Highly masculinized and younger males attain higher reproductive success in a social rodent

Highly masculinized and younger males attain higher reproductive success in a social rodent Abstract Alternative morphotypes have been reported in males of different taxa. In some mammals, highly masculinized and slightly masculinized males represent 2 opposite ends along a gradient of phenotypic variation in males. This phenotypical gradient originates during prenatal development. Laboratory studies have documented how highly and slightly masculinized males differ in several traits, including their reproductive success. However, the extent to which these reported differences materialize in natural populations remains unknown. We quantified the impact of male morphotype on male reproductive success in a natural population of Octodon degus, a highly social rodent. We assessed male morphotype through a continuous gradient of anogenital distance. We also tested the hypothesis that the social environment interacts with male morphotype to influence male reproductive success. We found that individual attributes, including masculinization level and age, impacted male reproductive success. Highly masculinized and younger males had greater reproductive success. Additionally, male body weight had a small magnitude but positive effect on male reproductive success. Male reproductive success was not affected by social attributes such as group composition. Thus, the number of males and females within a group did not affect male reproductive success, nor did the average male anogenital distance within a group. Our results support the hypothesis that the prenatal environment can result in long-term effects on individual life history and cause intrasexual phenotypical variation in natural populations. Our findings suggest that male phenotypical masculinization could be an adaptive trait, regardless of the social environment. INTRODUCTION Male phenotypic variation, characterized as the coexistence of 2 or more types of males in the same population (Engqvist and Taborsky 2016), occurs in wide range of taxa including mites, insects, crustaceans, fishes, amphibians, reptiles, and birds (Gross 1996). This variation may involve extreme differences in behavior, morphology, physiology, and reproductive traits. For example, conspicuous male phenotypes are associated with higher concentration of androgens and territorial tactics to monopolize reproduction. In contrast, inconspicuous male phenotypes are associated with low androgen concentrations and nonterritorial parasitic tactics (Knapp 2004; Taborsky et al. 2008; von Kuerthy et al. 2016). Previous studies on species with multiple male phenotypes indicate that alternative male morphotypes may be influenced by genetic or environmental mechanisms, the distribution of traits in a population could be discrete or continuous, and male morphotypes can exhibit alternative reproductive tactics linked to differences in fitness payoffs (Gross 1996; Moore et al. 1998; Rhen and Crews 2002; Knapp 2004; Taborsky et al. 2008). In male mammals, alternative morphotypes generally have been linked to different levels of male masculinization (Clark and Galef 1998; Kaiser and Sachser 2001; Rhen and Crews 2002; Kaiser et al. 2003; Roff et al. 2017). This masculinization gradient is a consequence of 2 proximate mechanisms. First, alternative male morphotypes may result from maternal stress responses during pregnancy, in which the male embryo/fetus is exposed to 1) a lower concentration of androgens, 2) less powerful androgens (e.g., DHEA), and/or 3) exposure to androgens but outside of the most sensitive period (Ward 1972; Ward and Weisz 1980; Kaiser et al. 2003). These conditions have been reported to produce a male with inconspicuous masculine traits (henceforth slightly masculinized male). A second mechanism corresponds to the intrauterine position phenomenon (IUP), where position of embryos/fetuses alters the male phenotype through the influence of androgens released by neighboring male siblings. A male embryo/fetus that would develop between 2 males would be exposed to higher concentration of androgens and therefore develop into an adult with exacerbated male traits (henceforth highly masculinized male). In contrast, a male embryo/fetus that would develop between 2 female siblings would be exposed to low androgen concentrations, and therefore develop into a slightly masculinized male (vom Saal 1989; Clark et al. 1992; vom Saal et al. 1999; Ophir and delBarco-Trillo 2007; Roff et al. 2017). Taken together, prenatal exposure to high or low concentrations of androgens may result in litter and population gradients of male offspring masculinization that persist through adulthood (Clark and Galef 1998; vom Saal et al. 1999). Highly masculinized and slightly masculinized males differ in several traits, including anogenital distance (Clark and Galef 1998; vom Saal et al. 1999; Ryan and Vandenbergh 2002; Ophir and delBarco-Trillo 2007). Given that prenatal exposure to androgens affects the development of perineal tissue, distance between the penis and anus is longer in males that were exposed to high concentrations of androgens, and shorter in males that were exposed to low concentrations of androgens (vom Saal 1989; Clark and Galef 1998; Hotchkiss and Vandenbergh 2005; Ophir and delBarco-Trillo 2007; Bánszegi et al. 2015). Thus, anogenital distance allows the noninvasive assessment of male masculinization level. Additionally, highly and slightly masculinized males also may differ in behavioral and reproductive traits such as aggressiveness, dominance status, sexual behavior, attractiveness to females, home range size, dispersal behavior, territorial scent-marking, parental behavior, hormone profiles, and weight/size of testicles and secondary sex organs (Drickamer 1996; Clark and Galef 1998; vom Saal et al. 1999; Kaiser and Sachser 2001; Ryan and Vandenbergh 2002; Kaiser et al. 2003; Ophir and delBarco-Trillo 2007; Godsall et al. 2014; Bánszegi et al. 2015). Our understanding of how masculinization influences reproductive success of males is limited to laboratory studies of a few domestic species. These studies point to effects on both fertility and fecundity. Slightly masculinized males of domestic mice (Mus musculus) and Mongolian gerbils (Meriones unguiculatus) exhibit a higher percentage of unsuccessful female impregnation (Crump and Chevins 1989; Clark and Galef 1998, 2000), and highly masculinized males of Mongolian gerbils and guinea pigs (Cavia aparea f. porcellus) produce more offspring than slightly masculinized males (Clark and Galef 1992; Guenther et al. 2014; Zimmermann et al. 2017). Intriguingly, the effect of male morphotype on reproductive success may also be influenced by intermale conflict, a component of social context. Thus, Kaiser and Sachser (2009) proposed that an aggressive tactic of some males (presumably highly masculinized) could be superior in low-density conditions, when intermale conflict is low. In contrast, a less-aggressive tactic of infantilized males could be superior in large mix-sex colonies, when intermale conflict is high. Under this scenario, an infantilized male could remain cryptic in the queue, avoiding harassment from more aggressive male (Sachser et al. 2011). Whenever aggressive males disappear, infantilized males would breed with female group mates (Sachser et al. 2011; Siegeler et al. 2011; Guenther et al. 2014; Zimmermann et al. 2017). In mammals, male body weight and age can affect male reproductive success (Clutton-Brock 1989; Raveh et al. 2010). For example, the impact of body mass on male reproductive success varies between mating systems (Emlen and Oring 1977; Clutton-Brock 1989; Clutton-Brock and Huchard 2013). Thus, in species in which the access to females is determined by male fighting abilities (interference competition) large males would have advantage (Emlen and Oring 1977; Clutton-Brock 1989; Zedrosser et al. 2007; Clutton-Brock and Huchard 2013). This prediction has been supported by studies on polygynous and sexual dimorphic ungulates and pinnipeds (Clutton-Brock 1989; Zedrosser et al. 2007; Clutton-Brock and Huchard 2013; Dubuc et al. 2014). However, in species with promiscuous mating systems and without sexual dimorphism, male searching abilities and experience (scramble competition) are more important determinants of male reproductive success (Schwagmeyer 1988; Clutton-Brock 1989; Schradin and Lindholm 2011). Older males are generally larger than, dominant over, and more experienced than younger males (Clutton-Brock 1989; Zedrosser et al. 2007). Thus, older males are expected to attain higher reproductive success than younger males (Clutton-Brock 1989). However, in species with fast life-history strategies and early senescence, natural selection could favor a high reproductive investment in early reproductive events (Hamel et al. 2010; Lemaître et al. 2015). In these species, younger individuals generally make a high investment in early life, thus compromising the reproductive performance and survivorship late in the life (Dobson and Oli 2007; Hamel et al. 2010; Lemaître et al. 2015). Herein, we use a 6-year dataset on a natural population of the caviomorph rodent Octodon degus to determine how male attributes (masculinization level, body weight, age) and social conditions (solitary vs. group-living) influences direct fitness of males. Degus live in social groups, where 0–3 adult males and 1–9 adult females communally nest and rear their offspring (Ebensperger et al. 2014; Ebensperger et al. 2016). Occasionally, degus also may live solitarily. Male and female degus are sexually mature at 6 months of age. However, in our study population, degus breed annually with the primary mating season taking place during the late, austral fall (June), when females begin to cycle with ascending photoperiod (Ebensperger et al. 2013). At that stage of the season, male and female yearlings are between 9 and 10 months of age. Most adult degus (85–90%) do not survive to their second year of age, which means that reproductive success during their first (and likely only) breeding event has a major impact on lifetime reproductive success (Ebensperger et al. 2013). Previous studies indicate that male degus exhibit alternative morphotypes, and where males displaying a territorial tactic also exhibit higher levels of testosterone compared with nonterritorial males (Soto-Gamboa et al. 2005). In captivity, males may be highly or slightly masculinized (Correa 2012). Moreover, both maternal stress and the IUP phenomenon were noted to contribute to this phenotypic variation (Correa 2012; Roff et al. 2017). A link between male morphotype (assessed through the anogenital distance analyses) and reproductive success has not yet been examined. In this study, we aimed to determine sources of variation in male reproductive success. First, we hypothesized that 1) male masculinization level, male body weight, and adult male age influences male reproductive success. This hypothesis predicts that (i) the number of offspring sired by males would increase with increasing anogenital distance. Additionally, we predicted that (ii) the number of offspring sired by males would increase with increasing body weight (iii) and that the number of offspring sired by adult males would decrease as the males grow old. Second, we hypothesized that 2) regardless of male morphotype, male reproductive success is influenced by attributes of its social group. This hypothesis predicts that (iv) the number of offspring sired by males would increase with increasing number of females (i.e., a measure of mating opportunities), and would decrease with increasing number of males (a measure of male–male competition over mates) in the social group. This hypothesis also predicts that (v) solitary living males or males living with other males exclusively would suffer a cost in terms of offspring sired relative to males nesting with one or more females. Our third hypothesis was that 3) social context modulates the effect of male morphotype on male reproductive success. Herein, we expected the effect of anogenital distance on male reproductive success would depend on the morphotypes (male anogenital distance lengths) of other male group members. This hypothesis predicts that (vi) males with long anogenital distances (highly masculinized) would sire more offspring in social groups in which the mean of anogenital distance of the group (i.e., a measure of similarity between male group mates in terms of anogenital distance) is relatively short whereas, the number of offspring sired by males with relatively long anogenital distance would decrease in social groups in which the mean of anogenital distance of the group is relatively long. MATERIAL AND METHODS Study population This study was conducted between 2010 and 2015 on a natural degu population located at the Estación Experimental Rinconada de Maipú (33°23′ S, 70°31′ W, altitude 495 m), a field station of the Universidad de Chile. This study area is characterized by a Mediterranean climate with cold, wet winters and warm, dry summers (di Castri and Hajek 1976). The site consisted of open areas with scattered shrubs (Proustia pungens, Acacia caven, and Baccharis spp.) that on average covered 14.5% of the field site (Ebensperger and Hurtado 2005). The total area examined at Rinconada was 2 ha and did not vary across years of study. Total degu density was 35 ± 18.3 degus/ha in winter and 43.5 ± 41.7 degus/ha in spring. Live trapping and telemetry Live trapping and telemetry were conducted in May to July (time encompassing mating) and in September and October (time encompassing parturition, lactation, and offspring weaning) of each year. Degus are diurnally active and remain in underground burrows overnight (Ebensperger et al. 2004). A burrow system was defined as a group of burrow openings surrounding a central location spanning 1–3 m in diameter where individuals were repeatedly found during night-time telemetry (Fulk 1976; Hayes et al. 2007). Ten traps (Tomahawk model 201, Tomahawk Live Trap Company, Tomahawk, WI) were used at each burrow system daily. Traps were set prior to the emergence of adults during morning hours (06:00 h). After 1.5 h, traps were closed until the next trapping day. The identity, location, sex, anogenital distance, body weight (weighed to the nearest 0.1 g), was determined for all captured degus. At first capture, each degu received ID coded tags on each ear (Monel 1005-1, National Band and Tag Co., Newport). Adults weighing more than 170 g were fitted with 6–7 g radio-collars (AVM Instrument Co., Colfax, CA) with unique pulse frequencies. From 2010 to 2015, 155 degus (including adult males and females) were radio-tracked during the winter season, while 193 degus (including adult males and females) were radio-tracked during the spring season ((Analyses based on 2008–2016 data from the Rinconada population indicate that radio-collars do not influence survival (L.A. Ebensperger, unpublished data, Supplementary Table S1 of Supplementary Data)). During night-time telemetry, degus were tracked to their home burrows via radio telemetry. Previous studies at Rinconada have confirmed that night-time locations represent underground nest sites (Ebensperger et al. 2004). Locations were determined once per night approximately 1 h before sunrise using LA 12-Q receivers (for radio collars tuned to 150.000–151.999 MHz frequency; AVM Instrument Co., Auburn, CA) and hand held, 3-element Yagi antennas (AVM instrument Co., Auburn, CA). Previous studies have verified these radiotelemetry methods as sufficient for determining group membership (Hayes et al. 2009). The number of burrow systems monitored was 59 ± 7 in winter and 55 ± 9 in spring. The number of days that each burrow system was trapped was 36 ± 3 in winter and 50 ± 8 in spring. The