172 PROBLEMS OF RELATIVE GROWTH 



are particular morphological expressions may continue through- 

 out life. He and his pupils have most thoroughly demon- 

 strated this for hydroid polyps and planarian worms. By this 

 work, certain important empirical laws have been established, 

 notably the fact that the first region to differentiate in re- 

 generation normally acts as a ' dominant ' region which has 

 a morphogenetic effect on the regions which differentiate later. 

 This has now been brought into line with the facts concern- 

 ing Spemann's ' organizer ' in Amphibian development (see 

 Santos, 1929). Further, such experiments as Stockard's cele- 

 brated production of cyclopia in Fundulus, cannot be inter- 

 preted except in terms of axial gradients. Many of Child's 

 empirical facts have been independently conformed by Abeloos 

 (1930), and the existence of a dominant region with morpho- 

 genetic effect by various workers, of whom Berrill (1931) is 

 the latest. (See also the work of Buchanan, p. 260.) 



In regard especially to hydroids and planarians, Child has 

 been able to show that the physiological gradients constitute a 

 true field system, e.g in regenerating fragments of Planaria the 

 gradient can be either steepened (e.g. by optimum tempera- 

 ture) or flattened (e.g. by cold or by narcotics) : and when 

 this is done, it is found that the dominant region whose activity 

 has been depressed induces smaller dependent organs, at a 

 smaller distance than normal from itself, while the converse 

 holds when it has been heightened. Analogous experiments 

 have been performed on developing eggs, by using stimulatory 

 or depressant drugs, or by applying temperature-gradients. 



The continuance of the gradient throughout life is shown 

 in many forms by their graded capacity for regeneration, e.g. 

 the head-frequency often decreases steadily in an antero- 

 posterior direction. There exists also a graded susceptibility 

 to poisons along the main axis. In forms which bud or divide 

 by transverse fission, the distance between the dominant 

 regions of old and new zooids appears to be determined by 

 the extent and steepness of the gradient. The production of 

 axial heteromorphosis in regeneration, such as biaxial heads 

 or tails, can also be satisfactorily interpreted in terms of the 

 gradient hypothesis. Child has attempted to explain these 

 facts in relation to differences in metabolic rate, but the proof 

 cannot yet be said to be conclusive. However, whether or 

 no this metabolic interpretation be correct, a set of impor- 

 tant empirical principles remain — notably that physiological 

 gradients exist in early stages of development, that they may 

 persist throughout life, that their slope and extent are deter- 



