On the other hand, the chances of the preservation of land snails in aqueous deposits are vastly less than those in favour of the preservation of aquatic species. The first Carboniferous species found[15] had been preserved in the very exceptional circumstances afforded by the existence of hollow trunks of Sigillariæ on the borders of the Coal formation flats, and the others subsequently found were in beds no doubt receiving the drainage of neighbouring land areas. Still it is not uncommon on the modern sea-shore, anywhere near the mouths of rivers, to find a few fresh-water shells here and there. The carbonaceous beds of the Trias, the fossil soils of the Portland series, the estuarine Wealden beds would seem to be as favourably situated as those of the coal formation for preserving land shells, though possibly the flora of the Mesozoic was less suitable for feeding such creatures than that of the Coal period, and they may consequently have become few and local. After all, perhaps more diligent collecting and more numerous collectors might succeed, and may succeed in the future, in filling this and similar gaps. [15] _Pupa vetusta_ of the Nova Scotia coal formation. It is a great mistake to suppose that discoveries of this kind are made by chance. It is only by the careful and painstaking examination of much material that the gaps in the geological record can be filled up, and I propose in the sequel of this article to note a few instances, in a country where the range of territory is altogether out of proportion to the number of observers, and which have come within my own knowledge. It was not altogether by accident that Sir C. Lyell and the writer discovered a few reptilian bones and a land-snail in breaking up portions of the material filling an erect Sigillaria in the South Joggins coal measures. We were engaged in a deliberate survey of the section, to ascertain as far as might be the conditions of accumulation of coal, and one point which occurred to us was to inquire as to the circumstances of preservation of stumps of forest trees in an erect position, to trace their roots into the soils on which they stood, and to ascertain the circumstances in which they had been buried, had decayed, and had been filled with mineral matter. It was in questioning these erect trees on such subjects and this not without some digging and hammering that we made the discovery referred to.