THE AXIAL GRADIENTS IX HYDROZOA. 423 



('04) that the time between cutting and completion of the pol\ p 

 is shorter the more apical the piece. But according to Lund the 

 time between the beginning and end of the outgrowth is the same 

 at all levels. By defining the regeneration period as the time 

 period during which the outgrowth is elongating, Lund is able 

 to reach the conclusion that the "rate of regeneration" is the 

 same at all levels. It may merely be pointed out that adopting 

 some particular definition in no wise alters the facts of the matter, 

 which are the same for Obelia as for other ccelenterates. Lund 

 is able to state that the rate of regeneration does not differ at 

 different levels only because his definition of the expression "rate 

 of regeneration" differs from that used by other workers. The 

 point raised by Lund that the time interval from the beginning 

 to the end of the elongation process is the same at all levels 

 may be correct but it does not seem to me to be proved by his 

 tables and graphs. However, it is difficult to come to any 

 decision on the matter, as neither the time of beginning of growth 

 of the apical pieces nor of completion of growth of the basal 

 pieces is given. It is very probable, nevertheless, that the chief 

 differences in regeneration rate at different levels lie in the early 

 part of the regeneration period. It is admitted by Lund and is 

 shown by his tables and graphs that the rate of elongation of the 

 outgrowth is faster the more apical the level. Billard ('04) had 

 previously made a similar observation ; he noted that the sum of 

 the lengths of the outgrowth at both ends of each piece is greater 

 the more apical the piece. In the face of his own data, Lund 

 still attempts to maintain the conclusion that the rate of elonga- 

 tion is the same at all levels on the assumption that the rate of 

 elongation decreases apico-basally because the mass of the pieces 

 decreases in the same direction. He assumes without any proof 

 whatever that the amount of tissue regenerated is proportional 

 to the mass of the piece. Such an assumption is incorrect. As 

 already pointed out considerable differences in length of pieces 

 have no effect on either the time required for regeneration or 

 the amount of tissue produced in that time; and such slight 

 differences in amount regenerated as are correlated with differ- 

 ences in diameter are counterbalanced by the more rapid rate of 

 regeneration of pieces of smaller diameter. All of the available 



