BY H. J. TTLLVAKD. 33 



To deal with all five of these possibilities would take us beyond 

 the limits of a single paper. Leaving aside entirely the con- 

 sideration of the functi(His of the general integument, as well as 

 that of the spiracles, which, even if they are really permeable, 

 can only be used for breathing in air, and not in water, we still 

 have three sets of organs demanding our attention. Specially 

 projecting folds of the rectum, assumed to have a i-espiratory 

 function in cases where direct experiments have shown the ex- 

 istence of regular movements of impulsion and expulsion of water 

 in the rectal cavity, have so far only been found in the case of a 

 few genera, though it does not follow that they may not be uni- 

 versally present in the sub-order, if carefully sought foi*. Paired 

 lateral processes or gills are known to be confined to a few genera 

 in the subfamilies Ejmllaginm and Thorince. There remain, 

 then, for the Zygoptera, the universally present organs known as 

 the caudal c/ills, clearly of the highest importance to a right 

 understanding of the respiratory processes of the larvte of this 

 sub-order. I have, therefore, decided that the first step neces- 

 sary in the solution of this latter problem is a careful study of 

 the morphology of these organs. This paper is confined entirely 

 to that one object, and will only touch upon other aspects of the 

 problem when their introduction appears to be necessary for the 

 development of the main purpose of the paper. 



I had originally intended to confine my study to Australian 

 forms. However, it soon became apparent that the triquetral 

 form of gill, which does not occur in any Australian genus, must 

 be included in any general study of these organs. I therefore 

 sought to obtain gills of this form from various European and 

 American correspondents. Dr. P. P. Calvert, of the University 

 of Pennsylvania, Philadelphia, U.S.A., very kindly provided me 

 with the required material, in the form of fixed larvae of Caloj)- 

 teryx and Hetcerina. To him, I desire to express my best thanks 

 and appreciation of his kindness. I desire, also, to thank Pro- 

 fessor W. A. Haswell, F.R.S., Professor of Biology in the Uni- 

 versity of Sydney, and Dr. S. J. Johnston, Lecturer in Biology, 

 for much valuable advice concerning the technical difficulties 

 which have been met with during the course of my work. 



