The genomic imprint of chromosomal inversions and demographic history in island populations of deer mice.

Populations that colonize islands experience novel selective pressures, fluctuations in size, and changes to their connectivity. Owing to their unique geographic setting, islands can function as natural laboratories in which to examine the interactions between demographic history and natural selection replicated across isolated populations. We used whole genome sequences of wild-caught deer mice (Peromyscus maniculatus) from two islands (Saturna and Pender) and one mainland location (Maple Ridge) in the Gulf Islands region of coastal British Columbia to investigate two primary determinants of genome-wide diversity: chromosomal inversions and non-equilibrium demographic history. We found that segregating inversions produce characteristic, large-scale distortions in allele frequencies and linkage disequilibrium that make it possible to identify and characterize them from short-read sequence data. Patterns of variation within and between karyotypes indicate that six inversion polymorphisms have been maintained by a shared history of balancing selection in both island and mainland populations. Whereas the estimated timing of contemporary population splits is consistent with the isolation of island populations from each other following the Last Glacial Maximum, ancestral island and mainland lineages are inferred to have diverged much earlier. These aspects of demographic history suggest that shared inversions existed long ago in a common ancestor or spread via limited gene flow between ancestral island and mainland lineages. Our results raise the possibility that inversions segregating among Gulf Islands populations are on similar evolutionary trajectories, providing a contrast to previous findings in mainland P. maniculatus and contributing to the emerging portrait of inversion evolution in this species.

The island syndrome hypothesis (ISH) stipulates that, as a result of local selection pressures and restricted gene flow, individuals from island populations should differ from individuals within mainland populations. Specifically, island populations are predicted to contain individuals that are larger, less aggressive, more sociable, and that invest more in their offspring. To date, tests of the ISH have mainly compared oceanic islands to continental sites, and rarely smaller spatial scales such as inland watersheds. Here, using a novel set of genome-wide SNP markers in wild deer mice (Peromyscus maniculatus) we conducted a genomic assessment of predictions underlying the ISH in an inland riverine island system: analysing island-mainland population structure, and quantifying heritability of phenotypes thought to underlie the ISH. We found clear genomic differentiation between the island and mainland populations and moderate to high marker-based heritability estimates for overall variation in traits previously found to differ in line with the ISH between mainland and island locations. FST outlier analyses highlighted 12 loci associated with differentiation between mainland and island populations. Together these results suggest that the island populations examined are on independent evolutionary trajectories, the traits considered have a genetic basis (rather than phenotypic variation being solely due to phenotypic plasticity). Coupled with the previous results showing significant phenotypic differentiation between the island and mainland groups in this system, this study suggests that the ISH can hold even on a small spatial scale.

Translocation and reintroduction are important tools for the conservation or recovery of species threatened with extinction in the wild. However, an understanding of the potential genetic consequences of mixing populations requires an understanding of the genetic variation within, and similarities among, donor and recipient populations. Genetic diversity was measured using two independent marker systems (microsatellites and AFLPs) for one island and four small remnant mainland populations of Setonix brachyurus, a threatened medium sized macropod restricted to fragmented habitat remnants and two off-shore islands in southwest Australia. Microsatellite diversity in the island population (Rs = 3.2, He = 71%) was similar to, or greater than, all mainland populations (Rs = 2.1–3.9, He = 34-71%). In contrast, AFLP diversity was significantly lower in the island population (PPL = 20.5; Hj = 0.118) compared to all mainland populations (mean PPL = 79.5–89.7; mean Hj = 0.23–0.29). Microsatellites differentiated all (mainland and island) populations from each other. However, AFLP only differentiated the island population from the mainland populations—all mainland populations were not significantly differentiated from each other for this marker. Given a known time since isolation of the island population from the mainland (6,000 years ago), and an overall more conservative rate of evolution of AFLP markers, our results are consistent with mainland populations fragmenting thousands of years ago (but <6,000 years), probably as a consequence of reduced rainfall and the constriction of the preferred mesic habitat of quokkas. Our results also support a recent history of severe population bottlenecks in mainland populations, and a long history of bottlenecks of the island population, but reflect a recent explosion in numbers since European occupation of the island. Our results indicate that translocation of island populations to supplement mainland populations would introduce genetically markedly differentiated, and possibly maladapted, individuals.

