308 FIELDS AND GRADIENTS 



But the precise method by which the dominant region exerts its 

 morphogenetic control over the rest of the field is still unknown. 



However, an experiment may be described here which not only 

 illustrates the importance of quantitative potential difference, but 

 also throws light on the problem of determination in regeneration. 

 In the fresh-water Annelid worm Lwnbriculus, if the hindmost fifth 

 of the body is cut off, a head will be regenerated from the front edge 

 of this piece in 90-95 per cent, of cases. In a second series of ex- 

 periments, a small piece containing two or three segments is cut off 

 in such a way that its anterior edge is at precisely the same level on 

 the long axis of the worm as the anterior edge of the whole hindmost 

 fifth in the first series. These small pieces of the second series only 

 regenerate a head in 20-30 per cent, of cases. It might be supposed 

 that this lack of power to develop a head was due to insufficiency of 

 material in the small piece, but this is not so. If a hindmost fifth 

 of the worm is cut off as before, and then, 20 hours later, a large 

 piece of this be removed so as to leave a piece identical in size and 

 in level with that used in the second series of experiments, it is 

 found that a head will be regenerated in 70 per cent, of cases.^ 



Lack of power to develop a head in the second series of experi- 

 ments is therefore not due to lack of material, for the pieces of the 

 third series are of the same size as those of the second, but can re- 

 generate a head almost as well as those of the first series. The only 

 difference between the pieces of the third and second series is that 

 for 20 hours the anterior end of the pieces of the third series has 

 been in continuity with the whole hindmost fifth of the worm, and 

 this period of time is apparently long enough for the qualitative 

 determination of a head to be effected, as in the first series. After 

 this determination, reduction in size of the piece does not hinder 

 head-production. The fact that the act of cutting raises the activity 

 of the old tissue in small pieces more than in large pieces where the 

 cuts are farther apart and the stimulation consequent upon them 

 has to act on a much larger mass of material. The anterior edge of 

 small pieces will therefore have more difficulty in obtaining the 

 necessary threshold potential difference for head-determination. 



Experiments in every way analogous to those just described on 

 Lumhriciilus have been performed on Planaria^ and with similar 



^ Hyman, 1916. 



