Behavioral Ecology Vol. 15 No. 4: 647-653
Behavioral Ecology vol. 15 no. 4 © International Society for Behavioral Ecology 2004; all rights reserved
Philopatry, kin clusters, and genetic relatedness in a population of woodrats (Neotoma macrotis)
a Museum of Vertebrate Zoology and Department of Integrative Biology, University of California, Berkeley, CA 94720–3160, USA<, and b Department of Biological Sciences, Campus Box 8007, Idaho State University, Pocatello, ID 83209, USA
Address correspondence to Marjorie Matocq. E-mail: matomarj{at}isu.edu
Received 13 January 2003; revised 16 June 2003; accepted 28 September 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Studies of highly kin-structured mammal societies have revealed the importance of natal philopatry in determining the distribution of genetic variation within populations. In comparison, the relationship between philopatry and genetic diversity within populations of moderately kin-structured societies has received relatively little attention. Previous studies of Neotoma macrotis have suggested that females form distinct kin clusters. Each kin cluster overlaps spatially with the home range(s) of one or more males that are not related to each other or to the females with which they are spatially associated. To examine interactions between philopatry and genetic structure in this apparently moderately kin-structured species, we characterized spatial and genetic relationships among individually marked females in a population of N. macrotis from central coastal California. Our field studies revealed that, contrary to expectation, females in this population were not strongly philopatric and spatially clustered females were not characterized by high levels of genetic relatedness. Nevertheless, genetic structure was evident within the study population; spatial and genetic distances among females were significantly correlated, suggesting that dispersal patterns influenced genetic structure even in the absence of marked female philopatry. Because females with overlapping spatial distributions were not typically closely related to one another, opportunities for the evolution of kin-selected social behavior (e.g., cooperative care of young) appear to be limited in this population.
Key words: Neotoma macrotis, Neotoma fuscipes, relatedness, female kin clusters, natal philopatry.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Kin groups are the core of many vertebrate social systems (Emlen, 1994
, 1995). Kin groups generally form owing to natal philopatry (Emlen, 1982, 1991; Solomon, 2003), with the result that individuals that share recent ancestry are spatially clustered in the habitat. This physical arrangement of closely related individuals is expected to impact significantly patterns of both social behavior and population genetic structure (Dobson et al., 1998; Sugg et al., 1996). For example, cooperative interactions that are maintained by kin selection (e.g., alloparental care: Hamilton, 1964; alarm calling: Sherman, 1981; but see Clutton-Brock, 2002) should occur primarily within clusters of closely related individuals. In terms of population genetic structure, the temporal persistence of kin clusters is expected to lead to decreased variation within but increased genetic variation among kin groups (Sugg et al., 1996). Thus, understanding the kin structure of free-living populations of vertebrates should yield important insights into the nature of social and genetic variation in these animals.
Among mammals, female kin groups occur in numerous taxa, including carnivores (lions: Packer, 1986), ungulates (zebras: Rubenstein, 1986), rodents (mice: Wilkinson and Baker, 1988), and primates (baboons: Dunbar, 1986). Although the behavioral consequences of female kin cluster formation have been examined for a variety of mammalian species and social systems (see Chapais et al., 2001; Dalton, 2000; Holekamp et al., 1997; Hoogland, 1995; Lacey et al., 1997; Michener, 1979, 1983), the genetic consequences of female kin group formation have received considerably less attention. Analyses of the degree to which kin group formation influences genetic structure have typically focused on highly social, strongly kin-structured mammalian species such as black-tailed prairie dogs (Chesser, 1983; Dobson et al., 1998), red howler monkeys (Pope, 1992, 1998), and naked mole-rats (Faulkes et al., 1997; Reeve et al., 1990). In comparison, relationships between social structure and genetic variation in less strongly kin-oriented societies have not been as extensively documented (Dobson et al., 1998). Comparative data from moderately kin-structured societies are important because they serve to delineate how social and genetic structure covary. What degree of kin clustering is required for specific patterns of social behavior to evolve? Does genetic structure persist in the absence of pronounced philopatry and kin cluster formation? Answering these questions should significantly improve our understanding of how social and genetic structure interact to shape the evolution of free-living populations of mammals.
To explore relationships between social and genetic structure in a moderately kin-structured mammal, we characterized spatial relationships and genetic variation among female woodrats in a population of Neotoma macrotis (recognized as a species distinct from N. fuscipes by Matocq (2002); the population studied in this report has been uniformly considered N. fuscipes in the prior literature). A distinctive feature of the biology of this species is the construction of stick "houses" (Linsdale and Tevis, 1951). Each adult N. macrotis occupies its own house, which it actively maintains through the addition of new sticks and vegetation (Linsdale and Tevis, 1951). A previous study of this species indicated that females rarely disperse more than 100 m from their natal house (Kelly, 1989), suggesting that philopatry and the formation of female kin groups may be common. These traits imply that N. macrotis may be moderately kin-structured, making them ideal for comparisons with the more strongly kin-structured species identified above.
