562 MONTGOMERY— MORPHOLOGY OF THE [April 24, 



nephridia and are in connection with the coelom (genital chamber), 

 and since the development of the nephridia is unknown. 



The mid-gut has also been demonstrated to be excretory 

 (Schimkewitsch, 1884). 



17. HiRUDINEA. 



Adult Meganephridia. — There is a series of separated pairs, less 

 numerous than the somites. Nephrostomes may be lacking as in 

 the case of five out of the seventeen pairs of Hirudo (McKim, 

 1895) the three most anterior pairs of Nephelis (Graf, 1893), ^^^^ 

 all of Branchellion (Bourne, 1884). Leuckart (1894) discovered 

 the anatomical connection of the nephridia with the nephrostomes, 

 and this has been corroborated by Voinov (1896), McKim, Graf 

 and Schultze (1883), in opposition to the results of Bolsius (1892) 

 that the " organes cilies " have no connection with the loop. But 

 even when they are connected there need not be an open communi- 

 cation between the two (Graf, 1899). When present the funnel 

 lies in the segment preceding that of the loop. The cavity of the 

 nephridia is much branched and intracellular. An excretory bladder 

 may be present as a part of the excretory duct, but this is lacking 

 in Clepsine. 



The nephridia arise from segmentally arranged mesoblastic 

 nephroblasts, that lie deep below the embryonic epidermis. Each 

 of these divides into two cells, the anterior of which gives rise to 

 the funnel and the posterior to a cord of cells that forms the secre- 

 tory portion of the loop ; the cavity into which the nephrostomes 

 open is a true coelom ; the excretory ducts and vesicles are ecto- 

 blastic ingrowths (Burger, 1891, 1894, 1902, Bergh, 1891, McKim, 

 1895). Burger is very positive with regard to the mesoblastic 

 origin of the nephridia, in opposition to the earlier view of Whit- 

 man (1887). 



Adult Plcctoncphridia. — Bourne (1884) first found net-like 

 nephridia in Branchellion, PontobdcUa and Piscicola; in Pontobdella 

 they consist of a network of canals extending from the ninth to the 

 nineteenth segment, with ten pairs of nephridiopores, while in 

 Branchellion they have only one pair of such openings. They have 

 been redescribed by Johansson (1898), and I am acquainted with 



