TIIR LIMTTATIOXS OF KEROSENE AS A LARVlCir>E. 291 



eJsewliciP (]91o), may leacli tlio hundredth dav. The larvae kept in the vessel from 

 whicli air was excluded behaved in an exactly similar way, and it seems probable that 

 tlie reason why none survived for months was mainly due to the fact that a compara- 

 tively small number of specimens was used and a centenarian did not happen to be 

 included. 



In this system the larvae alter a time became reconciled to the new conditions, 

 appeared no longer to be anxious to find the air surface, and settled down to brousmg 

 along the sides of the vessel. It was thought possible that lar%-ae thus habituated to a 

 submerged existence might be unable to resume the normal method of breathing. 

 One set of seven larvae of Stegomyia fasciata was therefore tested after four and a half 

 days and the larvae were found to be still quite able to respire through the siphon 

 tracheae. On being removed to a jar half-filled with water they quickly found the 

 surface and attached themselves to it in a natural manner. 



A few experiments were carried out with the larvae of two species of Cidex, namely 

 C. faligans and C. tlialassius. The larvae of C. faiigans did not survive a day in my 

 experiments, but the larvae of C. ihalassms seemed to be able to adapt themselves 

 completely and lived a fortnight, at the end of which period they were removed for 

 examination. It is of some interest that this species should be able to survive without 

 access to air, as it possesses very small anal papillae, and it suggests that the oxygen 

 absorption by submerged larvae of C. tJialassiiis is mainly through the general 

 cutaneous surface and only secondarily through the papillae. After a fortnight's 

 submergence no change in size or structure could be seen in the anal papillae of these 

 larvae. 



Pu]3ation of a submerged larva was observed only twice in all my experiments, 

 once in the apparatus described above, and once in a gauze-capped tube immersed 

 in a large vessel of water. On both occasions the pupa died immediately, and in one 

 case it did not succeed in liberating itself from the larval pelt. The inability of 

 the pupae to subsist by respiring only dissolved air is no doubt accounted for by the 

 fact that they possess an impermeable cuticle. 



My observations confirm those of Lima, to which reference has already been made. 

 This observer experimented with the larvae of Culexfatigans and Stegomyia fasciata . 

 as well as with other species that are not found on the Gold Coast. C. faiigans he 

 found survived submergence but a short time, but S. fasciata was able to adapt itself 

 to the unusual conditions and might live even so long as 53 days. The results of his 

 experiments convinced him " that mosquito larvae, while generally breathing mainly 

 free air by the two tracheae of the respiratory siphon, also respire the oxygen of the 

 air dissolved in water, the gaseous exchanges being made by the branchial leaflets 

 and also the general integument of the body." 



Sen, using a different method, found that the " larvae rarely lived for nu)re than 

 7 hours," and he therefore doubts the accuracy of Lima's results. His method was 

 however questionable, in view of the fact already referred to that it is oidy a small 

 proportion of the oxygen dissolved in water that is serviceable to the larvae. 



Sen has criticised Lima's method of carrying but his experiments because bubbles 

 occasionally appeared in the jars, and air might have been introduced, he thinks, 

 with food particles, or by the agitation of the v.ater wlien it was chauired. The 



