accidental or otherwise is doubtful; certainly there is no 
such evidence of organic union as is found in the case 
of Uranosphaera. The two most striking spicules have 
a spread of about 20 mm.; each has six rays, which are 
not arranged in a flat plane, but slope uniformly down- 
wards at about 30 degrees from horizontal. These two 
spicules are shown at D and E, the latter being the more 
symmetrical of the two. In the downward slope of the 
rays these spicules agree with those of Uranosphaera, 
particularly the many-rayed forms of Uranosphaera 
polyaster; they are much larger than any of the ordinary 
six-rayed- wall-spicules seen so far in Uranosphaera 
hexaster. 
VARIOUS ISOLATED SPONGE SPICULES. 
A number of these are illustrated in Fig 43. A is the 
largest yet found at the Ajax. The ray to the right is 
22 mm. long; the only other ray visible is the frag- 
mentary one to the left. B has a long ray of 12 mm. 
planted on a base of short curved rays, but insufficient 
of these latter remain to determine their original number. 
C shows one ray of 11 mm. with a slight sigmoidal 
curvature, also the fragmentary bases of at least four 
other rays; although too imperfect for complete deter- 
mination, the spicule might perhaps have had six 
uniform sigmoidal rays sloping downwards. D is a fine 
example of a symmetrical six-ray spicule with straight 
rays sloped uniformly downwards; two views are shown 
of this spicule. One or two other six-rayed spicules 
have been found in which the rays have less slope, half 
way between that figured and a plane surface. E is a 
delicate spicule with one long ray to which are attached 
short arms (probably four in the form of a cross), and 
another short arm proceeds from the point of junction 
nearly but not quite in the line of the long arm. FGH 
are three many-rayed spicules; it is, of course, impos- 
sible to say whether the rays were originally distributed 
in all directions of space or only over approximately a 
hemisphere. J is a pretty little spicule of unusual type, 
only the one specimen having been found; it has (or 
rather had) ten uniform rays, the lower surface being 
nearly plane, the upper bossed upwards in the centre; 
there is a possibility that a central ray may have pro- 
ceeded upwards from the boss, as there is a small circle 
of fracture at this point. 
Dr. R. J. Tillyard, who has seen some of the spicules 
and our sketches, writes: *I have been very interested in 
your sponge spicule problem and have had quite a lot of 
discussion with sponge experts on your fossils. At the 
beginning, they were all suspicious of them and con- 
sidered them to be pseudo-spicules . . . but we finally 
put the problem into the hands of the most expert 
authority on sponges that we know of, viz., Mr. E. F. 
Hallmann, B.Sc., who used to be a Linnean Macleay 
Fellow and worked on sponges. He also was doubtful 
at first, but has now written to me to say that he agress 
with you that they are actually sponge spicules, though 
very remarkable ones." 
11 
Class ARCHAEOCYATHI (CYATHOSPONGIA). 
Order ACANTHINOCYATHINA (Okulitch). 
Family ACANTHINOCYATHIDAE (Bedford). 
Acanthinocyathus Apertus Keser (See Part 1, p. 
4, Fig. 20.) 
Specimen with Sponge Spicules in Intervallum. 
This fragmentary specimen is shown in Fig. 44. One 
side of the main of the two fragments, A, shows the outer 
wall, which is irregularly thickened and obscured by 
secondary silicification. In contact with the outside of 
the wall is a large six-rayed spicule two rays of which 
have disappeared (probably at the commencement of 
work on the specimen). In order to preserve the remains 
of this spicule part of the matrix outside the wall was 
retained ;this part is shown by cross-hatching in the 
sketch. The upper part of this sketch shows a surface 
which cuts obliquely across the intervallum. (b) and 
central] cavity (a). "The sketch B shows the reverse side 
of the same fragment; silicified organisms are shown 
white, limestone matrix black. The oval space in the 
middle is the central cavity. After this sketch was made 
more of the matrix was carved away (D) and etched 
so as to expose the inner surface of the network of the 
inner wall (a); the connecting bars between outer and 
inner wall are also well seen (b). The rod-like elements 
of the inner wall are studded with numerous tubercles, 
a feature which had not before been noticed. 
In the intervallum are numerous and varied sponge 
spicules; none of of these are exposed within the central 
cavity. C shows another small fragment of the same 
specimen, the outer wall being here exposed from its 
inner surface. 
Sketch E shows a number of the sponge spicules from 
the intervallum of the specimen; they are drawn as 
exposed and allowance must be made for parts which 
have been lost and for parts which are still buried in 
the. matrix. 
a. Insufficient remains to determine number of rays. 
b. Apparently four rays in one convex surface and 
another ray perpendicular. c, e, f. Multiradiate. 
d. Perhaps six rays in one convex surface but insufficient 
to be sure. g. Apparently four (possibly five) rays in 
one plane and a ray of considerable length perpendicu- 
lar. h. Six rays in one plane (if another ray perpen- 
dicular to these; it is not exposed. j. Only two rays 
exposed, which suggest a gate of six rays in one 
plane. 
Are these spicules intergral parts of the fossil? or are 
they there by accident? In favor of the former alterna- 
live we have seldom found sponge spicules within the 
intervallum or central cavity of other Archaeos; also the 
fact that in this specimen the spicules appear confined to 
the intervallum and none in the central cavity, which 
