Galactose usually is ingested as lactose, which is composed of equimolar amounts of glucose and galactose. The contribution of galactose to the increase in glucose and insulin levels following ingestion of equimolar amounts of galactose and glucose, or lactose, has not been reported in people with non-insulin-dependent diabetes mellitus (NIDDM). Therefore, we studied the effects of galactose ingestion alone, as well as with glucose either independently or in the form of lactose, in subjects with untreated NIDDM. Eight male subjects with untreated NIDDM ingested 25 g glucose, 25 g galactose with or without 25 g glucose, or 50 g lactose as a breakfast meal in random sequence. They also received 50 g glucose on two occasions as a reference. Water only was given as a control meal. Plasma galactose, glucose, glucagon, α-amino nitrogen (AAN), nonesterified fatty acids (NEFA), and serum insulin and C-peptide concentrations were determined over a 5-hour period. The integrated area responses were quantified over the 5-hour period using the water control as a baseline. Following ingestion of 25 g galactose, the maximal increase in plasma galactose concentration was 1 mmol/L. The mean maximal increases in plasma galactose concentration following ingestion of 25 g galactose + 25 g glucose or following 50-g lactose meals were similar and were only 12% of that following ingestion of galactose alone (P < .05). The mean galactose area response over the water control for the 25-g galactose meal was 0.95 ± 0.31 mmol · h/L. That following ingestion of 25 g glucose + 25 g galactose or following the 50-g lactose meal was 0.17 ± 0.07 and 0.13 ± 0.05 mmol · h/L, respectively. Following ingestion of 25-g or 50-g glucose meals, the galactose area responses increased only slightly. The mean glucose area response following the 50-g glucose meals was 14.8 ± 2.5 mmol · h/L. Glucose area responses following ingestion of 25 g galactose, 25 g glucose, 25 g glucose + 25 g galactose, and 50 g lactose were 11%, 49%, 54%, and 60% of that observed following ingestion of 50 g glucose, respectively. The mean insulin area response following ingestion of the 50-g glucose meals was 965 ± 162 pmol · h/L. The insulin area responses observed with 25 g galactose, 25 g glucose, 25 g glucose + 25 g galactose, and 50 g lactose were 24%, 51%, 81%, and 85% of that observed with the 50-g glucose meals, respectively. The C-peptide data confirmed the insulin data. The glucagon concentration was unchanged after galactose ingestion and decreased after glucose ingestion. However, the decrease in the glucagon area response observed with 25 g galactose + 25 g glucose or 50 g lactose was less than that with ingestion of 25 g glucose alone. The latter suggests inhibition of the glucagon response to glucose by the added galactose. In conclusion, ingested galactose results in only a modest increase in plasma glucose concentration. The glucose area responses to galactose and glucose are additive. Oral galactose is a relatively potent insulin secretagogue, and the insulin response is also additive to that following glucose ingestion. Ingestion of glucose with galactose markedly reduces the increase in plasma galactose concentration. The mechanism of this effect remains to be defined.