number of radio-collared degus was 26 ± 10 in winter and 32 ± 8 in spring. The average number of night-time telemetry locations per radio-collared degu across all years of the study was 16 ± 0.6 in winter and 18 ± 1.4 in spring (data expressed in mean ± SD). Details per year are reported in Supplementary Table S2 of Supplementary Data. Sample size A total of 83 measures from 78 males were sampled in this study, where 5 males were sampled during 2 consecutive study years. We considered these additional measurements to be statistically independent because none of these 5 males remained in the same social group between years. Individual attributes: male anogenital distance, male body weight, and male age Male morphotype in terms of masculinization level was assessed through anogenital distance measures. Males that were close to the short end of the anogenital distance distribution were considered to as slightly masculinized males, while males that were close to the long end of this gradient represented highly masculinized males (Correa 2012). Anogenital distance was measured as the distance from the ventral anus commissure to the base of the genital papilla, as defined by Vandenbergh and Hugget (1995). We measured the anogenital distance of each adult male with a digital caliper (precision 0.1 mm) at every capture event during the mating season. All anogenital distance measurements were taken by the same observer across all 6 years of the study. To obtain a single measure per subject, all anogenital distance measures taken in the same season were averaged per male degu (Correa et al. 2016). Males that were present in 2 consecutive years had 2 estimations of average of anogenital distance, because between years, males grow. The number of anogenital measures per degu averaged 8.08 ± 4.48 (range: 2–23). Repeatability of anogenital distance and its correlation with male body weight were calculated according to the methods of Correa et al. (2016). Anogenital distance was used as a continuous predictor during all subsequent statistical analyses. Mean male body weight during the mating period was recorded from trapping data. The number of body weight measures per male during the mating period averaged 11 ± 5.84 (range: 2–25). Males in our study were either young of the year (age = 1, n = 66) or had survived until a second year (age = 2, n = 17). Five males were present in 2 consecutive years. Social group determination The main criterion used to assign individuals to social groups was the sharing of burrow systems at night. The sharing of burrow systems was determined by 1) burrow trapping during early morning activity and 2) night-time telemetry. To determine group composition, we first compiled a symmetric similarity matrix of pairwise associations of burrow locations of all adult degus during trapping and telemetry (Whitehead 2008). The association (overlap) between any 2 individuals was determined by dividing the number of early mornings that these individuals were captured at or tracked with radiotelemetry to the same burrow system, by the number of early mornings that both individuals were trapped or tracked with radiotelemetry on the same day (Ebensperger et al. 2004; Hayes et al. 2009). To determine social group composition, a hierarchical cluster analysis of the association matrix was conducted using SOCPROG software (Whitehead 2009). The fit of the data were analyzed using cophenetic correlation coefficients, correlations between the actual association indices, and the levels of clustering in the diagram. In this procedure, values above 0.8 indicate that hierarchical cluster analysis provided an effective representation of the data (Whitehead 2008). The maximum modularity criterion (Newman 2004) was used to cut off the dendrogram and define social groups. 83 males were included in this analysis because they were assigned to a well-defined social group. Additionally, 41 males were excluded from all analyses because they were not assigned to a well-defined social group. The total number of social groups (11 ± 5, for winter and 16 ± 7, for spring) and male and female group members monitored during winter of each year are reported in Supplementary Table S3 of Supplementary Data. Social group attributes Social group attributes recorded were 1) the number of adult females (a measure of male mating opportunities), 2) the number of adult males (a measure of male competition), and 3) the mean male anogenital distance within groups (i.e., a measure of similarity between male group members). These attributes were estimated once per mating season because degu social group composition is relatively stable across the breeding season (Ebensperger et al. 2009). Genetic methods and paternity analyses We genotyped a total of 1517 individuals captured between 2010 and 2015 (Supplementary Table S4 of Supplementary Data). Tissue samples (a 1 × 5 mm ear snip) were taken from each individual. Tissue samples were stored in ethanol 99% at 5–6 °C until analysis. DNA was extracted using the Reliaprep DNA animal tissue miniprep system kit (Promega) mouse tail protocol. DNA was eluted in 200 µl of nuclease-free water and stored at −20 °C. We worked with 10 microsatellite loci, 9 from O. degus (Quan et al. 2009) and one from S. cyanus (Schroeder et al. 2000). These loci were amplified via polymerase chain reaction (PCR), with the following protocol: 15 min at 94 °C for DNA denaturation, 30 cycles of a 1 min denaturation step at 94 °C, followed by 1 min of locus-specific annealing temperature (Supplementary Table S5 of Supplementary Data), 1 min at 72 °C for elongation, and a final elongation step of 10 min at 72 °C. For fragment analysis, the PCR products were mixed in 3 combinations (2 with 4 loci each and one with 2 loci). Each of these mixes was contrasted with an internal size standard and analyzed using an ABI Prism 3130Xl genetic analyzer and allele sizes were determined using the Genemapper software (Applied Biosystems). All loci amplified successfully and were polymorphic (Supplementary Table S5 of Supplementary Data). Genotypes for all individuals across years were complete with no missing data. We tested the Hardy-Weinberg observed and expected heterozygosity for each study year with CERVUS 3.0 software (Marshall et al. 1998). Deviations from Hardy-Weinberg expectations were detected in 5 out of 6 years (Supplementary Table S6 of Supplementary Data) and were not the consequence of null allele presence (all markers were checked for null alleles with Microchecker software, van Oosterhout et al. 2004). This finding was expected because our study population was open, nonpanmitic, and characterized by a relatively high level of genetic relatedness (Quirici et al. 2011; Davis et al. 2016). We used CERVUS 3.0 software (Marshall et al. 1998) to conduct paternity analyses, in which all offspring were checked against all potential fathers in the population. Confidence in parentage assignments calculations was made using the mean logarithm of odds (LOD) score option in CERVUS 3.0. Simulations were run for 10,000 cycles using allele frequency data from the entire population, with a genotyping error rate of 1% and under the assumption that 90% of the population was sampled. Paternity assignment analyses were made using strict (95%) confidence levels. Paternity was assigned when the following conditions were met: 1) the LOD score for the father-offspring pair was positive, 2) the father-offspring pair confidence level was significant, and 3) there was 0 or 1 mismatch. Paternity assignments with 2 or more mismatches were discarded. Supplementary Table S7 of the Supplementary Data displays the number of offspring that were or were not assigned to each father. Determination of reproductive success Using paternity analysis data, we determined the number of offspring sired for each male. This variable was measured at weaning, when offspring become trappable. Annual trapping ended when the daily capture rate of new offspring was less than 5% of all offspring captured (Correa et al. 2016). Statistical analyses An information theoretic approach (Burnham and Anderson 2002) was adopted to assess the inclusion of predictors in a model of reproductive success (hypothesis 1, prediction i, ii and iii; hypothesis 2, prediction iv; hypothesis 3, prediction vi). The overall procedure consisted of specifying a full model with all plausible fixed and random terms. During the first step, random effect components were retained whenever the variance explained by these components was higher or at least in the same order of magnitude compared with the residual error term. During the second step, we generated a list of competing models assembled from all possible combinations of fixed effects declared in the full model, and fit each model to the data. Models were ranked according to parsimony by means of AICc, and we selected the model with the lowest AIC value for further analysis and interpretation (see Supplementary Table S8 of the Supplementary Data). For hypotheses 1 (prediction i, ii and iii), 2 (prediction iv), and 3 (prediction vi), we used generalized linear mixed model to test whether individual and social group attributes independently influenced male reproductive success. Additionally, we included a mean male anogenital distance within group by male anogenital distance interaction. Thus, the initial model examined (model 1) was: Number of offspring sired = male anogenital distance + male mean body weight + male age (1 and 2 years old) + number of adult females + number of adult males + mean male anogenital distance within group + (anogenital distance x mean male anogenital distance within group). Year and the social group ID were added as random effects. The number of offspring sired by males was assumed to be Poisson distributed (no overdispersion was detected) and the link function used was the log-link. Outlier influence analysis and qq-plots revealed no outlier observations or departures from model assumptions. We used a generalized linear mixed model to examine how reproductive success would vary across different types of social groups (hypothesis 2, prediction v). Categories of social groups included male-only groups, multi-male/multi-female groups, and solitary males. Year and the social unit ID were added as random effects. The number of offspring sired was fit to a Poisson distribution (no overdispersion was detected). Goodness of fit for this model was obtained comparing against an intercept-only with the same random effect structure by means of Likelihood Ratio Test. We needed to examine the potential effect of type of social unit separately because solitary males lack all other social attributes included in the previous models. All statistical analyses were conducted in R 3.2.3. Generalized mixed models were fitted using library lme4 1.1–12 (Bates et al. 2015); best subset model selection was done using functions included in library MuMIn ver. 1.15.6 (Barton 2009); Variance inflation factor values were estimated using formulas in Zuur et al. (2009); and power analysis was done using library simr 1.0.2 (Green and MacLeod 2016). All libraries were accessed on March 2017. Ethical note Animal handling techniques and all protocols used in this study were approved and supervised by the Ethics Committee of the Pontificia Universidad Católica of Chile (CBB-155, 2012 resolution, supervised and approved 03/03/2015), and followed the Chilean Ethical Legislation (Permits 1–31/2009 [1956], 3881/2012, and 2826/2013 by the Servicio Agricola y Ganadero). Blood and tissue sampling were performed by well-trained veterinarians and biologists (C.L. and J.R.-E.). RESULTS Repeatability of male anogenital distance Across all 6 study years, population mean male anogenital distance was 9.5 ± 0.9 mm (SD) (range: 7.25 mm to 11.83 mm, n = 714 measurements from 78 males). Intraclass correlation coefficient (ICC) or intraseason repeatability of male anogenital distance was 0.81 (95% confidence interval: 0.75–0.86, n = 627 measurement for 78 males). Intraclass correlation coefficient (ICC) or interyear repeatability of male anogenital distance was 0.64 (95% confidence interval: 0.36–0.94, n = 87 measurement for 5 males). Male anogenital distance and male body weight were positively correlated (Pearson’s correlation r = 0.352; t = 3.389, P = 0.00108). However, this association did not result in model collinearity (correlation of fixed effects in glmm = 0.049). Types of social groups From 83 male measurements, 67 males were members of 41 multi-male/multi-female groups, 6 were members of male-only groups, and 10 were solitary individuals (Table 1). In multi-male/multi-female groups, the number of females averaged 2.1 ± 1.7 females per group (range: 1–9), while the number of males averaged 1.6 ± 1.0 males per group (range: 1–5). Mean male anogenital distance within multi-male/multi-female groups averaged 9.5 ± 1.0 (range: 7.3–11.5). The number of males in the only male groups was 2 ± 0 (3 pairs). Mean male anogenital distance within male-only groups averaged 10.2 ± 0.7 (range: 9.6–11.1). Table 1 Number and attributes of social groups studied during 2010–2015 Year 2010 2011 2012 2013 2014 2015 Number of groups (n) 5 4 2 9 15 9  Multi-male/multi-female groups 5 4 2 8 14 8  Male only groups 0 0 0 1 1 1 Group size  Mean 3 5 2 4 3 4  SD 1 1 0 3 1 3  Range 2–4 3–6 2–10 2–5 2–11 Number of females  Mean 2 2 1 2 2 3  SD 1 1 0 3 1 2  Range 1–3 1–3 0–9 0–4 0–6 Number of males  Mean 2 3 1 1 2 2  SD 1 2 0 1 1 1  Range 1–2 1–5 1–3 1–3 1–5 Number solitary males 1 3 0 4 1 1 Year 2010 2011 2012 2013 2014 2015 Number of groups (n) 5 4 2 9 15 9  Multi-male/multi-female groups 5 4 2 8 14 8  Male only groups 0 0 0 1 1 1 Group size  Mean 3 5 2 4 3 4  SD 1 1 0 3 1 3  Range 2–4 3–6 2–10 2–5 2–11 Number of females  Mean 2 2 1 2 2 3  SD 1 1 0 3 1 2  Range 1–3 1–3 0–9 0–4 0–6 Number of males  Mean 2 3 1 1 2 2  SD 1 2 0 1 1 1  Range 1–2 1–5 1–3 1–3 1–5 Number solitary males 1 3 0 4 1 1 View Large Table 1 Number and attributes of social groups studied during 2010–2015 Year 2010 2011 2012 2013 2014 2015 Number of groups (n) 5 4 2 9 15 9  Multi-male/multi-female groups 5 4 2 8 14 8  Male only groups 0 0 0 1 1 1 Group size  Mean 3 5 2 4 3 4  SD 1 1 0 3 1 3  Range 2–4 3–6 2–10 2–5 2–11 Number of females  Mean 2 2 1 2 2 3  SD 1 1 0 3 1 2  Range 1–3 1–3 0–9 0–4 0–6 Number of males  Mean 2 3 1 1 2 2  SD 1 2 0 1 1 1  Range 1–2 1–5 1–3 1–3 1–5 Number solitary males 1 3 0 4 1 1 Year 2010 2011 2012 2013 2014 2015 Number of groups (n) 5 4 2 9 15 9  Multi-male/multi-female groups 5 4 2 8 14 8  Male only groups 0 0 0 1 1 1 Group size  Mean 3 5 2 4 3 4  SD 1 1 0 3 1 3  Range 2–4 3–6 2–10 2–5 2–11 Number of females  Mean 2 2 1 2 2 3  SD 1 1 0 3 1 2  Range 1–3 1–3 0–9 0–4 0–6 Number of males  Mean 2 3 1 1 2 2  SD 1 2 0 1 1 1  Range 1–2 1–5 1–3 1–3 1–5 Number solitary males 1 3 0 4 1 1 View Large Number of offspring sired by males at weaning The best model explaining variation in male reproductive success included the main effects of 4 predictors and one random effect (Table 2). The fixed factors explained 23% of variation in the number of offspring sired by males. Together, the fixed factors and random factor (social group ID) explained 60% of variation in the number of offspring sired by males. The best model revealed a positive relationship between male anogenital distance and the number of offspring sired by males. Males with longer anogenital distances sired more offspring than males with shorter anogenital distances (Figure 1a), such that each 1 mm increase in anogenital distance would result in a 40% more offspring (i.e., e0.36 = 1.4). One-year-old males sired more offspring (mean = 3.83 ± 4.9) than 2-year-old males (mean = 0.94 ± 1.4) (Figure 1b), such that for every year a male ages, reproductive success would decrease by 13.5% fewer offspring (i.e., e−1.96 = 0.135). Similarly, heavier