GEODITES. 209 
-09 mm. in thickness, whilst in the largest form noticed they are 1 mm. in length 
by *25 mm. in thickness. The rays are relatively robust, gradually tapering, with 
blunted ends. 
These spicules were first noticed by the late Mr. Charles Moore, who states 
that they occur in such numbers in the Liassic limestones of South Wales as to 
constitute a material portion of the substance of the rock. As they are now of 
calcite Mr. Moore supposed that they belonged to a Calcisponge and placed them 
under the genus Grantia, Fleming, but Mr. Carter subsequently pointed out that 
from their form and proportions they must have been originally siliceous, and that 
they probably belonged to Pachastrella. The spicules are scattered irregularly in 
the grey liassic limestone, and weather out on the surface of the rock as minute 
whitish bodies, which can occasionally be obtained quite detached. In sections of 
limestone of the same age at Shepton Mallet these caltrop spicules are inter- 
mingled with acerate and trifid spicules belonging to other sponges (Pl. XIII, 
fig. 4). In all, the silica has now been replaced by granular calcite. The same 
forms also occur abundantly in beds of chert at Portland, similarly intermingled 
with other spicules. In this deposit they are siliceous, and sometimes they 
are found quite detached in a matrix of decayed chert. 
Distribution.—Lower lias conglomerate and limestone at  Brocastle, 
Cowbridge, Southerndown, Glamorganshire, South Wales; Shepton Mallet, 
Somersetshire; Portland beds, in the cherty bands below the ‘‘ Whit bed,” on 
the Isle of Portland, and at Upway near Weymouth (Geol. Surv. Museum, 
Jermyn Street). 
Genus.—GuopitEs, Carter. 
17. Guopitus, sp. (a). Plate XIII, fig. 5. 
Trifid spicules with stout, straight, elongate shaft, and simple head-rays 
projecting obliquely forwards. In the specimen figured the shaft is 2°25 mm. in 
length by °15 mm. in thickness, whilst the head-rays, of which only one is 
preserved entire, are 45 mm. in length by -1 mm. in thickness. These trifid 
spicules are fairly common in beds of Portland chert, associated with the 
Pachastrella spicules described above, but they are usually in a very fragmentary 
condition. ‘They belong to a common type of zone-spicule, and differ but slightly 
in form and size from the trifid spicules of Geodites antiquus, from the Carboni- 
ferous limestones of Scotland and North Wales (Part II, p. 150, Pl. V., figs. 3 a—d) ; 
similar forms likewise occur in sponges from the Upper Chalk, and in existing 
sponges as well. 
Distribution.—Portland beds; in the cherty bands below the ‘‘ Whit bed,’’ 
Isle of Portland. 
