104 



Finally with regard to the origin of the perihaemal spaces in 

 Asterina gibbosa, what is new in Goto' s account is not true. The 

 distal portions of the radial perihaemal canals are not formed by the 

 hollowing out of blocks of mesenchyme but by the hollowing out of 

 wedges of cells wich grow from the tips of the coelomic rudiments. 

 The lumen appears secondarily, it is true, and is not a direct prolon- 

 gation of the lumen of the coelomic pocket, but this is a matter of de- 



Place where the transverse 



septum has become 



absorbed 



Left Hijdrocoele 



Left posterior Goelom 



Anterior Coelom 



Bight Hydrocoele 



Bight posterior Coelom 



fig. 3. 



tail rather than of principle. Had Goto examined sections parallel to* 

 the disc which were accurately orientated, he could not, I think, have 

 given this mistaken account of the origin of these canals. — 



To sum up — 



A repeated study of the development of Asterina gibbosa has 

 established the following points: 



1) The coelom on each side of the larva is divided into an anterior 

 and a posterior part by a transverse septum. 



2) From the anterior coelom on each side a water-vascular rudi- 

 ment is budded off. 



3) The perihaemal spaces are coelomic in origin. 

 Montreal, Dec. 15. 1899. 



