Reference: Biol. Bull. 165: 330-341. (August, 1983) GEOGRAPHIC LIMITS AND LOCAL ZONATION: THE BARNACLES SEMIBALANUS (BALANUS) AND CHTHAMALUS IN NEW ENGLAND DAVID S. WETHEY DepartmentofBiologyandMarineScienceProgram, University of South Carolina, Columbia, SC 29208 ABSTRACT The interactions between the intertidal barnacles Semibalanus (Balanus) bal- anoides and Chthamalus fragilis were examined in order to determine whether the factors which influence local zonation in the intertidal also contribute to the estab- lishment of geographic limits. Both physical and biotic factors influence intertidal zonation at the northern limit of Chthamalus in New England. On sloping surfaces Semibalanus died at all shore levels higher than mid tide level, apparently as a result of desiccation associated with high summer temperatures. Chthamalus settlement occurred at all shore levels above mean tide level, and postsettlement mortality apparently restricts Chthamalus to high shore locations where Semibalanus growth and survival is inhibited. North of the northern limit of Chthamalus, Semibalanus does not suffer summer heat death, so it occupies the zone where Chthamalus would have a refuge from com- petition further south. The northern limit of Chthamalus is set not by factors directly related to cold acting on Chthamalus. Rather the northern limit appears to be set by cold which allows the dominant competitor to exclude Chthamalus from its refuge zone. South of the northern limit the competitor, Semibalanus, is excluded from the high shore by high summer temperatures. INTRODUCTION One of the goals of ecology is to determine the mechanisms responsible for the patterns of distribution and abundance of organisms. The rocky intertidal zone has been used very successfully to make experimental tests of a wide variety of hy- potheses about the organization of communities and the dynamics of assemblages of species. Much of this work has been designed to elucidate patterns of local dis- tribution and abundance, rather than large scale geographic patterns. Here I examine whether the same mechanisms that control local zonation are responsible for large scale geographic patterns, those of geographic limits of species. The strong physical gradient in the intertidal zone was long considered to be fully responsible for the zonation patterns observed. Upper and lower limits of distribution were thought to be set by physiological tolerances (Colman, 1933; He- watt, 1937; Doty, 1946). Upper limits on the shore are now known to be generally determined by physical factors. Foster (1969) and Hatton (1938) demonstrated that barnacles die if transplanted above their usual shore zone, and that both heat and moisture influence the rate of death. There is little field evidence that intolerance of submersion sets the lower limit of marine species in the intertidal zone. The Received 6 January 1983; accepted 23 May 1983. 330