xm, B.i Barber: The Transmission of Malaria 39 



is readily infected experimentally with formation of sporozoites 

 in the salivary glands, while type indefinitus is little susceptible 

 to experimental infection. Again the pond where the mosqui- 

 toes that were used in these experimental cases were obtained 

 was examined repeatedly during some six months subsequent to 

 these experiments, and nearly 99 per cent of the hundreds of 

 rossi larvae collected were of the Giles type. Such evidence, 

 taken by itself, would not be convincing, since the anopheline 

 fauna of a given pond often varies, especially if the water level 

 varies greatly, but the constancy of this type in the breeding place 

 over such a long period may have some value as confirmatory 

 evidence. 



Considering all the evidence, it is practically certain that type 

 Giles rossi infected the experimental cases, but one cannot wholly 

 exclude the possibility that one or more of the insects was of 

 type indefinitus. 



Both experimental cases — Chinese coolies — freely volunteered 

 to submit to those experiments, and both had seen enough of ma- 

 laria to be fully aware of the possible consequences of infection. 



No. 727, male, 26 years of age, born in China, had lived in the 

 Federated Malay States a little less than five years. About two 

 years previous to the experiments he had had one attack of fever 

 with rigors daily, the attack lasting about two weeks. He had 

 no history of fever since. The spleen was normal. He had 

 had secondary eruptions of syphilis about one year previously. 

 As shown in Table XVII, a daily blood examination showed no 

 parasites until the fourteenth day after exposure to infected 

 mosquitoes. 



On March 15 and 16 three attempts were made to in- 

 fect No. 727, but only five mosquitoes took blood. One of 

 these had oocysts in the mid-gut, but none had sporozoites in 

 gut or salivary glands. . On March 17 mosquitoes that had been 

 deprived of sugar for two days were applied and five out of 

 eleven exposed took blood. Two of the five had sporozoites 

 in the salivary glands. The others were negative, in both 

 gut and salivary glands. Dissections were made immediately 

 after feeding. These mosquitoes had been infected by crescent 

 carrier 537 (Table XI) nineteen to nineteen and one-half days 

 previously. Mosquitoes had been exposed twice to the crescent 

 carrier, and on the first exposure, the evening of February 26, 

 the percentage of crescents was 35.6. • 



Both of the sporozoite-bearing mosquitoes had well-developed 

 ova, a further indication that they were of type Giles, since, in 

 the comparative infection experiments (see discussion preceding 