The objectives of the current study are twofold. First, we characterize spatial relationships among individual females to determine the extent of kin cluster formation in this species. Based on the information provided by Kelly (1989), we predict that closely related females will occur in close spatial proximity, with most pairs of first-order female kin occupying adjacent houses and home ranges. Second, we explore how the spatial distribution of related females influences patterns of genetic variation in these animals. If our inferences regarding the kin structure of N. macrotis are correct, we predict that genetic variation in this species will be spatially structured, although distinctions between different kin clusters should be less extreme than those reported for highly kin-structured societies in which only close kin overlap spatially (e.g., prairie dogs). To test these predictions, we combine field data on the spatial distributions of individually marked females with microsatellite analyses of genetic variation in these animals. Our analyses reveal unexpected variation in spatial relationships among close female kin that is reflected in the genetic structure of the study population. These findings have important implications for our understanding of the role of natal philopatry and female kin associations in shaping population genetic structure.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study population
This study was conducted on a 580-m stretch of riparian/oak woodland along Big Creek at the Hastings Natural History Reservation in Carmel Valley, California. A 329-m portion of the site is densely vegetated by large individual oaks and/or willow thickets, whereas the remaining 249 m are more sparsely covered (Kelly, 1989
; Linsdale and Tevis, 1951). Average width of the riparian corridor is 26 m (N = 10; 2.8 SE). The current study site was included in the studies of both Linsdale and Tevis (1951) and Kelly (1989). In fact, many of the houses included in the current study were present during both previous investigations, which indicates that individual houses may persist for at least 60 years. Members of our study population of N. macrotis were captured by using Tomahawk live traps baited with peanut butter and oats. Two traps housed in a wooden shelter were placed near each woodrat house. Traps were set for three consecutive nights at approximately 8-week intervals from August 1997–March 2000. Traps were opened at sunset, checked within 4 h of sunset, and then checked again and closed at sunrise. During the 1998 and 1999 breeding seasons (February–September), livetrapping was increased to four to five nights per week; to avoid disrupting the mating activity of the study subjects, traps were checked within 5 h of sunset and then closed for the night.
At first capture, all animals were eartagged (Monel 100S, National Band and Tag Co.), sexed, and weighed. In addition, a small piece of ear pinnae was removed with sterile surgical scissors and stored in 95% ethanol for genetic analyses of relatedness. Weight and reproductive status were monitored throughout the breeding season, and mother-offspring relationships were established by transfer of fluorescent pigment powders between individuals (Ribble, 1991). To minimize inhalation of pigmented powder by the animals (Stapp et al., 1994), only a small amount of powder was applied to the area immediately surrounding the teats of lactating females.
Spatial relationships among females
Because only females in the study population were expected to be philopatric (Kelly, 1989), we focused our attention on members of this sex. Patterns of space use were determined from trapping records for all adult female woodrats resident on the site during the 1998 and 1999 breeding seasons. Capture locations for each animal were used to estimate home range size by using the Minimum Convex Polygon option of the Animal Movement extension of ARCVIEW 3.2 (ESRI Technology). Only data from females captured at a minimum of three different locations were included in this analysis.
Geographic distances among individuals were calculated based on primary house residency. Residency was first established by direct observation of which house was entered each time an individual was released after trapping. To validate this method, radiotelemetry was used to determine residency for a subset of animals in the study population. Specifically, eight individuals (five males, three females) were radiocollared and located throughout daylight hours in order to determine which house they occupied (Matocq MD and Good-Brummer L, unpublished data). For six of these eight individuals, assumptions regarding house residency were confirmed. In the remaining two cases, individuals were found to be occupying hollow logs adjacent to and within 1 m of the suspected house. Thus, consistently entering the same house immediately after release from a trap was taken as evidence that an animal was resident in that house.
Hierarchical spatial relationships among females were established by using the UPGMA clustering algorithm as implemented in the NEIGHBOR subroutine of the software package PHYLIP (version 3.5c) (Felsenstein, 1993). By using a simple clustering algorithm to analyze spatial distance between females, it was possible to identify the most spatially proximate groups of females. Clusters of females can be identified at a variety of spatial scales from nearest neighbors to progressively larger spatial groupings, thus facilitating the examination of the scale at which there may be a relationship between spatial and genetic relatedness. Phylograms were visualized with the TREEVIEW software (version 1.5) (Page, 1996). Single females that were isolated from their nearest neighbors by more than 30 m of open grassland were not included in the spatial clustering analysis. Although some females disperse more than 30 m from their natal area (Kelly, 1989; Linsdale and Tevis, 1951), individuals resident in isolated habitat patches were never trapped outside these patches, and conversely, nonresident females were never trapped in such patches. These data suggest that opportunities for residents of isolated patches to interact with other females were limited. Because we were interested in assessing spatial and genetic relationships among individuals that were likely to engage in interactions with one another, we excluded residents of isolated patches from our analyses.
