611. Yellow and blue are so dissimilar, that their contrast is always sufficiently great for their juxtaposition to be favourable, although the juxtaposed colours belong to different scales of yellow and blue. 2. _One colour juxtaposed with another which is not its complementary, may be improved, while the latter may be injured._ A blue, which is improved by a yellow, being placed beside a violet (blue rather than red) may lose some of its beauty, by becoming greenish; while the orange it adds to the violet, neutralising its excess of blue, rather improves than injures it. 3. _Two non-complementary colours may injure each other._ A violet and a blue are reciprocally injurious, when the first greens the second, and the latter neutralizes sufficient of the blue in the violet to make it appear _faded_. It might also happen that although the colours juxtaposed be modified, neither gaining nor losing in beauty, that the one may gain without the other losing; lastly, that the one may neither gain nor lose, while the other loses. 612. _In the association of two colours of equal tone, the depth of the tone may have some influence on the beauty of the association._ For example, a deep indigo-blue and an equally deep red, gain by juxtaposition; the first, by losing some violet, will become a pure blue, the second, in acquiring orange, will become brighter. If we take light tones of the same scales, it may happen that the blue will become too green to be good as a blue, and that the red, acquiring orange, will be too yellow to be a pure red. 613. _In the association of two coloured objects of tones very distant from each other, belonging to the same scale, or to scales more or less allied, the contrast of tone may have a favourable influence upon the beauty of the light tone_; because, in fact, if the latter is not a pure colour, its juxtaposition with the deep tone, upon the whole brightening it, will purify the colour from whatever grey it may have. 614. It is very necessary for the correction of our judgment of these principles on the binary associations of colours, not to lose sight of all that precedes, concerning colours that are “dead,” or without gloss, and that their combination be considered independently of the form of the objects presenting them, for the twofold reason, that _the glossiness of the coloured surfaces, and the form of the bodies which these surfaces bound_, may modify the effect of two associated colours; consequently, the analysis I have made of the optical effects of colours would be incomplete without speaking of the possible influence of these causes. _Influence of Gloss taken into consideration in the Effect of Contrast of two Colours._