142 THE EVOLUTION OF THE META20A 



I have shown it to be very probable that in the Ctenophora 

 the last remains of protonephridia have still been preserved, 

 even if their condition as well as their function have been 

 considerably changed. These remains have developed the form 

 of the so-called "ciliate rosettes" and they correspond, in my 

 opinion, to the terminal organs of the typical protonephridia 

 which have been grown over by the intestinal wall (Hadzi, 

 1957, see Fig. 24). In this way they have re-established their 

 contact with the external wcrld by way of an intestinal lumen 

 instead of the lost contact by way of pores. 



A very interesting case is that of Nematoidea. They are 

 characterized by the fact that they can no longer produce 

 undulipodia, not even in connection w4th androgametes. 

 ■Correspondingly there are no terminal organs in Nematoidea 

 so that here the excretory function has been taken over by 

 a device which consists of two cells. An eventual homology 

 of this device with protonephridia is more than questionable ; 

 it is in all probability a new substitute form. 



It should be briefly mentioned that in the Annelida among 

 Polymeria, where the general polymerization has also influenc- 

 ed the excretory organs that developed into metanephridia, 

 the nephridia have been reduced either to a sole pair or to 

 an unpaired nephridium. This has been partly due to their 

 transition to a semi-sessile or sedentary way of life, and partly 

 to their extensive diminution. This is true even to a greater 

 extent for Oligomeria that have adopted the sessile way of 

 life; in these we can find developed either individual nephridia 

 or none whatever. Even among the sessile Chordata, the 

 Tunicata, no real nephridia have been preserved. 



If we presuppose, though cannot, strictly speaking prove it, 

 that protonephridia have been secondarily lost in Cnidaria 

 because of their sessile way of life, we will find such an inter- 

 pretation strongly supported by analogies to numerous other 

 cases, where it is immediately clear that protonephridia had 

 been reduced or abandoned under similar circumstances. In all 

 these cases we find the emunctory part of the nephridial 



