OF THE FOSSIL PLANTS OF THE COAL-MEASURES. 
523 
cavity being '01. Its superficial protuberances are not drawn out into acute spines, 
but are short, obtuse tubercles, which are irregularly distributed over the surface of 
the hemisphere. Fig. 75 exhibits a combination of these short tubercles with elon¬ 
gated spines in the same individual, hence this specimen may be regarded as a largely 
modified example of the group which I have designated C. spinosa. 
The only additional example of the Welsh series to which I would call attention is 
that represented by fig. 80, but as seen under a one-sixth objective. It is enlarged 
190 diameters. It may belong to the type of S. spinosa, but that is not certain. Its 
importance is found in the minute but obvious foramination of its sphere-wall, a 
condition that readily explains the surface structure of such examples as fig. 67. 
Such are the objects which Professor Judd believes to be Carboniferous Radiolarians 
—a conclusion which neither I nor my experienced friend Henry Brady, F.B.S., are able 
to accept. In support of this determination I would call attention to some specimens, 
myriads of which constitute almost the mass of a “ Corniferous limestone ” from the 
Devonian beds of Kelly’s Island, U.S.A., for specimens of which I am indebted to Mr. 
Brady, and which we both believe to be closely related to the Welsh organisms. These 
objects have also been spherical bodies, having a diameter of from '05 to '04, Like 
the Welsh specimens, they are more opaque than the mean of the surrounding matrix, 
when viewed by transmitted light, and more brightly wdrite when examined by reflected 
light. The limestone consists almost entirely of perfect examples and fragments 
of these objects, the intervals between these being chiefly occupied by a translucent 
crystalline carbonate of lime. Each organism has been a hollow sphere. The sphere- 
wall has been much thicker in proportion to its entire diameter than is the case 
among the Welsh specimens. Externally, the transverse section of each sphere pre¬ 
sents an undulating outline, due to the intersection of prominences and ridges that 
characterise its surface. Sometimes these projections surround the entire section; 
but more frequently, as is the case with fig. 81, they are absent from limited 
portions of the periphery. Occasionally these ridges may be seen pursuing an oblique 
direction like the bands crossing the nucules of a Chara. The central cavity is 
always occupied by crystalline infiltrated carbonate of lime. Though the sphere- 
wall often exhibits a granular texture, I discover a radiating structure in a sufficient 
number of the specimens to convince me that, in this respect, they have closely 
resembled some of the Welsh objects. Since I cannot learn that this American form 
has received a name, it may be designated Calcisphcera robusta. Whilst I am 
thoroughly satisfied that these objects are not Radiolarians, it is not easy to say what 
they are. Like the Traquarice, they are altogether different from all known Radio¬ 
larians. The C. robusta constitutes the chief part of the material of the Corniferous 
limestone, which appears to have been as much indebted to them for its calcareous 
matter as the chalk is to the Foraminifera. Hence in this case any idea of a substitu¬ 
tion of calcareous matter for silica is out of the question. They are, and obviously 
have always been, calcareous organisms, and the Welsh examples present many 
