The Notice sur la succession des poissons fossiles dans la série des formations géologiques is a short synthetic essay in which Louis Agassiz sets out the stratigraphic distribution of fossil fishes as an ordered succession. Its subject is not the anatomy of individual species or the biography of a naturalist, but the use of fish remains for identifying and correlating geological formations.
Agassiz argues that fishes are especially valuable guides to the age of strata because their remains occur in many deposits and because species are sharply limited in time. He distinguishes species, which he treats as confined to particular formations, from genera, which may continue through several formations. This distinction allows him to use fish faunas both to recognize formations and to mark their boundaries.
The organizing framework is Agassiz’s division of fishes into four orders based on the structure of their scales: Placoids, Ganoids, Ctenoids, and Cycloids. The notice traces the relative abundance and first appearance of these orders through the geological series. In the oldest fossiliferous rocks, only Placoids and Ganoids are represented; the other orders are absent. Through the Palaeozoic formations, Ganoids and Placoids dominate, with ganoid fishes especially characteristic of the Devonian and Carboniferous. In the Mesozoic formations, Ganoids remain important, while Ctenoids and Cycloids enter the record at higher levels, with Ctenoids prominent in the Cretaceous and Cycloids especially characteristic of the Tertiary. The notice stresses that this is a succession of distinct assemblages rather than a smooth transformation of one species into another.
It then reviews the formations in stratigraphic order, noting for each the characteristic fish types and the absence of types that appear only later. The oldest rocks yield few remains; the Devonian and Carboniferous contain rich ganoid and placoid faunas; the Triassic and Jurassic show changes in the ganoid assemblage; the Cretaceous introduces new ctenoid and cycloid forms; and the Tertiary contains the first abundant cycloid faunas. From these changes Agassiz concludes that geological formations can be recognized by their fish remains and that the fish record displays a law of successive appearance and extinction. Old types disappear and new types appear at definite levels, with no species passing unchanged from one formation to the next.
The notice also makes explicit the methodological consequences of this position. Because genera may cross formation boundaries while species do not, Agassiz treats precise specific identification as essential: a genus alone is too long-lived to date a bed, whereas a species, once correctly determined, can serve as a marker for a particular formation. The fish fauna of a formation is therefore diagnostic as an assemblage—its overall composition, including the absence of types that appear only higher up, carries the chronological information.
A further theme is the relation between zoological rank and stratigraphic position. Agassiz does not present the succession as a simple march from “lower” to “higher” fishes; rather, he uses the scale-based orders to show that different major types dominate at different levels, with older formations characterized by types that are remote from living fishes and younger formations containing forms increasingly resembling the living fauna. This increasing resemblance is treated as a check on the order of formations, not as evidence that one species has been transformed into another.
The notice’s significance lies in turning Agassiz’s ichthyological classification into a stratigraphic instrument. It argues that the order of appearance and disappearance of fish types is general enough to correlate formations and to recognize boundaries where other fossils are scarce. At the same time, it keeps the question of chronological succession separate from the question of ancestry: the record shows distinct faunas replacing one another at definite levels, and the law it formulates is one of successive appearance and extinction rather than gradual derivation.
