144 F. H. STEWART. 



resemblauce to an unmodified skin-gland. Jagerskiold (10) has 

 pointed this out also, stating that physiologically at least the 

 ventral gland and the skin-glands are interchangeable. The 

 results given above agree with this; in the female of Onclio- 

 laimus vulgaris the functions of the ventral gland are 

 taken over presumably by the vulvar and lateral line glands, 

 as the animal reaches maturity. 



It is hardly necessary to insist on the structural resemblance 

 between the ventral gland on the one hand and the tail- 

 glands and glands of the lateral lines on the other. 



The Ccelom. 



It seems probable that the chief reason why a coelom has 

 not been recognised in Nematodes is to be found in the 

 manner in which the reproductive organs have been described. 

 The terms " testis " and " ovary " have been made to cover, 

 not only these organs themselves but also the spaces which 

 contain them, and although the spaces have of course been 

 recognised, sufficient morphological importance has not been 

 attached to them. 



The condition may be summarised as follows : In all Nema- 

 todes hitherto described there is to be found either a single 

 median space or two bilateral spaces, lined by a characteristic 

 flat epithelium. These spaces with their walls I have de- 

 scribed as testicular and ovarian regions. A specialised 

 l^atch of this epithelium forms the testis or ovary. From the 

 space, or spaces, a duct, or ducts, leads to the exterior : where 

 there are two ducts they unite before reaching the exterior. 

 In regard to origin, the cavity and, no doubt, the ducts as 

 well are mesodermal. This is the view of Jainmes (7), and in 

 Hallez's (6) figures it can be seen that the gonad tubes arise 

 from two groups of cells which in the early embryo lie in the 

 blastococle, one on either side of the gut. 



The cavities are, then, typical gonococls or protococloms 

 (11, pp. 35, 3G), and the ducts (unless they should prove to be 

 formed by ingrowth of e})idermis, which there is no reason to 

 expect) are typical gonoducts or coclomoducts. 



