, it will
give you as much help as you need in the linear expression of
ground-surface. Strive to get the retirement and succession of masses in
irregular ground: much may be done in this way by careful watching of
the perspective diminutions of its herbage, as well as by contour; and
much also by shadows. If you draw the shadows of leaves and tree trunks
on any undulating ground with entire carefulness, you will be surprised
to find how much they explain of the form and distance of the earth on
which they fall.

148. Passing then to skies, note that there is this great peculiarity
about sky subject, as distinguished from earth subject;--that the
clouds, not being much liable to man's interference, are always
beautifully arranged. You cannot be sure of this in any other features
of landscape. The rock on which the effect of a mountain scene
especially depends is always precisely that which the roadmaker blasts
or the landlord quarries; and the spot of green which Nature left with a
special purpose by her dark forest sides, and finished with her most
delicate grasses, is always that which the farmer plows or builds upon.
But the clouds, though we can hide them with smoke, and mix them with
poison, cannot be quarried nor built over, and they are always therefore
gloriously arranged; so gloriously, that unless you have notable powers
of memory you need not hope to approach the effect of any sky that
interests you. For both its grace and its glow depend upon the united
influence of every cloud within its compass: they all move and burn
together in a marvelous harmony; not a cloud of them is out of its
appointed place, or fails of its part in the choir: and if you are not
able to recollect (which in the case of a complicated sky it is
impossible you should) precisely the form and position of all the clouds
at a given moment, you cannot draw the sky at all; for the clouds will
not fit if you draw one part of them three or four minutes before
another.

149. You must try therefore to help what memory you have, by sketching
at the utmost possible speed the whole range of the clouds; marking, by
any shorthand or symbolic work you can hit upon, the peculiar character
of each, as transparent, or fleecy, or linear, or undulatory; giving
afterwards such completion to the parts as your recollection will enable
you to do. This, however, only when the sky is interesting from its
general aspect; at other times, do not try to draw all t