l:mbryolo3v of THALASSEMA MELLITA 187 



It may not be out of place to direct attention once more to 

 the striking similarities in the formation of the prototroch in the 

 larvae of annelids and molluscs. So detailed and precise are 

 these resemblances that the homology of this highly characteris- 

 tic larval organ seems to be quite as complete as is that of any 

 adult structure. Mead ('98) has stated this conclusion con- 

 cisely in the following sentence, taken from a Wood's Holl lec- 

 ture. "... The component cells of the prototroch are as 

 homologous in the various types studied in the same sense as 

 are the component bones in the skeleton of the vertebrate limb." 

 With this conclusion the cytogeny of the prototroch in Tlialas- 

 sema falls in line. Judging by what we know of the variations 

 that the vertebrate limb may undergo, Mead's comparison is 

 strengthened rather than weakened, I think, by the fact that in 

 some cases cells are added to, and in other cases what seem 

 typical cells are entirely absent from the prototroch. 



Cilia first appear with great regularity at four and one half 

 hours after fertilization and simultaneously on the prototroch and 

 rosette (Text-Fig. 3, A). Those on the former constitute a 

 complete band and are first short and delicate but gradually in- 

 crease in size and strength. On the rosette, however, they grow 

 very rapidly and soon are more than half as long as the embryo 

 itself. The cilia do not arise until about an hour after the forma- 

 tion of the primary prototroch and the rosette, but immediately 

 the after differentiation of the cross. Up to this time the embryo 

 is entirely separated from the ^^'g membrane, just as in Cerebra- 

 tulus(fL. B. Wilson, '00), but by the practically bilateral divisions 

 of protoblasts of the cross a very marked prominence arises at the 

 center of the upper hemisphere, as in Nereis. The rosette and 

 primary prototroch, accordingly, become pressed against the 

 Qgg membrane and thus permit its being punctured by the 



the older stages of Arenicola is ascribed by Child ('00) to the presence of yellow 

 granules. In Tkalassema, however, this color seems to be due to the fact that the 

 protoplasm of these cells holds the picric stain with tenacity. The prototroch, 

 wide at first, becomes narrower as the development proceeds and in the completely 

 differentiated trochophore consists of three rows of greatly elongated cells. The 

 first row consists of II cells, the second of ii cells and the third (incomplete in the 

 dorsal region) of 6 cells. 



