80 C. 0. Whitman 



is not improbable tliat one of the two modes of reproduction exbibited 

 in dipby genie individuai« may be introduced by fecundatiou (p.58 — 59). 

 In four instances niodified forms of Infusoriforms bave been found 

 in Nematogens (p. 59). Altbongb it is quite certain tbat these Infuso- 

 riforms did not originate in tbe Nematogens, it is not clear vs^liether they 

 are to be regarded as stragglers, or males (Van Beneden) that had pen- 

 etrated for the purpose of feeundating. The former supposition appears 

 to me tbe more probable. 



14. The first cleavage, introduced karyokinetically, splits the ver- 

 mific germ-cell into two unequal parts. The two-cell stage is followed 

 by one of three cells, and this by one of four. Three of the four cells 

 form a cap to the fourth (tìg. 71), and gradually envelope it by epibolic 

 expansion. A Gastrula is thus formed consisting of a small number of 

 ectodermic eells and a single central endodermic cell (figs. 76 and 77). 

 Periods of growth intervene between the successive cleavages, the indi- 

 viduai cells attainiug, each time, approximately the size of the original 

 germ-cell. 



The blastopore closes , and the multiplication of cells at this pole 

 soon leads to the pyriform embryo (figs. 80, 81, 87). The pointed end 

 of this embryo is an elougation from the blastoporal region ; the broad 

 end corresponds to the future cephalic pole. The endodermic cell lies at 

 first wholly in the broad end , but gradually elongates backward be- 

 tween the caudal cells. The whole number of cells seen in the adult 

 are present before the embryo attains its definitive form (p. 63). 



15. Two primary germ-cells arise at the two poles of the endo- 

 derm , the first (always the posterior) about the time the pyriform stage 

 is reached (fig. 86). The nucleus of the endoderm participates only in 

 the formation of these two cells, which give rise to other germ-cells by 

 division. The manner in which the germ-cells arise in the germogen 

 has not been ascertained ; but it may be safely assumed that the nucleus 

 of the germogen participates in their formation, at least in the first in- 

 stance. For remarks on endogenous cell-formation, see p. 65. 



16. Comparing the development of the vermiform embryo with 

 the development of the male Orthonectid (Julin), Dicyemids may be 

 Said to have a transient triploblastic stage, represented by an ectodermal 

 layer, one axial endodermic cell, and two mesodermic cells (primary 

 germ-cells) derived from the two poles of the endoderm. (Vid. p. 75.) 



