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



At regular intervals the rectal epithelium o£ the gill-basket is evaginated 

 into the interior to form one o£ the gill-folds. Their position and extent vary 

 in the different types of gill-basket ; but the structure of every type of fold 

 is essentially the same. The two walls of the evagination come into close 

 apposition over their whole extent, except for a comparatively small basal 

 portion, where they remain separate, enclosing a small space in which the 

 liypohrancMal tissue is developed, and into which the efferent or gill-trachese 

 pass from the exterior (Plate 22. figs. 25-27). The fine tracheal branches or 

 capillaries of the gill-fold are supported in the syncytium of one wall or the 

 other. Throughout the distal portion of the gill, where the syncytia of 

 the two walls fuse completely, the tracheal capillaries are seen to lie in the 

 fused mass, but usually slightly nearer to one side than to the other. Owing 

 to the great rapidity with which the air passes out from the capillaries after 

 the death of the larva (no trace of air can be detected therein two to three 

 hours after death) and to the absence of any spiral thread in these tiny 

 tracheoles, it is seldom that one can make out the lumen of any of the 

 capillaries in cross-section. But one finds, throughout the slender distal 

 portion of the gill-folds, in addition to the already-mentioned nuclei of the 

 epithelial matrix, a large number of tiny nuclei of very characteristic 

 structure, w^hich are undoubtedly the nuclei of the tracheal capillaries, since 

 they correspond very closely to the nuclei lying in the matrix of the larger 

 tracheae. They are small, usually rounded, and stain deeply with hsema- 

 tosylin, so that no very definite nuclear structure can be made out in them. 

 Frequently one notices a tiny clear space attached to one side of one of these 

 nuclei. This appears to me to represent the collapsed lumen of the capillary 

 vessel. Other small bodies noticeable in the syncytium of the distal portion 

 of the gill, but not present in the gills of all forms examined, are small 

 pigment-granules and tiny transparent globules of a highly refractive nature, 

 probably comj)osed of fat. 



The basal portion of each gill-fold, as has been already stated, broadens 

 out so that a space of greater or less extent is formed between the two walls. 

 Here the epithelium either of the anterior of the two walls, or of both, may 

 be broadened out into the form of a thickened basal pad. As these structures 

 vary much in shape and position, as well as in actual size and thickness, I 

 have postponed a general discussion of them until after the various types of 

 gill have been described (see p. 170). For the same reason I shall here only 

 mention the presence, in the space between the walls at the base of a gill, of 

 a peculiar mass of tissue whose origin and functions seem to be rather 

 doubtful. This is the tissue called by Chun and Faussek connective tissue, by 

 Sadones adipose tissue. As both these names are admittedly unsuitable, I 

 propose here for it the name hypohrancliial tissue. We may then define the 

 hypobranchial tissue of the gill-basket of Anisoptera larvse as the mass of 

 tissue lying in the basal space between the two walls of the gill-fold, and 



