Rana cascadae tadpoles aggregate with siblings: an experimental field study.

Previous laboratory studies have shown that Rana cascadae larvae preferentially associated with siblings over non-siblings in choice tests. This study, conducted during three consecutive summers, tests the hypothesis that R. cascadae larvae aggregate nonrandomly with respect to sibship in natural ponds. Pairs of sibships were reared in separate tanks or together in the same tank in the laboratory. Each sibship within a pair was then stained with neutral red or methylene blue dye and released together in a natural pond. Over a period of several days, aggregations of tadpoles within test ponds were repeatedly captured, censused for sibship composition, and released. In control tests, two groups of tadpoles from the same sibship were dyed different colors and released together. In total, 25 different tests were conducted using tadpoles from 31 sibships and 456 aggregations were sampled. The distribution of color compositions of aggregations in control tests did not differ from an expected random distribution. Color compositions of aggregations in experimental tests differed from controls and from an expected random distribution. Aggregations in these tests tended to be dominated by one of the two colors (sibships). We conclude that R. cascadae tadpoles recognize and prefer to aggregate with siblings in natural field conditions. Circumstances of early rearing (i.e., whether tadpoles were reared with siblings or in mixed sibling/non-sibling groups) had no influence on preferences to associate with siblings, but there was an inverse correlation between group size and sibling association.