584 E. W. MACBUIDE. 



just before the metamorphosis has commenced^ whilst text- 

 figure 4 shows the changes which have supervened during 

 the first period of metamorphosis. 



Whilst these changes in the relative growth of the internal 

 organs and in the external symmetry have been taking place, 

 new structures have been making their appearance. From 

 the inner wall of the left posterior ca3lom four outgrowths are 

 given off which alternate with the lobes of the left hydrocoele. 

 These outgrowths have thick walls and a narrow lumen. At 

 first their cavities are in open communication with the left 

 posterior ccelom (cf. fig. 49), but this communication is rapidly 

 narrowed (fig, 45), and soon eutirely cut off. A similar out- 

 growth is given off from the anterior ccelom (fig. 44 a, p,/i. 1 

 and 2), which insinuates itself between lobes 1 and 2 of the 

 hydrocoele, whilst those originating from the left posterior 

 cnclom intervene between lobes 2 and 3, 8 and 4, 4 and 5, 

 and (when the hydrocoele ring is completed) lobes 5 and 1 

 respectively. These outgrowths are tbe rudiments of the 

 perihsemal system of cavities and of the muscles and nerves 

 which originate from their walls. As already mentioned, the 

 tips of the hydrocoele lobes now appear as tentacles freely 

 projecting into the outer part of the stomodteum. Alternating 

 with these tentacles there appear inter-radial ridges of ecto- 

 derm (fig. 46, ej).), from which flaps grow out to the right and 

 left. The flaps of adjacent inter-radii meet one another over 

 the bases of the tentacles and in this way the epineural canals 

 or closed ambulacral grooves characteristic of the Ophiuroidea 

 are formed. I own that I was disappointed to find that in 

 this respect Ophiuroidea so closely resemble Echiuoidea, for 

 the development which I have just described is identical with 

 the pi'ocess of formation of the epineural ridges in Echinus 

 esculent us (19). We possess a fairly full geological history of 

 the evolution of Ophiuroidea out of Asteroidea, and of the 

 gradual conversion of the open ambulacral groove into a 

 canal by the meeting of the edges, and I expected to find 

 that the star-fish form would be definitely attained by the 

 larva before the groove became closed. 8ucli is, however, not 



