ECHINODERM LARV^. 471 



to be pelagic, the small number of species of these larvae, 

 four or five at most, which have come under my notice 

 is somewhat disappointing. On the other hand, no 

 Echinoderm larva with which I am acquainted has 

 appeared in such large numbers as that of Asterias rubens 

 (PL I, figs. 1-9), which I have been able to identify 

 satisfactorily by comparing those taken in the plankton 

 with others reared from eggs fertilised naturally in one 

 of the tanks of the Port Erin Aquarium. On several 

 occasions I have seen a few examples of the blastula 

 (fig. 1) and early gastrula (fig. 2) stages as early in the 

 year as the end of January ; but during the years 1907 to 

 1912 inclusive, the species appeared in great force about 

 the middle of February, when every stage, from the 

 gastrulating blastula to the early bipinnaria (figs. 2-9) 

 was obtainable in the plankton, and furnished abundant 

 material for a study of the earlier stages of development. 

 In February, 1913, careful search through a number of 

 tow-nettings failed to reveal a single specimen of this 

 larva, but it re-appeared at the same time in 1914. I 

 have met with the stages represented in figs. 8 and 9 in 

 large numbers in the early part of March ; but later than 

 this they have disappeared from view; and I have not 

 been able to trace the later bipinnaria and brachiolana 

 stages. The course of development of the early stages is 

 in complete accord with that of the larva of Asterias 

 vulgaris as described by Field (1). Fig. 4 represents, 

 from the ventral surface, an early bipinnaria stage, in 

 which the coelomic vesicles are seen budding to right and 

 left from the blind end of the archenteron. The latter 

 is bending ventrally, as shown in the slightly diagram- 

 matic optical section (fig. 5), to meet the stomodaeal 

 invagination (std.). The blastopore (hi.) is moving 

 forward towards the ventral surface of the larva. 



