Sixteen German Shepherd Dogs from 4 litters were IgA-deficient on the basis of at least 1 of 2 serum IgA determinations, and all had small intestinal bacterial overgrowth, as documented by quantitated small intestinal bacterial culture in another study. These dogs were fed 2 diets that differed principally in their protein source (chicken vs beef, milk, and wheat). All dogs were fed each diet for 2 weeks before the study began. Next, all dogs were fed the chicken-based diet for 2 months. Then, half the dogs (group 1) were randomly assigned to continue eating the chicken-based diet, while the other half (group 2) ate a diet containing beef, milk, and wheat proteins. The small intestine was biopsied at the beginning of the study and after dogs had eaten the assigned diet for 2 and 4 months. At 2 months, group-2 dogs had more colonic mucosal mast cells, but this difference did not persist at 4 months. At the end of the study (ie, 4 months), although all dogs were clinically normal, group-2 dogs had significantly (P = 0.010) decreased numbers of jejunal villus plasma cells. However, these histologic changes were not considered clinically important. There were no significant differences in blood eosinophil counts, serum trypsin-like immunoreactivity, or cobalamin, folate, or IgA concentration. Clinical differences were not detected between the 2 groups, before or after the study. Changes were seen in serum IgM and IgG concentrations. Although results of this study suggest that dietary protein may influence intestinal mucosal cell populations, there was no evidence that the protein sources in these 2 diets caused intestinal disease in these dogs under the conditions of this study.