Allozyme variation was investigated in 17 Japanese populations of Campanula punctata, ten from the Izu Islands and seven in the mainland (Honshu). The data indicate that there are two groups, a mainland group and an island one, and that the systematically problematic Oshima Island (northernmost Izu island) populations are closely related to those of the other islands. Nei's genetic identity values among islands and among mainland populations were 0.95 and 0.97, respectively, while the value between island and mainland populations was 0.84, suggesting that the island populations are an independent species. Total genetic variation was nearly the same among island and mainland populations. However, the apportionment of variation within and among populations was considerably different; 14% of gene diversity exists among mainland populations, while 31% of the diversity exists among island populations. Mean outcrossing rates of self‐incompatible mainland and Oshima populations are 0.62–0.79, supporting xenogamy; those in self‐compatible island populations are 0.37–0.57 in the northern islands, indicating a mixed mating system, and 0.16–0.25 in southern ones, indicating dominant inbreeding. Total genetic diversity in each island population decreased with distance from the mainland. Genetic and geological data suggest that the ancestors of insular populations were founded on northern islands in a relatively ancient period and that they dispersed progressively to the southern ones. Chromosome number (2n = 34) and isozyme numbers indicate gene duplications in this species, which suggests it is an ancient polyploid.

Sullivan (1977) reported important differences in the demography of island and mainland populations of the deer mouse ( Peromyscus maniculatus ) in British Columbia. In this paper we compare the behavioral characteristics of these insular populations with those of the mainland population. The results demonstrate that island deer mice show significantly fewer aggressive and defensive interactions than do deer mice from the mainland. There is no significant difference in general activity. It is suggested that a decrease in competition over food and the frustration of dispersal may have been important factors in the reduction of agonistic behaviors in the island populations.

Island vertebrates that are small on the mainland tend to be larger and exhibit tamer behavior than their mainland conspecifics-a combined set of characteristics known as "island syndrome." Such island-specific traits are often attributed to lower predation pressure on islands than on the mainland. While the morphology and behavior of island vertebrates has received significant attention, relatively few studies have compared physiological traits between island and mainland populations. Given that hormones such as glucocorticoids are involved in responding to predation, it is reasonable to test whether island populations that have demonstrated characteristics of island syndrome also have different glucocorticoid levels than their mainland conspecifics. Here, we used a long-term museum collection of deer mice (Peromyscus maniculatus) obtained from two archipelagos and two mainland regions to test the hypothesis that island syndrome includes changes in time-averaged corticosterone levels, as measured in hair. As predicted by island syndrome, deer mice from islands were structurally larger and heavier for their given structural size than their mainland conspecifics. When we compared size-matched individuals (i.e., holding size constant), hair corticosterone levels did not differ between island and mainland mice. However, corticosterone levels scaled positively with body mass and condition across our sample population. This led to a relative increase in corticosterone levels among median-sized mice from islands relative to median-sized mice from mainland populations. We conclude that island syndrome does include effects on deer mouse stress physiology but only indirectly through the evolution of larger body size in island populations.

Population differences in benzodiazepine sensitive male scent-induced analgesia in the deer mouse, Peromyscus maniculatus

If dispersal is reduced on islands, then the demography of island populations of deer mice should be different from that of mainland populations. Areas of 1.1 hectare were periodically cleared of mice on Samuel Island (206 ha) and Saturna Island (3, 102 ha) in the Gulf Islands of southwestern British Columbia. A similar experiment was conducted on the mainland at Maple Ridge, British Columbia. The average density of mice per hectare on Saturna (43.5) was 2° that on Samuel Island (22.0) and nearly 2.5° higher than that on the mainland (18.7). The reproductive rate, as measured by length of breeding season, number of successful pregnancies, proportion of breeding animals, and number of recruits surviving to breed, was much higher on Samuel Island that on either Saturna or the mainland. Survival was lowest on Samuel Island, with little difference between the mainland and Saturna Island. Mice on Samuel Island grew more than 5° faster than mainland animals, and Saturna growth rates were double those on the mainland. Dispersal (or colonization) rate was reduced on the two islands compared with that on the mainland. Recruitment of young animals occurred throughout the breeding season on the islands but was delayed until the end of breeding on the mainland. Seasonal changes in aggressiveness of the adult population may be sufficient but are not necessary to determine breeding density in deer mice. Regulatory processes in populations of Peromyscus maniculatus may be different on islands.

