A
697,994
BENSON
SCIENCE OF COLOUR

►
1. THE
VERSITY
IVERSITY
ARTES
↑ MICHIGAN ·
OF
OF
-
GAN
MICHIC
SE
SCIERTA
1817
LIBRARIES

!

THE COLOUR-CUBE
P
S
R
B
G
MANUAL
OF THE
زه
SCIENCE OF COLOUR
ON THE

TRUE THEORY OF THE COLOUR-SENSATIONS
AND THE
NATURAL SYSTEM.
冷
​(SITY OF MICRUIS
WITH A COLOURED FRONTISPIECE AND OTHER ILLUSTRATIONS.
T.
BY WILLIAM BENSON, ARCH™.
AUTHOR OF "PRINCIPLES OF THE SCIENCE OF COLOUR."
LONDON:
CHAPMAN & HALL, 193, PICCADILLY, W.
1871.
QC
495
B46
LONDON:
VINTON AND SON, PRINTERS, HAMPSTEAD ROAD, N.W.
Lately published, by the same Author,
QUARTO, PRICE 15s.
THE PRINCIPLES OF THE SCIENCE OF COLOUR,
CONCISELY STATED, TO AID AND PROMOTE THEIR USEFUL
APPLICATION IN THE DECORATIVE ARTS.
Chapman & Hall, 198, Piccadilly.
Besides other diagrams and coloured illustrations, this treatise contains
fifty-nine natural harmonies of from three to nine colours each, hand-
coloured in best water-colours, and laid down by a facile contrivance
which has allowed of a reduction of one half the original price. These
groups of colours are arranged in five plates to admit of being studied
singly or in conjunction with other parallel or cognate groups; to do
which to the best advantage, all other groups occurring on the same plate
should be concealed from view. They afford beautiful illustrations of the
principal gradations and contrasts of colour, and of many of the rules of
harmony mentioned in the Manual: and exemplify the endless elements
of colour-designs derivable from the Natural system of colour.
From the Athenæum, 8 August, 1868.
"We have received from Messrs. Chapman & Hall a capital treatise on "The Prin-
ciples of the Science of Colour" by William Benson, Architect. Mr. Benson recom-
mends the use of the prism as indispensable to those who wish to judge of colour
correctly, and refers to the method long ago devised by Lambert for finding intermediate
colours by the aid of a slip of polished glass-an easy way to discover the true mean.
Also among the branches of his subject to which Mr. Benson refers with unusual
perspicacity, is a definition of the natures and qualities of the secondary colours.
Besides these matters the student will find a new and very satisfactory system of colour
proposed which complies with the theory of the three elementary colour sensations.
The serviceableness of this plan, which follows a suggestion of Sir John Herschel in
his Treatise on Light,' will commend itself to artists and decorators who may desire
to bring the resources of science to the aid of their studies. To these we commend
the volume before us as at once scientific and artistic in its treatment and bearings.
It is the most valuable contribution to our knowledge of the subject which has ap-
peared for a very long time, and will receive the applause it well deserves.
The author announces for early publication a work on 'Colour Lore,' to recount
the principal theories, observations, and discoveries which have been made about
colours from the earliest ages to the present time, and to discuss some points which
have been imperfectly mastered. If Mr. Benson fulfils in the new work, the promise
of that which is now before us, he will be a very fortunate man, to whom the world of
students will owe a great obligation. Even general readers may find in such a book
as that which is announced an enormous mass of curious knowledge."
}
*
CONTENTS.
INTRODUCTION.
What the Science of Colour includes; its uses and advantages. Com-
mon misconceptions, their causes and consequences. Proofs and author-
ities. The nature of light, and the action of the eye.
CHAPTER I.
THE PRINCIPAL COLOURS, AND HOW TO SEE THEM.
1. Colour-sensations, simple and compound. 2. Hues, tints, and shades
of colour. 3. Excitement of colour-sensations. 4. Cause of variety of
hue. 5. Separation of mixed lights. 6. What light gives a complete
spectrum. 7. How to obtain the spectrum (figs. 1 and 2). 8. The pris-
matic colours. 9. What are the primary colours, and how to see them.
10. What are the secondary colours, and how to see them. 11. How to
see the primaries and secondaries in contrast (fig. 3). 12. How to see
some of the best combinations of prismatic colours in contrast (fig. 4).
13. How to see at once all the best combinations of prismatic colours
(fig. 5). 14. Value of such observations. 15. The best standard colours.
CHAPTER II.
THE COLOURS OF OBJECTS SELF-LUMINOUS OR ILLUMINATED.
16. Colours of incandescent solids. 17. Colours of incandescent vapours.
18. Effects of different kinds of illumination. 19. Causes of colours of
non-luminous bodies. 20. Colours from transmission of light. 21. Effect
of increasing the thickness of a translucent body. 22. Effect of super-
posing translucent bodies. 23. Colours from internal reflection. 24. Effect
of mixing pigments. 25. Effect of thickening a pigment. 26. Inferiority
of colours from internal reflection. 27. Effect of form and texture of
surface. 28. Colours from superficial reflection. 29. Colours from metal-
lic superficial reflection. 30. Effect of repeating reflections. 31. Colours
from reflection from small particles. 32. Colours from interference.
33. Colours from fluorescence. 34. How to test the colours of objects.
CHAPTER III.
INTERMEDIATE COLOURS, AND MEANS BETWEEN TWO OR MORE COLOURS.
35. How to find intermediate colours. 36. How to find the mean of
two colours (fig. 6). 37. How to see whether colours are complementary.
38. How to find the mean of several colours. 39. How to compare means
of different sets of colours. 40. Truth of results. 41. Means between
principal colours. 42. Shades of tints. 43. How to ascertain the real
hues of obscure colours.
CHAPTER IV.
THE NATURAL SYSTEM OF COLOURS AND ITS USE.
44. Construction of the colour-cube (fig. 7 and frontispiece). 45. Rec-
tilineal gradations. 46. Plane gradations. 47. Complementary plane
gradations. 48. Cognate plane gradations. 49. Complementary and
cognate rectilineal gradations. 50. Limits of colours in depth and clear-
ness. 51. Utility of the natural system.
iv
CHAPTER V.
OCULAR MODIFICATIONS OF COLOURS; THE SENSIBILITY OF THE EYE.
52. Variable sensibility of the eye 53. Law of variation. 54. Simul-
taneous contrast. 55. Effects of simultaneous contrast. 56. Accommo-
dation of the eye to light. 57. Sensibility of the eye under a given
illumination. 58. Number of distinguishable colours. 59. Effect of very
feeble light. 60. Effect of dazzling light.
CHAPTER VI.
THE HARMONY OF COLOUR.
61. Harmony the source of beauty. 62. Principles of harmony. 63.
Rule I. Each colour-sensation should be excited. 64. Rule II. Each
colour-sensation should be relieved. 65. Rule III. Kindred colours
associate well. 66. Rule IV. Light colours with dark associate well.
67. Rule V. Gradations should be true and regular. 68. Rule VI. Ends
of gradations should be equally striking. 69. Rule VII. Contrasting
colours should be equally striking. 70. Rule VIII. Gradations contrasts
and single colours should correspond. 71. Rule IX. Colours and groups
of colours should occupy appropriate breadths. 72. Rule X. Variety of
colour and treatment may increase beauty. 73. Conditions restrictive of
rules. 74. Necessary compromise of principles.
CHAPTER VII.
MENTAL EFFECTS OF COLOURS AND COLOUR-COMPOSITIONS,
75. Impressiveness of colour. 76. Liveliness of colour. 77. Warmth
of colour. 78. Effects of colours in colour-compositions. 79. Effects of
modes of composition. 80. Imaginary mental effects.
CHAPTER VIII.
PECULIAR AND DEFECTIVE COLOUR-VISION.
81. Dichromism. 82. Defective sensibility to red light. 83. Dichromism
with defective sensibility. 84. Effect of colour-designs in such cases.
ILLUSTRATIONS.
Frontispiece.
p. 5.
Two coloured diagrams of the colour-cube, in explanation of
the natural system of colours
Two diagrams, showing how to obtain the prismatic spectrum
Three diagrams, for exhibiting the prismatic colours and
their regular compounds when viewed through a prism, facing p. 10.
A diagram, showing how to find intermediate colours by a
slip of glass
A diagram of the cube of colours
p. 22.
p. 28.
INTRODUCTION.
·
THE Science of Colour teaches the nature and causes of
colours; their distinctions; their relations to each other;
their classification; the mental effects that attend them, and
the causes and laws of their harmony. It includes the modi-
fications of colour arising from the varying sensibility of the
eye, and also the peculiarities of colour-vision which are found
to exist in different persons. It enables the eye to distinguish
and recognize the different hues in their various tints and
shades, and to acquire a correct judgment about the real
constituents of colours, upon which their effect in any com-
position depends; it improves and directs the natural taste,
and aids the student in devising as well as appreciating good
compositions of colour.
The whole subject is full of interest and beauty, and an
acquaintance with it adds extremely to the pleasure to be
derived from contemplating natural objects and scenery, as
well as pictures and other works of art; and it has one ad-
vantage, perhaps, above any other branch of natural philo-
sophy; it teaches us in a very remarkable way to distinguish
between sensations and their causes, and not to judge hastily
according to the appearance of things. And as the chief
difficulty of the science seems to arise from this, it is proper
in the Introduction to call the student's attention especially to
it, and to guard against those misconceptions which might
otherwise hinder his progress.
First, then, the science teaches us that colours are purely
and simply sensations, excited by the action of light on the
nervous tissue of the retina which forms a screen at the back
of the eye; and a little observation and reflection is enough to
convince us that they must be so, since they are only seen by
the aid of light, and vary with different kinds of light. But
when we look at an object, we see so clearly the colour on its
surface, at a distance from us, and that colour under the
•
F
vi
..
ordinary light of day is so constantly the same, that we can
hardly imagine it anything but some inherent quality of the
surface, belonging to the object, just as much as the form which
we see. This is therefore the notion which unreflecting persons
usually have about colours, insomuch that it is not always
easy to make them understand the difference between colours
(properly so called) and the pigments or colouring matters
which go by the same names.
But suppose one has advanced beyond this primitive stage,
and knows that colours are caused by the light that enters the
eye; and suppose one has learnt also that there are innumer-
able different kinds of light, each distinguished by some
difference of colour, and that a difference of colour in two
objects certainly indicates some difference in the kinds of light
which they send to the eye; what then is more natural than
to infer that colour is an indication of some distinctive quality
residing in the light? Even eminent philosophers have hastily
assumed it to be so; and some have supposed that there
must be as many different species of light as there are distinct
elementary sensations of colour, which is inconsistent with what
is known of the nature of light. And it is still more common
for those who have not studied the science of colour, to think
that the colours of simple or homogeneous light (light that is
all of one kind) are themselves simple or homogeneous. Yet
our science tells us that all the colours of homogeneous lights
may be more or less compound sensations, and most of them
certainly are so; and that mixtures of different kinds of light
may have the same colours. For the colours of even certain
kinds of homogeneous light may be imitated by mixtures of
other kinds; and again the very same kinds of light may have
different colours, not only in different eyes, but also under
different circumstances in the same eye.
Hence we must adhere, contrary to all natural prejudices,
to the doctrine that colours are sensations, simple or compound;
and it must be clearly understood that when their names are
used as adjectives to describe any light, as red light, yellow
light, white light, they mean only that the light, whatever it
is, has the property of exciting in our eyes the sensation of
the colour red, yellow, white, or whatever it may be. In like
manner, when the same terms are applied to pigments or other
material objects, as red paint, yellow solution, white paper, they
must be understood to mean that the lights which the objects
!
vii
send to the eye, under the ordinary illumination of daylight,
produce those same sensations.
Again, our science teaches that some colours, such as whit、
and yellow, are compound sensations. A perfect white, for
example, it tells us, is a combination or mixture of three dis-
tinct sensations, namely, red, green, and blue, excited at
once in equal force on the same part of the retina. A perfect
yellow, it tells us again, is a combination or mixture of two of
these, red and green, in equal force. Yet, when we con-
template a white object, we perceive nothing whatever to
call up the idea of red, or of green, or of blue, or of any
other colour. It seems perfectly homogeneous, simple and
pure. And when we look at a yellow surface, it is by no
means easy (though perhaps not quite impossible under
certain circumstances) to imagine we see a combination or
mixture of red and green. Nevertheless, a careful investigation.
of the matter shows that these really are compound sensations
as asserted. For we can compound white light by throwing
together or mixing red, green, and blue lights, and yellow
light by throwing together or mixing red and green lights.*
We can also show that the light given out by the best white
and yellow objects contains a mixture of the same components†.
And we can produce the sensations of white and yellow by
mixing the sensations of red, green, and blue together, and
those of red and green together, without mixing the lights.‡
Hence, strange as it seems, it must be admitted that sensa-
tions apparently simple may really be compound; and their
components such as cannot be recognized in them.
To mention another instance, it is well known that if we mix
together two coloured solutions, or two transparent pigments,
or superpose two coloured glasses, against a white surface, we
produce a third colour. What more natural than to suppose
that this new colour is a combination or mixture of the colours
of the two solutions, pigments, or glasses separately? Yet,
our science says this is only approximately true in certain
• By mixing opaque coloured powders; by reflection from glass (as in Sect. 36); and
by actually throwing together beams of light that have passed through coloured
glasses.
+ By analysis with the prism (Sect. 34); by analysis with coloured glasses (Sect. 22).
This is evidently effected when colours are blended by the method of rotation
(Sect. 38); or by blending two separate sensations excited in the two eyes, as when we
