In intense exercise (>80% maximum oxygen uptake) a huge, up to 8-fold increase in glucose production (Ra) is tightly correlated to marked increases in plasma norepinephrine (NE) and epinephrine. Both Ra and glucose uptake (Rd) are enhanced, not reduced, during beta-adrenergic blockade in normal subjects. Beta-blockade also caused a greater fall in immunoreactive insulin (IRI) during exercise, which could, in turn, have increased Ra directly or via an increased glucagon/insulin ratio. To control for adrenergic effects on endogenous insulin secretion, we tested type 1 diabetic subjects (DM) made euglycemic by overnight i.v. insulin that was kept constant in rate during and after exercise. Their responses to postabsorptive cycle ergometer exercise at 85-87% maximum oxygen uptake for approximately 14 min were compared to those of similar male control (CP) subjects. Six DM and seven CP subjects received i.v. 150 microg/kg propranolol over 20 min, then 80 microg/kg x min from -30 min, during exercise and for 60 min during recovery. Plasma glucose increased from similar resting values to peaks of 6.8 mmol/L in DM and 6.5 mmol/L in CP, then returned to resting values in CP within 20 min, but in DM, remained higher than in CP from 8-60 min (P = 0.049). Ra rose rapidly until exhaustion, to 13.3 mg/kg x min in CP and 11.6 in DM (P = NS). Ra declined rapidly in recovery, although somewhat more slowly in DM (P = 0.013 from 2-15 min). The Rd increased to 10.6 in CP and 9.2 mg/kg x min in DM (P = NS), then declined similarly in early recovery, but remained higher in CP from 50-100 min (P = 0.05). The rises in plasma glucose during exercise in both groups were thus due to the increments in Rd less than those in Ra. The higher recovery glucose in DM was due to the slower decline in Ra and the lower Rd in later recovery. IRI was higher in DM than in CP before exercise (P = 0.011), and whereas it decreased in CP (P < 0.05), it increased approximately 2-fold in DM, thus being higher throughout exercise (P = 0.003). The glucagon/insulin ratio was unchanged in DM, but increased in CP during exercise (P = 0.002). NE showed a rapid, marked increment during exercise to peak values of 23.7 nmol/L in CP and 25.7 nmol/L in DM (P = NS), and epinephrine showed parallel responses. Both correlated significantly with the Ra responses. In summary, the Ra responses of both DM and CP during exercise were greater than those of control unblocked subjects (previously reported) despite higher IRI (all exogenous) in DM. This suggests an important contribution of direct alpha-adrenergic stimulation to this Ra effect.