ORIGIN OF THE MESODERM. 115 



the greater part of the ventral surface, in Peripatus it is 

 confined to the part of the ventral surface behind the anus. 



I have elsewhere (No. 49) stated my reasons for agreement 

 with Balfour's view, viz. that such a method of mesoderm 

 formation is probably to be regarded as a modification of 

 archeuteric diverticula, such as are found in Amphioxus, &c. 

 Whether the origin of mesoderm from the walls of archenteric 

 diverticula is a primitive process or not is open to grave 

 doubt. 



It seems to me there is a large body of embryological facts 

 which suggestj at any rate, the view that the mesoderm arose 

 as a result of the differentiation and rearrangement of certain 

 of the nuclei of the amoeboid central mass of the ancestral 

 parenchymella or gastrula; that is to say, the facts seem 

 to suggest the following as a possible general view of the origin 

 of the three layers of the Triploblastica. 



(a) Starting with a large multinucleated Protozoon, the 

 first advance consists in the differentiation of a cortical layer 

 of nuclei, and of the protoplasm governed by them, into a 

 peripheral layer or ectoderm. This layer was possibly of 

 a plastic nature, and allowed the protrusion of the central 

 mass at one or more points. The central mass would, in 

 consequence of its large size, probably be capable of arrang- 

 ing its vacuoles into a series of thoroughfares through itself 

 from one opening on the surface to another, so that the intro- 

 duction of nutritive matters to its deeper parts would be possible. 

 On the analogy of the Platyhelminth excretory system we may 

 imagine that the protoplasm of these tracts would acquire the 

 property of throwing out vibratile processes into this system of 

 channels for the purpose of assisting in an efi"ective circulation 

 of the external medium through the body. Such an animal 

 would consist, then, of an ectoderm and a central multinucleate 

 mass which, with Metschnikoflf, we may call the meso-endoderm. 



(b) The next change would consist in the differentiation of 

 the nuclei of the meso-endodermic mass into two kinds : 

 (a) those governing the protoplasm lining the diflferentiated 

 vacuoles ; and {b) the remainder, which would gradually dif- 



