238 O. W. HOLMES 



those which are never so propagated. This general result I 

 shall only support by the authority of Dr. Ramsbotham, who 

 gives, as the result of his experience, that the same symptoms 

 belong to what he calls the infectious and the sporadic forms 

 of the disease, and the opinion of Armstrong in his original 

 Essay. If others can show any such distinction, I leave it to 

 them to do it. But there are cases enough that show the 

 prevalence of the disease among the patients of a single prac- 

 titioner when it was in no degree epidemic, in the proper 

 sense of the term. I may refer to those of Mr. Roberton 

 and of Dr. Peirson, hereafter to be cited, as examples. 



2. I shall not enter into any dispute about the particular 

 mode of infection, whether it be by the atmosphere the 

 physician carries about him into the sick-chamber, or by the 

 direct application of the virus to the absorbing surfaces with 

 which his hand comes in contact. Many facts and opinions 

 are in favour of each of these modes of transmission. But 

 it is obvious that, in the majority of cases, it must be im- 

 possible to decide by which of these channels the disease is 

 conveyed, from the nature of the intercourse between the 

 physician and the patient. 



3. It is not pretended that the contagion of puerperal fever 

 must always be followed by the disease. It is true of all con- 

 tagious diseases that they frequently spare those who appear 

 to be fully submitted to their influence. Even the vaccine 

 virus, fresh from the subject, fails every day to produce its 

 legitimate effect, though every precaution is taken to in- 

 sure its action. This is still more remarkably the case 

 with scarlet fever and some other diseases. 



4. It is granted that the disease may be produced and 

 variously modified by many causes besides contagion, and 

 more especially by epidemic and endemic influences. But 

 this is not peculiar to the disease in question. There is no 

 doubt that smallpox is propagated to a great extent by 

 contagion, yet it goes through the same records of periodical 

 increase and diminution which have been remarked in puer- 

 peral fever. If the question is asked how we are to reconcile 

 the great variations in the mortality of puerperal fever in 

 different seasons and places with the supposition of con- 

 tagion, I will answer it by another question from Mr. Farr's 



