178 C. M. CHILD. 



I find it very difficult to obtain any definite conception of the 

 character of the hydranth-forming substance or of the manner in 

 which it acts. Apparently it is not responsible for the size of the 

 structure but only for its characteristic form. In the further 

 development and discussion of* the hypothesis a large number of 

 special assumptions are made to meet special cases. 



Loeb has recently asserted his belief in the existence of forma- 

 tive substances in connection with regulation and especially as 

 determining polarity. Loeb's view appears to resemble the older 

 Bonnet-Sachs hypothesis of the migrations of particular sub- 

 stances in particular directions. According to Loeb the formative 

 substances are nutritive in character. This hypothesis does not 

 appear to meet the objections to which other hypotheses are open 

 and introduces some new difficulties since conditions determining 

 the localization and distribution of the nutritive material do not 

 seem to be considered. 



On the other hand, various results have been obtained in work 

 along this line which hypotheses of formative substances do not 

 seem to me to account for. To mention only a few of these, in 

 Leptoplana the larger the portion cut off posteriorly the larger 

 the amount of new tissue formed at the posterior end of the piece 

 remaining. Thus a piece comprising only the anterior third or 

 fourth of the body will produce without being fed five or six 

 times as much new tissue as the anterior four fifths. Yet the 

 " tail-forming substances ' and nutritive substances must be 

 much more abundant at the posterior end of the longer piece 

 than at that of the shorter. Again, of two pieces of Leptoplana 

 with posterior ends at the same level one containing the cephalic 

 ganglia produces a larger tail than one without them. Why 

 should this difference occur, since the tail-forming substances are 

 the same in amount at the level of the cut and the amount of 

 nutritive substances must be approximately the same in both 

 cases. In Ccstoplana the new pharynx appears at a given level 

 in a piece containing the ganglia and farther anteriorly in a piece 

 from the same region without the ganglia. 



Objection may be made to the citation of these cases here on the 

 ground that they are not pertinent to the matter in hand, in that 

 they are not connected with the problem of polarity. It seems 



