Decembeb 2, 1904.] 



SCIENCE. 



745 



are at first much shortened, perhaps even 

 undeveloped. Thus there is no building 

 outwards of the new organs from the cut 

 edge, but, on the contrary, the differentia- 

 tion is in large part centripetal in direc- 

 tion. Pfliiger thought that the regenera- 

 tion of a new part may take place in some- 

 what the same way that a crystal grows in 

 a saturated solution, but the case that I 

 have just given shows that this conception 

 will not explain the main facts,* although 

 it is not to be denied that to a limited ex- 

 tent the old organs along the cut edge may 

 have some influence on the formation of 

 the new organs besides that of proliferat- 

 ing new cells. 



The same points that are illustrated by 

 the last case are even more strikingly 

 shown in a piece cut off far out to one side. 

 In this case the structures exposed along 

 the cut edge are not median structures, yet 

 the new median organs are here also laid 

 down between the old and the new tissues. 

 The old parts in this case can not be sup- 

 posed to determine the formation of the 

 median organs, as was possible in the 

 former instance, but the entire process of 

 differentiation in the new material is 

 rather centripetal, i. e., from the surface 

 inwards. At least, if the actual differen- 

 tiation is not in point of time from without 

 inwards, the influences that determine the 

 extent and kind of organs that are laid 

 down must be thought of as acting in this 

 way. 



In these pieces that regenerate laterally a 

 new head appears at the anterior end of 

 the new material and a new tail at the 

 posterior end. The position of the pharynx 

 in lateral pieces removed from different 

 levels of the body gives a clue to one, at 

 least, of the internal factors that must be 

 at work. If the lateral piece is from the 



* Pflilger's view does not explain those cases in 

 the regeneration of a head or tail when the new 

 part is shorter tlian the part removed. 



anterior region of the worm, the new 

 pharynx develops near its posterior end; 

 if the piece is from the middle of the worm 

 the new pharynx develops near the middle; 

 and if the piece Is from the posterior region 

 the new pharynx develops near the an- 

 terior end. Thus the location of the 

 pharynx gives us a clue to certain condi- 

 tions present in the new part. An analysis 

 of the results leads to the following con- 

 clusions : 



1. The new material is at every level 

 totipotent, as shown by the fact that a 

 new head will fonu near the anterior end 

 of a piece at whatever level the piece has 

 been removed. 



2. The new material although totipotent 

 is not homogeneous, or, more technically, 

 not isotropic, as shown, for example, by the 

 position of the pharynx. We must con- 

 elude from this that the material is some- 

 what different at every level, and that this 

 difference corresponds in kind to the char- 

 acter of the body at each level. Conse- 

 quently there is in every piece a gradation 

 in the new material from before backwards 

 that gives us the phenomenon that we call 

 polarity. With this difference, or polarity, 

 as a basis the centripetal influence, acting 

 from the surface inwards, determines the 

 organization of the new part.* The action 

 of this centripetal influence is on the new 

 part as a whole, and determines the rela- 

 tive location of each organ. 



By means of these three assumptions — 

 of totipotence, of heterotropy and of organ- 

 ization-power — we can explain the main 

 features in the result. Each assumption 

 is, moreover, a direct deduction from an 

 experiment or observation !f 



This view will also account for the de- 



* This idea of the action from without inwards 

 was formulated by Morgan in 1899, and by 

 Driesch more fully in 1900. 



t The same explanation applies to the develop- 



