BY R. J. TILLYARD. 413 



as to the true respiratory value of the caudal gills and general 

 integument. 



It remains true, in spite of the fact that rectal respiration has 

 now been shown to be carried on both in newly-hatchefl and in 

 well-grown Zygopterid larvj^, that no trn.r trachi'dl (/ills are 

 developed in the rectum of these larvae The diffusion of COo 

 outwards into the rectal water-supply, and the diffusion of air 

 inwards, is undoubtedly effected tlirougli the rectal epithelimn 

 n-itlnmt any sjiecialisation of the tracheal supply of that rpgion, and 

 the exchange must be regarded as being primarily one between 

 the blood of the larva and the water in the rectum, cjuite irres- 

 pective of the tracheal system, which is bound to benefit sooner 

 or later by the change. Where the Anisoptera differ from the 

 Zygoptera is in the fact that a set of definitely new structures, 

 viz., trachpal ijilh, of great complexity and efficacy, have been 

 developed in the rectum in such a manner as to do away with 

 the need for an}' other method of respiration, and to bring the 

 necessary oxygen direct from the water to the tracheal system, 

 by means of the thousands of minute capillary loops which lie in 

 the projecting gill-folds. The thickened folds of rectal epithelium 

 in Zygopterid larvpp are structurally not homologous with the 

 gills of Anisopterid larvse, but rather with their basal pads, to 

 which they bear a close morphological resemblance, especially in 

 the more archaic types. It would seem, therefore, highly pro- 

 bable that the elimination of CO^ may be one of the principal 

 functions of these pads, seeiilg that, before special tracheal gills 

 were developed, such elimination must have proceeded outwards 

 pari passu with the entry of air inwards, by diffusion. 



In conclusion, we would suggest that there is some definite 

 correlation between the general asthenogenetic trend of develop- 

 ment of the Zygoptera as a whole, and their failure to develop 

 any single set of highly efficient respiratory organs in the larval 

 stage. They cannot compete on equal terms with the highl}-- 

 endowed Anisopterid larvie in this respect. Inherent weakness 

 in the respiratory system is as fatal to the development of a 

 strong evolutionary line as inherent inability to obtain food easily. 



