746 



SCIENCE. 



[N. S. Vol. XX. No. 518. 



velopment of a head at the anterior end 

 and a tail at the posterior end of a cross 

 piece of planaria, or of lumbriculna, or 

 of hydra. In the planarians a head forms 

 on the anterior end, because, on my view, 

 the new material there is continuous with 

 the new material throughout the old part 

 and the difference in the character of the 

 parts at each level gives the basis on which 

 the organizing power acts. Similarly for 

 the tail development at the posterior end. 

 In Liimbriculus the same conditions hold. 

 In both Cases the old tissue may be so little 

 difEerentiated that its regional differences 

 may also furnish the basis for the organ- 

 izing power to act. Very short cross-pieces 

 of Planaria maculata often produce a head 

 at each end. This result may be due, as 

 I have explained elsewhere,* to the absence 

 of sufficient difference between the ends on 

 which as a basis the organization of the 

 posterior end can take place. It must be 

 assumed in this case that the external and 

 internal factors, in the new part, have a 

 stronger influence in calling forth a new 

 head than a new tail— much as in Tiibu- 

 laria. In hydra no new tissue is produced 

 at the cut ends, but the old part molds itself 

 into the typical form. Here the old tissues 

 are so little fixed that the organizing power 

 acting along the lines of regional difference 

 molds the old part into a new whole. In 

 this respect the difference between hydra 

 and the other forms is merely relative. 



My view differs in many points from the 

 stuff hypothesis. It assumes no specific 

 stuff apart from the living material, and 

 consequently it makes no assumption of 

 the migration of such stuffs in multicellular 

 organisms ;t it assumes that the material 



* ' The Control of Heteromorphosis in Planaria 

 maculata,' Roux's Archiv, XVII., 04. 



f rt has been shown in the case of the egg of 

 many forms (frog, sea-urehin, crepidula, ascidian) 

 that movements of the protoplasm may occur. 

 The differentiation that results appears to follow, 

 to some extent at least, the regional differences in 



at each level has in addition to its toti- 

 potence something also of the material 

 basis' characteristic of that level. Lastly, 

 my view takes into account the organizing 

 power of the living material which builds 

 up its structure independently of cell 

 boundaries on the basis of the totipotence 

 and heterotropy of the new part. It is 

 needless to point out further differences of 

 minor importance. 



Another example of lateral regeneration 

 brings into the foreground the character 

 of the organizing factors that are at work. 

 If the arm of a salamander is cut off near 

 the body new material appears over the cut 

 end, and while the new material is still 

 relatively small in amount (compared with 

 the amount removed) a new limb, includ- 

 ing parts of all the structures, is laid down. 

 The humerus does not complete itself, and 

 then the other parts form in order of suc- 

 cession from within outwards, but simul- 

 taneously the new material is propor- 

 tionally subdivided or segregated into the 

 typical elements. Subsequently the new 

 parts all grow larger and longer, vintil the 

 new limb reaches the size of the one on the 

 other side. 



The migration hypothesis is not needed in 

 this case where no alternative exists to ac- 

 count for the location of the new part, and 

 it is helpless to explain the phenomenon of 

 segregation that takes place when the dif- 



the protoplasm. Here we are not dealing with 

 formative stuffs, in the original sense of the term, 

 but with differences in the kinds of protoplasm, 

 or perhaps only with quantitative differences, 

 which become relegated to different cells. It is 

 presumably the same differences of protoplasm in 

 the cells of the fully formed animal that furnishes 

 the basis for the regional differences in the new 

 and old material in antero-posterior regenera- 

 tion, etc., and which gives the basis on which 

 the organizing changes go forward. In multi- 

 cellular forms there is no extensive evidence of 

 migration from cell to cell of the different kinds 

 of protoplasmic materials, nor any necessity of 

 making the assumption that they do migrate. 



