Relationship between the Social Structure and Potential Reproductive Success in Muroid Rodents (Rodentia, Myomorpha)

In many systematic groups of mammalian species, the evolution of sociality leads to the formation of large social groups (group-size evolution). In rodents, however, the most social species live in family groups consisting, as a rule, of a fairly small number of individuals. The family-group lifestyle of some terrestrial sciurids (Sciuridae) and voles (Microtus) results in fewer breeding adult females, a decreased litter size, and a longer time before the first reproduction. Statistical analysis of published data performed in this study showed that the transition to the family-group lifestyle leads to decreased potential reproductive success in many muroid rodents: females produce fewer litters and, consequently, fewer pups during the breeding season. However, the potential reproductive success in muroid rodents with a family-group lifestyle could be increased due to delayed offspring dispersal (who spend the severe winter period in family groups) and extensive cooperation in foraging and in defending their territory and food reserves, as well as in care-giving activities. It can be assumed that females of these species produce fewer offspring, yet of “better quality.” Ultimately, social rodent species may gain inclusive fitness benefits because of the higher fitness of their offspring as compared to solitary dwellers and gregarious species.


INTRODUCTION
According to existing concepts, the evolution of societies is aimed at providing certain advantages to individuals uniting in groups (Alexander, 1974). In other words, a group lifestyle, which is inevitably associated with such negative factors as an exacerbation of intraspecific competition and an increase in the risk of the spread of infectious and parasitic diseases, nevertheless, should be associated with an increase in fitness, the measure of which is reproductive success (Wolff and Sherman, 2007).
In many taxonomic groups of mammals (including primates, carnivores, and ungulates), the evolution of sociality is directed towards the formation of large groups (group-size evolution) (Lee, 1994;Grove, 2012). However, in rodents, the most social species are those with family groups, which usually unite a relatively small number of individuals (Gromov, 2013(Gromov, , 2017. The family-group lifestyle of the majority of rodent species is characterized by long-lasting pairbonds, active participation of males in raising offspring, and a complex social organization, which is expressed in a hierarchical system of relations, the distribution of behavioral roles, differentiation of reproduction, and widespread cooperation. In other words, the evolution of sociality in rodents is closely associ-ated with the formation of a complex social structure and social organization (Blumstein and Armitage, 1998).
However, the increasing complexity of social organization, as was shown in some studies (in particular, on terrestrial sciurids (Sciuridae)), has an adverse effect on individual fitness, because the reproductive potential of the most social species decreases: the proportion of adult females participating in breeding decreases, the number of pups in litters decreases, and the age of females starting to breed for the first time increases (Blumstein and Armitage, 1998;Armitage, 2007). A decrease in the number of pups in litters was also noted in some species of voles of the genus Microtus (M. pinetorum and M. ochrogaster), which are characterized by a family-group lifestyle (Innes, 1978). In view of the above, a question arises as to whether a decrease in the reproductive potential is a consequence of the transition to a family-group lifestyle in other rodent species as well.
Analysis of published data shows that, in addition to large terrestrial sciurids (Cynomys spp. and Marmota spp.), as well as some species of the tribe Hystricomorpha, the family-group lifestyle is also typical of some representatives of muroid rodents (Gromov, 2008). No one has previously tested whether there is a relationship between social structure and reproductive potential in these rodents. In this regard, in this study, an attempt was made to compare the reproductive potential in a number of Myomorpha species belonging to different categories of sociality (in other words, with different types of the spatial-and-ethological population structure, SEPS) (Gromov, 2008). In rodents, four main types of SEPSs can be distinguished: type I (solitary dwellers), type II (a system of aggregations of adult heterosexual individuals), type III (societies with weakly consolidated family groups), and type IV (societies with structured family groups). According to this classification, species with type IV SEPS are characterized by the most complex social structure and, therefore, are at the top of sociality.

MATERIALS AND METHODS
The material for this study was published data on the reproductive potential of 45 species of muroid rodents (Tables 1, 2), namely, the average number of litters per adult female in the breeding season and the average number of pups in a litter. Due to the heterogeneity of the published data (the number of litters, as well as the number of pups in the litter, varies in different sources), the arithmetic mean value of the parameters studied was taken as the basis for statistical analysis (shown in parentheses in Tables 1 and 2). On the basis of these data, an additional index was calculated-the average number of pups produced by the female during the breeding season, which was defined as the product of two values -the average number of pups in a litter and the average number of litters.
The data obtained were processed by statistical methods of analysis of variance (one-way ANOVA) and regression analyzes using Pearson's correlation coefficient (Sokal and Rohlf, 1995).

RESULTS AND DISCUSSION
Analysis of variance of the data shown in Tables 1 and 2 allows us to postulate that there is a strong correlation between the types of SEPS (and, accordingly, the social structure) and the reproduction indices such as the average number of litters and the total number of pups produced by the female during the breeding season: for the first index, F = 6.568, df = 3.41, p < 0.001; for the second index, F = 6.803, df = 3.41, p < 0.001.
The results of the regression analysis are completely consistent with the results of ANOVA: in the series of species compared, upon the transition from type I SEPS to type IV SEPS, the number of litters of females, as well as the total number of pups produced by one female during the breeding season, decreases (Figs. 1, 2).
The regression equation for the second index (the total number of pups (N) produced by one female during the breeding season) is expressed by the formula N = 24.974 -3.431x (1-4), where 1-4 is the SEPS type (R = -0.56, F = 20.061, df = 1.43, p < 0.001).
On the basis of the calculations, it can be asserted that, in the weakly social species with type I SEPS, females during the breeding season produce, on average, 3.5 litters containing, on average, 21.5 pups. In the social species with type IV SEPS, these indices are 2.2 and 11.3, respectively. Thus, the transition to the family-group lifestyle in muroid rodents, indeed, is associated with a decrease in reproductive potential: females produce fewer litters and, accordingly, fewer pups per breeding season.
Conversely, in the species with a family-group social organization and delayed dispersal of young animals, the rates of reproductive success may be higher. This is favored by the delay in the dispersal of young animals that survive the harsh winter period in a family group, as well as by the widespread cooperation, including the protection of the territory from other conspecific individuals, foraging, protection of food reserves, and raising offspring, in which not only parents but also grown immature individuals from older litters take an active part (Gromov, 2013;Solomon and Keane, 2007). Under these conditions, the mortality of young animals is reduced to a minimum and, accordingly, the total fitness of breeding individuals increases.
In many species with a family-group social organization, the dispersal of the major part of the young individuals takes place only after the first wintering, and in some species of marmots (Marmota spp.) and beavers (Castor spp.), it takes place in the third or even fourth year of life (Barash, 1974;Suntsov, 1981;Dezhkin et al., 1986;Armitage, 2007). In this case, young individuals are prepared for dispersal and independent existence much better than in the species with early  Kambulin, 1941;Dubrovskii, 1978;Sokolov et al., 1983 Meriones unguiculatus 2-3 (2.5) 4.5-6.5 (5.5) Leont'ev, 1954;Khamaganov, 1954;Lapin, 1989  dispersal of young animals. Thus, it can be postulated that the species with a family-group social organization produce fewer offspring, yet of a "better quality." The low mortality rate of young individuals in family groups and their delayed dispersal undoubtedly enhance the individual fitness. Eventually, the inclusive fitness (Hamilton, 1964) in the social species may be higher than in solitary dwellers. However, special studies are required to confirm this assumption.

COMPLIANCE WITH ETHICAL STANDARDS
The authors declare that they have no conflict of interest. This article does not contain any studies involving animals or human participants performed by any of the authors.