The high dose intravenous glucose tolerance test and concurrent immunoreactive serum insulin and glucagon levels were measured and the results related to the presence or absence of pancreatic insular amyloid in 16 cats, seven of which were known to be diabetic. Control values for all parameters were established using seven additional clinicopathologically normal cats. Nine of the 16 cats had normal fasting blood glucose levels (less than 120 mg/dl) and impaired glucose tolerance. These cats had attenuated (3/9) or normal (6/9) 0 to 5 minute glucose-stimulated insulin secretion, rising 45 to 60 minute insulin secretion (7/9), low mean insulin/glucose ratio, and normal mean serum glucagon. Three of the nine cats with impaired glucose tolerance had insular amyloidosis. These three cats had significantly higher mean blood glucose levels during the glucose tolerance test than did cats with impaired glucose tolerance and no insular amyloid deposits. Also, these three cats accounted for three of the four longest glucose disappearance one-half times (T 1/2 s), three of the four lowest glucose disappearance coefficients, and three of the four lowest 0 to 5 minute insulin responses. The seven diabetic cats (fasting blood glucose levels greater than 120 mg/dl) had either low to low normal (6/7) or above normal (1/7) fasting insulin levels, no insulin response to intravenous glucose stimulation (6/7), and elevated mean serum glucagon levels. Insular amyloid was present in six of the seven diabetic cats. Three diabetic cats with marked insular amyloid deposits had glucose disappearance T 1/2 and K (coefficient) values, serum insulin levels, serum glucagon levels, and insulin/glucose ratios which were not significantly different from the other three diabetic cats with slight to moderate insular amyloidosis. These results confirm a strong association between the occurrence, but not the extent of insular amyloidosis and diabetes mellitus in adult diabetic cats, although amyloid replacement of pancreatic islets does not appear to be the primary diabetogenic event. Rather, these results appear to be consistent with our hypothesis that insular amyloid deposition is a morphologic marker of primary B-cell dysfunction that is basic to the pathogenesis of the diabetic condition, and is reflected clinically by impaired glucose tolerance.