No. 1.] ACTINIARIA OF THE BAHAMAS. 39 
retraction of the tentacles by the folding in of the upper part 
of the column over them. In transverse sections the sphincter 
muscle of Dzscosoma is oval and attached to the column wall by 
a distinct pedicle, the mesogloea of which breaks up quickly 
into a number of processes, although towards the inner side a 
prolongation of it, much reduced in size, is continued on towards 
the extremity of the swelling, processes arising from it in a 
somewhat pinnate manner. A little below the point of attach- 
ment of the pedicle to the column wall are a series of well- 
marked muscle processes, succeeding which comes a short in- 
terval in which all such processes are wanting ; below this they 
again make their appearance, being now relatively small, how- 
ever, and extend the rest of the way down the column. 
The ectoderm of the column is raised into slight elevations 
(Pl. III., Fig. 16), not sufficiently large, however, to be plainly 
noticeable to the naked eye, or to give a tuberculate appearance 
to the column. They are produced by solid conical elevations 
of the mesogloea, and are very numerous and closely packed, 
the ectoderm covering them being quite undifferentiated and 
resembling in structure that which covers the walls in the 
intervals between them. The endoderm cells throughout con- 
tain numerous “ yellow cells.” 
The tentacles are short finger-like processes, and are very 
numerous in the larger specimens, their number approaching in 
some cases 600. All the tentacles of each radial series commu- 
nicate with the same intraseptal space, but the number in each 
series is subject to considerable variation, and I was not able to 
discover any law regulating this number. Some of the series 
were very evidently longer than the others, extending nearly to 
the mouth, but the length of these rows varied, as also did the 
number of shorter rows between successive longer ones, there 
being in some cases only two or three, in other cases five or six. 
I endeavored by carefully dissecting away in succession the 
mesenterial pairs, and the tentacles communicating with their 
intra-mesenterial spaces, to determine the relation between the 
number of tentacles, and the grade of the septum ; but the process 
was one of considerable difficulty, and, so far as I carried it, 
yielded no definite results. In structure the tentacles resemble 
closely those found in members of the sub-tribe Actinine. The 
ectoderm was exceedingly richly supplied with nematocysts, 
