuals. Jaeger (1990) argued that smaller, lightly armoured prey (such as springtails) are the preferred prey of P. cinereus during wet foraging periods, but that larger, heavily armoured prey (such as ants) are consumed on dry days and nights when foraging activity is limited. Salamanders prefer lightly armoured prey items because they pass rapidly through the gut, thus maximizing net energy gained per unit time (Jaeger 1990). Because we obtained our diet samples on a single day, the differences in diet between the two colour morphs suggest that unstriped salamanders preyed on less profitable prey compared to striped salamanders. Previous studies have shown that male–female pairs of P. cinereus share a single territory even outside the courtship season (Jaeger et al. 1995). Single pairs are maintained through a form of social monogamy (Gillette et al. 2000) enforced through sexual intimidation by males (Jaeger et al. 2002) and by females (Prosen, Jaeger & Hucko 2006). Thus, we predict that high-quality males and females should be paired with high-quality mates. The size differences between pairs that we report suggest that striped (red) males have greater success attracting larger and presumably more fecund females. Such size differences could arise from differences in rates at which reproductive maturity is reached, if striped females reach maturity at a larger size than do unstriped females. Under this scenario striped males would gain access to larger females as a byproduct of assortative mating by colour. An alternative possibility is that striped males have diets and territories that are attractive to females. Female red-backed salamanders are able to assess the diet of conspecifics through the examination of faecal material (faecal squashing; Walls et al. 1989; Karuzas, Maerz & Madison 2004) and are presumably attracted to areas where nutritious prey are readily available. Our diet data suggest that red striped salamanders prey on more nutritious prey than do unstriped salamanders and we speculate that this may make striped males, and/or their territories, more attractive to females. At our study site, male and female salamanders of the same colour morphology are found together more often than expected, providing evidence of positive assortative mating in this population of P. cinereus. Positive assortative mating can arise through a number of mechanisms (Jawor et al. 2003). It can occur when one or both sexes attempt to gain access to the highest-quality mates which results automatically in low-quality individuals pairing with one another because of competitive exclusion from high-quality mates (Johnstone, Reynolds & Deutsch 1996). Intrasexual competition for territories can also result in a positive pattern of mating assortment (Creighton 2001). In this case, males and females of similar competitive ability might occupy the same cover object and associated territory. Intrasexual competition is well documented in P. cinereus, where males in some populations compete for high-quality territories and expel other males from those territories (Jaeger 1984; Mathis 1991). Females are attracted to male territories and will attempt to expel intruding females as well (Lang & Jaeger 2000). At our site, the smallest females were found with unstriped males. Although we do not have sufficient data to analyse diet differences by sex, it is possible that assortative mating in this population results from high-quality diet cues of striped males attracting high-quality females (most of which are striped). These females are then able to exclude intruding females from the shared territory so that only the largest unstriped females gained access to striped males (Fig. 1). © 2008 The Authors. Journal compilation © 2008 British Ecological Society, Journal of Animal Ecology, 77, 646–653 652 C. D. Anthony et al. Whatever the mechanism involved, assortative mating in an ecologically differentiated, polymorphic species provides the potential for reduced gene flow between ecomorphs and subsequent genetic divergence, even in sympatry (Hendry et al. 2000; Jones et al. 2003). For example, studies by Whiteman & Semlitsch (2005) and Whiteman, Krenz & Semlitsch (2006) have provided evidence that reproductive isolation