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



Arising also from this junction, a strong trachea, the latero-ventral branch 

 (Ivh), passes posteriad to the region of the anus. From this trachea the 

 remaining four efferents are given off to the ventral gill-fold. The last 

 efferent of all appears to be a straight continuation o£ Ivh, and gives off 

 numerous secondary tracheae to the ventral gill-fold. The anal branch (an) 

 comes off from Ivb at the point where we consider the last efferent to begin, 

 curves inwards towards the middle line, and then passes to the end of the 

 tenth segment. 



During the movements of respiration, the oblique branch o swings to and 

 fro, so that its angle of inclination to the dorsal trunk constantly varies. 

 Hence the position of the first junction (^i) is always altering. Sometimes 

 it is seen to lie just under the suture between segments 7 and 8 (text-fig. 19), 

 sometimes it appears in the anterior portion of segment 7 (text-fig. 18) . The 

 structure and size of this oblique branch strongly suggests that both the 

 visceral trunk and the latero-ventral branch are true outgrowths of the main 

 dorsal system, and probably receive their air in the first instance via the 

 oblique branch. I was not, however, able to observe this taking place during 

 the very short period in which the air was travelling along the tracheae. 



When we turn to the lateral gills (text-fig. 19) we can see that each gill- 

 fold is supplied by two sets of efferent tracheae, just as in the case of the 

 dorsal and ventral gill-folds. In the case of the latero-dorsal gill-folds, these 

 sets are both derived from the dorsal trunks, but in different manners. My 

 observations on their distribution are not so complete as I could wish, but I 

 was able to make out the following points. From the point where the large 

 postero-median branch (j^mlt) comes off, the dorsal trunk runs posteriad 

 almost to the anal end of the larva. This point I have termed the second 

 junction (^2). It marks also the branching off of the oblique branch (0) from 

 the main dorsal trunk. One set of efferent trachese, numbering eight or 

 nine, comes off from the whole length of the main dorsal trunk in this region 

 to supply the dorso-lateral gill-fold, while a second appears to branch off 

 from the efferent trachese of the dorsal gill-fold, close to their points of 

 origin. Owing to the great foreshortening of these trachese in lateral view^ 

 their exact distribution was difficult to make out ; while, in dorsal view, the 

 laterally-lying position of the latero-dorsal gill-folds made it impossible to 

 study their points of entry into the gill at all. 



With regard to the latero-ventral gill-folds^ the same difficulties were met 

 with, but in a less degree, as the amount of foreshortening was less. A set 

 of seven or eight efferent trachese enters these gills on their dorsal side, and 

 arises from the dorsal trunk. They seem to be united near their bases with 

 the lower set of efferents to the latero-dorsal gill-folds — that is to say, the 

 dorsal trunk on each side gives origin to a set of trachese, each of which 

 branches into two near its base, one branch passing to the latero-dorsal gill- 

 fold, the other to the latero-ventral. The other set of trachese supplying 



