This is, however, an illusion. A long time ago it was proved by Laplace, that tides are caused by the great weight of a liquid. If the ocean were filled with mercury instead of water, the tides would reach a formidable height. Tides, then, do exist in the atmosphere, but in comparison much less than in the water, because the air is so much lighter. It happens that we do not live on the surface of the atmosphere, but in the lowest strata of this airy sea; and in these strata, where the weather manifests itself, the effect of the tides in the upper air is so insignificant, that nothing of it has yet been discovered in spite of most diligent barometer observations. Learned men have had such a respect for this popular belief, that thorough observations and investigations have been made in order to settle the question. Those investigations were of three kinds: 1st. What influence with regard to heat and cold has the nearness or remoteness of the moon upon our weather? 2d. What influence has the same upon rain or dryness in the atmosphere? 3d. Has the change of the moon any bearing upon the variability of our weather? For the reply to these questions, some naturalists have made use of the minutest observations for a period of nearly forty years; during which time the temperature, pressure, and moisture of the air have been measured daily. These observations have been scrupulously examined; the conclusion arrived at is, that the moon is not quite without influence upon the state of our atmosphere; but this influence is so very small, that it is not brought to bear at all on meteorology. When the moon is nearest to the earth, it is certainly a little colder than when she is farther off; but the decrease of heat amounts in the average scarcely to one-fifth of a degree, and this is a quantity entirely imperceptible in our weather. As to rain, it is a little less frequent in the time of the moon's greatest distance from the earth; but this difference, too, is imperceptibly small. In one thousand rain-storms there are four hundred and eighty-eight during the moon's greatest distance, five hundred and twelve during her nearest. As to the pressure of the air, it is during the moon's greater distance somewhat greater than when she is nearer, but this difference is still less than the preceding ones, so much so that a common barometer does not even indicate it. The most thorough investigations have been made about the influence of the waxing and waning moon upon the weather, because it was on this subject that the greatest illusion prevailed. The result here is likewise, that scarcely any difference exists, and that it is a mere superstition for people to maintain, that when the moon changes, the weather changes also. The change in the moon, moreover, does not take place all of a sudden, but with great regularity from day to day, from minute to minute; while the weather, especially with us, changes often very abruptly.