can occur in sympatry between facultatively paedomorphic and metamorphic mole salamanders (Ambystoma talpoideum). For these salamanders, ecological differentiation is profound because adults occupy distinctly different niches and are exposed to very different selective regimes. Hence, divergence might occur even in the face of relatively high frequency of intermorph pairings. Red-backed salamanders at our site occupy the same habitat but still show some degree of ecological differentiation. They exhibit morph-specific differences in prey use even though they are exposed to a similar terrestrial prey base, and they exhibit differences in seasonal activity. In the northern part of its geographical range, P. cinereus might be especially prone to niche expansion and sympatric divergence. Compared to other species of Plethodon, P. cinereus has had greater success invading previously glaciated areas (Petranka 1998) where it is often the most common species (Burton & Likens 1975). At our field site, P. cinereus make up greater than 99% of all caudate species observed. Species-poor environments characterized by a lack of interspecific competitors and an abundance of available niches are thought to favour divergent selection (Robinson 2000; Bolnick et al. 2003). For example, the evolution of trophic specialization between adjacent upland and lowland populations of P. cinereus in the postglacial north-eastern United States may have been facilitated by a lack of interspecific competition (Maerz et al. 2006). Our estimates of dietary niche breadth (2·70 for striped and 2·39 for unstriped) are consistent with this model. These estimates are higher than those reported for other populations of P. cinereus (range 2·08–2·18; Jaeger 1981) and the magnitude of difference that we report between the two morphs at our locality (0·31) is higher than that reported among localities for P. cinereus (Jaeger 1981), suggesting an overall expansion of dietary niche coupled with limited separation between the two colour morphs. At our study site, 38·2% of pairings were between striped and unstriped salamanders and we doubt that, at this level of gene flow, genetic divergence is occurring unless strong selective differences between the two morphs exist. With continued global climate change, we may begin to see such differences in selective regime between the two morphs. For example, Gibbs & Karraker (2006) report a 6% decrease in relative abundance of striped P. cinereus over the last century and they attribute this decline to forest disturbance and increasing global temperatures. Changing global temperatures could provide an unexploited niche available to either unstriped salamanders as microhabitats warm or striped salamanders as winters become increasingly mild. Future studies that establish rates of intermorph gene flow or examine behavioural components of mate choice (Whiteman et al. 2006) would be useful in determining the likelihood of future divergence in polymorphic red-backed salamanders. Acknowledgements We thank J. Keiper of The Cleveland Museum of Natural History and The Ohio Conservation Alliance for funding portions of this study. CMH was supported by a NSF Doctoral Dissertation Improvement grant (DEB-0608239) and a Cleveland State University Doctoral Dissertation Research Expense award. CDA was supported by a John Carroll University faculty summer fellowship and a George Grauel Faculty Fellowship. The Cuyahoga Valley National Park granted permission to conduct fieldwork and provided access to the Woodlake Environmental Field Station. Fieldwork was conducted under National Park Service scientific research permit number CUVA-2004-SCI0010 and work with live animals was conducted in accordance with JCU IACUC protocol number JCU503 and CSU IACUC protocol number 2604WAL-AS. We also thank J. Johansen and M. Walton for statistical advice. We gratefully acknowledge O. Lockhart, K. Thompson and M. Walton for flushing salamander guts and identifying and measuring invertebrates. J. Loew assisted in field site set-up and data collection. Comments by D. Adams and an anonymous reviewer greatly improved our paper. The work presented here complies with the current laws of the United States and of the state of Ohio. References