88 



where cases had occurred or rats had died was tested by 

 placing a guinea-pig there overnight. The guinea-pig was- 

 subsequently segregated, and in 31 cases it died from plague. 



Experiments were made in which two similar animals 

 monkeys, guinea-pigs, or rats the one protected by gauze 

 or sticky paper, the other not protected, were placed in 

 suspected houses. Of 92 experiments, 15 unprotected 

 animals died, none of the protected. 



On 96 occasions fleas taken off guinea-pigs which had 

 spent the night in a suspected house were transferred to a 

 healthy animal. In 26 cases the second animal died from 

 plague, although not infrequently the guinea-pig from which 

 they had been removed escaped infection. The conclusion 

 that in a plague-infected house the infection is due to- 

 infected rat fleas, and not to an infection of the soil or the 

 air, seems to me abundantly justified. 



Fleas as tJie Agents of Transmission from Rat to Man. 



In considering how far the results just detailed can be- 

 applied to man we enter at once upon less secure territory, 

 because it is impossible to put conclusions to the test of 

 experiment. The only measure of the correctness of such an 

 interpretation is its adequacy to interpret all the known 

 epidemiological facts concerning the spread of plague. 



The justice of applying the results of these animal experi- 

 ments has been severely criticised in several quarters, and 

 particularly by Galli-Valerio (1907), on the ground that rat 

 fleas do not bite man. It is true that, speaking generally, 

 different animals harbour specific parasites ; but the more 

 we learn upon this subject the clearer it becomes that such 

 specificity is not so sharply defined as was imagined, and 

 that a particular flea, although commonly confined to a few 

 hosts of allied species, may frequently be found in consider- 

 able numbers upon quite different animals. Moreover, it has 

 become abundantly clear that a flea, although exhibiting a 

 decided preference for one species, will, if hungry, betake 

 itself to animals of widely different type in the absence of 

 its proper host. 



The common flea found on rats in India and other warm 

 parts of the world 5 is X. eheopis. This flea is a non-combed 

 species, and not unlike the human flea Pulex irritans in 

 appearance. That JL. cJieopis readily feeds on man was 

 observed by Tidswell (1903), Gauthier and Raybaud (1903), 

 and Listen (1905). The Commission for the Investigation 

 of Plague in India kept X. eheopis alive for four weeks upon 

 an exclusively human diet, and my colleague, Mr. Bacot, has 

 bred them for years on the same regimen. 



5 Distribution of rat fleas has recently been dealt with by Dr. Chick 

 and myself in the Journal of Hygiene, vol. ii., p. 129, 1911. 



