Rofe—Pentremites, &c. 249 
those of Pla/ycrinus; but, instead of the semilunar depression for 
the attachment of the arms in that genus, there is in Pentremites a 
deep excavation, or sinus, down the middle of each plate, giving it 
somewhat the appearance of a fork with two broad prongs. The 
upper sides of the adjoining radial plates are connected by a quadri- 
lateral plate, fitted into the more or less obtuse angle formed at their 
junction (in this strongly resembling the plates on the dome of Cya- 
thocrinus above noticed); whilst the two other sides of this plate 
form continuations of the sinus nearly to the summit. So far, 
undoubtedly, Pentremites may be considered to be allied to the 
Crinoidea ; but beyond this they appear to differ; for, instead of 
the long branching arms attached to the radial plates of Crinoidea, 
there is only a row of short fimbriated tentacles, ranging along each 
side of the above sinus, nearly meeting in its centre, and forming 
the so-called ‘ pseudambulacral area.’ The removal of the tentacles 
(which, however, are not often found im setw) leaves a groove down 
each side of the sinus; and at the bottom of this a row of minute 
pores is seen. On making a cross section of a Pentremite, it is 
found that these pores are openings into one or more flat tubes, vary- 
ing in number in different species, which pass from the base of the 
sinus to the apex, and are attached nearly at right angles to the 
plates on each side of the sinus by one edge; whilst the inner edge, 
which projects into the visceral cavity of the animal, expands so 
as to form a nearly circular tube as seen at fig. 7, which is a section 
of Pentremites ellipticus. Varieties of this species and of P. orbi- 
cularis are the most commonly found in this district; and they all 
have only one tube to each side of the sinus, and appear to have in 
all respects similar organization. As these flat tubes approach the 
summit, they become yvradually narrower; and at the apex of the 
quadrilateral plate the expanded parts of the tubes from the adjoin- 
ing sinus (and not from the same sinus) meet, and form one of the 
summit-openings characteristic of thisgenus. These flat tubes may 
be called the ‘ pore-tubes ;’ and the interval between those in each 
sinus is filled by what M. F. Roemer calls the ‘lancet-plate,’ which 
forms a bed for the tentacles, and is slightly indented to receive 
them; but in many species of Pentremite, if not in all, this lancet- 
plate is in reality a compound plate. formed of two contiguous plates, 
extending from the bottom of the sinus to the top, and, there turning 
right and left round the summit-openings, they pass down the adjoin- 
ing sinus, to form half its lancet-plate, leaving at the apex of the 
body a pentagonal aperture, supposed to be the mouth. In seme 
weathered specimens, the two parts of the lancet-plate are separate; 
and in many they appear to meet only at the top and bottom of the 
cross section, leaving a lozenge- shaped opening between SDS as 
shown at a, fig. 7. 
Possibly we should not be very far wrong if we were to suggest 
that the tentacles of Pentremites represent the arms and tentacles of 
Crinoidea, and the internal tubes of the former those under the 
dome of the latter; and in that case Pentremites would be nearly 
allied to Crinoidea. 
