CLASSIFICATION 841 



20 minutes at 15 lb., or by absorbing out the anti-bacillary agglutinins from a 

 mixed serum. No attempt was made in this work to study the relationship of 

 different members of the group to each other, but in a paper published shortly 

 after Howie and Cruickshank's, Lamanna (19406) brought evidence to show that 

 B. suhtilis, B. vulgatus, and B. mesentericus along with B. agri, belonging to the 

 small-celled species, could be separated serologically from each other by means 

 of their spore antigens (see also Lamanna 1942). Separation of members of the 

 large-celled species was less successful, but a broad distinction could be drawn 

 between B. cereus and B. megatherium. 



Pathogenicity. — Speaking generally, we may say that the anthrax bacillus is 

 highly pathogenic for most animals ; and that most other members are non- 

 pathogenic for all animals. This statement, however, must be qualified. Under 

 natural conditions the anthrax bacillus gives rise to disease in man, cattle, sheep, 

 and certain other of the domesticated animals ; under experimental conditions 

 it is pathogenic for the laboratory animals. Under natural conditions other 

 species of Bacillus rarely give rise to disease, but an exception must be made for 

 B. suhtilis, which may cause severe eye lesions, notably iridocyclitis and pan- 

 ophthalmitis (Axenfeld 1908), and which may occasionally invade the blood stream 

 of patients whose powers of resistance are lowered by the attack of some fatal 

 disease (Sweany and Pinner 1925). Occasionally too, infections such as menin- 

 gitis and pneumonia may be due to pseudoanthrax bacilli (Senge 1913, Wilamowski 

 1912). There is evidence that some members, when allowed to grow excessively 

 in food, may produce toxic substances capable of giving rise on ingestion by man 

 to gastro-enteritis (see Chapter 72). Under experimental conditions, the pseudo- 

 anthrax bacilli are non-pathogenic for all laboratory animals except mice, and 

 for these animals only when injected intraperitoneally in a large dose — 1-3 loopfuls 

 of an agar culture. B. megafJierium is, however, definitely toxic, and is able to 

 kill guinea-pigs injected intraperitoneally in less than 24 hours. This is due to 

 the formation of a htemolysin. The fact that some pseudoanthrax bacilli may 

 on occasion prove pathogenic to man and animals, and that after long subculture 

 B. anthracis may lose its virulence for laboratory animals, suggests that there 

 may be a gradual transition from the non-pathogenic to the pathogenic state 

 (but see p. 845). A pseudoanthrax bacillus, described as B. tropicus, has been 

 isolated from mice inoculated with the blood of patients suffering from " coastal 

 fever " in Queensland, but its relation to the disease is doubtful (see Heaslip 

 1941). 



Classification. — This is very difficult, and any classification adopted is bound to 

 be arbitrary. Some authors divide the group on the basis of motility, others on the 

 character and situation of the spore, others on cultural characteristics, and others 

 on several properties taken together. To each of these methods there are objections, 

 and agreement is still far from being reached. In two of the most recent studies, 

 for example, the criteria used for classification differ widely, with necessarily 

 contradictory results. De Soriano (1935), who made a systematic study of 206 

 strains, proposes a classification based primarily on whether or not the spore 

 causes deformation of the bacillus, and secondarily on morphological, cultural and 

 l)iochemical characteristics. Lamanna (1940a), on the other hand, who studied 

 105 strains, suggests a classification based primarily on the size of the bacterial 

 cell, and secondarily on the mode of germination of the spore. Thus he dis- 

 tinguishes first of all between small-celled and large-celled species. The small- 



