BY DR. R. VON LEXDENFELD. 215 



in the Protoplasme, that I separate them in accordance with Claus 

 from the other Hydromedusee. 



The process described above, by which the Medusae became 

 more and more rudimentary, was apparently advantageous only 

 for one series of Hydromedusae, whilst in another series we find 

 the Hydroid colony becoming more and more rudimentary 

 inasmuch as the nourishment is more and more assimilated by the 

 Medusa stage and less by the Polypes. 



This process which is indicated in Zygodactyla may go so far 

 that the Polype stages are finally done away with altogether, and 

 a Medusa is produced direct from the ovum without the inter- 

 mediate stage of a Polype. In this manner the change of 

 generation is done away with altogether and we have Medusa? 

 before us, which, although probably descended from Hydroid 

 ancestors, show no trace of such a descendance, and now only the 

 comparative anatomy of these forms proves that they belong to 

 our Hydroniedusae. Such Medusae are very numerous. Clatcs 

 combines all these forms in his suborder Trachomedusse, which 

 corresponds with our fourth suborder Trachoniedusinse. 



The Trachotnedusinse comprise Haeckel's Trachomedusae and 

 Narkomedusee, which correspond to our families Trachomedusidae 

 and Narkomedusidae. 



It appears to me that the total neglect of those Hydromedusai, 

 which do not produce free Medusae, in Haeckel's System der 

 Medusen is not true to nature. If Haeckel considers the 

 Lucernaridae as part of the Acraspaedaa which they doubtlessly 

 are, one would also expect that he would include the Hydroid 

 Polypes in his Craspedotae. And I think I shall follow out 

 Haeckel's ideas better by placing them there than by separating 

 them from those Hydromedusae which produce free zoids. 



The difference between Craspedotae and Acraspedae is so great 

 that it seems advantageous to separate these two groups entirely. 

 In this point I agree with Claus. The great similarity between 

 the two is only analogy and not homology. At all events the 

 Hydroid Zoophytes are much more nearly related to the 

 Ciaspedote Medusae than the latter to the Acraspedae. 

 15 