Genetic analyses of female relatedness
DNA was isolated by using a standard phenol-chloroform extraction followed by ethanol precipitation (Sambrook et al., 1989). Individuals were genotyped for six microsatellite loci under the conditions described in Matocq (2001). Four loci—Nfu1, Nfu4, Nfu5, and Nfu6—were developed specifically for our study species (Matocq, 2001). The remaining two loci, Nma10 and Nma11, were originally designed for Neotoma magister (Castleberry et al., 2000). Fluorescently labeled PCR products were electrophoresed on 6% denaturing acrylamide gels on an ABI 377 (Applied Biosystems Inc.). Fragment sizes were analyzed and genotypes were assigned by using the GENESCAN and GENOTYPER software (Applied Biosystems Inc.).
Deviations from Hardy-Weinberg equilibrium were evaluated by using a modified version of Fisher's Exact test (Guo and Thompson, 1992) as implemented in ARLEQUIN 2.000 (Schneider et al., 2000). Linkage disequilibrium among loci was evaluated by using the likelihood-ratio test described by Slatkin and Excoffier (1996). Pairwise relatedness among females was calculated by using the RELATEDNESS software package (version 5.0) (Queller and Goodnight, 1989). Correlations between geographic distance and genetic relatedness were conducted by using regression analyses as implemented by STATVIEW 5.0 (SAS Institute, Inc.). Throughout the text, means are reported ±1 SE.
Finally, if the female population was subdivided into multiple independent breeding groups as predicted by Kelly (1989), then either inbreeding or the Wahlund effect should produce significantly positive values for the inbreeding coefficient FIS (Wright, 1931). To determine whether there was significant substructure within the female population, we estimated FIS by using the FSTAT software (Goudet, 1999).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Spatial relationships among females
A total of 28 adult females were resident on the study site in 1998, with 26 resident on the site in 1999. Ten (38.5%) of the 26 females present in 1999 had also been present in 1998. Because the identities of neighboring females and the dimensions of individual home ranges changed substantially across seasons, individuals present in both years of the study were included in analyses for 1998 and 1999. Female population density was 22.1 individuals/hectare at the start of the 1998 breeding season, versus 19.5 individuals/hectare at the start of the 1999 breeding season. Eighty-seven woodrat houses were present on the site at the start of this study; no new houses were built and only three houses in trees disintegrated during the course of data collection. Considering animals of both sexes and assuming that each house was occupied by a single adult, this yielded approximately 60% of houses on the site that were occupied during the study. Although some individuals may have occasionally used more than one house, only the primary residence of each adult is considered here.
On average, adjacent houses (N = 87) were located 7.5 ± 0.6 m from each other. Average female home range size was 181.9 ± 22.1 m2 (range = 4.8–545.2), and home ranges included an average of 5.8 ± 0.3 houses (range = 3–10; N = 47). Estimates based on trapping data typically underestimate the full extent of space use (Cranford, 1977). Trapping estimates should, however, reflect at least the core portion of the area (i.e., the area encompassing an individual's house) used by a female.
UPGMA cluster analysis of the pairwise distances among females revealed spatial groupings of females at several hierarchical levels. Six apparent spatial clusters of females (2–5 females per cluster) are depicted in Figure 1. Kelly (1989) reported seven spatially defined groups of females in the same portion of the study site. Although Kelly did not present data regarding the exact geographic extent of the putative kin clusters that he identified, it is not unreasonable to assume that spatially defined groups would be fairly consistent across time if woodrat houses are naturally clustered in space and if these groups of houses are maintained across generations. Nonetheless, because it is unclear at which spatial scale, if any, there is high relatedness among females in this population, we examined average relatedness among females at multiple spatial scales (see below).
|
In 1999, the mean distance between the houses of nearest neighbor females (regardless of relatedness) was 19.1 ± 2.5 m (range = 3.5–50.2; N = 26). In contrast, the distance between individual females and their spatially nearest first-order relative was 45.3 ± 23.1 m (N = 13). The distance between first-order relatives ranged from 8–311 m; at both extremes, this was the distance between houses occupied by a female and her adult daughter. Although females were, on average, twice as far from their spatially nearest first-order relative than from their nearest female neighbor (regardless of relatedness), there was no significant difference between the distances (Mann-Whitney U = 143; N = 13, 26; p =.44). The known first-order kin in this analysis were separated by an average of 7.5 ± 2.7 (range = 0–23) unoccupied houses, and six (60%) of these 10 pairs of females were separated by at least one unrelated female. Estimates of the number of unoccupied houses are based on maximum occupancy during the breeding season and, thus, represent conservative estimates of the number of vacant houses between individuals.
