1 5 2 
Cactogorgia is more densely spiculose, firmer, and more rigid than Siphonogorgia; but it 
resembles that genus in (a) the complete retractility of the anthocodise within prominent verrucse, 
( 6 ) the definite '•‘crown and point” arrangement of the spicules on the anthocodise, and (c) the 
spindle shape of the spicules, except on the tentacles, where they are flat and scale-like. 
9. As to the peculiar genus Dactylonephthya Thomson and Henderson, it seems nearer 
Alcyonidae than Siphonogorgidse, and its separation off from the latter is indicated by the 
absence of distinct verrucse or anthocodise. It was included amongst the Siphonogorgiaceae owing 
to the presence of large spindle-shaped spicules in the canal walls. These, however, are not 
extremely numerous. 
Thus the genera of the sub-family stand as follows: 
Siphonogorgia) . , , . . . . 
_ _ , > Dubiously separable: in our opinion, not separable. 
Chironeph thy a \ * 1 r r 
Paranephthya. Markedly distinguished by foliaceous spicules, and scaly appearance when dried ; 
in part = Capnella (Nephthyidae). 
Scleronephthya. Externally resembling Dendronephthya, with hints of an irregular axis. 
Lemnalia. Lacking the Siphonogorgia rigidity, with numerous delicate spicules in the canal walls. 
Stereacanthia. Externally resembling some species of Eunephthya, with stalked polyps, and no 
definite separation between crown and stalk. 
Agaricoides. Quite unique in its pedicelled anthocodiae, with octagonal disc-like expansions. 
Cactogorgia. With polyps borne mainly on the margins of flattened, rigid lobes. 
Criticism of Specific Characters. 
In distinguishing species of Siphonogorgia it seems quite clear that there are many 
differences of little specific importance. 
(a) Little importance can be attached to colour. 
Thus, among eight specimens of S. variabilis described by Thomson and Henderson 
1909, the following combinations of colours occured : (1) White to greyish; polyps white. (2) Pale 
brown merging into purplish-red; polyps -bright yellow. (3) Yellow; polyps yellow. (4) White 
merging into dull red in the branches; polyps white. (5) Yellow; polyps bright red. (6) Creamy- 
white; polyps bright red. (7) Brown; polyps dark red. (8) Sulphur yellow; polyps red. 
Again the colours of the general coenenchyma of three specimens of S. mirabilis are 
described by Thomson and Henderson as being, (1) creamy-white, (2) red, and (3) almost coral- 
red, merging into orange red towards the apex. A specimen from the Siboga expedition is 
coral red, but merges into creamy white at the ends of the branches and on the verrucae. At 
the same time there are cases of a constant colour-scheme so far as is at present known. 
(b) Little importance can be attached to the size of spicules, except where they are markedly 
larger than is usual, as in Y. macrospina , Y. macrospiculata , and Y. media ; or markedly smaller 
than is usual, as in Y. mirabilis. 
The proportion of length to breadth may however be diagnostic, where the general 
Siphonogorgia type of slender spindles is deviated from, as in Y. mirabilis , where the spicules 
are almost as broad as long. 
