68 THE EVOLUTION OF THE METAZOA 



form of one or several circlets. It would take us too long to 

 give here a complete survey of all animal groups which have 

 polypean forms. The species of whole large groups can show 

 polypean forms (this is the case with Cnidaria and Entoprocta 

 among the Ameria), or the same thing is true for at least a large 

 part of them (Annelida). Oligomeria are mainly polypoid ani- 

 mals. Even among the Chordoma we can find in the benthonic 

 Ascidia a clear inclination to develop polypean characters. Im- 

 portant from our standpoint, however, is the knowledge that 

 a sessile radially symmetric polypean form had been repeatedly 

 developed from a freely moving bilaterally symmetric form, 

 and that this polypean form itself has been later frequently 

 changed into a freely moving form, either one that creeps over 

 the sea bottom, e.g. Echinodermata, or one that can swim 

 freely in water, e.g. medusae of the Cnidaria. 



The Validity of Watson's Rule for the Invertebrates 



In our attempt to place the origin of Cnidaria on a new, 

 rational basis we can use the experiences of researchers in Verte- 

 brata who have certainly had to solve much easier problems— 

 even if we do not intend to enter here in to the whole enormous 

 problem of the origin of new animal types, i.e. macro-evo- 

 lution. The question is whether Watson's rule (de Beer, 1954) 

 of a mosaic -like evolution can be considered valid also in the 

 case of the "lower" invertebrates. A detailed study of the 

 Archaeopteryx (a third specimen of this animal has recently been 

 found) has led de Beer to develop the ideas which had 

 been first formulated by D. M. S. Watson and according to 

 which there is a divergent evolution in the subclasses (types) 

 of Vertebrata and a simultaneous mosaic-like mixing of old 

 and new characteristics. In this way a new combined form can 

 be created with a simultaneous change of the way of life, which 

 is stil able to live in the milieu. Thus, there had been no general 

 and parallel change of characteristics, (in this case of the 