males were associated with a larger number offspring sired (according to our directional hypothesis predictions), such that a 1 g increase in body weight would result in 1% more offspring (i.e., e0.009 = 1.009) (Figure 1c). In contrast, mean anogenital distance within groups did not predict the number of offspring sired by males. None of the social attributes and neither the interaction of male anogenital distance by mean anogenital distance within social groups were included in the best model, implying no modulatory role of social attributes on male morphotype. Table 2 Effects of individual male attributes and social group attributes on the number of offspring sired by males Fixed Factors Estimate Standard error Z P VIF (Intercept) 0.03 1.625 0.018 0.986 Male anogenital distance 0.36 0.152 2.337 0.019 1.88 Male age (2) −1.96 0.404 −4.856 <0.001 1.18 Male body weight 0.01 0.005 1.760 0.074 1.34 Mean anogenital distance group −0.42 0.240 −1.760 0.078 2.08 Random Factors σ2 Social group ID 0.962 σ2 error 1.080 Number of social groups 54 Number of males 83 Family Poisson (log) Fixed Factors Estimate Standard error Z P VIF (Intercept) 0.03 1.625 0.018 0.986 Male anogenital distance 0.36 0.152 2.337 0.019 1.88 Male age (2) −1.96 0.404 −4.856 <0.001 1.18 Male body weight 0.01 0.005 1.760 0.074 1.34 Mean anogenital distance group −0.42 0.240 −1.760 0.078 2.08 Random Factors σ2 Social group ID 0.962 σ2 error 1.080 Number of social groups 54 Number of males 83 Family Poisson (log) Terms reported are those from the most parsimonious model retained after using a best subset model selection routine based on Akaike Information Criteria. Significance was defined as P < 0.05. VIF = Variance inflation factor. View Large Table 2 Effects of individual male attributes and social group attributes on the number of offspring sired by males Fixed Factors Estimate Standard error Z P VIF (Intercept) 0.03 1.625 0.018 0.986 Male anogenital distance 0.36 0.152 2.337 0.019 1.88 Male age (2) −1.96 0.404 −4.856 <0.001 1.18 Male body weight 0.01 0.005 1.760 0.074 1.34 Mean anogenital distance group −0.42 0.240 −1.760 0.078 2.08 Random Factors σ2 Social group ID 0.962 σ2 error 1.080 Number of social groups 54 Number of males 83 Family Poisson (log) Fixed Factors Estimate Standard error Z P VIF (Intercept) 0.03 1.625 0.018 0.986 Male anogenital distance 0.36 0.152 2.337 0.019 1.88 Male age (2) −1.96 0.404 −4.856 <0.001 1.18 Male body weight 0.01 0.005 1.760 0.074 1.34 Mean anogenital distance group −0.42 0.240 −1.760 0.078 2.08 Random Factors σ2 Social group ID 0.962 σ2 error 1.080 Number of social groups 54 Number of males 83 Family Poisson (log) Terms reported are those from the most parsimonious model retained after using a best subset model selection routine based on Akaike Information Criteria. Significance was defined as P < 0.05. VIF = Variance inflation factor. View Large Figure 1 View largeDownload slide Male individual attributes and their effects on the number of offspring sired by males at weaning (n = 83). (a) Positive relationship between male anogenital distance and the number of offspring sired by males. (b) Effects of male age on the number offspring sired by males. (c) Positive relationship between male body weight and the number of offspring sired by males (mean ± 1SE). Figure 1 View largeDownload slide Male individual attributes and their effects on the number of offspring sired by males at weaning (n = 83). (a) Positive relationship between male anogenital distance and the number of offspring sired by males. (b) Effects of male age on the number offspring sired by males. (c) Positive relationship between male body weight and the number of offspring sired by males (mean ± 1SE). Social group type and the number of offspring sired at weaning The number of offspring sired did not differ across different types of social groups, including multi-male/multi-female groups, male-only groups, and solitary males. Males in multi-male/multi-female groups tended to sire more offspring than male-only groups and solitary males, but this was not statistically significant (Likelihood Ratio Test χ2 = 5.39, P = 0.0677, df = 4; Figure 2). Figure 2 View largeDownload slide Social group type (solitary males n = 10; male-only groups n = 6; multi-male/multi-female groups n = 67) and its effects on the number of offspring sired by males at weaning. Data are presented as mean ± 1SE. Figure 2 View largeDownload slide Social group type (solitary males n = 10; male-only groups n = 6; multi-male/multi-female groups n = 67) and its effects on the number of offspring sired by males at weaning. Data are presented as mean ± 1SE. DISCUSSION We assessed how male individual attributes, social group attributes and their interaction influence male reproductive success. Male reproductive success increased with male anogenital distance, but decreased with male age. In agreement with our prediction, male masculinization level was the most important predictor of male reproductive success. This finding suggests intriguing parallels with a previous study on Mongolian gerbils and studies on domestic guinea pigs. In particular, highly masculinized male Mongolian gerbils sired significantly more offspring than slightly masculinized males in captivity (Clark et al. 1992). Likewise, in captivity, infantilized guinea pig males sired fewer offspring than aggressive males (Guenther et al. 2014; Zimmermann et al. 2017). Thus, our current study on degus strongly supports that male phenotypical masculinization level has consequences on male reproductive success under natural conditions, and where highly masculinized males may attain a fitness advantage. Higher reproductive success in highly masculinized male degus, is not influenced for female effects. Previously, we found that in degus, heavier females are more fertile and masculinized females wean heavier pups (Correa et al. 2016) however, male degus did not select a particular type of female (based on her anogenital distance or body mass) to mate, but seek to mate with as many females as possible (L.A. Ebensperger, unpublished data). The masculinization gradient in males has been studied in several species of rodents (Ward 1972; vom Saal 1989; Drickamer 1996; Ryan and Vandenbergh 2002; Ophir and delBarco-Trillo 2007; Siegeler et al. 2011; Godsall et al. 2014) and in domestic rabbits (Bánszegi et al. 2015). These studies have revealed how masculinization may result in other phenotypic differences with potential consequences on reproductive success. For example, highly masculinized males exhibit heavier/bigger testes and ventral scent glands, larger seminal vesicles and chin glands, and more developed genital musculature than slightly masculinized males (Clark and Galef 1998; vom Saal et al. 1999; Ryan and Vandenbergh 2002; Bánszegi et al. 2015). Masculinization also results in behavioral differences. In addition to being more aggressive (Ryan and Vandenbergh 2002), highly masculinized males scent mark more frequently (Bánszegi et al. 2015), disperse more often, and have larger home ranges than less slightly masculinized males (Drickamer 1996; Godsall et al. 2014). Highly masculinized males are sexually more active, are more attractive to females (Ophir and delBarco-Trillo 2007), mount females at higher rates, and exhibit longer time to ejaculation than slightly masculinized males (Ward 1972; Clark and Galef 1998; vom Saal et al. 1999). Together, these features suggest that masculinization during prenatal develop can have profound impacts on the reproductive success of adult males. Subsequent studies in degus are needed to address how some of these morphological and behavioral characteristics of males may determine the natural differences in reproductive success recorded among male morphotypes in this study. An intriguing question from our study is “under what conditions slightly masculinized males may attain a fitness advantage in the population?” Previous studies on laboratory-reared rodents suggest at least 3 potential scenarios. First, slightly masculinized male Mongolian gerbils are more attentive to offspring and remain with their mate during postpartum oestrus (Clark et al. 1997). The presence of these kinds of males’ decreases maternal costs and increases female fecundity during postpartum estrus. Second, Clark and Galef (2000) reported the existence of a subgroup of slightly masculinized males that exhibit no sexual interest in females and fail to mate. However, these “asexual” males provide exceptionally high quality of parental care, a tactic that may benefit these males if nesting with genetically related females (Clark and Galef 2000). Third, infantilized male guinea pigs exhibit low aggressiveness, a trait that may reduce costs of living in large, complex social systems. These males might be using a queuing strategy before attaining social dominance status (Kaiser and Sachser 2009; Sachser et al. 2011; Siegeler et al. 2011; Guenther et al. 2014; Zimmermann et al. 2017). Female degus exhibit postpartum oestrus, yet this second breeding does not always result in a second litter per year in our study population (Ebensperger et al. 2013, but see Previtali et al. 2010). During years in which females produce 2 litters, second litters represent no more than 9% of total offspring produced during lifetime (L.A. Ebensperger, unpublished data). These conditions suggest a queuing strategy is unlikely to be profitable to slightly masculinized male degus. On the other hand, the observation that male degus exhibit parental care in captivity (Ebensperger et al. 2010), suggests an alternative way in which slightly masculinized males could attain indirect fitness benefits through rearing the offspring of related females, as suggested by Clark and Galef (2000). Our next step will be to evaluate the hypothesis of male indirect fitness benefits through the provision of paternal care. This will be done through studies in captivity that evaluate the paternal behavior of males of different levels of masculinization, and through studies in a natural population, evaluating the level of relatedness between adult males and females that share same burrow system during parental care period. The observation of a negative relationship between age and reproductive success is not surprising. In our study population, most adults (90%) do not survive to their second year of life, implying these rodents are ecologically semelparous (Ebensperger et al. 2013; Correa et al. 2016). In captivity, where degus can reproduce successfully until 4 years of age, reproductive output decreases in a sustained way from the first to the fourth reproductive event (Correa 2012). A comparative study of rodents suggests that high mortality rates can modify reproductive strategies, from slow to fast life-history strategies and that fast life-history species tend to have precocial offspring (Dobson and Oli 2007). The ecological and reproductive characteristics of degus are consistent with these findings. Under these conditions, male and female degus likely invest most reproductive effort in their first reproductive event, despite the trade-off between current and future reproduction (Hamel et al. 2010; Lemaître et al. 2015; Correa et al. 2016). Some support to this possibility comes from the breeding strategy of female degus. Females with higher cortisol (a measure of reproductive effort) levels during main lactation season produce more surviving offspring, but reduce their ability to produce a second litter (Ebensperger et al. 2011, 2013). The observation that interannual survivorship of adult males is lower than that of females (Ebensperger et al. 2009) suggests males are experiencing similar pressures to invest more during their first (and likely the only) breeding event. According with our original expectation, male reproductive success of degus was influenced by body weight, but the magnitude of the effect was extremely small. This result is partially in agreement with evidence from wild cavies, a closely related species in which heavier and territorial males sire 80% of offspring (Asher et al. 2008). Differences in the magnitude of the male body weight effect on fitness between degus and wild cavies are in agreement with what would be expected from species with different mating systems. As suggested previously (Clutton-Brock 1989; Schradin and Lindholm 2011), male body weight is expected to be a more important predictor of male reproductive success in polygynous species (i.e., wild cavies) than in promiscuous counterparts (i.e., degus). Male body weight is thought to be important for males to attain higher reproductive success when conditions allow monopolization of clumped females. Conversely, in species with promiscuous mating systems male searching ability is expected to be more important to locate relatively dispersed females (Schwagmeyer 1988; Clutton-Brock 1989; Schradin and Lindholm 2011). In agreement with these expectations, male African striped mice (Rhabdomys pumilio) may change their mating tactic depending on the female grouping pattern (dispersed, grouped) (Schradin et al. 2009; Schradin et al. 2010; Schradin and Lindholm 2011). Our study revealed that individual attributes have a larger impact than social attributes on the reproductive success of male degus. This result parallels findings for females of the same population (Correa et al. 2016). In male and female degus, masculinization level is positively related to number of offspring sired and offspring body mass at weaning. Together, these findings indicate how mammalian phenotypical masculinization might be adaptive regardless of social context. This scenario differs from that was suggested by previous studies in other social rodents such as cavies, where success of infantilized/aggressive males would be contingent upon social environment (Kaiser and Sachser 2009; Sachser et al. 2011; Siegeler et al. 2011; Guenther et al. 2014; Zimmermann et al. 2017). SUPPLEMENTARY MATERIAL Supplementary data are available at Behavioral Ecology online. FUNDING This work was supported by a FONDECYT grant (3130567 and 11170222 to L.A.C); FONDECYT grant (1090302, 1130091, and 1170409 to L.A.E); CONICYT PhD thesis grant (21120244 to A.L-P); and NSF OISE grant (0853719 and 1261026 to L.D.H). We are grateful to Gioconda Peralta, Lucie Jaugeon, Loreto Carrasco, and Sylvain Faugeron from the Molecular Diversity Laboratory, Pontificia Universidad Católica de Chile, for their valuable help and advice with molecular analyses. We are indebted to the Universidad de Chile, particularly to Marcelo Orellana Reyes and Rosa Peralta (Field Station Administrators), for providing facilities during field work at Rinconada. We also thank David Véliz for help with molecular data analyses and Carolyn Bauer for providing extremely useful suggestions on a previous version of this manuscript. Data accessibility: Analyses reported in this article can be reproduced using the data provided by Correa et al. (2018). Handling editor: Anna Lindholm REFERENCES Asher M , Lippmann T , Epplen JT , Kraus C , Trillmich F , Sachser N . 2008 . Large males dominate: ecology, social organization and mating system of wild cavies, the ancestors of the Guinea pig . Behav Ecol Sociobiol . 62 : 1509 – 1521 . Google Scholar CrossRef Search ADS Bánszegi O , Szenczi P , Dúcs A , Hudson R , Altbäcker V . 2015 . Long-term under-masculinization in male rabbits due to maternal stress is reversed by prenatal administration of testosterone . Behav Processes . 115 : 156 – 162 . Google Scholar CrossRef Search ADS PubMed Barton K . 2009 . MuMIn: Multi-model inference . Available from: http://r-forge. r- project.org/projects/mumin. Bates D , Maechler M , Bolker B , Walker S . 2015 . Fitting linear mixed-effects models using lme4 . J Stat Sofw . 67 : 1 – 48 . Burnham KP , Anderson DR . 