94 



into the cavity of the pouch which at this stage is being constricted 

 from the free end of the archenteron. The only way I can account 

 for the presence of this cellular mass is that it is the core of a once 

 solid archenteron. 



Larvae twelve hours older than those just described show two 

 bulges a little posterior to the median transverse line, one on the left 

 the other on the right side. The blastopore has closed and the posterior 

 end of the archenteron has broken away from the ectoderm and lies 

 loose in the larvae except for its support by the mesenchyme which 

 fills all the space not occupied by other organs. The anterior end of 

 the archenteron has bent over and fused with an invagination of the 

 ectoderm thus forming the oesophagus and mouth, the latter being 

 situated at a point on the ventral surface about equidistant from the 

 anterior and posterior ends of the larva. 



Encircling the oesophagus lies the already five rayed but horseshoe 

 shaped hydrôcoel with ray number three pointing directly toward 

 the anterior end of the larva. Lying on top of the stomach and con- 

 nected with it , is the pouch which will form finally the Hypogastric 

 enterocoel of Goto 1898. This large pouch has arisen through a longi- 

 tudinal constriction of the archenteron, and as stated above is in this 

 stage not entirely cut off. 



On the left side at the point where the hydrocoel and enterocoel 

 are in open connection, the pore canal arises and passes towards the 

 dorsal surface but its walls have not fused with the ectoderm in larvae 

 of this stage. Dorsal to the stomach lies the pouch which will ultima- 

 tely form the epigastric enterocoel of Goto. Figure 1 is a reconstruction 

 of a larva at this stage, and shows it with the ventral half of the ecto- 

 derm removed and the oesophagus cut across just ventral to the hydro- 

 coel. Figure 2 is a section through such a larva at the plane indicated 

 by the line a b on Figure 1, and shows the connection of the stomach 

 and hypogastric enterocoel pouch. In a sixty-hour larva the external 

 appearance has quite changed, for instead of the coat of cilia, w^e find 

 the cilia restricted to four transverse bands. The anterior end of the 

 larva, which will enter directly into the formation of the star, is very 

 much widened and gives to the whole a club shape. Encircling the 

 mouth are five groups of rounded elevations of the ectoderm, three 

 in each group. They are caused by the pushing out of the end ten- 

 tacle and first pair of foot tentacles which in this stage are quite large, 

 behind these is the much smaller second pair of foot tentacles. 



Figure 3 is the ventral view of the larva just described and Fi- 

 gure 4 is a reconstruction of the anatomy of the same. 



In a five day embryo, the ectoderm lying under the water ring 



