The four cultures which form the basis of this communication were recovered from peculiar cases of primary cervical adenitis in man, three of which terminated fatally of disseminated acute miliary tuberculosis in four to six weeks.
A careful comparative study shows that Culture II corresponds closely with the "human" and Culture IV with the "bovine" type of tubercle bacilli; while Cultures I and III present variations from the standard types and are to be retarded as "intermediate" or "atypical" forms.
Culture I is of unusual interest because of its remarkable variations. The clinical picture of the case, the rapid course of the infection, the enormous number of the bacilli in the tissue, their tendency to occur in "heaps" like the leprosy bacillus, the high degree of virulence alike for rabbits and guinea-pigs, the production of lesions in chickens, the case of cultivation and the prolonged viability under unfavorable conditions, all mark the organism as a decided atypical form of tubercle bacillus in man. The prolonged viability, the production of lesions in the chicken and the great profusion of bacillary growth in the tissues would indicate an avian type. Though for years the reaction curve was atypical it has since changed completely to the "avian" curve. In this connection it is of interest to note that L. Rabinowitsch (3) states that she has isolated avian tubercle bacilli from two cases of tuberculosis in man.
Cultures II and III undoubtedly belong to the human type of the tubercle family though they were under cultivation and were repeatedly tested upon glycerine broth over a period of months before their identity was definitely established.
Culture IV completely corresponds in growth and reaction in glycerine bouillon to the bovine strain; however, it manifests a low degree of virulence for rabbits which is exceptional for bovine cultures.
The old belief that bovine bacilli are more slender and beaded in the tissues and are thicker and shorter in culture than the human type, I have not been able to confirm. The morphological characters of the different cultures here reported were so inconstant that no reliance could be placed on this feature as an aid in differentiation. Outside of the animal body it would seem that the differences in size and character of the individual bacilli depend largely on the kind and reaction of the medium, whilst in the animal body they are influenced by their situation and the resistance of the host.
The nature of the growth of these tubercle cultures varies for the same culture even under apparently identical conditions. The character of the growth was never an indication of the type of the culture, It was common to obtain two distinct types of growth on the same flask of bouillon, i. e., a portion of the surface would be covered with a heavy and uniformly granular layer of closely packed wax-like colonies twice the size of an ordinary pin's head, while the other portion would be a dense homogeneous layer with the typical depressed blisters.
The rapidity of growth also varied greatly for the same culture. Often in a series of twelve or more bouillon flasks which were prepared alike and inoculated with the same culture, some would cover the surface in eight days to two weeks, others would take four to six weeks, still others two to three months. It was thought in the beginning of the work that this variation might depend on the amount of oxygen within the flask or on the change in reaction in the bouillon, but further tests proved that neither of these influenced the rate of growth in any way. It would occur in loosely corked flasks as well as in those that were sealed, and in flasks where the reaction was neutral, acid or slightly alkaline. It would seem that these changes are by no means specific for any group of the tubercle bacilli but a property possessed by them all.
The growth of the cultures on solid medium showed approximately the same variation as that from the surface of the glycerine bouillon. The wax-like colonies described by L. Rabinowitsch (3) as characteristic for avian tubercle bacilli were noted at times for all of the cultures. On the modified egg mixture the growth was always more rapid and profuse than on any other medium. I found this egg medium more certain than any other for the direct recovery of the tubercle bacillus from the tissues. Where it was desired to recover the culture from the animal tissues with certainty and celerity it had no equal. Occasionally in seven days after the inoculation of the tuberculous gland material the growth was sufficiently advanced to transplant to the bouillon flasks.
The glycerine bouillon test serves admirably to distinguish between the human, bovine and avian types of tubercle bacilli. It is also of value in the determination of degrees of adaptation in man for bacilli of the lower host-species, and in the recognition of "intermediate" types. The test to be of differential value requires repeated application and careful control over a period of months. Some freshly isolated cultures may produce their specific reaction curve in glycerine bouillon within a few weeks. On the other hand the same culture may fail to give its characteristic reaction or any alteration in the glycerine bouillon for several months though the growth has been luxuriant and complete.
The rise in acidity that occurs in glycerine bouillon for the human type of tubercle bacilli is due to a specific action on the glycerine of the products of disintegration of bacilli (autolysis); with the bovine and avian types the products of bacillary disintegration have no action on the glycerine. The fall in acidity which occurs for all three types of the tubercle bacillus is due to the products of metabolic activity.