This section is from the book "Arts & Crafts Magazine Vol1-2", by Hutchinson & Company.
If you do not understand thoroughly the principles of perspective, be very careful about admitting objects in your sketch that do not belong to nature itself. Buildings, bridges, etc, must not be introduced if they are prone to betray you into error. You might go on for a long time sketching natural scenery without making use of mathematical perspective; still, if you have little or no knowledge of the subject, you are likely to meet with embarrassments. There is no excuse for deficiency in this direction, for a few weeks of intelligent application to the subject will make you fairly competent.
All landscape involves aerial perspective, but this is not comprehended by means of rules, but by faithful observation and critical perception.
There is a common dogma among most modern writers on colour that "warm colours" (reds and yellows) "approach" or express nearness, and "cold colours" (blue and grey) "retire" or express distance. Ruskin positively denies this. Colours, as such, he declares, are absolutely inexpressive as to distance. It is their quality (as depth, delicacy, etc.) which expresses distance, not their tint. "A blue bandbox set on the same shelf with a yellow one will not look an inch further off, but a red or orange cloud in the upper sky will always appear to be beyond a blue cloud close to us, as it is in reality. It is quite true that in certain objects blue is a sign of distance; but it is not because blue is a retiring colour, but because the mist in the air is blue, and therefore any warm colour which has not strength enough to pierce the mist is lost or subdued in the blue; but blue is no more, on this account, a 'retiring colour' than brown is a retiring colour, because when stones are seen through brown water, the deeper they lie the browner they look; or than yellow is a retiring colour, because when objects are seen through a London fog, the further off they are the yellower they look. Neither blue nor yellow nor red can have, as such, the smallest power of expressing either nearness or distance; they express them only under the peculiar circumstances which render them at the moment or in that place signs of nearness or distance. Thus, vivid orange in an orange is a sign of nearness, for if you put the orange a great way off, its colour will not look so bright; but vivid orange in sky is a sign of distance, because you cannot get the colour of orange in a cloud near you. So purple in a violet or a hyacinth is a sign of nearness, because the closer you look at them the more purple you see. But purple in a mountain is a sign of distance, because a mountain close to you is not purple but green or grey. "It may, indeed, be generally assumed that a tender or pale colour will more or less express distance, and a powerful or dark colour nearness; but even this is not always so. Heathery hills will usually give a pale and tender purple near and an intense and dark purple far away; the rose colour of sunset on snow is pale on the snow at your feet, deep and full on the snow in the distance; and the green of a Swiss lake is pale in the clear waves on the beach, but intense as an emerald in the sunstreak six miles from the shore. And in any case, when the foreground is in strong light, with much water about it, "or white surface, casting intense reflections, all its colours may be perfectly delicate, pale and faint: while the distance, when it is in shadow, may relieve the whole foreground with intense darks of purple, blue green or ultramarine blue. So that, on the whole, it is quite hopeless and absurd to expect any help from laws of aerial perspective." There is, however, one law about distance which Mr. Ruskin thinks has some claims to be considered a constant one - namely, that dulness and heaviness of colour are more or less indicative of nearness. All distant colour is pure colour: it may not be bright, but it is clear and lovely, not opaque nor soiled; for the air and light coming between us and any earthy or imperfect colour purify or harmonize it. R. Jervis.
Take care of your sketches. Date each as you make it, for nothing is more interesting than the review of your progress by means of a consecutive series of such work. Artists up to a certain point are proverbially reckless in regard to their sketches. But as time brings wisdom, they regret their carelessness, and arrange them with care, often when most of them are injured beyond redemption,