CONTROL OF HEAD-FORM IN PLANARIA 105 



fused cephalic lobes may appear at the tip as in the teratomorphic 

 head. Histological examination shows that this outgrowth is 

 really a head, for i1 contains a rudimentary cephahc ganglion, 

 and its behavior also indicates its characteristic in most cases. 



In the headless form (fig. 7) the new tissue simply fills in the 

 contracted cut surface and does not grow out beyond the con- 

 tour of the piece and no ganglion is present. 



These different types of anterior end are in reality somewhat 

 arbitrary groupings of the members of a graded series between 

 the normal head and the headless condition and represent dif- 

 ferent degrees of development of the cephalic gangha. In stand- 

 ardized material they occur with a characteristic frequency 

 which depends as I have already shown on length of piece and 

 region of the bodj from which it is taken (Child, '11 a, '11 b). 

 But this frequency can also be altered and controlled in a great 

 variety of ways both through the physiological condition of the 

 worms and by the action of external factors. In this and fol- 

 lowing papers some of the methods of control and their re- 

 sults are described. 



III. CONTROL AND MODIFICATION OF HEAD-FORM AND 

 HEAD-FREQUENCY BY MEANS OF CYANIDE 



The following tables give in percentages some of the character- 

 istic results obtained with KCN, the conditions of experiment 

 being gii^en in connection with the table in each case. Experi- 

 ence has shown that with properly standardized material and 

 conditions the limit of error in regions of the first zooid is not 

 greater than 10 per cent in lots of fifty pieces each and is probably 

 often less than this. In other words, differences of more than 

 ten per cent between control and experimental lots are cer- 

 tainly the result of the experimental conditions and not of un- 

 controlled factors. Since the differences to which attention is 

 called are usually greater than ten per cent there can be no doubt 

 of their significance. In the region of the posterior zooids the 

 error is greater because the lengths of different zooids differ 

 somewhat in different individuals and a piece from this region 



