[FE]A band and not a line, because R. is unstable to the impact of a considerable range of light-vibrations.
[FF]Mr. Chattock has kindly supplied me with the following note:—
"Readings at the violet end were taken at the extremity of the lavender rays, at the point where the faint band of lavender light seemed to end off about half-way across the field of view (the cross-wires being invisible).
"At the red end the cross-wires were always visible, and were in each case set to the point where the top horizontal edge of the spectrum lost its definition.
"Other things equal, the 'red' readings should be more reliable than the violet, therefore, from the greater definiteness of the point observed, and the means of observing it. But against this has to be set off the fact that the extreme violet rays were spread out by the prism used more than eight times as much as the red rays.
"In any case, the wide differences observed in the 'red' readings are much greater than could have been due to misunderstanding or careless observation—as shown by setting the instrument to maximum and minimum readings, and noting the very obvious difference between them apparent to a normal eye. The same conclusion is rather borne out by the closer (average) agreement between the two eyes of the same individual than between those of different persons.
"The source of light was the central portion of an ordinary Argand burner."
[FG]The variations above indicated throw light on a fact to which Lord Rayleigh has directed attention. The yellow of the spectrum may be matched by a blending of spectral red and spectral green; but the proportions in which these spectral colours must be mixed differ for different individuals. The complementary colours for different individuals are also not precisely the same.
[FH]"Colour-Vision and Colour-Blindness," R. Brudenell Carter (Nature, vol. xlii. p. 56).
[FI]Journal of Marine Biological Association, New Series, vol. i. Nos. 2 and 3. His experiments with regard to the colour-sense in fishes gave, for the most part, negative results.