1904.] NATURAL SCIENCES OF PHILADELPHIA. 727 



X. — The Phylogenetic Relationships of Dinophilus in the 

 Light of Its Early Development. 



In this chapter I wish to briefly compare the early development of 

 Dinophilus with that of other groups, and to consider what light a 

 study of its cell-lineage sheds on its systematic position. That the 

 study of cell-lineage is of value in determining relationship has been 

 abundantly shown by the results accomplished in this particular branch 

 of zoological research. These have in a most striking degree corrobo- 

 rated the results attained by the study of comparative anatomy, and 

 have further shown that in a large number of forms, representative of 

 large and important groups, the characters of the various forms as 

 manifested in the cleavage are as constant as the anatomical characters, 

 and must therefore be as truly inherited. Furthermore, since coeno- 

 genetic changes may be supposed to affect the later stage of develop- 

 ment first, we may expect to find the earlier stages retaining longer their 

 primitive characters, although even the earliest stages have been 

 affected by precocious segregation and are no longer highly primitive. 

 The study of the early development cannot be regarded as a sure or 

 certain guide in determining relationship in every case, yet it may, 

 I think, be very properly called to aid in the determination of the 

 relationships of doubtful forms. In the case of Dinophilus, the cleav- 

 age has presented such startling and accurate resemblances to the 

 chsetopod annelids that it seems impossible that they do not indicate 

 relationship, for so many and minute correspondences could hardly 

 have arisen independently. I have already stated these resemblances 

 separately, but wish here to bring them together, in order that their 

 force may be more apparent. 



In the first place the manner of origin of the germ layers — ectoderm 

 arising from the first three quartettes, mesoderm from the left posterior 

 member of the fourth quartette, and entoderm from the remaining 

 cells — brings Dinophilus into a list of forms already large and still 

 increasing, containing members of the Lamellibranchia, Gasteropoda, 

 Polychaeta and Echiuridse. In common with many Annelida and 

 Mollusca the larger part (or all) of the ectoderm of the trunk is 

 derived from one cell, the posterior member of the second quartette. 

 A little later the resemblances to the annelids become more marked. 

 In the development of the second preoral ciliated band of Dinophilus 

 is recognized the prototroch of the trochophore, and the earliest rudi- 

 ment of the brain appears at the same point as the annelid "Scheitel- 

 platte." There are, however, in the cleavages themselves resem- 

 blances to the cleavage of the polychajte annelids which are most 



