Further Examination of the Cell-Theory 201 



of half -size." It thus seems that the mosaic hypothesis for 

 the frog is partly true. Under some circumstances the 

 cells seem more or less like the stones in a mosaic, under 

 other circumstances however they do not. Later work, par- 

 ticularly by Roux, Curt Ziegler, Morgan and Ellen Torelle, 

 has in general confirmed this much modified form of the 

 hypothesis in the case of the frog. It has at the same time 

 shown how much more complicated the whole matter is than 

 Roux's original simple statement would lead one to suppose. 

 But the frog is not the only animal in which the cells of 

 the early embryo present something of the mosaic character. 

 ('. Chun discovered that one of the cells of the two-celled 

 stage of a Ctenophore would develop as a half-embryo if 

 isolated from its mate. Driesch and Morgan confirmed this 

 discovery in general, but pointed out certain details of 

 structure of the resulting organisms which make the latter 

 depart quite fundamentally from half-organisms in a strict 

 sense. For instance, a true ectoderm covered over the side 

 that would be the cut surface were two half-animals to be 

 produced by halving a whole one with a knife. Further, 

 some of the internal organs, notably the endodermal pockets, 

 were not merely what they would be in a half-animal, but 

 were as much like those of a whole animal. The normal 

 animal has four, so a typical half-animal would have two, 

 but the half-animals produced from isolated blastomeres 

 had three. Several later investigations on ctenophore eggs 

 have been carried out, the upshot of all being that with 

 very decided reservations the cells of the young embryos 

 of these animals may be looked upon as the "stones in a 

 mosaic work." A few other kinds of animals, for example 

 the mollusc IlijfinufDni obsoleta show something of the same 

 sort of developmental capacity. 3 



