130 ME. K. J. TILLYAKD ON THE RECTAL BREATHING-APPARATUS 



connections, but without making any notable addition to our knowledge o£ 

 these orgaias. Miall (16), East (7), and Latter (11) have contributed popular 

 accounts of the same structures. 



The greatest advance in our knowledge of the subject is undoubtedly due 

 to Sadones (25), who^ working under Professor Grilson (10), produced in 1896 

 a classical memoir on the whole digestive tract of the larva of Lihellula 

 depressa. The outstanding merit of this work is the careful and accurate 

 study of the histology of the rectal epithelium, and the part played by it in 

 the formation of the gill-lamellse. Sadones's original contributions to the 

 subject include (1) the discovery of the basal pads of the lamellse ; (2) the 

 discovery of the small tubercles on the lamellae, which prevent them from 

 lying too closely upon one another ; (3) the demonstration of the complete 

 continuity of the rectal epithelium as a fine syncytial matrix beneath the 

 whole branchial cuticle ; and (4) the fixing of the definite position of the 

 tracheal capillary loops as running in this syncytium. To him, also, we owe 

 a very clear criticism of Chunks description and figure. Sadones, however, 

 seems to us to have forsaken firm ground when he turns to the physiological 

 aspect of the subject. His theories concerning the absorption of oxygen and 

 the elimination of carbon dioxide, whether they be ultimately proved correct 

 or not, show a distinct lack of understanding of the closed traclieal system 

 as it exists in Odonate nymphs, and are quite unsupported by any of that 

 definite evidence which the importance of the subject demands. 



After Sadones, the subject remained untouched for sixteen years (except 

 for a short account by Scott (26), who added nothing new to our knowledge) 

 until K-is (23) published, in a short but very valuable paper, the results of his 

 comparative studies on the rectal gills of a number of European Anisopterid 

 larvae. Ris described the gill-structures in the following genera : — Cor- 

 dulegastei\ Gomplms, Onycliogomphus, B^^achi/tron, JEsclina, Anax, Cordidia, 

 Ofthetrum, Lihellula, and Sympeirum. In these ten genera he recognizes 

 six distinct types of gill-formation. He also shows how these structures are 

 found to be both more morphologically complex and physiologically perfect 

 as we pass from genus to genus along the phylogenetic tree, as it has been 

 already constructed for us on the venational characters of the imagines. Eis 

 made a very beautiful series of microphotographs of the various tvpes of 

 gills ; but these were unfortunately never published, owing to the difficulty 

 of tinding a process which would reproduce them satisfactorily. 



Following on all these authors, it now remains for me to extend the 

 methods and observations of Ris to our rich Australian fauna, and also to 

 attempt the more difficult histological study of each new type, on the lines 

 adopted by Sadones in his classical memoir on Lihellula depressa. These two 

 objects achieved, there remain some interesting phylogenetic considerations 

 to be given, which will be found included in this paper. 



