1. In the intradermic test virulence of “morphological diphtheria bacilli,” 0·2 c.c. containing 100 million organisms is injected. The control guinea-pig receives 500 units of antitoxin the night prior to the test. The test animal receives 125 units six hours after the injection of the cultures. A known virulent organism is always injected as a control; 12 to 20 cultures can be tested on one guinea-pig.Of 2368 swabs received, 1388 (59 per cent.) showed morphological C. diphtheriae. Fromthese 1286 pure cultures were isolated (92·7 per cent.). In 234 positive swabs from cases of clinical diphtheria, in a series examined under very favourable conditions, morphological C. diphtheriae were obtained in 231 (98·7 per cent.) and of these 229 (99·1 per cent.) were virulent. In 758 positive cultures from carriers morphological C. diphtheriae were isolated in 715 (94 per cent.); 50 per cent. were virulent C. diphtheriae, 39 per cent. avirulent, 3·6 per cent. of Hofmann group, 0·3 per cent. Xerosis and 6.4 per cent. unclassified diphtheroids.3. Only one culture of intermediate virulence was found, i.e. uncertain reactions occurred with a dose of 100 million organisms while clear reactions occurred with 1000 million.4. Out of 626 positive cultures from cases of diphtheria, both virulent and avirulent bacilli were isolated in two instances, once from the throat and once from the nose.5. A monovalent antitoxin (Park Williams 8) nenutralised all of 819 virulent strains.6. Amongst 39 cultures obtained from the ear in post-diphtherial conditions, virulent C. diphtheriae were isolated fourteen times.7. Impure primary (“field”) cultures gave reliable results (94 per cent.) when more than one-third of the organisms seen in the smear made from the overnight culture were morphologically C. diphtheriae. If fewer C. diphtheriae were present in the smear th e primary culture gave uncertain results.