146 MR. R. J. TILLYARD ON THE RECTAL BREATHING-APPARATUS 



Arrangement of the Tracliece in the Gill. 



Plate 18. fig. 1 shows the arrangement of the tracheae in a portion of a 

 main longitudinal gill-fold of Austrogomphus ocJtraceus, which, previous to 

 being photographed, had been smoothed out almost into one plane by means 

 o£ a camel's-hair brush. The gill-trachese leave the secondary efferents 

 along the line of the base of the gill a little to left and right of it alternately. 

 Each undulation is usually supplied by four or five of these gill-trachese, 

 which are of gross calibre run far up into the gill-fold, their distal ends 

 reaching fairly close up to the free edge of the fold. From each gill- 

 trachea short side-branches are given off, most abundantly towards its 

 distal end. These branches quickly break up into tufts or bundles of 

 capillaries, which also seem to be most numerous towards the distal ends 

 of the gill-trachese. The capillaries diverge as they leave the branches, and 

 spread out all over the gill-fold, running slantwise at all angles towards the 

 free edge of the gill. Here they all turn over i^i complete loops. It is 

 possible under a high power, by careful manipulation of the fine adjustment, 

 to follow the course of a single capillary, and to note two interesting facts in 

 connection with it. First, those capillaries which arise near the distal 

 end of a gill-trachea, after forming their loops, run back to enter branches of 

 an adjoining gill-trachea which are situated more proximally to the gill-base. 

 Secondly, those capillaries which arise so as to run along one gill-wall, after 

 forming their loops, descend along the opposite gill-wall. As the two gill- 

 walls are in very close apposition, this change of position can only be 

 accurately judged by the use of the fine adjustment. By examining the free 

 edge of the gill-fold in Plate 18. fig. 1, about the middle of the picture, both 

 the above conditions can be made out fairly well. 



In the cross-folds, the gill-tracheae, short branches, and capillaries are very 

 similar to those in the main folds, but the gill-tracheae are shorter, and do 

 not extend so fur into them. They are thus mainly filled with capillaries 

 which arise and end more at the same level than do those in the main fold. 

 In Plate 18. fig. 1, three cross-folds can be seen flattened down below the 

 main fold, while portions of two intermediate cross-folds lying above the 

 main fold can also be made out. 



Numher of Capillaries in the Gill-hasket. 

 In Austrogomphus ochraceus I estimate the number of complete loops 

 belonging to each secondary efferent trachea at 100 or over. There are 

 6 longitudinal main folds^ each of which is thrown into about 12 principal 

 undulations, each of which receives approximately the whole of the 

 capillaries from one secondary efferent trachea and its branches. Thus 

 we get a total of 100 x 6 x 12, or 7200 complete capillary loops for the six 

 main folds. Allowing a total of 50 per cent, extra capillaries for the twelve 