see double images of two different coloured spots, one of each colour overlapping,
viii
cases; in most cases the third colour must differ widely from
the colour that would result from the mixture of the two
colours. Thus, green has been commonly supposed to be a
mixture of yellow and blue, because it is easily produced by a
mixture of yellow and blue pigments; yet it is easy to show that
the true mixture of those two colours is not green but gray;
and it is easy also to explain why green is produced by the
mixture of such pigments.
It is equally natural to suppose that by making a coloured
solution stronger or thicker, or by laying on a thicker wash of
a pigment, we increase the strength or intensity of the colour,
without other alteration. It has even been proposed to use a
combination of hollow glass wedges, filled with coloured
solutions, as a means of measuring the intensity of a colour
according to the thicknesses of the solutions which give a
colour to match it; and experiments made in this way are the
basis of what is commonly called Field's doctrine of chromatic
equivalents. Yet the truth is easily shown that to thicken a
solution or transparent pigment materially alters the hue as
well as darkens its colour; and that it is absurd in any case
to make the thickness of a solution the measure of the strength
of its colour.*
Once more, as we notice the brilliant whiteness of the sun,
and of other intensely luminous bodies, and also the whiteness
of many illuminated objects, as clouds, snow, linen, paper,
chalk, far brighter than other colours, and the white superficial
reflection that takes place from even dark-coloured objects,
quite independently of their proper colours; and as we find,
too, that a white illumination is necessary to bring out all the
various hues of objects, we easily connect the ideas of white-
ness and of light, and infer that other colours are something
essentially different from white, and perhaps think, as some
have done, that they rather partake of the nature of darkness.
Other phenomena added to these, such as the blueness of
the light reflected from very small reflecting particles in a
transparent medium (which is only noticed when viewed
against a dark background), and the prevailing redness of the
* In truth colouring matters act by destroying some kinds of light faster than
others; so that the lights that escape from thick washes of a transparent pigment
cannot be the same in quantity or quality; and the light that escapes from two pig-
ments is no more a mixture of the lights that escape from both of them separately,
than the grain which falls through both of two different sieves is a mixture of that
which falls through each one of them alone (Sections 20-25),
ix
light which is transmitted through most bodies imperfectly
transparent, have led, not only ancient philosophers, but also
ingenious moderns, like Goethe, to suppose that bluish colours
are constituted by darkness seen through a white or light
medium, and reddish colours by white or light seen through
a dark medium.
But our science utterly contradicts all such notions, con-
straining us to believe that white only differs from other
colours in containing all the simple colour-sensations in equal
strength. It makes no difficulty in explaining the remarkable
phenomena alluded to above; showing us on the one hand
that bodies just heated enough to be luminous give out red
light, and when more heated give out other kinds also, and all
more and more, so that their light becomes not only brighter,
but also whiter; and on the other hand, that bodies which
reflect all kinds of light equally, often reflect all very power-
fully, so that white objects are commonly bright; whilst the
blueness of the sky, and the redness of the sun through a
fog, and thousands of other effects, are necessary consequences
of the various ways in which matter acts upon light.
Whenever, therefore, the term "light" is used (as it often is)
to express a sensation, it must be understood as a general term,
including each and all of the colour-sensations, and not as
denoting any other sensation excited by the action of light,
for there is none beside red, green, blue, and their compounds.
The importance of understanding these points is great; since
whole theories and systems of colour have been founded upon
erroneous conclusions hastily drawn from natural phenomena,
correctly observed, but not understood; and it is not easy to
deduce reliable and useful rules in colour-designing from such
defective principles. Besides this, a most beautiful branch of
natural philosophy has been neglected and discredited in
consequence of them. Through the rashness of several
esteemed writers in advancing crude speculations about colour,
and through ignorance of what has been really discovered,
many are disinclined to entertain the subject at all. They
either think that nothing certain is known about it, or that all
theories will prove equally useless in art; or else are wedded
to preconceived opinions which they are unwilling to disturb.
Hence a manual which aims at freeing the subject from
previous errors and imperfections, should call special attention
to the proofs of its doctrines (so far as they are not universally
X
accepted), and the authorities that support them, as well as
the advantages which will attend their adoption.
Now it may be safely asserted that by a little intelligent
study of the facts revealed in those simple and beautiful
experiments which are described in this treatise,* any person
who has the use of his eyes, and believes the grand proposi-
tions about light which Newton established when he first
unfolded to the world the meaning of the prismatic spectrum,
may very quickly satisfy himself of the truth of the doctrine.
And it may be safely asserted, too, that when the matter is
once understood, no one who takes interest in the harmony of
colour will feel any wish to revert from light to darkness, or
regret the few hours of interesting study which enabled him
to master a charming science, hitherto generally neglected
by those who should profit most by it.
But accurate researches for determining the colour-sensations
excited by the prismatic rays are not wanting; and they fully
establish the doctrine. Those recently made in this country
by Mr. James Clerk Maxwell have just now been confirmed
by others made on another plan by Herr J. J. Müller in
Germany.† Experiments by Sir John F. W. Herschel, and
by Professor Helmholtz, might also be cited in confirmation.
For these investigations have distinctly proved that both white
and all possible colours can be produced by mixtures of the
best prismatic red, green, and blue, and by mixtures of no
other three colours but these; quite contrary to the common
red-yellow-blue theory. But that this common theory is utterly
wrong is obvious from the simplest experiments, which prove
that yellow consists, as before mentioned, of the sensations of
red and green excited together, and is exactly complementary
to blue. So that if red and blue are simple colours, yellow
must be compound.‡
• See especially Sections 9-13, 34, 35, 36, 38, 39.
+ Transactions of Royal Society, 1860, and Poggendorff's Annalen, March and
April 1870.
The true doctrine of the three colour-sensations was clearly stated at the begin-
ning of this century by that great philosopher Thomas Young (Natural Philosophy,
Lecture xxxvii.), and has lately been explained in a popular manner by Sir John
F. W. Herschel (Essay on Light in "Good Words" for 1865, which has also been
reprinted in his Familiar Lectures), as well as by Mr. Maxwell (Lecture at the Royal
Institution, 17 May, 1861). Notwithstanding the seeming difficulty of reconciling it
with first appearances, the neglect it has hitherto met from artists is strange, so
great and so evident is its superiority, for all the purposes of art, over the conven-
tional system that has been substituted for it.
In several letters to the "Builder" (in the latter part of 1868), and in a paper read
at the Institute of Architects (Sess. Paper No. 7, 1869), the author himself has
endeavoured to expound the proofs and uses of the science of colour. Since then, he
xi
Another advantage of learning the true nature and relations
of colours will be found in the improvement to be derived from
the intelligent study of the rich compositions of colour pre-
sented in natural scenery and natural objects, and also in
works of art. This will greatly increase the value of those
illustrated works on colour-decoration and ornament which
have been so abundantly produced of late years. For those
who would study these to profit, must not only feel that some
combinations of colour are preferable to others; they must
also know the reasons why; otherwise they can but make
empirical attempts to improve. Now a false theory mystifies
the matter, and even if sound rules are hit upon, often hinders
the right application of them; but true knowledge will ever
improve the judgment, and direct the hand; and if "a thing
of beauty is a joy for ever," it is doubly so when the under-
standing co-operates with the feeling.
The nature of light itself is perhaps not strictly a part of
the science of colour; but it adds much to the interest of the
subject to know that the wonderful medium through which
we have communication with the utmost parts of the universe,
the all-pervading and perfectly elastic ether, receives and
transmits in waves of vibration, with inconceivable velocity
in every direction, all sorts of shocks, however minute, arising
amongst atoms of matter, just as air does with those arising
from the tap of a hammer or the ring of a bell; and that
these etherial waves, when they are of the kind called waves
of transverse vibration, and their time or period falls within
certain limits, constitute light;—the different kinds of light
differing essentially in their wave-times alone.
has met with a treatise on colour, prepared at the suggestion of the Directors of the
Imperial Austrian Museum of Art and Manufacture, by Ernest Brücke, Professor of
Physiology in the University of Vienna, and which has been translated into French
(Des couleurs au point de vue physique, physiologique, artistique et industriel, par le
Dr. Ernest Brücke. Traduit d'allemand par J. Schutzenberger, Paris, 1866). The
progress of the science of colour on the continent appears from the fact that in this
work the whole subject is treated in a scientific manner, and what is wrong in the
common theories is rejected, though no complete system of colour is proposed. Field's
doctrine of chromatic equivalents, which has long held too much authority in England,
Professor Brücke declares "is false from beginning to end, and owes its origin to an
incorrect interpretation of natural phenomena.' Sir J. F. W. Herschel, after his
life-long labours for science, may be allowed to speak with authority here. In a letter to
the author, he tersely describes the prevailing opinions about light and colours
as "idola, which, from a jargon, have fixed themselves into a doctrine." It may be
long before that doctrine loses its hold on the public; in lectures, and even in new
treatises on Natural Philosophy, its truth is still sometimes assumed by persons who
ought to know better; but at least those who direct the education of students in art
should take care not to fall behind the age, or allow errors repudiated in Germany
and France to remain unchallenged in England.
xii
Reflection and refraction, with all their complicated and
beautiful phenomena, are but natural consequences of such
waves meeting with different media-portions of space in
which the etherial density and elasticity are modified, so that
the waves traverse them with altered velocity. The equally
marvellous phenomena of interference, which led to the dis-
covery of the nature of light, and the actual measurement of
the lengths of the luminous waves, arise from the action of
one wave upon another, when two either concur to increase
the disturbance, or oppose each other to quiet it.
How the minute wave-lengths, and still more inconceivably
short wave-times, of the different kinds of light have been
measured, and what the results of the measurements are,
must be left to works on optics to recount. It is enough to
note here that their periods are all so short that there is time,
even in the fraction of a second during which the sensation
excited by a luminous wave continues, for a succession of
innumerable separate waves of any kinds of light to follow
each other along the same line or ray; so that we may have
lights of all kinds mingled together in any proportions, and
making any conceivable mixture of their separate colours at
the same moment on the same part of the retina.
The student should of course be so far acquainted with the
action of the eye, as to know that the light which proceeds.
from the various objects in the field of view is caused by it to
fall regularly upon corresponding parts of the retina, so as to
form upon it a picture of the whole scene, like that produced
in a camera obscura; and that when the direction of the eye'
is changed, this picture shifts its position, so that the part of
the sensitive screen which was before affected by the light
from one object, is now affected by the light from another, or
perhaps by none at all. It should be noticed, too, that the
sensations excited on corresponding parts of the retinas of the
two eyes become blended together by the union of the two
branches of the optic nerve.
J
MANUAL
OF
THE SCIENCE OF COLOUR.
CHAPTER I.
THE PRINCIPAL COLOURS, AND HOW TO SEE THEM.
1. Colour-sensations, simple and compound. The eye
is so constituted in the great majority of persons as to
be capable of three simple or elementary sensations
of colour, which are best described by the terms red,
green, and blue.*
The simple sensations are never excited separately,
with perfect purity, on any part of the retina, but
always accompany each other in greater or less degree,
causing the endless variety of colour which we observe.
When all are excited at once with equal intensity on
the same part of the retina, the result is the sensation
of white.
When one of them is excited in less degree than the
others, the two predominant sensations form a compound
or mixed sensation, accompanied with a mixture of white
proportionate to the degree in which the third is excited.
These binary compounds are―
(1) Seagreen, which consists of green and blue in equal
strength, and passes through seagreen-green into green,
when the blue is weakened, and through seagreen-blue
or azure into blue, when the red-is weakened.
*As these and all other names of colours are commonly applied to
many different hues, it should be observed that the particular hues here
intended are those of the deepest prismatic red, green, and blue, as will
be explained further on. Cases of peculiar and of defective colour-vision
are noticed at the close of this treatise.

