THK LIMITATIONS OF KEROSENE AS A LARVICfDE. 280 



Theoretically, then, the thoroughly aerated water in the large bottle should contain 

 enough oxygen in solution to supply the requirements of one fully developed larva 

 for 68 days. There are, however, many reasons why the period of survival should 

 be shorter than that theoretically possible. For example, it is not natural for the 

 lai-va to respire in this manner, and tho fully developed individuals are not able to 

 adapt themselves to the uimatural conditions so well as are younger larvae. Also the 

 water, although cut oft' from the air, is not sterile and organisms grow in it, and must 

 absorb some of the oxygen. It is on the.se organisms that the larA^ae feed. 



AVhen young larvae are used, they are found to adapt themselves much more easily, 

 and if they do not actually survive the length of time they theoretically might, they 

 nevertheless live for many days, grow in size, moult and develop fully. 



In order to have studied this (juestion iidecjuately it would have been necessary to 

 determine the respiratoiy exchange of each species of larva dealt with, as well as the 

 variations occurring under dift'erent conditions of temperature, etc. As it was not 

 possible to do this with the ap]iaratus available, the subject was not pursued further, 

 but even these incomplete observations support Lima's conclusion that larvae when 

 denied access to the air can live only in well-aerated water. 



The Survival of Mosquito Larvae Submerged in Slowly-running Water. 



Having ascertained in these preliminary experiments that S.fasciata larvae are able 

 to live a long time respiring only dissolved oxygen, an attempt was made to reproduce 

 more nearly the conditions present in a pool or stream in which the water is not held 

 fast until its oxygen is exhausted, but is able to circulate more or less freely. The 

 apparatus used is shown in figure 1. It was composed of a cylindrical glass tube 

 (AB) of 700 cc. capacity, firmly corked at both ends, and provided with an inlet at 

 C a)Kl an outlet at D. A large glass funnel was attached to the inlet by a rubber tube, 

 and into this funnel water from a tap was allowed to drip. To the outlet a short 

 rubber tube fitted with a sciew clamp (E) was attached, by means of which the flow of 

 water could be regulated. The actual vessel used by me was the water-jacket tube of a 

 Liebig condenser. In order to ensure the oxygenation of the water, it was delivered 

 as a spray from a finely pointed glass tube (G) fixed about two feet aliove the funnel.* 

 When an experiment was to be started, the end A of the vessel was lowered and water 

 allowed to run in, the clamp E was tightened and the vessel filled up, the larva or 

 larvae introduced at B, the end A lowered until the water overflowed at B, and the 

 cork at this end was then thrust home. In this way air bubbles were completely 

 avoided. The inlet and outlet had of course to be blocked by loose pads of wool, 

 cylinders of gauze, or some such contrivance, so as to prevent the escape of the 

 larvae. 



Under these conditions, so far as could he observed, the lar\'ae of S(e<juii(//iafascia(a 

 lived in comfort. Young larvae grew and moulted in a normal manner, and both 

 you)ig and fully developed larvae moved freely about the vessel feeding, iMit develo]v 

 ment was arrested at the fourth or final phase and pupation did not as a rule take place. 

 No food was supplied to the larvae, but they apparently found sufficient in the 

 circulating water and on the sides of the vessel, where no doubt bacteria and fungi 



* In tho diagran) the downward portion of this tube has boon drawn much too long ; 

 tho end (G) should be two inches above wliore it is shown. 



