Geographic variation in body size and sexual dimorphism of the short‐nosed fruit bat (Cynopterus sphinx) was investigated in peninsular India. Bats were sampled at 12 localities along a 1200 km latitudinal transect that paralleled the eastern flanks of the Western Ghats. The geographic pattern of variation in external morphology of C. sphinx conforms to the predictions of Bergmann's Rule, as indicated by a steep, monotonic cline of increasing body size from south to north. This study represents one of the first conclusively documented examples of Bergmann's Rule in a tropical mammal and confirms that latitudinal clines in body size are not exclusively restricted to temperate zone homeotherms. Body size was indexed by a multivariate axis derived from principal components analysis of linear measurements that summarize body and wing dimensions. Additionally, length of forearm was used as a univariate index of structural size to examine geographic variation in a more inclusive sample of bats across the latitudinal transect. Multivariate and univariate size metrics were strongly and positively correlated with body mass, and exhibited highly concordant patterns of clinal variation. Stepwise multiple regression on climatological variables revealed that increasing size of male and female C. sphinx was associated with decreasing minimum temperature, increasing relative humidity, and increasing seasonality. Although patterns of geographic size variation were highly concordant between the sexes, C. sphinx also exhibited a latitudinal cline in the magnitude and direction of sexual size dimorphism. The size differential reversed direction across the latitudinal gradient, as males averaged larger in the north, and females averaged larger in the south. The degree of female‐biased size dimorphism across the transect was negatively correlated with body size of both sexes. Canonical discriminant analysis revealed that male‐ and female‐biased size dimorphism were based on contrasting sets of external characters. Available data on geographic variation in the degree of polygyny in C. sphinx suggests that sexual selection on male size may play a role in determining the geographic pattern of sexual size dimorphism.

Patterns of dispersion and site fidelity were investigated in a tent-roosting population of the short-nosed fruit bat, Cynopterus sphinx (Megachiroptera), in southern India. A local population of C. sphinx occupied diurnal roosts in a variable subset of 45 stem tents constructed within the dense foliage of mast trees (Polyalthia longifolia). Individually marked tent-roosting bats were visually censused over the course of a 38-d interval spanning the postpartum oestrus period. On any given day, 33.3–85.7% (mean = 60.8%, SD = 14.2) of adult males roosted singly, with the remainder holding harems of 1–10 breeding females (mean = 3.01, SD = 0.79). Average harem sex ratio was 2.8-fold higher than the adult sex ratio of the total tent-roosting population within the study area, indicating the potential for a high variance in male mating success within a single breeding season. Bats of both sexes typically occupied one primary tent, interspersed with shorter periods of residency in alternate tents. Males exhibited a significantly higher degree of roost fidelity than females. Some females roosted sequentially with different males and with different combinations of females, whereas others remained continuously associated with a single male and/or particular female roostmates over the duration of the census period. There were no statistically significant relationships between physical characteristics of tents and rates of occupancy by males or females. Intermittent transfers by females between groups suggest that the defence of diurnal roosts by males represents a more profitable mating strategy than the direct defence of compositionally labile female groups.

The short‐nosed fruit bat Cynopterus sphinx (Chiroptera: Pteropodidae) constructs shelters by severing stems of the curtain creeper, Vernonia scandens, and stems and leaves of the mast tree, Polyalthia longifolia, creating partially enclosed cavities (stem tents) in which to roost. Our observations indicate that the construction and maintenance of stem tents are primarily, if not exclusively, the behaviour of single males. A stem tent is formed in V. scandens when a single male C. sphinx severs up to 300 small‐ to medium‐sized stems creating a partially flattened, bell‐shaped cavity, and in P. longifolia when a male severs a few medium‐ to small‐sized branches and many leaf petioles, creating an entry/exit portal and space in which to roost. A tent constructed in V. scandens is completed in approximately 30 d, whereas one in P. longifolia is completed in about 50 d. Stem‐tent construction takes place mostly at night, but some stem chewing occurs in late afternoon. At night a stem tent is occupied by a single male, whereas females are usually absent. During the day the number of bats occupying completed tents is highly variable, ranging from two to 19 females (and their pups) and a single adult male. Tent construction is annually bimodal, which corresponds to a biannual breeding season. A dominant male sometimes deposits saliva on branches inside his tent cavity and actively defends this space from intrusions by other males. Both behaviours suggest forms of scent marking and territorial display. Our observations indicate that dominant males construct tents, recruit females and then defend the tents (and their female occupants) for the purpose of gaining reproductive access. The variance in harem group size indicates that some tent‐making males are more successful than others in recruiting females.

No abstract