PORITES. 7 



clavaria by being more openly forked, and that clavaria had more club-shaped branches. 

 Further, the figures have had a bewildering effect, for they have been supposed to represent 

 the general structure of Porites, whereas the specimen figured is an extreme West Indian 

 form, and it would be difficult to find another like it, certainly nowhere in the Indo-Pacific 

 area. Only a set of diagrams such as we attempted in Vol. V. p. 14,* can convey to us some 

 distant idea of the structure of Porites. Therefore the French figure, excellent as it may be as 

 the type of fureata, is calculated to damp the ardour of every investigator at the very outset of 

 his studies on Porites. I well remember that it was a standing enigma to myself for years, 

 that is, until I discovered its place in the system. 



Milne-Edwards and Haime's arbitrary treatment of Lamarck's astraoides has already been 

 referred to, and was moreover described, as far as it could be unravelled, in Vol. V. p. 244. 



The next work to be noted is that on the Corals of the Antilles (1860), by Duchassaing 

 and Michelotti. It referred chiefly to forms collected at the Islands St. Thomas and Guadaloupe, 

 although other localities are occasionally mentioned. In this work the Porites received but 

 scant attention, the genus being dismissed in little over a page. 



In addition to "les eispeces connues " (!), that i.s, clavaria and fureata 'La.iwa.YC^, Jlexuosa 

 Dana, divaricata Lesueur, they found five f which they thought to be new. 



In 1864, the same authors published a supplement to their earlier work, and in this the 

 Porites are more fully treated, and the classification of the genus fundamentally revised. 

 Further examination had showed them new and striking differences, so that even the genus 

 was split into three subgenera — Porites for the branching forms ; Neopoi'ites for encrusting 

 forms, which were massive ancl tended to have a knob-like columella on what was often a solid 

 basal floor ; Cosmoporites for encrusting forms which are thin, creeping, and, for the most part, 

 have a columellar tangle {"columella laxa"). The more branching or approximately 

 branching forms are rescued from " les especes connues," and given names of their own.t while 

 one is identified as of the same species as Lesueur's flabel.liformis. 



It is important to note that every part of this work of Duchassaing and Michelotti is 

 useful except those in which the authors identified " known species." The vicious custom was 

 (and, alas ! is) in vogue of calling a specimen by some old name, and giving neither figure nor 

 description. The specimen which Lesueur thought was specifically identical with Lamarck's 

 clavaria, he figured and described ; we can all see that it was not elavaria, and no harm is done. 

 But not 30 with these authors, and we are left almost entirely to conjecture as to the characters 

 of the specimen which they called clavaria. The probability that it was not Lamarck's would 

 have been very great had we not means of ascertaining this fact for certain, in that the authors 

 arranged their forms into morphological divisions, and the supposed clavarias were placed 



* Reproduced below, p. 139. 



f That is, four encrusting forms, which were called superficialis (see p. 59), inceiia (see p. 60), 

 guadalupensis (see p. 45), and agaricus (see p. 105), while the only branching form was called 

 solanderi (see p. 57). The rest of their collection was apparently distributed among the " known 



% Valida (see p. 56), plumieri (see p. 58), and inacrocephalu (see p. 89). 



