IV. 2 INTERACTIONS OF THE PARTS 273 



and the suggestion is made by Herbst that these movements are 

 to be interpreted as so many responses to stimuli, which are then 

 not merely directive but formative, the stimuli being either 

 external or exerted by other parts of the body. Thus the 

 spreading of the blastoderm over the yolk may be a case of 

 oxygenotaxis, the movement of cells of the Frog's blastula 

 towards one another described by Roux l as Cytotropismus may 

 be due to a mutual chemotactic stimulus, the migration of cells 

 to the surface of the Arthropod ovum is perhaps oxygenotactic, 

 the vitellophags are chemotactic, the application of mesoderm 

 cells to the outgrowing axis cylinders of nerve fibres, of connective 

 tissue and muscle-cells to blood-vessels, of the choroidal cells to 

 the optic cup (they are absent ventrally when the choroid fissure 

 fails to close), the reunion of separated blastomeres in Triclads 

 and others, may all be tactic phenomena of a positive kind, while 

 the dispersal of the elements of a complex may be negative. 

 Similarly, the outgrowth and anastomoses of nerves, glands, 

 ducts, the concrescence of layers may be tropisms of various sorts. 



The suggested stimuli in some cases proceed from the external 

 environment, in others they are exerted by one part on another. 

 The former have already been discussed, but it may be pointed 

 out again that the effect called forth depends both on the stimulus 

 and on the reacting organism or organ. Neither in these cases 

 nor in any others is the production of specific form due to the 

 action of an external agency upon a homogeneous material. For 

 even in those instances where the fate of some apparently in- 

 different tissue seems to be decided by the action of such an 

 agency as in the production of stems from whichever side of an 

 Antennularia stock happens to be uppermost (Loeb) it is evident 

 that all that the external agent can do is to decide between 

 a limited number of definite alternatives in respect of each of 

 which the parts of the organism are equipotential. Responses 

 to stimuli of the second kind possibly play an important part 

 in ontogeny. Unfortunately, the experimental evidence for this 

 is at present somewhat meagre. 



In the fish Fundulus the yolk-sac is deeply pigmented, the 

 pigment-cells being especially closely aggregated round the 



1 See above, Ch. II. 1, p. 55. 