Because genealogical relationships among females were not known for 1998, we used genetic data to infer first-order relationships among animals resident on the study site during that year; these analyses revealed six pairs of first-order female kin (average r =.528 ±.03). The mean distance between a female and her spatially nearest first-order kin in 1998 was 62.3 ± 6.9 m (range = 36.5–98.7; N = 10). In contrast, the mean distance between pairs of nearest neighbor females was 13.1 ± 1.2 m (range = 6.7–30.7; N = 28). This value was significantly smaller than was the mean distance between females and their spatially closest first-order relative (Mann-Whitney U = 0; N = 10, 28; p <.01). Pairs of related females were separated by an average of 5.7 ± 1.7 (range = 2–13) unoccupied houses, and five (83%) of the six related pairs were separated by at least one unrelated female. Thus, data from 1998 and 1999 are similar in suggesting that first-order relatives are not always each other's nearest neighbors.
Female philopatry and natal dispersal
Of the 26 breeding females resident on the site during the 1999 season, nine (34.6%) had been born on the site during the previous spring and summer. Of the remaining 17 females, 10 (38.5%) had been breeding adults on the study site during the previous year, whereas seven (26.9%) had dispersed onto the site after breeding ended in 1998. Only two adult resident females were known to have dispersed during the study: one moved 42 m (two houses away, one unoccupied), and the other moved 142 m (14 houses away, six unoccupied).
Of the 19 females born on the site in the 1998 breeding season, nine (47%) remained on the site to breed in 1999. Five (55%) of the nine females were philopatric, taking up residence either in the house in which they were born or in one immediately adjacent to (within 20 m of) their natal house. The other four yearling females (44%) dispersed to take up residence at houses 64, 82, 244, and 322 m away. These females were separated from their natal house by an average of 13 ± 4.3 unoccupied houses.
For the seven females that immigrated to the site, the three nearest locations from which they could have come were 40, 60, and 137 m away from the study site, across areas of open grassland. Four additional instances of dispersal by subadult females come from observations made at the beginning of the study and just before its completion. Two subadult females that were resident on the site when the study began dispersed 65 and 137 m before becoming reproductively active. Although the study terminated before establishment of breeding areas for the 2000 season, two of the females born in 1999 had already dispersed 122 and 145 m from their natal house. Thus, of the 88 females monitored during this study, six (6.8%) dispersed over 100 m from their natal house.
Genetic relatedness and population structure
The six loci examined were highly polymorphic, with nine to 17 alleles detected per locus (Table 1). One locus, Nfu6, deviated significantly from Hardy-Weinberg equilibrium owing to the presence of a null allele(s), the existence of which was confirmed by comparing the genotypes of several known mother-offspring pairs (Matocq MD, 2004). Null or nonamplifying alleles can cause heterozygotes to be scored as homozygotes (Pemberton et al. 1995), with the result that, if an offspring inherits a nonamplifying allele from its mother, the pair would appear to be unrelated at that locus. Because null alleles can contribute to underestimates of relatedness, locus Nfu6 was excluded from subsequent analyses. With the exclusion of locus Nfu6, heterozygosity at the five loci used to calculate relatedness ranged from 0.84–0.93 (Table 1).
|
All known mother-offspring pairs (N = 78) had an average pairwise relatedness of 0.50 ± 0.02. All known full-sib pairs (N = 24) had an average pairwise relatedness of 0.50 ± 0.03. Average pairwise relatedness of all adult females resident on the study site in 1998 was –0.006 ± 0.01 (N = 378 pairs); in 1999, this value was –0.021 ± 0.01 (N = 325 pairs). Average pairwise relatedness between nearest female neighbors (as determined from the locations of occupied houses) was 0.08 ± 0.04 in 1998 (N = 28) and 0.07 ± 0.05 in 1999 (N = 26). The average pairwise relatedness of nearest neighbors was not significantly different from that of randomly selected pairs of females in either 1998 or 1999 (N = 26 pairs each year; Mann-Whitney U tests, all p >.05). Thus, pairs of nearest neighbors were not typically composed of closely related females.
