1. In a cross-sectional study, 32 insulin-treated diabetic patients had elevated low shear (27·1 vs 22·1 mPa s, P < 0·05) and high shear blood viscosity (5·4 vs 5·0 mPa s, P < 0·05) when compared with 10 non-diabetic controls. After correction to 45% packed cell volume, the abnormality had a tendency to be greater in patients with proliferative (mean low shear viscosity, 30·8 mPa s) than background (29·2 mPa s) or nil/minimal retinopathy (27·6 mPa s, 0·05 < P < 0·07, permutational trend test). 2. The fibrinogen levels were higher in the diabetic group ( P < 0·05) and correlated with the low shear blood viscosity ( r s = 0·38, P < 0·05). 3. In a prospective study, 74 insulin-treated diabetic patients with background retinopathy were randomized into two groups. Thirty-six patients were on attempted improved therapy (A group); in these the mean glycosylated haemoglobin (Hb A 1c ) fell within 1 year (11·6 to 10·1%, P < 0·001). Both the corrected low shear blood and plasma viscosity fell similarly ( P < 0·001). The fall was greater than in those patients who were kept on usual therapy (U group) and whose glycosylated haemoglobin did not change significantly (11·7 to 11·4%) over the year. 4. The effect of diabetes on blood viscosity may not be a direct pathological factor, as the same increased viscosity would be produced by a mean increase of 1·7% in packed cell volume, compared with a population range of 14% packed cell volume.

1. Hypoglycaemia induces glucose intolerance in normal subjects—the ‘Somogyi effect’–and may be responsible for some instances of ‘brittle diabetes’. This effect may be mediated through the growth hormone response to hypoglycaemia, but other possible hormonal mechanisms have not been excluded. 2. Paired 2-h oral glucose tolerance tests have been carried out 2 h after both (a) i.v. saline (control day) and (b) i.v. insulin (test day) in four normal subjects and twenty-seven patients with pituitary and adrenal under- and over-activity. Plasma glucose, insulin, Cortisol and growth hormone have been estimated at half-hourly intervals during the 4 h of each study. 3. A significant Somogyi effect is usually seen only when there is a growth hormone response to insulin-induced hypoglycaemia; hypopituitary subjects do not show the effect. 4. There is a correlation between the extent of the Somogyi effect and the growth hormone response to insulin, using a simple derived index to represent each function ( P < 0·05). 5. Insulin secretion in normal subjects tends to be higher on the test than the control day, making inhibition of insulin release an unlikely primary mechanism. 6. The presence or absence of a Somogyi effect could not be related to insulin-induced changes in plasma Cortisol values. 7. Examination of individual cases supports the view that, while growth hormone may be mainly reponsible for the Somogyi effect, yet it is sometimes difficult to explain the effect without invoking other endocrine mechanisms.