Species Packing in Desmognathus Salamanders: Experimental Demonstration of Predation and Competition

Five years ago, experiments were proposed to determine whether competition or predation is the principal ecological factor in the organization of the community of Desmognathus salamanders. The experiments were conducted in the Nantahala Mountains of North Carolina over a four-year period, 1981-1984. There were four species present: D. quadramaculatus, the largest and most aquatic; D. monticola, the next largest, a stream-bank species; D. ochrophaeus, smaller and largely terrestrial; and D. aeneus, the smallest and most terrestrial of the four. A block design of replicate plots along streams was used. Removing a smaller species (D. ochrophaeus) was expected to cause an increase in the D. monticola or D. quadramaculatus populations only if competition were the important interaction. Removing D. monticola individuals was expected to increase the abundance of D. ochrophaeus, whether competition or predation was important, but it would increase D. quadramaculatus numbers only in the case of competition. The outcome largely supported the hypothesis that predation is the important organizing force. When D. ochrophaeus individuals were removed, the abundance of both larger species decreased significantly, relative to their abundance on control plots. Removing D. monticola salamanders led to the expected increase in the abundance of D. ochrophaeus, thus completing the predictions of the predation hypothesis. There was a small but statistically significant increase in the numbers of D. quadramaculatus, a result that confirmed the existence of competition. I conclude that D. monticola and D. quadramaculatus compete for food and that D. ochrophaeus is an important item in their diets. The importance of experimentation and prior publication of research designs is stressed.