214 PROP. HU5LBT ON THE 



epiblast or not. Kowalewsky's evidence, however, is in favour of 

 the origin of the muscles directly from the cells of the mesoblastic 

 diverticula. 



The brilliant investigations of Johannes Miiller upon the de- 

 velopment of the Echinodermata, confirmed in their general fea- 

 tures by all subsequent observers, have proved, first, that the 

 ciliated embryonic Gastraa (the primitive alimentary canal of 

 which is formed by involution of a vesicular blastoderm), to which 

 the egg of all ordinary Echinoderms gives rise, acquires a mouth 

 by the formation of an aperture in the body-wall distinct from 

 the primitive aperture of the Gastrcea, so thafc, in this respect, it 

 differs from all Coelenterata ; secondly, that the embryo thus pro- 

 vided with mouth, stomach, intestine, and anus acquires a com- 

 pletely bilateral symmetry ; thirdly, that the cilia with which it is 

 primitively covered become restricted to one or more circlets, 

 some of which encircle the axis of the body, or a line drawn from 

 the oral to the anal apertures ; and, fourthly, that within this bi- 

 laterally symmetrical larva or JEcliinopcedium, as it may be called, 

 the more or less completely radiate Echinoderm is developed by 

 a process of internal modification. 



Miiller believed that the first step in this process was the in- 

 growth of a diverticulum of the integument, as a hollow process, 

 out of which the ambulacral vascular system of the Echinoderm 

 took its rise. He did not attempt to explain the origin of the 

 so-called blood-vascular system (or pseudhsemal vessels), nor of 

 the perivisceral cavity. Miiller' s conclusions remained michal- 

 lenged until 1864, when Prof. Alexander Agassiz took up the 

 question afresh, and, in a remarkable paper on the development 

 of the genus Asteracanthion, detailed the observations which led 

 him to believe that the ambulacral vessels do not arise by involu^ 

 tion of the external integument, but that they commence as two 

 primitively symmetrical diverticula of the stomach (the "wiirst- 

 formige Korper" of Miiller), one of which becomes connected 

 with the exterior by an opening (the " dorsal pore " observed by 

 Miiller, and considered by him to be the origin of the ambulacral 

 vessels), and gives rise to the ambulacral vessels, the ambulacral 

 region of the body of the Echinoderm being modelled upon it ; 

 while, upon the other gastric sac, the antambulacral wall of the 

 starfish-body is similarly modelled. Both gastric sacs early be- 

 ' come completely separated from the stomach of the Uchino- 

 pcedium, and open into one another, so as to form a single horse- 



