10 GENERAL REMARKS. 



are, as in Madrepora, usually broader than the others. In Montipora the septa are sometimes 

 all more or less rudimentary and trabeculate, but in some species they are undoubtedly 

 lamellar, and the directive septa may then be very broad or may even become fused together 

 in the middle line. A bilateral arrangement of parts is thus as well-marked by the directive 

 septa of Anacropora and some species of Montipora as in Madrepora. 



Colony -formation. — Ridley in 1884* discussed the mode of budding in ikfafi?repor« and 

 Montipora, and considered that there is a fundamental difference between the two types, 

 dependent on the terminality or non-terminality of the distal corallite. He pointed out that 

 Isopora, Studer, a subgenus of Madrepora, is not without axial corallites as had been supposed, 

 but that it is provided with several instead of one. In both Madrepora and Montipora there 

 is a more or less abundant trabeculate coenenchyma. In Madrepora the budding is 

 centrifugal, i. e. new buds arise below the axial corallite. In Montipora the apex consists of 

 undifferentiated coenenchyma and new buds are added above those existing, i. e. centripetally, 

 He compared the condition to determinate or indeterminate inflorescence. The mode of 

 budding in Anacropora is the same as in the genus Montipora. Ridley therefore suggests 

 the foundation of two subfamilies, Madreporinae and Montiporinse, with characters based on 

 this distinction. The terms centrifugal and centripetal do not appear to express accurately 

 the precise modes of budding to which they are applied, and it would probably have been 

 better had Ridley employed the botanical terms determinate and indeterminate to express the 

 distinction in the case of branching species. In foliate species of Montipora the budding is 

 centrifugal, not centripetal, seeing that new corallites are added at the periphery. In 

 branched specimens of Madrepora the buds arise around, and are indirectly connected with an 

 elongate corallite forming the axis of each branch, and extending from its point of origin to 

 the apex, where it always projects more or less. This corallite, often spoken of as the parent 

 corallite, is usually of larger diameter than the others, and often exhibits a better-developed 

 series of septa. It is usually termed the apical corallite (" Endkelche " by the Germans), but 

 axial corallite seems much more appropriate ; the part of it which is " apical," and recognizable 

 in surface- view, is only an insignificant part of its whole length. 



Although the types of budding indicated by Ridley form an essential distinction between 

 Madrepora and Montipora, the type characteristic of Madrepora is confined to branches formed 

 by the living colony during its growth ; in other situations the buds are formed in a similar 

 manner to those of Montipora. In specimens which form incrustations (and all are incrusting 

 in the first instance), new corallites are added peripherally from an undifferentiated mass of 

 tissue which projects beyond existing corallites. It is only when certain of the corallites increase 

 in length and thickness, and indicate the first formation of branches by the development of 

 buds around them, that the form of budding characteristic of Madrepora comes into operation. 

 Frequently both types of budding take place at the same time in one colony ; the one leads to 

 branch-formation, the other to marginal or basal extension. One not infrequently meets 



• " The Classiflcatory Value of Growth and Budding of the Madreporidse," Ann. Mag. N. H. 1884, 

 vol. xiii. pp. 284-201, pi. xi. 



