68 EICHAED EVANS. 



loop, is not differentiated, but when the secondary commu- 

 nication has been formed the canals in question are lined re- 

 spectively by cells which are quite different in character. 

 The portion situated between the receptaculum seminis and 

 the ovary is lined by short cells with no definite cell outlines, 

 and with their nuclei placed either in the centre or near their 

 free end ; but the portion situated on the other side of the 

 receptaculum seminis is lined with columnar cells, which 

 possess sharp cell outlines and nuclei placed at their base 

 (PL 9, fig. 26 e). This difference is probably prophetic of 

 the different functions which these lining cells have to per- 

 form later on in life; the columnar cells of the uterine part 

 having to provide the developing young with the enormous 

 amount of food material necessary to enable them to grow to 

 the unusual length of twenty-seven millimetres. 



The Male Grenital Ducts. — The male genital ducts de- 

 velop in the same way as the oviducts, up to the stage at 

 which they acquire an opening to the exterior, except that 

 their inner ends do not unite with each other (PI. 8, figs. 

 19 a and 19 h). After the formation of the external pore, the 

 short median portion, which alone is ectodermal, elongates 

 and forms a loop, which is usually placed on the left side of 

 the rectum, in the middle chamber of the body-cavity. Oc- 

 casionally, however^ the loop in q uestion may pass under the 

 left nerve-cord. In the former case the point of union of 

 the vasa deferentia to the common duct is placed in the 

 median chamber, and the right vas deferens does not pass 

 under the left nerve-cord; but in the latter case the point of 

 union of the two ducts is drawn to the left side of the cor- 

 responding nerve-cord, and the right vas deferens passes 

 under both cords, just in front of the external pore. It seems 

 from Balfour's description of P. cap en sis (1) and von 

 Kennel's account of P. edwardsii (4) that the condition least 

 prevalent in the genus Eoperipatus is almost universal in 

 these species. It would seem that there is no doubt that the 

 condition described by Balfour and von Kennel is the derived 

 one; the primitive condition being the one found in the 



