166 A. E. Verrill — Bermudian and West Indian Reef Corals. 



The most remg-rkable specimen that I have studied is now figured 

 (pi. xxxii, fig, 1). It is preserved in the Museum of Yale University. 



In most parts it is a typical specimen of variety pahnata. But 

 growing out of the upper side of one of its palmate fronds there is a 

 cluster of typical branches of the variety prolifera. The two forms 

 are in perfect continuity and there is no evidence of injury or other 

 physical cause for this abrupt alteration in the character of the 

 growth at this particular place. No. 6621. 



Many specimens of var. palmata have the distal ends of some of 

 the fronds divided into digitate branches of variety prolifera, but 

 in such cases the change is gradual. Such subvarieties may be 

 designated as palmato-prolifera, for convenience. 



Var. palmata, when growing vigorously, often produces small, 

 ascending, or incipient branchlets over the whole or part of its upper 

 surface, which is then ver}^ uneven. Some of these branchlets some- 

 times become 75 to lOO™'" long, and agree with prolifera. The large 

 specimen from the Bahamas, in the American Museum, figured b}' 

 Whitfield (op. cit.) is one of this kind. I have named this sub- 

 variety, surcido-palmata. M. cornuta D. and M. seems to have been 

 based on a specimen of this kind. 



Specimens intermediate between variety cervicornis and variety 

 prolifera are to be found in many American collections, but I have 

 never seen specimens clearly intermediate between p)almata and cer- 

 vicornis, though such probably exist. They seem to be the extremes 

 of the variations in form. 



Y ariety f'djell urn grows Wke palr/tata, but forms much thinner 

 fronds than usual. 



Many specimens occur, especially in the Bahamas, intermediate 

 between flabellurn and pjrolifera. In some of these there may be 

 on one side of the same clump, broad frond like branches of the 

 flabelliform type, while on the other side digitate clusters of 2)'>'0- 

 lifera may occur ; or a flabelliform branch may end in free digita- 

 tions ; or free branches, proximally of the prolifera form, may, 

 farther out, coalesce into a flat frond, and distally may again split 

 up into prolifera branchlets. The American Museum* New York, 

 has a good series of such intermediate forms, from the Bahamas, 

 (coll. R. P. Whitfield). 



For these intermediate forms, I use the name flahello-prolifera. 



