106 



THE AMERICAN NATURALIST 



[Vol XLI 



all the cells which are utilized in every structure which is meta- 

 merically repeated, the material for the new somite not bein^ 

 budded from any pre-existing somite, but always just in front of 

 the hinder end of the body. 



This applies strictly to all cases which are known to me in the 

 arthropods as well as to most of the annelids; but in a few of the 

 latter group modifications occur in the process which have great 

 interest for us. As is well known in a number of annelids asex- 

 ual reproduction by transverse division occurs. At one or more 

 points in the body a new head may develop with the eyes, append- 

 ages, etc., characteristic of the anterior end of the worm, these 

 features arising from a somite which in its earlier stages is appar- 

 ently normal and like its fellows on either side. Then, just in 

 front of this new head the worm divides and two worms, each 

 with fewer somites than the original one, are produced and from 

 this time onward lead an independent existence. 



Of these only the anterior worm need now be considered. After 

 the separation the segment which was just in front of the new head 

 of course becomes the terminal somite of the new worm. The 

 worm now increases in length and the new somites are formed by 

 material cut off from the terminal somite which thus must have 

 within it the equivalent of the teloblasts of the embryo. 



From these facts it seems logically to follow that at least certain 

 somites in the body have the potentialities of forming material 

 for additional somites and must contain within tlicni the same 

 physi(.loui(;d possibilities as the original teloblasts from which 

 they arise. In other words, in the annelid before the heginning 



was located at more than one point in the body, but it was not 

 exercised until after the asexual reproduction was well advanced. 



In the case of the Naides the somites thus ])ro(Iuce(l are all 

 similar in character but in such instances as ProTnla, where heter- 



by the formation of new somites which ditler in kind. 



I'he ap{)lication of these facts to the variolr^ types of meristie 

 variation which occur in annelids n<-ed not be (hsi nss,.,| hc,.,.^ h^t 

 J think it is apparent that they will in ])art e\|)laiii some of them. 

 I do not mean to say that they reveal first causes but they do point 



