He refers in an appreciative way to the first special treatise on fossil shells ever published, that of an Englishman named Brander,[84] who collected the shells "out of the cliffs by the sea-coast between Christ Church and Lymington, but more especially about the cliffs by the village of Hordwell," where the strata are filled with these fossils. Lamarck, working upon collections of tertiary shells from Grignon and also from Courtagnon near Reims, with the aid of Brander's work showed that these beds, not known to be Eocene, extended into Hampshire, England; thus being the first to correlate by their fossils, though in a limited way to be sure, the tertiary beds of France with those of England. How he at a later period (1805) regarded fossils and their relations to geology may be seen in his later memoirs, _Sur les Fossiles des environs de Paris_.[85] "The determination of the characters, both generic and specific, of animals of which we find the fossil remains in almost all the dry parts of the continents and large islands of our globe will be, from several points of view, a thing extremely useful to the progress of natural history. At the outset, the more this determination is advanced, the more will it tend to complete our knowledge in regard to the species which exist in nature and of those which have existed, as it is true that some of them have been lost, as we have reason to believe, at least as concerns the large animals. Moreover, this same determination will be singularly advantageous for the advancement of geology; for the fossil remains in question may be considered, from their nature, their condition, and their situation, as authentic monuments of the revolutions which the surface of our globe has undergone, and they can throw a strong light on the nature and character of these revolutions." This series of papers on the fossils of the Paris tertiary basin extended through the first eight volumes of the _Annales_, and were gathered into a volume published in 1806. In his descriptions his work was comparative, the fossil species being compared with their living representatives. The thirty plates, containing 483 figures representing 184 species (exclusive of those figured by Brard), were afterwards published, with the explanations, but not the descriptions, as a separate volume in 1823.[86] This (the text published in 1806) is the first truly scientific palæontological work ever published, preceding Cuvier's _Ossemens fossiles_ by six years. When we consider Lamarck's--at his time unrivalled--knowledge of molluscs, his philosophical treatment of the relations of the study of fossils to geology, his correlation of the tertiary beds of England with those of France, and his comparative descriptions of the fossil forms represented by the existing shells, it seems not unreasonable to regard him as the founder of invertebrate palæontology, as Cuvier was of vertebrate or mammalian palæontology.