Sudden transitions, by producing too much effect, the lights being too light, and the darks too dark, produce a hard, dry, a staring, and a vulgar appearance, for want of neutralizing their qualities, and bringing the parts more in union with each other. This overwrought manner is principally the cause of that common look so identified with the modern French school, the effect of too much relief.
On the other hand, nothing but flatness and insipidity is the result of too softly blending and uniting the light with the shadow, and the parts with one another, without that distinction and solidity constituting the arrangement that should bring the near and the remote together, in the treatment of the intermediate relations.
Light should be so skilfully woven into the shadow, as not to prejudice, but assist its depth by its intrusion; this is of most essential consequence.
It is not necessary that the light should come in at one side of the picture, nor pass out at the other, as has been asserted. It is, perhaps, better to attach ourselves to no particular theory: few theorists are good painters; their works, in general, bear a contradictory proportion to the opinions set forth in their speculations.
Sketching light and shade from nature (with a single colour, or a stump), teaches us to profit by every circumstance, natural or accidental. And these sketches, studied at home, teach us, in turn, at once to compose, and to extend the sphere of our observation;—it carries us to the doctrine of probable possibilities; and invests the meanest subject with attraction; the most infinite variety becomes simplicity upon these terms.
The light and shade of a picture should never bear the same proportions; it should, in all instances, differ materially in quantity; a repetition of forms should always bear a different proportion in size, the one having a decided superiority over the other, or, the inevitable consequences will be, confusion.
Unconnected lights and shadows, that are too much defined, will have a bald, a chequered, or draught-board appearance.
In sketching from nature, I usually commence by rubbing in the effect first, and adding the details, or features of the scene, afterwards; mostly beginning with the centre, or else the point of sight, and working outwards, and upwards, and downwards, to the sides of the picture. But this can only come of extensive practice, or, at least, a power of grasping the whole at once.
I have said that the first and principal part of art is Composition, or placing things together appropriately; the situation, motion, and expression of the figures; their shapes, and lights, and shadows, according. A perfect outline is of most consequence, and can only be acquired by study. Next to this, the situation, colour, and quantity of shadows; these being infinite, may be variously managed. At the same time, it requires much more observation and study to shade a picture, than to merely draw the lines of it. No fixed rule can be given for this; but, after having got the outline free and flowing, endeavour, by various trials, on other bits of paper, to leave the masses of shadow and light broad, so as to convey an appearance of space and extent. In the infinite gradations of shade, and the blending of them, nature has no determined law.
Objects out of doors, which receive the general light of the sky, and where the surrounding air gives light on all sides, will look altogether different from the same objects drawn and shaded in a room, which would give dark shadows where in nature there are none. ([Plate 3, fig. 2.])