572 ON INCREASE IN COMPLEXITY [pt. iii 



mination may from one point of view be regarded as the point 

 where the curve of decreasing totipotence reaches the base-line. For 

 totipotence persisting into later stages, such as blastula and gastrula, 

 demonstrates its presence by the capacity of any piece of tissue to 

 develop into structures which it never would have done if it had 

 been left to itself The name usually given to this first-period in- 

 determinateness is "plasticity". Thus before gastrulation Spemann 

 found that a piece of epidermis normally destined to fold in to form 

 the ectodermal nerve-cord can be exchanged for a piece of ordinary 

 epidermis with the result that the presumptive nerve-cord will turn 

 into skin and the presumptive skin into nerve-cord. Such plasticity 

 occurs all over the embryo before gastrulation, and Spemann was 

 able, in some very beautiful experiments, to work with the dark- 

 coloured Triton taeniatus and the light-coloured Triton cristatus to- 

 gether so that dichromic grafted embryos were produced. But if one 

 attempts to make such transformations as these after gastrulation 

 has been completed, one meets with no success whatever — even 

 though no differentiation at all is visible, the hidden process has 

 determined the fate of the cells. 



In the preliminary plastic cells even the germ layers are inter- 

 changeable. Pieces of presumptive ectoderm can be planted just 

 beneath the dorsal lip of the blastopore, and being carried in as 

 invagination proceeds can afterwards be recognised as taking part 

 in the constitution of mesoderm or endoderm. In other words they 

 behave "Ortsgemass", i.e. according to the arrangements of their 

 host, and not "Herkunftsgemass", i.e. according to the arrange- 

 ments of their origin. Such pieces can be earmarked by the use 

 of vital stains, as Goodale in his work against the concrescence 

 theory was the first to show, and Vogt has in more recent years done 

 many experiments of this nature. The converse, namely, that pre- 

 sumptive endoderm or mesoderm can give rise to ectoderm, has been 

 demonstrated by Mangold, and similar remarks hold with respect 

 to the limb buds. 



The fact that the organiser of one kind of Triton can bring about 

 its effects in another kind of Triton naturally led to the question of 

 what taxonomic distance was necessary to prevent the action of the 

 organiser. Geinitz, working on this point, found that the organiser 

 of even very different species such as Amblystoma or Bombinator could 

 be grafted into Triton embryos with efficient results. The grafts might 