2002 . Model selection and multi-model inference: a practical information-theoretic approach ( 2nd ed .). New York (USA) : Springer-Verlag . di Castri F , Hajek E . 1976 . Bioclimatología de Chile . Santiago (Chile) : Pontificia Universidad Católica de Chile . Clark MM , Desousa D , Vonk J , Galef BG Jr . 1997 . Parenting and potency: alternative routes to reproductive success in male Mongolian gerbils . Anim Behav . 54 : 635 – 642 . Google Scholar CrossRef Search ADS PubMed Clark MM , Galef BG Jr . 1998 . Effects on intrauterine position on the behavior and genital morphology of litter bearing rodents . Dev Neuropsychol . 14 : 197 – 211 . Google Scholar CrossRef Search ADS Clark MM , Galef BG Jr . 2000 . Why some male Mongolian gerbils may help at the nest: testosterone, asexuality and alloparenting . Anim Behav . 59 : 801 – 806 . Google Scholar CrossRef Search ADS PubMed Clark MM , Tucker L , Galef BG Jr . 1992 . Stud and dud males: intrauterine position effects on the reproductive success of male gerbils . Anim Behav . 43 : 215 – 221 . Google Scholar CrossRef Search ADS Clutton-Brock TH . 1989 . Mammalian mating systems . Proc R Soc Lond B Biol Sci . 236 : 339 – 372 . Google Scholar CrossRef Search ADS PubMed Clutton-Brock TH , Huchard E . 2013 . Social competition and selection in males and females . Philos Trans R Soc Lond B Biol Sci . 368 : 20130074 . Google Scholar CrossRef Search ADS PubMed Correa LA . 2012 . Mecanismos de regulación del tamaño de camada y razón de sexos en Octodon degus: efectos de la alostasis prenatal en la variación fenotípica de las crías y sus consecuencias en la estabilidad de los grupos sociales . PhD tesis, Valdivia : Universidad Austral de Chile . Correa LA , León C , Ramírez-Estrada J , Ly-Prieto A , Abades S , Hayes LD , Soto-Gamboa M , Ebensperger LA . 2018 . Data from: masculinized and younger males attain higher reproductive success in a social rodent . Dryad Digital Repository . http://dx.doi.org/10.5061/dryad.cb11j Correa LA , León C , Ramírez-Estrada J , Soto-Gamboa M , Sepúlveda RD , Ebensperger LA . 2016 . Masculinized females produce heavier offspring in a group living rodent . J Anim Ecol . 85 : 1552 – 1562 . Google Scholar CrossRef Search ADS PubMed Crump CJ , Chevins PF . 1989 . Prenatal stress reduces fertility of male offspring in mice, without affecting their adult testosterone levels . Horm Behav . 23 : 333 – 343 . Google Scholar CrossRef Search ADS PubMed Dobson FS , Oli MK . 2007 . Fast and slow life histories of rodents . In: Wolff JO , Sherman PW , editors. Rodent Societies an ecological and evolutionary perspective . Chicago (USA) : The University of Chicago Press . p. 99 – 105 . Davis GT , Vásquez RA , Poulin E , Oda E , Bazán-León EA , Ebensperger LA , Hayes LD . 2016 . Octodon degus kin and social structure . J. Mamm . 97 : 361 – 372 . Google Scholar CrossRef Search ADS Drickamer LC . 1996 . Intrauterine position and anogenital distance in house mice: consequences under field conditions . Anim Behav . 51 : 925 – 934 . Google Scholar CrossRef Search ADS Dubuc C , Ruiz-Lambides A , Widdig A . 2014 . Variance in male lifetime reproductive success and estimation of the degree of polygyny in a primate . Behav Ecol . 25 : 878 – 889 . Google Scholar CrossRef Search ADS PubMed Ebensperger LA , Chesh AS , Castro RA , Ortiz- Tolhuysen L , Quirici V , Burger JR , Hayes LD . 2009 . Instability rules social groups in the communal breeder rodent Octodon degus . Ethology . 115 : 540 – 554 . Google Scholar CrossRef Search ADS Ebensperger LA , Correa LA , León C , Ramírez-Estrada J , Abades S , Villegas Á , Hayes LD . 2016 . The modulating role of group stability on fitness effects of group size is different in females and males of a communally rearing rodent . J Anim Ecol . 85 : 1502 – 1515 . Google Scholar CrossRef Search ADS PubMed Ebensperger LA , Hurtado MJ . 2005 . Seasonal changes in the time budget of degus, Octodon degus . Behaviour . 142 : 91 – 112 . Google Scholar CrossRef Search ADS Ebensperger LA , Hurtado MJ , Soto-Gamboa M , Lacey EA , Chang AT . 2004 . Communal nesting and kinship in degus (Octodon degus) . Naturwissenschaften . 91 : 391 – 395 . Google Scholar CrossRef Search ADS PubMed Ebensperger LA , Ramírez-Estrada J , León C , Castro RA , Tolhuysen LO , Sobrero R , Quirici V , Burger JR , Soto-Gamboa M , Hayes LD . 2011 . Sociality, glucocorticoids and direct fitness in the communally rearing rodent, Octodon degus . Horm Behav . 60 : 346 – 352 . Google Scholar CrossRef Search ADS PubMed Ebensperger LA , Ramírez-Otarola N , León C , Ortiz ME , Croxatto HB . 2010 . Early fitness consequences and hormonal correlates of parental behaviour in the social rodent, Octodon degus . Physiol Behav . 101 : 509 – 517 . Google Scholar CrossRef Search ADS PubMed Ebensperger LA , Tapia D , Ramírez-Estrada J , León C , Soto-Gamboa M , Hayes LD . 2013 . Fecal cortisol levels predict breeding but not survival of females in the short-lived rodent, Octodon degus . Gen Comp Endocrinol . 186 : 164 – 171 . Google Scholar CrossRef Search ADS PubMed Ebensperger LA , Villegas A , Abades S , Hayes LD . 2014 . Mean ecological conditions modulate the effects of group living and communal rearing on offspring production and survival . Behav Ecol . 25 : 862 – 870 . Google Scholar CrossRef Search ADS Emlen ST , Oring LW . 1977 . Ecology, sexual selection, and the evolution of mating systems . Science . 197 : 215 – 223 . Google Scholar CrossRef Search ADS PubMed Engqvist L , Taborsky M . 2016 . The evolution of genetic and conditional alternative reproductive tactics . Proc Biol Sci . 283 : 20152945 . Google Scholar CrossRef Search ADS PubMed Fulk GW . 1976 . Notes on the activity, reproduction, and social behavior of Octodon degus . J Mamm . 57 : 495 – 505 . Google Scholar CrossRef Search ADS Godsall B , Coulson T , Malo AF . 2014 . From physiology to space use: energy reserves and androgenization explain home-range size variation in a woodland rodent . J Anim Ecol . 83 : 126 – 135 . Google Scholar CrossRef Search ADS PubMed Green P , MacLeod CJ . 2016 . SIMR: an R package for power analysis of generalized linear mixed models by simulation . Meth Ecol Evol . 7 : 493 – 498 . Google Scholar CrossRef Search ADS Gross MR . 1996 . Alternative reproductive strategies and tactics: diversity within sexes . Trends Ecol Evol . 11 : 92 – 98 . Google Scholar CrossRef Search ADS PubMed Guenther A , Kowalski G , von Engelhardt N . 2014 . Prenatal social conditions shape offspring adult phenotype and reproductive success . Behav Ecol Sociobiol . 68 : 1661 – 1667 . Google Scholar CrossRef Search ADS Hamel S , Gaillard JM , Yoccoz NG , Loison A , Bonenfant C , Descamps S . 2010 . Fitness costs of reproduction depend on life speed: empirical evidence from mammalian populations . Ecol Lett . 13 : 915 – 935 . Google Scholar CrossRef Search ADS PubMed Hayes LD , Chesh AS , Castro RA , Ortiz-Tolhuysen L , Burger JR , Bhattacharjee J , Ebensperger LA . 2009 . Fitness consequences of group living in the degu Octodon degus, a plural breeder rodent with communal care . Anim Behav . 78 : 131 – 139 . Google Scholar CrossRef Search ADS Hayes LD , Chesh AS , Ebensperger LA . 2007 . Ecological predictors of range areas and use of burrow system in the diurnal rodent, Octodon degus . Ethology . 113 : 155 – 165 . Google Scholar CrossRef Search ADS Hotchkiss AK , Vandenbergh JG . 2005 . The anogenital distance index of mice (Mus musculus domesticus): an analysis . Contemp Top Lab Anim Sci . 44 : 46 – 48 . Google Scholar PubMed Kaiser S , Kruijver FP , Straub RH , Sachser N , Swaab DF . 2003 . Early social stress in male Guinea-pigs changes social behaviour, and autonomic and neuroendocrine functions . J Neuroendocrinol . 15 : 761 – 769 . Google Scholar CrossRef Search ADS PubMed Kaiser S , Sachser N . 2009 . Effects of prenatal social stress on offspring development pathology or adaptation ?. Curr Dir Psychol Sci . 18 : 118 – 121 . Google Scholar CrossRef Search ADS Kaiser S , Sachser N . 2001 . Social stress during pregnancy and lactation affects in guinea pigs the male offsprings’ endocrine status and infantilizes their behaviour . Psychoneuroendocrinology . 26 : 503 – 519 . Google Scholar CrossRef Search ADS PubMed Knapp R . 2004 . Endocrine mediation of vertebrate male alternative reproductive tactics: the next generation of studies . Integr Comp Biol . 43 : 658 – 668 . Google Scholar CrossRef Search ADS von Kuerthy C , Ros AF , Taborsky M . 2016 . Androgen responses to reproductive competition of males pursuing either fixed or plastic alternative reproductive tactics . J Exp Biol . 219 : 3544 – 3553 . Google Scholar CrossRef Search ADS PubMed Lemaître JF , Berger V , Bonenfant C , Douhard M , Gamelon M , Plard F , Gaillard JM . 2015 . Early-late life trade-offs and the evolution of ageing in the wild . Proc Biol Sci . 282 : 20150209 . Google Scholar CrossRef Search ADS PubMed Marshall TC , Slate J , Kruuk LE , Pemberton JM . 1998 . Statistical confidence for likelihood-based paternity inference in natural populations . Mol Ecol . 7 : 639 – 655 . Google Scholar CrossRef Search ADS PubMed Moore MC , Hews DK , Knapp R . 1998 . Hormonal control and evolution of alternative male phenotypes: generalizations of models for sexual differentiation . Am Zool . 38 : 133 – 151 . Google Scholar CrossRef Search ADS Newman ME . 2004 . Analysis of weighted networks . Phys Rev E Stat Nonlin Soft Matter Phys . 70 : 056131 . Google Scholar CrossRef Search ADS PubMed van Oosterhout L , Hutchinson WF , Wills DPM , Shipley P . 2004 . Micro-checker: software to identifying and correcting genotyping errors in microsatellite data . Mol Ecol Not . 4 : 535 – 538 . Google Scholar CrossRef Search ADS Ophir AG , Delbarco-Trillo J . 2007 . Anogenital distance predicts female choice and male potency in prairie voles . Physiol Behav . 92 : 533 – 540 . Google Scholar CrossRef Search ADS PubMed Previtali MA , Meserve PL , Kelt DA , Milstead WB , Gutierrez JR . 2010 . Effects of more frequent and prolonged El Niño events on life-history parameters of the degu, a long-lived and slow-reproducing rodent . Conserv Biol . 24 : 18 – 28 . Google Scholar CrossRef Search ADS PubMed Quan YF , Macmanes MD , Ebensperger LA , Lacey EA , Hayes LD . 2009 . Isolation and characterization of polymorphic microsatellite loci from Octodon degus . Mol Ecol Resour . 9 : 999 – 1001 . Google Scholar CrossRef Search ADS PubMed Quirici V , Faugeron S , Hayes LD , Ebensperger LA . 2011 . Absence of kin structure in a population of the group-living rodent Octodon degus . Behav Ecol . 22 : 248 – 254 . Google Scholar CrossRef Search ADS Raveh S , Heg D , Dobson FS , Coltman DW , Gorrell JC , Balmer A , Neuhausa P . 2010 . Mating order and reproductive success in male Columbian ground squirrels (Urocitellus columbianus) . Behav Ecol . 21 : 537 – 547 . Google Scholar CrossRef Search ADS Rhen T , Crews D . 2002 . Variation in reproductive behaviour within a sex: neural systems and endocrine activation . J Neuroendocrinol . 14 : 517 – 531 . Google Scholar CrossRef Search ADS PubMed Roff DA , Wolak ME , Correa LA , Soto-Gamboa M . 2017 . The phenotypic correlates and quantitative genetics of masculinization in the rodent, Octodon degus . Heredity . 119 : 136 – 141 . Google Scholar CrossRef Search ADS PubMed Ryan BC , Vandenbergh JG . 2002 . Intrauterine position effects . Neurosci Biobehav Rev . 26 : 665 – 678 . Google Scholar CrossRef Search ADS PubMed vom Saal FS . 1989 . Sexual differentiation in litter-bearing mammals: influence of sex of adjacent fetuses in utero . J Anim Sci . 67 : 1824 – 1840 . Google Scholar CrossRef Search ADS PubMed vom Saal FS , Clark MM , Galef BG , Drickamer LC , Vandenbergh JG . 1999 . Intrauterine position phenomenon . In: Knobil E , Neil JD , editors. Encyclopedia of Reproduction Vol. 2 . New York (USA) : Academic Press . p. 893 – 900 . Google Scholar CrossRef Search ADS Sachser N , Hennessy MB , Kaiser S . 2011 . Adaptive modulation of behavioural profiles by social stress during early phases of life and adolescence . Neurosci Biobehav Rev . 35 : 1518 – 1533 . Google Scholar CrossRef Search ADS PubMed Schroeder JW , Honeycutt RL , Rooney AP , Han G , Begall S , Gallardo MH . 2000 . Microsatellites from the South American coruro, Spalacopus cyanus . Mol Ecol . 9 : 1447 – 1449 . Google Scholar CrossRef Search ADS PubMed Schradin C , König B , Pillay N . 2010 . Reproductive competition favours solitary living while ecological constraints impose group-living in African striped mice . J Anim Ecol . 79 : 515 – 521 . Google Scholar CrossRef Search ADS PubMed Schradin C , Lindholm AK . 2011 . Relative fitness of alternative male reproductive tactics in a mammal varies between years . J Anim Ecol . 80 : 908 – 917 . Google Scholar CrossRef Search ADS PubMed Schradin C , Scantlebury M , Pillay N , König B . 2009 . Testosterone levels in dominant sociable males are lower than in solitary roamers: physiological differences between three male reproductive tactics in a sociably flexible mammal . Am Nat . 173 : 376 – 388 . Google Scholar CrossRef Search ADS PubMed Schwagmeyer PL . 1988 . Scramble competition polygyny in an asocial mammal: male mobility and mating success . Am Nat . 131 : 885 – 892 . Google Scholar CrossRef Search ADS Siegeler K , Sachser N , Kaiser S . 2011 . The social environment during pregnancy and lactation shapes the behavioral and hormonal profile of male offspring in wild cavies . Dev Psychobiol . 53 : 575 – 584 . Google Scholar CrossRef Search ADS PubMed Soto-Gamboa M , Villalón M , Bozinovic F . 2005 . Social cues and hormone levels in male Octodon degus (Rodentia): a field test of the Challenge Hypothesis . Horm Behav . 47 : 311 – 318 . Google Scholar CrossRef Search ADS PubMed Taborsky M , Oliveira RF , Brockmann HJ . 2008 . The evolution of alternative reproductive tactics: concepts and questions . In: Oliveira RF , Taborsky M , Brockmann HJ , editors. Alternative reproductive tactics: an Integrative Approach . Cambridge (UK) : Cambridge University Press . p. 1 – 22 . Google Scholar CrossRef Search ADS Vandenbergh JG , Huggett CL . 1995 . Mother’s prior intrauterine position affects the sex ratio of her offspring in house mice . Proc Natl Acad Sci USA . 91 : 11055 – 11059 . Google Scholar CrossRef Search ADS Ward IL . 1972 . Prenatal stress feminizes and demasculinizes the behavior of males . Science . 175 : 82 – 84 . Google Scholar CrossRef Search ADS PubMed Ward IL , Weisz J . 1980 . Maternal stress alters plasma testosterone in fetal males . Science . 207 : 328 – 329 . Google Scholar CrossRef Search ADS PubMed Whitehead H . 2008 . Analyzing animal societies: quantitative methods for vertebrate social analysis . Chicago (USA) : Chicago University Press . Google Scholar CrossRef Search ADS Whitehead H . 2009 . SOCPROG programs. Analyzing animal social structures . Behav Ecol Sociobiol . 63 : 765 – 778 . Google Scholar CrossRef Search ADS Zedrosser A , Bellemain E , Taberlet P , Swenson JE . 2007 . Genetic estimates of annual reproductive success in male brown bears: the effects of body size, age, internal relatedness and population density . J Anim Ecol . 76 : 368 – 375 . Google Scholar CrossRef Search ADS PubMed Zimmermann TD , Kaiser S , Hennessy MB , Sachser N . 2017 . Adaptive shaping of the behavioural and neuroendocrine phenotype during adolescence . Proc Roy Soc Lond B . 284 : 20162784 . Google Scholar CrossRef Search ADS Zuur AF , Ieno EN , Walker N , Saveliev AA , Smith GM . 2009 . Mixed effects models and extensions in ecology with R . New York (USA) : Springer-Verlag . Google Scholar CrossRef Search ADS © The Author(s) 2018. Published by Oxford University Press on behalf of the International Society for Behavioral Ecology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Behavioral Ecology Oxford University Press

Highly masculinized and younger males attain higher reproductive success in a social rodent

Loading next page...
