Volcanoes of Auvergne, 53 



from Auvergne appears to agree very closely — there being a pre- 

 dominance, according to Messrs Croizet, Bravard, and Pomel, of 

 species not known to exist at present, with an intermixture of a few 

 others undistinguishable from quadrupeds now inhabiting Europe. 

 Among the landslioUs associated with the bones, were found Cyclo- 

 stoma elegansy Clausilia rugosa, Helix hortensis, H. nemoralis, H, 

 lapicida Siiid H. ohvoluta — all recent, and all, with the exception of 

 the last, now found in the immediate neighbourhood. Mr Lyell 

 thinks it probable that the deposit of red argillaceous sand under the 

 lava containing these remains, was derived chiefly from volcanic 

 matter, which the eruption of Tartaret threw out, and that the fossil 

 animals perished by floods occasioned by that outburst. That a similar 

 fauna continued to live in Auvergne after the latest eruptions, is in- 

 ferred from the discovery of the remains of many of the same group 

 of animals — spermophiius, lepus, castor and others, in the clefts of a 

 lava-current as modern as that of Tartaret, observed at Aubier, near 

 Clermont. This fauna, so diflerent as a whole from that now living 

 in Europe, evidently inhabited Auvergne when the valley of the Couze 

 had been excavated down to the same level as that over which the 

 lava of Tartaret flowed : yet its antiquity must be extremely great — 

 the gradual dying-out of species, and the introduction of new ones 

 taking place, according to Mr Lyell's views, with extreme slowness. 

 The fact that the shells belonged all to living species (which possibly 

 might not hold good if a larger number were obtained) affords no 

 presumption against an indefinitely remote origin, as compared to the 

 periods of history and tradition, because the lecturer has shewn that 

 the ravine of the Niagara (" Travels in JST. America," vol. i. ch. 2), 

 and the Delta of the Mississippi (Reports of the Brit. Assoc, for 

 1846), both of which must have required an enormous period for 

 their formation, are, nevertheless, posterior in date to deposits full of 

 the recent land and fresh-water shells of North America, associated 

 with the remains of quadi'upeds, nearly all of which are now extinct. 

 It was next shewn that all the volcanoes of the modern class of 

 which the Puy de Tartaret is a type, were not formed at once, for the 

 lavas of some (as for example, at Champheix, in the same valley of 

 the Couze) stand at a greater height above the actual river-courses, 

 and repose on ancient alluvium formed when the valleys were shal- 

 lower. To allow time for the ejection of these numerous cones and 

 lava-currents, of which there are several hundreds in Central France, 

 we require a long series of ages, all subsequent to the miocene period, 

 to which another class of monuments of anterior date are referable — 

 as, for example, the bone-bearing alluviums alternating with volcanic 

 formations (pumiceous and trachytic) of Mont Perrier, to which a 

 distinct fauna (of the genera mastodon, elephant, hippopotamus, tapir, 

 &c.) belongs. Some of the valleys cut out of the still more ancient 

 lacustrine strata were only half eroded to their present depth in the 



