The italian habit of having a low calorie breakfast suggests an insulin treatment made up of two daily injections of soluble insulin before lunch and dinner, adding at bed time an injection of retard insulin. Aim of this study was to verify whether Monotard or Ultratard proved more effective as retard insulin in this regimen. Sixteen insulin-dependent diabetics treated with the above-mentioned insulin regimen were studied for 16 weeks. They were divided in two groups: group 1 was treated initially with Ultratard and group 2 with Monotard, with a cross-over after 8 weeks. The metabolic control achieved with the treatment was considered good with either retard insulin used. HbA1c, fructosamine, body weight and glycemic values measured after 8 and 16 weeds failed to reveal any differences between the treatment with Monotard on Ultratard. Either Monotard or Ultratard can be successfully used as bedtime retard insulin in this multiple injection insulin treatment.