As indicated above (Figure 1), distinct clusters of females could be identified at several spatial scales. Because we had no predefined spatial criteria for determining the limits of female kin clusters, we estimated the average relatedness of females in all spatial groupings that emerged from the UPGMA analysis. These values are reported in Figure 1 for each branch of the phylogram connecting particular clusters of females. Average pairwise r values for a spatially clustered set of females ranged from –0.230 to 0.670, indicating a range of genetic relationships among spatially associated females. Data for 1998 are not shown in Figure 1, but the results obtained are similar, with average pairwise r values among spatially clustered females ranging from –0.290 to 0.660.
In a system characterized by extreme philopatry, we would expect a strong negative correlation between degree of relatedness and interhouse distance, with only highly related females overlapping at the smallest spatial scales. In contrast, in a system in which dispersal distance is random with respect to natal site, we would expect no relationship between genetic and spatial distance. Our data suggest that N. macrotis falls between these extremes, with 22% of first-order relatives living in adjacent houses but with some being separated from each other by distances of more than 100 m (approximately 20% the length of the study site) as well as by unoccupied houses. Regressing the estimated coefficient of relatedness between a pair of females on the distance between the houses occupied by those individuals revealed a significant negative relationship between genetic relatedness and distance (data from 1998 and 1999 combined; r2 =.032, F = 23.1, df = 701, p <.0001) (Figure 2). Despite statistical significance, the small r2 value indicates a minimal relationship between spatial distance and genetic relatedness in this population.
|
When averaged across all loci, FIS among females resident in the study population in 1998 was –0.040 (0.013); in 1999 this values was –0.023 (0.018). Neither average value of FIS was significantly different from zero (Bonferroni corrected p =.01), indicating no significant genetic substructure within the female population.
Female reproductive success
To determine if females that lived near close relatives achieved greater direct fitness than did those not associated with close kin, we compared the number of pups reared to weaning by females whose nearest neighbors were first-order relatives (r 
.5) to the number of pups reared to weaning by females that did not live adjacent to kin. Combining data across years, a total of 12 (22%) females lived adjacent to a first-order relative. When data from both years were combined, there was no significant association between the percentage of females that reared at least one pup to weaning and the presence of close kin (Fisher's Exact test, p =.5061). Similarly, the number of pups reared to weaning did not differ between females living near close kin (1.75 ± 0.55; N = 12) and those living near more distantly or unrelated animals (1.41 ± 0.22; N = 42; Mann-Whitney U = 236.5; N = 12, 42; p =.747). Thus, the close physical proximity of first-order kin did not appear to confer a reproductive advantage on females in the study population.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Contrary to expectation, females in our study population were not strongly philopatric. The mean distance between houses of known first-order relatives was greater than that between houses of nearest neighbors, and typically, several unrelated females were resident in houses located between those occupied by closely related individuals. Microsatellite analyses revealed that the mean coefficient of relatedness between females occupying adjacent houses was not significantly different from the background degree of relatedness among females in the study population. Finally, although 47% of the females born on the study site remained resident there as breeding adults, only slightly more than half of these animals (55%) were resident in or adjacent to their natal house as adults. Collectively, these data indicate that although some females were philopatric, philopatry is not the rule in this population of N. macrotis.
Variation in female philopatry
Our data, similar to those of Linsdale and Tevis (1951), do not indicate that female N. macrotis at the Hastings Reservation are strongly philopatric. This is in contrast to the findings of Kelly (1989), who suggested that females in this population are philopatric, resulting in the formation of distinct kin clusters. In the current study, 7% of females dispersed more than 100 m. This same dispersal distance was reported by Linsdale and Tevis (1951) for 15.6% of females between 1938–1942 and 4.1% of females between 1946–1947 (from Kelly, 1989). Kelly (1989), however, found that only 2% of females dispersed more than 100 m. Although these data do not provide an objective criterion for identifying philopatry, they suggest considerable temporal variation in the extent to which related females are spatially clustered in the habitat. Reasons for this temporal variation in the degree of natal philopatry are unclear. Methodological differences are unlikely to explain this variation, because all three studies were based on measured dispersal distances for marked females.
Philopatry is thought to occur when ecological factors constrain opportunities for natal dispersal (Emlen, 1982, 1991; Koenig et al., 1992, Solomon, 2003). Thus, higher densities resulting from a larger population and/or fewer available houses could contribute to variation in natal dispersal among years. However, adult population sizes at the beginning of the breeding season appear to have been fairly similar between the current study and that of Kelly (1989), with approximately 35–40 individuals on comparable areas of the study site. Also, there appears to have been a similar number of houses available on the study site, with 87 in the current study and approximately 91 in the comparable portion of Kelly's study site from 1986–1988. Although it remains unclear what variables contribute to variation in dispersal tendencies across years, changes in habitat quality (e.g., increased food availability associated with oak masting; Kelly, 1989) and intense fluctuations in weather (e.g., El Niño events) may be important in determining dispersal patterns in this population.
