1907.] Development of Ophiothrix fragilis. 443 



with a high vacuolated crest were obtained from artificial fertilisation. It 

 appears to be clear that we must assume that eggs are capable of fertilisation 

 before they are quite ripe, and that the fact that eggs can be fertilised is no proof 

 that they are fully ripe or that the resulting development is normal. This 

 conclusion has, I think, a somewhat important bearing on the experimental 

 studies for which the eggs of ; Echinodermata have furnished the material. 

 Notably the statements which some authors have made about obtaining 

 ripe eggs from sea-urchins like Strongylocentrotus all the year round must be 

 received with great caution. 



When engaged in the study of the development of Echinus esculentus, 

 I noticed that if one attempted to fertilise the eggs of half-grown specimens, 

 larvae with imperfectly developed arms resulted, which only lived a few days. 



The fact, however, that the abnormal development of Ophiothrix fragilis has 

 a considerable resemblance to the normal development of Ophiura brevis is of 

 interest, as showing how far reaching in its influence on the subsequent 

 development is the condition of the egg at the moment of fertilisation, and 

 the idea is suggested that here, perhaps, is to be found the origin of variations. 



We left the normal development at the moment when the ccelom had been 

 completely separated from the gut and had been divided into right and left 

 halves. The ciliated band had been formed, and its first two outgrowths, the 

 postero-lateral arms, had appeared. The stomodseum had joined the gut. 



The madreporic pore is now formed by the union of an ectodermic ingrowth 

 with a short outgrowth from the left coelom. The left ccelom is now distin- 

 guishable from the right by its more distinct lumen — an indication of the 

 fact that it gives rise to the so-called ampulla of the stone-canal. I have 

 searched in vain for any trace of a right pore -canal such as has been found 

 by Grave in the case of some ophiurid larvse. From both cceloms, muscle 

 fibres are developed which surround the oesophagus of the larva and endow 

 it with the power of peristaltic motion. The other larval arms then appear, 

 first the antero-lateral pair at the sides of the mouth and then the post-oral 

 pair at the sides of the anus. The coelom now undergoes transverse division 

 on both sides into anterior and posterior cceloms, the division on the left side 

 being completed before that on the right (fig. 2). The posterior ccelom on 

 both sides is at first a solid nodule of cells which later acquires a lumen. 

 Next, the last pair of ciliated arms, the postero-dorsal, appear, and then the 

 anterior coelom on both sides again prepares for division. As before, a 

 nodule of cells buds out from the posterior end, but this time the division 

 is not completed. On both sides the posterior portion, which will form the 

 middle division of the coelom, is not completely separated from the anterior 

 division. On both sides it becomes hollow. On the left side it buds out 



VOL. LXXIX. — B. 2 K 



