AND
Marine Biological Laboratoyv, Woods Hole, Massachuserts 02543 U S AAbsrruct. Because of the narure of their substratum, the sessile invertebrate species of the marine epifaunal comrnunity living on rocks occur in discrete patterns of distribution. The rocks are finite patches or habitat islands with a limited space for colonization and growth. Such a system is ideal for studying the parameters affecting the distribution of species within a comrriunity. Also, because of the small size and immobility of the adults, the system is also ideal for studying the pattern of change in species composition and diversity within a community. This study used multiple series of manipulated experimental plates, which both duplicated natural rock surfaces and could be compared with sanlples of the rocks, to investigate the developmental and distributional processes of this community.Five major factors were found to be important to both the development of the comrnunity and its distribution on the rocks: ( I ) the selecriviry of the metamorphosing larvae as to site of attachment; ( 2 ) the seawnal fluctuation in larval abundances; (3) the biological interactions within and between species; ( 4 ) the size of rock substrata; and ( 5 ) the physical disturbance of the substrata (rock turnover). Initially, tne developmental process can be uncoupled fiom the effects of substrate size and disturbance. Predation is relatively unimportant as a biological interaction within this conrrnunity, but the species can be ranked according to their ability to compere for the availablespace on a substratum. This ranking implies a type of successional sequence in the development of the community: however, the sequence is greatly affected by historical components. The colonization of a substratum is directly dependent upon the abundance of settling larvae, which in turn is a function of seasonality and selectivity. The eventual competitive outcome and development of the community will depend upon which species have immigrated onro the substratum and is thus dependent upon history. The process is, therefore, open ended: colonization will be highly variable and change seasonally and, although one species may eventually dominate the substratum, it may be one of nine different species depending upon the individual history of that area.The frequency with which a substrarum is disturbed (with the resultant extinction of its fauna) is a function of wave force and is inversely proportional to both the size of the substratum and the depth at which it occurs. Disturbance will determine when a substratum is initially exposed for colonization and how long it will have for development. In the shallow subtidal (mean low water to -2.5 m ) , the frequent disturbance of small rocks will came them to support less than their equilibrium number of species and their fauna will reflect immediate larval abundances. Large rocks will remain stable for long periods of time and will usually be dominated by a single species. Intermediate-sized rocks (1 to 10 dm") will remain ...