DYNAMICS OF MORPHOGENESIS 105 



decreases as the level of the piece becomes more posterior in the 

 body; and finally the relative amount of regeneration in the stricter 

 sense as compared with redifferentiation increases at the anterior 

 end and decreases at the posterior end as the level of the piece 

 becomes more posterior. 



These statements hold good only for pieces within the limits of 

 a single zooid. With the passage from the posterior end of one 

 zooid to the anterior end of another there is a marked change in 

 the axial factor or gradient : the pieces from the 'anterior' region 

 of the second zooid, for example, show more anterior charac- 

 teristics than those from the posterior region of the first zooid, 

 which originally lay just in front of them (Child, '11 c, '11 f). 



Of the four regional differences in regulation along the main 

 axis which were mentioned above, the first, the rate difference, 

 the third, the difference in length of the prepharyngeal region, 

 and the fourth, the difference in proportion of regeneration and 

 redifferentiation, are all obviously quantitative in character. 

 The second regional difference as stated above consists in the 

 longer pieces of a difference in the size of the head, also obviously 

 a quantitative difference, but in the shorter pieces of a difference 

 in the character of the head, which, according to some points of 

 view, must be qualitative. As a matter of fact, these differences 

 are also primarily quantitative, for it has been possible to pro- 

 duce experimentally all these axial differences in the character of 

 the head by quantitative changes in external factors, e. g., tem- 

 perature (Child, '11 d; Child and McKie, '11). Experiments of 

 this sort will be considered in another connection. 



In short, the differences in the regulation of similar pieces from 

 different regions along the main axis indicate the existence of a 

 quantitative gradient or gradients of some sort along the axis, 

 but they do not give any evidence for the existence of qualitative 

 .axial differences, i.e., of a graded distribution of substances such 

 as Morgan has at various times suggested. 



It will be shown, however, in section C below and in later papers 

 that the matter is not as simple as it appears from the comparison 

 of pieces in sequence under constant conditions. As a matter of 

 fact a metabolic gradient does exist along the main axi^ of the 



