572 MONTGOMERY— MORPHOLOGY OF THE [Ap.il 24, 



portion of the primitive nephridium. There is no ectodermal in- 

 vagination," hut the terminal portion comes from a region w^here 

 mesoblast and ectoblast join, probably from a region that v\^as orig- 

 inally ectoblastic. 



The work of Meyer, I^raipont and Woltereck shows that Hat- 

 schek (1878) was entirely wrong in deriving the nephridia from a 

 continuous anlage, and in stating the adult nephridia of Polygordius 

 arise as branches of longitudinal ducts of larval nephridia. 



Larval Nephridia. — There is one pair of these in Polymnia 

 (Meyer, 1887), ^^ch with a long flagellum placed upon the outer 

 surface of the closed inner end, on which region follows a loop com- 

 posed of two cells and then an excretory canal with intercellular 

 cavity ; these persist until the first definitive nephridia function. In 

 Psygmobranchus (Meyer, 1888) there is also one pair, each com- 

 posed of two cells and probably without internal opening, that open 

 on the ectoblast and do not touch the mesoblast ; they belong to the 

 first somite (that just behind the metastomium). Meyer (1887 J 

 has figured the larval nephridia of Nereis as internally closed 

 canals; Hatschek (1885) finds this structural relation in Eupomatus, 

 and holds the nephridia to be mesoblastic. In Hydroides the head 

 kidney opens into the proctodaeum (Wilson, 1890). Drasche 

 (1884, Poniatoccros) held the head kidneys to have funnels, and to 

 be mesoblastic. 



The larva about which there has been the most discussion is that 

 of Polygordius. For P. neapolitanus Hatschek (1878) described 

 the branched head kidney as having open nephrostomes and being 

 joined by longitudinal canals with the trunk nephridia, a condition 

 that has led to manifold comparisons with platodan relations. But 

 Fraipont (1888) and Meyer (1901) found that such longitudinal 

 canals do not exist, and that the inner ends of these tubes do not 

 possess funnels but are beset with slender cells (solenocytes) that 

 project into the blastocoel. Meyer described also a second pair 

 of larval nephridia behind these, which differ from trunk nephridia 

 only in the lack of funnels. Then Woltereck (1905) in disagree- 

 ment with these writers states that the two-branched first pair of 

 larval nephridia belong to the second somite, are mesenchymatous 

 and degenerate entirely ; while the second larval nephridium belongs 