Adams, D.C. & Rohlf, J.F. (2000) Ecological character displacement in Plethodon: biomechanical differences found from a geometric morphometric study. Proceedings of the National Academy of Sciences, 97, 4106–4111. Anderson, M.T. & Mathis, A. (1999) Diets of two sympatric neotropical salamanders, Bolitoglossa mexicana and B. rufescens, with notes on reproduction for B. rufescens. Journal of Herpetology, 33, 601–607. Arif, S., Adams, D.C. & Wicknick, J.A. (2007) Bioclimatic modeling, morphology, and behavior reveal alternative mechanisms regulating the distributions of two parapatric salamander species. Evolutionary Ecology Research, 9, 843–854. Bagnoli, F. & Guardiani, C. (2005) Sympatric speciation through assortative mating in a long-range cellular automaton. Cellular Automata, Proceedings Lecture Notes in Computer Science, 3305, 405–414. Bolnick, D.I., Svanbäck, R., Fordyce, J.A., Yang, L.H., Davis, J.M., Hulsey, C.D. & Forister, M.L. (2003) The ecology of individuals: incidence and implications of individual specialization. American Naturalist, 161, 1–28. Brodie, E.D. Jr & Brodie, E.D. III (1980) Differential avoidance of mimetic salamanders by free-ranging birds. Science, 208, 81–182. Burger, J.W. (1935) Plethodon cinereus (Green) in eastern Pennsylvania and New Jersey. American Naturalist, 64, 578–586. Burton, T.M. & Likens, G.E. (1975) Salamander populations and biomass in the Hubbard Brook Experimental Forest, New Hampshire. Copeia, 1975, 541–546. Cassell, R.W. & Jones, M.P. (2005) Syntopic occurrence of the erythristic morph of Plethodon cinereus and Notophthalmus viridescens in Pennsylvania Northeastern Naturalist, 12, 169–172. Creighton, E. (2001) Mate acquisition in the European blackbird and its implications for sexual strategies. Ethology, Ecology and Evolution, 3, 247 – 260. Danley, P.D., Markert, F.A., Arnegard, M.E. & Kocher, T.D. (2000) Divergence with gene flow in the rock-dwelling cichlids of Lake Malawi. Evolution, 54, 725–1737. Dieckmann, U. & Doebeli, M. (1999) On the origin of species by sympatric speciation. Nature, 400, 354–357. Feder, M.E. (1982) Thermal ecology of neotropical lungless salamanders (Amphibia: Plethodontidae): environmental temperatures and behavioral responses. Ecology, 63, 1665–1674. Feder, M.E. & Pough, F.H. (1975) Temperature selection by the red-backed salamanders, Plethodon cinereus (Green) (Caudata: Plethodontidae). Comparative Biochemistry and Physiology, 50A, 91–98. Fraser, D.F. (1976) Empirical evaluation of the hypothesis of food competition in salamanders of the genus Plethodon. Ecology, 57, 458–471. Gall, S.B., Anthony, C.D. & Wicknick, J.A. (2003) Behavioral interactions between salamanders and beetles indicate a guild relationship. American Midland Naturalist, 149, 363–374. Gibbs, J.P. & Karraker, N.E. (2006) Effects of warming condition in eastern North American forests on red-backed salamander morphology. Conservation Biology, 20, 913–917. Gillette, J.R., Jaeger, R.G. & Peterson, M.G. (2000) Social monogamy in a territorial salamander. Animal Behaviour, 59, 241–1250. Greer, A.E. (1973) Adaptive significance of the color phases of the red backed salamander. Yearbook of the American Philosophical Society, 973, 308–309. © 2008 The Authors. Journal compilation © 2008 British Ecological Society, Journal of Animal Ecology, 77, 646–653 Ecological separation in a salamander 653 Hendry, A.P., Wenburg, J.K., Bentzen, P., Volk, E.C. & Quinn, T.P. (2000) Rapid evolution of reproductive isolation in the wild: evidence from introduced salmon. Science, 290, 516–518. Hickerson, C.M., Anthony, C.D. & Wicknick, J.A. (2004) Behavioral interactions between salamanders and centipedes: competition in divergent taxa. Behavioural Ecology, 15, 679–686. Highton, R. (1959) The inheritance of the color phases of Plethodon cinereus. Copeia, 1959, 33–37. Highton, R. (1975) Geographic variation in genetic dominance of the color morphs of the red-backed salamander, Plethodon cinereus. Genetics, 80, 363– 374. Hollander, J., Lindegarth, M. & Johannesson, K. (2005) Local adaptation but not geographical separation promotes assortative mating in a snail. Animal Behavior, 70, 1209–1219. Jaeger, R.G. (1981) Diet diversity and clutch size of aquatic and terrestrial