COLOUR. 
790 
liave devoted a principal part of their fortunes and their 
lives to a practical inveftigation of this fubjebl, Among 
thefe are the refpedlei* njimes of Dufay, Hellot, Macquer, 
Fourcroy, and Barthollet, from whofe united labours our 
treatifes on this l'peculation are compiled, and for which 
feethe-articles Chemistry, Pigment, and Dying.—• 
For the theory of natural or primitive colours, fee the ar¬ 
ticle Chromatics, p.524 of this volume. 
Concerning the nji/ion of colours, or that the fame co¬ 
lours appear the fame to all men, we have lately had a 
very extraordinary proof to the contrary. This is mani- 
felted in a memoir, tranfmitted to the Manchefter Philo- 
fophical Society, in 1798, by Mr. John Dalton, who had 
long fufpedted fome peculiarity in his own vilion, with 
refpedl to colours. To him pink and blue appeared nearly 
allied, pink and red fcarcely at all. In the autumn of 
J792, however, he was convinced that la is vifion was not 
like that of the generality of men, by viewing the flower 
of the geranium zonale by candle-light. The flower is 
faid to be pink : but to the author it appeared by day Iky 
blue, and by candlelight a red; a colour which to him 
forms a ftrong contraft to blue. Remarking this phseno- 
menon to his friends, they likewife obferved the flower, 
but all agreed {excepting his brother) that its colour in 
candle-light did not differ materially from its colour in 
day-light. Having thus afcertained a peculiarity in his 
own vilion, lie did not enter into any inveftigation of the 
fubjedl until two years afterward ; and the following are 
the particulars and the refult of this invelfigation; 
“ It may be proper to obferve that I am fhort-fighted. 
Concave glafles about five inches focus fuit me bell. I 
can fee diltinftly at a proper diftance ; and am feldom 
hurt by too much or too little light; nor yet with long 
applications. My obl'ervations began with the folar fpec- 
trum, or coloured image of the fun, exhibited in a dark 
room by means of a glads prifm. I found that perfons in 
general diftinguifli fix kinds of colour in the folar image; 
namely, red, orange, yellow, ' green, blue, and purple. 
Newton, indeed, divides the purple into indigo and vio¬ 
let ; but the difference between him and others is merely 
nominal. To me it is quite otherwise : I fee only two, 
or at moll three diffinilions. Thefe I fhould call yellow 
and blue; oryellow, blue, and purple. My yellow com¬ 
prehends the red, orange, yellow, and green, of others, 
and mv blue and purple coincide with theirs. That 
part of the image which others call red, appears to me 
little more than a fhade, or defedt of light; after that the 
orange, yellow', and green, feem one colour, which de- 
fcends pretty uniformly from an intenfe to a rare yellow, 
making what f fhould call different fliades of yellow. The 
difference between the green part and the blue part is 
very ffriktng to my eye : they feem to be ftrongly con- 
tralted. That between the blue and purple is much iefs 
fo. The purple appears to'be blue, much darkened and 
condenfed. In viewing the flame of a candle by night 
through the prifm, the appearances are pretty much the 
fame, except that the red extremity of the image appears 
more vivid than that of the folar image. I now proceed 
to fcate the refults of my obl'ervations on the colours of 
bodies in general, whether natural or artificial, both by 
day-light and candle-light. I moftly ufed ribbons for the 
artificial colours. 
Red,. by day-light .—Under this head I includecrimfon, 
fcariet, red, and pink. All crimfons appear to me to con- 
fill chiefly of dark blue ; but many of them feem to have 
a ftrong tinge of dark brown. I have keen fpecimens of 
crimfon, claret, and mud, which were very nearly alike.' 
Crimfon has a grave appearance, being the reverse of every 
' fhewy and fplendid colour. Woollen yarn died crimfon 
or dark blue is the fame to me. Pink feems to be com- 
.pofed of nine parts of light blue, and one of red, or fome 
colour which has no other effedt than to make the light 
Ulue appear dull and faded a little. Pink and light blue, 
therefore, compared together, are to be diftinguifhed no 
©therwile than as a fplendid colour from one that has 
loft a little of its fplendour. Befides the pinks, roles, &c. 
of the gardens, the following Britifh flora appear to me 
blue ; namely, Statice Armeria, Trifolium pratenfe. 
