| Body size is one of the key features of organisms, and variation in size is one of the most striking characteristics of life. However, phenotypic traits do not vary independently, but instead reflect webs of developmental, physiological and functional interactions of varying strengths. Allometry is the study of the relationship of body size to anatomy, physiology, ecology and finally behavior. Allometries could constrain phenotypic evolution by forcing evolving species along fixed trajectories. Alternatively, allometric relationships may result from natural selection for functional optimization. As a result, the relationship between allometry and phenotypic evolution is very tight, which is an important and popular topic in evolutionary ecology. The study of allometric and size scaling relationships have well flourished in most biological fields, such as physiological function, forage behavior and ecological traits, but have lagged in the area of animal behavior, especially for phenotypes correlating with courtship display, such as sexual size dimorphism(SSD) and acoustic communication signals. In this case, it will be one-sided for understanding deeply the relationship between allometry and phenotypic evolution.Rensch’s rule describes the allometry of SSD: when males are larger than females, SSD increases with increasing body size, but when females are larger than males, SSD decreases in larger species. Rensch’s rule is well supported for taxa that exhibit malebiased SSD or mixed SSD, but patterns of allometry among taxa with female-biased size dimorphism are mixed, there is evidence both for and against the rule. The majority of mammals exhibit male-biased SSD, bats, however, primarily display female-biased sexual size dimorphism. Whether patterns of SSD in bats conform to Rensch’s rule and whether male body size was more variable, were little known.Horseshoe bats emit high duty cycle calls including a prominent constant frequency(CF) component with most energy concentrated in the second harmonic. Echolocation calls play an important in the progress of sexual selection and speciation. Moreover, each species of horseshoe bats may evolve a ‘private bandwidth’ for species recognition and communication, thus the diversity of echolocation calls in Rhinolophidae is very high. The allometry hypothesis has been put forward to explain the evolution of diversity in echolocation frequencies of horseshoe bats. Allometry hypothesis predicts that the echolocation call frequencies of horseshoe bats normally decline as body size increases. However, in previous studies, body mass or FA has been commonly used to scale the allometric relationship between echolocation frequencies and body size in Rhinolophidae. Little is known about the relationship between echolocation frequency and morphological traits mediating directly echolocation calls’ production or reception(e.g. nasal capsule, cochlea or pinna).In this dissertation, we selected bats in Rhinolophidae as research object. The allometry and evolution of SSD and echolocation calls in horseshoe bats were investigated by comparative study at intraspecific and interspecific levels on a broad geographical scale. The details are as followed.1. We performed major axis linear regression to examined pattern of variation in SSD among 23 populations in Rhinolophus ferrumequinum on a global scale. Moreover, in order to clarify the causes of pattern of SSD variation, the relationship between body size and latitude were also tested. The results showed SSD was consistently femalebiased in greater horseshoe bat. The slope of major axis regression of log10(male) on log10(female) was significantly different from 1, suggesting variation in patterns of SSD among greater horseshoe bat populations is consistent with Rensch’s rule indicating that males were the more variable sex. Forearm length for both sexes of greater horseshoe bats was significantly negatively correlated with latitude, and males displayed a slightly but nonsignificant steeper latitudinal cline in body size than females. The results suggested that sex-specific latitudinal variation in body size may not be an important contributing factor to Rensch’s rule in Rhinolophus ferrumequinum.2. In chapter 3, we collected a comprehensive dataset for 45 species of horseshoe bats including body mass, forearm length, baculum size and habitat type by literature consulting and field sampling. We found that Rhinolophdae family exhibit predominantly female-biased SSD. The size of both sexes as well as SSD exhibits substantial differences among species. Variation in degree of dimorphism exhibited an allometric pattern. Nonetheless, we found no support for Rensch’s rule in family Rhinolophidae irrespective of the measure of SSD(body mass vs. forearm length) and computation method(conventional vs. phylogenetically informed). The fecundity selection on female size is stronger but the intensity and direction of sexual selective pressures acting upon males of each species were different due to horseshoe bats occupy the wide niches and have complex mating system, which may contribute to the inconformity of Rensch’s rule among horseshoe bats species. Parsimony ancestral reconstruction results showed median size and female-biased SSD as ancestral state(male = 48.37 mm; female = 48.88 mm; SSD = 1.009). There was a strong and significant relationship between baculum length and SSD, but the relationship was not significant when baculum length adjusted by male body size. The result indicated interspecific variation in SSD of horseshoe bats appears not to be associated with baculum size. Interestingly, we found that habitat type constitutes a strong predictor of SSD. Species inhabiting other habitats exhibit a higher mean size dimorphism than species inhabiting forest habitat. However, the effect of habitat type on SSD becomed non-significant when we considered the phylogeny history, suggesting that differences in SSD among horseshoe bats may attribute to their evolutionary history rather than an adaptation to different habitat use.3. In chapter 4, we have tested the allometry hypothesis regarding the evolution of echolocation in 12 horseshoe bats distributed in China by integrating echolocation call frequencies, phylogeny, forearm length, cochlea size, pinna size and nasal capsule size by multiple regressions. Our results showed no significant correlation between echolocation call resting frequency and forearm length across these species without considering the phylogeny of the group. However, ear length, ear width and nasal capsule size were significantly and negatively correlated with echolocation call frequencies among the studied species. These findings clarified that the allometric relationship between echolocation and morphological phenotypes. The fit between the linear regression model and the data was much better for the relationships between echolocation call frequencies and pinna or nasal capsule size than for the relationship between echolocation call frequencies and forearm length. These results illustrated that phenotypes mediating directly echolocation calls’ production or reception(pinna and nasal capsule size) were better predictors of echolocation call frequencies than forearm length. Therefore, selective pressures acting upon both echolocation calls and morphology, and promote changes in the phenotype with different evolution rate, which may drive the evolution of diversity of echolocation calls in horseshoe bats.In conclusion, in this dissertation, we found evidence to support Rensch’s rule at intraspecific level but no evidence to support Rensch’s rule at interspecific level. In addition, Phenotypes mediating directly echolocation calls’ production or reception were better predictors of echolocation call frequencies than forearm length. These works highlight the importance of allometry in studying the phenotypic evolution. The results will be helpful to uncover the mechanisms of signal divergence, speciation and the maintenance of biodiversity, and promoting species conservation in bats. |
PDF Full Text Request