B
et
2
(2) Pink, which consists of blue and red in equal
strength, and passes through pink-blue or puce into
blue, when the red is weakened, and through pink-red
or cerise into red, when the blue is weakened.
(3) Yellow, which consists of red and green in equal
strength, and passes through yellow-red or orange into
red, when the green is weakened, and through yellow-
green into green, when the red is weakened.
When two of the simple sensations are equally excited
in less degree than the third, we have of course the pre-
dominant sensation, together with a mixture of white
proportionate to the degree in which the other two are
excited.
The simple sensations are called primary colours, or
primaries; and their binary compounds, secondary
colours, or secondaries.
White, the peerless ternary sensation, though in com-
mon language sometimes distinguished from colours,
is in truth the chief of all. Black, on the other hand,
though no positive colour at all, must yet for convenience
sake be included amongst them, as the lowest limit of
all colours.
2. Hues, tints, and shades of colour. The absolute
excess or predominance of some one or two of the simple
sensations over the rest, as above described, is called
the hue of a colour; and the greater such excess, the
stronger the hue.
The whole series of hues is best remembered by con-
ceiving the six gradations from red to yellow, from
yellow to green, from green to seagreen, from seagreen
to blue, from blue to pink, and from pink back again
to red, arranged in order and in equal strength on the
sides of a hexagon, as indicated in the outer row of
coloured spots in each of the diagrams on the frontis-
piece. A similar arrangement is frequently described
under the name of the chromatic circle; but the hexagou
is the preferable form, because it distinguishes the six
points from which new gradations start. White and
black, being perfectly neutral, or without any hue, form
no part of any chromatic circle.

3
If the deficiency of each sensation in a colour is
diminished in proportion, until all the three are brought
to the full and equal intensity which they possess in
white, the colour changes by a regular gradation into
white, without any variation of hue. The lighter
colours thus produced are often called tints of their
respective hues.
The brightness of a colour is the sum of the intensities
of all the simple sensations of which it consists. The
proportionate diminution of these from the most power-
ful that the eye is capable of down to the absence of all
sensation, produce all the shades of colours, from the
brightest down to black. 2:
3. Excitement of colour-sensations. Colour-sensations
may be slightly and promiscuously excited by various ·
means, as by pressure or an electric shock, and even
without any external application, as seen in the unde-
fined images that seem to float before the eye in the
dark. But the action of light from without is incom-
parably the most powerful and effective means of
producing them; and it is for the admission of light
in proper quantities, and its regular distribution on the
surface of the retina, according to the direction from
which it comes, that the transparent parts of the eye
are designed.
4. Cause of variety of hue. Each of the innumera-
ble different kinds of light, in each degree of intensity,
produces a peculiar mixture of the three simple sensa-
tions of colour, one or two always predominating more
or less over the other.
When two or more of the different kinds of light
enter the eye together, or in very rapid succession,
they produce an effect compounded of the effects which
the separate lights would produce; in other words they
make us see a colour compounded of the separate colours
which we should have seen had the same lights come
separately.
5. Separation of mixed homogeneous lights. In con-
sequence of the rays of each kind of light being
differently bent or refracted in passing from one trans-
B 2
4
parent substance into another, a ray of mixed light going
through the angle of a prism is always divided into
the several kinds of light of which it consists; and
these, being all bent at different angles from their
original direction, when they enter the eye fall upon
different parts of the retina, and excite each its proper
sensation of colour apart from the rest. Thus is pro-
duced the splendid series of colours known as the
prismatic colours, and seen in the solar spectrum, or,
with less purity, in the rainbow; and these are the
colours of the several different kinds of light, arranged
in order of the refrangibility of the lights that produce
them. A knowledge of them is of great importance,
and can only be obtained by actual observation. For
they far surpass the best representations that can be
made by pigments or dyes.
6. What light gives a complete spectrum. An uninter-
rupted spectrum, exhibiting a complete series of the
prismatic colours in sufficient intensity to make them all
distinct, can only be obtained from the brilliant white
light given out by any intensely heated solid, such as
the lime in the oxy-hydrogen light, or the carbon in
the electric light.
Such, for all purposes of the study of colour, is the
light of the sun, either direct, or reflected from a bright
white surface; though when accurately examined its
spectrum is found to be interrupted by innumerable
narrow dark lines, indicating rays which are wanting,
having doubtless been absorbed or lost in traversing
the solar atmosphere or our own.
7. Hon to obtain the spectrum. To obtain a spectrum
sufficiently brilliant and pure for studying the colours, the
readiest way is to view a bright and narrow white stripe
through a good prism. The stripe may be a portion of
a bright white cloud seen through a slit in a dark
screen, or it may be a piece of silver wire, or a slip of
white paper, laid across a dark cavity, and brightly
illuminated. The more brilliant it is, the narrower
it may be, or the greater the distance at which it
may be viewed, without too much diminishing the
5
light; and therefore the more nearly pure the spec-
trum will be.
Fig. 1 shows how the
prism may best be placed
to view the spectrum of
the light of a white sky
through a horizontal slit
in a shutter.
Fig. 2 shows how it
may be placed to view a
white stripe on a dark
ground.
The dotted lines con-
verging to the eye in the
direction of the rays
emerging from the prism
are intended to indicate
more distinctly the direc-
tion in which the spec-
trum will appear.
The refracting angle
of the prism should
always be parallel to
the bright line, and
should be so directed
that the light may enter
and emerge at nearly
equal angles with the
sides of the prism, as
indicated in the figures.
This position deflects
the light as little as
possible from its ori-
ginal direction. It is
easily found by slowly
turning the prism one
way or the other on an
axis parallel to its edge,
till the spectrum becomes
stationary.

BG
B
Fig. 2
Fig. 1.

ཡ་
G
R
B
6
When a sunbeam enters a dark room through a slit
in a shutter, and passes through a prism held parallel
to the slit, the brilliancy of the spectrum (too great
for direct vision) allows it to be received on a white
screen. In this way its splendid colours may be seen
by many spectators at once, the white surface reflecting
it in every direction.
8. The prismatic colours. On the first view, a good
spectrum appears to consist of three bands of colour,
red, green, and blue: the first including all the less
refrangible rays, the second all those of mean refrangi-
bility, and the third all the more refrangible rays.
But on a closer inspection it appears that the colours
change gradually into each other, and the principal
varieties are stated in the following list :-
Red band
Dark Red.
Bright Red or Scarlet.
Yellow-red or Orange.
Yellow.
Yellow-green.
Green band
Green.
Seagreen-green.
Seagreen.
Seagreen-blue or Azure.
Blue band
Blue.
Dark Blue.
Very dark Violet.
1
The last of these is scarcely visible unless the spectrum
is produced from very strong light. It should be
noticed that the rays of the red band are most con-
densed, and those of the blue band least condensed,
because the glass disperses or spreads out the former
much less than the latter. This makes the red band
narrower and brighter, and the blue band wider and
darker, than they otherwise would be. The brightest
part of the spectrum is between the yellow and green;
but if all the rays were equally dispersed it would be
about the middle of the green band.
my
C
9. What are the primary colours, and how to see them.
The best prismatic red, green, and blue, near the begin-
ning, middle, and end of the spectrum, are so much
deeper (or stronger in hue in proportion to their bright-
ness) than the colours which lie between them in the
spectrum, that by their mixture all the latter may be
produced in the full strength in which they are found
in the spectrum, whilst they themselves cannot be
produced by the mixture of any other rays. Now the
colours of all objects must be mixtures of prismatic
colours; therefore, since these are compounded of the
prismatic red, green, and blue, it follows that all possible
colours are so compounded, and that the prismatic red,
green, and blue give the nearest possible approach to the
three simple sensations of colour. For the sake of
shortness therefore these may be called the primary
colours.
To see them in their greatest strength, we should look
through the prism at a white band of a moderate
breadth upon a black ground; the colours so produced
are much stronger than those of the pure prismatic rays,
produced by looking at a very narrow line of the same
white, because numbers of those rays very nearly
resembling each other in colour are thereby thrown.
together. It is easy to adjust the distance of the
prism so as to produce the strongest green which is
possible out of the light that comes from the white
band, and the same position of the prism will give
also the strongest red, and very nearly the strongest
blue.*
10. What are the secondary colours, and how to see
them. When we view in like manner through the
prism a black band upon a white ground, we obtain
a spectrum that is exactly complementary to the spec-
trum of a white band of the same width on a black
ground. In other words, the colour of any part of the
one spectrum contains exactly what the colour of the
*Owing to the blue rays being more dispersed than the green, the
distance of the prism from the object should be rather less to throw
together all the blue rays.
8
corresponding part of the other spectrum wants, to
make up the full white. For as the white band would
exactly cover the black band, and reduce the whole to
a uniform white, so would the spectrum of the white
band cover that of the black, and every colour in it
would reduce the corresponding colour in the other to
white. If therefore the one exhibits the three primary
colours in greatest strength, the other must give, op-
posite to each, a combination of the other two primaries
in greatest strength. These may therefore be called
the secondary colours.
The seagreen which we thus obtain is produced by
throwing together all the prismatic rays of the blue
and green band without those of the red band; the pink,
by throwing together all the red and blue, without the
green band; the yellow, by throwing together all the
red and green, without the blue band. Each has a
brightness equal to the sum of both the primaries by
which it is constituted.
11. How to see the primaries and secondaries in con-
trast. The primary and secondary spectra in the two
preceding sections are best seen side by side, by making
the black and white bands join, as in Fig. 3. By this
means the complementary colours are brought into
juxtaposition, and by the aid of the prism the eye
instantly perceives the true complementary to any one
of the splendid series of colours on the right or on the
left of the central line. Care should be taken to hold
the prism parallel to the white and black bands, and
opposite to the centre of the diagram, that the comple-
mentary bands may join end to end without confusion.
12. How to see some of the best combinations of the
prismatic colours in contrast. By viewing through a
prism an edge of white against black in a line with an
edge of black against white, as in fig. 4, the beautiful
series of colours produced by throwing together larger
and larger continuous parcels of the prismatic rays,
beginning from the red end of the spectrum, may be
seen in juxtaposition with their perfect complementaries,
formed by throwing together smaller and smaller parcels

