66 C. M. CHILD 



The contrast in head-frequency between the long and short pieces 

 is strikingly shown in these two lots. It must be remembered 

 that in the two cases the level of the body concerned in head- 

 formation is the same, or as nearly as possible the same, yet in 

 the long pieces 100 per cent of heads are formed and in the short 

 pieces none. It may be added further that in pieces of inter- 

 mediate length all possible intermediate percentages of head- 

 frequency are found. 



Comparison of Lots I to IV of the short pieces shows essentially 

 the same result as table 1 above. In Lot I all remain headless, 

 in Lot II, 68 per cent, in Lot III, 32 per cent and in Lot IV, 4 

 per cent. The table also shows that not only does the head- 

 frequency increase but that a marked approach to the normal 

 in the character of the head occurs, as the length of time during 

 which the short piece remains as an anterior portion of the long 

 piece increases. In Lot II, 20 per cent of the pieces form heads 

 with eyes, in Lot III, 40 per cent, and in Lot IV, 96 per cent. 



The only possible conclusion from these two series is that the 

 factors which determine whether an isolated piece shall give rise 

 to a head or not, begin to act almost immediately after the opera- 

 tion; that within 3 hours after section, the determination between 

 'head' and 'headless' has already occurred in about 50 per cent 

 of a lot of pieces under the usual conditions and that this determi- 

 nation has occurred in practically 100 per cent within twelve 

 hours after section. 



It also appears from the two tables that the determination of 

 the character of the head formed occurs somewhat later than 

 the determination whether a head shall be formed or not. At 

 3 hours and even at 6 hours, anophthalmic as well as normal 

 and teratophthalmic forms appear in considerable percentages 

 (table 2) and even at 12 hours the percentages of normal heads 

 is only half that in the controls. At 18 hours, however, most 

 heads are determined as normal (Lot IV, table 1). 



These and other similar series in which worms of approxi- 

 mately the same size and physiological condition were used and 

 in which external conditions are as nearly as possible uniform, 

 all give similar results. Under these conditions, head-formation 