 
/lp/ou_press/highly-masculinized-and-younger-males-attain-higher-reproductive-lI40OA3jsK
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the International Society for Behavioral Ecology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
1045-2249
eISSN
1465-7279
D.O.I.
10.1093/beheco/ary015
Publisher site
See Article on Publisher Site

Abstract

Abstract Alternative morphotypes have been reported in males of different taxa. In some mammals, highly masculinized and slightly masculinized males represent 2 opposite ends along a gradient of phenotypic variation in males. This phenotypical gradient originates during prenatal development. Laboratory studies have documented how highly and slightly masculinized males differ in several traits, including their reproductive success. However, the extent to which these reported differences materialize in natural populations remains unknown. We quantified the impact of male morphotype on male reproductive success in a natural population of Octodon degus, a highly social rodent. We assessed male morphotype through a continuous gradient of anogenital distance. We also tested the hypothesis that the social environment interacts with male morphotype to influence male reproductive success. We found that individual attributes, including masculinization level and age, impacted male reproductive success. Highly masculinized and younger males had greater reproductive success. Additionally, male body weight had a small magnitude but positive effect on male reproductive success. Male reproductive success was not affected by social attributes such as group composition. Thus, the number of males and females within a group did not affect male reproductive success, nor did the average male anogenital distance within a group. Our results support the hypothesis that the prenatal environment can result in long-term effects on individual life history and cause intrasexual phenotypical variation in natural populations. Our findings suggest that male phenotypical masculinization could be an adaptive trait, regardless of the social environment. INTRODUCTION Male phenotypic variation, characterized as the coexistence of 2 or more types of males in the same population (Engqvist and Taborsky 2016), occurs in wide range of taxa including mites, insects, crustaceans, fishes, amphibians, reptiles, and birds (Gross 1996). This variation may involve extreme differences in behavior, morphology, physiology, and reproductive traits. For example, conspicuous male phenotypes are associated with higher concentration of androgens and territorial tactics to monopolize reproduction. In contrast, inconspicuous male phenotypes are associated with low androgen concentrations and nonterritorial parasitic tactics (Knapp 2004; Taborsky et al. 2008; von Kuerthy et al. 2016). Previous studies on species with multiple male phenotypes indicate that alternative male morphotypes may be influenced by genetic or environmental mechanisms, the distribution of traits in a population could be discrete or continuous, and male morphotypes can exhibit alternative reproductive tactics linked to differences in fitness payoffs (Gross 1996; Moore et al. 1998; Rhen and Crews 2002; Knapp 2004; Taborsky et al. 2008). In male mammals, alternative morphotypes generally have been linked to different levels of male masculinization (Clark and Galef 1998; Kaiser and Sachser 2001; Rhen and Crews 2002; Kaiser et al. 2003; Roff et al. 2017). This masculinization gradient is a consequence of 2 proximate mechanisms. First, alternative male morphotypes may result from maternal stress responses during pregnancy, in which the male embryo/fetus is exposed to 1) a lower concentration of androgens, 2) less powerful androgens (e.g., DHEA), and/or 3) exposure to androgens but outside of the most sensitive period (Ward 1972; Ward and Weisz 1980; Kaiser et al. 2003). These conditions have been reported to produce a male with inconspicuous masculine traits (henceforth slightly masculinized male). A second mechanism corresponds to the intrauterine position phenomenon (IUP), where position of embryos/fetuses alters the male phenotype through the influence of androgens released by neighboring male siblings. A male embryo/fetus that would develop between 2 males would be exposed to higher concentration of androgens and therefore develop into an adult with exacerbated male traits (henceforth highly masculinized male). In contrast, a male embryo/fetus that would develop between 2 female siblings would be exposed to low androgen concentrations, and therefore develop into a slightly masculinized male (vom Saal 1989; Clark et al. 1992; vom Saal et al. 1999; Ophir and delBarco-Trillo 2007; Roff et al. 2017). Taken together, prenatal exposure to high or low concentrations of androgens may result in litter and population gradients of male offspring masculinization that persist through adulthood (Clark and Galef 1998; vom Saal et al. 1999). Highly masculinized and slightly masculinized males differ in several traits, including anogenital distance (Clark and Galef 1998; vom Saal et al. 1999; Ryan and Vandenbergh 2002; Ophir and delBarco-Trillo 2007). Given that prenatal exposure to androgens affects the development of perineal tissue, distance between the penis and anus is longer in males that were exposed to high concentrations of androgens, and shorter in males that were exposed to low concentrations of androgens (vom Saal 1989; Clark and Galef 1998; Hotchkiss and Vandenbergh 2005; Ophir and delBarco-Trillo 2007; Bánszegi et al. 2015). Thus, anogenital distance allows the noninvasive assessment of male masculinization level. Additionally, highly and slightly masculinized males also may differ in behavioral and reproductive traits such as aggressiveness, dominance status, sexual behavior, attractiveness to females, home range size, dispersal behavior, territorial scent-marking, parental behavior, hormone profiles, and weight/size of testicles and secondary sex organs (Drickamer 1996; Clark and Galef 1998; vom Saal et al. 1999; Kaiser and Sachser 2001; Ryan and Vandenbergh 2002; Kaiser et al. 2003; Ophir and delBarco-Trillo 2007; Godsall et al. 2014; Bánszegi et al. 2015). Our understanding of how masculinization influences reproductive success of males is limited to laboratory studies of a few domestic species. These studies point to effects on both fertility and fecundity. Slightly masculinized males of domestic mice (Mus musculus) and Mongolian gerbils (Meriones unguiculatus) exhibit a higher percentage of unsuccessful female impregnation (Crump and Chevins 1989; Clark and Galef 1998, 2000), and highly masculinized males of Mongolian gerbils and guinea pigs (Cavia aparea f. porcellus) produce more offspring than slightly masculinized males (Clark and Galef 1992; Guenther et al. 2014; Zimmermann et al. 2017). Intriguingly, the effect of male morphotype on reproductive success may also be influenced by intermale conflict, a component of social context. Thus, Kaiser and Sachser (2009) proposed that an aggressive tactic of some males (presumably highly masculinized) could be superior in low-density conditions, when intermale conflict is low. In contrast, a less-aggressive tactic of infantilized males could be superior in large mix-sex colonies, when intermale conflict is high. Under this scenario, an infantilized male could remain cryptic in the queue, avoiding harassment from more aggressive male (Sachser et al. 2011). Whenever aggressive males disappear, infantilized males would breed with female group mates (Sachser et al. 2011; Siegeler et al. 2011; Guenther et al. 2014; Zimmermann et al. 2017). In mammals, male body weight and age can affect male reproductive success (Clutton-Brock 1989; Raveh et al. 2010). For example, the impact of body mass on male reproductive success varies between mating systems (Emlen and Oring 1977; Clutton-Brock 1989; Clutton-Brock and Huchard 2013). Thus, in species in which the access to females is determined by male fighting abilities (interference competition) large males would have advantage (Emlen and Oring 1977; Clutton-Brock 1989; Zedrosser et al. 2007; Clutton-Brock and Huchard 2013). This prediction has been supported by studies on polygynous and sexual dimorphic ungulates and pinnipeds (Clutton-Brock 1989; Zedrosser et al. 2007; Clutton-Brock and Huchard 2013; Dubuc et al. 2014). However, in species with promiscuous mating systems and without sexual dimorphism, male searching abilities and experience (scramble competition) are more important determinants of male reproductive success (Schwagmeyer 1988; Clutton-Brock 1989; Schradin and Lindholm 2011). Older males are generally larger than, dominant over, and more experienced than younger males (Clutton-Brock 1989; Zedrosser et al. 2007). Thus, older males are expected to attain higher reproductive success than younger males (Clutton-Brock 1989). However, in species with fast life-history strategies and early senescence, natural selection could favor a high reproductive investment in early reproductive events (Hamel et al. 2010; Lemaître et al. 2015). In these species, younger individuals generally make a high investment in early life, thus compromising the reproductive performance and survivorship late in the life (Dobson and Oli 2007; Hamel et al. 2010; Lemaître et al. 2015). Herein, we use a 6-year dataset on a natural population of the caviomorph rodent Octodon degus to determine how male attributes (masculinization level, body weight, age) and social conditions (solitary vs. group-living) influences direct fitness of males. Degus live in social groups, where 0–3 adult males and 1–9 adult females communally nest and rear their offspring (Ebensperger et al. 2014; Ebensperger et al. 2016). Occasionally, degus also may live solitarily. Male and female degus are sexually mature at 6 months of age. However, in our study population, degus breed annually with the primary mating season taking place during the late, austral fall (June), when females begin to cycle with ascending photoperiod (Ebensperger et al. 2013). At that stage of the season, male and female yearlings are between 9 and 10 months of age. Most adult degus (85–90%) do not survive to their second year of age, which means that reproductive success during their first (and likely only) breeding event has a major impact on lifetime reproductive success (Ebensperger et al. 2013). Previous studies indicate that male degus exhibit alternative morphotypes, and where males displaying a territorial tactic also exhibit higher levels of testosterone compared with nonterritorial males (Soto-Gamboa et al. 2005). In captivity, males may be highly or slightly masculinized (Correa 2012). Moreover, both maternal stress and the IUP phenomenon were noted to contribute to this phenotypic variation (Correa 2012; Roff et al. 2017). A link between male morphotype (assessed through the anogenital distance analyses) and reproductive success has not yet been examined. In this study, we aimed to determine sources of variation in male reproductive success. First, we hypothesized that 1) male masculinization level, male body weight, and adult male age influences male reproductive success. This hypothesis predicts that (i) the number of offspring sired by males would increase with increasing anogenital distance. Additionally, we predicted that (ii) the number of offspring sired by males would increase with increasing body weight (iii) and that the number of offspring sired by adult males would decrease as the males grow old. Second, we hypothesized that 2) regardless of male morphotype, male reproductive success is influenced by attributes of its social group. This hypothesis predicts that (iv) the number of offspring sired by males would increase with increasing number of females (i.e., a measure of mating opportunities), and would decrease with increasing number of males (a measure of male–male competition over mates) in the social group. This hypothesis also predicts that (v) solitary living males or males living with other males exclusively would suffer a cost in terms of offspring sired relative to males nesting with one or more females. Our third hypothesis was that 3) social context modulates the effect of male morphotype on male reproductive success. Herein, we expected the effect of anogenital distance on male reproductive success would depend on the morphotypes (male anogenital distance lengths) of other male group members. This hypothesis predicts that (vi) males with long anogenital distances (highly masculinized) would sire more offspring in social groups in which the mean of anogenital distance of the group (i.e., a measure of similarity between male group mates in terms of anogenital distance) is relatively short whereas, the number of offspring sired by males with relatively long anogenital distance would decrease in social groups in which the mean of anogenital distance of the group is relatively long. MATERIAL AND METHODS Study population This study was conducted between 2010 and 2015 on a natural degu population located at the Estación Experimental Rinconada de Maipú (33°23′ S, 70°31′ W, altitude 495 m), a field station of the Universidad de Chile. This study area is characterized by a Mediterranean climate with cold, wet winters and warm, dry summers (di Castri and Hajek 1976). The site consisted of open areas with scattered shrubs (Proustia pungens, Acacia caven, and Baccharis spp.) that on average covered 14.5% of the field site (Ebensperger and Hurtado 2005). The total area examined at Rinconada was 2 ha and did not vary across years of study. Total degu density was 35 ± 18.3 degus/ha in winter and 43.5 ± 41.7 degus/ha in spring. Live trapping and telemetry Live trapping and telemetry were conducted in May to July (time encompassing mating) and in September and October (time encompassing parturition, lactation, and offspring weaning) of each year. Degus are diurnally active and remain in underground burrows overnight (Ebensperger et al. 2004). A burrow system was defined as a group of burrow openings surrounding a central location spanning 1–3 m in diameter where individuals were repeatedly found during night-time telemetry (Fulk 1976; Hayes et al. 2007). Ten traps (Tomahawk model 201, Tomahawk Live Trap Company, Tomahawk, WI) were used at each burrow system daily. Traps were set prior to the emergence of adults during morning hours (06:00 h). After 1.5 h, traps were closed until the next trapping day. The identity, location, sex, anogenital distance, body weight (weighed to the nearest 0.1 g), was determined for all captured degus. At first capture, each degu received ID coded tags on each ear (Monel 1005-1, National Band and Tag Co., Newport). Adults weighing more than 170 g were fitted with 6–7 g radio-collars (AVM Instrument Co., Colfax, CA) with unique pulse frequencies. From 2010 to 2015, 155 degus (including adult males and females) were radio-tracked during the winter season, while 193 degus (including adult males and females) were radio-tracked during the spring season ((Analyses based on 2008–2016 data from the Rinconada population indicate that radio-collars do not influence survival (L.A. Ebensperger, unpublished data, Supplementary Table S1 of Supplementary Data)). During night-time telemetry, degus were tracked to their home burrows via radio telemetry. Previous studies at Rinconada have confirmed that night-time locations represent underground nest sites (Ebensperger et al. 2004). Locations were determined once per night approximately 1 h before sunrise using LA 12-Q receivers (for radio collars tuned to 150.000–151.999 MHz frequency; AVM Instrument Co., Auburn, CA) and hand held, 3-element Yagi antennas (AVM instrument Co., Auburn, CA). Previous studies have verified these radiotelemetry methods as sufficient for determining group membership (Hayes et al. 2009). The number of burrow systems monitored was 