Correspondence 63 
and two lateral dorsal plates of the same species of Tracostens clarki. 
There are two mandibles set with teeth in position close and sharp, round 
in section, smooth, and about one-eight of an inch long. The head-shield 
shows the orbits, four inches in diameter, complete. There are also 
fragments of a second with the head bones separate and four tubercular 
plates, each tvibercle of which ends in a spinous point. The upper pre- 
maxillarj, part of which only is present, ends in two hooklike teeth. 
The largest tuberculated plate measures seven inches by nine, and the 
specimen is about fifteen inches over all in Avidth. The most remarka- 
ble "find," however, of 1887, consists of two fossils so different from any- 
thing yet known as apparently to be the types of a new family. The first 
of these has a very elongated narrow body, resembling that of a gar-pike 
in general form. The snout is sharp, with a projecting tip and narrow, 
with long slender jaw bones, set with minute teeth of cladodont type, but 
very small. The front of the mandible is furnished with a large cutting 
tooth. The body was covered with very narrow, lozenge-shaped ganoid 
scales. Only two fins are present, set well forwai-d, behind the head, 
with rounded ends and very strong rays. This specimen measures seven 
inches from tip to tip of these rays. Posteriorly, the body which is about 
twenty inches long tapers slightly with indications of a small caudal fin, 
but the actual end has not been preserved. 
The second fossil is broader and shorter, with heavier jaws, stronger, 
coarser, and more numerous fin-rays, and measures eight inches over 
the two fins, by twenty-four inches in length. It shows some signs of 
an ossified spinal column. The ganoid scales have projecting points 
which fit into grooves in those adjoining them. Of this form there are 
several specimens, showing the head and fins. The range in size of the 
species, is shown by one of these, which measures twenty-four inches 
across the fins, each of which is nine inches long. The hinder end, which 
is well preserved, apparently terminates in an expanded and laterally 
flattened tail (as the flukes of the whale), showing some indication of a 
caudal fin. 
Dr. Clark has also obtained two crania of a Dinichthys, apparently 
new, and four mandibles, (one nearly perfect,) two premaxillary teeth 
five inches long, and four lateral or cutting teeth four and a half inches 
long; a pair of ventral plates, a pair of clavicles (.'), a pair of unknown 
bones found with the rest, and a large number of plates, probably belong- 
ing to D. TirelU or to D. Herzert, some of which have been entirely un- 
known hitherto. 
In the same collection is a new {}) placoderm with mandible about 
twenty inches long and dorsal plates measuring nine inches by seven, 
having a very short crest and no neck, thus differing from those of 
Dinichthys. 
All the above were found in the Cleveland shale of the Ohio Geological 
Survey. Those yet to be mentioned came from the Cuyahoga shale and 
are (1) a fossil about two and a half feet long, the jaws of which show 
cladodont teeth, but this, if not new, has not yet been identified. (2.) A 
