TUBERCULIN REACTION 625 



found in apparently healthy persons and cattle. The conditions under- 

 which a negative reaction may occur in the presence of tuberculosis 

 are, however, fairly well understood, and physicians should be thoroughly 

 acquainted with these. Likewise most instances in which a positive 

 reaction was observed in the apparent absence of tuberculosis have 

 usually narrowed down to the fact that the lesion was so small or so 

 situated as to escape detection, and, indeed, this has been shown so 

 conclusively by autopsies that, in the presence of a tuberculin reaction, 

 on the autopsy must rest the burden of proof and blame. When we 

 realize how small a lesion may produce hypersensitiveness, it will readily 

 be understood how easily the clinician and pathologist may fail to detect 

 the lesion. 



A large part of our knowledge regarding the specificity of the tu- 

 berculin reaction has been gained from veterinary practice, as the results 

 of a test in an animal could immediately be controlled by the autopsy 

 findings. Thus Fraenkel 1 collected from the literature 8000 carefully 

 observed instances, and found only from 2 to 3 per cent, of differences 

 between the result of the tuberculin test and of the autopsy. Voges, 2 

 in 7327 instances, noted 2.7 per cent, of contradictions. Kuhnau, 3 

 Bang, 4 and von Behring 5 speak of similar experiences. 



It has long been known that the prevalence of tuberculous findings 

 anatomically far exceeded the number of cases recognized clinically. 

 Among cattle, anatomic tuberculosis is found in from 12 to 25 per 

 cent., and about 80 per cent, or more react to tuberculin. In many of 

 the latter, however, the disease does not progress, but, on the contrary, 

 tends to recede. 



Similar conditions exist in human pathology. That tuberculosis is 

 very frequent among adults is now well known. The figures of Nageli 6 

 and Burkhardt 7 showed that the increasing frequency of tuberculous 

 infection with advancing years reached over 90 per cent, among those 

 past the eighteenth year; these figures are now well corroborated. Ham- 

 burger, 8 in the published results of an analysis of 848 autopsies on chil- 

 dren, showed that tuberculosis was in evidence in 40 per cent., increasing 



1 Zeitschr. f. Tuberk., 1900, i, 291. 



2 "Tuberculin und Organismus," Jena, 1905, 77. 



3 Berl. tierarztl. Wochenschr., 1899, 78. 



4 Sixth Internation. Congress on Tuberculosis, 1908, 211. 



5 Beit. z. exper. Therap., 1905, x, 1. 



6 Virch. Arch. f. path. Anat., 1900, clx, 426. 



7 Zeitschr. f . Hyg. u. Infectionsk., 1906, liii, 139. 



8 Wien. klin. Wochenschr., 1907, xx, 1070. 



