83 



the eye. In every case tuberculosis and tubercle bacilli 

 were found in the infected animal. 



Having spent two years in the completion of this work, 

 amid all the facilities of the imperial laboratory ; having 

 meanwhile permitted himself no public intimation of the 

 same, Koch quietly announced his results at a regular 

 meeting of a medical society, with as little ostentation as 

 if he had merely appropriated a chapter from Ziemssen. 

 One whose knowledge of bacteria and of disease is not 

 such as to permit a technical appreciation of Koch's 

 work, cannot help seeing in the unobtrusive, systematic, 

 and undeviating work of two years, and in the modest 

 announcement of the result, that Koch's work is not to 

 be classed with that of Klebs. or Letzerich, or even Pas- 

 teur. I would call your attention to the fact that Koch's 

 assertion embodies not a theory, but simply an ocular 

 demonstration. If a man is seen to plunge a knife into 

 the heart of another the killing is a fact, not a theory ; 

 if Koch saw tuberculosis invariably follow the introduc- 

 tion of isolated bacilli, the relation of cause and effect is a 

 fact, not a theory. There is only one possible escape (I 

 use this word intentionally out of regard for the prejudices 

 of many friends) from the conclusion that the bacillus 

 causes tuberculosis ; and this forlorn hope is the possi- 

 bility that Koch did not see what he says he saw that 

 he made some vital error of observation. This is of 

 course possible, though if true it will be the first error 

 that the most searching scrutiny could ever detect in his 

 observations ; that it is improbable is evident. 



If we accept Koch's observations as accurate, there is 

 only one conclusion that these bacilli cause tubercu- 

 losis. For here the conclusion and the observation are 

 identical ; this is not a deduction, but a demonstration. 



And how shall it be decided that this work is or is not 

 free from errors of observation ? Certainly not by say- 

 ing that it cannot be so ; not by exhuming Niemeyer's 

 buried argument that tuberculosis is not infectious ; but 

 simply and solely through the repetition, by competent 

 observers, of the same work. Until such repetition shall 



