A mating system is an important life history for animals dealing with changing environments. Population density affects the plasticity of a mating system and subsequently the family structure of animals, but its impacts on mating systems and social structures are rarely investigated by using molecular markers in field conditions. In this study, using microsatellite genetic markers, we examined the changes in the social and genetic mating system and family structure of Brandt’s voles in the grassland of Inner Mongolia, China, under low-, medium-, and high-density enclosures (each enclosure 0.48-ha with 4 replicates.) We found, that with the increase in population density of the founder voles introduced into the enclosure in early spring, both sexes increased their number of genetic mating partners, while males increased their social partners, resulting in a more promiscuous mating system. The number of genetic fathers and mothers per family, the number of social offspring per founder male or female, and the proportion of extra-group offspring increased with increased density, indicating that the response of the family structure to density change which is consistent with the change in mating system observed in our study. Both male and female voles had multiple mates, but males had a larger number of social and genetic mates, the number of social and genetic offspring, and the number of social and genetic families. Our study highlights the significance of density-dependent plasticity of a mating system and family structure in affecting the population change of small mammals.

The study site had pre-constructed twenty-four 0.48-ha enclosures (80 × 60 m) with galvanized iron sheets extending 1 m below the ground’s surface and 1.4 m above the surface to prevent escaping, intrusion and movement of burrowing rodents into, out of, and between enclosures (Li et al., 2016). A raptor‐proof nylon netting (10 cm mesh size) covered the top of each enclosure to obstruct avian predators. The integrity of each enclosure’s construction was regularly checked and maintained.

Twelve enclosures were randomly assigned to one of three treatments that differed in founder population size: Low Density (6 ♂:6 ♀), Medium Density (12 ♂:12 ♀) and High Density (18 ♂:18 ♀). Each treatment had four replicates. The density level was based on a previous test in which 13-15 pairs of male and female voles were released into each enclosure in April (Li et al., 2016). The highest population density of an enclosure was recorded in one of the high-density enclosures at 138 individuals by the end of the breeding season (autumn) which translates to 287.5 voles/per ha. It was lower than the highest density recorded in natural conditions of the same area (590-2300 voles/ha) (Zhong et al., 2007; Wan et al., 2014).

We captured overwintering adults of Brandt’s voles weighing ≥ 40 g for enclosure tests. The experimental voles were trapped and caught from the wild (See section for animal trapping below). They were sexed, weighed, and injected with a unique PIT (Passive Integrated Transponder) tag under their skin using the Mini RFID Glass Capsule Tag (1.25 × 7 mm) in the laboratory at the field station where they were kept and observed for five successive days before they were released into the experimental enclosures. The voles were raised in the laboratory under the condition of natural light and room temperature. Rodent chow and water were supplied ad libitum.

On April 17, 2018, a total of 288 voles (144 females) were introduced into three different density settings as the founding population. We randomly chose one male and one female vole and released them into one burrow. Before their release, we verified their sex, and weight, scanned their PIT codes, and collected a tail sample (2-3 mm) using sanitized and sharp scissors for DNA analysis following the “Institutional Animal Use and Care Committee Guidebook, Institute of Zoology, Chinese Academy of Sciences (CAS IAUCC)”. We released the paired voles close to an opening of a burrow and recorded their released location. The enclosures contained pre-existing burrow systems from the previous year’s experiment and were divided into a grid of 70 grid cells. We used these cells to record and track the movement of all voles in the enclosure. Before releasing the founder voles, all voles living in the enclosures from previous studies were removed using 80 live traps in each enclosure. Later, a survey found that most of them lived in burrows close to each other, not in the same burrow where they were released, indicating they re-selected their partners. 

Data can be opened using Microsoft Excel and other similar software such as LibreOffice Calc.