This thesis shows that there is a latitudinal gradient in the structure of rocky intertidal communities along the Pacific coast of North America. Specifically, a marked transition in percentage cover and recruitment of sessile invertebrates occurs near Cape Blanco in southern Oregon. This transition coincides with a shift in oceanographic regime: upwelling is strong and persistent from Cape Blanco south to Point Conception, California, and it is weak and intermittent north of Cape Blanco. Mathematical models linking larval transport during upwelling with benthic species interactions confirm that these oceanographic differences should produce the differences in abundances that exist between California and Oregon. The models also confirm that small-scale species removal experiments should have larger effects where upwelling is weak than where it is strong. This explains the tendency for species removals to have larger effects in Washington and Oregon than in central California. Furthermore, the models indicate that differences in upwelling produce population-level differences in the strength of species interactions that mirror the differences in local experimental effects. The latter results are obtained by developing a method of approximating process-based ecological models with classical Lotka-Volterra models. Such approximations make it possible to translate models tailored for particular ecological systems into the language of general ecological theory, and they explicitly characterize the functional dependency of community properties, such as interaction strength, on ecosystem properties, such as the oceanographic processes that transport larvae. These findings put geographical differences in experimental results into an oceanographic context, and they offer a resolution to a longstanding controversy about the generality of strong species interactions in intertidal communities.

While gradients in recruitment within California were small, relative to the transition that occurred near Cape Blanco, nonsignificant latitudinal trends within central and northern California suggest that previous studies in this region suggesting both increasing and decreasing recruitment with latitude can be synthesized. Specifically, correlations of recruitment with latitude indicate that recruitment rates are higher between San Francisco and Monterey Bay than they are either north or south of this region. This suggests that mesoscale transport processes are relatively more important determinants of recruitment patterns than overall latitudinal trends in upwelling within central and northern California.

During the course of these studies, large increases in recruitment of intertidal barnacles were observed during the onset of the 1997-98 El Nino. Across nine sites spanning five degrees of latitude in central and northern California, recruitment rates were significantly higher during May-September 1997 than during the corresponding months of the previous year. These increases were unusually large at several sites. Onshore and poleward wind anomalies coincided with these increases, indicating that increased onshore transport of larvae was a likely cause of the high recruitment.