Decembeb 1, 1905.] 



SCIENCE. 



695 



fluence, and the insects will always tend to 

 persecute most those houses which lie in 

 the immediate vicinity of their breeding 

 pool. Even when there are many pools 

 scattered about among the houses, there is 

 no reason why, after feeding, the mos- 

 quitoes will go to one rather than to an- 

 other; and the result must be that in gen- 

 eral they will tend to remain where they 

 were. 



Self-evident as this argument may now 



and drain away all the pools to the right of 

 it, leaving all those to the left of it intact. 

 Then all the insects on the left of the line 

 must be natives of that part ; and all those 

 on the right of it must be immigrants 

 which have crossed over the line from the 

 left. How many mosquitoes will there now 

 be on the right side, compared with those 

 on the left side? The following diagram 

 will enable us to consider this question 

 more conveniently. 



UNDRAINED COUNTRY 



BOUNDARY 



Y 



NORMAL DENSITY 

 X- 



FA 



t-L/A/G 



■£{%> 



■'SJK 



HALF DENSITY 



-y 



-L 



DRAINED COUNTRY 





ZERO DENSITY 



DiAGEAM II. Curve of falling mosquito-density due to drainage on right of boundary. 

 -L are the limit of migration on either side of the boundary. 



L and 



appear, it is not understood by many who 

 write on the subject and who seem to think 

 that mosquitoes radiate from a center and 

 shoot forever onward into all parts of the 

 country as rays of light do. Accepting 

 this fallacy without ciuestion, they argue 

 that it is useless to drain local breeding 

 pools because of the influx of mosquitoes 

 from without. Such an influx certainly 

 always exists; but I shall now endeavor to 

 show that it can not generally compensate 

 for local destruction. 



Let us consider a tract of country over 

 which numbers of mosquito-breeding pools 

 are scattered, with houses and other feed- 

 ing places lying among them. Suppose 

 w^e draw a straight line across this country 



First, examine the state of affairs before 

 the drainage was effected. We may sup- 

 pose that mosquitoes were then breeding 

 fairly uniformly over the whole country, 

 and that their density was much the same 

 on both sides of the line. A certain amount 

 of migration across the line, both from 

 right to left and from left to right, must 

 always have been going on; and since the 

 density was equal on both sides, this mi- 

 gration must also have been equal and 

 opposite — that is, as many emigrants must 

 have been constantly passing from right to 

 left as from left to right. Now, after the 

 drainage has been effected, the following 

 changes occur. The insects breed as be- 

 fore on the left of the line, and some con- 



