226 



differs from B. mollis in spicular characters and from B. philippl in the smaller 

 size of the polyps (Whitelegge). Thus the question of species resolves itself 

 into a question of spicules. In our specimens we find all the types of spicules 

 which go to characterise the three species under discussion. We would there- 

 fore adopt the oldest name and rank the present forms as intermediate or 

 linking types. 



A beautiful specimen of this species in the Wood-Mason Collection con- 

 firms our opinion that Bebryce phUippi cannot rank as a species distinct from 

 Behryce inollix. The spicular characters are not so different as to justify such 

 a demarcation, while externally little or no ditterence is found. The colony is 

 complete, and is 95 mm. in height and 60 mm. in breadth. The branching is in 

 one plane, and the branches arise mainly at right angles though they may after- 

 wards diverge from this mode of growth ; they are almost uniform in diameter. 

 The co3nenchyma is thick and arenaceous. The verructe are not very promi- 

 nent but give an undulating appearance to the surface of the cojnenchyma. 

 In these and other respects this specimen agrees with those just described. 



The intimate relationship of the three species and our sj^ecimens may best 

 be summed up in the following comparative table : — 



