Renal arteriolar diameters were measured by using vascular casts in the remnant kidneys of subtotally nephrectomized rats. Two weeks after nephrectomy, both afferent and efferent arterioles were dilated, with development of glomerular hypertrophy. Thirteen weeks after nephrectomy, the afferent arteriole was dilated further, while the efferent arterioles became constricted. Glomerular hypertrophy was augmented. At this stage, the systemic pressure was elevated, with development of marked glomerular sclerosis. Throughout the experiment, captopril lowered the systemic pressure, suppressed glomerular hypertrophy, and limited glomerular damage. Both the afferent and efferent arterioles were dilated further. Changes in the arteriolar diameters and the systemic blood pressure suggested that elevated glomerular pressure existed in the remnant kidney and did not exist in the captopril-treated remnant kidney throughout the experiment. Hydralazine and trichloromethiazide therapy lowered the systemic pressure and maintained the arteriolar diameters and glomerular size at normal levels. However, at the late stage, afferent and efferent arterioles were dilated, with development of glomerular hypertrophy and severe glomerular sclerosis. The results suggested that elevated glomerular pressure was involved in development of glomerular sclerosis. However, factors other than hemodynamics should be considered in the pathogenesis of glomerular sclerosis. A direct causal relationship between glomerular hypertrophy and glomerular damage was not shown because captopril ultimately limited glomerular sclerosis despite glomerular hypertrophy at the early stage, and hydralazine and trichloromethiazide therapy did not ultimately ameliorate the glomerular sclerosis despite normal glomerular size at the early stage.