52 MADREPORARIA. 



under iIr' old fasliioiu'd method would luive been certainly classed together as a species, one 

 being made the type and the other a variety, though which sliould ])o which would have rested 

 upon the arbitraiy choice of the worker. This method is, to me, wholly unscientific and 

 consequently indefensible. That the two are closely related no one can deny ; they not only 

 belong to the same genus, but they have special characters in common. Both are small 

 encrusting colonies ; both seem to begin as ragged, flaky layers of skeletal network, with very 

 irreguhu- calicles opening in it. But then they diverge ; in one, the skeletal network persists 

 as so many horizontal layers running into one another and perforated by neat, round holes, 

 with no ai)parently trabecular formation ; in the other, trabecular and horizontal elements 

 become more equally developed, and the network is more uniformly filamentous. In this 

 latter, owing to the development of trabecular the walls of the calicles rise into surging ridges, 

 and the whole aspect of the surface changes. These then are, as far as we can yet see, the 

 chief morphological differences. What we want to know is their meaning and explanation. 

 If we wi'ite down that one is a species, and the other a variety of it, we are guessing at the very 

 thing we wish to find out. That is not the only objection to such methods, for we should have 

 not only guessing, but, worse still, guessing in terms which we cannot define, and thus 

 enveloping the facts in a cloud instead of stating them as simply and as clearly as possible for 

 the use of the students who come after us. 



«• 1906. 1. 1. 21. 



34. Porites Nevis Island 1. {P. Nevis prima.) 

 [Nevis Island,* coll. Lesueur ; ? ] 

 Syn. Porites davaria Lesueur, Mem. du Mas. vi. (1820) p. 289, pi. .xvii. fig. 17. 



The original description is as follows : — 



" P. dichotomo-ramulosa ; ramvlis rraxdn, auh clavatis, obsolete compi-essis ; stcllis latis, 

 planuldlis, contiguis, superficial ilms." 



The tentacles of the animals are said to be whitish, but quite white at the tips and 

 arranged round a reddish (burnt sienna) disk, which rises into a short cone ; the mouth is 

 bordered with white. The protruded animal was not seen to ri.se higher than about half the 

 diameter of the calicle. Its sides were ribbed or furrowed. The corallum rises 15 cm. high, 

 but in spite of its size it is very fragile. With its crowded, intertwined branches it covers 

 great areas. 



Lesueur says that it is not the same as Ellis and Solander's coral,t because the living layer 

 is confined to the tops of the stems, whereas in the Madrcpora Porites figured by Ellis and 

 Solander it extends over the whole corallum. The fragility of the coral is not a known 

 character of Lamarck's type of davaria, see p. 81. Lesueur's figure shows quite an unusual 

 growth-form ; I have never seen a Porites like it. And it is worth noting that its possession 



* In the little harl)Our which faces " St. Eustache." 

 t Zoophytes (1786) p. 172, pi. xlvii. fig. 1. 



