NEW INTERPRETATION OF CNIDARIA 325 



area between the skin and the intestinal layers. In the second 

 case the middle layer develops as a pure derivative of the 

 primitive intestinal wall (the entomesoderm) usually by means 

 of a formation of diverticula growing out of the intestinal wall. 

 This formation of diverticula partly resembles the formation 

 of lodges that are developed in anthopolyps together with 

 their sarcosepta. Attempts have also been made to derive the 

 formation of the ecterocoele from that of the enterocoele. This 

 situation becomes even more complicated in the ontogenies 

 of numerous higher animal groups, above all of the Arthropoda 

 and of the Vertebrata; thus in the human embryo the mesoderm 

 develops from not less than five different Anlagen. The human 

 body is also developed during the ontogeny from four and even 

 more layers. 



These differences in the type of development of the mesoderm 

 which lead to the formation of secondary body cavities 

 (coeloms), and the differences that can be observed in the type 

 of development of the intestinal orifices— they all occur in the 

 ontogenies —even if they are not radical, have served as a basis 

 for the general classification of the animal groups (Ectero- 

 coelia or Protostomia, and EnterocoeUa or Deuterostomia). We 

 must consider this as an unfortunate solution. The point 

 where this supposed separation or bifurcation of the two large 

 groups had taken place is assigned characteristically by 

 various zoologists into different levels. 



The question must also be raised whether it can a priori be 

 probable that the Eumetazoa, which are considered as a homo- 

 geneous group whose body layers are believed to be mutually 

 homologous, can all have such a common basic part, as the 

 middle body layer is supposed to be, and at the same time that 

 such a part had evolved not only ontogenetically but also 

 phylogenetically along tw^o essential different ways, so that the 

 same form is not even mutually homologous. It seems to us 

 much more acceptable to consider that the middle body layer 

 is the same everywhere, that it is therefore always a homologous 

 formation, whose ontogenetic development, however, has 

 22 



