THE POSSIBLE TRANSMITTER 195 



that the insect transmitting it must be a relatively 

 hearty feeder ; the insect must also be a fairly " long- 

 tongued " one in order to reach the parasite. Another 

 factor needing consideration was the way in which 

 Kala-azar repeatedly occurred in some houses, while 

 neighbouring ones remained free of it. This sug- 

 gested that the insect carrier was one which would 

 return to its old haunts and would remain there until 

 new victims were available for food. Captain Patton, 

 working in Madras, had considered all these problems, 

 and by a process of exclusion concluded that the 

 bed-bugs, extremely common in Madras, were the 

 most likely insects to be concerned. Their wide 

 distribution, voracious feeding, retreating into hiding, 

 and possession of moderately long sucking apparatus 

 were all in their favour. The natives also aided in 

 their ubiquity by their reluctance to destroy vermin, 

 which, being merely deposited in the street after 

 capture, promptly returned into the house whence 

 they had been ejected, or entered one in the immediate 

 neighbourhood. 



Prolonged search was made by Patton of all the 

 blood-sucking insects found in infected areas. They 

 included head lice, body lice, and ticks that frequented 

 the person or the dwelling, the gnat Culex fatigans, 

 the mosquitoes Neocelliastephensi and Stegomyia ingens, 

 and the tick Ornithodorus savignyi,which were common 

 pests indoors and outdoors alike. Some of the insects 

 were found to contain natural flagellates, even when 

 they were bred in the laboratory. When they were 

 fed on patients, in no case did the Leishman-Donovan 

 bodies undergo any development, nor were these 



