igoS.J EXCRETORY ORGANS OF METAZOA. 551 



two excretory canals, found in all the others (though the relations 

 are not yet known for the Polyclades). 



3. Nemertini. 



From the comprehensive treatment given by Biirger (1895), 

 based largely upon his own researches, it follows that the excretory 

 organs are as a rule in the form of two main canals parallel with 

 the lateral blood vessels and not communicating together; each 

 opens to the exterior of the body by one, or more rarely by a series 

 of several (up to about forty), excretory ducts; the main canals are 

 usually restricted to the region of the stomach, but in some genera 

 they extend the length of the body. From them proceed delicate 

 capillaries that terminate blindly in multicellular " Endkolbchen " ; 

 the latter may project into the walls of blood vessels, but (contrary 

 to the earlier opinion of Oudemans) there is no open communica- 

 tion of any portion of the nephridia with these vessels or other in- 

 ternal cavities. In the freshwater Stichostemnia I showed (1897) 

 that an unusual condition obtains, in that in the adult instead of a 

 single canal on each side there is a series of them, some with and 

 some without excretory canals; and then Bohmig (1898) demon- 

 strated that the latter are produced by a secondary segmentation 

 of originally continuous ducts. ^ Punnett (1900) and Coe (1906) 

 found in Tcsniosoma besides excretory pores opening on the surface 

 of the body others that connect with the oesophagus ; the latter are 

 clearly embryonic ducts persisting in the adult. 



The larvae do not possess special excretory organs. The defini- 

 tive ones arise, according to Burger, as a pair of hollow evaginations 

 of the ectoblastic stomodseum of the larva, soon abstrict from the 

 oesophagus and then open into the amniotic cavity at a ventral point 

 near the mouth, a position quite different from that of the adult 

 excretory pores. The origin of the latter is not known, and 



^ I had described the terminal bulbs of this genus as closed from the 

 capillaries, with an internal cuticular lining but no flame, while Bohmig 

 found them essentially as described by Burger except that each consists of 

 usually not more than two cells. I have recently had opportunity to ex- 

 amine living material and to compare it with my former sections, and find 

 I had overlooked the true flame cells and that Bohmig had described them 

 correctly. Each terminal bulb consists of from one to five cells. 