Adaptive bases for natal dispersal
Reasons for natal dispersal in mammals are poorly understood. Frequently proposed hypotheses include inbreeding avoidance, competition with close relatives, and variation in territory (house) quality (Dobson, 1982; Greenwood, 1980). This study did not attempt to test these hypotheses directly, and further research is needed to determine why the tendency to be philopatric differs among individuals, leading some females to disperse further from their natal area than would appear necessary based on house occupancy.
One factor that was not considered in the current study that may contribute to female dispersal patterns is the behavior of males. Breeding females never lived adjacent to their fathers; this pattern appeared to be owing to movements by daughters, rather than to dispersal by males. Five males remained at or near the same house during the 1998 and 1999 seasons (Matocq MD, unpublished data). In these cases, their female offspring dispersed to other areas on the study site or disappeared from the study site. If spatial segregation of breeding males and their adult daughters is important (e.g., to avoid inbreeding), then dispersal by subadult females may be influenced by patterns of adult male territoriality. Additional field studies are needed to identify the conditions under which adult males and their daughters disperse.
Another factor that was not considered during this study is house quality. In our study population, dispersing females often moved more than the apparent minimum distance required to obtain a house, as evidenced by the number of unoccupied houses between first-order relatives. In a number of cases, females moving into an area with multiple vacant houses took up residency in the house that had been most recently occupied by another woodrat (Matocq MD, personal observation), suggesting that some houses are preferred over others. Factors such as cache quality (Post, 1992), which would presumably be highest in recently occupied houses, may influence house choice. In addition to affecting which house is chosen by a dispersing individual, house (and territory) quality may also contribute to the need for females to leave their natal area. In three cases in which individual females dispersed from their natal area, their mother and a female sibling from the same litter remained at the natal site. This may indicate a limit to the number of females that can occupy certain territories. Further studies are required to assess the role of house and territory quality in shaping the dispersal tendencies of individual females.
Philopatry, genetic structure, and social behavior
How does the mixture of female natal dispersal and female natal philopatry reported here influence the genetic structure of the study population? Because of the combination of a high annual rate of turnover among females resident on the study site, frequent female natal dispersal, and multiple paternity within and among litters reared by the same female (Matocq, 2004), a relatively small proportion of adult females are related at the 0.5 level, and an even smaller proportion are both closely related and spatially proximate. A significant pattern of isolation by distance, however, was evident within the population, suggesting that female movements are not random with respect to spatial distribution. This finding may reflect relatively short-term local patterns of female association, such as the tendency for most close relatives that lived adjacent to one another to be mothers and their yearling daughters. At the same time, it may reflect a relatively consistent mean dispersal distance among females such that, over larger spatial scales, distances between closely related females tend to be roughly normal in distribution. Regardless of the underlying mechanism, our genetic analyses suggest that female movements can influence genetic structure even in the absence of marked natal philopatry.
With regard to social behavior, the relatively low average relatedness among neighboring females suggests that the potential for the evolution of complex kin-selected patterns of social behavior in this population is also low. The formation of temporally-stable kin groups is generally thought to be a prerequisite for kin selection (Chesser, 1991), yet such groups were not revealed by our analyses of either nearest-neighbor relationships or the distribution of alleles within individuals (FIS). Whether this represents the typical kin structure of our study population or a transient condition resulting from fluctuations in environmental conditions, the absence of stable kin groups is expected to influence the evolution of social behavior in these animals. Although cooperation may occur among unrelated females (Clutton-Brock, 2002), it is expected to be less common than cooperation among close kin. Thus, patterns of female dispersal also have important implications for the social structure of N. macrotis. Future studies of this population will combine molecular analyses with detailed behavioral observations of social interactions to explore relationships between individual movements, kin structure, and the social system of N. macrotis.
