176 MK, R. J. TILLYARD ON THE RECTAL BREATHING-APPARATUS 



text-fig. 12, which is drawn from a photograph o£ an opened and cleared 

 gill-basket of Diplacodes hcematodes, and also to text-fig. 11 [Syntliernis), it 

 will be seen that this is by no means the case. On the contrary, each pad 

 lies at the outer anterior extremity of the basal line of insertion of each 

 lamella, and hence it is as far as possible from the axis of symmetry. This 

 can also be easily verified in transverse sections, where the pads are aiwaj'S 

 found lying on the external sides of their lamellae (Plate 22. fig. 27). Now, 

 since the basal line of insertion of each lamella lies in a slanting direction 

 to the longitudinal axis of the gill-basket, with its inner end considerably 

 more anterior than its outer, it follows that the external position of the pad 

 in cross-sections indicates an actual position at the far end of the lamella 

 and on its anterior side. Sadones^s error obviously arose from his failing to 

 distinguish clearly the two sets of lamellae belonging to one double series. 

 He has, by error, selected two adjacent hemibranchs from left and right 

 respectively of two adjacent double series. Hence his axis of symmetry is 

 really a line midway between two double series, and the position of the basal 

 pads is correspondingly misdescribed. 



Plate 22. fig. 27 shows a transverse section through one single series of 

 lamellae in Cordidepliya pygma;a, enlarged from a portion of the complete 

 section shown in fig. 24. Apart from their shape, the structure of the pads 

 closely resembles that already described in Austrogomphus. The pad is 

 more thickened than in jEsclma, and the meganuclei are quite as elongated 

 as those of Austrocfomj^lius. 



Sadones mentions the occurrence, in the basal pads of Libellula, of an 

 occasional small nucleus, considerably smaller than the micronuclei, lying- 

 close up to the cuticle. I have also noticed similar nuclei in the pads of 

 other Libellulidse, but I may add that they are of very rare occurrence. 

 One of them is shown in fig. 27, rm^. 



The very striking shape and colour of the basal pads in Libellulidae is 

 worthy of remark. On opening a gill-basket, the pads appear to be rows of 

 buttons, of a rich transparent orange-brown colour, surrounded, in Austro- 

 cordidia, by a ring of very dark brown. It is somewhat doubtful whether 

 this colour actually belongs to the pad or to the underlying hypobrauchial 

 tissue. But, judging by its regularity of shape, I am inclined to attribute 

 it to the pad itself, since the masses of hypobrauchial tissue are of far 

 less regular shape, and often project outwards below the bases of the 

 pads. 



The shape of the pad appears different according to the point from which 

 it is viewed. When the gill-basket is everted, the pads at either end of a 

 hemibranch are viewed somewhat end-on and appear as very elongate ovals. 

 Towards the middle of the rows the convexity of the basket may throw the 

 pads upward so that a more direct view is obtained. Viewed thus from 

 above, they appear either circular or very convexly triquetral. It will thus 



