Patchiness in the distribution of organisms in space and time is a fundamental characteristic of most natural populations and communities. The dynamics of individual patches are determined by the interaction of a variety of biotic and abiotic factors. However, given the complexity and large-scale spatial and temporal patchiness of most natural systems, it is often difficult to examine multiple processes in a single study. In this respect, intertidal communities on rocky shores are a particularly tractable system because they are relatively simple and the component organisms are generally small, and many are sessile or sedentary.

Mussels are common on rocky shores in aggregations that range in size from patches a few centimeters in diameter to large beds containing meter-scale gaps. My thesis examines mussel (Mytilus trossulus, M. edulis) assemblages in two intertidal habitats, tidepools and the surrounding emergent rock, on a wave-exposed rocky shore near Halifax, Nova Scotia, Canada. Mussels in tidepools generally occur in centimeter-scale patches, whereas those on emergent rock form more extensive beds with centimeter-scale gaps. Tidepools are potentially important refuges from stressful environmental conditions on emergent rock, which may explain differences in the structure and dynamics of species assemblages between the two habitats. In these two habitats, I examined both the formation of mussel aggregations through colonization by settling larvae and larger postlarval individuals and the structure and dynamics of established mussel assemblages.

More than 96% of mussels colonizing the natural substratum were too large to be settling larvae, indicating the importance of post-settlement dispersal of mussels. Patterns of colonization after 5-16 mo reflected patterns measured at sampling intervals of 2-7 d, suggesting that spatial variation in initial colonization is important in determining the abundance and distribution of mussels. I examined the relative roles of supply of colonists and substratum type in determining colonization patterns by comparing colonization rates on artificial collectors and natural substrata. Colonization rate on natural substrata was related to a suite of biological (macroalgal and barnacle cover) and physical factors (water flux, tidal height).

I examined the effects of predation by the whelk Nucella lapillus on established mussel assemblages by manipulating the densities of whelk recruits and post-recruits. Reduction of the density of whelk post-recruits influenced the percentage cover and size distribution of mussels and had a greater effect on emergent rock than in tidepools, reflecting between-habitat differences in density of whelk post-recruits. I investigated the influence of rates of recruitment, immigration, mortality, and growth by transplanting artificially constructed mussel patches to both habitats. In addition, I monitored the movement of tagged mussels and estimated the probabilities of wave dislodgment of mussels from measurements of water velocity and attachment strength. The structure and dynamics of mussel patches were influenced by all of these processes. However, physical processes (wave dislodgment) were more important than biological processes (predation) in determining the structure and dynamics of mussel assemblages on this shore.

Over the time scale of these studies, the rates and patterns of colonization, growth, mortality, and wave dislodgment of mussels could not account for the observed differences in distribution and abundance between tidepools and emergent rock. Patterns of distribution and abundance of mussels develop slowly on this shore due to the slow growth rate of individuals. The results of this thesis extend our understanding of the structure and dynamics of mussel assemblages by demonstrating the importance of wave disturbance as a cause of redistribution as well as loss of mussels. Dislodgment and reattachment of large mussels have not been recognized as a major determinant of the structure and dynamics of mussel assemblages on hard substrata.