132 PALEOZOIC PALEONTOLOGY. 
mm. in length, whose transverse outline, so far as preserved, is semi- 
elliptical. The complete cross-section seems to have been ovate in 
outline, with the siphuncle near the broader side, though this cannot 
be certainly determined from the specimen. In this specimen several 
sutures are preserved, which are .5 mm. apart. In the specimen 
exhibiting the longitudinal section the sutures are slightly more 
distant and the shell is shghtly curved. It is not certain that both 
these specimens belong to the same species, but neither of them is 
well enough preserved for identification. 
ARTHROPODA. 
TRILOBITA. 
ISOTELUS CANALIS Whitf. 
Plate III., Figs. 5-6. 
1886. Asaphus canalis Whitf., Bull. Am. Mus. Nat. Hist., vol. I, p. 
336, pl. 34, figs. 1-8. 
1889. Asaphus canalis Whitf., Bull. Am. Mus. Nat. Hist., vol. IL., 
p-. 64, pls. 11-12. 
This species is represented only by fragments of the free cheeks, 
genal spines and pygidia. The largest fragmentary free cheek ob- 
served indicates an individual with a breadth of 50 mm. or more. 
This specimen, with a fragmentary pygidium of similar size, has been 
illustrated. The species resembles J. gigas of the Trenton limestone, 
but is proportionally much broader, the head and pygidium being 
nearly semi-circular in outline. It also differs from J. gigas in the 
presence of well-developed genal spines, in this respect resembling 
I. megistos. Broken fragments of these spines are the most common 
portion of the trilobite preserved in the limestone at Columbia. Some 
of the specimens, with no part of the remaining portion of the free 
cheeks preserved, are 30 to 40 mm. in length, and the individuals to 
which these large spines belonged must have been at least twice as 
large as the specimens illustrated, as large, in fact, as the nearly 
complete specimen illustrated by Whitfield from Fort Cassin, on Lake 
