Dustin (1910) has repeated the investigation of Allen on Chrysemys tnarginata, and now confirms Allen's observations regarding the ento dermal origin of the extra-regional germ-cells in this form. Dustin still claims, however, that in this form also there is a secondary peri toneal source of origin of germ-cells or "gonocytes." But his illustra tions do not seem to support this conclusion. The so-called "secondary gonocytes" as illustrated are smaller than certain other germ-cells des ignated "primary gonocytes," but no unequivocal transitions appear between the mesenchyme cells of the sexual gland or its peritoneal cells and the secondary gonocytes.
In a postscript (p. 33) Allen (1911) records preliminary results in the case of Nedurus and Amblystoma similar to those reported by Dustin for Triton, namely, an actual mesodermal origin of sex-cells. Allen disposes of the conflicting views regarding the origin of germ-cells in amphibia by the claim that in the urodeles the usual source of origin of the germ-cells is the mesoderm; that in the anurans the source of origin is the entoderm; at the same time, he admits the possibility of exceptions.
A similar mesodermal origin of the sex-cell is described also for the teleost Lophius (Dodds, 1910). Wheeler's results on Petromyzon (1899) indicate that the germ-cells may become included in the meso derm at the time it separates from the entoderm. Jarvis (1908) reports an entodermal origin of the primordial germ-cells also in the horned toad, Phrynosoma cornutum.
Gatenby, however, admits " that some germ-cells possibly migrate from the entoderm of the yolk-sac," but "feels sure that during life very large additional reinforcements of germ-cells arise in the epithe lium of the gonad of Amphibia" (p. 276). His criticism, then, of the work of King, Allen, Beard, and others who hold to the view of a genetic independence between ccelomic epithelium and germ-cells is that they have studied only the earlier periods of the ontogeny. But
in the male chick Swift has now (1916) examined also the young post incubation stages, as late as 10 days, and still finds all the evidence in accord with the segregation view.
Gatenby describes the tadpole ovary as formed of (1) germ-cells of peritoneal origin; (2) germ-cells of retro-peritoneal origin; (3) germ cells of endoderm (yolk-sac) origin. He inclines to believe that the germ-cells of the yolk-sac are all laid at the first spawning. He de scribes and illustrates intermediate stages between peritoneal cells and young germ-cells, but one is forced to the criticism that his figures give no adequate support to his claim of a transformation of mesoderm-cells into germ-cells.
In the lizard embryo (Lacerta agilis), according to von Berenberg Gossler, the migration of the "primordial germ-cells" is not so strongly limited to the early developmental stages as described, e. g., for Chrysemys; and the cells (" entodermal wandering cells") are relatively much more numerous; nor do any of them disintegrate; they become converted into mesoderm-cells. "The migration of cells out of the entoderm in Lacerta has no other significance than a dilatory develop ment of mesoderm from entoderm" (pp. 247-248). He extends this interpretation to cover the so-called extra-regional genital cells of reptiles and mammals. As mesodermal cells they may possibly metamorphose into cells of the ccelomic epithelium, as earlier main tained by Hoffman. They wander from the entoderm largely to the location of the future pronephric (Wolffian) duct, to the construction of the caudal portion of which they greatly contribute. Occasional cells are found also in the somatopleure. None of the cells while within the mesoderm were found in process of mitosis. All these cells are believed to become changed into mesoderm-cells; thus they apparently disappear, but from some of those scattered among the cells of the peritoneal epithelium definitive germ-cells arise. This interpretation eases the difficulty of accepting the view of certain investigators (e. g., Firket, Dustin, Gatenby, Felix) that both primary and secondary genital cells are formed, the primary becoming displaced through " degeneration " by the secondary. The whole genital gland would then arise from mesoderm, as maintained by some of the earlier investigators.