8073 - Laserfiche WebLink

Home Browse Search

... 1't1i 001-- PETER S. PETRAITIS Ecology, Vol. 60, No.4 ",= -L NlH"/L nu. 1 u leasure is: R. = (H'm.. - H'..J/(H'm.. - H'm'.)' i 1!~ n"J substitute the appropriate summation for / jj' It Then [L n,,(\og CJ - log suJ 1/ [L nu(\og Pu - log suJ 1. ing, (I - Rol L nulog Su + R. L nulog Pu = L nulog CJ. (I) Subtract ~ nulog Pu from both sides, right hand side of Eq. I is log ')I; (I - Rol L nu(log Su + log P,J = log y. (2) Divide by ~ nu and note right hand side of Eq. 2 becomes log G. Thus (1- Rol ( L sulog Pu - H'm.J = log G. Ecology, 60(4), 1979, pp. 711_7l8 @ 1979 by the E~loSical Society of America ~ol~ COMMUNITY ORGANIZATION IN FISHES AS INDICATED BY MORPHOLOGICAL FEATURES! A. JOHN GATZ, JR.' Department of Zoology. Duke Universi!y, Durham. North Carolina 27706 USA Abstract. This study uses morphological data telating to many aspects of the niche to study community organization in assemblages of stream fishes from three separate drainages. Mean values for all morphological features were compared statistically for coeKisting species pairs, and those features which had significantly different means were assumed to indicate ecological differentiation that exists between the species under consideration. Average niche overlap as defined by the per- centage of morphological features not differing significantly between species pairs was constant for these assemblages regardless of a nearly twofold difference in number of species present. Similarly, the average Euclidean distance between the centers of morphologically defined niches for all sympatric common species was approximately the same for all three assemblages. Moreover, the distribution of Euclidean distances between species was the same in all three drainages and different than a distribution that would result from a random division of resources. Together, !hese results indicate that consistent patterns do exist in the organization of these stream communities, and that the species currently living together are not a random assortment of species and do not randomly divide their resources. The presence of higher numbers of specie' in a community seems no! to be accompanied by increased packing of the niches, but rather by an occupation of more total niche space by the community as a whole. Key words: communities; competition; eco-morph%gy; niches; organization; packing; parti- tioning. INTRODUCTION "Natural communities are products of evolution, and evolution may have produced nonrandom assem- blages of interacting species. . ." (Krebs 1978:505). The purpose of this research is to examine community organization in several communities of fishes living in freshwater streams. and to determine whether or not there is evidence that evolution has indeed produced "nonrandom assemblages of interacting species." In order to accomplish this purpose, operationally defined niches for species are essential. Although Hutchinson's (1957) originally defined N-dimensional hypervolume niche may be too abstract to be opera- tionally useful. as Pianka (1974: 197) suggests. this same basic concept with a slight modification can work. The modification, which was suggested by Hutchinson (1968: 177) in his discussion of the con- version of taxonomic space to niche space, is to use the morphological characteristics of a species to define a hypervolume. An implicit assumption underlies such usage of mor- phologically defined niches. The assumption is that the utilization of resources is constrained by, or at least correlated with, the phenotypes of the species. This seems probable for the highly diverse fish fauna under study. Substantial documentation (e.g.. Nilsson 1955. 1960,1963,1%5,1%7. Andrusak and Northcote 1971. Schutz and Northcote 1972, Griffith 1974, Werner and I Manuscript received II August 1977: accepted 30 No- vember 1978. 'Present address: Department of Zoology. Ohio Wesleyan University, Delaware. Ohio 43015 USA. Hall 1976, among others) indicates that in mixed species situations where interspecific competition pre- dominates, fishes specialize in their habits according !o their morphological adaptations; these same species, however, may adopt generalized habits in monospecies cultures if intraspecific competition is severe. MATERIALS AND METHODS Fishes and morphology Assemblages of species in three streams in the Pied- mont of North Carolina were used: East Prong Little Yadkin (Pee Dee drainage), Mud Creek (Cape Fear drainage), and Maho Creek (Roanoke drainage). Fish- es were collected using straight seines 1.83 by 1.22 m or 3.05 by 1.22 m in size, and preserved immediately after capture in 10% formalin. Six to nine collections were made throughout all seasons at each of four to six stations along every stream. All species collected more than once over the 2 yr of sampling (August 1972-August 1974) were considered. Morphological features were studied on a sample of 10 individuals of average adult size in these streams. Animals collected at as many different seasons of the year and as many different collecting stations as possible were utilized in order to avoid local or temporal bias in the results due to my use of a fairly small sample size. The use of only older age classes was a simplification because feeding habits and other ecological interactions un- doubtedly change with age. Moreover, it simplified the morphological analysis because the individuals used were already beyond the period of strongest allometric I Iii! !I, :'