|
ACKNOWLEDGEMENTS |
|---|
We gratefully acknowledge the assistance of a number of individuals without whom this research would not have been possible. For logistical support and hospitality, we thank researchers at the Hastings Natural History Reservation, especially M. Stromberg, W. Koenig, J. Dickinson, J. Haydock, D. Gubernick, and M. Kalcounis-Rueppel. For assistance in the field, we thank C. Feldman, L. Good-Brummer, and A. Minn. For introducing her to woodrats and for unlimited encouragement and support, M.D.M. gives special thanks to J. Patton. We also give special thanks to Patrick Kelly for support in this work and in many other aspects of M.D.M.'s dissertation research. A previous version of this manuscript was improved by comments from J. Patton and two anonymous reviewers. This research was supported by an NSF Dissertation Improvement Grant, the Museum of Vertebrate Zoology, the Department of Integrative Biology, U.C. Berkeley, the Mildred Mathias Fund, Sigma Xi, the American Society of Mammalogists, and the American Museum of Natural History.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Castleberry SB, King TL, Wood PB, Ford WM, 2000. Microsatellite DNA markers for the study of Allegheny woodrat (Neotoma magister) populations and cross-species amplification in the genus Neotoma. Mol Ecol 9:824-825.[CrossRef][Medline]
Chapais B, Savard L, Gauthier C, 2001. Kin selection and the distribution of altruism in relation to degree of kinship in Japanese macaques (Macaca fuscata). Behav Ecol Sociobiol 49:493-502.[CrossRef]
Chesser RK, 1983. Genetic variability within and among populations of black-tailed prairie dogs. Evolution 37:320-331.[CrossRef]
Chesser RK, 1991. Influence of gene flow and breeding tactics on gene diversity within populations. Genetics 129:573-583.[Abstract]
Clutton-Brock TH, 2002. Breeding together: kin selection and mutualism in cooperative vertebrates. Science 296:69-72.
Cranford JA, 1977. Home range and habitat utilization by Neotoma fuscipes as determined by radiotelemetry. J Mammal 58:165-172.[CrossRef]
Dalton CL, 2000. Effects of female kin groups on reproduction and demography in the gray-tailed vole, Microtus canicaudus. Oikos 90:153-159.[CrossRef]
Dobson FS, 1982. Competition for mated and predominant juvenile male dispersal in mammals. Anim Behav 30:1183-1192.[CrossRef]
Dobson FS, Chesser RK, Hoogland JL, Sugg DW, Foltz DW, 1998. Breeding groups and gene dynamics in a socially structured population of prairie dogs. J Mammal 79:671-680.[CrossRef]
Dunbar RIM, 1986. The social ecology of gelada baboons. In: Ecological aspects of social evolution (Rubenstein DI, Wrangham RW, eds). Princeton, New Jersey: Princeton University Press; 332–351.
Emlen ST, 1982. The evolution of helping: an ecological constraints model. Am Nat 119:29-39.[CrossRef][ISI]
Emlen ST, 1991. Evolution of cooperative breeding in birds and mammals. In: Behavioral ecology (Krebs JR, Davies NB, eds). Oxford: Blackwell; 301–337.
Emlen ST, 1994. Benefits, constraints, and the evolution of the family. Trends Ecol Evol 9:282-285.[CrossRef]
Emlen ST, 1995. An evolutionary theory of the family. Proc Natl Acad Sci USA 92:8092-8099.
Faulkes CG, Abbott DH, O'Brien HP, Lau L, Roy MR, Wayne RK, Bruford MW, 1997. Micro- and macro-geographic genetic structure of colonies of naked mole-rats, Heterocephalus glaber. Mol Ecol 6:615-628.[CrossRef][Medline]
Felsenstein J, 1993. PHYLIP (Phylogeny Inference Package), version 3.5c. Seattle: Washington: Department of Genetics, University of Washington.
Goudet J, 1995. FSTAT: a computer program to calculate F-statistics. J Hered 86:485-486.
Greenwood PJ, 1980. Mating systems, philopatry and dispersal in birds and mammals. Anim Behav 28:1140-1162.[CrossRef]
Guo SW, Thompson EA, 1992. Performing the exact test of Hardy-Weinberg proportion for multiple alleles. Biometrics 48:361-372.[CrossRef][ISI][Medline]
Hamilton WD, 1964. The genetical evolution of social behaviour. J Theor Biol 7:1-52.[CrossRef][ISI][Medline]
Holekamp KE, Cooper SM, Katona CI, Berry NA, Frank LG, Smale L, 1997. Patterns of association among female spotted hyenas (Crocuta crocuta). J Mammal 78:55-64.[CrossRef]
Hoogland JL, 1995. The Black-tailed prairie dog: social life of a burrowing mammal. Chicago: University of Chicago Press.
Kelly PA, 1989. Population ecology and social organization of dusky-footed woodrats, Neotoma fuscipes (PhD Dissertation). Berkeley, California: University of California.
Koenig WD, Patelka FA, Carmen WJ, Mumme R, Stanback MT, 1992. The evolution of delayed dispersal in cooperative breeders. Q Rev Biol 67:111-150.[CrossRef][Medline]
Lacey EA, Braude SH, Wieczorek JR, 1997. Burrow sharing by colonial tuco-tucos (Ctenomys sociabilis). J Mammal 78:556-562.[CrossRef]
Linsdale JM, Tevis LP, 1951. The dusky-footed woodrat. Berkeley, California: University of California Press.
