Studies were done to characterize the bladder dysfunction associated with diabetes mellitus and to distinguish between changes occurring from increased diuresis and autonomic neuropathy. Four experimental conditions were compared: control, 4-wk-streptozocin-induced diabetes, sucrose feeding (diuretic), and galactose feeding (diuretic and sugar alcohol). A 10-fold increase in urine output and 25-50% increases in bladder weight, protein content, and DNA content were observed in all noncontrol treatment groups. Compliance properties were studied by measuring the intravesicular pressure as the bladder was infused with buffer in vitro. All treated bladders exhibited a reduction in plateau pressure and an increase in fluid capacity. Thus, diuresis results in an increased bladder size, which correlates with an alteration of compliance properties. Nervous system control in anesthetized rats was examined by monitoring contractions as the bladder was infused with buffer. Three distinct patterns of response were observed: normal, diabetic, and diuretic (galactose and sucrose treatments). The difference between responses in diuretic and diabetic animals suggests the presence of a diabetes-induced alteration in nerve regulation of the bladder. Reserpine pretreatment of control or diuretic models produced marked changes in the pattern of contractions, whereas pretreatment of diabetic rats had only modest effects. This suggests that diabetic bladders were lacking sympathetic control before the drug treatment. When rats treated for 4 wk with galactose were removed from this diet for 4 wk before testing, the bladders responded similarly to controls. This observation, coupled with the fact that galactose did not produce the same response as diabetes in the in vivo experiment, suggests that the galactose model does not produce irreversible functional neuropathies.