salamanders. Oecologia, 48, 190–193. Jaeger, R.G. (1984) Agonistic behavior of the red-backed salamander. Copeia, 1984, 309 –314. Jaeger, R.G. (1990) Territorial salamanders evaluate size and chitinous content of arthropod prey. Behavioral Mechanisms of Food Selection (ed. R.N. Hughes), pp. 111–126. NATO ASI Series, Subseries G. Ecological Sciences. Springer-Verlag, Heidelberg. Jaeger, R.G., Gillette, J. & Cooper, R. (2002) Sexual coercion in a territorial salamander: males punish socially polyandrous female partners. Animal Behaviour, 63, 871–877. Jaeger, R.G., Wicknick, J.A., Griffis, M.R. & Anthony, C.D. (1995) Socioecology of a terrestrial salamander: juveniles enter adult territories during stressful foraging periods. Ecology, 76, 533–543. Jawor, J.M., Linville, S.U., Beall, S.M. & Breitwisch, R. (2003) Assortative mating by multiple ornaments in northern cardinals (Cardinalis cardinalis). Behavioural Ecology, 14, 515–520. Johnstone, R.A., Reynolds, J.D. & Deutsch, J.C. (1996) Mutual mate choice and sex differences in choosiness. Evolution, 50, 1382–1392. Jones, A.G., Moore, G.I., Kvarnemo, C., Walker, D. & Avise, J.C. (2003) Sympatric speciation as a consequence of male pregnancy in seahorses. Proceedings of the National Academy of Sciences USA, 100, 6598–6603. Karuzas, J.M., Maerz, J.C. & Madison, D.M. (2004) An alternative hypothesis for the primary function of a proposed mate assessment behaviour in redbacked salamanders. Animal Behaviour, 68, 489–494. Kondrashov, A.S. & Mina, M.V. (1986) Sympatric speciation – when is it possible? Biological Journal of the Linnean Society, 27, 201–223. Lang, C. & Jaeger, R.G. (2000) Defense of territories by male-female pairs in the red-backed salamander (Plethodon cinereus). Copeia, 2000, 169–177. Lotter, F. (1978) Reproductive ecology of the salamander Plethodon cinereus (Amphibia, Urodela, Plethodontidae) in Connecticut. Journal of Herpetology, 2, 231–236. Lotter, F. & Scott, N.J. (1977) Correlation between climate and distribution of the color morphs of the salamander Plethodon cinereus. Copeia, 1977, 681– 690. Maerz, J.C. & Madison, D.M. (2000) Environmental variation in territorial behavior in a terrestrial salamander. The Biology of Plethodontid Salamanders (eds R.C. Bruce, R.G. Jaeger & L.D. Houck), pp. 395–406. Kluwer Academic/Plenum Publishing, New York, NY. Maerz, J.C., Myers, E.M. & Adams, D.C. (2006) Trophic polymorphism in a terrestrial salamander. Evolutionary Ecology Research, 8, 23–35. Maglia, A.M. (1996) Ontogeny and feeding ecology of the red-backed salamander, Plethodon cinereus. Copeia, 1996, 576–586. Marsh, D.M. & Goicochea, M.A. (2003) Monitoring terrestrial salamanders: biases caused by intense sampling and choice of cover objects. Journal of Herpetology, 37, 460–466. Mathis, A. (1990) Territoriality in a terrestrial salamander: the influence of resource quality and body size. Behaviour, 112, 162–175. Mathis, A. (1991) Territories of male and female terrestrial salamanders: costs, benefits, and intersexual spatial associations. Oecologia, 86, 433–440. Mathis, A., Jaeger, R.G., Keen, W.H., Ducey, P.K. & Buchanan, B.W. (1995) Aggression and Territoriality by Salamanders and a Comparison with the Territorial Behavior of Frogs Amphibian Biology, vol. 2. Social Behavior (eds H. Heatwole & B.K. Sullivan), pp. 633–676. Surry Beatty and Sons, Chipping Norton, NSW, Australia. Maynard Smith, J. (1966) Sympatric speciation. American Naturalist, 100, 637–650. Mitchell, J.C., Wicknick, J.A. & Anthony, C.D. (1996) Effects of timber harvesting practices on peaks of otter salamander (Plethodon hubrichti) populations. Amphibian and Reptile Conservation, 1, 15–19. Moreno, G. (1989) Behavioral and physiological differentiation between the color morphs of the salamander, Plethodon cinereus. Journal of Herpetology, 3, 335–341. Nagel, J.W. (1977) Life history of the red-backed salamander, Plethodon cinereus, in northeastern Tennessee. Herpetologica, 33, 13–18. Nosil, P. (2004) Reproductive isolation caused by visual predation on migrants between divergent environments. Proceedings of the Royal Society of London, Series B, Biological Sciences, 271, 1521–1528. Peterson, M.G. (2000) Alternative life styles in a terrestrial salamander: do females preferentially associate with each other? The Biology of Plethodontid Salamanders (eds R. Bruce, R.G. Jaeger & L. Houck), pp. 417–429. Plenum Publishers, New York. Petranka, J.W. (1998) Salamanders of the United States and Canada. Smithsonian Press, Washington. Petruzzi, E.E., Niewiarowski, P.H. & Moore, F.B.