1-month-old Sprague-Dawley rats were fed synthetic diets containing 55% sucrose (SU) or starch (ST) as the sole source of carbohydrate for 2, 3 or 8 months. Both the ST and SU fed rats gained weight normally, but SU fed rats had enlarged kidneys. A higher yield of glomerular basement membrane (GBM) was recovered from 9-month-old SU rats. An increase in the hydroxylated amino acid content was found in GBM prepared from SU fed rats and the glycine content was also higher. The increase in the hydroxylation of lysine was accompanied by increased glycosylation and there was 30% more Glc-Gal-Hyl present in GBM from 9-month-old SU rats. GBM was solubilised with sodium dodecyl sulphate (SDS) and 2-mercaptoethanol and subjected to electrophoresis on 5% polyacrylamide gels. There was an apparent fall in the intensity of the bands with molecular weights greater than 200,000 and a concomitant rise in low molecular weight components (50,000–100,000) in GBM from 4-month-old SU rats. These differences between ST and SU membrane were accentuated when the membranes from 9-month-old rats were compared. No significant differences were found in the glucosyl transferase activities of renal cortical homogenates prepared from 3-month-old SU and ST rats, but the activities in SU rats were significantly higher at 4 and 9 months. The feeding of SU-rich diets to rats induces a number of biochemical changes in the kidney which are similar to those found in diabetes. The feeding of SU diets provides a useful animal model with which to study the effect of dietary carbohydrate on renal GBM. SU should not be included in diets fed to diabetic rats because of the similarity of some of its effects and those seen in chemically induced diabetes.