ने
9
of the rays beginning from the violet end. These colours,
like those mentioned in the last section, deserve careful
study. The following lists show how the two comple-
mentary series are constituted :-
R+G+B= White
R+G+B=Light Yellow B
0
Black
Dark Blue
R+G
Yellow
B
Blue
·R + G
Yellow-red
B+9
Seagreen-blue
Ꭱ
R
2
Red
B+G
Seagreen
Dark Red
0
Black
B+G+= Light Seagreen
B+G+R=White
The middle colour on each side is about half as
bright as the full white.
13. How to see at once all the best combinations of
prismatic colours. Another beautiful and instructive.
experiment with the prism exhibits at one view all
possible combinations of continuous parcels of the
prismatic rays. This is effected by viewing through
the prism a triangular spot of white upon a black
ground, and a similar spot of black upon a white ground
as in fig. 5, the prism being held at such a distance
that small spots of white and black shall appear re-
spectively on the left hand edges of the two spectra.
The first of these spectra being formed by the partial
overlapping of an infinite number of triangles of all
the prismatic colours, unequally displaced by the
prism, must contain the colours of all possible com-
binations of the prismatic rays in single parcels, large
or small; while the other must contain in each corre-
sponding part the complementary colour proper to the
combination of all the remaining rays.
14. Value of such observations. In the experiments
mentioned in the last three sections the opposite colours
are perfect complementaries; each pair when thrown
together making the white from which the spectra are
produced. It follows therefore
(1) That every colour seen in these experiments
equals its opposite colour in strength of hue.
B 3
10
4
(2) That if any of the colours exceed the mean
brightness (that is, half the brightness of the white), its
opposite colour falls as much below it.
(3) And therefore that the depth of each of the colours
is perfectly matched by the clearness of its opposite
colour-the depth of a colour depending on the propor-
tion which its strength of hue bears to its brightness,
and the clearness of a colour depending upon the
proportion which its strength of hue bears to its dark-
ness, or defect from the brightness of the white.
Thus, the study of the colours exhibited in these
experiments, is not only valuable for teaching us the
colours that are perfectly complementary in hue and
in brightness, but also what is perhaps quite as im-
portant, the colours that match in depth and clearness.
15. The best standard colours. The prismatic colours
and their regular compounds produced in these experi-
ments, are far superior in depth and clearness to the
best colours of pigments and other natural objects
resembling them in hue. The cause of this will be
shown in the next chapter. They therefore form better,
as well as more certain and uniform and more easily
available standards of colour than can otherwise be
obtained. Thus, the colours of the best collections of
the red, green, and blue rays may be taken as the
standard primaries, and the seagreen, pink, and yellow
produced by throwing these together in pairs, may be
taken as the standard secondaries. Still it must be
remembered that even these are but approximations
to the unattainable simple and binary compound sen-
sations of colour, which, together with perfect white
and black, constitute the eight principal colours.
...

Fig.3
$
Fiorf
Fio:5
:
!

11
•
CHAPTER II.
THE COLOURS OF OBJECTS, SELF-LUMINOUS OR
ILLUMINATED.
16. Colours of incandescent solids. All bodies, not
in the state of gas or vapour, when so heated as to be-
gin to be luminous, emit first the extreme red rays of
the spectrum, and as their heat increases emit rays of
higher and higher refrangibility. Hence their colour
is at first red, as seen in iron beginning to glow; then
orange, bright yellow, and still brighter pale yellow, as
seen in iron more strongly heated, in burning coals, or
in flames in which a stream of incandescent particles. of
solid carbon is continually pouring forth; and lastly
(when the substance is so strongly heated that it emits
all kinds of light abundantly) it becomes a dazzling
white, as seen in burning magnesium, or in the lime in
the oxy-hydrogen flame, or in the carbon of the electric
light.
17. Colours of incandescent vapours. Gaseous sub-
stances, or vapours, in a state of incandescence, emit
only some peculiar kinds of light; so that each is dis-
tinguished by its own colour, and their light when
analysed by the prism presents a spectrum of one or
more bright lines alone. Thus the light given out by
burning vapour of alcohol or spirits of wine is a pale blue,
but if common salt is put on the wick, the incandescent
vapour of the salt gives out in addition a bright yellow
light which belongs entirely to a narrow line in the

12
yellow part of the spectrum, and is distinctive of the
metal sodium contained in the salt.
In like manner potassium imparts a purple light,
compounded of red and blue rays; lithium a brilliant
red light; copper a green light. Several such substances,
which are easily vaporized, and whose glowing vapours
are remarkable for their colours, are used to make
coloured flames and fireworks.
18. Effects of different kinds of illumination. Objects
which are not self-luminous can only be seen by that
portion of the light of some luminous body which they
reflect or transmit to the eye. Their colour therefore
depends (1) on the kinds of light with which they are
illuminated, and (2) on the proportions in which they
send those kinds of light to the eye; for while one body
may send to the eye all kinds of incident light in the
same proportion, another (having the property of re-
flecting or transmitting some kinds of light more freely
than others) sends them in different proportions.
Bodies which send to the eye all kinds of light in
like proportion, always appear of the same sort of colour
as the light which illuminates them: thus, for instance,
they are white by daylight, yellow by candlelight, red
by a red light; their brightness depending, of course,
on the quantity of light which they send to the eye.
But bodies which send to the eye a larger proportion
of some kinds of light than of others, often differ widely
from the colour of the incident light. Thus things appear
of different colours, though illuminated with the same
white light of the sun; some sending red rays to the
eyes in greatest proportion, others green, and so on.
Such objects also may have a strong hue by one sort
of illumination and be almost black by another; as things
deep blue by daylight seem black by candlelight, from
its deficiency in deep blue light. Or again, they may
change their colour remarkably; as what is seagreen or
pink by daylight appears green or red by candlelight,
and what is white by daylight appears yellow by candle-
light, all from the same cause. Hence also objects may

