BY R. J. TILLYARD. 425 



tissue is not clearly understood, but it seems to be a derivative 

 of the fat-body, and obviously lies in the hsemoccele. In fact, 

 it practically blocks out the circulation of the blood from the 

 gill, in which definite blood-spaces do not really exist, except 

 around the larger tracheae near the base. In the lamellae itself, 

 although the two walls of the gill can be slightly forced apart by 

 the artificial injection of fluid (thus proving the continuity of 

 the haemocoele into this organ), yet no circulation of blood- 

 corpuscles has ever been observed; and, so far as ray own observa- 

 tions go, such circulation would be impossible, owing to the fact 

 that the two walls of the gill are, in the natural state, practically 

 in contact everywhere except at the base. In other forms, where 

 tracheae of considerable calibre penetrate some distance into the 

 gill, blood-spaces are met with further in (6s) around these 

 tracheae. In the Papillate type, the blood-spaces follow the axial 

 trachea of the papilla almost to its tip. But in no case is there 

 any noticeable circulation of blood, and that fluid does not pla}^ 

 any important part in the problem of oxygen-caiTying. 



The gill is covered everywhere by a very fine transparent 

 cuticle (cu), representing the inthna or chitinous lining of the 

 undifferentiated rectal epithelium. Beneath the cuticle lies the 

 gill-epithelium {ep). This is a thin layer of very flattened cells, 

 all fused together into a syncytium in which the nuclei (nitj are 

 clearly visible at somewhat irregular intervals, and placed rather 

 far apart. The mauve colour of the lamellae of H. tau is due to 

 the presence of very fine granules of purplish pigment {fg) 

 scattered in this syncytium. 



The arrangement of the gill-tracheae can be gathered from 

 Fig.l. Almost the whole of the lamella proper carries only 

 capillaries (ca/>). These are very numerous and excessively tine 

 tubes, with no trace of a spiral thread. It is very important to 

 note that each capillary forms a" complete loop, passing up one 

 side of the gill, crossing over beneath the rounded distal border 

 of the lamella, and descending along the other side, finally joinino- 

 up with other capillaries to enter a larger branch which unites, at 

 last, with the branch from which the capillary took its rise. 

 Under a high power, the capillaries can be seen to lie actually 



