ANATOMY AND CLASSIFICATION OF THE ARENICOLTD^. 519 



third of A. cristata is collected into vessels whicli open into 

 the parietal vessel or into the nephridial vessel. The remain- 

 ing nephridia of these three species return blood to the sub- 

 intestinal vessel (PI. 24, figs. 26, 30). 



The blood whicli passes through the nephridia of A* 

 Grubii, and all except the last two nephridia of A. ecau- 

 data, is returned to the longitudinal nephridial vessels, which 

 are much more highly developed in these two species than in 

 the other three species (PI. 25, figs. 44, 45, and PI. 26, figs. 

 53, 54). 



Nephridia of Post-larval Stages. — A. marina. — 

 As in the adult, there are six pairs of nephridia, opening 

 near the fourth to the ninth neuropodia. The first nephri- 

 dium is in some cases smaller than the others (see Benham, 

 1893, p. 52). We can confirm Benham's account of the 

 structure of these organs of the post-larva in all particulars. 

 The nephridium is a simple, slightly bent tube, opening 

 anteriorly into the coelom by a small round or oval aperture, 

 which is 'quite simple, i.e. there is no funnel. The nephi'i- 

 dium is of almost uniform diameter throughout, and its 

 walls are formed of a single layer of columnar or cubical 

 ciliated cells covered by an exceedingly delicate layer 

 of coelomic epithelium. In our specimen (4'5 mm. long) 

 the cells of the nephridium are apparently all alike, but 

 Benham found in a post-larva {6'8 mm. long) probably a 

 little older than ours that the cells of the anterior part of 

 the nephridium contain excretory granules Avhich are not 

 present in the cells of the posterior portion. In our specimen 

 4" 5 mm. long the nephridia are '12 mm. to *15 mm. in length, 



along its course from the ventral vessel to llie neplirostome. It is generally 

 whitish, but sometimes pinkish in colour in life, and may readily be mistaken 

 (as it has also been by Milne Edwards and Cosmovici) for a blood-vessel with 

 thick walls. Such, however, is not the case ; it is a strand of connective tissue 

 homologous with the strand occurring in exactly the same position in A. 

 cristata and A. Grubii. Thus the first efi"erent vessel passing to the sub- 

 intestinal vessel in A. marina, is in the segment behind the one which bears 

 this deceptive connective-tissue strand, i. e. in the first gill-bearing segment. 



