Discussion
In the present study, high-density populations had lower recruitment and proportion of reproductive conditions than low-density populations, and the number of founders was negatively associated with recruitment rates and the proportion of reproductive voles. Because offspring born in enclosures were moved to the laboratory to use to examine the effects of density-induced maternal stress on offspring phenotype (Bian et al., 2015; Yang et al., 2018), the negative relationship between founder numbers and reproductive traits was only due to the suppressive effects of founder numbers on reproduction. In addition, because the enclosures were isolated from common vole predators, predator-induced density dependence on demographic processes was excluded from our study. Therefore, our results concluded that high-density can induce the effect of density dependent reproduction, which corroborates other studies on vole populations as described in the introduction section, indicating a universal of density-dependent reproduction in the population of microtine rodents.
In our previous papers we have reported that high-density population in both years had higher FCM level than low-density population (Bian et al., 2015; Yang et al., 2018). In the present study, we also revealed that FCM levels were positively associated with founder numbers, which also corroborates recent studies on other mammalian species (Boonstra & Boag, 1992; Novikov & Moshkin, 1998; Viblanc et al., 2014). However, Harper and Austad (2004) and Charbonnel et al. (2008) did not find a positive correlation between density and FCM levels in water voles and red-backed voles (Clethrionomys gapperi ). In those studies, the faeces were sampled in both breeding and non-breeding season (Harper & Austad, 2004; Charbonnel et al., 2008) and some samples were from individuals of different ages and reproduction conditions (Harper & Austad, 2004). However, our experiment was performed during the breeding season and the faecal samples of pregnant individuals were not collected, due to pregnancy naturally raising glucocorticoid levels in most mammals, not necessarily as a result of stress but for developmental and energetic reasons (Boonstra & Boag, 1992; Edwards & Boonstra, 2018; Edwards et al., 2019). In addition, our previous study has validated the effectiveness of detecting corticosterone levels in the faeces of root voles (He et al., 2013). Therefore, our experiment excluded the confounding effects of reproductive condition, seasonality, and the effects of trapping/handling stress on FCM levels and the differential between our and those results may be due to these confounding factors.
In addition, for unknown reasons, when founder voles were introduced into enclosures to establish an enclosed population, a few voles died during the two-week period of acclimation to their new environments. At the first trapping session in 2012 and 2015, the mean number of founders in high-density treatments was 54.5 and 53.8 voles per enclosure (363 and 358 vole ha-1), respectively. In this study area, Sun et al. (2002) reported that population density of root vole was 356 vole ha-1 in October in the habitat with mainlyE. nutan s where grazing activities were limited. Moreover, Rodd and Boonstra (1984) reported that the high density of meadow voles (M. pennsylvanicus ) population in abandoned farmland reached 469 vole ha-1 in October. Although we realise that the high-density treatments (400 vole ha-1) were the approximate peak density of natural vole populations in this study area, it is not an unseen natural extreme level. Thus, our findings of correlation between FCM levels and density can represent what happens in a natural vole population.
In semelparous and partially semelparous species (i.e., Australian semelparous marsupial, the arctic ground squirrel Spermophilus parryii plesius ) and dominant individuals in social species and cooperative breeders (i.e., the African wild dog Lycaon pictus , dwarf mongoose Helogale parvula , and grey wolf Canis lupus ), dominance was strongly correlated with reproductive success. However, dominance was also associated with heightened glucocorticoid levels, which reflected either a classical trade-off of reproductive success for potential survival during short mating periods (2–3 weeks), only one breeding period followed by programmed death (see review by Boonstra, 2005), or an adaptive stress response for competing intensely for access to females and maintaining dominant status by frequent physical aggression and challenges (see review by Sapolsky, 2005). Conversely, iteroparous voles breed continuously throughout the breeding season, and antagonistic interaction is not particularly severe. For example, breeding male meadow voles do not appear to be engaging in costly aggressive acts to assert dominance or access females (Edwards et al., 2019). Dominant status is maintained by cues rather than physical combat; subordinates tend to have the highest indices of stress. Thus, for these iteroparous species, elevated corticosterone induced by high population density can suppress reproduction (Wingfield & Sapolsky, 2003; Boonstra, 2005). In fact, antagonistic behaviour is not the only factor that induces stress responses; increased population or breeding density can also lead to an increase in parasite load, attraction of predators, and food shortage. These factors can induce stress responses in individuals (Creel et al., 2013). Thus, the density-induced stress responses are an additive stress effect of intrinsic and extrinsic factors, reflecting the biological cost of cumulative stress responses (Goymann & Wingfield, 2004) rather than a single factor effect (i.e., antagonistic behaviour).
In the present study, we found that the influence of female- and male-founder vole numbers on the proportion of reproductive condition was mediated through FCM levels in 2012, and that female-founder number indirectly negatively affected recruitment through FCM levels in 2015. Although the influence of FCM levels on reproductive traits had different pathways in both years, we still conclude that density-induced stress participates in density-dependent reproduction; also, density-induced stress is one of the factors generating density-dependent reproduction effects.
In conclusion, in the present study, high density generated density-dependent reproduction and increased FCM levels of founders. The negative effect of high density on reproduction was partly through its positive effects on FCM levels of founder voles. Thus our results provide the first evidence that density-induced stress is one ecological factor generating density-dependent reproduction effects in vole populations.
Author contribution : Jianghui Bian and Yan Wu conceived and designed the experiments. Shouyang Du, Yanbin Yang and Yifan Cao performed the experiments. Jianghui Bian and Guozhen Shang analyzed the data. Guozhen Shang and Jianghui Bian wrote the manuscript. Yan Wu provided editorial advice.
Acknowledgements: We thank Hui He, Xuheng Nie, and Xin Zhang for their assistance in the field. This work was funded by the National Natural Science Foundation of China (Grant Numbers 31870397, and 31570421), the Natural Science Foundation of Qinghai Province (Grant Number 2018-ZJ-906), and the Strategic Priority Research Program of the Chinese Academy of Sciences (Grant/Award Number XDA2005010406). The use of animals in this study was in accordance with the guidelines of the regulations of experiments on animals and was approved by the Animal Ethics and Welfare Committee of the Northwest Institute of Plateau Biology, Chinese Academy of Science.