From the foregoing description it is evident that when rabbits are inoculated intravenously with equal amounts of tubercle bacilli of bovine and human type respectively, they are subject to an immediate reaction in the form of an interstitial pulmonary exudation, which, being of equal severity and character does not serve to distinguish the type.
There is an hyperplasia of the lymphoid tissue which is much more pronounced in the bovine series and which may distinguish this type. Opinion on this point may well be reserved until other typical cultures are examined for this response.
The two types are sharply distinguished by the behavior of the tubercle bacillus and by the progression of tubercle formation. Tubercles are formed by both types and for about I week after inoculation they are not distinguishable. Progressively thereafter those formed as a response to the bovine bacillus become more numerous; they caseate and become conglomerate, finally coming to occupy the major part of the pulmonary tissue and its associated lymph nodes. With the human type the tubercles do not progress to caseation, do not become more numerous after their first well defined formation, and finally tend to disappear. The human type bacillus does not multiply considerably, if at all, and disappears early. The bovine bacillus suffers little or no restraint in growth and finally multiplies enormously. It seems clear that so far as histologic evidence goes the fundamental difference in the reaction of the rabbit to the two types of tubercle bacilli is referable to the ability of the animal to restrain the growth of the human type or to the prevalence of conditions which permit a most vigorous multiplication of the bovine type. The initial cellular responses seem to be qualitatively of the same order and their quantitative distinctions are for the most part developed coincidently with the manifest growth of the bovine type bacillus.
If we undertake to state the observed results in the terminology of immunity we can say only that the histologic picture discloses a difference in the rate of bacillary multiplication which suggests that a difference in the physiologic requirements for growth of the two types of bacilli is satisfied or unsatisfied, in the respective cases, by the rabbit as host; or on the other hand, that there is a positive growth-restraining action exerted with efficiency against bacilli of the human type. It is evident that the present observations furnish no points of discrimination between these alternatives. There is, however, an occasional result of the injection of human type bacilli into rabbits (not seen in this series) which offers a suggestion.
When animals so injected are allowed to live for 2 or 3 months, the lungs at autopsy not infrequently present a few nodules of large size, often 1 cm. in diameter, which are found to be well encapsulated, soft, caseous masses. These often contain large numbers of tubercle bacilli. Since we know nothing of the particular conditions which give rise to these rather exceptional formations it is impossible to draw general conclusions from them, but they do suggest that the rabbit is not lacking in the food materials required by the human type bacillus; and that if the more usual suppression of this type is due to failure of its essential nutritives, it is rather a question of the distribution within the animal than an absence which is responsible. The usual result would then appear to be due to a positive growth-restraining action exerted against the human type bacillus.
Certain other points of interest in the histologic picture described are worthy of comment.
The lymphocytes do not appear as active cells in any preponderant way in either series and they are much less in evidence in the immune case (human type) than in the non-immune (bovine type). This might suggest that the activity of this cell type is a response to infection rather than that it furnished an effective preexisting barrier against infection in this particular case. If the lymphocytes were the most important agents in the immune reaction, it might be expected that they would show an immediate sharp response in the human series.
The large mononuclear type of cell is clearly most closely related physically to the tubercle bacillus within the body of the rabbit and this without distinction as to bacillary type. Foci of these cells are the loci of the disappearing bacilli of human type, and in either the active or necrotic state similar cell collections are the site of the most vigorous multiplication of the bovine bacilli. These cells undoubtedly stand in a central position in any picture which can be drawn of experimental tuberculosis in the rabbit and deserve as a consequence all of the very considerable attention they have received at the hands of numerous observers in recent years.
It has been quite usual of late to consider that the whole of the essential reaction of the animal against tubercle bacilli is carried by the cells of the mononuclear series, either lymphocytes or large mononuclears according to the predilections of the observer. We cannot, however, entirely overlook the presence in very large numbers of polymorphonuclear leucocytes, both amphophilic and eosinophilic, in this experimental series. They are much less prominent in the animals injected with the killed culture and hence can hardly be neglected on the assumption that they are merely a part of a reaction to an indifferent foreign body They are in large measure a reaction to the living organism: whether a primary and direct or a secondary, indirect consequence of its presence we are unable to decide.
These cells are not massed in any regular relationship to the well formed tubercles or to the clusters of mononuclear cells initiating tubercle formation. They are also very much less abundant in the very severe late lesions of the bovine type where enormous numbers of bacilli are enclosed in the tubercles. It seems possible that the polymorphonuclears are a response to the living free tubercle bacilli as contrasted with either the dead bacilli or the living bacilli segregated in mononuclear cell clusters or in tubercles. They would appear also to be related to something apart from the bacillus itself, either a diffusion or disintegration product, since phagocytosis of bacilli, or the presence of bacilli in close physical relation to polymorphonuclear leucocytes, is so infrequent in general as not to have been observed in this series of experiments.