Interspecific Associations in a Woodland Stream

Publisher Website
Google Scholar
Abstract
Patterns of association of benthic macroinvertebrates were examined within a riffle of New Hope Creek, North Carolina, USA. The 16 most common species [Brachyptera fasciata, Ephemerella deficiens, E. doris, Isonychia bicolor, Stenonema annexum, S. rubrum, Cheumatopsyche sp., Chimarra atterima, Hydropysche sp., Simulium sp., Empididae, Allocapnia sp., Taeniopteryx maura, Tipula abdominalis, Acroneuria abnormis and Isoperla sp.] were tested for co-occurrences. Chi-square contingency tests on simultaneous presence and absence of 2 spp. at a time were performed. The strength of the association and its direction were determined for significantly associated species pairs using the V statistic. The patterns of association among 120 spp. pairs are presented for each of 3 mineral substrate types (cobbles, pebbles and gravel) and leaf packs (5 g of Cornus florida leaves) attached to them. These data were collected over a 3-mo. period. The results show that significant interspecific association on a given substrate was found for 18.3% or less of the species pairs. The species which strongly associated on 1 substrate type differed from those associationg on another substrate. The presence or absence of leaf packs on a mineral substrate patch, which did not affect most species densities, did affect the co-occurrence patterns among species. The presence of leaf packs reduced the number of significantly associated species pairs on all substrates. In all cases in mineral substrates, negative associates were rare. The proportion of negative associations approximated what one would predict by chance alone. This suggests that competition may not be an important factor in stream communities. A different approach to the analysis of lotic communities is presented and several hypotheses about community structure in streams are suggested. Substratum type is a major governing factor in stream community structure and dynamics.