13
sometimes appear of different colours under one kind of
illumination, and of the same colour under another, as
those which are white and pale yellow by daylight,
appear both of one yellow by candlelight.
19. Causes of colours of non-luminous bodies. When
light falls on the surface of any object, part of it enters
the body, and part is reflected without entering. Of
that part which enters, part is absorbed or lost within
the body, and the remainder is either transmitted
through the body and emerges on the other side, or is
reflected back again from within the body, and emerges
on the side on which it entered.
The endlessly varied colours and appearances of non-
luminous bodies arise from the different degrees and
ways in which they thus affect the light that illumi-
nates them. In noticing the principal varieties, it will
be supposed, unless the contrary is expressly stated,
that the illuminating light is white and consists of a
mixture of all kinds of light.
20. Colours from transmission of light. First, colours
are produced by light transmitted through an object.
These depend on the quantities of the different
kinds of incident light which escape reflection both
from the outer surfaces of the body and from within
it, and also escape absorption in the body. The power
of reflection being generally the same for all kinds of
light, differences of colour mainly arise from peculiar
powers of absorption. Some bodies, like air, water, or
clear white glass, absorb no kind of light sufficiently
to make a noticeable alteration in the colour when the
thickness is moderate.* Others, like a solution of
verdigris, absorb the red rays most, and therefore
appear seagreen; others absorb the green most, and
therefore appear pink; others the blue, and therefore
* But even the very clearest of these will be found to have their peculiar
colour when the thickness is sufficiently increased; for there is no pon-
derable substance which does not absorb light at all. That the atmosphere
obstructs the passage of all kinds, and of some more than others, is evi-
denced by the dimness and redness of the setting sun, seen through a
thickness of many hundred miles.
14
appear yellow. Others again absorb equally both the
green and blue, and appear red; or absorb equally the
blue and red, and appear green; or absorb equally the
red and the green, and appear blue. Lastly, some
absorb nearly all the rays with such great power that,
like black glass or common ink, they are almost totally
opaque except in small thicknesses.
Two varieties of bodies giving colour by transmitted
light should be noticed:-(1) Such as transmit light
regularly, which, as they allow the forms of bodies to
be seen through them, are called transparent. (2)
Such as transmit it irregularly, scattering it more or
less by internal reflection, and are therefore only trans-
lucent.
In order to see the colour of any such translucent
body by transmitted light, it is necessary to view it
against a white sky, or other sufficiently bright white
surface, no light being allowed to fall on the side next
to the eye.
21. Effect of increasing the thickness of a translucent
body. Owing to the difference of the rates at which
bodies absorb the different kinds of light, the colours
of translucent bodies usually vary with the thickness
of the body. Thus a solution of potassium bichromate
gives first a pale yellow; in greater thickness, a golden
yellow; then orange, approaching at last to dark red.
One of copper sulphate varies in like manner from
bluish seagreen to dark blue. But those substances
which almost destroy at once all the prismatic rays
except a few to which they are very transparent, do
not perceptibly change their colour except by becoming
darker, when their thickness is increased; as may be
seen by looking through one, two, three, or more
thicknesses of the common red glass which owes its
colour to copper oxide, and is very opaque to all but
some of the red rays.
The colour of a translucent body generally approaches
nearer to some particular prismatic colour as the thick-
ness of the body increases; namely, the colour of that
kind of light which traverses its substance most freely.
15
In that case, therefore, they usually become deeper and
more striking to the eye.
*
22. Effect of superposing translucent bodies. On these
principles it will be easily seen that the colour of two
such bodies superposed must not be regarded as a
mixture of their separate colours, but as merely the
colour of those prismatic rays which happen to traverse
both the bodies. Thus a white sky, viewed through a
seagreen glass and a yellow glass laid together, gives
a good green, but much darker than the seagreen or
the yellow of the separate glasses; for each of these
glasses allows the green light to pass through, whilst
the red component of the white light is destroyed by
the seagreen glass, and the blue component by the
yellow glass. In like manner, pink and yellow glasses
together give a red colour, and pink and seagreen
glasses a blue colour.
The composition of colours may often be shown thus.
If a yellow object, for instance, appears as bright
through a red glass, or through a green glass, as a
white one does, but appears black through a blue
glass, it shows that the light of the yellow object has
in its composition all the red and all the green, but
none of the blue light of the white object.
•
23. Colours from internal reflection. Secondly,
colours are produced by light reflected from within the
substance of the object.
This is the commonest case, including the largest
number of organic and inorganic substances. A large
part of the illuminating light enters the body, and
after traversing some small depth of its substance, the
part of it which escapes absorption is reflected; the
part of this reflected light which again escapes absorp-
* In consequence of this, some have supposed the strength of the colour
is increased by increasing the thickness; as if there were more blue, for
instance, in the colour of a thick solution of copper sulphate than in that
of a thin one. But it must be remembered that the absolute quantity of
blue is greater in the colour of the thin solution, though it is there com-
bined with a quantity of green, from rays which are lost in the thick
solution.
16
tion on its return, emerges to enter the eye. Thus
these colours, like those of translucent bodies, are
caused by the absorption of light. Indeed those pig-
ments and dyes which give their colour when laid over
white substances, may be regarded almost entirely as
translucent substances, since their colour is mainly due
to light that passes through them to the white reflecting
substance beneath, and then returns through them again
to the eye, the remainder of the incident light being
extinguished within the substance. In these cases the
stronger the reflecting white, and the more transparent
the pigment or dye, the brighter the resulting colour;
as seen in the brilliant effects produced by transparent
varnishes laid over burnished metal.
24. Effect of mixing pigments. When bodies of this
kind, such as soluble or finely-ground pigments, are
mixed together or laid in washes one over another,
their combined effect is like that of superposed glasses
of the same colours; but the more nearly the pigments
approach to the condition of opaque powders, the less
the resemblance. Thus, a mixture of Prussian blue
and gamboge produce a very good green; whilst cobalt
blue and chrome yellow powders, which are more
opaque, give a colour approaching to gray, which is
more truly the mixture of their separate colours.
25. Effect of thickening a pigment. The colours
given by such substances in very thin layers are
deepened and altered in hue by increase of thickness,
for the same reason as those of translucent bodies.
Thus gamboge in thin washes is yellow; in thicker,
orange carmine in thin washes is nearly pink; in
thicker, nearly red: Prussian blue in thin washes is
seagreen-blue; in thicker, blue.
26. Inferiority of colours from internal reflection.
The colours of bodies seen in this way cannot equal
in depth and richness those which may be derived
from transmitted light. Not only is their light in
general scattered, and therefore much weaker than
that which transparent bodies may transmit when the
sun or other brilliant white light is viewed directly
17
through them; but they are also necessarily more or
less mixed with the white light of superficial reflection,
which is always considerable, and often so great as to
overpower the peculiar colour derived from the light
internally reflected. Thus the colours of a painting
are not well discernible from a point of view in which
the eye receives also a large quantity of light reflected
from its surface. Hence to see objects of this kind to
best advantage they should be viewed in some direction
very divergent from the direction in which the greatest
quantity of incident light is reflected, so that their
proper colours may be as little as possible diluted.
The effect of a transparent varnish, laid over a
coloured surface which scatters light by a superficial
reflection, is to deepen the proper colour from most
points of view by preventing reflection from the original
surface, and producing on itself a glossy surface which
reflects light regularly, instead of scattering it in all
directions.
27. Effect of the form and texture of a surface. The
colours of all such bodies are also much deepened
when the texture or form of the body is such as to
cause repeated reflections from one part of it to another.
Hence the depth of colour which may be seen in vel-
vets and woollen stuffs and in the hollows and interstices
between the petals of flowers, which is lost when the
parts are viewed separately. The effect is the same as
if the object were illuminated by light of its own
proper colour; the superficial reflection being tinged,
as well as the internal reflection being deepened in
colour by the repetition of the process of absorption.
But in general the hue is also altered more or less by
this means, from the cause which modifies the colours
of translucent bodies when their thickness is increased;
namely, the unequal rates of absorption for the dif-
ferent kinds of light.
28. Colours from superficial reflection. Thirdly,
colours are produced by light reflected from the outer
surface of an object.
18
Endless varieties in the appearance of bodies are
produced by reflection, according to the form and
nature of the surface, the reflecting power of the sub-
stance, and the nature and direction of the illumination.
In superficial reflection the quantity of light reflected
increases considerably with the obliquity of the inci-
dent light; but this is more apparent with polished
surfaces than with those which scatter the light in
all directions. Two kinds of superficial reflection
must be noticed.
Ordinary superficial reflection is almost wholly the
same for all kinds of light, and therefore dilutes with
white the colours produced in other ways. Its effect
is best observed in bodies which powerfully absorb all
light which enters them, as black marble, rough or
polished, where the superficial not being complicated
with the internal reflection, it is easy to see that the
hue of the reflected light is the same as that of the
illuminating light.
29. Colours from metallic superficial reflection. Certain
bodies, however, especially metals, possess an extra-
ordinary power of superficial reflection, attended with
almost total opacity. The reflective power of silver
(the most intensely reflective of all bodies) is very
nearly equal for all kinds of light; and this is also the
case with most metals, but not with all. Thus gold
and copper reflect yellow and red light most powerfully.
A few non-metallic substances also have the same
property, either for all, or for some kinds of light.
Plumbago for instance gives a gray metallic reflection;
compressed indigo gives a red one; potassium perman-
ganate a green one (as may be seen in its crystals, and
even in a strong solution of it, which is a beautiful pink
by transmitted light and very opaque to the green rays).
The splendid colours of some birds and insects,
distinguished by their peculiar metallic appearance, and
by the richness of their hues increasing when the
reflected light increases, instead of being more and
more diluted by white, are also due, in part at least, to
this kind of superficial reflection.
19
30. Effect of repeating reflections. Colours produced
by superficial reflection are deepened and altered in hue
by causing the light to be repeatedly reflected from one
part of the surface of the body to another. Thus the
inside of a gold cup gives a deep orange red, instead of
yellow, like a plane surface of the same metal; and
even pure silver under like circumstances gives a pale
yellow, though on a plane surface it shows no per-
ceptible tendency to any hue.
The colour to which such bodies tend under these
conditions is of course the colour of that kind of light
which they most copiously reflect.
31. Colours from reflection from small particles.
Fourthly, colours arise from a peculiar kind of reflect-
ion which takes place from small foreign particles
included in a transparent medium, whether gaseous
liquid or solid.
When the particles are very small the only kinds of
light they reflect are those which are most refrangible;
and when the particles are larger they reflect more
and more of the other kinds also. Hence the colour
of the reflection is deep blue in the first case, then
becomes more azure and paler, until, when the particles
are large enough, it is white; and the light thus re-
flected is always scattered in all directions.
Hence the beautiful blue of the sky, which is reflected
from extremely minute particles of water produced by
the condensation of aqueous vapour near the uttermost
limit of the atmosphere; larger particles of the same
kind in the clouds, reflecting white. The blue of the
lower part of the atmosphere, the blue of wood smoke,
of diluted milk, and of some kinds of opalescent stones
and glass and other natural objects (viewed against a
black background), is attributable to like causes.
consequence of such objects scattering so much of the
incident blue, or blue and seagreen light, is of course
that they appear by transmitted light of a yellow,
orange, or reddish hue.*
The
* See Professor Tyndall on this. Philosophical Transactions, 1869.
2
*
20
32. Colours from interference. Fifthly, variable
colours are produced in a great variety of ways by what
is called the interference of the waves of light, whereby
certain kinds of light are destroyed, and other kinds
are increased in intensity, on some parts of the surface
of a body; while the reverse takes place on other parts,
according to the direction in which the surface is viewed.
The two most common groups included in this class
are (1) the colours of finely-striated surfaces, such as
those of mother-of-pearl, which have been successfully
imitated by engraving lines extremely narrow and close
together upon steel or glass; (2) the colours of thin
plates such as those of soap-bubbles, of the wings and
scales of some insects and fishes, of the coat of dross
that forms on the surface of some metals, and the
iridescence often seen on ancient glass. Such colours
may be produced in great beauty by reflecting light
from the thin layer of air between two pieces of clean
polished glass pressed hard together. Looking through
the thin layer of air pale colours may also be seen,
complementary to those of the reflected light.*

33. Colours from fluorescence. Sixthly, colours are
produced by what is called fluorescence, a property
which certain bodies have of emitting new light of
certain kinds when they are acted on by rays of other
kinds.
Greenish fluor spar, for instance, when placed in
sunshine emits a peculiar blue luminosity, from which
the term is derived. The like occurs with a decoction
of the bark of the horse-chesnut and of the ash tree,
and with a solution of quinine. Uranium glass exhibits
a yellow-green fluorescence, and a tincture of the green
colouring matter of leaves, a red fluorescence. Objects
which have this property may be regarded as self-lumi-
* By transmitting what is called polarized light through a plate of mica,
selenite, or other substance which possesses the property of double re-
fraction, interference-colours are produced in great beauty, which by
inaking the plate revolve through a quarter of a circle change into their
perfect complementaries through gray, as from red to seagreen, green to
pink, blue to yellow; but the necessary apparatus is somewhat costly,
and the full explanation lengthy.
21
nous in some degree, under the action of those rays
which excite it; especially as these are often invisible
rays, more refrangible than the extreme violet rays.
*
34. How to test the colours of objects. The compo-
sition of the colours of natural bodies of all sorts may
be readily tested by looking through a prism at a stripe
of the object continuous with a stripe of white paper
of the same breadth, over a perfectly black ground.
The rays which the object sends to the eye will appear
in their proper places in the spectrum of its stripe,
whilst the rays which it does not send, or in which it is
defective as compared with white, will be discovered by
comparing this spectrum with the adjoining spectrum
of the stripe of white paper.
If the colour to be examined belongs to a translucent
substance, it may be placed behind part of a slit in
an opaque screen, and the light of a white sky be
transmitted through it. If it belong to a pigment, a
portion of a stripe of white paper may be covered with
the pigment (taking care to cover well the edges, and
to make the ends square) and the stripe, placed over
a dark cavity, may then be viewed through the prism.
The colours of flowers may be examined in like manner
by laying small rectangular portions of the leaves on
a stripe of white paper of the same width.
Such experiments afford an admirable means of
instructing the eye to judge correctly of the constitu-
tion of all sorts of colours. The method may also be
advantageously used by those who seek to improve the
manufacture of pigments, or to make good mixtures
of pigments. The more nearly the pigment or mixture
reflects some continuous parcel of the prismatic rays
without diminution, while it totally destroys the rest,
the more striking is its colour.
* Fluorescence was first explained by Prof. G. G. Stokes (Phil. Trans.
1852). It has been detected in many bodies, but its effect is generally
too small to be noticed. Possibly the red tinge in the extreme violet rays
of the spectrum may arise from fluorescence of the retina.
1
22
CHAPTER III.
INTERMEDIATE COLOURS, AND MEANS BETWEEN TWO OR
MORE GIVEN COLOURS.
35. How to find intermediate colours. The colours
intermediate between one given colour and another
may be shown in the following way :—

Fig. 6
Lay a spot of each of the two colours upon an
horizontal surface, and erect midway between them
a slip of clean polished glass as represented in fig. 6
23
One of the spots may then be seen reflected from that
part of the glass through which the other is seen, and
their colours are mixed or blended together. When
the eye is placed low, the mixture contains most of the
colour of the far spot, and very little of the colour of
the near spot; when high, the reverse; because the
more obliquely the rays fall upon the glass the more
they are reflected, and the less they are transmitted.
If the eye is raised therefore from a low point to a high
one near the plane of the glass, the mixture will
change through the whole gradation from the colour of
the far spot to that of the near one.
If, instead of spots, stripes of the two colours are
laid, meeting each other end to end at the glass, so
that the one will be reflected from the part of the glass
through which the other is seen, the whole gradation
may be viewed at once.
In all these experiments care should be taken that
the two spots or stripes are equally illuminated; and it
is best to lay them on a neutral ground.
36. How to find the mean of two colours. The exact
mean between two given colours (that is, the colour
which is produced by combining half the light reflected
by one of the coloured surfaces with half of that
reflected by the other) is easily found by the same
method. For, if spaces of white and black are put
on either side of spots of the given colours opposite to
each other (as shown in the same fig. 6), then where
the white reflected by the glass is exactly as strong as
the transmitted white, so as to produce the same gray
on each side of the blended colours, we may be sure
that the resulting colours contain half of each of the
given colours.
37. How to see whether colours are complementary.
If we desire to try whether two colours are perfect
complementaries, we have not only to find (1) whether
they neutralize each other on the glass in some position
of the eye, but (2) whether they do so at that particular
position of the eye in which half of each is thrown
together, and (3) whether they then produce a gray.

