63 
in diameter. These from their frequent union have assumed other 
forms also, and particulai’ly that approaching to an oval, resulting 
from the formation of one trunk, by the junction of two or three 
branches. 
The superior termination of these branches appears from one or 
two, which have remained unchanged in their form, at one end of the 
specimen, to have been a concave star of a very elegant figure, formed 
by perpendicular plates converging from the circumference. The stars 
are also plainly seen on the polished surface of the specimen. It is 
there seen that the perpendicular plates, some much larger than 
the rest, reach in a straight line to the centre, and that in the trian- 
gular space between every two of these, are smaller plates which 
pass from the circumference in different lengths towards the centre. 
These smaller plates are decussated by other perpendicular plates 
concentrically disposed, and thereby forming a species of net-work 
within each small triangular cavity. This structure, not exactly in- 
deed expressed in the engraving, gives to the stars a peculiar richness of 
appearance, which is still farther increased by the alternately greater 
and less projection of the ends of the straight plates beyond the peri- 
phery of the circle formed by the most external of the concentric 
plates. 
The fossil madrepores, with distinct, finely striated branches, termi- 
nating with tlie madreporean star, have been considered by the earlier 
oryctologists as petrified reeds, (junci lapidei ). Thus Mercatus gives]^the 
representation of a madreporite of this kind under the designation of 
juncus lapideus ; but Lancisius remarks that this stony reed of Mercatus 
differs very little from millepora Imperati. 
Madreporites bearing this reedy appearance are found in several parts 
of this island. They are frequently met with in Derbyshire, of two inches 
in thickness, and upwards of two feet in length. 
A small specimen of this kind of fossil madrepore from Derbyshire, of 
rather a conical form, is figured at Plate VI. Fig. 2. Sufficient of the 
