Abstract:The tails of tadpoles of anuran amphibians play an important role in anti-predator behavior, as they can confer powerful swimming abilities or become a lure for shifting a predator's attack from the body to the tail. We used Rana zhenhaiensis (Anura:Ranidae) tadpoles in the Gosner stages from G26 to G29 as a model system to evaluate the locomotor cost of tail loss in tadpoles. Two experimental groups were created based on the degree of tail-removal, defined as the ratio between the length of the tail removed and length of the tail left intact. The first experimental group, with slight tail loss (tail loss group 1) comprised manipulated tadpoles with less than 30% tail loss, and a mean value of 22% (standard deviation [SD]=0.06, N=30) tail loss; the other experimental group (tail loss group 2) comprised individuals with tail loss of more than 30%, and a mean value of 43% (SD=0.10, N=44) tail loss. The swimming speed of each individual was measured before and after tail removal. We used a digital video camera to record the swimming performance of each individual in a 50-cm straight lane. The video files were then examined using the software program Ulead Video Studio to determine the fastest swimming speed over a 10-cm interval. The control group, 16 individuals with intact tails, was measured for swimming speed to be compared to that of the tadpoles in the two experimental groups. We used repeated-measures ANOVA to analyze the swimming speed data, which indicated that tail loss affected the swimming speed of R. zhenhaiensis tadpoles, but only in some individuals with more than 30% tail loss. These results suggested that slight tail injury may not severely affect the swimming speed of R. zhenhaiensis tadpoles. Swimming speed positively correlated with tail length in individuals, both before and after tail-removal, and the swimming speed was significantly greater after tail loss. Therefore, we speculate that the swimming speeds of tadpoles with tail injuries may be partly compensated by increases in the frequency of tail and body twisting. Tail loss occurred more frequently in the distal portion of the tail tip in R. zhenhaiensis tadpoles collected from the field, and 47% of captured tadpoles had different levels of tail injury; moreover, we used the G-test to analyze different levels of tail injury in wild population of R. zhenhaiensis tadpoles, and found significant differences between our experimental groups with two different levels of tail injury. More than 94% of tadpoles have slight tail injuries, but not different among different developmental stages in the natural ponds. Thus, we conclude that R. zhenhaiensis tadpoles with tail injuries did not incur substantial locomotor costs in the wild.