For the past several years “bacterial tuberculoioidosis” has caused severe losses among young fishes of Seriola quinqueradiata and S. purpurascens cultured in various parts of Japan. The name “bacterial tuberculoidosis” was proposed by KUBOTA et al (1970) on th basis of the symptomatology and histopathology of the disease, though the causative agent responsible for it was not isolated. The present authors succeeded repeatedly in isolating the causative bacterium in pure culture from lesions of spleen and kidney of affected fish which were obtained on the occasions of outbreaks of the disease in Miyazaki, Oita and Ehime Prefectures during the period from 1969 to 1970. The bacterium when inolulated into Seriola caused the same pathological features as seen in naturally affected fish and killed them. The bacterium grew well on Brain heart infusion agar containing fresh human or rabbit blood and Brain heart infusion broth containing rabbit serum, both with 3% NaCl, but very poorly on ordinary nutrient agar or broth with 3% NaCl. In 2-day old cultures, grown on blood agar, cells were straight or slightly curved rods of 0.6 by 1.6 microns in the average size and arranged singly, in pairs or in chains. They were Gram-positive in young cultures but changed into Gram-negative after about 18 hours generally. Cells exhibited pleomorphism and had metachromatic granulus at the both ends of the cells. They did not form spores and were non-motile in semisolid agar madia. Capsules were absent. The bacerium was a facultative anaerobe. Nitrates were not reduced and neither indole (Kovac's method) nor hydrogen sulfide was produced. The oxydase test, methyl red test and catalass test were all positive, and the Voges-Proskauer test was negative. Gelatin was not liquefied. Agar Plates containing rabbit erythrocytes showed alpha-hemolysis. Acid was produced from glucose, mannose, levurose and galactose, but other sugars tested were not fermented. The bacterium was definitely sensitive to penicillin. On the basis of these observations, it is assumed that this organism is more closely related to the genus Corynebacterium than to any other known genera.