Grazers affect prey abundance negatively through grazing and positively through releasing nutrients and substrates. In the present study, the relative importance of these positive and negative effects of grazers on bacterial abundance was examined seasonally at 2.5 m and 30 m depths in Lake Biwa. We used a method by Elser and Goldman to estimate in situ growth rate, grazing rate, and the fraction of growth rate that was stimulated by grazers smaller than 100 m. At 2.5 m, the bacterial grazing rate showed clear seasonal changes with higher values in summer. At 30 m, bacterial grazing was significant but consistently low throughout the year. The fraction of the growth rate stimulated by grazers was changed with season and depth. At 2.5 m, bacteria compensated 30-40% of grazing loss by using deficient nutrients released by the grazers from May to October. During that period, bacterial growth rate was severely limited by phosphorus, and it is most likely that phosphorus release by grazers was stimulating bacteria. At 30 m, however, the bacterial growth rate was not stimulated by the grazers in the majority of the cases. Bacterial growth rate was limited by low temperature at this depth, so nutrient return from grazers does not have a beneficial effect on bacterial growth. These results indicate that the relative importance of positive and negative effects of grazers on bacterial abundance differs seasonally and vertically, and that positive effect becomes important when bacterial growth is limited by nutrients. In a given habitat, bacterial abundance changes less than one order of magnitude in spite of high growth rates. The present study suggests that grazers contribute to numerical stability of bacterial assemblages not only through grazing but also by regenerating deficient nutrients.