﻿168 MADEEPORARIA. 



but in the absence of protuberant calicles. The stock was also much larger than any of 

 those known. While this coral appears to belong to this sub-family, its generic affinities 

 are somewhat uncertain. 



Including Eidley's and Quelch's types, there are at the time of writing twenty-two 

 specimens, several of which are mere fragments, in the National Collection. Examination 

 of these has resulted in the establishment of two new species, making six in all. An 

 account of the morphological results has already been published in the Ann. and Mag. 

 Nat. Hist., XX. (1897) p. 127. 



II. MORPHOLOGICAL. 



The Anacroporse agree in all important structural features with branching Montipores. 

 There is the same axial streaming layer, which is frequently laminate. This axial layer 

 forms in both * the growing ccenenchymatous tips of the branches, and in it the young 

 calicles first appear. A cortical layer is developed, but seldom attains any great thickness, 

 hence it is frequently raised into protuberances by the grovting calicles. Whereas, in the 

 calicles which do not raise the cortical layer, the septa have the same characters as those 

 of Montipora, being stunted to mere vertical rows of spines, in the more developed protuberant 

 calicles the septa may be laminate structures, the radial prolongations of which even project 

 down the sides of the protuberances as true costae. 



Summarising, then, the points in which Anacropora differs from Montipora we have, 

 in all, four distingmshing characters : — 



(1) The method of branching. 



(2) The protuberance of the calicles consequent on 



(3) The thinness of the cortical layer. 



(4) The development of laminate radial structures as septa and costse in the protuberant 

 calicles. 



The method of hranching. — This is very characteristic. The rising stem gives off a 

 branch at a generally wide angle, itself bending away from the brancL Both these may 

 again branch and rebranch. The result is the formation, not of rising tree-like stocks with 

 the branches directed mainly upwards, but of closely matted tangles comparatively low down 

 near the ground and complicated by frequent fusions. The meshes in such a tangle tend 

 to be angular, although curvings of the branches and stems may obscure this latter character- 

 istic. This matting or tangling is very pronounced in Eidley's original species A.forbesi, in 

 Quelch's A. gracilis, and A. solida, and in one of the new species here established, A. reptans. 

 It is not so clear in Eehberg's A. spinosa, or in the other new species here described, A. erecta. 

 In these two latter the comparative smaUness of the diverging angle of the branches causes 

 the stocks to approach the arborescent. 



* Mr. Ridley was not aware that this was also the case in Montipora, and made it one of the 

 distinguishing features between Montipora and Anacropora, 



