mals, birds, or reptiles; but when comparative anatomy became better understood, it was perceived that their relations to mammals and birds were only in external forms, while the essential features of their structure were undeniably reptilian. Every one has heard of flying dragons, reptiles which, like "flying fishes" and "flying squirrels," are able partially to sustain themselves in the air by means of parachute-like expansions from their bodies. But in the Pterodactyls were true aerial reptiles, as bats are genuine flying mammals (see Fig. 72). The Pterodactyl, in the length of its neck and form of its head, resembled a bird. The trunk and tail were like those of a quadruped. The numerous conical recurved teeth were formed after the Saurian type. The anterior extremities were constructed after the character of bats, the last finger having been greatly elongated, and adapted for supporting a membranous wing, the impression of which is sometimes preserved in connection with the bones. We know twenty species of this remarkable order, all Old-World marvels save a single pair of long finger-bones found at Phoenixville in Pennsylvania. Some were no larger than a snipe, while others were capable of expanding their wings to a breadth of sixteen feet. Along the valley of the Connecticut River, from the neighborhood of New Haven to the northern part of Massachusetts, is a brownish-red sandstone, resting mostly in horizontal beds, which have been extensively quarried for building purposes. On the banks of the river at Portland, opposite Middletown, are excavations several acres in extent, which have been in progress more than a hundred years. Thousands of ship-loads have been sent down the Connecticut, and built into the aristocratic brown stone fronts of New York. This formation furnished a valuable resource to the earliest settlers of Connecticut. Their names and virtues are commemorated on the brown stone slabs still standing in the oldest cemeteries. The sheets of this formation were spread out in an elongated depression in the surface of the older and underlying formations. On each side of this belt of horizontal sandstones we reach a limiting wall of tilted gneisses and bubbling granites. These were the land while the waters of Long Island Sound stretched through an estuary up to New Hampshire, and received there the waters of the embryo Connecticut. If the student of the world's history will go to the Portland quarries, he will see, spread over the ground in the vicinity of some of the offices, slabs large and small, bearing traces like the imprints of the feet of birds. These track-bearing layers of the rock are found at all depths in the quarry. The formation is generally believed to belong to the later Triassic or earliest Jurassic (Fig. 73). The ornithic character of the footprints has been strenuously argued by Dr. Deane, the discoverer, and Professor Hitchcock, the first describer of these ichnolites. This opinion has been supported by the weight of such names as Buckland, Lyell, Mantell, and Forbes; but all observations hitherto made on the distribution of organic types through geological time tend toward the general principle that every class-type of vertebrates, and every ordinal type of invertebrates, has been introduced upon the earth in the line of succession indicated by its rank, and there is an à priori improbability of the existence of so high a type of organization as we find in birds-and birds of the size that these must have been-at a time when the reptilian type had scarcely reached its culmination. Moreover, the Pterodactyls have made us acquainted with the existence and characters of bipedal reptiles in the very age when the bipedal footprints of the Connecticut sandstone were impressed. It should be noted, also, |