Matocq MD, 2001. Characterization of microsatellite loci in the dusky-footed woodrat, Neotoma fuscipes. Mol Ecol Notes 1:194-196.[CrossRef]
Matocq MD, 2002. Morphological and molecular analysis of a contact zone in the Neotoma fuscipes species complex. J Mammal 83:866-883.[CrossRef]
Matocq MD, 2004. Reproductive success and effective population size in woodrats (Neotoma macrotis). Molecular Ecology (in press).
Michener GR, 1979. Spatial relationships and social organization of adult Richardson's ground squirrels. Can J Zool 57:125-139.
Michener GR, 1983. Kin identification, matriarchies, and the evolution of sociality in ground-dwelling sciurids. In: Advances in the study of mammalian behavior (Eisenberg JF, Kleiman DG, eds.), special publication 7 of the American Society of Mammalogists. Lawrence, Kansas: Allen Press, 528–572.
Packer C, 1986. The ecology of sociality in felids. In: Ecological aspects of social evolution (Rubenstein DI, Wrangham RW, eds). Princeton, New Jersey: Princeton University Press, 429–451.
Page RDM, 1996. TREEVIEW: an application to display phylogenetic trees on personal computers. Comput Appl Biosci 12:357-358.
Pemberton JM, Slate J, Bancroft DR, Barrett JA, 1995. Nonamplifying alleles at microsatellite loci: a caution for parentage and population studies. Mol Ecol 4:249-252.[Medline]
Pope TR, 1992. The influence of dispersal patterns and mating system on genetic differentiation within and between populations of the red howler monkey (Alouatta seniculus). Evolution 46:1112-1128.[CrossRef]
Pope TR, 1998. Effects of demographic change on group kin structure and gene dynamics of populations of red howling monkeys. J Mammal 79:692-712.[CrossRef]
Post DM, 1992. Change in nutrient content of foods stored by Eastern woodrats (Neotoma floridana). J Mammal 73:835-839.[CrossRef]
Queller DC, Goodnight KF, 1989. Estimating relatedness using genetic markers. Evolution 43:258-275.[CrossRef][ISI]
Reeve HK, Westneat DF, Noon WA, Sherman PW, Aquadro CF, 1990. DNA "fingerprinting" reveals high levels of inbreeding in colonies of the eusocial naked mole-rat. Nature 358:147-149.
Ribble DO, 1991. The monogamous mating system of Peromyscus californicus as revealed by DNA fingerprinting. Behav Ecol Sociobiol 29:161-166.[CrossRef]
Rubenstein DW, 1986. Ecology of sociality in horses and zebras. In: Ecological aspects of social evolution (Rubenstein DI, Wrangham RW, eds). Princeton, New Jersey: Princeton University Press, 282–302.
Sambrook J, Fritsch EF, Maniatis T, 1989. Molecular cloning—a laboratory manual. Cold Springs Harbor, New York: Cold Springs Harbor Laboratory Press.
Schneider S, Roessli D, Excoffier L, 2000. Arlequin ver. 2.000: A software for population genetics data analysis. Geneva: Genetics and Biometry Laboratory, University of Geneva.
Sherman PW, 1981. Kinship, demography, and Belding's ground squirrel nepotism. Behav Ecol Sociobiol 8:251-259.[CrossRef]
Slatkin M, Excoffier L, 1996. Testing for linkage disequilibrium in genotypic data using the expectation-maximization algorithm. Heredity 76:377-383.
Solomon NG, 2004. A reexamination of factors that influence natal philopatry in rodents. J Mammal 84:1182-1197.[CrossRef]
Stapp P, Young JK, VandeWoude S, Van Horne B, 1994. An evaluation of the pathological effects of fluorescent powder on deer mice (Peromyscus maniculatus). J Mammal 75:704-709.[CrossRef]
Sugg DW, Chesser RK, Dobson FS, Hoogland JL, 1996. Population genetics meets behavioral ecology. Trends Ecol Evol 11:338-342.[CrossRef]
Wilkinson GS, Baker AEM, 1988. Communal nesting among genetically similar house mice. Ethology 77:103-114.
Wright S, 1931. Evolution in mendelian populations. Genetics 16:97-159.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Csillery, T. Johnson, D. Beraldi, T. Clutton-Brock, D. Coltman, B. Hansson, G. Spong, and J. M. Pemberton Performance of Marker-Based Relatedness Estimators in Natural Populations of Outbred Vertebrates Genetics, August 1, 2006; 173(4): 2091 - 2101. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Randall, K. Rogovin, P. G. Parker, and J. A. Eimes Flexible social structure of a desert rodent, Rhombomys opimus: philopatry, kinship, and ecological constraints Behav. Ecol., November 1, 2005; 16(6): 961 - 973. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||