NEW INTERPRETATION OF CNIDARIA 261 



First, we see that larvae which belong to large, and not in- 

 frequently even to very large, animal groups, can have a very 

 similar form and structure. Secondly, the organization of these 

 larvae is almost always simpler (exceptions, however, can be 

 found almost ever}^vhere in the field of biology); larvae are 

 therefore believed to be closer to their supposed ancestors. 

 Thirdly, not a few cases have been known where there can 

 be no doubt that certain characteristic properties can be found 

 in larvae, or in embryos, that are also know^n to occur in 

 their supposed ancestors; while at the same time these proper- 

 ties can be found no longer, or at least no longer function, 

 in the corresponding recent grown-up forms. 



We do not intend to make here a detailed analysis of the 

 problem as to what can be understood as a larval stage. It is 

 certain that the distinction must be made of very different 

 orders of larvae (primary, secondary, possibly even younger), 

 i.e. larvae can be developed from various stages in the deve- 

 lopment of embryo. This takes place in such a way that any 

 such stage begins to evolve in its own direction, and it adopts 

 its own system of feeding which is usually not parasitic. There 

 are, however, also some parasitic larvae; the problem remains 

 open whether there exist any sessile larvae: and it seems that 

 such larvae can be found among the Echinodermata, the Penta- 

 crinus stage of the Comatula. A necessary transformation 

 by means of which the animal endeavours to reach the organ- 

 ization and the way of life of the adult form is the consequ- 

 ence of such a development. A radical metamorphosis fre- 

 quently becomes necessary in these cases. 



We are finally able, after prolonged and repeated discussions, 

 to make a more uniform evaluation of the morphologic 

 significance of larvae and to use it correctly for phyletic 

 purposes. At all events, it can be stated that similar larvae had 

 been evolved quite independently along several lines of evolu- 

 tion by sessile sea animals. Convergence has played an 

 important role; it was brought about by the adaptation to 

 a planktonic w^ay of life. It can be maintained that one of the 

 18 



