Individual dispersal plays an important role in preserving genetic diversity in density-fluctuating populations of arvicoline rodents. When habitats are fragmented and dispersal between habitats is severely constrained, genetic diversity can be lost. Here, I investigated whether genetic diversity in the gray-sided vole Myodes rufocanus was preserved in an intensive farming region in Japan, where voles inhabited isolated windbreak forests along the borders of plowed lands. Genetic structure was examined in 673 vole samples (330 in spring and 343 in fall) collected at 34 windbreak forests located 0.35–20 km apart. A part of the control region (425 bp) of mitochondrial DNA (mtDNA) was sequenced in 673 voles, yielding 76 haplotypes. Genetic differentiation of maternally inherited mtDNA among trapping sites was markedly lower in males than in females in both seasons, indicating strong male-biased dispersal. Genotypes at six microsatellite DNA loci were determined in 494 voles (245 in spring and 249 in fall) from 18 trapping sites, and loci harbored 16–24 alleles. The mean number of alleles per locus (allelic diversity) at trapping sites was positively correlated with the number of examined individuals (density) in both seasons, and the relationship was very similar to that of a previous study performed in much less fragmented populations. Genetic differentiation of microsatellite DNA among trapping sites decreased considerably from spring to fall. In a STRUCTURE analysis with a most probable cluster number of two, closer trapping sites showed more similar mean values of cluster admixture proportions. The present findings indicate that gene flow among isolated windbreak forests, which occurred mainly by dispersal of males, was not restrained in this intensive farming region. Furthermore, the results suggest that genetic diversity in the study population was preserved as well as in less fragmented populations.

1. Tissue samples were collected two times (spring and fall) in 2007 at 34 trapping sites in Tokachi Plain in Hokkaido, Japan. During each trapping session, 30 Sherman-type traps were arranged in a 3 × 10 grid pattern at an interval of 10 m (~0.3 ha) for 2 nights (i.e., a total of 60 trap-nights per session). Upon capture, three toes (one per foot) of each vole were clipped, sex was determined, and the vole was released.

2. Genomic DNA was extracted using the conventional phenol-chloroform method. 

3-1. For mtDNA haplotype data, partial nucleotide sequences of the mtDNA control region were sequenced as described previously (de Guia et al., 2007). Here, the most variable part (425 bp) was sequenced using the internal primer HIP8 (de Guia et al., 2007).                              

   de Guia, A. P., Saitoh, T., Ishibashi, Y., & Abe, H. (2007). Taxonomic status of the vole in Daikoku Island, Hokkaido, Japan: examination based on morphology and genetics. Mammal Study, 32(1), 33–44. doi:10.3106/1348-6160(2007)32[33:TSOTVI]2.0.CO;2

3-2. For microsatellite genotype data, six microsatellite loci (MSCRB-01, -04, -07, -09, -11, and -13) were genotyped as described previously (Ishibashi and Takahashi, 2021).

   Ishibashi Y, Takahashi K. (2021) Role of individual dispersal in genetic resilience in fluctuating populations of the gray-sided vole Myodes rufocanus. Ecology and Evolution, 11(7), 3407–3421. doi:10.1002/ece3.7300.

Microsoft Excel, CLUSTALW, Network