146 



On examining Milne-Edwards' s ' Histoire des Corallaires,' I was 

 pleased to find that my views on this point were in accordance with 

 that naturalist, he having divided the family into three sections, 

 which he calls respectively Smooth Actinia, Warty Actinia, and 

 Perforated Actinia, or with pores, each being equal to one of Mr. 

 Gosse's genera and my subdivisions. The true generic characters 

 of the ActiniadcB are to be found (as I have already mentioned) in 

 the integuments and in the tentacula ; these latter, although varying 

 with age, are yet constant in form and number in adults of the same 

 species. From these characters amongst others, I had divided the 

 family into eight genera (exclusive of Anthea and Adamsia), of 

 which the types were Actinia equina, A. gemmacea, A. clavata, A. 

 bellis, A. viduata, A. dianthns, A. parasitica and A. coriacea ; and 

 these I named respectively Actinia, Bnnodes, Cyrtactis, Heliactis, 

 Sagartia, Actiniloba, Aster, and Cribrina. My divisions and genera 

 were formed from examinations of British species only, and I was 

 not a little surprised to find that they so closely accorded with 

 Milne-Edwards' s labours, founded as they are on an examination of 

 all described species. The only genus containing a British species 

 that I have not verified is his genus Dysactis, in which he includes 

 A. biserialis. This species, however, was described by the late 

 Professor Forbes as occurring plentifully on the shores of the island 

 of Herm ; I have therefore admitted it as a genus on the authority 

 of Milne-Edwards. The British species are distributed amongst six 

 of his genera, namely, Metridium, Actinia, Par actis, Dysactis, Cereus, 

 and Adamsia ; of these, Actinia is the same as mine, Paractis I have 

 used in lieu of Sagartia, and Dysactis I also admit unaltered — they 

 are represented by Actinia equina, Actinia viduata and Actinia bise- 

 rialis. My genus Actiniloba is the same as his genus Metridium, and 

 includes Actinia dianthus. I retain Actiniloba, as being the most cha- 

 racteristic name. His genus Cereus includes A. coriacea, A. gemmacea, 

 A. clavata and A. bellis ; these species, it was quite evident, could not 

 be included in one genus ; indeed M. -Edwards divides his genus Cereus 

 into several sections, A. coriacea being in one section, A. gemmacea 

 and A. clavata in a second, and A. bellis in the third. Restricting, 

 therefore, his genus Cereus to A. gemmacea, instead of retaining Mr. 

 Gos3e's name Bunodes, I retained my division of the remainder of the 

 species under the generic names of Cribrina, Cyrtactis and Heliactis. 

 His genus Adamsia includes Actinia parasitica and Adamsia palliata, 

 two very dissimilar species. No one, I am sure, who has ever seen 

 these Polypes would imagine for an instant that they can possibly 

 belong to the same genus ; and indeed the fact that Adamsia palliata 

 secretes a horny base, an incipient polypidom, induces* me to suggest 

 the necessity of forming on its characters a distinct family. I have 

 therefore formed a genus for the reception of Actinia parasitica, and 

 withdrawing the suggested name Sagartia from the genus that will 

 now stand as Paractis, I have transferred it to this genus. The 

 genus Adamsia I at present retain under the family ActiniadcB ; and 

 without entering into the question in this paper as to its proper 

 position, I would suggest that it will eventually be placed between 