59 ± 7 in winter and 55 ± 9 in spring. The number of days that each burrow system was trapped was 36 ± 3 in winter and 50 ± 8 in spring. The number of radio-collared degus was 26 ± 10 in winter and 32 ± 8 in spring. The average number of night-time telemetry locations per radio-collared degu across all years of the study was 16 ± 0.6 in winter and 18 ± 1.4 in spring (data expressed in mean ± SD). Details per year are reported in Supplementary Table S2 of Supplementary Data. Sample size A total of 83 measures from 78 males were sampled in this study, where 5 males were sampled during 2 consecutive study years. We considered these additional measurements to be statistically independent because none of these 5 males remained in the same social group between years. Individual attributes: male anogenital distance, male body weight, and male age Male morphotype in terms of masculinization level was assessed through anogenital distance measures. Males that were close to the short end of the anogenital distance distribution were considered to as slightly masculinized males, while males that were close to the long end of this gradient represented highly masculinized males (Correa 2012). Anogenital distance was measured as the distance from the ventral anus commissure to the base of the genital papilla, as defined by Vandenbergh and Hugget (1995). We measured the anogenital distance of each adult male with a digital caliper (precision 0.1 mm) at every capture event during the mating season. All anogenital distance measurements were taken by the same observer across all 6 years of the study. To obtain a single measure per subject, all anogenital distance measures taken in the same season were averaged per male degu (Correa et al. 2016). Males that were present in 2 consecutive years had 2 estimations of average of anogenital distance, because between years, males grow. The number of anogenital measures per degu averaged 8.08 ± 4.48 (range: 2–23). Repeatability of anogenital distance and its correlation with male body weight were calculated according to the methods of Correa et al. (2016). Anogenital distance was used as a continuous predictor during all subsequent statistical analyses. Mean male body weight during the mating period was recorded from trapping data. The number of body weight measures per male during the mating period averaged 11 ± 5.84 (range: 2–25). Males in our study were either young of the year (age = 1, n = 66) or had survived until a second year (age = 2, n = 17). Five males were present in 2 consecutive years. Social group determination The main criterion used to assign individuals to social groups was the sharing of burrow systems at night. The sharing of burrow systems was determined by 1) burrow trapping during early morning activity and 2) night-time telemetry. To determine group composition, we first compiled a symmetric similarity matrix of pairwise associations of burrow locations of all adult degus during trapping and telemetry (Whitehead 2008). The association (overlap) between any 2 individuals was determined by dividing the number of early mornings that these individuals were captured at or tracked with radiotelemetry to the same burrow system, by the number of early mornings that both individuals were trapped or tracked with radiotelemetry on the same day (Ebensperger et al. 2004; Hayes et al. 2009). To determine social group composition, a hierarchical cluster analysis of the association matrix was conducted using SOCPROG software (Whitehead 2009). The fit of the data were analyzed using cophenetic correlation coefficients, correlations between the actual association indices, and the levels of clustering in the diagram. In this procedure, values above 0.8 indicate that hierarchical cluster analysis provided an effective representation of the data (Whitehead 2008). The maximum modularity criterion (Newman 2004) was used to cut off the dendrogram and define social groups. 83 males were included in this analysis because they were assigned to a well-defined social group. Additionally, 41 males were excluded from all analyses because they were not assigned to a well-defined social group. The total number of social groups (11 ± 5, for winter and 16 ± 7, for spring) and male and female group members monitored during winter of each year are reported in Supplementary Table S3 of Supplementary Data. Social group attributes Social group attributes recorded were 1) the number of adult females (a measure of male mating opportunities), 2) the number of adult males (a measure of male competition), and 3) the mean male anogenital distance within groups (i.e., a measure of similarity between male group members). These attributes were estimated once per mating season because degu social group composition is relatively stable across the breeding season (Ebensperger et al. 2009). Genetic methods and paternity analyses We genotyped a total of 1517 individuals captured between 2010 and 2015 (Supplementary Table S4 of Supplementary Data). Tissue samples (a 1 × 5 mm ear snip) were taken from each individual. Tissue samples were stored in ethanol 99% at 5–6 °C until analysis. DNA was extracted using the Reliaprep DNA animal tissue miniprep system kit (Promega) mouse tail protocol. DNA was eluted in 200 µl of nuclease-free water and stored at −20 °C. We worked with 10 microsatellite loci, 9 from O. degus (Quan et al. 2009) and one from S. cyanus (Schroeder et al. 2000). These loci were amplified via polymerase chain reaction (PCR), with the following protocol: 15 min at 94 °C for DNA denaturation, 30 cycles of a 1 min denaturation step at 94 °C, followed by 1 min of locus-specific annealing temperature (Supplementary Table S5 of Supplementary Data), 1 min at 72 °C for elongation, and a final elongation step of 10 min at 72 °C. For fragment analysis, the PCR products were mixed in 3 combinations (2 with 4 loci each and one with 2 loci). Each of these mixes was contrasted with an internal size standard and analyzed using an ABI Prism 3130Xl genetic analyzer and allele sizes were determined using the Genemapper software (Applied Biosystems). All loci amplified successfully and were polymorphic (Supplementary Table S5 of Supplementary Data). Genotypes for all individuals across years were complete with no missing data. We tested the Hardy-Weinberg observed and expected heterozygosity for each study year with CERVUS 3.0 software (Marshall et al. 1998). Deviations from Hardy-Weinberg expectations were detected in 5 out of 6 years (Supplementary Table S6 of Supplementary Data) and were not the consequence of null allele presence (all markers were checked for null alleles with Microchecker software, van Oosterhout et al. 2004). This finding was expected because our study population was open, nonpanmitic, and characterized by a relatively high level of genetic relatedness (Quirici et al. 2011; Davis et al. 2016). We used CERVUS 3.0 software (Marshall et al. 1998) to conduct paternity analyses, in which all offspring were checked against all potential fathers in the population. Confidence in parentage assignments calculations was made using the mean logarithm of odds (LOD) score option in CERVUS 3.0. Simulations were run for 10,000 cycles using allele frequency data from the entire population, with a genotyping error rate of 1% and under the assumption that 90% of the population was sampled. Paternity assignment analyses were made using strict (95%) confidence levels. Paternity was assigned when the following conditions were met: 1) the LOD score for the father-offspring pair was positive, 2) the father-offspring pair confidence level was significant, and 3) there was 0 or 1 mismatch. Paternity assignments with 2 or more mismatches were discarded. Supplementary Table S7 of the Supplementary Data displays the number of offspring that were or were not assigned to each father. Determination of reproductive success Using paternity analysis data, we determined the number of offspring sired for each male. This variable was measured at weaning, when offspring become trappable. Annual trapping ended when the daily capture rate of new offspring was less than 5% of all offspring captured (Correa et al. 2016). Statistical analyses An information theoretic approach (Burnham and Anderson 2002) was adopted to assess the inclusion of predictors in a model of reproductive success (hypothesis 1, prediction i, ii and iii; hypothesis 2, prediction iv; hypothesis 3, prediction vi). The overall procedure consisted of specifying a full model with all plausible fixed and random terms. During the first step, random effect components were retained whenever the variance explained by these components was higher or at least in the same order of magnitude compared with the residual error term. During the second step, we generated a list of competing models assembled from all possible combinations of fixed effects declared in the full model, and fit each model to the data. Models were ranked according to parsimony by means of AICc, and we selected the model with the lowest AIC value for further analysis and interpretation (see Supplementary Table S8 of the Supplementary Data). For hypotheses 1 (prediction i, ii and iii), 2 (prediction iv), and 3 (prediction vi), we used generalized linear mixed model to test whether individual and social group attributes independently influenced male reproductive success. Additionally, we included a mean male anogenital distance within group by male anogenital distance interaction. Thus, the initial model examined (model 1) was: Number of offspring sired = male anogenital distance + male mean body weight + male age (1 and 2 years old) + number of adult females + number of adult males + mean male anogenital distance within group + (anogenital distance x mean male anogenital distance within group). Year and the social group ID were added as random effects. The number of offspring sired by males was assumed to be Poisson distributed (no overdispersion was detected) and the link function used was the log-link. Outlier influence analysis and qq-plots revealed no outlier observations or departures from model assumptions. We used a generalized linear mixed model to examine how reproductive success would vary across different types of social groups (hypothesis 2, prediction v). Categories of social groups included male-only groups, multi-male/multi-female groups, and solitary males. Year and the social unit ID were added as random effects. The number of offspring sired was fit to a Poisson distribution (no overdispersion was detected). Goodness of fit for this model was obtained comparing against an intercept-only with the same random effect structure by means of Likelihood Ratio Test. We needed to examine the potential effect of type of social unit separately because solitary males lack all other social attributes included in the previous models. All statistical analyses were conducted in R 3.2.3. Generalized mixed models were fitted using library lme4 1.1–12 (Bates et al. 2015); best subset model selection was done using functions included in library MuMIn ver. 1.15.6 (Barton 2009); Variance inflation factor values were estimated using formulas in Zuur et al. (2009); and power analysis was done using library simr 1.0.2 (Green and MacLeod 2016). All libraries were accessed on March 2017. Ethical note Animal handling techniques and all protocols used in this study were approved and supervised by the Ethics Committee of the Pontificia Universidad Católica of Chile (CBB-155, 2012 resolution, supervised and approved 03/03/2015), and followed the Chilean Ethical Legislation (Permits 1–31/2009 [1956], 3881/2012, and 2826/2013 by the Servicio Agricola y Ganadero). Blood and tissue sampling were performed by well-trained veterinarians and biologists (C.L. and J.R.-E.). RESULTS Repeatability of male anogenital distance Across all 6 study years, population mean male anogenital distance was 9.5 ± 0.9 mm (SD) (range: 7.25 mm to 11.83 mm, n = 714 measurements from 78 males). Intraclass correlation coefficient (ICC) or intraseason repeatability of male anogenital distance was 0.81 (95% confidence interval: 0.75–0.86, n = 627 measurement for 78 males). Intraclass correlation coefficient (ICC) or interyear repeatability of male anogenital distance was 0.64 (95% confidence interval: 0.36–0.94, n = 87 measurement for 5 males). Male anogenital distance and male body weight were positively correlated (Pearson’s correlation r = 0.352; t = 3.389, P = 0.00108). However, this association did not result in model collinearity (correlation of fixed effects in glmm = 0.049). Types of social groups From 83 male measurements, 67 males were members of 41 multi-male/multi-female groups, 6 were members of male-only groups, and 10 were solitary individuals (Table 1). In multi-male/multi-female groups, the number of females averaged 2.1 ± 1.7 females per group (range: 1–9), while the number of males averaged 1.6 ± 1.0 males per group (range: 1–5). Mean male anogenital distance within multi-male/multi-female groups averaged 9.5 ± 1.0 (range: 7.3–11.5). The number of males in the only male groups was 2 ± 0 (3 pairs). Mean male anogenital distance within male-only groups averaged 10.2 ± 0.7 (range: 9.6–11.1). Table 1 Number and attributes of social groups studied during 2010–2015 Year 2010 2011 2012 2013 2014 2015 Number of groups (n) 5 4 2 9 15 9  Multi-male/multi-female groups 5 4 2 8 14 8  Male only groups 0 0 0 1 1 1 Group size  Mean 3 5 2 4 3 4  SD 1 1 0 3 1 3  Range 2–4 3–6 2–10 2–5 2–11 Number of females  Mean 2 2 1 2 2 3  SD 1 1 0 3 1 2  Range 1–3 1–3 0–9 0–4 0–6 Number of males  Mean 2 3 1 1 2 2  SD 1 2 0 1 1 1  Range 1–2 1–5 1–3 1–3 1–5 Number solitary males 1 3 0 4 1 1 Year 2010 2011 2012 2013 2014 2015 Number of groups (n) 5 4 2 9 15 9  Multi-male/multi-female groups 5 4 2 8 14 8  Male only groups 0 0 0 1 1 1 Group size  Mean 3 5 2 4 3 4  SD 1 1 0 3 1 3  Range 2–4 3–6 2–10 2–5 2–11 Number of females  Mean 2 2 1 2 2 3  SD 1 1 0 3 1 2  Range 1–3 1–3 0–9 0–4 0–6 Number of males  Mean 2 3 1 1 2 2  SD 1 2 0 1 1 1  Range 1–2 1–5 1–3 1–3 1–5 Number solitary males 1 3 0 4 1 1 View Large Table 1 Number and attributes of social groups studied during 2010–2015 Year 2010 2011 2012 2013 2014 2015 Number of groups (n) 5 4 2 9 15 9  Multi-male/multi-female groups 5 4 2 8 14 8  Male only groups 0 0 0 1 1 1 Group size  Mean 3 5 2 4 3 4  SD 1 1 0 3 1 3  Range 2–4 3–6 2–10 2–5 2–11 Number of females  Mean 2 2 1 2 2 3  SD 1 1 0 3 1 2  Range 1–3 1–3 0–9 0–4 0–6 Number of males  Mean 2 3 1 1 2 2  SD 1 2 0 1 1 1  Range 1–2 1–5 1–3 1–3 1–5 Number solitary males 1 3 0 4 1 1 Year 2010 2011 2012 2013 2014 2015 Number of groups (n) 5 4 2 9 15 9  Multi-male/multi-female groups 5 4 2 8 14 8  Male only groups 0 0 0 1 1 1 Group size  Mean 3 5 2 4 3 4  SD 1 1 0 3 1 3  Range 2–4 3–6 2–10 2–5 2–11 Number of females  Mean 2 2 1 2 2 3  SD 1 1 0 3 1 2  Range 1–3 1–3 0–9 0–4 0–6 Number of males  Mean 2 3 1 1 2 2  SD 1 2 0 1 1 1  Range 1–2 1–5 1–3 1–3 1–5 Number solitary males 1 3 0 4 1 1 View Large Number of offspring sired by males at weaning The best model explaining variation in male reproductive success included the main effects of 4 predictors and one random effect (Table 2). The fixed factors explained 23% of variation in the number of offspring sired by males. Together, the fixed factors and random factor (social group ID) explained 60% of variation in the number of offspring sired by males. The best model revealed a positive relationship between male anogenital distance and the number of offspring sired by males. Males with longer anogenital distances sired more offspring than males with shorter anogenital distances (Figure 1a), such that each 1 mm increase in anogenital distance would result in a 40% more offspring (i.e., e0.36 = 1.4). One-year-old males sired more offspring (mean = 3.83 ± 4.9) than 2-year-old males (mean = 0.94 ± 1.4) (Figure 1b), such that for every year