Granule color may be an important factor influencing avian exposure to granular pesticides when they are mistakenly picked up as grit, and is clone of the most easily altered characteristics of pesticide granules. We studied colored grit use by house sparrows (Passer domesticus) and northern bobwhites (Colinus virginianus) by offering captive birds a grit mixture consisting of equal amounts of 8 colors (red, brown, yellow, green, blue, black, white, clear), either on a light-brown or a dark-brown soil background. After 7 days, gizzards of both species consistently contained more yellow and white particles and fewer black and blue particles than expected by chance (sparrows: P < 0.025; bobwhites: P < 0.001). Soil background color was not an important influence on grit color selection by either species. To examine the influence of food color on grit color selection, we repeated the experiments (on dark soil only), using birds maintained on food dyed to match 3 of the 8 grit colors (red, yellow, blue). Grit color use again was nonrandom (P < 0.001). Overall, house sparrows preferred brown, yellow, and white grit, and northern bobwhites preferred yellow and green house sparrows preferred brown, yellow, and white grit, and northern bobwhites preferred yellow and green grit. Black again received little use by both species. Food color did not influence grit color selection by house sparrows, but was associated with small but statistically significant differences in the use of black and clear grit by northern bobwhites. Our results suggest that the colors black andd blue should be tested further if the goal is to design pesticide granules in colors unattractive to birds. On the other hand, if the goal is to deter avian consumption of pesticide-treated granules by applying a mixture of unattractively-colores, treated particles and attractively-colored, untreated ones, then yellow (and perhaps white and green) should receive further study as colors likely to attract birds.