Objective To investigate the effects of regular insulin (RI)on duodenal smooth muscle in diabetic mice. Methods Diabetes mellitus (DM) model was established by intraperitoneal injection of 150 mg/kg streptozotocin (STZ) in male BALB/c mice. The model mice were divided into DM group and DM treated with RI group with 6 each. Meanwhile, 6 normal mice were served as controls. The mice in treatment group were intraperitoneally injected with 40 U/kg of RI daily.Whereas the mice in DM and control groups were intraperitoneally injected with phosphate buffer solution (pH = 7. 40). After 6 weeks, the small intestinal transit rate of mice was determined by lavage of Indian ink. Interstitial cells of cajal (ICC) in duodenal myenteric plexus were counted using immunohistochemical staining. Slow waves of duodenal smooth muscle cells were recorded with intracellular recordings. Data were analysed by SPSS 17.0 software, and comparisons among three groups were done using LSD test. Results After intervention for 6 months, the clinical presentations,such as more water and food intake and polyuria, were improved in treatment group. The body weight was increased in treatment group [(23.33±3.13) g] compared with DM group [(15.42±1.40) g,P＜0.01] ,but dereased compared with control group [(26.78 ± 2.09) g, P＜0.05]. The level of blood glucose in DM group was significantly higher than that in control and treatment groups(P＜0.01). Small intestine transmission rate was significantly reduced in DM group than that in control and treatment groups (P＜0.01), but it was slower in treatment group than that in control group (P＜ 0. 01 ). Immunohistochemical study showed that the number of c-kit positive cells reduced obviously in DM group than that in control group and treatment group (P＜0.05), whereas it was lower in treatment group than that in control group (P ＜ 0.05). The slow wave frequency and amplitude of duodenal smooth muscle cells in DM group were reduced when compared with control and treatment groups (P＜0.01) and both were lower in treatment group than that in control group (P＜0. 01 ). Conclusion The findings indicate that DM mice have gastrointestinal dysmotility and exogenous insulin may improve small intestinal dysmotility in DM mice.