We have, in fact, very striking evidence in Gilbert White’s account of the cold weather of December, 1784, which he specially describes as “very extraordinary,” to show that neither our severe nor our average winter weather can differ materially from that which people experienced in the eighteenth century. “In the evening of December 9,” he says, “the air began to be so very sharp that we thought it would be curious to attend to the motions of a thermometer; we therefore hung out two, one made by Martin and one by Dolland” (_sic_, presumably Dollond), “which soon began to show us what we were to expect; for by ten o’clock they fell to twenty-one, and at eleven to four, when we went to bed. On the 10th, in the morning the quicksilver in Dolland’s glass was down to half a degree below zero, and that of Martin’s, which was absurdly graduated only to four degrees above zero, sank quite into the brass guard of the ball, so that when the weather became most interesting this was useless. On the 10th, at eleven at night, though the air was perfectly still, Dolland’s glass went down to one degree below zero!” The note of exclamation is White’s. He goes on to speak of “this strange severity of the weather,” which was not exceeded that winter, or at any time during the twenty-four years of White’s observations. Within the last quarter of a century, the thermometer, on more than one occasion, has shown two or three degrees below zero. Certainly the winters cannot be supposed to have been ordinarily severer than ours in the latter half of the last century, when we find that thermometers, by well-known instrument makers, were so constructed as to indicate no lower temperature than four degrees above zero. Let us return, after this somewhat long digression, to the levelling action of rain and rivers.