57 



mata . there is not sufficient evidence to decide. It may 

 be r'ossible that both of these factors act ir, deterr-inin^ the 

 shar^e of the segienting -lass of cells. And doubtless the 

 metibraneless character of the ejy.c^ nlays a -art in thene 

 phenomena 



PLANULA. 



V/hen segmentation is coTdete a solid efrbryo is forr.ed 

 v^hich r.ay at first be called a ncrula. Small snaoes occur 

 soroeti'-es between the blastcneres during the di^fferent 

 cleavage stages, but they are sooner or later obliterated 

 by the crov/ding together of the cells, A central cleavage 

 cavity v/hich is later transformed into a blastocoele is not 

 forred; consequently a true blastula does not exist in 

 the development of Turritcnsis , In this resnect it differs 

 very "-arl^edly from S tor^otoca and the majority of hydro- 

 medusae of which the develonment has been studied, in 

 which a definite blastocoele is formed that becomes filled 



