lib

ph na xa ve Trinity Bay, Canada. Capelin larvae were abundant during all surveys except that of July 1985, when spawning was late as a result of cooler environmental conditions. Abundance of larvae was positively correlated with subsequent recruitment and inversely correlated with the time interval between northeasterly winds following spawning. Larval transport was mainly across (from northwest to southeast) and out of the bay, away from the spawning beaches. Larval transport itself was positively correlated with both the intensity of Ekman transport and a cumulative measure of wind-forcing in July, but it was negatively correlated with a measure of variability in wind speed and direction. The wind indices were related to measures of recruitment, but their slopes were opposite in sign to those observed between them and larval transport. There was no obvious relationship between larval transport and recruit-ment, i.e. transport of capelin larvae out of Trinity Bay was not a necessary requirement for successful recruitment. Our data are consistent with the hypothesis that wind-generated turbulence in the upper layers of the water column can modulate survival and recruitment