1. Two experiments were carried out. The first with five normal male subjects was placebo controlled and single blind, each subject being studied on two occasions. Lower body subatmospheric pressure (LBSP) was used to assess the cardiovascular effects of graded hypovolaemia before and during either a hyperinsulinaemic, euglycaemic clamp or a placebo clamp using 0.9% (w/v) NaCl only. 2. During hyperinsulinaemia, resting systolic blood pressure rose and was accompanied by forearm vasodilatation. Forearm blood flow (FABF) and heart rate (HR) were higher at each level of LBSP during than before hyperinsulinaemia. In addition, hyperinsulinaemia was accompanied by a small increase in noradrenaline, but packed cell volume did not change. 3. In the second experiment, the effects of a hyperinsulinaemic euglycaemic clamp on the cardiovascular responses to LBSP were assessed in seven diabetic subjects with peripheral and autonomic neuropathy. 4. In contrast to the normal subjects, there was a slight fall in systolic blood pressure during the clamp but no effect was noted on HR or FABF. Mean arterial blood pressure was lower at each level of LBSP during hyperinsulinaemia compared with the pre-elamp period. Packed cell volume fell during the clamp and plasma noradrenaline rose. In one of the diabetic subjects, a precipitous fall in blood pressure occurred during hyperinsulinaemia when LBSP of 10 mmHg (1.3 kPa) was applied, this manoeuvre having been well tolerated before the clamp. 5. The mode of action of hyperinsulinaemia is not clear, but there was, however, no evidence that a fall in plasma volume had occurred.