TYPHOID-FEVER POISON. 521 



stoned up, belonging to Otto Schmidt. All of these houses were in- 

 vaded by the fever. But it is necessary to account for the contamina- 

 tion of the water. Here comes in the fact that gave me the key to the 

 mystery. On the 20th of September, as I have already stated, there 

 occurred a severe thunderstorm. It was so violent that a large amount 

 of the rainfall was carried off as surface-wash. It did considerable 

 damage to the gardens, and, among other things, filled up to overflow- 

 ing the vault attached to Otto Schmidt's house. Here was deposited 

 all the excrementitious matter from this severe case of the disease. It 

 was easy to observe, from the appearance of a board walk, of the grass 

 and ground, that the overflow had scattered material of this kind in the 

 neighborhood of the well. Whether it ran into the well, or filtered 

 through the ground, it is difficult to say ; either could have occurred, 

 or possibly both. 



Now observe what occurred. Fourteen days after the shower, the 

 4th of October, a case of the disease broke out in the house g. Twenty 

 days after, the first case was taken down in the house c ; thirty and 

 thirty-six days later, two cases in c?, and two cases at intervals of six 

 and twelve days later in the house A. These were the families habit- 

 ually dependent on the Schmidt well for their supply of water. 



Admitting that I am right in tracing these cases to the contami- 

 nation of the well ^, by the shower of the 20th of September, we have 

 left two families in which occurred five cases of the fever who did not 

 make use of this well. On questioning the people in the houses e and,/", 

 I learned that they crossed the road and drew water from the well s. 

 This missing link in my chain of cause and effect made a halt in my 

 investigations. I had fancied that I was so near success that I took 

 this disappointment as a sort of personal matter, and was half inclined 

 to give the whole thing up. I made another effort ; I crossed the road 

 and investigated the well and its owner. There was no fault to be 

 found with the well, which was a " drive well " ; but the owner told me 

 that it occasionally became dry, and was in that condition before and 

 for a week or two after the great shower of September. Back I went 

 to my fever-houses, and exploded this question like a bomb-shell in the 

 midst of their inhabitants : " Where did you go for water when the 

 Bogert well was dry ? " " Oh, we went to INIr. Schmidt's well." 



After this testimony let us trace the course of the invasion a step 

 further. The fever appeared among the inmates of the house/ on the 

 20th of October, and of the house e on the 28th. Had we no further 

 evidence, here is sufficient to render probable that these cases were due 

 to drinking the infected water from the Schmidt well. But we may 

 strengthen it by negative evidence. The inhabitants of the uninfected 

 group of houses on the opposite side of the street at M never used 

 the suspected well. When the well s became dry they resorted to the 

 well at X. There is one other group of uninfected houses, two of them 

 belonging to the same block as the infected houses, and one situated on 

 VOL. XIV. — 34 



