20 IVILSOX. 



to expect that the annehds will ultimately be shown to agree with 

 the mollusks in showing reminiscences of the ancestral mode of 

 development in the double origin of the mesoblast. 



Returning now to the mollusks, Wierzejski, in a recent pre- 

 hminary paper (1897) states very explicitly that in PJiysa a 

 part of the mesoblast is derived from two cells of the tJiird 

 quartet/ This result, if well founded, giv^es good reason to 

 suspect that the third quartet may give rise to mesoblast in 

 some of the polyclades, as Lang has maintained for Discoccelis. 

 In Lcptoplana I have sought carefully for evidence of such a 

 process, but thus far without success. This negative result is, 

 however, inconclusive owing to the difficulty of tracing the later 

 history of the individual cells. The first division of the third 

 quartet is vertical to the surface (Fig. 6, C^ and in later stages I 

 have thus far found no evidence that a delamination of meso- 

 blast occurs. Soon after the delamination of mesoblast in the 

 second quartet, all of the ectoblast-cells forming the lips of the 

 blastopore become much flattened (Fig. 6, F)^ while the ecto- 

 blast-cap rapidly extends downward, the blastopore finally clos- 

 ing at or near the lower pole. In these stages the outlines of 

 the thin ecoblast-cells are very difficult to see, either in life or in 

 preparations, owing to the confusion produced by the underlying 

 deutoplasm-spheres, now much increased in size, on which they 

 are moulded. The mesoblast now forms four groups of 

 rounded granular cells conspicuously seen through the trans- 

 parent outer cells. A study of the successive stages proves that 

 the greater number of these are derivatives of the second quartet ; 

 but the possibility remains that some additions may have been 

 made from the third quartet. 



From the foregoing account it appears that the '' mesoblast " 

 of the polyclade is derived from the ectoblast ; and it may, I 

 think, be taken as a fair working hypothesis that this " meso- 

 blast " is represented in the mollusks, and probably also in 

 some annelids by cells ('' larval mesenchyme," etc.) derived from 

 the second quartet {IJnio^ Crcpidida, Aj'iciaQ)) or perhaps in 



1 Confirmed by Holmes in the case of Pianorbis since the aI)ove was written. 

 See Science, VI, No. 154. 



