170 ‘THE ATLANTIC. [CHAP. UI. 
The singular observation was afterward made by Mr. Buch- 
anan, that this black or brown substance which inecrusted the 
coral, and appeared to pass into and to form its bases of attach- 
ment, consisted of almost pure black oxide of manganese. The 
whole of the coral was dead, and appeared to have been so for 
along time. It was so fresh in its texture that it was scarcely 
possible to suppose that it was subfossil, although, from the com- 
paratively great depth at which it was found, and the many 
evidences of volcanic action over the whole of this region, one 
could scarcely avoid speculating whether it might not have 
lived at a higher level, and been carried into its present posi- 
tion by a subsidence at the sea-bottom. 
Attached to the branches of the coral there were several spec- 
imens of a magnificent sponge belonging to the HrxacrineL- 
LIp#. One specimen, consisting of two individuals united to- 
gether by their bases, is 60 centimetres across, and has very 
much the appearance of the large example of the tinder-fungus 
attached to the trunk of a tree (Fig. 38). Both surfaces of the 
sponge are covered with a delicate net-work of square meshes 
closely resembling that of //yalonema, and formed by spicules 
of almost the same patterns. The sponge is bordered by a 
fringe of fine spicules, and from the base a large brush of 
strong, glassy, anchoring spicules project, fixing it to its place 
of attachment. The form of the barbed end of the anchor- 
ing spicules is as yet unique among sponges. ‘Two wide, com- 
pressed flukes form an anchor very much like that of one of 
the skin-spicules of Synapta. The sponge when brought up 
was of a delicate cream color. It was necessary to steep it in 
fresh water to free it from salt, and the color changed to a 
leaden gray. A number of small examples of the sponge, 
some of them not much beyond the condition of gemmules, 
were found attached to the larger specimens and to branches 
of the coral, so that we have an opportunity of studying the 
earlier stages of its development. 