24
equal in darkness to the mean between black and
white. The first condition being fulfilled merely shows
that their hues are complementary in kind; the second,
that they are also of equal strength; the third alone
proves that the lights of the two thrown together
would make up the full white, and that they are there-
fore complementary in light and shade, as well as in hue.
38. How to find the mean of several colours. The
colour which is the mixture of two, three, or more
colours, in any given proportions, may be found by
covering a circular disc with the given colours in sectors
of the required proportion, and giving it a rapid rotation
on its centre in the manner of a top.
39. How to compare means of different sets of colours.
The mean of a number of given colours may be com-
pared in this way with the mean of any others.
Let a
circle be drawn midway between the centre and the
circumference of the rotating disc, and let the part of
the disc outside this circle be covered in the proper
proportion with the first set of colours, and the part
inside with the second set. When the disc revolves,
the means of the two sets will at once be seen in
juxtaposition.
Thus, to see whether some given red, green, and blue
will neutralize each other to a gray one third of the
brightness of white, cover a third of the outer part of
the disc with each of them, and let one third of the
inner part be white and the rest be black. If the
colours are perfectly correct the whole disc when
revolving should appear one uniform gray.
40. Truth of results. The means of two colours
found by the method of rotation (half of the rotating
disc being covered with each of them) perfectly agree
with the results shown by the slip of glass. Those
pairs of colours, also, which are found by experiments
made with a slip of glass or a rotating disc to have a
gray for their mean colour, always correspond in hue
with colours which appear by experiments with the
prism to be complementary with each other. Indeed,
the more nearly we imitate with pigments the colours

25
of the prismatic complementaries, the more nearly do
the colours of our imitations neutralize each other in
equal proportions on the glass or the rotating disc.
Thus the colours of vermilion and verdigris, of emerald
green and rose madder, of cobalt blue and king's
yellow, respectively, which approximate to the standard
prismatic primary and secondary colours, produce in
either of those ways a nearly perfect gray.
41. Means between principal colours. It is useful to
notice the effects of mixing thus the principal colours
in equal quantities, as in the following instances:-
The means between pairs of full primaries are dark
secondaries, the same as the means between the full
secondaries and black; namely, dark seagreen, between
green and blue; dark pink, or crimson-purple, between
blue and red; dark yellow, or olive-green, between red
and green.
The means between pairs of full secondaries are
light primaries, the same as the means between the
full primaries and white; namely, light red, between
pink and yellow; light green, between yellow and
seagreen; light blue, between seagreen and pink;
which three means are respectively complementary to
the former three.
The means between the full primaries and black,
and those between the opposite full secondaries and
white, are also perfect complementaries; namely, dark
red, and light seagreen; dark green, and light pink;
dark blue, and light yellow.
The means between the several full primaries and
each of their cognate secondaries are all rich hues of
medium brightness; namely, yellow-red, or orange,
between red and yellow; yellow-green, between green
and yellow; seagreen-green, between green and sea-
green; seagreen-blue, or azure, between blue and
green; pink-blue, or puce, between blue and pink ;
pink-red, or cerise, between red and pink; the former
three of which are respectively complementary to the
latter three.
a combination of Hive, red.
white tred: light red
愛​ダームール
​C
...
26
The means between the full primaries and their
opposite secondaries are of course all the same gray
as the mean bet n black and white.
The mean or average colour of any set of three pri-
maries of equal strength, is a gray of their mean
brightness; and the same is true of any set of three
secondaries, and indeed of any set of colours which are
equidistant on any chromatic circle, that is, equidistant
in hue and equal in strength.
42. Shades of tints. Nature abounds with objects
whose colours are compounded of the three simple
colours in not very unequal quantities; and which
therefore do not differ much from a gray. These are
called tertiaries, from the character of their compo-
sition. Among them may be mentioned maroons,
russets, browns, citrines, olive-greens, sage-greens,
stone, slate, and lavender colours, in all their endless
varieties, light and dark.
All such grayish colours are in fact shades of tints,
that is, intermediates between some clear light colour
and black; or they may be equally well regarded as
tints of shades, that is intermediates between some
deep dark colour and white.
43. How to ascertain the real hues of obscure colours.
The methods suggested for blending colours are not
only useful in finding the complementary of any given
colour, or the general tone of any composition of
colours; they also enable us to detect the real hues
of those obscure colours in which the predominance of
one or more of the simple sensations over the rest is
too small to be easily recognized.
For it is generally easy to see by inspection to which
of the six principal gradations the hue of the given
colour belongs; that is, whether it lies between pink
and red, or red and yellow; between yellow and green,
or green and seagreen; or between seagreen and blue,
or blue and pink. Let then a stripe of paper, coloured
with the gradation to which the hue belongs, be laid
parallel to the foot of the vertical slip of glass, and be
27
reflected against a neutral ground on the other side of
the glass varying in darkness from white to black. It
will be possible so to adjust the posit of the eye,
and the distances of the coloured stripe and of the
neutral surface from the glass, that the obscure colour
in question shall be exactly imitated by the mixture
of the strong colour of some part of the stripe with the
dark or light gray of some part of the neutral surface;
and thus the real nature of the colour in question may
be determined.*
CHAPTER IV.
THE NATURAL SYSTEM OF COLOURS AND ITS USE.
44. Construction of the colour-cube. Since all colours
are combinations of three simple sensations, any one of
which may vary in intensity from nothing up to the
highest degree which the organ of sight can bear, all
colours may be supposed to be arranged in the form of
a cube in the following manner :—
Let three sides meeting. in one corner of the cube be
the sides of no red, no green, no blue, and at right
angles with every part of these sides respectively let
red green and blue be supposed to increase in intensity
up to the strength in which they exist in a full white
in the opposite three sides, which will therefore be the
sides of full red, full green, and full blue. Hence the
* Instead of thus dulling a strong colour to match the obscure one,
the object may sometimes be attained by throwing a strong white light
by a lense or reflector upon a spot of the obscure coloured surface, so
strengthening its hue till it matches as nearly as possible the colour of
some part of the less illuminated gradation of colours.
c 2
28
colour belonging to any given point in the cube will be
compounded of red green and blue in the proportions
of the distances of that point from the three sides of
no red, no green, no blue.
Thus if Z in fig. 7 is the corner in which the three
initial sides meet, it will be the place of black (the
zero of all colour); and if ZGSB is the side of no red,
ZBPR the side of no green, and ZRYG the side of no
blue, then it easily follows from the construction of the
cube that R, G, B, the corners next to Z, will be the
places of full red, full green, and full blue, and that
S, P, Y, the corners respectively opposite to R, G, B,
will be the places of full sea-green, full pink, and full
yellow. Also that W, the corner opposite to Z, will be
the place of full white; and that the centre being half
the distance of W from each of the three initial sides,
will be the place of a gray of half the brightness of
the full white. In like manner the composition of the
colour of any point may be found when its proportionate
distances from the initial sides are known.

W
Fig 7
Y
P
S
R
G
B
N<
The frontispiece contains two perspective views of
the colour-cube, representing the principal colours and
29
their means, twenty-seven in all. The lower diagram
shows the three sides adjoining to the corner of black,
the upper diagram shows the three opposite sides,
reversed in position so that the complementary colours
may be vertically over each other. These diagrams
should of course be viewed by the white light of day;
but even then the failure of the pigments to give a
true representation of the colours intended will be
very apparent. Some deviate considerably in hue;
and their depth and clearness is unequal, though pains
have been taken to use the best, and in the most
advantageous thickness, so as to give some idea of
what the six sides of a perfect colour-cube should be.
45. Rectilineal gradations. From the construction
of this cube it also follows that every straight line
drawn through it, will pass through a uniform gradation
of colours from end to end; and also that the gradations
along all parallel lines will be of the same nature,
being produced by the regular addition or substraction
of some one and the same colour. Thus, the gradations
along ZR, and all lines parallel to it (fig. 7), are pro-
duced by regular increase of red alone; the gradations
along ZY, and all lines parallel to it, are produced by
the regular and equal increase of both red and green,
or of yellow; the gradations along ZW, and all lines
parallel to it, are produced by the regular and equal
increase of red green and blue, or of white; again, the
gradations from R to G, and on all lines parallel to it,
are produced by the regular and equal decrease of red
and increase of green; and the gradations from R to S,
and on all lines parallel to it, are produced by the
regular and equal decrease of red and increase of both
green and blue.
Hence, the middle point between the places of any
two given colours, is the place of their mean colour;
and the colours at the extremity of any line passing
through the centre of the cube, and bisected by it,
are perfect complementaries.
46. Plane gradations. Again, the colours that belong
to any plane section of the cube will vary according to

30
certain laws in every direction; and those of all parallel
sections, varying, as they must, in the same manner in
the corresponding directions, will have certain relations
to each other. Thus, the plane ZRYG will present an
orderly group of all colours in which there is no blue;
the parallel plane BPWS will give the same colours
modified by the addition of the full blue to each of
them; and a parallel plane half way between these
two would give them modified by the addition of half
the full blue to each.
Again, any section at right angles with the line
ZW (which may be called the principal axis of the
cube) would present an orderly group of all the colours
possessing some equal degree of brightness or lumin-
osity.
In all sections that can be made through the colour-
cube, the central colour must be the mean or average
of all the colours, both in hue and brightness, and
must therefore express the general tone or effect of the
whole of that peculiar natural arrangement or harmony
of colours which the section presents.
47. Complementary plane gradations. If any section
is taken not through the centre of the cube, it is easy
to see that another section may be made parallel to it
at the same distance on the other side of the centre,
which will contain all the colours perfectly comple-
mentary to those of the first section.
Thus, a section at right angles with the principal
axis of the cube (ZW) taken near the corner of black,
will pass through the dark primaries and their inter-
mediates; and a corresponding parallel section near the
corner of white will pass through the complementary
light secondaries and their intermediates. Again, the
colours of a section near and even with the edge ZR
will in like manner be complementary to those of the
corresponding section near the edge WS; and so on.
48. Cognate plane gradations. Again, as all parts of
a cube are arranged symmetrically on a ternary system
around the principal axis (ZW), any section that is not
taken at right angles with the principal axis may be

