THE ANATOMY OF THE MADREPOEARIA. 13 



are so crowded with zooxanthellae, owing presumably to the 

 amount of nutriment supplied to them, that the tissues are 

 much obscured. 



The stomatodaeum is invaginated to a considerable depth 

 into the future polyp cavity before it is perforated for com- 

 munication between the coelenteron and the exterior, and also 

 apparently before any mesenteries are formed. The cavity 

 into which it is invaginated is already of considerable diameter, 

 and larger than the ordinary canals of the colony ; though 

 smaller than that of a fully formed polyp, at that point it is 

 probably never enlarged by reabsorption of coral, but its con- 

 tinuation upwards by future growth of the polyp possesses a 

 gradually increasing diameter. 



In a young polyp in which the stomatodaeum was invagi- 

 nated, but not yet perforated below, the latter appeared to be 

 supported by tissue surrounding the future septa, just as the 

 external body wall is supported by tissue enclosing the costse. 

 In sections below the stomatodaeum, and unconnected with it, 

 were seen two small mesenteries with filaments, which appeared 

 to be growing upwards towards the stomatodaeum, and to have 

 not yet joined it. It is therefore possible that these grow 

 upwards from the canal system, and are formed quite inde- 

 pendently of the rest of the polyp. This view is further 

 supported by the observation that, in sections quite at the top 

 of a branch, above the plane of any lateral polyps, occur in 

 the canals one, sometimes two, little mesenteries with fila- 

 ments, which I believe to be growing upwards towards the 

 sites of future polyps. They appear to take rise, near the 

 cavity of the apical polyp, from the wall of the canals. 



In the only other stage of development from which any 

 observations could be made six mesenteries had appeared ; of 

 these the two furthest from the axis carried muscles on the 

 outer faces, though it does not necessarily follow that they 

 were the abaxial *' directives '' of the adult. The muscles of 

 the other two pairs were not sufficiently developed to allow of 

 their arrangement being recognised. 



Conclusion. — From M. Durvillei, the present species is 



