480 EDWIN S. GOODRICH. 



open the fine branches ending in flame cells, and from which 

 pass segmental ducts to the exterior, corresponding to the 

 segmentally arranged gonads. It is by the breaking up of 

 such a system into separate organs that Lang would derive the 

 nephridia of the higher Ccelomata. 



Unfortunately, we know little about the origin of the ne- 

 phridia in this group. Lang has described, in Discocelis, 

 paired ingrowths of the epiblast, which he believes give rise to 

 the nephridia (70). This observation strongly supports his 

 theory as to the phylogenetic derivation of the nephridia from 

 epidermal glands ; and, indeed, it seems pretty certain that an 

 incipient excretory organ to be efficient must have been 

 derived from, or at all events situated close to, the surface 

 layer in order to get rid of its excretory products. 



It is in the Planarians, a group undoubtedly primitive^ in 

 some respects, that we should expect to discover the ccelora in 

 its first stages of development, and, in fact, we do seem to be able 

 to trace it from its first appearance. In some Accela (Grafi", 

 42, 44), and other simple forms, the gonads consist merely of 

 the genital cells lying freely in the parenchyma. In others, 

 these cells become surrounded by an epithelium formed by the 

 adjacent cells; the epithelial sacs, one on either side of the body, 

 may then become hollow, while the wall grows out to form two 

 tubes, the genitalducts (peritoneal funnels). Another important 

 stage is presented by these organs in Gun da segmentata 

 (Lang, 69). Here the genital follicles are repeated segmentally, 

 the first pair being ovarian, the rest testicular sacs. If these 



' One of the most useful lessons of modern research has been to teach us 

 with what great care the word "primitive" should be applied to auy group of 

 existing animals. A few years ago naturalists readily derived one group of 

 living animals directly from another, apparently more primitive ; but their 

 genealogical trees are novi^ becoming reduced to bushes, in which the branches 

 spring from a common base. Nevertheless, it is true that certain groups 

 may retain, either in their general organisation or in some particular details, 

 characteristics of the ancestors from which they have diverged. The Plana- 

 rians, with their complicated nephridial and genital apparatus, their deeply- 

 sunk nervous system, yet generally archaic plan of structure, are ft striking 

 case in point. 



