SYNOPSIS Five hundred specimens of urine have been examined for pyuria and bacteriuria, the leucocytes being stained by the Sternheimer-Malbin method. Most urines contained either less than 1 or more than 10 leucocytes per cmm; a few specimens contained 1 to 10 cells per cmm, whatever their viable bacterial count. The presence of leucocytes in urine was usually related to the bacterial count, pyuria being commonest in urines showing 'significant bacteriuria'. However, urinary tract instrumentation caused pyuria in the absence of infection. Leucocytes with nuclei staining blue by the Sternheimer-Malbin technique were considered to be indicative of active inflammation, but the incidence of such cells appeared to be a reflection of the total leucocyte count of the specimen rather than of its viable bacterial count. In the majority of cases the diagnosis of infection can be made on the basis of the bacterial count and the degree of pyuria. The staining technique appears to have a limited use, restricted to the interpretation of cases in which the results of culture and conventional leucocyte counts are ambiguous.In a clinical bacteriology department, pathologists are frequently confronted with the problem of assessing the significance of pyuria and bacteriuria. The definition by Kass (1956) of a level of 'significant bacteriuria' has done much to clarify the differentiation of infection from contamination, and some form of quantitative or semi-quantitative culture is now standard practice in most laboratories. Nevertheless, in a proportion of patients, the 'significant' level of 105 organisms per millilitre of urine will not be observed, despite the presence of active infection, since such factors as frequency of micturition or diuresis may prevent the multiplication of bacteria in the bladder which is necessary for attainment of the significant level (Roberts, Robinson, and Beard, 1967). In hospital bacteriology, the situation is further complicated by the fact that many patients may be receiving antibacterial therapy, and the drugs may reach such high concentrations in the urine that bacterial growth on culture media is inhibited. Indeed certain authors (Wear, 1966;Bradley, Crowley, and Darrell, 1967) question the wisdom of too rigid an adherence to the criterion of 105 organisms per millilitre in the interpretation of urine cultures.