2 
Diagnosis. Corallum massive hemispherical, irregularly spherical or 
encrusting. Corallites polygonal or circular; walls composed of fused 
septal elements, with pores on vertical and horizontal rows, alternating 
with the septa. Septa typically twenty, wedge-shaped, radiating, of equal 
length, short, usually extending one-fifth or less of the diameter into the 
corallite, inclined slightly upward. Tabulse well developed, complete and 
incomplete. No dissepiments. A coenenchyme may or may not be present; 
if it is, twenty “ costse radiate from the corallites. Buds arise from 
between corallites. 
Remarks. No types were mentioned by Billings. Lindstrom in 1883 
selected C. anticostiensis as the genotype; thus the specimen C. anticos- 
tiensis Billings, “West side of Gamache bay, Anticosti island; Division 1, 
Anticosti group; collected by T. C. Weston,” accepted by the Geological 
Survey, Canada, as Billings’ type of this species and in its collection at 
Ottawa (G.S.C. 2267), becomes the genolectotype. Miller’s selection of 
C. canadensis as genotype in 1889, which was followed by Bassler (1915), 
is thus anticipated. 
The meaning of Billings’ genus is quite apparent from his diagnosis 
and descriptions, and the diagnosis given above differs only in those points 
that can be discerned by more recent methods of study. Billings states 
(page 425) that the corallites are “ perforated as in Favosites and their 
outside striated by imperfectly developed costse.” It will be shown below 
that the perforation of the walls differs from that in the Favositidae, and 
although “ costse ” are present in some members of the genus they are not 
found in all and, therefore, their mention has no place in the generic 
diagnosis. The original diagnosis has, further: “ R,adiating septa (in the 
species at present known) about twenty-four.” The writer is able to 
state that in all the specimens of this genus that he has examined the 
number of septa is constantly twenty. A count of the “ costse ” (continu- 
ous with the septa) in the genolectotype, which Billings must have had 
before him, nowhere reveals more than this number. It is not always 
simple to count the septa at the surface, this is best done in thin sections, 
and the fact that some of the material that Billings examined was silicified 
probably also increased the inaccuracy. 
Synonymy. Nicholson in 1874 described the genus Columnopora 
“ for the reception of a single species of coral, of which I have seen two 
well-preserved examples — one collected from the Hudson River Group of 
Canada by Mr. George Jennings Hinde; the other obtained from the 
Cincinnati Group (Hudson River formation) by Mr. U. P. James” (253). 
The former of these came from Credit river, Ontario, and the latter from 
near Cincinnati, Ohio. In 1875 the same facts were again published in 
two papers by Nicholson, but more material had come to his notice. 
Foerste (1916, 293) states, “The first figured specimen of this species 
(Figure 8 in the Paleontology of Ohio) at present forms No. 216 in the 
James collection in the Museum at Chicago University.” Unfortunately 
the writer has been unable either to borrow or to see the original of this. 
However, he has sectioned and examined the topotype material in the 
^ The use of the term “ costa ” is explained in the footnote on page 22. 
