GENEEA or THE ALCTONAEIA STOLONIPEEA. 331 



Notwithstanding this fact, Sars (26) described a new species as Bhizoccenia filiformis 

 with completely retractile polypes, and von Koch (17), in describing Dana's old species, 

 Bhizoxenia rosea, says the polypes are extraordinarily contractile. 



Without giving more examples illustrating the fearful state of confusion of the 

 group at the present time, I will merely express my opinion that all the old classifi- 

 cations must now be definitely abandoned, and a new one be formed to take their place. 

 In order to do this it is necessary to criticize the genera as they now stand. 



Since the publication of the ' Histoire Naturelle des Coralliaires,' several genera 

 have been added, so that we have now thirteen genera in all (omitting the non- 

 colonial forms Haimeia, Hartea, and Monoxenia). 



The Genera of the Stolonifera. 



In the report on the Alcyonaria of H.M.S. ' Challenger,' Studer and Perceval 

 Wright (30) include the following sixteen genera in the family Cornulariidse : — 



9. Coelogorgia, Milne-Edwards. 



10. Cyathopodium, Verrill. 



11. Scleranthelia, Studer. 



1. Cornularia, Lamarck. 



2. RMzoxenia, Ehrenberg. 



3. Clavularia, Quoy & Gaimard. 



4. Sarcodictyon, Forbes. 



5. Anthelia, Savigny. 



6. Gymnosarca, Saville Kent. 



7. Cornulariella, Verrill. 



8. Telesto, Lamouroux. 



12. Anthopodimn, Verrill. 



13. Sympodium, Ehrenberg. 



14. Erythropodium, Kolliker. 



15. Callipodium, Verrill. 



16. Pseudogorgia, Kolliker. 



These sixteen genera fall into three groups: — 



(1) Those that have been thoroughly well described and figured, and can be readily 



identified as separate genera belonging to the Stolonifera, namely : Cornu- 

 laria, Clavularia, and Sympodium. 



(2) Those that have been only imperfectly described, and had better be incorpo- 



rated in the other genera : Bhizoxenia, Sarcodictyon, Anthelia, Gymnosarca, 

 Cornulariella, and Cyathopodium. 



(3) Those that do not come within the limits of my definition of the Stolonifera, 



and should be placed in other suborders : Telesto, Coelogorgia, Scleranthelia. 

 Erythropodium, Pseudogorgia, Anthopodimn, and Callipodium. 

 The three genera Bhizoxenia, Sarcodictyon, and Anthelia have been established for 

 many years, and I feel some hesitation in proposing that they should£now be abolished. 

 Increased knowledge of the varieties of the species of Clavularia, however, shows that 

 it is a matter of impossibility to draw any hard-and-fast lines between the forms described 

 under these generic names and some of the species of Clavularia. 



The o'enus Bhizoxenia was established by Ehrenberg (5) in 1834 for those forms 

 allied to Cornularia and Clavularia, but difiering from them in the non-retractility of 



