THE EDWARD SIA-STAGE OF LEBEUNIA. 271 



outgrowths are very numerous and vesicle-like, and the polyps 

 float freely, characters never met with in the form under con- 

 sideration. 



Within the past year, Prof. A. E. Yerrill (1898), without 

 experience of either form, has cast doubts upon the identity of 

 the Jamaican specimen with the true Soplophoria coralligens, 

 and also with the conclusion that it may be but a young form 

 of Leorunia. 



Prom the above facts, however, no other conclusion than 

 that it is identical with Wilson's species seems possible, and in 

 the fuller description yet to be published I hope to show its 

 agreement with the Jamaican Lehrunia. Meanwhile, it will be 

 sufficient to refer to it as Lehrunia coralligens (H. Y. Wilson). 



The adult polyps are hermaphrodite. Ova and spermaria 

 occur in considerable numbers closely associated in the same 

 mesentery. 



II. External Characters. 



In living polyps the larvae were to be seen through the some- 

 what transparent tissues of the parent moving about within the 

 ccelenteron, and especially within the tentacular cavities. As 

 many as five were contained within a single tentacle. 



So far as could be observed, they possessed little or no motion 

 of their own, but allowed themselves to be carried along by the 

 circulation of the coelenteric fluid. By a peculiar jerking action 

 of the adult three or four larvae at a time would be shot out 

 through the mouth for some distance, and then sink to the 

 bottom of the vessel. Here they would lie unattached for some 

 time in any position, though usually on their side. Afterwards 

 slow movements, both translatory and rotatory, would commence, 

 but these at no stage became pronounced. Individuals might 

 come to rest at apparently any point of their surface and again 

 commence their swimming movements. 



When first extruded the larvae measured about 1 mm. inleneth 

 and were of a pyriform shape (PI. 18. fig. 1), divided by shallow 

 grooves into eight, nearly equal, longitudinal regions. Sometimes 

 they would take the form represented in fig. 2, or become short 

 and cake-like as in fig. 3. Viewed along the edge, the colourless 

 ectoderm was easily distinguished from the endoderm, the latter 

 dark brown from the presence of numerous zooxanthellae. In 



21* 



