42 RepoTt on an Experimental Investigation on 
of four were placed on the other field, 5. The third lot re- 
mained in the pasture they had been on from February. Lot I. 
— After three weeks on sewage-grass one was noticed to be 
slightly ill at night, and died at 8 A.M. on the following day. 
The remaining four were taken off this grass, three of them 
were slightly ill, two " scouring badly, one passing blood from 
the womb." After three or four days they were well. They 
were kept from the sewage fields a fortnight, then the four of 
Lot L and the four of Lot IL were put together, and were on 
the sewage pastures 12 hours each day. In about 10 days, 
heifer No. 2 died, after 14 hours' apparent illness. This was 
the only remaining one of Lot L which had not been ill. 
Heifer No. 3 was one of the second lot which had been on 
these fields 12 hours each day for about 10 days, and was not 
ill for 4^ days after it had been removed from the second field. 
Lot in. remained perfectly well ; they had not been on the 
sewage meadows. Another sheep died there of anthrax on 
July 30th. 
These facts seem to require one or two words of explanation 
to bring out more clearly the inferences deduced from them. 
At first sight it does not clearly appear why only some of the 
animals were affected. Nor can it be seen why the sheep were 
affected in less proportion than the cattle. But certain facts 
must be borne in mind. The first is that the sheep in wet 
meadows rarely drink from running water, and as a matter of 
fact it was noticed that those sheep which cropped the thick 
grass at the edges of the runnels were those which died. Most 
of them would not take it there, but some seemed to have a 
preference for it. In the same way some of the cattle would 
not, especially at first, drink the water flowing in these channels, 
but after a time they grew accustomed to it, and took it in 
preference to other water. It is probably to this fact, as well 
as to the long intermission in the supply of the infected sewage, 
that the escape of the majority of Lot II. is due. A con- 
siderable interval was known to have elapsed, during which 
no washing of Van mohair was carried on, and the near 
coincidence of the recurrence to the use of that material, and 
the occurrence of cases in men and in cattle, is very striking. 
On the other hand, if any other local cause of infection from 
previous cases had been operative, there can be little doubt that 
the mode of occurrence of the cases would have been very 
different ; and there is thus a strong body of evidence that the 
disease was directly due to infection from the mohair. 
To this farm three of the animals which had been inoculated, 
A, C, and D, were sent on July 14th, and placed at once on 
the fields in which the previous cases had occurred. They 
