Advantage of saliva analysis are the ease of sample collection and that samples can be collected more frequently with much less stress on the patient. The objective of the present study was to comparatively evaluate the concentrations of saliva and fasting serum glucose in both normal and diabetic subjects. The mean salivary glucose level in diabetic patients was $15.66{pm}17.1$ mg/dl and $1.78{pm}1.72$ mg/dl (P = 0.0006) in the control group. The mean fasting serum glucose level in diabetic patients was $202.12{pm}66.91$ mg/dl, while that in the control group was $94.21{pm}14.97$ mg/dl (P < 0.0001). The 0.95 degree of correlation between salivary and fasting serum glucose could be demonstrated. The concentration of salivary and fasting serum glucose was not significant different betweeen the measurements for male and female. In the oral glucose tolerance test (75g), the glucose concentration in saliva progressively increased during the first 30 minutes of the test and then progressively decreased, reaching at minutes 120 ~ 180 lowest point as like fasting serum glucose concentration. We can conclude that salivary glucose concentration was significantly higher in the diabetic subjects and that there was significant correlation between salivary and fasting serum glucose concentration. Measurement of salivary glucose could be a useful test having good correlation between salivary and fasting serum glucose concentration.