The effect of mammalian insulin on plasma glucose levels was studied in the paedogenic urodele amphibian, Necturus maculosus. Blood was collected from a cannula implanted in the dorsal aorta via the gastric artery and plasma glucose was estimated by the glucose oxidase method. Plasma glucose levels of untreated animals were 31.6 f 1.6 mg% at the completion of the operation and then declined to 14.1 +-2.5 mg% on the third postoperative day; the levels remained relatively stable from the third through the fifth day when sampling was terminated. Subsequent experiments began on the third post-operative day. "Normal" plasma glucose levels were judged to be too low to demonstrate a statistically significant insulin effect and a glucose tolerance study was performed to elevate plasma glucose. Plasma glucose levels were 9.3 2 1.6 mg% before glucose administration and were 68.8 k 4.2 mg% at 30 minutes after the infusion via the cannula of 25 mg glucose per 100 gm body weight. Plasma glucose levels declined 52% at 3 hours, 75% at 6 hours, and to approximately the pre-treatment levels at 24 hours after the infusion. The effect of insulin on the decay curve of infused glucose was determined. Control animals received glucose and the hormone-free carrier; experimental animals received glucose and 10 I U insulin per 100 gm body weight. Plasma glucose levels were not significantly lower than those of the controls at 30 minutes after the infusion, but were significantly lower at all subsequent sampling times (3, 6, 9, 12, 24, 48 hours) and were less than 3 mg% from 12 through 48 hours; no hypoglycemic symptoms were noted even in animals without detectable plasma glucose. The data indicate that plasma glucose is not a required source of energy in the mud puppy.