Evaluation of the Effect of Ecological Factors on Cannibalism in Pelophylax bedriagae (Pallas, 1771) (Amphibia: Anura)

Marsh frog, Pelophylax bedriagae tadpoles habituate in temporary wetlands where intraspecific aggression has been observed. In this study, we examined the independent and interactive effects of predator cues, water level and density on head width size and head, tail and whole body cannibalism of larval P. bedriagae. We designed a 2×2×3 factorial experiment, by crossing two levels of predatory cues (presence/absence), two levels of density (low/ high) and three levels of water (low/ high/ decreasing) that were carried out within 29 weeks. Despite using conditions that might facilitate cannibalism (i.e., the presence of predator cues, low water level, and high density), a low rate of cannibalism was observed in all treatments. Independently, the highest percentage of total cannibalism was indicated in the presence of predator cues (2.26 %), high water level (2.04 %) and high density (2.24 %). Interactively, the highest rate of total cannibalism was recorded for larvae reared in the high density/absence of predator cues/low water level treatment (3.55%). In contrast, the lowest rate of total cannibalism was observed in the both the low density/presence of predator cues/decreasing water level and low-density/presence of predator cues/high water level treatments (0.44%). The results of the MANOVA indicated that predator cues, water level and density independently and interactively had not significant effect on rate of total cannibalism. At the end of experiment, the maximum size of head width was measured in low-density/absence of predator cues/high water level treatment (9.60±0.39mm), and there was no significant relationship between the maximum head width and the highest rate of cannibalism. According to the results of this experiment, it seems that P. bedriagae larvae may function to avoid the risk of cannibalism by conspecific.