The nomenclature of the fishes of the Old Red Sandstone of Great Britain, with the exception of the Cephalaspidæ, revised some years ago by Professor Lankester, is at present in a very unsatisfactory state. A vary large number of the species named by Agassiz, as well as by McCoy, were undoubtedly founded upon deceptive characters, due partly to different modes of preservation in different rocks, partly also to those apparent variations in external form, which are inevitable in such ancient fossil fishes devoid of a fully ossified internal framework, without which the original outline cannot be expected to be constantly preserved. In specimens from one locality the external ganoid surface of the scales may be well shown, in those from another it may be constantly hidden or obscured, while the proportional measurements in the very same species may vary infinitely, by the fish being lengthened out, or shortened up by changes, which have occurred after death or during the consolidation of the enclosing rock. These and kindred sources of fallacy can only be guarded against by long experience in deciphering such remains, coupled with the examination of an immense number of specimens.