Based on data from our annual lakewide bottom trawl survey of Lake Michigan, we determined that density of ninespine sticklebacks Pungitius pungitius increased from an average of 0.234 kg/ha during 1973–1995 to an average of 1.318 kg/ha during 1996–2007. This greater-than-fivefold increase in density coincided with the dreissenid mussel invasion of Lake Michigan. Intervention analysis revealed that ninespine stickleback density in Lake Michigan significantly increased between the two time periods. In contrast, based on data from our annual bottom trawl survey of U.S. waters of Lake Superior, ninespine stickleback density decreased from an average of 0.133 kg/ha during 1978–1999 to an average of only 0.026 kg/ha during 2000–2007. This greater-than-fivefold density decrease, which was found to be significant via intervention analysis, coincided with population recovery for both lean and fat morphotypes of lake trout Salvelinus namaycush in Lake Superior. In contrast to Lake Michigan, dreissenid mussels have not invaded Lake Superior on a lakewide basis. Thus, a comparison of these two lakes indicated that the increase in ninespine stickleback abundance in Lake Michigan was most likely attributable to the dreissenid mussel invasion. In addition, based on our correlation analysis, alewives Alosa pseudoharengus did not have an adverse effect on ninespine stickleback abundance in Lake Michigan. Perhaps the recent increase in biomass of green algae Cladophora spp. associated with the dreissenid mussel invasion improved spawning habitat quality for ninespine sticklebacks and led to their stepwise abundance increase in Lake Michigan beginning in 1996