The rufous hare-wallaby (Lagorchestes hirsutus) is now extinct in the wild on mainland Australia, but survives in captivity. However, endemic populations persist on Bernier and Dorre Islands, Western Australia. This study aimed to compare the genetic diversity and differentiation amongst three remaining rufous hare-wallaby populations using mitochondrial DNA (mtDNA) (cytochrome b, control region) sequence data and nuclear (microsatellite) markers. Levels of microsatellite diversity were low in both island populations but high in the captive mainland population. Levels of mtDNA diversity were low in all three populations. The mainland and island populations of L. hirsutus were found to be significantly differentiated for both microsatellite and mtDNA data, but the two island populations were significantly differentiated only for the microsatellite data. This pattern of differentiation is not consistent with the recognition of two separate island subspecies, but we recommend that the mainland and island populations be regarded as separate subspecies. The low diversity of the island populations and differentiation between island and mainland populations presents both challenges and opportunities for future management.

To understand the effects of island and mainland on life history traits (e.g. longevity, age at maturity, body size and body mass) of the smooth newt Lissotriton vulgaris (Linnaeus, 1758), we generated data on age and body measurements for an island (Bozcaada) and a mainland (Çanakkale) population in northwestern Turkey. Age was determinated by skeletochronology. The maximum life span was 4 years in the island population and 5 years in the mainland population. Age at maturity was estimated to be 2–3 years for both sexes and the populations. While mean snout-venth length (SVL) was calculated as 35.56 mm for females and 31.95 mm for males in the mainland population, it was found to be 32.83 mm for females and 31.78 mm for males in the island population. Females from the mainland population were found to be significantly larger and heavier than those of the island population. A significant positive correlation was found between SVL and age for only males in the mainland population. Unexpectedly, age was not correlated with body size for the island individuals. Since there were hardly any demographic studies on any Urodela species comparing island and mainland populations, in this paper we contribute to the literature on this subject.

Island populations are an interesting dichotomy in conservation biology. On the one hand, they can be a refuge for species where mainland populations have been decimated by loss of habitat and predation by exotic predators. On the other hand, island populations usually have reduced genetic diversity and are more susceptible to extinction through genetic and demographic processes. Genetic variation and morphological characters were measured for island and mainland populations of Parantechinus apicalis, small dasyurid marsupials, restricted to southwest Australia. Genetic variation at seven microsatellite loci revealed low levels of heterozygosity (He= 0.20−0.44) and high levels of inbreeding (Fe= 0.40−0.72) in island populations compared with the mainland population (He= 0.73). A nested clade analysis revealed that allopatric fragmentation was probably responsible for the association between geographical location and control region haplotypes, which is consistent with the isolation of populations on islands and indicative of two main populations of P. apicalis representing separate conservation units for management. While these results are typical of many island populations, they have important implications in terms of the conservation of threatened species in Australia and around the world, where island populations are a common source of founders for captive breeding and translocation to mainland sites.

BackgroundBody size exerts a strong influence on the physiology, morphology, ecology, and evolution of other life history traits in vertebrates. We compared the morphometry and allometry of two lizard species (Anolis nebulosus and Aspidoscelis lineattissima) occurring on mainland and island populations on the Pacific Coast of Mexico in order to understand the effect of an insular environment on body size and other morphometric structures.ResultsResults showed that both males and females of A. nebulosus from San Pancho Island were larger in body size than those from the mainland. Moreover, males of A. lineattissima from Cocinas Island exhibited larger forms of most measured morphometric traits than those from the mainland, whereas females from both island and mainland populations did not differ in body size or in other morphometric traits analyzed. Multivariate allometric coefficients of males and females of A. nebulosus from island and mainland populations showed a lower percentage of positive allometries than in A. lineattissima, probably because the former species is highly sedentary. Island populations of both species exhibit male-biased sexual dimorphisms in body size and size-adjusted morphometric traits. In contrast to the mainland population, morphometric comparisons of body size-adjusted traits showed that male A. lineattissima were larger than females only in head length, head width, forearm length, and tibia length, whereas in A. nebulosus, sexual dimorphism was observed just in HL.ConclusionsThis study supports the hypothesis (island rule) that vertebrates on islands are larger than those of conspecifics on the mainland. In addition, sexual dimorphism observed between males and females of both species and populations could be associated with allometric growth (positive or negative) from some morphometric structures, as well as differences in the growth rates of these organisms.

