Systematic application of renal biopsy studies is a relatively new strategy aimed at improving the understanding of disease processes in man. However, much has been learned from the quantitative study of such protocol biopsy materials. We have determined that renal structure and function are closely associated in IDDM patients, albeit that substantial structural changes may be present before any functional abnormalities are detectable. It is now known that the closest correlate of renal function (GFR, albuminuria, blood pressure) in IDDM is mesangial fractional volume expansion [Vv(Mes/glom)] but that vascular, glomerulosclerotic and interstitial lesions also play a role. We have learned that appropriately spaced biopsies, several years apart, provide a more dynamic view of a diabetic nephropathy (DN). Using paired biopsies 5 years apart it was learned that changes in albuminuria are more closely related to increasing Vv(Mes/glom) than changes in glomerular basement membrane (GBM) width or Vv(Int/cortex). Recent studies of tubular basement membranes (TBM) indicate that DN lesions involve these structures with equivalent changes to those seen in glomeruli and this is a strong argument against the hemodynamic hypothesis. It has been demonstrated that changes in renal structures over time can be used as an endpoint for clinical trials and using such a design, we learned patients under tight glycemic control prevents the early lesions of diabetes in the renal allograft compared to patients maintained with standard glycemic control. Similar biopsy studies are now being done to test whether other therapies can slow the rate of development of important lesions of DN at the earlier stages of diabetic renal disease, before functional outcomes would be anticipated. Also, from sequential biopsy studies we have learned that the established glomerular lesions of DN in IDDM are reversible by successful pancreas transplantation. Although not evident at 5 years, reversal is seen after 10 years of normoglycemia. Structural studies of IDDM sibling pairs are indicating concordances in the severity and in the patterns of lesions. These studies are strongly supportive of a genetic basis for DN risk. Studies of NIDDM patients indicate greater complexity to the nephropathology indicating variable pathogenetic mechanisms. Immunohistochemical and in situ hybridization studies of renal biopsy material in IDDM are helping to unravel the basic nature of this disorder. Further, all biology and molecular biology studies of cultured cells derived from IDDM patients whose diabetic nephropathy lesions are defined by renal biopsy are expanding our investigative options. In summary, the renal biopsy has greatly increased our understanding of DN and, having raised many questions, is guiding our future research directions.