Factors organizing rocky intertidal communities of New England: Herbivory and predation in sheltered bays

In New England, U.S.A., shores exposed to severe wave action are dominated by the common blue mussel Mytilus edulis L. while moderately protected areas are covered with perennial algae. It is thought that algae are limited by mussels which are a superior competitor. Because the effectiveness of predators is inhibited by wave activity, it is assumed that the rate of predation, which varies across this environmental gradient, accounts for the observed distribution of mussels and algae.Shores along sheltered bays appear to be an exception to this pattern and this study addresses some of the possible causes. In New England bays, mussels and barnacles Semibalanus balanoides (L.) are the most common organisms on the solid surfaces in the lower intertidal zone. Perennial macroalgae, such as Chondrus crispas Stackhouse and Fucus vesiculosus L., are rare. The distribution and abundance of species differs from that on moderately protected shores and is similar to very exposed shores which are dominated by mussels and barnacles.Herbivory by the common periwinkle Littorina littorea (L.) limits the abundance of F. vesiculosus and indirectly affects the success of mussels. During 4 years of experimental manipulations, F. vesiculosus rarely recruited in the presence of periwinkles but dominated experimental surfaces if periwinkles were excluded. When experimental surfaces with F. vesiculosus, which had been protected from herbivory for > 1 year, were exposed to natural conditions, herbivores cleared most of the surfaces within several months. Recruitment by barnacles and mussels was higher when periwinkles were excluded. However, the effect of periwinkles on mussels was indirect; the snails reduced barnacle success and thus reduced mussel recruitment which was enhanced by the surface irregularities provided by barnacles.The occurrence of mussels in sheltered bays is not due to a lack of predators. Predators were commonly seen at all sites. Most mussels on experimental surfaces were removed <4 wk when surfaces were exposed to natural levels of predation. Experiments do not provide an explanation for the occurrence of mussels, although the enhancement of mussel recruitment by barnacles suggests that the availability of settlement sites may be important.