584 Child, Driescli's harmonic equipotential systems in form -regulation. 



the length of the primordium occourred with decrease in the length 

 of the piece, and that this decrease in length of the primordium 

 amounted to about 50 ^/^ when the length of the piece was double 

 the length of the typical hydranth-primordium. He also showed 

 that the two tentacle-areas of the primordium maintain approximate 

 proportionality during this decrease. He did not show, however, 

 that the length of the hydranth-primordium decreases proportionally 

 to the length of the piece, or even approximately proportionally, 

 yet he asserts, as quoted above, that in pieces below a given length 

 the lengths of the whole primordia in different pieces are approxi- 

 mately proportional to the lengths of the respective pieces. His 

 experiments showed simply that in pieces below a given length de- 

 crease in length of the piece is accompanied by decrease in the 

 length of the primordium formed in it, but they did not even show 

 that such decrease was approximately proportional; in fact mea- 

 surements show clearly that it is not proportional, though Driesch 

 did not make use of the facts which concern this point. 



In my study of Tubularia I found from a considerable number 

 of measurements that the length of the hydranth-primordium de- 

 creases less rapidly than the length of the stem in all cases 

 (Child, 1907 f, p. 289, Table III). Moreover, I found, on exami- 

 nation of Driescli's measurements, that they showed exactly the 

 same thing. I have given the results of Driescli's measurements 

 in Table IV of my paper (Child, 1907 f, p. 289). They show very 

 clearly for all groups compared that the length of the hydranth- 

 primordium decreases less rapidly than the length of the piece. 



In Driescli's recent criticism of my work (Driesch, 1908^), 



1) So far as a critical consideration of this paper is desirable it will be found 

 in the following pages. Some considerable part of it, " however, consists either of 

 reassertion of the authors views, without actual criticism of the results of my work, 

 or of condemnation of that work, and to these portions no reply is necessary. 



In support of my earlier criticism (Child, 1907 e) that Driesch had failed 

 at various points to state both the results of my experiments and my conclusions 

 correctly, it is perhaps worth noting that Driesch's second attempt to state the 

 results of my work on Leptoplana (Child, 1904a, 1904b) is a failure like his first. 

 His first statement was: ,,Nach Child nimmt bei Leptoplana das Eegenerations- 

 vermögen nach hinten zu ab" (Driesch, 1905b, p. 694). I called attention to 

 the incorrectness of this statement (Child, 1907c, p. 142), and pointed out that my 

 experiments on Leptoplana showed that posterior regeneration is qualitatively com- 

 plete at all levels (I should have added, posterior to the cephalic ganglia), but is 

 quantitatively incomplete in the absence of food: that anterior rsgeneration, however, 

 is always qualitatively incomplete at all levels posterior to the cephalic ganglia, as 

 I had stated in my original paper. In his reply to my criticisms Driesch corrects 

 his earlier statement as follows: „Nach Child nimmt bei Leptoplana das Ver- 

 mögen, die qualitativ richtigen Regenerate quantitativ vollständig auszubilden, nach 

 hinten zu ab, wenn keine Nahrung gereicht wird (Driesch, 1908, p. 408). This 

 Statement is not even correct for posterior regeneration, for as a matter of fact 

 posterior regeneration in the absence of food is much more nearly complete quanti- 



