468 ALFRED GIBBS BOUENE. 



merely rounded (I.e., c.) and others much flattened (I.e., a.), 

 they present a nucleus [n.) and very often a lumen (/.) ; these 

 cells represent either separate funnels or more probably 

 lobes of a simpler funnel. They are set upon the walls of 

 a dilatation corresponding to the dilatations in Clepsine and 

 Pontobdella, and as in those genera containing a sort of 

 debris. In Clepsine we have seen a very simple funnel to 

 be present, two cells taking part in its formation, in Pontob- 

 della more cells take part in its formation, and it becomes 

 somewhat lobed, and in Nephelis and Trocheta this is 

 carried to a still greater extent, and it seems very probable 

 that the structure found in Hirudo is the same thing as the 

 funnels of the other genera, only that this division into lobes 

 has been carried to the extreme, and at the same time, although 

 it is a point exceedingly difficult of determination, the commu- 

 nication between the lumen of the tubules and the coelomic 

 sinus seems to have been obliterated. It is interesting to note 

 in this connection that the nephridium of Hirudo possesses a 

 much fuller blood-supply than the nephridium of the genera 

 which present a better developed funnel. 



Following upon the funnel is the '' testis lobe " (woodcut, 

 fig. 7, b — c). This is throughout a spongy mass, the ductules 

 being very irregularly arranged. At c the cells of the *' testis 

 lobe " are continuous with those of the " main lobe " (c — d). 

 It is in these cells that the ductules become so remarkably 

 branched, the main ductule receiving collecting ductules (Cf. 

 my figure in ' Quart. Journ. Micr. Sci.,' vol. xx, PL XXIV, 

 fig. 5). The ductules collect together, those from the ^^erminal 

 portion {d — e) as well as those from the main portion (c — e) , 

 at e ', the lumen simply perforates the three or four rows of 

 cells which constitute the portion e — f} At / these cells join 



' The term " recurrent lobe," which I proposed for this portion, may still 

 be used, for although we now know the lumen to be simply continued from 

 funnel to external aperture without any recurrent portion in the sense in which 

 I previously used the term, yet the ductules in question piss from e, to the 

 apex g, and then return to e. I now speak of the lobe e—f,&s the "recurrent 

 lobe;" the ductules running from e to g, as the "recurrent ductules," aud 

 the portion of the duct returning from ^^^ to e as the " recurrent duct." 