The introduction of species into new ecosystems, especially in small and isolated regions such as islands, offers an excellent opportunity to answer questions of the evolutionary processes occurring in natural conditions on a scale that could never be achieved in laboratory conditions. In this study, we examined the Mexican red rump tarantula Brachypelma vagans Ausserer (Mygalomorphae: Theraphosidae), a species that was introduced to Cozumel Island, Mexico, 40 years ago. This introduction provides an exceptional model to study effects such as morphological variation between island populations and those on the mainland in open habitats facing the island. Intraspecific variation related to the color polymorphism was compared. The aim of this study was to determine the phenotypic differences between continental populations of B. vagans and the introduced population on Cozumel Island. Phenotypic difference was evaluated using two approaches: 1) comparison of the morphometric measurements of adult and juvenile individuals at the local scale and between continental and island populations, and 2) comparison of individual color polymorphism between mainland and island populations. Two locations were sampled within the continental part of the Yucatan peninsula and two on the island of Cozumel. The number of samples analyzed at each site was 30 individuals. The morphometric results showed significant differences between continental and island populations, with bigger individuals on the island. In addition, three new variations of the typical color pattern of B. vagans recorded so far were observed. This study opens the door to further investigations to elucidate the origin of the phenotypic variation of the isolated individuals on Cozumel Island. Also, the widest range of color morphs found for a tarantula species is reported.

The systematic affinities of Boullanger Island and Western Australian mainland populations of the Grey-bellied Dunnart Sminthopsis griseoventer were investigated using allozyme electrophoresis and phylogenetic analysis of a 404 bp region of the mitochondrial DNA (mtDNA) control region. Forty-six allozyme loci were screened for variation and found to be monomorphic in S. griseoventer from both Boullanger Island and one mainland population. Low levels of variation were also detected in the control region sequence, with just one haplotype observed among eight island individuals and three haplotypes among 10 mainland individuals, each differing at between two and four nucleotide sites (0.5?1.0% divergence). Phylogenetic analyses using maximum parsimony of control region sequence from mainland and island taxa, and four species of the "murina complex", S. aitkeni, S. murina, S. gilberti and S. dolichura, indicated that the island and mainland taxa formed a monophyletic group to the exclusion of the other "murina complex" species, but were paraphyletic at the level of the individual haplotypes. These results are consistent with the hypothesis that mainland and island populations of S. griseoventer comprise a single species, and suggest that there has been no long-term barrier to gene flow between these populations. Analyses of molecular variation provided evidence the island population represents a separate management unit for conservation, but are insufficient to determine whether there has been inbreeding or a recent bottleneck in the island population.

Island populations of small land vertebrates frequently exhibit insular gigantism, presenting with larger body sizes in comparison to mainland counterparts. While insular gigantism has been observed globally, the effects of biogeographic and ecological factors on body size in island systems are not well understood. Here, we examine the biogeographic and ecological associations of insular gigantism. Deer mice (Peromyscus maniculatus) were live trapped and body mass measured on six of the Gulf Islands and the nearby mainland of British Columbia, Canada. Biogeographic measures of land area and island distance from the mainland and the ecological measure of predator species richness were used in piecewise structural equation modelling to identify associations with insular gigantism. We found evidence of insular gigantism in the Gulf Islands system, with island mice having a larger mean body mass than mainland populations. Land area was positively associated with predator species richness, and predator species richness had a strong negative effect on deer mouse body mass, resulting in the observed pattern of insular gigantism. The concurrent analysis of biogeographic and ecological factors contributes to a better understanding of the evolution of insular gigantism in small vertebrates and its juxtaposition to the phenomenon of insular dwarfism of large vertebrates.