Interactions Between the Salamander Siren Intermedia and the Keystone Predator Notophthalmus Viridescens

Abstract

Is the ability to function as a keystone predator a property of a species or an emergent property of the community? We discussed this question in a temporary—pond community where the broken—striped newt Notophthalmus viridescens dorsalis is known to act as a keystone predator on larval anurans. We independently manipulated the initial density of adult Notophthalmus (two or four) and the presence or absence of one adult Siren intermedia in a set of 20 artificial ponds to determine if this additional predator affected the ability of Notophthalmus to function as a keystone predator. Each pond received a diverse assemblage of larval anuran prey: a spring—breeding assemblage of 275 Rana utricularia, 100 Pseudacris crucifer and 25 Bufo americanus tadpoles, and a summer assemblage of 200 Hyla chrysoscelis and 150 Gastrophryne carolinensis tadpoles. Two additional pond received neither predator, to assay the outcome of competition among the tadpoles. The effect of the salamander Siren on Notophthalmus survival and fecundity depended on the density of Notophthalmus. At low newt density Siren had no effect on survival, but at high newt density competition reduced the survival and growth rates of Notophthalmus. The density of Notophthalmus and the presence of Siren interacted to determine the fecundity of Notophthalmus. At low newt density Siren reduced Notophthalmus reproductive success by preying on larvae. At high newt density Siren indirectly enhanced reproductive success by reducing survival of adults, thus releasing larvae from intraspecific competition and cannibalism. Notophthalmus density had no effect on adult Siren survival or growth rate in this experiment, but other evidence indicates that competition with Notophthalmus reduces the growth rates of Siren. Notophthalmus acted as a keystone predator on the assemblage of spring—breeding anurans; it increased the number of metamorphs of the weak competitor Pseudacris crucifer by releasing them from interspecific competition. Siren preyed on tadpoles in a nonselective manner. An additive model was sufficient to describe the effects of Siren and initial density of Notophthalmus on the structure of the tadpole assemblage. In our system of experimental ponds, the strong interaction between Siren and Notophthalmus density did not extend its effects to lower trophic levels, and the direct and indirect effects of Siren did not alter the role of Notophthalmus as a keystone predator.