31
accompanied with two other sections, so disposed that
the three sections shall all bear the same relation to
the three primaries respectively; that is, to the three
initial sides.
Thus, ZRWS, the section which contains the red
and seagreen colours, is cognate with the sections
ZGWP and ZBWY, which contain the green and pink
colours, and the blue and yellow colours.
And moreover, if any such section is not symmetri-
cally taken with reference to either of the three initial
or terminal sides, it may be accompanied with another
section, deviating as much from symmetry as the first
does but in the opposite direction; and then, instead
of three, there will be three pairs of cognate sections;
and they will all represent certain natural arrangements
or harmonies of colours having certain relations with
each other.
49. Complementary and cognate rectilineal gradations.
It is obvious too that the regular gradation of colours
represented in the cube by any straight line (except the
principal axis) must have three or six cognate grada-
tions which may easily be found by noting the places
where the line intercepts the sides of the cube; and
that all these must have their corresponding comple-
mentary gradations; so that sets of gradations may
always be found bearing certain relations to each other.
For instance, the following six (see fig. 7) would
be cognate gradations:-
From the middle of ZR to the middle of RY;
From the same to the middle of RP;
From the middle of ZG to the middle of GS;
From the same to the middle of GY;
From the middle of ZB to the middle of BP;
From the same to the middle of BS;
(that is, from dark red to orange, from dark red to
cerise, from dark green to seagreen-green, from dark
green to yellow-green, from dark blue to puce, and
from dark blue to azure).
And the following six would be respectively com-
plementary to them:-
32
1
From the middle of WS to the middle of SB;
From the same to the middle of SG;
From the middle of WP to the middle of PR;
From the same to the middle of PB;
From the middle of WY to the middle of YG ;
From the same to the middle of YR;
(that is, from light seagreen to azure, from light sea-
green to seagreen-green, from light pink to cerise,
from light pink to puce, from light yellow to yellow-
green, and from light yellow to orange).
Also the single colour represented by any given point
in the cube (not being in the principal axis) has in like
manner its cognate colours, and all these have their
complementary colours; so that regular sets of single
colours may also be suggested and easily found by the
aid of the colour-cube.
50. Limits of colours in depth and clearness. Only
colours in which some one or more of the colour-
sensations is not at all excited, and which belong
therefore to one or other of the three initial sides of
the cube, may be regarded as perfect in depth; while
colours in which one or more of the colour-sensations is
fully excited, and which therefore belong to one or
other of the three terminal sides, may be regarded as
perfect in clearness.
In practice, however, it must be remembered that
(excepting black and white) we cannot get those theo-
retical pure or perfect colours which would find their
places on the surface of the colour-cube; but the
deeper a colour is, the nearer its place in the cube to
one of the three initial sides, and the clearer it is, the
nearer to one of the three terminal sides. Hence, as
the colours produced by the extreme rays of the red and
blue bands of the spectrum are the deepest possible,
we may approach nearer to the surface of the cube on
the side of no green, to which belong the blues, violets,
purples, crimsons, and reds, than to any other initial
side. Again, as the perfect complementaries of the
deepest colours must be the clearest colours, we may
33
approach nearer to the opposite side of full green, to
which belong the yellows, greens, and seagreens, than
to any other terminal side. This perfectly accords
with what is observed in the colours of pigments,
flowers, insects, birds, and other natural objects; the
best blues, purples, crimsons, and reds are unsurpassed
in depth, the best yellows, light greens, and seagreens
are unsurpassed in clearness.
51. Utility of the natural system. The advantage of
studying the natural system of colour contained in the
colour-cube is not only the infinite variety of different
gradations and arrangements of colour which it suggests,
but also the endless symmetrical combinations of them
which it will enable the designer to invent with a preci-
sion and completeness that otherwise would be impos-
sible; besides the facility with which by its aid the
true affinities of colours may be borne in mind, together
with their practical limits in depth and clearness.
By means of a model, which may be formed by
colouring a small cube of paper, it will be easy to
trace out the colours belonging to any lines and
planes which may be conceived to intersect the cube,
as mentioned in sections 45-49. Thus many beautiful
natural harmonies of colour (as they may be called)
may be studied, besides the six which are represented
in the frontispiece.*
* In another treatise, "The Principles of the Science of Colour," this
part of the subject is illustrated by diagrams of natural harmonies of
colours taken from about sixty different sections of the cube, and arranged
so that either each single group of colours, or each set of parallel or of
cognate groups may be viewed together. Methods of forming innumerable
regular compositions out of such natural harmonies, are also suggested
in the same work.
c 3
34
CHAPTER V.
OCULAR MODIFICATIONS OF COLOURS; THE SENSIBILITY
OF THE EYE TO LIGHT.
52. Variable sensibility of the eye. The excitement
of a colour-sensation diminishes for the time the
capacity of the retina for the further excitement of
that sensation; and any rays of light which then enter
the eye do not produce that particular sensation so
powerfully as they would do if the eye was not so
excited. This defect of sensibility continues for some
short period after its cause has ceased to act.
For example, after looking fixedly for some time at
a very brilliant red spot against a black background, if
we look suddenly at a dark red surface, that part of
the retina which received the impression of the bright
red will be hardly sensible to the faint red light which
it now receives, and a nearly black spot will therefore
appear on the dark red surface. Or if in the same
case we look at a dark yellow or olive-green surface, a
spot of dark green will appear upon it; if we look at
a purple surface, a spot of dark blue will appear; if
at a gray surface, a spot of dark seagreen: all obvious
effects of the same cause, the reduced sensibility of
the eye to red light. And if strong green or blue is
substituted for the exciting red, and the eye suddenly

i
35
:
directed to a surface of some other colour, corre-
sponding effects follow; an "ocular spectrum," as it
is called, being at once seen, whose colour is that which
would be produced by subtracting a portion of green
or blue from the true colour of the surface we look at.
53. Law of variation. The loss of sensibility thus
caused is proportionate to the intensity of the excite-
ment; so that if the three simple colour-sensations
are excited all at once in unequal degrees (as in fact
they always are when we look at any bright object that
is not white), the loss of sensibility will be greatest for
the sensation most powerfully excited, and least for the
sensation least excited.
Thus if we look at a brilliant light orange surface
(such as a bright gold coin) on a dull gray background,
and then suddenly remove it without moving the eye,
we see in its place a darkish spot on the gray, tinged
with the colour complementary to the orange. For
the eye which was before most strongly excited to the
sensation of red, rather less so to that of yellow, and
still less so to that of blue, is now least sensible to the
red component of the gray, rather more sensible to
the green, and most sensible to the blue; it therefore
sees, instead of gray, a dull seagreen-blue, comple-
mentary in hue to the orange.
54. Simultaneous contrast. As the eye is ever
glancing forwards and backwards from one part of
the objects viewed to another, the consequence of the
variableness of its sensibility is that the different
colours of neighbouring parts always more or less
modify each other. For they are sure to be some
mixture of the three sensations; and the proportions
of the mixture must become altered according to the
nature of the previous excitement of the eye.
These modifications of colour are commonly referred.
to as the effects of "simultaneous contrast."
55. Effects of simultaneous contrast. The general
effect of simultaneous contrast is to make the differ-
ence of two colours in brightness and hue appear
greater than it really is.
36
Thus a bright white subtracts whiteness from neigh-
bouring objects and causes their colours to appear
darker and deeper than they really are, but without
altering their hue. In like manner,
Strong seagreen subtracts seagreen; and therefore
dulls another seagreen, and tends to alter
Seagreen-blue to Blue,
Blue to Pink-blue,
Pink-blue to Pink,
Pink to Pink-red,
Pink-red to Red,
Seagreen-green to Green,
Green to Yellow-green,
Yellow-green to Yellow,
Yellow to Yellow-red,
Yellow-red to Red.
It deepens red, and tinges gray with red.
Strong pink subtracts pink; and therefore dulls
another pink, and tends to alter
Pink-red to Red,
Red to Yellow-red,
Yellow-red to Yellow,
Yellow to Yellow-green,
Yellow-green to Green,
Pink-blue to Blue,
Blue to Seagreen-blue,
Seagreen-blue to Seagreen,
Seagreen to Seagreen-green,
Seagreen-green to Green.
It deepens green, and tinges gray with green.
Strong yellow subtracts yellow; and therefore dulls
another yellow, and tends to alter
Yellow-green to Green,
Green to Seagreen-green,
Seagreen-green to Seagreen,
Seagreen to Seagreen-blue,
Seagreen-blue to Blue,
Yellow-red to Red,
Red to Pink-red,
Pink-red to Pink,
Pink to Pink-blue,
Pink-blue to Blue.
It deepens blue, and tinges gray with blue.
Strong red subtracts red; and therefore dulls another
red, and tends to alter
Yellow-red to Yellow,
Yellow to Yellow-green,
Yellow-green to Green,
Green to Seagreen-green,
Seagreen-green to Seagreen,
It clears seagreen, and tinges
Pink-red to Pink,
Pink to Pink-blue,
Pink-blue to Blue,
Blue to Seagreen-blue,
Seagreen-blue to Seagreen.
gray with seagreen.
37
Strong green subtracts green; and therefore dulls
another green, and tends to alter
Seagreen-green to Seagreen,
Seagreen to Seagreen-blue,
Yellow-green to Yellow,
Yellow to Yellow-red,
Seagreen-blue to Blue,
Blue to Pink-blue,
Yellow-red to Red,
Red to Pink-red,
Pink-blue to Pink,
Pink-red to Pink.
It clears pink, and tinges gray with pink.
Strong blue subtracts blue; and therefore dulls
another blue, and tends to alter
Pink-blue to Pink,
Pink to Pink-red,
Pink-red to Red,
Red to Yellow-red,
Yellow-red to Yellow,
Seagreen-blue to Seagreen,
Seagreen to Seagreen-green,
Seagreen-green to Green,
Green to Yellow-green,
Yellow-Green to Yellow.
It clears yellow, and tinges gray with yellow.
Black of course seems to brighten white, and to
make all other colours lighter and clearer; that is, as
compared with what they would appear if seen in
juxtaposition with any light colour.*
56. Accommodation of the eye to light. There is a
certain degree of brightness which is on the average
most agreeable and suitable to each person's sight. If
we take the sensibility of the eye while contemplating
the mean of all colours, or the middle gray, belonging
to that degree of brightness, as its regular or normal
state, there are two ways in which that wonderful
organ tends to diminish the difference in the intensity
of the sensations that would otherwise arise from a
difference in the quantity of light:—
(1) By the diminution of sensibility under increased
excitement, and the increase of sensibility under di-
minished excitement, which has been already discussed
in this chapter;

* See Chevreul's treatise on this subject. His chromatic circle, how-
ever, and complementary colours require correction from the red-yellow-
blue to the red-green-blue system.
Wor!
38
(2) By the contraction of the pupil of the eye
which always follows an increase of light, and the
enlargement which follows a diminution of light. This
self-adjustment of the area of the circle that admits
light to the eye makes that area under a very bright
light less than a tenth part of what it is in the dark.
But though the effects of changes in the illumina-
tion of objects are thus in some degree lessened, yet
the range of difference in the intensity of the colour-
sensations which the eye can receive without injury
is extremely large. The brightness of white paper
(for example) illuminated by the direct beams of
the sun must be hundreds of times greater than its
brightness in the shade, even supposing the eye to
have accommodated itself to the conditions of each
case.
57. Sensibility of the eye under a given illumination.
With light of a moderate brightness, a good eye is
just sensible of an increase or decrease of about one
hundredth part of that brightness. This is best
shown by illuminating a white screen with two equal
lamps, placing one lamp near the screen, and removing
the other to such a distance that a shadow cast by it
upon the screen first becomes imperceptible (the eye
being shaded from the lamps). If the distances of the
two lamps from the screen are then measured it will
be found that the latter lamp is about ten times further
than the former, showing that a difference of less
than one hundredth part of the light (namely, of the
brightest light by which the eye is at the time affected)
is imperceptible.
This rule of course does not hold good when the
brightness is so great as to be dazzling to the eye, or
so small as to be insufficient for distinct vision.
In
approaching either of these extremes the eye is neces-
sarily less capable of being sensibly affected by small
changes of the intensity of the light. But for a very
wide range of brightness, as for instance from that of
white paper in sunshine down to that of white paper
shaded from the sun and great part of the sky, nearly
39
the same number of steps may be distinguished at
once between white and black, the eye being affected
by the white in each case and by nothing brighter.
Hence, as soon as the eye has adapted itself to the
circumstances in which it is placed it distinguishes
shades and hues of colour with nearly equal power
under very different degrees of illumination; and the
importance is evident of shading the eye from any
glaring light, whenever we would discriminate the
colours of objects that send nearly the same lights to
the eye.*
58. Number of distinguishable colours. It would
seem then that if the full white with which the eye
is affected is of any moderate brightness, about one
hundred gradations may be distinguished at once
between it and black; and as the eye is no doubt
equally sensible to changes in each single colour-
sensation, it may be estimated that the whole number
of perceptible variations of colour would be about
one million (the cube of one hundred), were it possible
to excite any one of these sensations by itself to the
degree in which they are all excited together in white.
But this is so far from being possible that probably
not a tenth part of that number of colours could ever
be really produced, distinguishable at once under any
given intensity of white light.†
59. Effect of very feeble light. When the eye is very
feebly excited, it seems not to receive exactly the same
mixture of sensations from any kind of light as it does.
under a moderately strong illumination; and the dif-
ference is not the same with different kinds of light.
For such red and blue surfaces as by full daylight
appear equally bright, are no longer so when viewed
by a very dim light of the same character. The red
Experiments have been made by Fechner and other continental
philosophers on this subject.
+ This appears from a comparison of the best observations which have
been made on the brightness of the prismatic rays, with their respective
strength of hue as determined by Mr. Maxwell.
40
becomes darker than the blue, as if the one were mixed
with black and the other with white. This effect may
be best observed if a spot of one colour is put upon
the other as a ground, or if spots of both are put near
together on a gray ground as nearly as possible equally
luminous with them: and as it appears not only when
the light is diminished, but when the spots are viewed
with nearly closed eyes, it must be an ocular effect.
Hence the changes sometimes noticed in the relative
brightness of deep reds and blues in a picture under
different degrees of illumination. But it is certain that
their relative strengths of hue do not change; other-
wise what is pure white in the brighter light would no
longer be so in the dim light.*
60. Effect of dazzling light. On the other hand
when the eye is very strongly excited by any kind of
light, as by any of the prismatic rays in dazzling
brightness, their colours seem all to approach towards
white, as if the retina became incapable of receiving
the particular sensations which they most powerfully
excite in much greater degree than the other sensations,
which are doubtless always more or less mixed with
the principal ones. Thus the sun's disc viewed through
any kind of coloured glass which does not too greatly
obscure it, appears nearly white; more so at least than
any white cloud viewed through the same. Yet the
glass transmits similar light in both cases, as may be
seen by putting white paper in the transmitted sun-
beam; so that it is evident the effect is purely ocular.
*In trying these effects care should be taken that the illuminating
light varies only in strength; for slight changes in its hue, from altered
proportions of its component lights, may considerably affect the bright-
ness of colours in the way mentioned in § 18. Thus the approach of
sunset, the clouding over of a blue sky, or even a haze coming on, affects
the colours of some pigments in a remarkable manner, by diminishing
the proportion of certain prismatic rays in the light which they reflect.

