154 



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



secondary efferent trachea gives off branches, which branch again and again 

 "until the whole leaf is supplied with an immense number of capillaries. 

 The free edge of each leaf is much folded in and out, giving it the appear- 

 ^anice seen in a crinkled rhubarb or cabbage leaf. Fat-globules and dense 

 purplish-brown pigment are plentifully distributed, not only in the basal 

 swellings of the rectal epithelium, but also in the leaves themselves. The 

 thickness and complexity of the formation, together with the dense pigmen- 

 tation, make it impossible to do justice to its appearance in a photograph. 



The leaves ox folire arising in two longitudinal rows from one holobranch 

 tend to diverge distiilly. But the foliae from right and left halves of two 

 consecutive holobranchs do not come into very close contact, because each 

 folia assumes a direction nearly transverse to the long axis of the gill-basket. 

 Thus all the folise come to project into the cavity at approximately equal 

 intervals apart, the distal end or tip of each lying so as just to interpose 



Text-fig. 6. — Aisehna brevistyla, 

 Ramb. 

 Portions of three hemibranclis from tlie 

 anterior portion of the everted gill- 

 basket of a larva at about the 10th instar. 

 Normal Foliate Sub-type. Semi-dia- 

 grammatic, drawn from the freshly- 

 opened gill-basket. ( X 50.) 



Text-fig. 7. — JEschna brevi- 

 styla, Ramb. 

 Portions of two hemibranchs 

 from the posterior narrow por- 

 tion of the gill-basket of the 

 same larva. Normal Foliate 

 Sub-type. Semi-diagrammatic, 

 drawn from the freshlj^-opened 

 gill-basket. (XoO.) 



itself into the space left between two consecutive foliee of the next hemi- 

 branch (text-fig. 6). Hence, on everting a gill-basket of ^^sclina by a 

 longitudinal cut^ one notices at first twelve rows of leaves arrnnged in six 

 sets in which the distal ends of the leaves face one another, while the ti'acheal 

 axes appear to diverge from the region of a longitudinal axis which is the 

 true axis of symmetry of each holobranch. The two hemibranchs really 

 belonging to one holobranch have their folise arranged so as to diverge away 

 from one another distally. 



Arrangement of the Tracliece. — If one of the folise be dissected out, and its 

 free border smoothed out by means of a camel's-hair brush, it will be seen 



