192 Garter — Fossil Sponge- Spicules from Ben Bulben. 
number of rays varies from 6, 12, 18 to 24 (figs. 4-7), according to the amount 
of division of the extremities of six arms. The stellates vary m size from 
l-75th to 5-24ths of an inch (fig. 1) in diameter ; and the smallest are not 
only the most numerous and have the greatest number of rays, but, as the 
latter often arise from a division of the arm close to the centre, they acquire 
the appearance of globular little stars ; while, where the arms are a little more 
extended, they often present the appearance of a “ Maltese cross.” But the 
most striking feature of this spicule is that, from the smallest to. the largest, 
each ray is spiriform ; that is, its surface presents a spiral inflation m which 
the coils are more or less numerous, extending from the base to the apex of 
the ray (figs. 2 and 3) ; where they are most numerous they, of course, are 
more transverse, and then appear like separate annulations ; while the most 
remarkable difference in them is confined to the rays of the largest stellates, 
where, towards the base, the spiral line of inflation becomes broken up into 
short portions (fig. 3 b), which look very much, from their alternate arrange- 
ment in adjoining coils, as if this had been produced by another spiral groove 
pursuing an opposite direction (that is, across the original inflation). 
This fossil spicule is incomparably more numerous than the other spicular 
forms accompanying it ; so that, together with its peculiarities being confined 
to a stellate form, I am compelled to think that it must have belonged to a 
species of Holasterella which, if found in situ hereafter (that is, forming the 
entire sponge), will be like H. conferta ; hence I have much pleasure in naming 
it Holasterella Wrightii, after its discoverer. 
The next spicule in frequency appears to have been a hexactinellid, with 
the sixth or external arm not produced as in the large surface-spicules of the 
Sarcohexactinellida, in which four arms are extended over the surface laterally, 
while the fifth, like the shaft of a nail, goes vertically inwards (fig. 8).. With 
this, and also equally plentiful, are the free ends of “ anchoring -spicules 
terminated by four much-recurved spines or flukes situated opposite each 
other, and still attached to a small fragment of the shaft (fig. 9), longer and 
larger portions of which may be observed in the “chert,” both longitudinally 
and in transverse sections under fracture, indicating that with the anchoring 
ends they are the fragmentary remains of what originally were anchoring- or 
cord-spicules of a Hyalonema or Sarcohexactinellid sponge. 
Tolerably numerous also are Lithistid spicules, especially one like a tripod, 
in which the centre is convex and smooth, while the three legs, bending out- 
wards and downwards, end respectively in expanded concave feet, which seem 
to show that they were once applied to similar surfaces on adjoining spicules 
(figs 10 and 11). This in all probability was a surface-spicule like those of 
Corallistes aculeata (‘ Annals,’ 1880, vol. vi. pi. vii. fig. 45). Fragments of 
other Lithistid spicules are also present, such as the dendritically branched 
surface-spicule (fig. 12), and the shafted one of the fully developed or internal 
structure (fig. 13). 
