Fishes were assessed in Guffin, Chaumount, and Black River bays in northeastern Lake Ontario with a 7.9-m (headrope) bottom trawl during late September and early October, 1978 to 1997. Fish density declined in the early 1990s with sharp declines in abundance of spottail shiner (Notropis hudsonius), trout-perch (Percopsis omiscomaycus), and johnny darter (Etheostoma nigrum) occurring in 1993 to 1995. Rising numbers of piscivores, walleye (Stizostedion vitreum) and double-crested cormorant (Phalacrocorax auritus), increased predation pressure, presumably acting in concert with oligotrophication to lower fish density, particularly after 1991 when large numbers of adult alewife (Alosa pseudoharengus) no longer migrated to the northeast basin in spring. Annual mortality of yellow perch (Perca flavescens) from age 2 to 5 rose from 33% in 1980–83 to 65% in 1992–95 and was positively related to piscivore numbers (P = 0.01, r = 0.96, n = 5). Annual mortality of yellow perch from age 0 to 2 also peaked in 1992–95. Abundance of yellow perch YOY in fall varied 40 fold and was not related to water warming in spring (P = 0.45, r = −0.19, n = 18) but was negatively related to the abundance of adult alewives in spring (P = 0.04, r = −0.49, n = 18). Although yellow perch produced moderate to strong year classes each year during 1991–95, stock size failed to increase because of rapidly accelerating mortality. Fully 85% of the variation in mean length of yellow perch YOY was explained by a multiple regression model which included YOY abundance, mean total phosphorus, and cumulative degree days > 13.5°C (P < 0.01, n = 15). Abundance of white perch (Morone americana) YOY varied nearly 200 fold and was not related to water warming or spring alewife abundance (P > 0.15). Variation in mean length of white perch YOY was related to cumulative degree days > 15°C (P < 0.01, r = 0.69).