45 



by these points. The posterodorsal rods have just appeared; they are 

 fenestrated. The dorsal arch has a very long posterior process; the fact 

 of its appearance at this comparatively young stage would seem to indicate 

 that the arms, except the postoral ones (— they are broken in the spe- 

 cimen, so that their exact length cannot be ascertained; but they are 

 fairly long, anyhow — ) are, upon the whole, short. The anterolateral 



Fig. 11. Skeleton of the larva of Toxopneusies roseus (?). ^^Vi- Letters as in fig. 5. 



arms (they are broken, but somewhat restored in the figure) are evidently 

 short. The posterior transverse rod has not yet appeared. The body 

 skeleton is nearly smooth (Fig. 11). 



Pseudocentrotus depressus (A. Ag.). 



As the breeding season of this species did not occur during my stay 

 in Japan (April — July), I could not study its development myself. Being, 

 however, very anxious to learn, whether the larva of this species would 

 prove to be of the same type as the other Toxopneustids, as I expected, 

 or perhaps of the Echinus-Strongylocentrotus type, I asked my Japanese 

 colleagues to do me the service to try to rear the larva, when its breeding 

 season came on. Dr. Fujita, then Assistant at the Biological Station 

 of Misaki, kindly undertook to rear the larva, when in the end of Oc- 

 tober ripe specimens of the species were found. The preserved larvae 

 were sent to me and received in February 1916 in Panama. On exami- 

 ning them I found my expectations confirmed; the body skeleton formed 

 a basket structure as in the Toxopneustidae (i. e. most of them, Lgtechi- 

 nus forming a remarkable exception). The closer study of the larvae, 

 which were reared only till the first stage (the oldest being 10 days old), 



