Objectives. To investigate if low birth weight as a consequence of intrauterine malnutrition is a risk factor for the later development of diabetic nephropathy. Design and subjects. In a case-control set-up a group of type 1 diabetic subjects with diabetic nephropathy (n = 51) and a matched control group with normal kidney function (n = 51) were compared. Diabetic nephropathy and normal kidney function were defined as urinary albumin excretion rate above 200 μg min-1 and below 20 μg min-1, respectively. The birth weights were all obtained from the midwives' original records. Setting. The patients were identified from a population-based study of chronic diabetic complications in the Funen County, Denmark. Main outcomes. Birth weights according to the presence of diabetic nephropathy. Results. The median (10-90 percentile) birth weights were 3600 g (2960-4274) in the group with diabetic nephropathy and 3600 g (2880-4220) in the group without nephropathy, P = 0.52. In the lower quartile of birth weights the median (10-90 percentile) birth weights were 3000 g (2780-3200) in the group with nephropathy versus 2850 g (2250-3175) in the group without nephropathy, P = 0.07. In the upper quartile the median (10-90 percentile) birth weights were 4225 g (4000-4741) in the nephropathy group and 4000 g in the group without nephropathy, P = 0.13. We found no significant correlation between birth weights and log urinary albumin excretion rate (r = 0.148, P = 0.14) and no difference in the number of patients with nephropathy in the lower versus upper quartiles of birth weights. Conclusion. We found no evidence of low birth weight as a risk factor for the development of diabetic nephropathy.