28 Modes op Infection. 



Next, many flies were kept in the wire-netting stable 

 with the healthy and affected horses and allowed to bite 

 the animals freely. Excepting a case of doubtful infec- 

 tion (Experimental horse No. 472) among the healthy 

 horses, the transmission of the disease was not so marked 

 as in the mixed pasturing and in wooden enclosures. 

 Hence the stable-flies cannot be taken as the true trans- 

 mitter of the disease. 



F. Horse-flies (rabamc^ae)— These are most abundant 

 in the pastures and attack horses violently. In the stables 

 they are scarce. It must especially be mentioned that 

 the spread of the disease coincides in time with the ap- 

 pearance of this insect. Therefore at the outset of the 

 experiments we caught many horse-flies and tried to 

 let them bite healthy horses, but we soon discovered 

 it extremely difficult to keep them alive for any 

 length of time. Moreover, we could not force them to 

 suck blood out of the patients, though all the possible 

 artificial means were tried. Not having been able to 

 overcome these difficulties, we could not, to our regret, 

 domonstrate experimentally that the horse-fly is an 

 actual transmitter of the diesease. But deducing from 

 the fact that the infection takes place both in the mixed 

 pasturing and in the wood enclosures at the period of 

 the appearance of horse-flies and that other blood-suck- 

 ing insects are not likely to be the virus-transmitters, 

 as was already described, we canot escape from the 

 conclusion that the infection of the disease is accom- 

 plished through the agency of the horse-fly. 



Several species of the horse-fly occur in the pasture, 

 where the foregoing experiments were performed. The 

 time of appearance differs in different species. 



Chrysopus japonicus, Wied. The first part of May— 

 the first part of June. 



