Differentiation between glomerular and non-glomerular haematuria by observation of the changes in red cell morphology using phase-contrast microscopy is a well established technique. However, the method is not widely accepted in clinical practice because of controversy regarding the minimum percentage of dysmorphic red cells required to diagnose glomerular aetiology, as well as the need for specialized microscopes. Recently, a glomerular-specific morphological alteration of red cells has been described, which has the form of a doughnut shape with one or more blebs and which is termed the "G1" cell. In the present double-blind prospective study 250 urine samples were examined without any knowledge of diagnosis. Haematuria was detected in 122 cases. The type of haematuria was characterized by counting dysmorphic cells and G1 cells separately, in each case using a phase-contrast microscope as well as an ordinary bright-field microscope with and without staining of urinary sediments. The results were later correlated with the confirmed diagnosis. The study showed that the G1 cell is more specific than the dysmorphic cell for the diagnosis of glomerular haematuria. Evaluation of both dysmorphic red cells and G1 cells can be done using bright-field microscopy with 100% specificity and sensitivities of 82 and 100%, respectively. It has been concluded that the ordinary bright-field microscope can be used for the diagnosis of glomerular haematuria with an efficiency similar to that of a phase-contrast microscope.