Introduction
Small mammals in the Northern Hemisphere often have a broad spectrum of population dynamics and seasonal fluctuations. Any pattern of population dynamics is caused by changes in demographic parameters (Inchausti et al., 2009). Thus, knowledge regarding the factors influencing demographic parameters is of great importance for understanding population dynamics.
In iteroparities such as microtine rodents, density dependence in reproduction plays an important role in stabilising population dynamics via immediate negative feedback from population density to reproductive output (Ostfeld et al., 1993). Although research regarding density dependence has often been criticised (Krebs, 2002), density-dependent reproduction does occur in seasonal territorial microtine rodents with strong social effects on demographic parameters during breeding season, such as the grey-sided vole (Clethrionomys rufocanus ) (Saitoh, 1981), meadow vole (Microtus pennsylvanicus ) (Ostfeld et al., 1993), water vole (Arvicola terrestris ) (Saucy, 1994), common vole (Microtus arvalis ) (Inchausti et al., 2009) and bank vole (Clethrionomys glareolus ) (Koskela et al., 1999) and Yangtze vole (M. fortis calamorum ) (Zhang et al., 2010). The mechanisms underlying density-dependent reproduction are associated with direct and indirect interactions among individuals, such as consumptive and non-consumptive competition (Rödel et al., 2004), as well as social suppression for juvenile maturation (Saitoh, 1981).
Christian (1971) suggested that density-dependent reproduction might be the result of increased adrenocorticotropic hormone secretion in response to increased density and decreased or suppressed reproduction. However, Christian’s research has been criticised because it was largely performed on captive animals in a laboratory setting (Krebs & Myers, 1974; Lee & McDonald, 1985); moreover, there has been limited evidence supporting their hypothesis. Nonetheless, studies have shown that factors known to influence density dependence of reproduction, such as social interference and antagonistic interactions, are normally modulators of individual stress responses in the population and act as stressors to activate the hypothalamic-pituitary-adrenal axis and subsequent secretion of glucocorticoids (GC), a critical hormone in stress response (Sheriff et al., 2009). Studies on free-living, non-social, territorial vertebrates demonstrated that population density is usually positively associated with GC levels (see review by Creel et al., 2013, but see Harper & Austad, 2004; Kuznetsov et al., 2004; Charbonnel et al., 2008). In voles specially, increase in population density was associated with an increase in corticosterone levels (Boonstra & Boag, 1992; Novikov & Moshkin, 1998; Bian et al., 2015). For some social mammals living in groups, dominant individuals have higher reproductive success and heightened GCs level (see review by Sapolsky, 2005); however, evidence from laboratory studies and free-ranging iteroparous territorial rodents have shown that chronically elevated GC concentrations inhibit the gonadal axis function and lead to a negative effect on reproduction (see review by Creel et al., 2013). Thus, density-dependent reproduction in microtine rodents may be the result of elevated corticosterone concentrations in response to increased density.
In the present study, we investigated the effects of density-induced stress on reproduction via manipulation of the population density in the root vole (Microtus oeconomus ). This is part of a larger multifaceted project examining the effects of density-induced maternal stress on population dynamics. Parental enclosures for breeding offspring were established by introducing six and thirty adults of each sex in low- and high-density enclosures, respectively, in 2012 and 2015. Our previous papers have reported that high parental population density was approximately three times higher compared with low parental population density at the end of the experimental period in 2012 (Bian et al., 2015). Moreover, the population under high-density conditions showed higher mean faecal corticosterone metabolite (FCM) levels than the population under low-density conditions in both 2012 (Bian et al., 2015) and 2015 (Yang et al., 2018). In this study, we present unpublished data from the same parental populations in 2012 and 2015, which include recruitment and proportion of reproductive conditions and the numbers of male and female founders in both years. We separately tested the difference in reproductive traits between high-and low-density parental populations, and analysed how the number of founders and FCM levels affected reproduction by the recursive model in the structural equation model (SEM). We aimed to test the hypothesis that density-dependent reproduction in voles may be due to the suppressive effects of density-induced stress on reproduction. We predicted that a high-density population would have lower recruitment and the proportion of reproductive condition, and the effects of founder number on both reproductive traits would be mediated by FCM levels.