570 MONTGOMERY— MORPHOLOGY OF THE [April 24, 



19. POLYCH^TA. 

 Adult Nephridia. — There is usually one pair to each trunk seg- 

 ment. In the Phyllodocidse, Glyceridse and Nephthyidse their inner 

 ends are closed and the loops are beset with solenocytes, each of 

 which is a cell projecting into the body cavity " containing a deeply- 

 staining rounded or oval nucleus, attached by a sort of neck to the 

 extremity of a thin tube which opens at its opposite end into the 

 lumen of the nephridial canal . . . Working inside the tube and 

 attached at its distal end is a single long flagellum, which passes far 

 down the nephridial canal " (Goodrich, 1900). In the other families 

 the inner end is open to the coelom, with the exception of Poly- 

 gordius (Hempelmann, 1906) where the first pair is closed. The 

 nephridiopores usually open separately. Each pair of nephridia 

 stands in relation to two segments in Archiannelids, Alciopidae, 

 Typhloscolecidge, certain Nereids (Eisig, 1887), Terebelloids and 

 Cirratulidae (Meyer, 1887), Aphroditidae (Darboux, 1900) and 

 Disomidse (Allen, 1904) ; in the other families, therefore in the 

 majority of species, to only a single segment. Some of the main 

 deviations from this type are the following: 



(a) In Capitellids each nephridium may have several nephro- 

 stomes, there may be several pairs to a segment and they may dis- 

 charge into the skin and not on the surface of the latter (Eisig, 

 1887). In Laiiicc and Ploimia the fourth segment possesses two 

 pairs (Meyer, 1887). 



{h) In the Terebelloid Lanicc conchUcga the three anterior pairs 

 of nephridia connect with a pair of longitudinal canals from each 

 of which a single nephridiopore discharges on the surface; while 

 the four following pairs of nephridia open into a longer pair of 

 posterior canals which end blindly at about the sixteenth thoracal 

 segment, and each of which discharges by four nephridiopores. 

 Ploimia presents quite similar relations. ^leyer (1887) who de- 

 scribed these conditions holds it probable that the longitudinal canals 

 are formed secondarily by a meeting and fusion of separate 

 nephridial loops, incipient stages of which are to be noted in other 

 genera. Also in Ozvenia (Gilson, 1894) do the nephridia open into 

 longitudinal canals, that are here described as formed by an infold- 

 ing of the epidermis. 



