No trace of the family has yet been detected in even the Tertiary rocks. [Illustration: Fig. 57 PTERICHTHYS OBLONGUS. (One half nat. size.)] Of the ganoids of the second age of vertebrate existence,--that of the Old Red Sandstone,--some were remarkable for the strangeness of their forms, and some for constituting links of connection which no longer exist in nature, between the ganoid and placoid orders. The Acanth family, which ceased with the Coal Measures, was characterized, especially in its Old Red species, by a combination of traits common to both orders; and among the extremer forms, in which Palæontologists for a time failed to detect that of the fish at all, we reckon those of the genera Coccosteus, Pterichthys, and Cephalaspis. The more aberrant genera, however, even while they consisted each of several species, were comparatively short lived. The Coccosteus and Cephalaspis were restricted to but one formation apiece; while the Pterichthys, which appears for the first time in the lower deposits of the Old Red Sandstone, becomes extinct at its close. On the other hand, some of the genera that exemplified the general type of their class were extremely long lived. The Celacanths were reproduced in many various species, from the times of the Lower Old Red Sandstone to those of the Chalk; and the Cestracions, which appear in the Upper Ludlow Rocks as the oldest of fishes, continue in at least one species to exist still. It would almost seem as if some such law influenced the destiny of genera in this ichthyic class, as that which we find so often exemplified in our species. The dwarf, or giant, or deformed person, is seldom a long liver;--all the more remarkable instances of longevity have been furnished by individuals cast in the ordinary mould and proportions of the species. Not a few of these primordial ganoids wore, however, of the highest rank and standing ever exemplified by their class; and we find Agassiz boldly assigning a reason for their superiority to their successors, important for the fact which it embodies, and worthy, as coming from him, of our most respectful attention. "It is plain," we find him saying, "that before the class of reptiles was introduced upon our globe, the fishes, being then the only representatives of the type of vertebrata, were invested with the characters of a higher order, embodying, as it were, a prospective view of a higher development in another class, which was introduced as a distinct type only at a later period; and from that time the reptilian character, which had been so prominent in the oldest fishes, was gradually reduced, till in more recent periods, and in the present creation, the fishes lost all this herpetological relationship, and were at last endowed with characters which contrast as much, when compared with those of reptiles, as they agreed closely in the beginning. Lepidosteus alone reminds us in our time of these old-fashioned characters of the class of fishes as it was