-G. (2006) The role of thermal niche selection in maintenance of a colour polymorphism in redback salamanders (Plethodon cinereus). Frontiers in Zoology, 3, 10. Pfingsten, R.A. & Walker, C.F. (1978) Some nearly all black populations of Plethodon cinereus (Amphibia, Urodela, Plethodontidae) in Northern Ohio. Journal of Herpetology, 12, 163–167. Powell, R., Parmerlee, J.S. Jr, Rice, M.A. & Smith, D.D. (1990) Ecological observations of Hemidactylus brookii haitianus Meerwarth (Sauria: Gekkonidae) from Hispaniola. Caribbean Journal of Science, 26, 67–70. Prosen, E.D., Jaeger, R.G. & Hucko, J.A. (2006) Sexual coercion in the salamander Plethodon cinereus: is it merely a result of familiarity? Herpetologica, 62, 10–18. Quinn, V.S. & Graves, B.M. (1999) Space use in response to conspecifics by the red-backed salamander (Plethodon cinereus, Plethodontidae, Caudata). Ethology, 105, 993–1002. Rice, W.R. & Salt, G.W. (1990) The evolution of reproductive isolation as a correlated character under sympatric conditions – experimental evidence. Evolution, 44, 1140–1152. Robinson, B.W. (2000) Trade offs in habitat-specific foraging efficiency and the nascent adaptive divergence of sticklebacks in lakes. Behaviour, 137, 865–888. Schluter, D. (2001) Ecology and the origin of species. Trends in Ecology and Evolution, 16, 372–380. Schulte-Hostedde, A.I., Zinner, B., Miller, J.S. & Hickling, G.J. (2005) Restitution of mass-size residuals: validating body condition indices. Ecology, 86, 155–163. Skúlason, S., Snorrason, S.S., Ota, D. & Noakes, D.L.G. (1993) Genetically based differences in foraging behaviour among sympatric morphs of arctic charr (Pisces: Salmonidae). Animal Behaviour, 45, 1179–1192. Smith, T.B. & Skúlason, S. (1996) Evolutionary significance of resource polymorphisms in fish, amphibians, and birds. Annual Review of Ecology and Systematics, 27, 111–133. Spotila, J.R. (1972) Role of temperature and water in the ecology of lungless salamanders. Ecological Monographs, 42, 95–125. Takahashi, T. (2004) Morphological and genetic distinctness of rock and shellbed dwelling Telmatochromis (Teleostei, Cichlidae) in the south of Lake Tanganyika suggest the existence of two species. Journal of Fish Biology, 65, 419–435. Test, F.H. (1952) Spread of the black phase of the red-backed salamander in Michigan. Evolution, 6, 197–203. Tilley, S.G., Lundrigan, B.L. & Brower, L.P. (1982) Erythrism and mimicry in the salamander Plethodon cinereus. Herpetologica, 38, 409–417. Venesky, M.D. & Anthony, C.D. (2007) Antipredator adaptations and predator avoidance for two color morphs of the Eastern Red-backed Salamander, Plethodon cinereus. Herpetologica, 63, 450–458. Vernberg, F.J. (1953) Hibernation studies of two species of salamanders, Plethodon cinereus and Eurycea bislineata. Ecology, 34, 55–62. Walls, S.C., Mathis, A., Jaeger, R.G. & Gergits, W.F. (1989) Male salamanders with high quality diets have faeces attractive to females. Animal Behaviour, 38, 546–548. West-Eberhard, M.J. (1986) Alternative adaptations, speciation, and phylogeny (a review). Proceedings of the National Academy of Sciences USA, 83, 1388–1392. Whiteman, H.H., Krenz, J.D. & Semlitsch, R.D. (2006) Intermorph breeding and the potential for reproductive isolation in polymorphic mole salamanders (Ambystoma talpoideum). Behavioural Ecology Sociobiology, 60, 52–61. Whiteman, H.H. & Semlitsch, R.D. (2005) Asymmetric reproductive isolation among polymorphic salamanders. Biological Journal of the Linnean Society, 86, 265–281. Williams, E.E., Highton, R. & Cooper, D.M. (1968) Breakdown of polymorphism of the red-backed salamander on Long Island. Evolution, 22, 76–86. Zweifel, R.G. (1957) Studies on the critical thermal maxima of salamanders. Ecology, 38, 64–69. Received 5 October 2007; accepted 5 February 2008 Handling Editor: Andre Gilburn © 2008 The Authors. Journal compilation © 2008 British Ecological Society, Journal of Animal Ecology, 77, 646–653