80 subjects with liver cirrhosis selected for this study were divided into 2 groups, 24 cases were considered to have combinated with primary diabetes and the remaining 56 cases were not. The latter was classified into groups of probable diabetics, chemical diabetics, clinical diabetics, and non-diabetics according to the diagnosis from GTT. The results of tolbutamide test obtained were compared not only with each group with and without primary diabetes, but also with pathologic stage of the liver. The results were as tfollow: 1. Of 56 cases of liver cirrhosis wihout primary diabetes, 12 cases (21.5% ) revealed the diabetic curve in the tolbutamide test, and remaining 33 cases (58.9% ) were normal, on the other hand, of 24 cases combinated with primary diabetes, 21 cases (87.5% ), 3 cases (12.5% ), and no cases, respectively. 2. Observing the incidence of the abnormality of TTT in the subjects with liver cirrhosis from the view of the diagnosis of GTT, 11.8% of 18 cases with non-diabetic GTT, 47.0% of 17 cases with probable GTT, 61.9% of 21 cases with diabetic GTT, showed no abnormality, while, 24 cases combinated with primary diabetes were all found to be abnormal. Although the incidences described were high by about 10% than that found in each group of primary diabetics without liver cirrhosis, the frequency of the cases with delayed recovery in each group of liver cirrhosis was high by about 10% than those of primary diabetics. 3. Although the mean blood glucose curve following tolbutamide infusion in the subjects with liver cirrhosis without combination of primary diabetes situated between those of normal controls and of clinical diabetics, however, no significant difference could be observed in the mean blood glucose curve comparing the groups of liver cirrhosis with non-diabetic, probable, chemical diabetic GTT, and of combination with primary diabetes with those without liver cirrhosis. 4. Not only initial fall and subsequent rise in the blood glucose level, but also the duration requiring for reaching the minimum value following tolbutamide infusion were similarly delayed in both subjects with liver cirrhosis and primary diabetics without liver cirrhosis, and these delays became greater proportional to the grade of the impairment of their GTT. 5. It was impossible to find any significant correlation between the results of TTT and the duration of liver disease, the results of liver function test, or the histological changes in the liver, however, the frequency of impairment of TTT was greater in the subjects with ascites as compaired to that without it. 6. The incidence of abnormality of tolbutamide test (TTT) in the subjects with impaired prednisolone-primed glucose tolerance was significantly greater than that of the cases without impairment of the test.7. No significant correlation was observed between the results of TTT and of secretin test. 8. The response of immunoreactive insulin following tolbutamide infusion in the subjects with liver cirrhosis was higher than those of diabetics or normal subjects. 9. It was concluded from these results that the abnormal glucose tolerance found in the subjects with liver cirrhosis could not be distinguished by using the TTT from that derived from pancreatic factors.