A series of 166 American Indian renal biopsy specimens from 1971 to 1989 showed a very high proportion with mesangial proliferative glomerulonephritis with mesangial immunoglobulin deposition (Ig-pos mesGN). This disease comprised 68.7% of all the biopsies and 83.8% of all primary GN, proportions much greater than those (23.5% and 37.7%, respectively) of a local contemporaneous biopsy series from non-Indians (P < 0.001). These proportions and the extrapolated population-based incidence rates of mesGN are the highest yet described in any population. Males and females were equally represented in the Indian Ig-pos mesGN series. Biopsy was most commonly performed in early adulthood, but duration of suspected disease prior to biopsy was often many years. Mesangial glomerulonephritis often occurred in family clusters. It was occasionally associated with rashes, arthralgias, and/or a history of alcohol abuse. Due to different surveillance and biopsy practices, the spectrum of severity was different in Zunis and Navajos, allowing examination of clinicopathologic correlations over a broad range of disease. Early disease was manifest by microscopic hematuria alone, but rates of severe disease, with hypertension, heavy proteinuria, and renal insufficiency, were very high. Clinical severity increased with age, with extension of pathology beyond the mesangium, and with scarring and vascular change. Changing patterns of deposition of mesangial immunoglobulin and of electron-dense deposits in sequential biopsy specimens, recurrence of disease in kidney transplants, and biopsy diagnosis of disease in asymptomatic relatives of afflicted subjects were all observed. Five-year renal failure rates were estimated at 41%, much higher than in most other series. This disease constituted most biopsied cases of GN-end stage renal disease in Navajos, and largely determined the young age of Navajo GN-end stage renal disease subjects compared with their US-wide counterparts. To reconcile the diversity of clinical and morphologic findings in such homogeneous ethnic groups, and in individual families within such groups, we propose a unifying hypothesis for a broader spectrum of mesGN than traditionally defined by immunoglobulin subtype deposition. We also argue against a major role for IgA aggregates or immune complexes in initiating or propagating the mesangial inflammatory process, and propose that mesangial pathology of another, independent cause might be primary. Hyperinsulinemia, or insulin resistance, which is common in these populations and in other transitional populations with similar GN and ESRD patterns, might be one such mesangiopathic factor.