1884. ] in the Silurian Rocks of North America. 1225 
of the rock in the county but in only a few places. Probably if 
it were more accessible they would prove to be more abundant. 
They are seldom in a good state of preservation, but fragments 
capable of exhibiting the structure are abundant in some places. 
From the above facts it may be inferred that the Pennsylvania 
fossil, Palæaspis, is somewhat older than the unique Scaphaspis 
ludensis of the English Lower Ludlow, and therefore consider- 
ably older than the specimens of Scaphaspis and Cyathaspis ? that 
have been found in the Upper Ludlow. 
A thousand feet lower down, in the middle of the red shale, I 
have met with a thin bed full of comminuted fish-scales or shields 
resembling in many points those of Paleaspis. Their matrix 
presents some resemblance to coprolites, but their condition is 
too imperfect to allow of description. 
Nor is this quite the whole. Five hundred feet lower still, be- 
low beds indisputably of Clinton age, as determined by their fos- 
sils, among which are Calymene clintont, Beyrichia lata, Hormo- 
ceras vertebratum, lies the well-known iron sandstone of the Clin- 
ton group of Pennsylvania. In this is a thin layer thickly charged 
_ with comminuted scales in much better condition than those in 
the red shale. With these occurs a spine somewhat like those 
from the Bloomfield sandstone, and which I have named Ouchus 
clintoni. : 
Associated with these are great numbers of small pellets which 
Present all the characters of coprolites, and may be referred almost 
with certainty to that group of objects, being in great part com- 
Posed of phosphate of lime. 
: It is evident that in these fossils we have the most ancient rel- 
ics of vertebrate life yet known from any part of the world, At 
the same time, though carrying the existence of fishes down 
_ through at least 1500 feet of strata and back through a cone 
‘ponding lapse of time, it must be added that these fossils do lit- 
tle or nothing to bridge over the chasm existing between a 
vertebrates and the invertebrates. Odd and aberrant from oa 
type as they are, they cannot logically be excluded from the class 
of fish, and though we know nothing of their soft parts, it = ar 
Sonable to believe they were arranged as in their nearest allies 
Which we know anything, Cephalaspis and Coccosteus. 
_ Yet on the other hand, though these particular fossi ete 
‘Mo light on the connection between vertebrates and “Ms ean 
. > an indisputable fact that the whole group to which they 
Se pid oor E oe ais hl el met Pe Eee ee r et Sr Te I R ed eS! ec ae -_ 
