h a sketch will probably secure a higher completion
at last, than if half an hour had been spent in getting a true outline
before beginning.

48. In doing this, however, take care not to get the drawing too dark.
In order to ascertain what the shades of it really are, cut a round
hole, about half the size of a pea, in a piece of white paper the color
of that you use to draw on. Hold this bit of paper with the hole in it,
between you and your stone; and pass the paper backwards and forwards,
so as to see the different portions of the stone (or other subject)
through the hole. You will find that, thus, the circular hole looks like
one of the patches of color you have been accustomed to match, only
changing in depth as it lets different pieces of the stone be seen
through it. You will be able thus actually to _match_ the color of the
stone at any part of it, by tinting the paper beside the circular
opening. And you will find that this opening never looks quite _black_,
but that all the roundings of the stone are given by subdued grays.[10]

49. You will probably find, also, that some parts of the stone, or of
the paper it lies on, look luminous through the opening; so that the
little circle then tells as a light spot instead of a dark spot. When
this is so, you cannot imitate it, for you have no means of getting
light brighter than white paper: but by holding the paper more sloped
towards the light, you will find that many parts of the stone, which
before looked light through the hole, then look dark through it; and if
you can place the paper in such a position that every part of the stone
looks slightly dark, the little hole will tell always as a spot of
shade, and if your drawing is put in the same light, you can imitate or
match every gradation. You will be amazed to find, under these
circumstances, how slight the differences of tint are, by which, through
infinite delicacy of gradation, Nature can express form.

If any part of your subject will obstinately show itself as a light
through the hole, that part you need not hope to imitate. Leave it
white; you can do no more.

50. When you have done the best you can to get the general form, proceed
to finish, by imitating the texture and all the cracks and stains of the
stone as closely as you can; and note, in doing this, that cracks or
fissures of any kind, whether between stones in walls, or in the grain
of timber or rocks, or in any of the thousand other conditions t