Many studies attempt to examine the causes of recruitment variation in fishes and relate these to biotic and abiotic factors. Hollowed [1992] lists 47 such studies along the coast of the northeast Pacific. There are two schools of thought regarding the utility of the correlative approach. It can be considered futile because of biases, measurement error, and the near certainty of spurious correlations [ Walters and Collie, 1988]. Others claim that these studies provide information on patterns that lead to testable hypothesis [ Kope and Botsford, 1990]. Recent studies [ Tyler, 1992; Hollowed, 1992] advocate the use of correlative studies with the constraint that the analysis be based on a sound conceptual framework and judicious use of statistical methods. This approach was applied to a recruitment time series for adult pollock for the period of 1962-1989 and egg and larval series for 1981-1989 [ Megrey et al., 1995]. The physical time series included estimates of precipitation, sea-level atmospheric pressure, wind-driven mixing, transport and water properties. Statistical techniques suggest that recruitment, as well as indices of age-0 and age-1 abundance, were related to precipitation, an index of sea level pressure, and wind mixing. Given the time period and phasing of the physical factors, the results imply that stratification can influence behavior of and predation on juveniles, and that increased baroclinicity influences larval survival through its impact on eddy generation and wind mixing. The former implication supports previous studies of juvenile behavior [ Olla and Davis, 1995], while the latter two have been inferred from analysis of biophysical observations.