CELL-LIXEAGE. 15 



show a remarkable similarity, in all the forms, up to a certain point. 

 In morphological value, however, the micromere-quartets of the 

 polyclade appear to differ radically from those of the annelid- 

 moll usk type. In the former the first quartet is described as 

 giving rise to the entire ectoblast, while the second and third 

 quartets are mesoblastic.^ In the latter, on the other hand, 

 these same three quartets give rise to ectoblast, while, as stated 

 above, the main mass of the mesoblast is derived from a single 

 cell (the posterior) of a fourth quartet of which the other three 

 cells form entoblast (Fig. 5, B). If a fifth quartet is formed it 

 is invariably entoblastic (Fig. 2, A). 



At the time attention was first called to these differences it 

 seemed hopeless to reconcile them. Later researches showed, 

 however, that the discrepancy was not so great as it seemed. 

 Lillie first discovered in 1895 that in the lamellibranch Unio one 

 cell (the left) of the second quartet give rise to mesoblastic ele- 

 ments (the "larval mesenchyme")^ and more recently Conklin 

 has found a similar derivation of mesoblast-cells from three cells 

 (right, left and anterior) of this quartet in the gasteropod Crc- 

 pidiila} 



It is clear that these interesting discoveries partially bridge 

 the gap between the polyclade and the other forms ; though 

 how great it still remains may be judged from the fact that 

 Conklin still regarded the differences as " very great, perhaps 

 irreconcilable," ^ while Mead, in a still more recent work on the 

 cell-lineage of annelids, is forced into a position of skepticism 

 regarding Lang's whole account of the origin of mesoblast in 

 the polyclade.'^ 



For these and other reasons a re-examination of the early de- 

 velopment of polyclades has become in the highest degree de- 

 sirable. After a search extending through several years, I have 

 at length succeeded in finding a form very favorable for this 

 purpose — a species of Lcptoplana ^ having eggs that are large 



iLang, 1884. *Crepidu/a, p. 196. 



2 Unio, p. 24. 5 1897, p. 289. 



3 Crepidiila, p. 150. 



6 An undetermined species found in great profusion at Port Townsend, Washing- 

 ton, on Puget Sound. 



