70 THE PALEONTOLOGY OF MINNESOTA. 
[Anomalospongia. 
as I believed originally, perhaps triradiate, with a central opening, as in Brachio- 
spongia. Neither view is supported by positive evidence, so that for the present it 
seems best to leave that point entirely open. 
Two of the specimens are depressed-conical in form, one about 50 mm. in 
length, the other only 22 mm. The larger is 20 mm. wide at the large end, its mar- 
gins nearly parallel in the upper half and converging rapidly in the lower half. The 
extreme end may have been closed and pointed, but as both specimens are defective 
here, it would not be safe to assume that it was. Indeed, it is perhaps just as likely 
that a small opening existed in the extremity. The smaller specimen is very nearly 
a duplicate of the lower half of the larger. 
The relative length and disposition of the three horizontal rays are probably 
specific peculiarities, hence are mentioned in the generic diagnosis merely in a gen- 
eral way. In A. reticulata they have a definite form and arrangement, in part, very 
likely controlled by the arrangement of the vertical rays. The latter form straight 
or curved transverse and diagonally intersecting rows, generally very regular, and 
when the horizontal rays are removed by attrition (seemingly a common occurrence) 
they appear as sub-hexagonal rounded knobs, in most cases with ten or eleven in5 mm. 
transversely. In three fragments, otherwise apparently identical with the others, 
the parts are smaller, and in these there are thirteen in that space. Their length, 
and consequently the thickness of the sponge, is commonly about 2.7 mm., but varies 
between the observed extremes of 2.0 and 3.4 mm. 
Since working out the nature of the fossil and its spicular elements, I can detect 
more or less clear evidences of the horizontal rays on most of the specimens.* In 
many the exposed rounded end of the vertical ray preserves a triradiate impression 
of the horizontal rays. In others the rays themselves are preserved but so much 
pressed that their extension beyond the impressed boundary line between the verti- 
cal rays is not to be made out. In the best preserved fragments, however, their 
entire extent, overlapping, and general construction, is shown in as clear a manner 
as can be hoped for in such delicate structures. From the last specimens, a small 
portion of the surface of one of which is represented by fig. 14, it appears that one 
of the horizontal rays is a trifle longer than the other two rays. It is also the one 
most prominent and oftenest seen, aud overlaps except near its extremity. This may 
be called the longitudinal ray, since it lies parallel with the length of the conical 
specimens, while the two others are oblique. When the surface is partly obscured 
by adhering matrix, the first ray alone is likely to be seen clearly. Viewed through 
*These remains of the horizontal rays were noticed by me in the 1878 work on the species, but their nature was misin- 
terpreted because of my erroneous belief that the affinities of the fossils were to be sought for among the Echinodermata. 
Hence the statement in the original description that there is “‘a minute pit on the top for the articulation of two very fine 
and small spines.” 
o 