a male ages, reproductive success would decrease by 13.5% fewer offspring (i.e., e−1.96 = 0.135). Similarly, heavier males were associated with a larger number offspring sired (according to our directional hypothesis predictions), such that a 1 g increase in body weight would result in 1% more offspring (i.e., e0.009 = 1.009) (Figure 1c). In contrast, mean anogenital distance within groups did not predict the number of offspring sired by males. None of the social attributes and neither the interaction of male anogenital distance by mean anogenital distance within social groups were included in the best model, implying no modulatory role of social attributes on male morphotype. Table 2 Effects of individual male attributes and social group attributes on the number of offspring sired by males Fixed Factors Estimate Standard error Z P VIF (Intercept) 0.03 1.625 0.018 0.986 Male anogenital distance 0.36 0.152 2.337 0.019 1.88 Male age (2) −1.96 0.404 −4.856 <0.001 1.18 Male body weight 0.01 0.005 1.760 0.074 1.34 Mean anogenital distance group −0.42 0.240 −1.760 0.078 2.08 Random Factors σ2 Social group ID 0.962 σ2 error 1.080 Number of social groups 54 Number of males 83 Family Poisson (log) Fixed Factors Estimate Standard error Z P VIF (Intercept) 0.03 1.625 0.018 0.986 Male anogenital distance 0.36 0.152 2.337 0.019 1.88 Male age (2) −1.96 0.404 −4.856 <0.001 1.18 Male body weight 0.01 0.005 1.760 0.074 1.34 Mean anogenital distance group −0.42 0.240 −1.760 0.078 2.08 Random Factors σ2 Social group ID 0.962 σ2 error 1.080 Number of social groups 54 Number of males 83 Family Poisson (log) Terms reported are those from the most parsimonious model retained after using a best subset model selection routine based on Akaike Information Criteria. Significance was defined as P < 0.05. VIF = Variance inflation factor. View Large Table 2 Effects of individual male attributes and social group attributes on the number of offspring sired by males Fixed Factors Estimate Standard error Z P VIF (Intercept) 0.03 1.625 0.018 0.986 Male anogenital distance 0.36 0.152 2.337 0.019 1.88 Male age (2) −1.96 0.404 −4.856 <0.001 1.18 Male body weight 0.01 0.005 1.760 0.074 1.34 Mean anogenital distance group −0.42 0.240 −1.760 0.078 2.08 Random Factors σ2 Social group ID 0.962 σ2 error 1.080 Number of social groups 54 Number of males 83 Family Poisson (log) Fixed Factors Estimate Standard error Z P VIF (Intercept) 0.03 1.625 0.018 0.986 Male anogenital distance 0.36 0.152 2.337 0.019 1.88 Male age (2) −1.96 0.404 −4.856 <0.001 1.18 Male body weight 0.01 0.005 1.760 0.074 1.34 Mean anogenital distance group −0.42 0.240 −1.760 0.078 2.08 Random Factors σ2 Social group ID 0.962 σ2 error 1.080 Number of social groups 54 Number of males 83 Family Poisson (log) Terms reported are those from the most parsimonious model retained after using a best subset model selection routine based on Akaike Information Criteria. Significance was defined as P < 0.05. VIF = Variance inflation factor. View Large Figure 1 View largeDownload slide Male individual attributes and their effects on the number of offspring sired by males at weaning (n = 83). (a) Positive relationship between male anogenital distance and the number of offspring sired by males. (b) Effects of male age on the number offspring sired by males. (c) Positive relationship between male body weight and the number of offspring sired by males (mean ± 1SE). Figure 1 View largeDownload slide Male individual attributes and their effects on the number of offspring sired by males at weaning (n = 83). (a) Positive relationship between male anogenital distance and the number of offspring sired by males. (b) Effects of male age on the number offspring sired by males. (c) Positive relationship between male body weight and the number of offspring sired by males (mean ± 1SE). Social group type and the number of offspring sired at weaning The number of offspring sired did not differ across different types of social groups, including multi-male/multi-female groups, male-only groups, and solitary males. Males in multi-male/multi-female groups tended to sire more offspring than male-only groups and solitary males, but this was not statistically significant (Likelihood Ratio Test χ2 = 5.39, P = 0.0677, df = 4; Figure 2). Figure 2 View largeDownload slide Social group type (solitary males n = 10; male-only groups n = 6; multi-male/multi-female groups n = 67) and its effects on the number of offspring sired by males at weaning. Data are presented as mean ± 1SE. Figure 2 View largeDownload slide Social group type (solitary males n = 10; male-only groups n = 6; multi-male/multi-female groups n = 67) and its effects on the number of offspring sired by males at weaning. Data are presented as mean ± 1SE. DISCUSSION We assessed how male individual attributes, social group attributes and their interaction influence male reproductive success. Male reproductive success increased with male anogenital distance, but decreased with male age. In agreement with our prediction, male masculinization level was the most important predictor of male reproductive success. This finding suggests intriguing parallels with a previous study on Mongolian gerbils and studies on domestic guinea pigs. In particular, highly masculinized male Mongolian gerbils sired significantly more offspring than slightly masculinized males in captivity (Clark et al. 1992). Likewise, in captivity, infantilized guinea pig males sired fewer offspring than aggressive males (Guenther et al. 2014; Zimmermann et al. 2017). Thus, our current study on degus strongly supports that male phenotypical masculinization level has consequences on male reproductive success under natural conditions, and where highly masculinized males may attain a fitness advantage. Higher reproductive success in highly masculinized male degus, is not influenced for female effects. Previously, we found that in degus, heavier females are more fertile and masculinized females wean heavier pups (Correa et al. 2016) however, male degus did not select a particular type of female (based on her anogenital distance or body mass) to mate, but seek to mate with as many females as possible (L.A. Ebensperger, unpublished data). The masculinization gradient in males has been studied in several species of rodents (Ward 1972; vom Saal 1989; Drickamer 1996; Ryan and Vandenbergh 2002; Ophir and delBarco-Trillo 2007; Siegeler et al. 2011; Godsall et al. 2014) and in domestic rabbits (Bánszegi et al. 2015). These studies have revealed how masculinization may result in other phenotypic differences with potential consequences on reproductive success. For example, highly masculinized males exhibit heavier/bigger testes and ventral scent glands, larger seminal vesicles and chin glands, and more developed genital musculature than slightly masculinized males (Clark and Galef 1998; vom Saal et al. 1999; Ryan and Vandenbergh 2002; Bánszegi et al. 2015). Masculinization also results in behavioral differences. In addition to being more aggressive (Ryan and Vandenbergh 2002), highly masculinized males scent mark more frequently (Bánszegi et al. 2015), disperse more often, and have larger home ranges than less slightly masculinized males (Drickamer 1996; Godsall et al. 2014). Highly masculinized males are sexually more active, are more attractive to females (Ophir and delBarco-Trillo 2007), mount females at higher rates, and exhibit longer time to ejaculation than slightly masculinized males (Ward 1972; Clark and Galef 1998; vom Saal et al. 1999). Together, these features suggest that masculinization during prenatal develop can have profound impacts on the reproductive success of adult males. Subsequent studies in degus are needed to address how some of these morphological and behavioral characteristics of males may determine the natural differences in reproductive success recorded among male morphotypes in this study. An intriguing question from our study is “under what conditions slightly masculinized males may attain a fitness advantage in the population?” Previous studies on laboratory-reared rodents suggest at least 3 potential scenarios. First, slightly masculinized male Mongolian gerbils are more attentive to offspring and remain with their mate during postpartum oestrus (Clark et al. 1997). The presence of these kinds of males’ decreases maternal costs and increases female fecundity during postpartum estrus. Second, Clark and Galef (2000) reported the existence of a subgroup of slightly masculinized males that exhibit no sexual interest in females and fail to mate. However, these “asexual” males provide exceptionally high quality of parental care, a tactic that may benefit these males if nesting with genetically related females (Clark and Galef 2000). Third, infantilized male guinea pigs exhibit low aggressiveness, a trait that may reduce costs of living in large, complex social systems. These males might be using a queuing strategy before attaining social dominance status (Kaiser and Sachser 2009; Sachser et al. 2011; Siegeler et al. 2011; Guenther et al. 2014; Zimmermann et al. 2017). Female degus exhibit postpartum oestrus, yet this second breeding does not always result in a second litter per year in our study population (Ebensperger et al. 2013, but see Previtali et al. 2010). During years in which females produce 2 litters, second litters represent no more than 9% of total offspring produced during lifetime (L.A. Ebensperger, unpublished data). These conditions suggest a queuing strategy is unlikely to be profitable to slightly masculinized male degus. On the other hand, the observation that male degus exhibit parental care in captivity (Ebensperger et al. 2010), suggests an alternative way in which slightly masculinized males could attain indirect fitness benefits through rearing the offspring of related females, as suggested by Clark and Galef (2000). Our next step will be to evaluate the hypothesis of male indirect fitness benefits through the provision of paternal care. This will be done through studies in captivity that evaluate the paternal behavior of males of different levels of masculinization, and through studies in a natural population, evaluating the level of relatedness between adult males and females that share same burrow system during parental care period. The observation of a negative relationship between age and reproductive success is not surprising. In our study population, most adults (90%) do not survive to their second year of life, implying these rodents are ecologically semelparous (Ebensperger et al. 2013; Correa et al. 2016). In captivity, where degus can reproduce successfully until 4 years of age, reproductive output decreases in a sustained way from the first to the fourth reproductive event (Correa 2012). A comparative study of rodents suggests that high mortality rates can modify reproductive strategies, from slow to fast life-history strategies and that fast life-history species tend to have precocial offspring (Dobson and Oli 2007). The ecological and reproductive characteristics of degus are consistent with these findings. Under these conditions, male and female degus likely invest most reproductive effort in their first reproductive event, despite the trade-off between current and future reproduction (Hamel et al. 2010; Lemaître et al. 2015; Correa et al. 2016). Some support to this possibility comes from the breeding strategy of female degus. Females with higher cortisol (a measure of reproductive effort) levels during main lactation season produce more surviving offspring, but reduce their ability to produce a second litter (Ebensperger et al. 2011, 2013). The observation that interannual survivorship of adult males is lower than that of females (Ebensperger et al. 2009) suggests males are experiencing similar pressures to invest more during their first (and likely the only) breeding event. According with our original expectation, male reproductive success of degus was influenced by body weight, but the magnitude of the effect was extremely small. This result is partially in agreement with evidence from wild cavies, a closely related species in which heavier and territorial males sire 80% of offspring (Asher et al. 2008). Differences in the magnitude of the male body weight effect on fitness between degus and wild cavies are in agreement with what would be expected from species with different mating systems. As suggested previously (Clutton-Brock 1989; Schradin and Lindholm 2011), male body weight is expected to be a more important predictor of male reproductive success in polygynous species (i.e., wild cavies) than in promiscuous counterparts (i.e., degus). Male body weight is thought to be important for males to attain higher reproductive success when conditions allow monopolization of clumped females. Conversely, in species with promiscuous mating systems male searching ability is expected to be more important to locate relatively dispersed females (Schwagmeyer 1988; Clutton-Brock 1989; Schradin and Lindholm 2011). In agreement with these expectations, male African striped mice (Rhabdomys pumilio) may change their mating tactic depending on the female grouping pattern (dispersed, grouped) (Schradin et al. 2009; Schradin et al. 2010; Schradin and Lindholm 2011). Our study revealed that individual attributes have a larger impact than social attributes on the reproductive success of male degus. This result parallels findings for females of the same population (Correa et al. 2016). In male and female degus, masculinization level is positively related to number of offspring sired and offspring body mass at weaning. Together, these findings indicate how mammalian phenotypical masculinization might be adaptive regardless of social context. This scenario differs from that was suggested by previous studies in other social rodents such as cavies, where success of infantilized/aggressive males would be contingent upon social environment (Kaiser and Sachser 2009; Sachser et al. 2011; Siegeler et al. 2011; Guenther et al. 2014; Zimmermann et al. 2017). SUPPLEMENTARY MATERIAL Supplementary data are available at Behavioral Ecology online. FUNDING This work was supported by a FONDECYT grant (3130567 and 11170222 to L.A.C); FONDECYT grant (1090302, 1130091, and 1170409 to L.A.E); CONICYT PhD thesis grant (21120244 to A.L-P); and NSF OISE grant (0853719 and 1261026 to L.D.H). We are grateful to Gioconda Peralta, Lucie Jaugeon, Loreto Carrasco, and Sylvain Faugeron from the Molecular Diversity Laboratory, Pontificia Universidad Católica de Chile, for their valuable help and advice with molecular analyses. We are indebted to the Universidad de Chile, particularly to Marcelo Orellana Reyes and Rosa Peralta (Field Station Administrators), for providing facilities during field work at Rinconada. We also thank David Véliz for help with molecular data analyses and Carolyn Bauer for providing extremely useful suggestions on a previous version of this manuscript. Data accessibility: Analyses reported in this article can be reproduced using the data provided by Correa et al. (2018). Handling editor: Anna Lindholm REFERENCES Asher M , Lippmann T , Epplen JT , Kraus C , Trillmich F , Sachser N . 2008 . Large males dominate: ecology, social organization and mating system of wild cavies, the ancestors of the Guinea pig . Behav Ecol Sociobiol . 62 : 1509 – 1521 . Google Scholar CrossRef Search ADS Bánszegi O , Szenczi P , Dúcs A , Hudson R , Altbäcker V . 2015 . Long-term under-masculinization in male rabbits due to maternal stress is reversed by prenatal administration of testosterone . Behav Processes . 115 : 156 – 162 . Google Scholar CrossRef Search ADS PubMed Barton K . 2009 . MuMIn: Multi-model inference . Available from: http://r-forge. r- project.org/projects/mumin. Bates D , Maechler M , Bolker B , Walker S . 2015 . Fitting linear mixed-effects models using lme4 . J Stat Sofw . 