§ 15. Its rapidity, however, observe, does not account for its visibility, and this is one of the questions I cannot clearly solve, unless I were sure of the nature of the vesicular vapor. When our breath becomes visible on a frosty day, it is easily enough understood that the moisture which was invisible, carried by the warm air from the lungs, becomes visible when condensed or precipitated by the surrounding chill; but one does not see why air passing over a moist surface quite as cold as itself should take up one particle of water more than it can conveniently--that is to say, invisibly--carry. Whenever you _see_ vapor, you may not inaccurately consider the air as having got more than it can properly hold, and dropping some. Now it is easily understood how it should take up much in the lungs, and let some of it fall when it is pinched by the frost outside; but why should it overload itself there on the hills, when it is at perfect liberty to fly away as soon as it likes, and come back for more? I do not see my way well in this. I do not see it clearly, even through the wet cloth. I shall leave all the embarrassment of the matter, however, to my reader, contenting myself, as usual, with the actual fact, that the hill-side air does behave in this covetous and unreasonable manner; and that, in consequence, when the weather is breaking (and sometimes, provokingly, when it is not), phantom clouds form and rise in sudden crowds of wild and spectral imagery along all the far succession of the hill-slopes and ravines. [Illustration: FIG. 91.] § 16. There is this distinction, however, between the clouds that form during the rain and after it. In the worst weather, the rain-cloud keeps rather high, and is unbroken; but when there is a disposition in the rain to relax, every now and then a sudden company of white clouds will form quite low down (in Chamouni or Grindelwald, and such high districts, even down to the bottom of the valley), which will remain, perhaps, for ten minutes, filling all the air, then disappear as suddenly as they came, leaving the gray upper cloud and steady rain to their work. These "clouds of relaxation," if we may so call them, are usually flaky and horizontal, sometimes tending to the silky cirrus, yet showing no fine forms of drift; but when the rain has passed, and the air is getting warm, forms the true clearing cloud, in wreaths that ascend continually with a slow circling motion, melting as they rise. The woodcut, Fig. 91, is a rude note of it floating more quietly from the hill of the Superga, the church (nearly as large as St. Paul's) appearing above, and thus showing the scale of the wreath. [Illustration: FIG. 92.]