16 BULLETIN OF THE 



To my mind it is unquestionable that the parent cells of the rhabditi 

 are of ectodermic origin, as first suggested by Hallez ('79, p. 7). It is 

 only in Triclads and in Rhabdocoels that the mother cells lie in the 

 deeper tissues, and we know so little about the embryology of these 

 gi'oups that we cannot tell just how the passage from the exterior takes 

 place. I have endeavored to show that the cells have a connection with 

 the hypodermis in the earliest stages of their development, long before 

 they show any traces of rhabditi, but whether the cells pass from the 

 hypodermis through the basement membrane, or are separated from the 

 hypodermis before the formation of such a structure, I cannot say. The 

 epidermis of embryos of Mesostomum was found by Graff ('82, p. 56) 

 to be filled with rhabditi, while he could find no traces of the sub-hypo- 

 dermal parent cells so prominent and abundant in the adult. In Poly- 

 clads, the development of the rhabditi is in my opinion identical with 

 that in Triclads ; but in the former the parent cells lie permanently in 

 the hypodermis, whereas in the latter they sink down below that layer, 

 where greater opportunity for growth is afforded. The condition found 

 in Polyclads, therefore, I believe to be the primitive one. 



Another mode of origin of the parent cells of the rhabditi has been 

 proposed by Loman ('87, p. 69), who considers them to be modified 

 connective-tissue cells that migrate from their original positions in the 

 mesenchyma and pass bodily through the basement membrane, and 

 come to lie eventually between the cells of the hypodermis ; or, in the 

 words of the author, "Nach meiner Meinung sind die Stabchenzellen 

 niesenchymatose Gebilde, die eine factische AVanderung durch dass sie 

 umgebende Bindgewebe unternehmen, wahrend ihr Inhalt sich zu den 

 fadenformigen Stabchen ausbildet. Endlich treten sie durch die Basal- 

 membran (wovon spater die Eede sein wii-d), drangen sich zwischen die 

 Zellen der Oberhaut," etc. Thus according to Loman the parent cells 

 form a part of the hypodermis, and only differ from the conditions 

 found in Polyclads in that their epidermal position is a secondary one. 

 Loman presents no evidence, and in the face of the facts here presented 

 his position is untenable. 



Rhabditi are being constantly discharged from the epidermis during 

 the life of the individual, and provision must be made for their renewal. 

 Parent cells are therefore being continually produced to supply the 

 steady demand of the epidermis for rhabditi. The evidence of this lies 

 in the fact that in individuals of all ages these cells are found in all 

 stages of development. lijima ('84, p. 373) says " es sicher scheint, 

 dass die Rhabditen nicht ausgestossen werden." If the rods are not dis- 



