244 GASTRULA-ARRESTED EMBRYOS 



amphibian presumptive notochord is self-differentiating with re- 

 spect to the morphogenetic movement of convergent extension, 

 for notochordal explants from a young gastrula will elongate in 

 the proper axis within a reasonable time, and so will similar ex- 

 plants mounted suitably upon an endodermal substrate. 



On the other hand, hybrid presumptive notochord explants 

 l)ehave abnormally ( Gregg and Klein, 1955 ) ; instead of elongat- 

 ing, they merely round up into balls of tissue and remain that way 

 as long as they are cultured. Now, if autonomous convergent 

 extension of notochordal tissue is a necessary concomitant of nor- 

 mal gastrulation, its failure to occur constitutes a sufiBcient con- 

 dition for the nonoccurrence of normal gastrulation. Thus, in its 

 failure, we have a partial explanation for the peculiar course of 

 post-blastula hybrid development. But we also have a generation 

 of further puzzles. It will be recalled that the initial value of the 

 epiboly angle for hybrid embryos (138°) is much greater than 

 that for R. pipiens embryos (106°). Moore, as we have seen, re- 

 gards this as evidence that considerable epiboly precedes dorsal 

 lip formation in hybrid embryos. But that epiboly is one compo- 

 nent (extension) of convergent extension. If the latter is not au- 

 tonomous in hybrid embryos, then the question arises, how is the 

 initial pre-invagination epiboly accomplished? At present, there 

 is no clear-cut answer. 



It is evident, from Moore's description, that a hybrid invagi- 

 nates enough to move its presumptive head endoderm in under- 

 neath its presumptive chordamesoderm. When this relationship 

 is mimicked in vitro by mounting a bit of hybrid chordameso- 

 derm on an endodermal substratum, the chordamesoderm not 

 only fails to elongate normally, it also spreads upon the substra- 

 tum, in all directions, like ectoderm ( Gregg and Klein, 1955 ) . If 

 this sort of thing happens in situ, then it is easy to see why the 

 notochordal convergence necessary for normal gastrulation does 

 not occur, and why invagination is brought to a sudden halt. 



Thus, by considering the behavior of explant systems in rela- 

 tion to known features of hybrid development, we can construct 

 a fragmentary explanation in morphological terms for some of the 

 peculiarities of that development. We turn now to consider some 



