Why is it that the Matterhorn is subject to these abominable variations of weather? The ready answer is, “Oh, the mountain is so isolated; it attracts the clouds.” This is not a sufficient answer. Although the mountain _is_ isolated, it is not so much more isolated than the neighbouring peaks that it should gather clouds when none of the others do so. It will not at all account for the cloud to which I refer, which is not formed by an aggregation of smaller, stray clouds drawn together from a distance (as scum collects round a log in the water), but is created against the mountain itself, and springs into existence where no clouds were seen before. It is formed and hangs chiefly against the southern sides, and particularly against the south-eastern side. It frequently does not envelop the summit, and rarely extends down to the Glacier du Lion, and to the Glacier du Mont Cervin below. It forms in the finest weather; on cloudless and windless days. I conceive that we should look to differences of temperature rather than to the height or isolation of the mountain for an explanation. I am inclined to attribute the disturbances which occur in the atmosphere of the southern sides of the Matterhorn on fine days,(88) principally to the fact that the mountain is a _rock_ mountain; that it receives a great amount of heat,(89) and is not only warmer itself, but is surrounded by an atmosphere of a higher temperature than such peaks as the Weisshorn and the Lyskamm, which are eminently _snow_ mountains. In certain states of the atmosphere its temperature may be tolerably uniform over wide areas and to great elevations. I have known the thermometer to show 70° in the shade at the top of an Alpine peak more than 13,000 feet high, and but a very few degrees higher 6000 or 7000 feet lower. At other times, there will be a difference of forty or fifty degrees (Faht.) between two stations, the higher not more than 6000 or 7000 feet above the lower. Provided that the temperature was uniform, or nearly so, on all sides of the Matterhorn, and to a considerable distance above its summit, no clouds would be likely to form upon it. But if the atmosphere immediately surrounding it is warmer than the contiguous strata, a local “courant ascendant” must necessarily be generated; and portions of the cooler superincumbent (or circumjacent) air will naturally be attracted towards the mountain, where they will speedily condense the moisture of the warm air in contact with it. I cannot explain the downrushes of cold air which occur on it, when all the rest of the neighbourhood appears to be tranquil, in any other way. The clouds are produced by the contact of two strata of air (of widely different temperatures) charged with invisible moisture, as surely as certain colourless fluids produce a white, turbid liquid, when mixed together. The order has been—wind of a low temperature—mist—rain—snow or hail.(90)