SUMMARY: This paper describes an evaluation of the whole cell agglutination test for tuberculosist, carried out as Army Medical Directorate Research (AMDR) project 221. A !total of eighty-three cases, where the consultants were able to give a final diagnosis, are considered. The test was found to give inaccurate results in 39 per cent of cases under clinical investigation, also to show an 18 per cent false positive result in ante-natal patients who were established as not suffering from tuberculous disease. Apart from the erratic and unreliable results obtained with sera from women patients generally, the results presented here show that the test is too inaccurate to be of clinical value in early disease.