BY R. J. TILLYARD. 67 



the wall of the ventral blood-canal, on the side farthest from the 

 plane of symmetry of the gill-system. I'he other, or dorsal, 

 branch divides into two a little further on, one branch continu- 

 ing a fairly straight course, a little outside and dorsad of the 

 level of the nerve vl, to enter the base of the lateral gill of its 

 own side, as the dorsaf nerve of the lateral yiU {dl, dl'j — a nerve 

 considerably smaller than ?7, but placed in a similar position to 

 it upon the wall of the dorsal blood-canal. 



The second branch passes rapidly upwards and inwards over 

 the anal opening, and branches into two just before entering the 

 base of the median gill. Its two branches form the dorsal and 

 ventral nerves of the median y ill {dm, vm) respectively, of its own 

 side. A corresponding pair of nerves enters the median gill 

 from the other side. Thus, the nerve-supply of the median gill, 

 like that of the median ocellus, is a double one, symmetrical 

 about the plane of symmetry of the larva. 



In sections of the gills of Synlestes and Austrolestes, I can find 

 no trace of the two ventral nerves of the median gill. It would 

 appear, therefore, that these nerves are suppressed, in the family 

 Lestidce. On the other hand, it is easy to trace several strong 

 nerves arising from the dorsal nerves at various levels along the 

 gill-length, and passing closely around one or other of the two 

 main longitudinal tracheae, to supply the ventral portion of the 

 gill-blade. 



The structure of one of the main dorsal or ventral gill-nerves 

 may be seen from Plate vi., figs. 44, 45. A cross-section passes 

 through a considerable number of parallel nerve-fibres, the nuclei 

 of which are found at fairly frequent intervals along the course 

 of the nerve. Frequently the same section will intercept two 

 nuclei, as in Plate vi., fig. 45, and occasionally three. The nuclei 

 occur in all positions, either near the centre, or upon the circum- 

 ference, but seldom project beyond the regular periphery of the 

 nerve. 



The main gill-nerves retain their position either in, or very 

 close to, the walls of the main blood-canals, throughout the 

 whole length of the gill. Only in the Saccoid Gill-type are they 

 removed any distance from these canals. In this type, also, their 



