148 



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



rectum, and having the tracheal branch as a kind of axis within it. These 

 branches o£ the gill-trachese will be called the a.vial or papillar trachece 

 rtext-fig. 4, ax). The papillae, though actually arising thus in groups o£ five 

 or six, each group corresponding to a single " wave " in the original undulate 

 type, are so numerous and closely set as to appear like a continuous series. 

 The papillae are developed, not only along the course of the main gill-folds, 

 but also along the cross-folds. Hence, on opening the gill-basket longi- 

 tudinally, one sees a mass of papillae of an intense shining whiteness, pro- 

 jecting apparently from all points of the surface. 



Text-fig. 8. — Hemigomphus heteroditus, 

 Selys. 

 Posterior portion of a main fold from the 

 everted gill-basket of a nearly full-grown 

 larva, viewed somewhat in profile. Cross- 

 folds omitted. Papillate Tyj)e. Semi-dia- 

 grammatic, drawn from the freshly 



Text-fig. 4. — HemiffOiJijjhus heteroditus, 

 Selys. 

 Anterior portion (about one-fourth) of 

 a papilla from the same larva. ( X 600.) 

 ax, axial or papillar trachea ; cA, chi- 

 tinous rods ; tl, loops of tracheal capil- 

 laries. 



opened gill -basket. (X 30.) 



Structure of the FapiUce. — Each papilla consists of a simple finger-like 

 eversion of the wall of the gill-fold, containing no pigment. Its chitinous 

 cuticle, when examined under a high power, is seen to be beset with 

 numerous short transverse chitinous rods, which give it a finely ribbed 

 appearance (text-fig. 4, c/i). The use of these little chitinous projections 

 seems to be to prevent the papillae from clinging together, and thus to 



