抄録
The classification of the subspecies of Pasteurella pestis has been made from a viewpoint of biological reactions on one hand and from a viewpoint of virulence on the other hand. The former viewpoint is, for example, represented by the classification employing the glycerol-fermentation test or by that depending on the host animal species from which a certain type of these organisms is preferentially recovered (Tumansky, 1957, 1958) . The latter viewpoint, which is thought to be a more practical one, has been taken by Bhatnagar (1940), Seal (1951 a, b) and Korobkova (1956), who classified P. pestis into virulent, avirulent protective and avirulent non-protective groups of strains on the basis of virulence and immunogenic properties to various experimental animals.
Jawetz and Meyer (1943) distinguished two groups of avirulent strains: the first group strains (A 1122, E.V. 51) possess a large amount of envelope antigen, i.e. Fraction I of Meyer (1950), and are also protective for mice; the second group strains (TRU, 14) are poor in envelope antigen, but protective for guinea pigs. Another group of strains which is avirulent and non-protective both for mice and guinea pigs was reported by Bhatnagar (1940), Seal (1951 a) and Amies (1951) . Kasuga (1945) described that an avirulent variant, strain MII40, had a protective power for guinea pigs and human beings from lethal infections with fully virulent plague bacilli, even though it had no detectable amount of envelope antigen.
Considering those results, the present author proposes at first the following scheme for classification of the groups or varieties of strains of P. pestis from the point of virulence and immunogenicity.
1. Virulent protective strains.
2. Avirulent protective strains with envelope antigen.
3. Avirulent protective strains without envelope antigen.
4. Avirulent non-protective strains.
As will be shown in the present series of papers, the interrelationship between these different groups of strains has been explained to a certain extent and some complemental corrections will be added in their conclusions. It is most conceivable from my results, that every strain of P. pestis can take any phase of the above varieties depending on their life history, in other words, they may represent each of phases of the evolutional process of plague bacilli, where the environmental factors seem to play important roles.
The present paper deals with an atypical form of P. pestis, which may be considered to occupy the fifth place in the above mentioned scheme of classification, with respect to lacking in the plague murine toxin which is contained in the other varieties. Evidences will be presented that this form has peculiar colonial and morphological characteristics. This form will be designated as the third colonial form of P. pestis hereafter, because it should be regarded as one of the phases of colonial variation such as classical smooth and rough forms, which have been customarily accepted as the representative colonial forms. Incidentally, it may be noteworthy to mention here that many practical laboratory workers dealing with plague bacilli have mistakenly recognized this form of organisms as contaminants other than P. pestis.
