In connection with microbiological findings in two patients, a report on septicemia caused by C. tropicalis is given. C. tropicalis is second in frequency to C. albicans in man. Because for the different biological properties of Candida species in each case of isolation of such fungi from clinical specimens an identification of the species is obligatory, because the generalizing diagnosis of "Candida mycosis" may be insufficient for the detection of epidemiological and pathogenetic associations. The possible primary resistance of C. tropicalis against the antimycotic 5-Fluorocytosine (Ancotil) is discussed. It is shown that the antigen similarity within the most important Candida species as e.g. between C. albicans and C. tropicalis may cause a false interpretation of serological results concerning the causative agent of the infection. For a safe inactivation of C. tropicalis on human skin, cleaning with 70% isopropyl alcohol for ten seconds is proposed. Because C. tropicalis is entering the human body preferably through food it is discussed whether a microbiological control of food for patients predisposed for such infections should be performed.