1899.] ASTRvEID COE.ALS FROM THE SOUTH PACIFIC. 735 



(Journ. Linn. Soc, Zool. xviii. p. 1, 1885) appears to be absolutely 

 useless. The subfamilies and alliances are purely artificial, and 

 even for the genera in a single alliance there is no distinct phraseo- 

 logy. Indeed, genera and alhances are constantly described in 

 different but absolutely synonymous terms. That work, too, is not 

 to be depended upon in so far as it refers to the authorities for 

 the different genera. Indeed most of the errors of Milne-Edwards 

 and Haime, corrected by Verrill, Klunzinger and others, appear in 

 its pages. The Fungid and Perforate families are, however, often 

 well described, and the work is a useful key to their genera. 



Throughout the work I have made a liberal use of the Coral 

 Gallery of the British Museum, and I have profited greatly by 

 the advice of Prof. Jeffrey Bell and Mr. Bernard. 



Genus Euphtllia. 



Euphyllia, Dana, Zooph. p. 157 (1848). 



Euphyllia, Milne-Edwards & Haime, Cor. ii. p. 191 (1857). 



The characters separating the genera Euphyllia and Caulastrcea 

 do not seem to me of sufficient iuiportance for the retention of 

 the latter genus. The genus Eusmilia, further, is very doubtfully 

 distinct. The characters separating the subfamilies Astrceidce 

 ccespiiosce and Astrceidce coitfluentes of Martin Duncan, when 

 practically examined, are found to be merely artificial and 

 worthless. 



1. Euphtllia glabeescens Chamisso. 



CaryopJiyllia glahrescens, Chamisso & Eysenhardt, Nov. Act. 

 Curios. Nat. x. p. 369 (1821). 



Euphyllia glahrescens, Dana, Zooph. p. 163 (1848). 



Euphyllia glahrescens, Milne-Edwards & Haime, Cor. ii. p. 192 

 (1857). 



I have referred pieces from ten several colonies to this species, 

 from which Euphyllia gaimarcli (Milne-Edwards & Haime) is very 

 doubtfully distinct. 



The colony forms round clumps of corallites, which spring from 

 the same stem, the branching being dichotomous. The branches 

 vary greatly in length between their divisions, in one colony 

 dividing every 12 mm., and in another having a length of nearly 

 40 mm. between the divisions. The whole mode of growth varies 

 enormously, and appears to depend on the extent to which the 

 bases and sides of the corallites are bored into and incrusted by 

 different organisms, and doubtless also on the position of growth 

 in respect to food-supply, &c. 



The corallites naturally vary as well in respect to size, shape, 

 and depth, being somewhat larger and always more distant, with 

 deeper axial fossfe, in the more vigorously growdng colonies. In 

 one colony the corallites are about 4 mm. apart, and the separate 

 calices (showing no trace of any division) vary up to 1"6 cm. in 

 diameter, while in another colony they are •9-1*5 cm. apart by 



