312 Transactions of the Society. 



has never been yet considered or referred to. Taking the dimen- 

 sions of the Bacteria to be, diameter ' 5 /i, which is a fraction less 

 than the actual measurement, and the length to be 2 diameters, 

 which is undoubtedly under the average, a very simple calculation 

 shows that in a drop, taken as the 16th part of a cubic centimetre, 

 there would be 250,000,000,000 (two hundred and fifty thousand 

 million), or just a quarter of a billion ; this would be when the 

 blood was entirely filled with, or rather replaced by a solid mass of 

 Bacteria, leaving no space at all for the blood-corpuscles and but 

 little for the plasma ; and this is the utmost number which a drop 

 could contain. I think it is evident, therefore, that there is some 

 fundamental error in Davaine's statement and in that of those who 

 have followed him, on this point. I have endeavoured directly to 

 enumerate the number of Bacteria present in difierent portions of 

 blood, but I cannot pretend to have succeeded with even approxi- 

 mate accuracy ; the greatest number I could enumerate or estimate 

 was a few millions in a drop. 



Another point of special interest in this affection is the asserted 

 increase in the infective virulence of septicaemic blood in successive 

 generations of transmitted infection. This theory was explicitly 

 maintained by Coze and Feltz, but Davaine's statements on the 

 subject have been somewhat misunderstood, for although he asserted 

 this in the fullest extent at first, he ultimately qualified the state- 

 ment in some measure by showing that the maximum of viru- 

 lence is reached very early ; subsequent observers overlooked this 

 qualification, and repeated and even improved upon Davaine's 

 original statements. This question has again lately attracted 

 attention in connection with the relation of micro-organisms to 

 disease, and the sensational and, were they to be credited, appalling 

 statements that have been made, and even supported, by high 

 authority, asserting a transformation of physiological species in 

 some of the lower organisms, which hypothesis, it was supposed, 

 might be connected with or account for an increase in infective 

 virulence in the organisms present in septicaemic blood in successive 

 generations. On this point I shall only say that I have found in 

 a long series of experiments recently made, that although the 

 infectivity of such blood may be slightly variable, there is no such 

 thing as progressive increase of virulence in successive generations ; 

 the blood of the first generation is actively infective in the millionth 

 or the lOO-milhonth of a drop, or less, and it is not, and indeed 

 for the reasons already stated, cannot be infective in very much 

 smaller quantities, in the 25th nor any succeeding generations, 

 nor is there any shortening of the incubation period, which in the 

 large majority of cases is remarkably constant, ranging from 

 twenty-one to twenty-four hours. 



The relation of the organisms here described to the disease in 