67 : 1 – 48 . Burnham KP , Anderson DR . 2002 . Model selection and multi-model inference: a practical information-theoretic approach ( 2nd ed .). New York (USA) : Springer-Verlag . di Castri F , Hajek E . 1976 . Bioclimatología de Chile . Santiago (Chile) : Pontificia Universidad Católica de Chile . Clark MM , Desousa D , Vonk J , Galef BG Jr . 1997 . Parenting and potency: alternative routes to reproductive success in male Mongolian gerbils . Anim Behav . 54 : 635 – 642 . Google Scholar CrossRef Search ADS PubMed Clark MM , Galef BG Jr . 1998 . Effects on intrauterine position on the behavior and genital morphology of litter bearing rodents . Dev Neuropsychol . 14 : 197 – 211 . Google Scholar CrossRef Search ADS Clark MM , Galef BG Jr . 2000 . Why some male Mongolian gerbils may help at the nest: testosterone, asexuality and alloparenting . Anim Behav . 59 : 801 – 806 . Google Scholar CrossRef Search ADS PubMed Clark MM , Tucker L , Galef BG Jr . 1992 . Stud and dud males: intrauterine position effects on the reproductive success of male gerbils . Anim Behav . 43 : 215 – 221 . Google Scholar CrossRef Search ADS Clutton-Brock TH . 1989 . Mammalian mating systems . Proc R Soc Lond B Biol Sci . 236 : 339 – 372 . Google Scholar CrossRef Search ADS PubMed Clutton-Brock TH , Huchard E . 2013 . Social competition and selection in males and females . Philos Trans R Soc Lond B Biol Sci . 368 : 20130074 . Google Scholar CrossRef Search ADS PubMed Correa LA . 2012 . Mecanismos de regulación del tamaño de camada y razón de sexos en Octodon degus: efectos de la alostasis prenatal en la variación fenotípica de las crías y sus consecuencias en la estabilidad de los grupos sociales . PhD tesis, Valdivia : Universidad Austral de Chile . Correa LA , León C , Ramírez-Estrada J , Ly-Prieto A , Abades S , Hayes LD , Soto-Gamboa M , Ebensperger LA . 2018 . Data from: masculinized and younger males attain higher reproductive success in a social rodent . Dryad Digital Repository . http://dx.doi.org/10.5061/dryad.cb11j Correa LA , León C , Ramírez-Estrada J , Soto-Gamboa M , Sepúlveda RD , Ebensperger LA . 2016 . Masculinized females produce heavier offspring in a group living rodent . J Anim Ecol . 85 : 1552 – 1562 . Google Scholar CrossRef Search ADS PubMed Crump CJ , Chevins PF . 1989 . Prenatal stress reduces fertility of male offspring in mice, without affecting their adult testosterone levels . Horm Behav . 23 : 333 – 343 . Google Scholar CrossRef Search ADS PubMed Dobson FS , Oli MK . 2007 . Fast and slow life histories of rodents . In: Wolff JO , Sherman PW , editors. Rodent Societies an ecological and evolutionary perspective . Chicago (USA) : The University of Chicago Press . p. 99 – 105 . Davis GT , Vásquez RA , Poulin E , Oda E , Bazán-León EA , Ebensperger LA , Hayes LD . 2016 . Octodon degus kin and social structure . J. Mamm . 97 : 361 – 372 . Google Scholar CrossRef Search ADS Drickamer LC . 1996 . Intrauterine position and anogenital distance in house mice: consequences under field conditions . Anim Behav . 51 : 925 – 934 . Google Scholar CrossRef Search ADS Dubuc C , Ruiz-Lambides A , Widdig A . 2014 . Variance in male lifetime reproductive success and estimation of the degree of polygyny in a primate . Behav Ecol . 25 : 878 – 889 . Google Scholar CrossRef Search ADS PubMed Ebensperger LA , Chesh AS , Castro RA , Ortiz- Tolhuysen L , Quirici V , Burger JR , Hayes LD . 2009 . Instability rules social groups in the communal breeder rodent Octodon degus . Ethology . 115 : 540 – 554 . Google Scholar CrossRef Search ADS Ebensperger LA , Correa LA , León C , Ramírez-Estrada J , Abades S , Villegas Á , Hayes LD . 2016 . The modulating role of group stability on fitness effects of group size is different in females and males of a communally rearing rodent . J Anim Ecol . 85 : 1502 – 1515 . Google Scholar CrossRef Search ADS PubMed Ebensperger LA , Hurtado MJ . 2005 . Seasonal changes in the time budget of degus, Octodon degus . Behaviour . 142 : 91 – 112 . Google Scholar CrossRef Search ADS Ebensperger LA , Hurtado MJ , Soto-Gamboa M , Lacey EA , Chang AT . 2004 . Communal nesting and kinship in degus (Octodon degus) . Naturwissenschaften . 91 : 391 – 395 . Google Scholar CrossRef Search ADS PubMed Ebensperger LA , Ramírez-Estrada J , León C , Castro RA , Tolhuysen LO , Sobrero R , Quirici V , Burger JR , Soto-Gamboa M , Hayes LD . 2011 . Sociality, glucocorticoids and direct fitness in the communally rearing rodent, Octodon degus . Horm Behav . 60 : 346 – 352 . Google Scholar CrossRef Search ADS PubMed Ebensperger LA , Ramírez-Otarola N , León C , Ortiz ME , Croxatto HB . 2010 . Early fitness consequences and hormonal correlates of parental behaviour in the social rodent, Octodon degus . Physiol Behav . 101 : 509 – 517 . Google Scholar CrossRef Search ADS PubMed Ebensperger LA , Tapia D , Ramírez-Estrada J , León C , Soto-Gamboa M , Hayes LD . 2013 . Fecal cortisol levels predict breeding but not survival of females in the short-lived rodent, Octodon degus . Gen Comp Endocrinol . 186 : 164 – 171 . Google Scholar CrossRef Search ADS PubMed Ebensperger LA , Villegas A , Abades S , Hayes LD . 2014 . Mean ecological conditions modulate the effects of group living and communal rearing on offspring production and survival . Behav Ecol . 25 : 862 – 870 . Google Scholar CrossRef Search ADS Emlen ST , Oring LW . 1977 . Ecology, sexual selection, and the evolution of mating systems . Science . 197 : 215 – 223 . Google Scholar CrossRef Search ADS PubMed Engqvist L , Taborsky M . 2016 . The evolution of genetic and conditional alternative reproductive tactics . Proc Biol Sci . 283 : 20152945 . Google Scholar CrossRef Search ADS PubMed Fulk GW . 1976 . Notes on the activity, reproduction, and social behavior of Octodon degus . J Mamm . 57 : 495 – 505 . Google Scholar CrossRef Search ADS Godsall B , Coulson T , Malo AF . 2014 . From physiology to space use: energy reserves and androgenization explain home-range size variation in a woodland rodent . J Anim Ecol . 83 : 126 – 135 . Google Scholar CrossRef Search ADS PubMed Green P , MacLeod CJ . 2016 . SIMR: an R package for power analysis of generalized linear mixed models by simulation . Meth Ecol Evol . 7 : 493 – 498 . Google Scholar CrossRef Search ADS Gross MR . 1996 . Alternative reproductive strategies and tactics: diversity within sexes . Trends Ecol Evol . 11 : 92 – 98 . Google Scholar CrossRef Search ADS PubMed Guenther A , Kowalski G , von Engelhardt N . 2014 . Prenatal social conditions shape offspring adult phenotype and reproductive success . Behav Ecol Sociobiol . 68 : 1661 – 1667 . Google Scholar CrossRef Search ADS Hamel S , Gaillard JM , Yoccoz NG , Loison A , Bonenfant C , Descamps S . 2010 . Fitness costs of reproduction depend on life speed: empirical evidence from mammalian populations . Ecol Lett . 13 : 915 – 935 . Google Scholar CrossRef Search ADS PubMed Hayes LD , Chesh AS , Castro RA , Ortiz-Tolhuysen L , Burger JR , Bhattacharjee J , Ebensperger LA . 2009 . Fitness consequences of group living in the degu Octodon degus, a plural breeder rodent with communal care . Anim Behav . 78 : 131 – 139 . Google Scholar CrossRef Search ADS Hayes LD , Chesh AS , Ebensperger LA . 2007 . Ecological predictors of range areas and use of burrow system in the diurnal rodent, Octodon degus . Ethology . 113 : 155 – 165 . Google Scholar CrossRef Search ADS Hotchkiss AK , Vandenbergh JG . 2005 . The anogenital distance index of mice (Mus musculus domesticus): an analysis . Contemp Top Lab Anim Sci . 44 : 46 – 48 . Google Scholar PubMed Kaiser S , Kruijver FP , Straub RH , Sachser N , Swaab DF . 2003 . Early social stress in male Guinea-pigs changes social behaviour, and autonomic and neuroendocrine functions . J Neuroendocrinol . 15 : 761 – 769 . Google Scholar CrossRef Search ADS PubMed Kaiser S , Sachser N . 2009 . Effects of prenatal social stress on offspring development pathology or adaptation ?. Curr Dir Psychol Sci . 18 : 118 – 121 . Google Scholar CrossRef Search ADS Kaiser S , Sachser N . 2001 . Social stress during pregnancy and lactation affects in guinea pigs the male offsprings’ endocrine status and infantilizes their behaviour . Psychoneuroendocrinology . 26 : 503 – 519 . Google Scholar CrossRef Search ADS PubMed Knapp R . 2004 . Endocrine mediation of vertebrate male alternative reproductive tactics: the next generation of studies . Integr Comp Biol . 43 : 658 – 668 . Google Scholar CrossRef Search ADS von Kuerthy C , Ros AF , Taborsky M . 2016 . Androgen responses to reproductive competition of males pursuing either fixed or plastic alternative reproductive tactics . J Exp Biol . 219 : 3544 – 3553 . Google Scholar CrossRef Search ADS PubMed Lemaître JF , Berger V , Bonenfant C , Douhard M , Gamelon M , Plard F , Gaillard JM . 2015 . Early-late life trade-offs and the evolution of ageing in the wild . Proc Biol Sci . 282 : 20150209 . Google Scholar CrossRef Search ADS PubMed Marshall TC , Slate J , Kruuk LE , Pemberton JM . 1998 . Statistical confidence for likelihood-based paternity inference in natural populations . Mol Ecol . 7 : 639 – 655 . Google Scholar CrossRef Search ADS PubMed Moore MC , Hews DK , Knapp R . 1998 . Hormonal control and evolution of alternative male phenotypes: generalizations of models for sexual differentiation . Am Zool . 38 : 133 – 151 . Google Scholar CrossRef Search ADS Newman ME . 2004 . Analysis of weighted networks . Phys Rev E Stat Nonlin Soft Matter Phys . 70 : 056131 . Google Scholar CrossRef Search ADS PubMed van Oosterhout L , Hutchinson WF , Wills DPM , Shipley P . 2004 . Micro-checker: software to identifying and correcting genotyping errors in microsatellite data . Mol Ecol Not . 4 : 535 – 538 . Google Scholar CrossRef Search ADS Ophir AG , Delbarco-Trillo J . 2007 . Anogenital distance predicts female choice and male potency in prairie voles . Physiol Behav . 92 : 533 – 540 . Google Scholar CrossRef Search ADS PubMed Previtali MA , Meserve PL , Kelt DA , Milstead WB , Gutierrez JR . 2010 . Effects of more frequent and prolonged El Niño events on life-history parameters of the degu, a long-lived and slow-reproducing rodent . Conserv Biol . 24 : 18 – 28 . Google Scholar CrossRef Search ADS PubMed Quan YF , Macmanes MD , Ebensperger LA , Lacey EA , Hayes LD . 2009 . Isolation and characterization of polymorphic microsatellite loci from Octodon degus . Mol Ecol Resour . 9 : 999 – 1001 . Google Scholar CrossRef Search ADS PubMed Quirici V , Faugeron S , Hayes LD , Ebensperger LA . 2011 . Absence of kin structure in a population of the group-living rodent Octodon degus . Behav Ecol . 22 : 248 – 254 . Google Scholar CrossRef Search ADS Raveh S , Heg D , Dobson FS , Coltman DW , Gorrell JC , Balmer A , Neuhausa P . 2010 . Mating order and reproductive success in male Columbian ground squirrels (Urocitellus columbianus) . Behav Ecol . 21 : 537 – 547 . Google Scholar CrossRef Search ADS Rhen T , Crews D . 2002 . Variation in reproductive behaviour within a sex: neural systems and endocrine activation . J Neuroendocrinol . 14 : 517 – 531 . Google Scholar CrossRef Search ADS PubMed Roff DA , Wolak ME , Correa LA , Soto-Gamboa M . 2017 . The phenotypic correlates and quantitative genetics of masculinization in the rodent, Octodon degus . Heredity . 119 : 136 – 141 . Google Scholar CrossRef Search ADS PubMed Ryan BC , Vandenbergh JG . 2002 . Intrauterine position effects . Neurosci Biobehav Rev . 26 : 665 – 678 . Google Scholar CrossRef Search ADS PubMed vom Saal FS . 1989 . Sexual differentiation in litter-bearing mammals: influence of sex of adjacent fetuses in utero . J Anim Sci . 67 : 1824 – 1840 . Google Scholar CrossRef Search ADS PubMed vom Saal FS , Clark MM , Galef BG , Drickamer LC , Vandenbergh JG . 1999 . Intrauterine position phenomenon . In: Knobil E , Neil JD , editors. Encyclopedia of Reproduction Vol. 2 . New York (USA) : Academic Press . p. 893 – 900 . Google Scholar CrossRef Search ADS Sachser N , Hennessy MB , Kaiser S . 2011 . Adaptive modulation of behavioural profiles by social stress during early phases of life and adolescence . Neurosci Biobehav Rev . 35 : 1518 – 1533 . Google Scholar CrossRef Search ADS PubMed Schroeder JW , Honeycutt RL , Rooney AP , Han G , Begall S , Gallardo MH . 2000 . Microsatellites from the South American coruro, Spalacopus cyanus . Mol Ecol . 9 : 1447 – 1449 . Google Scholar CrossRef Search ADS PubMed Schradin C , König B , Pillay N . 2010 . Reproductive competition favours solitary living while ecological constraints impose group-living in African striped mice . J Anim Ecol . 79 : 515 – 521 . Google Scholar CrossRef Search ADS PubMed Schradin C , Lindholm AK . 2011 . Relative fitness of alternative male reproductive tactics in a mammal varies between years . J Anim Ecol . 80 : 908 – 917 . Google Scholar CrossRef Search ADS PubMed Schradin C , Scantlebury M , Pillay N , König B . 2009 . Testosterone levels in dominant sociable males are lower than in solitary roamers: physiological differences between three male reproductive tactics in a sociably flexible mammal . Am Nat . 173 : 376 – 388 . Google Scholar CrossRef Search ADS PubMed Schwagmeyer PL . 1988 . Scramble competition polygyny in an asocial mammal: male mobility and mating success . Am Nat . 131 : 885 – 892 . Google Scholar CrossRef Search ADS Siegeler K , Sachser N , Kaiser S . 2011 . The social environment during pregnancy and lactation shapes the behavioral and hormonal profile of male offspring in wild cavies . Dev Psychobiol . 53 : 575 – 584 . Google Scholar CrossRef Search ADS PubMed Soto-Gamboa M , Villalón M , Bozinovic F . 2005 . Social cues and hormone levels in male Octodon degus (Rodentia): a field test of the Challenge Hypothesis . Horm Behav . 47 : 311 – 318 . Google Scholar CrossRef Search ADS PubMed Taborsky M , Oliveira RF , Brockmann HJ . 2008 . The evolution of alternative reproductive tactics: concepts and questions . In: Oliveira RF , Taborsky M , Brockmann HJ , editors. Alternative reproductive tactics: an Integrative Approach . Cambridge (UK) : Cambridge University Press . p. 1 – 22 . Google Scholar CrossRef Search ADS Vandenbergh JG , Huggett CL . 1995 . Mother’s prior intrauterine position affects the sex ratio of her offspring in house mice . Proc Natl Acad Sci USA . 91 : 11055 – 11059 . Google Scholar CrossRef Search ADS Ward IL . 1972 . Prenatal stress feminizes and demasculinizes the behavior of males . Science . 175 : 82 – 84 . Google Scholar CrossRef Search ADS PubMed Ward IL , Weisz J . 1980 . Maternal stress alters plasma testosterone in fetal males . Science . 207 : 328 – 329 . Google Scholar CrossRef Search ADS PubMed Whitehead H . 2008 . Analyzing animal societies: quantitative methods for vertebrate social analysis . Chicago (USA) : Chicago University Press . Google Scholar CrossRef Search ADS Whitehead H . 2009 . SOCPROG programs. Analyzing animal social structures . Behav Ecol Sociobiol . 63 : 765 – 778 . Google Scholar CrossRef Search ADS Zedrosser A , Bellemain E , Taberlet P , Swenson JE . 2007 . Genetic estimates of annual reproductive success in male brown bears: the effects of body size, age, internal relatedness and population density . J Anim Ecol . 76 : 368 – 375 . Google Scholar CrossRef Search ADS PubMed Zimmermann TD , Kaiser S , Hennessy MB , Sachser N . 2017 . Adaptive shaping of the behavioural and neuroendocrine phenotype during adolescence . Proc Roy Soc Lond B . 284 : 20162784 . Google Scholar CrossRef Search ADS Zuur AF , Ieno EN , Walker N , Saveliev AA , Smith GM . 2009 . Mixed effects models and extensions in ecology with R . New York (USA) : Springer-Verlag . Google Scholar CrossRef Search ADS © The Author(s) 2018. Published by Oxford University Press on behalf of the International Society for Behavioral Ecology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Behavioral EcologyOxford University Press

Published: Feb 21, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off