588 E. W. MACBKIDE. 



formed on the left side of the larva is practically " mid- 

 dorsal '' with respect to the median sagittal plane of the 

 larva and to the larval mouth. 



The madreporic pore in these two groups attains its adult 

 position before the right hydrococle is formed, so that when 

 the latter appears it is close to the madreporic pore. 



With the disappearance of all the larval arms except the 

 two postero-lateral, the third and last stage of the meta- 

 morphosis is begun. Its external appearance is shown in 

 fig. 16, and a frontal section is given in fig. 48. The adult 

 arms have become longer and are folded ventrally over the 

 disc — this is not general among Ophiuroidea, but is character- 

 istic of Ophiothrix fragilis. The oral spines which even- 

 tually give rise to the so-called " teeth " have already 

 appeared (fig. 16, sp.). In the internal anatomy the great 

 point to be noted is the appearance of the peri-oral coelom 

 (fig. 48, peri.), as an outgrowth from the left posterior 

 coelom. The peri-oral coelom closely surrounds the oeso- 

 phagus (figs. 51 and 52, peri.), and corresponds to the 

 similarly-named space in the larva of Asterina gibbosa. 

 But whereas in Asterina gibbosa the wall of this space 

 gives rise to the ten retractor muscles which restore the 

 everted stomach to its place on the conclusion of a meal, in 

 Ophiothrix the space remains as an apparently functionless 

 vestige throughout life, as the stomach of an Ophiurid is not 

 eversible. 



The formation of the primitive germ cells commences at this 

 period. As shown in the sagittal section represented in 

 iig. 51, they make their appearance as a series of large uuclei 

 in the wall of the left posterior coelom overlying the stone- 

 canal (fig. 51, (jen.). The further history of these cells has 

 been worked out by me in my paper on Amphiura 

 squamata (17), to which the reader is referred. A section, 

 however, of one of the embryos of Amphiura squamata is 

 given in fig. 52. The single layer of cells shown in fig. 51 is 

 here seen to have become a projecting nodule (fig. 52, gen.), 

 the first rudiment, in fact, of the string called the genital 



