PT.ra,sECT.3] ON INCREASE IN COMPLEXITY 603 



by Assheton, who referred to them as primary and secondary centres 

 of cell-proHferation, and this point of view has been adopted by many 

 embryologists, e.g. Eccleshymer; Adelmann; and Kingsbury. 



In addition to these researches on embryos, Child has studied 

 the gradients during the development of the sea-urchin egg {Arbacia 

 punctulata), those of polychates {Nereis, Chaetopterus and Arenicola) and 

 those of an ascidian [Corella willmeriana) . In these cases, the primary 

 simple apicobasal gradient of the fertilised egg-cell was succeeded 

 by a double gradient resulting from the appearance of a zone of 

 high susceptibility at the posterior end of the embryo. Mention of 

 the work of Bellamy on the amphibian egg brings us back to our 

 starting-point, namely, the recent investigations on the organiser and 

 the mechanics of amphibian development. There is no need to 

 give a detailed description of Bellamy's results on the frog embryo, 

 for they resembled in many ways those of Hyman on the brook 

 lamprey. But they may be briefly summarised for the purpose of 

 comparing them with the work of Spemann and his school. 



Bellamy found that in the unfertilised amphibian zgg the beginnings 

 of polarity were to be found in the position in the ovary. He observed 

 by injections and by actual observation of blood-flow that the blood- 

 vessels to the eggs in the oogonia pass arterially over the pigmented 

 part of the egg and venously over the unpigmented part. It is more 

 than probable that the first polarity of the egg arises because the 

 animal pole is that point on the surface of the egg which happens 

 to be most well supplied with a capillary network, not that which 

 happens to be attached to the ovary by the pedicle. The initial 

 physiological gradient, therefore, would seem to be a matter of 

 position in the ovary. It may be mentioned here that very similar 

 conclusions were come to by Lillie for Chaetopterus and Sternapsis 

 eggs, by Child for Phialidium (hydromedusa) eggs, and by Boveri and 

 Jenkinson for Strongylocentrotus eggs. 



The early stages of development in the frog's egg are very resistant 

 to toxic agents. But it was possible to show that the fertilised but 

 undivided egg began to disintegrate at the animal pole and the 

 degeneration passed downwards and outwards with a special bias 

 towards the grey crescent. Much the same state of affairs was 

 seen in the 4-cell stage, but in the morula stages there are two 

 zones of high susceptibility, the second one appearing just above 

 the grey crescent, and contributing to the general spread down- 



