of such a sky
as that represented in the second diagram, so far as it can be
abstracted or conveyed by painting at all, implies the total
absence of any pervading warmth of tint, such as artists usually
call 'tone.' Every tint must be the purest possible, and above all
the white. Partly, lest you should think, from my treatment of
these two phases of effect, that I am insensible to the quality of
tone,--and partly to complete the representation of states of
weather undefiled by plague-cloud, yet capable of the most solemn
dignity in saddening color, I show you, Diagram 3, the record of an
autumn twilight of the year 1845,--sketched while I was changing
horses between Verona and Brescia. The distant sky in this drawing
is in the glowing calm which is always taken by the great Italian
painters for the background of their sacred pictures; a broad field
of cloud is advancing upon it overhead, and meeting others
enlarging in the distance; these are rain-clouds, which will
certainly close over the clear sky, and bring on rain before
midnight: but there is no power in them to pollute the sky beyond
and above them: they do not darken the air, nor defile it, nor in
any way mingle with it; their edges are burnished by the sun like
the edges of golden shields, and their advancing march is as
deliberate and majestic as the fading of the twilight itself into a
darkness full of stars.

These three instances are all I have time to give of the former
conditions of serene weather, and of non-electric rain-cloud. But I
must yet, to complete the sequence of my subject, show you one
example of a good, old-fashioned, healthy, and mighty, storm.

In Diagram 4, Mr. Severn has beautifully enlarged my sketch of a
July thundercloud of the year 1858, on the Alps of the Val
d'Aosta, seen from Turin, that is to say, some twenty-five or
thirty miles distant. You see that no mistake is possible here
about what is good weather and what bad, or which is cloud and
which is sky; but I show you this sketch especially to give you the
scale of heights for such clouds in the atmosphere. These thunder
cumuli entirely _hide_ the higher Alps. It does not, however,
follow that they have buried them, for most of their own aspect of
height is owing to the approach of their nearer masses; but at all
events, you have cumulus there rising from its base, at about three
thousand feet above the plain, to a good ten thousand in the air.

White cirri, in reality par