Cling round the aerial bow with prisms bright, And pleased untwist the sevenfold threads of light.

CANTO I. l. 117.

The manner in which the rainbow is produced was in some measure understood before Sir Isaac Newton had discovered his theory of colours. The first person who expressly shewed the rainbow to be formed by the reflection of the sunbeams from drops of falling rain was Antonio de Dominis. This was afterwards more fully and distinctly explained by Des Cartes. But what caused the diversity of its colours was not then understood; it was reserved for the immortal Newton to discover that the rays of light consisted of seven combined colours of different refrangibility, which could be seperated at pleasure by a wedge of glass. Pemberton's View of Newton.

Sir Isaac Newton discovered that the prismatic spectrum was composed of seven colours in the following proportions, violet 80, indigo 40, blue 60, green 60, yellow 48, orange 27, red 45. If all these colours be painted on a circular card in the proportions above mentioned, and the card be rapidly whirled on its center, they produce in the eye the sensation of white. And any one of these colours may be imitated by painting a card with the two colours which are contiguous to it, in the same proportions as in the spectrum, and whirling them in the same manner. My ingenious friend, Mr. Galton of Birmingham, ascertained in this manner by a set of experiments the following propositions; the truth of which he had preconceived from the above data.

1. Any colour in the prismatic spectrum may be imitated by a mixture of the two colours contiguous to it.

2. If any three successive colours in the prismatic spectrum are mixed, they compose only the second or middlemost colour.

3. If any four succesive colours in the prismatic spectrum be mixed, a tint similar to a mixture of the second and third colours will be produced, but not precisely the same, because they are not in the same proportion.

4. If beginning with any colour in the circular spectrum, you take of the second colour a quantity equal to the first, second, and third; and add to that the fifth colour, equal in quantity to the fourth, fifth, and sixth; and with these combine the seventh colour in the proportion it exists in the spectrum, white will be produced. Because the first, second, and third, compose only the second; and the fourth, fifth, and sixth, compose only the fifth; therefore if the seventh be added, the same effect is produced, as if all the seven were employed.

5. Beginning with any colour in the circular spectrum, if you take a tint composed of a certain proportion of the second and third, (equal in quantity to the first, second, third, and fourth,) and add to this the sixth colour equal in quantity to the fifth, sixth, and seventh, white will be produced.

From these curious experiments of Mr. Galton many phenomena in the chemical changes of colours may probably become better understood; especially if, as I suppose, the same theory must apply to transmitted colours, as to reflected ones. Thus it is well known, that if the glass of mangonese, which is a tint probably composed of violet and indigo, be mixed in a certain proportion with the glass of lead, which is yellow; that the mixture becomes transparent. Now from Mr. Galton's experiments it appears, that in reflected colours such a mixture would produce white, that is, the same as if all the colours were reflected. And therefore in transmitted colours the same circumstances must produce transparency, that is, the same as if all the colours were transmitted. For the particles, which constitute the glass of mangonese will transmit red, violet, indigo, and blue; and those of the glass of lead will transmit orange, yellow, and green; hence all the primary colours by a mixture of these glasses become transmitted, that is, the glass becomes transparent.

Mr. Galton has further observed that five successive prismatic colours may be combined in such proportions as to produce but one colour, a circumstance which might be of consequence in the art of painting. For if you begin at any part of the circular spectrum above described, and take the first, second, and third colours in the proportions in which they exist in the spectrum; these will compose only the second colour equal in quantity to the first, second, and third; add to these the third, fourth, and fifth in the proportion they exist in the spectrum, and these will produce the fourth colour equal in quantity to the third, fourth, and fifth. Consequently this is precisely the same thing, as mixing the second and fourth colours only; which mixture would only produce the third colour. Therefore if you combine the first, second, fourth, and fifth in the proportions in which they exist in the spectrum, with double the quantity of the third colour, this third colour will be produced. It is probable that many of the unexpected changes in mixing colours on a painter's easle, as well as in more fluid chemical mixtures, may depend on these principles rather than on a new arrangement or combination of their minute particles.

Mr. Galton further observes, that white may universally be produced by the combination of one prismatic colour, and a tint intermediate to two others. Which tint may be distinguished by a name compounded of the two colours, to which it is intermediate. Thus white is produced by a mixture of red with blue-green. Of orange with indigo-blue. Of Yellow with violet-indigo. Of green with red-violet. Of blue with Orange-red. Of indigo with yellow-orange. Of violet with green-yellow. Which he further remarks exactly coincides with the theory and facts mentioned by Dr. Robert Darwin of Shrewsbury in his account of ocular spectra; who has shewn that when one of these contrasted colours has been long viewed, a spectrum or appearance of the other becomes visible in the fatigued eye. Philos. Trans. Vol. LXXVI. for the year 1786.

These experiments of Mr. Galton might much assist the copper-plate printers of callicoes and papers in colours; as three colours or more might be produced by two copper-plates. Thus suppose some yellow figures were put on by the first plate, and upon some parts of these yellow figures and on other parts of the ground blue was laid on by another copper-plate. The three colours of yellow, blue, and green might be produced; as green leaves with yellow and blue flowers.