ON THE CCELOM, GENITAL DUCTS, AND NEPHRIDIA. 501 



which communicate with the exterior. The gonads develop 

 from the wall of the follicles ; in the lower forms a large 

 number are fertile, but in the higher forms the genital products 

 become restricted to a very limited region. 



No true nephridia have been discovered in this group ; for, 

 until the development of the interesting segmental tubules 

 described by Weiss (103) and Boveri (14) in Amphioxus is known, 

 it is not possible to decide for certain on their nature. 



Hemichorda or Enteropneusta. — In the anterior or pro- 

 boscis region is a coelomic cavity which appears to represent a 

 fused pair of follicles (fig. 23), as is evidenced by the fact that 

 it communicates with the exterior by paired peritoneal funnels 

 (ciliated proboscis pores) in Balanoglossus Kupfferi and 

 B. canadensis, and occasionally in Ptychodera minuta and 

 B. Kowalevskii (Spengel, 96 and 97), and by its development 

 as abilobed sac in B. Kowalevskii (Bateson, 2).^ The collar 

 region contains a second pair of coelomic follicles, also provided 

 each with a peritoneal funnel or collar pore (Bateson, 2; Morgan, 

 82; Spengel, 97). Behind these is a third pair of follicles 

 which do not develop funnels, but become of large size and 

 encroach on neighbouring segments, sending back blind poste- 

 rior prolongations. The posterior region contains a series of 

 fertile coelomic follicles, the gonads (fig. 23), each provided with 

 a peritoneal funnel leading to the exterior.- Although the 

 metamerism of this region is not very definitely pronounced, 

 possibly having become obscured through degeneration, the 

 genital follicles bear a remarkable resemblance, both in their 

 structural relations and in their arrangement, to those of the 

 Nemertines, as has already been noticed by Schimkewitsch (93 

 and 93a). As in the Nemertines, so also in the Enteropneusta, 

 several genital follicles may occur in one segment. 



1 Compare the development of the corresponding anterior pair of cavities in 

 Amphioxus (Hatschek, 49). 



^ If this explanation be correct, the gill-slits of Balanoglossus are meta- 

 meric and interseguiental as in other Vertebrates. If not correct, we have an 

 almost incredible state of things in which the genital cells are shed into a 

 cavity which is not the ccelom. 



