398 THE EVOLUTION OF THE METAZOA 



perigastrocoele which had afterwards evolved by way of oligo- 

 merization into a uniform cavity. Both these types of evolution 

 are a priori possible and probable. 



A comparison of all the methods of coelom formation 

 shows— we exclude here in principle all the tubular and 

 cavernous formations which had evolved either by way of 

 an infolding of the skin, as is the case with the tracheal 

 system of the land-inhabiting Arthropoda, or by way of a 

 folding of the skin layer, as is the case with the peribranchial 

 cavity of the Chordata— that all these "hollow spaces" (tu- 

 bules or cavities) emerge in those parts of the mesohyl where 

 w^e can find either streaming of the fluid or rhythmic move- 

 ments of the contractile organs. The first group includes the 

 phenomena of the blood vessel system and the system of 

 lymphatic and w^ater ducts; the second group includes the 

 rhynchocoele, the cardiocoele and, obviously, the perigastro- 

 coele. A third group consists of the gonocoeles in the gonads 

 together with their gonoduct, and the glands wdth central 

 cavities and their secretory canals. 



Thus the coelom cavities develop above all in combination 

 with the frequently repeated or rhythmical movements of the 

 internal muscular organs. We think that it was the peristalsis 

 of the intestinal tube— after a mesenchymal muscularis had 

 been added to the entodermal tubes — which led to the 

 formation of a perigastrocoele. At the same time the intestine 

 became lengthened as a consequence of a lengthening of the 

 entire body. Another consequence of this lengthening of the 

 body was also a radical change of the type of movement 

 which now took place by means of wave -like zigzag contrac- 

 tions of the skin muscle tube which progressed from 

 the front of the animal backw^ards. Such an interpretation has 

 already been suggested by various other zoologists and it is 

 therefore not an invention of my ow^n. 



We must now^ study as closely as possible the factual ma- 

 terial and try to interpret it correctly in order to be able to 

 determine w^hich of the tw^o possibilities is most feasible. So 



