Intelligence and Miscellaneous Articles. 307 



of the complementary colours : the intervals are very variable (from 

 1 to 19) ; and it seems impossible to deduce from them any im- 

 portant consequences. Helmholtz* also, who specially studied 

 their distribution in the spectrum, could not trace any law. This 

 irregularity will disappear when the results are discussed, if we 

 take as our guide the preceding definitions. The chromatic circle 

 upon which my experiments were made consists of three sections, 

 the limits of which are formed by the red, the yellow, and the blue. 

 These three sections are not necessarily equidistant from one an- 

 other ; but the twenty-four colours of which each is formed fulfil 

 that condition, as I have demonstrated in the first Note. 



The following discussion, based on the assumption of equidis- 

 tance, will not give rigorously exact results unless it be referred to a 

 single section only. "With this aim I select that which extends from 

 the yellow to the blue, passing through the green, because in this 

 portion of the chromatic circle the greatest number of complemen- 

 tary colours are assembled. The green, placed between the yellow 

 and the blue, which by their mixture produce only the sensation of 

 white, necessarily corresponds to a fundamental sensation. To fix 

 its exact position (for we have the choice among eighteen colours of 

 that name), I remark that with a second colour, at present un- 

 known, but certainly situated between the red and the orange- 

 yellow, and with a third colour, situated between the blue and the 

 violet-blue, it must give the whole of the colours of the chromatic 

 circle, at the same time producing the least possible sensation of 

 white. It will therefore be situated at an equal distance from the 

 complementaries of these two groups of colours. 



At the first glance it might be believed that the question, being 

 stated in such wide terms, would admit of several solutions. It is 

 nothing of the kind. The complements of the nineteen colours 

 extending from the red to the orange-yellow occupy in the circle 

 only four consecutive numbers, or from the fourth green to the first 

 blue-green. The complements of the blue to the violet-blue extend 

 from the yellow to the fourth yellow. The colour which occupies 

 the middle between the two systems is the third or fourth green- 

 yellow. I cannot here give the process employed to determine the 

 positions of the other two fundamental sensations ; I confine 

 myself to indicating the result. 



The three colours are (to within -^) the orange, the third green- 

 yellow, and the third blue : they are equidistant from one another 

 in the circle ; and so are their complementaries — the first yellow, 

 the blue-green, and the violet. The anomalies which I have pointed 



* HeLmholtz compared the wave-lengths of the complementary colours 

 (Joe. cit. p. 365), in order to discover the law of their distribution. He 

 did not, aud could not, arrive at any result. The quality possessed by two 

 colours of being complementary is of a purely physiological order ; while 

 the wave-lengths (that is to say, the refrangibilities) are physical proper- 

 ties, which depend on the intimate nature of light : they would not cease 

 to exist ; even if we were deprived of an organ for their perception. 



