Abstract
Perch Perca fluviatilis recruitment (based on the catch per unit effort, CPUE, of 2 or 3 year male perch) varied greatly from 1941 to 1995 in Windermere, U.K., with year class strengths strongly synchronous between the two basins of this temperate, mesotrophic lake. Statistically significant modified Ricker (Saila‐Lorda) stock‐recruitment curves could be fitted to data from three of four sampling sites, while the presence of widespread disease and late summer water temperature were also important explanatory variables. Multiple‐regression analyses revealed no influences of pike Esox lucius recruitment, zooplankton abundance, or the climatic influences ( i.e . North Atlantic Oscillation and displacement of the summer position of the Gulf Stream) on perch recruitment over and above associated influences from temperature. Fitted models realistically explained up to c . 60% of the observed variation in perch recruitment.