or as we hear of a sheep with five legs, and a kid with two heads. But these exceptions, from their comparative rarity, only make the laws more evident. When, therefore, we find, in various parts of our country, beds of shells in such a state of preservation as to lead us to believe that the animals must have lived and died where their remains are now to be seen, we justly infer that the districts where they occur must at one period have been submerged. If the shells belong to fresh-water species, it is plain that they occur on the site of an old lake. If they are marine, we conclude that the localities where they are found no matter how high above the sea must formerly have stood greatly lower, so as to form the ocean bed. To proceed one step further. If the shells are of a southern type, that is, if they belong to species[9] which are known to exist only in wanner seas than our own, we pronounce that at a former period the latitudes of Great Britain must have enjoyed a more temperate and genial climate, so as to allow southern shells to have a wider range northwards. If, on the other hand, they are of an arctic or boreal type, we in the same way infer that our latitudes were once marked by a severer temperature than they now possess, so as to permit northern shells to range farther southwards. This reasoning is strictly correct, and the truth involved forms the basis of all inquiries into the former condition of the earth and its inhabitants. [Footnote 8: Two of these birds (_Neopron pecnopterus_) are stated to have been seen near Kilve, in Somersetshire, in October 1825. One was shot, the other escaped.] [Footnote 9: There is not a little difficulty in reasoning satisfactorily as to climatal conditions, from the distribution of kindred forms. Even in a single genus there may be a wide range of geographical distribution, so that mere generic identity is not always a safe guide. Thus, the elephant now flourishes in tropical countries, but in the glacial period a long-haired species was abundant in the frozen north. I have above restricted myself entirely to _species_ whose habits and geographical distribution are already sufficiently known.] The evidence furnished by the northern shells in the boulder-clay series is, accordingly, of the most unmistakable kind. These organisms tell us that at the time they lived our country lay sunk beneath a sea, such as that of Iceland and the North Cape, over which many an iceberg must have journeyed, and thus they corroborate our conclusions, derived independently from the deep clay and boulder beds and the striated rock-surfaces, as to the glacial origin of the boulder-clay. CHAPTER III.