596 E. \V. i\IACBI?IDE. 



be inferred tliat here, too, these terminal tentacles when they 

 appear, project at first into the atrium. To the stomodajinn 

 or atrium, containing the primary tentacles, found in Ophiu- 

 roidea, Holothuroidea, and Crinoidea may be compared tlie 

 amniotic cavity of Echinoidea and we must suppose that in 

 the last-named group the original stomoda3um is divided into 

 two parts — one developed in the middle line in order to 

 function for the larva, the other in the adult position to 

 protect the first tentacles. 



As to Asteroidea, there occurs in the test-books a statement, 

 attributed to Agassiz, that in someBipinnaria3 the larval mouth 

 persists and moves to the left. Later researches on the de- 

 velopment of Bipiunaria by Bury and Goto have failed to 

 confirm this statement, but considering of how few types the 

 development is known, it is quite possible that in some form 

 the larval mouth does persist, and the detailed study of the 

 development of such a form would be of rare interest. In the 

 meantime we can only conclude that as the groups of Asteroidea 

 and Ophiuroidea diverged from one another, the Asteroid 

 larva, whilst retaining the fixed stage, lost the primitive 

 features of the mouth, Avhereas the Ophiuroid larva retained 

 these features Avhilst losing the fixed stage, just as the lower 

 insects have lost the caterpillar larva although the higher 

 insects have retained it. 



Summary, 



The principal points Avhich have been brought out in this 

 paper may be summarised as follows : 



(1) The early development of Ophiothrix fragilis varies 

 with the condition of the egg at the moment of fertilisation, 

 and the development of the unripe egg resembles in certain 

 features that of Ophiura brevis. 



(2) The coelom originates as a single vesicle from the apex 

 of the archenteron, and this appears to be true for all classes 

 of Echinoderms. 



(3) The segmentation of the ccolom proceeds along the 

 same lines as those already elucidated for Asteroidea and 



