400 THE EVOLUTION OF THE METAZOA 



planktonic larvae, and to a polymerization of various organs. 

 One could even say that if the Nemertinea had not their 

 characteristic proboscis and only slight indications of a peri- 

 gastrocoele we could easily consider them as Archiannelida. 

 These are not Polymeria, they are genuine Ameria which had 

 probably evolved from some ancestors which stood close 

 to the Turbellaria and which had served as a starting point 

 for the evolution of the primitive Annelida and therefore of 

 all the Polymeria. The perigastrocoele thus evolved, as did 

 all the coeloms, from some "sHts" that had developed in the 

 middle layer, in the mesohyl (mesoderm), and which were 

 filled with the cell fluid or with the body fluid; and not from 

 some sack-like formations of the intestine. All connection 

 (such connections are not numerous) of coeloms with the 

 intestine are secondary phenomena. Those zoologists will there- 

 fore oppose my interpretation (and such zoologists are quite 

 numerous) who firmly accept the enterocoele theory, in spite 

 of the fact that they meet wdth difficulties which cannot be 

 overcome as soon as they try to derive the conditions that 

 can be observed in the Polymera (Ecterocoelia) from those 

 that can be found in the Oligomeria (Enterocoeha). 



It must be admitted that it is possible to imagine a state 

 with a uniform perigastrocoele as the first step in the develop- 

 ment towards a segmented perigastrocoele. Yet at the same 

 time w^e see that among the recent groups of the Eumetazoa 

 there is not a single group, at least as far as I know, which 

 would represent such a state. Animals which show a state that 

 is at least rather similar to that with a uniform, general peri- 

 gastrocoele can be found on the periphery of the Polymeria 

 only, and above all among the OUgomeria. As regards the 

 reduction of segmentation we can find as early as in the 

 Annelida, species which are more or less atypical, and there- 

 fore aberrant, and which clearly show a kind of a retrogression 

 of the perigastrocoele system. Such a reduction is even more 

 advanced in the Sipunculoidea and in the Echiuroidea. It is 

 therefore completely wrong to place the Endoprocta close 