Lychnis Flos-cuculi, Lychnis dioica, and many of the 
Gerania. The colour of a florid complexion appears to 
me that of a dull, opaque, blackifh blue, upon a white 
ground. A folution of lulphat of iron in the tindlure of 
galls (that is, dilute black ink) upon white paper, gives 
a colour much refembling that of a florid complexion. 
It has no refemblance of the colour of blood. Red and 
fcariet form a genus with me totally different from pink. 
My idea of red I obtain from vermilion, minium, feal- 
ing-wax, wafers, a foldier’s uniform, &c. Thefe feem 
to have no blue whatever in them. Scarlet has a more 
fplendid appearances than red. Blood appears to me red ; 
but it differs much from the articles mentioned above. 
It is much more dull, and to me is not unlike that co¬ 
lour called bottle-green. Stockings fpotted with blood 
or with dirt would icarely be diftiriguilhable. By candle¬ 
light, red and fcariet appear much more vivid than by 
day. Crimfon lofes its blue and becomes yellowifh red. 
Pink is by far the mod changed ; indeed it forms an ex¬ 
cellent contraft to what it is by day. No blue now ap¬ 
pears ; yellow has taken its place. Pink, by candle¬ 
light, feems to be three parts yellow and one red, or a 
reddifh yellow. The blue, however, is iefs mixed by 
day than the yellow by night. Red, and particularly 
fcariet, is a fuperb colour by candle-light: but by day 
fome reds are the leaft fhewy imaginable : I fhould call 
them dark drabs. 
Orange and Yellow, by day-light and candle-light .— 
I do not find that I differ materially from other perfons 
in regard to thele colours. I have fometimes leen per- 
lons hefitate whether a'thing was white or yellow by cau¬ 
dle-light, when to me there was no doubt at all. 
Green, by day-light .—I take my ffandard idea from 
grafs. This appears to me very little different from red. 
The face of a laurel-leaf (Pi unus Laurocerafus) is a good 
match to a flick of red iealing-wax ; and the back of the 
leaf anfwers to the lighter red of wafers. Hence it will 
be immediately concluded, that I fee either red or green, 
or both, different from other people. The fail is, I be¬ 
lieve that they both appear different tome from what they 
do to others. Green and orange have much affinity alio. 
Apple-green is the molt plealmg kind tome; and any 
other that has a tinge of yellow appears to advantage. I 
can diftinguifli the different vegetable greens one from 
another as well as moll people; and thole which are 
nearly alike, or very uniike, to others, are fo to me. A 
decodlion of boliea tea, a folution of liver of fulphur, 
ale, &c. which others call brown, appear to me green. 
Green woollen cloth, l’uch as is ufed to cover tables, ap¬ 
pears to me a dull, dark, brownifh red, colour. A mixture 
of two parts mud and one red would come near it. It 
refembles a red foil juft turned up by the plough. When 
this kind of cloth lofes its colour, as other people lav, 
and turns yellow, then it appears to me a pleafant green. 
Very light green paper, filk, &c. is white to me. By candle¬ 
light, 1 agree with others, that it is difticuit to diltin- 
guiih greens from blues; but, with me, the greens only 
are altered and made to approach the blues. It is the 
rtal greens only that are altered in my eye ; and not fuch 
as I confound with them by day-light, as the brown li¬ 
quids above-mentioned, which are not at.all tinged with 
blue by candle light, but are the fame as by day, except 
that they are paler. 
Blue, by day-light and candle-light .— I apprehend this 
colour appears very nearly the lame to me ‘as to other 
people. 
Purple, by day-light and candle-light .—This feems to 
me a flight modification of blue. I Jeldom fail to diitin- 
guilh purple from blue ; but fhould hardly fufpedt pur¬ 
ple to lie a compound of blue and red. The difference 
between day-light and candie-light is not material.” Mr. 
Dalton then proceeds to make iome mifcellaneous obfer- 
vations. 