1
41
CHAPTER VI.
THE HARMONY OF COLOUR.
61. Harmony the source of beauty. When several
colour-compositions of like character are compared,
whether they consist of different colours or of the
same colours differently arranged, it is in general seen
at once that some are more pleasing than others.
Nor can there be a doubt but that, whether we con-
template works of art, as paintings, mosaics, stained
glass, mural decorations, hangings, dresses, furniture,
porcelain, jewellery, or natural objects, as the grand
scenery of land, sea, and sky, woods, rocks, meadows,
plants, leaves, flowers, animals, birds, insects, the vast
majority of those who see colours alike are agreed in
the objects of their preference in this respect.
Whatever it may be in the association of colours
that causes a composition to be thus generally felt to
excel in beauty, may be included under the general
name of harmony; for the term implies the right
fitting together of the colours to produce an agreeable
effect.
62. Principles of harmony. Colour-compositions
are endlessly varied in character. They may have
only two or three colours, or they may have many;
the colours may meet in sudden contrast, or blend
into each other by imperceptible gradations; the dif-
42
ferent hues may be bordered with other hues, or with
white, gray, or black; the colours may be the deepest
or the clearest possible, or they may be grayish and
nearly neutral in hue; they may be dark or light, or
a mixture of all sorts. And as some of all these
kinds are found to excel in beauty, it would seem that
harmony, the cause of beauty, consists in something
independent of all such circumstances.
The following rules, derived from a careful examina-
tion of different examples, are founded mainly on the
facts (1) that the excitation of each colour-sensation
up to a certain degree is attended with pleasure, and
also its cessation again after a certain interval; and
(2) that when we contemplate a colour-composition we
almost unconsciously direct the eye to its different
parts in succession, so that that portion of the retina
which is endowed with the property of distinct vision
(and perhaps every other part as well) becomes affected
with or relieved from the different colour-sensations
according as the object to which it is directed sends or
does not send to the eye the rays which excite such
sensation.
63. RULE I. Each colour-sensation should be excited.
A composition which is complete in itself should affect
the eye equally with the three colour-sensations, either
jointly or separately. In other words there should be
a balance of bue in the whole composition, and its
mean colour, or general tone, should be some shade of
white.
To illustrate this in the simplest manner, we may
compare the effect of filling the whole field of view
successively with several grounds of different hues, or
neutral, but all of some moderate brightness. At first
they all may appear equally pleasing as a change from
darkness; but we dwell longest with pleasure on one
that is neutral, and the eye is less fatigued when all
the three sensations are excited equally to produce the
given brightness, than when one or two of them are
more strongly excited in excess of the third.
43
64. RULE II. Each colour-sensation should be relieved.
A composition complete in itself should afford relief to
the eye at proper intervals from every colour-sensation
which it excites, either together or separately. Such
relief may be afforded at once, as by the intervention
of black; or by several steps, as by one or more shades
intervening, or by a regular gradation. The darker
any colour is the less it requires relief.
This may be illustrated in the simplest manner by
comparing the effect of a uniform surface of some
given colour, as red, with that of a surface in which
the same colour is presented in parallel stripes, leaving
equal stripes of black between them. In the first
there is no alteration of excitement and relief; in the
second, alterations are afforded whenever the eye shifts
transversely to the stripes, as it is sure to do; and the
superiority of the latter is instantly apparent.
Such alternations of any positive colour with black
are always beautiful, whether the colour is (as nearly
as possible) the simple red, green, or blue, or a com-
bination of some two of them, as seagreen, pink, or
yellow, or of all three in pure white, or any mixture
intermediate between these; and whether the colour
contrasts directly with the black, or changes into it by
distinct steps, or by insensible degrees.
The change from any positive colour to a darker
shade of the same, or to another positive colour which
does not contain the former, must of course be attended
with some measure of relief; but in consequence of
the large mixture of all three sensations in the colours
of natural objects, this relief is far from perfect, except
in such cases as the change from one dark deep pri-
mary to another, or to its opposite secondary.
Thus, blue upon white affords relief from the red and
green in the white; yellow upon white affords relief
from the blue in the white; red upon yellow affords.
relief from the green in the yellow; dark green upon
bright green affords relief from part of the bright
green; and so on. Such combinations may have a
pleasing effect though the relief is not perfect; but if
44
two colours are very strong in hue and bright they
are sure to contain so much in common that the relief
is hardly noticeable.
It is obvious that every colour-composition which
possesses the advantage of relief must necessarily
possess the advantage of excitement with it.*
65. RULE III. Kindred colours associate well. Colours
nearly related to each other (none of them exciting any
one of the colour-sensations in a much higher degree
than the others) have always a good effect in juxta-
position.
Thus very dark colours, whatever their hue, are all
congruous with black and with each other; very light
colours again, whatever their hue, are all congruous
with white and with each other; the colours which
differ little in hue from the primary red, green, or
blue, or the secondary seagreen, pink, or yellow, are
congruous respectively with those colours, and with
each other; and the colours which differ little from
the mean gray, are congruous with it and with each
other.
The peculiar beauty of associations of kindred colours
may be illustrated by compositions of dark red, dark
green, and dark blue upon a black ground, or by
compositions of light seagreen, light pink, and light
yellow upon a white ground; or again by compositions
in which a dark blue, a seagreen-blue, and a pink-blue,
appear together; or in which a light yellow, a yellow-
red, and a yellow-green, appear together.
Ássociations of this kind are extremely common in
nature. They are seen, for instance, in the various hues
and shades of crimson or pink in roses, arising not
It must be remembered that unless a colour-composition is intended
to occupy the whole field of view for some length of time, the full ob-
servance of this and of the preceding rule is neither needful nor always
even desirable. For in such cases the eye may obtain excitement and
relief from surrounding objects and the artist is left free to work for any
effect he wishes to produce, especially if he can insure the composition
being properly accompanied.
45
only from the different colours of their parts, but from
the reflections and shades produced between the petals;
in the varieties of blue exhibited by the iris, convol-
vulus, and larkspur; or of yellow by the daffodil and
primrose. They appear in the folds of coloured fabrics;
in washes of pigments varying in thickness; in stones;
in all sorts of polished woods; in the sky and clouds
around the setting sun; and in the charming grayish
secondary hues which play on the surface of mother-
of-pearl, or on the feathers of doves.
A certain degree of congruity is secured between all
colours by the circumstance that is impossible to excite
any of the simple sensations of colour in any high
degree without a large mixture of the rest.
66. RULE IV. Light colours with dark associate well.
Colours which differ mainly in light and shade, what-
ever their hues may be, have always a good effect in
juxtaposition, especially if the quantity of unneutral-
ized light is but small in both of them.
Thus complementaries and other colours which differ
widely from each other in hue, may often be placed in
juxtaposition with great advantage, and mutually im-
prove each other, if these conditions are observed; as may
be seen with a dark deep primary and its opposite pale
clear secondary, such as french blue and lemon yellow.
The most striking contrasts between colours of different
hues in flowers are of this kind, as that between purple
and yellow in the pansy; between dark red and light
orange in the coreopsis, dark blue and lighter crimson
in some varieties of the fuchsia, and the dark and light
colours of sweet peas and tulips.
But when bright colours that are complementary to
each other, or differ widely in hue, are placed together,
they have in general a less agreeable effect. There is
sometimes a certain harshness in the association of
such colours, which is commonly called a discord; as
if the eye experienced a difficulty in accommodating
itself to the sudden change of one powerful predomi-
nant hue into another. When colours of this kind

46
come much together in a composition it is not easy to
avoid a gaudy or glaring effect.
67. RULE V. Gradations should be true and regular.
When a series of colours are so related that the eye in
passing from end to end becomes less excited by one
or more of the colour-sensations and more excited by
the rest, the eye is best pleased when the colours form
a true and regular gradation, changing uniformly, or
in equal steps. In such a case, the middle colour of
the gradation will be the mean between the terminal
colours, and each intermediate colour the mean between
the colours on either side of it.
The peculiar charm that attends a perfectly true and
regular gradation of colour is very remarkable, whether
the gradation is from some colour to its complementary,
through gray, or merely to black, or to white; or
whether it is from one hue to some collateral hue,
distant or nearly related. But, except between kin-
dred colours (of which kind some tolerably perfect
examples may be found in the vegetable and animal
world), they are rare in nature, and require some skill
to produce in pigments.
68. RULE VI. Ends of gradations should be equally
striking. Whenever a single gradation of colours
constitutes a colour-composition, or an independent
part of one, its extreme colours should be equally
striking to the eye, so as to appear to balance or
sustain each other.
To do this, it is not necessary that they should be
equal either in brightness or in strength of hue; it is
sufficient if they seem equal in depth or in clearness,
or if the depth of one should seem matched by the
clearness of the other. That the peculiar beauty of
a balanced gradation is independent of the balance
of hue referred to in Rule I., is evident from its
being found in gradations in which there is no ba-
lance of hue, such as one from a deep red to a clear
yellow or to white. That it is independent of contrast
47
with any ground colour, is proved by its appearing
whether the gradation is viewed against a perfect
black, a pure white, a medium gray, or any coloured
ground whatever. Different grounds must, however,
affect the application of this rule by inducing ocular
modifications of the colours in respect of their depth
or clearness.
When the gradation is to be regarded in conjunction
with other parts of the composition, the balance neces-
sary to satisfy the eye may be obtained from them,
and any incompleteness in itself is not necessarily
matter of objection.
It may be added that any intermediate colour in a
gradation, that is not the mean between the extreme
colours, seems also to require, though in a less degree,
the support of a corresponding colour of equal depth
and clearness in the opposite part of the gradation;
and the observance of this condition enhances the
regularity of the gradation.
69. RULE VII. Contrasting colours should be equally
striking. Two colours opposed to each other, without
a gradation between them, and independent of others,
should sustain or balance each other in depth and
clearness.
This is, in fact, but the extreme case of all those
to which the previous rule applies. It will be found
fulfilled in many of the most beautifully coloured
natural objects, as flowers and birds and insects, in
which such contrasts occur.
70. RULE VIII. Gradations, contrasts, and single
colours should correspond. There should be a corre-
spondence or equivalence between the gradations and
contrasts which occur in the different parts of the
composition. If there is only one colour which is
striking for its depth or clearness, that colour should
occupy the middle parts to which the eye is naturally
mainly directed, so that it may form balancing grada-
tions and contrasts with all the less striking colours

48
around it; or else it should surround the less striking
colours as a background to them, so as to form the
like gradations and contrasts with the less striking
colours towards the middle. At any rate it should be
disposed so as to secure in some way or other a certain
symmetry of colours in the different parts of the com-
position.
The advantage of this may be easily seen by com-
paring the effect of a composition in which this rule
is attended to with that of another of the same colours
with which it is disregarded.
Again, if there are two or more striking colours,
they should be disposed so as to balance each other
across or around the central parts of the composition,
either by one or both of the colours being repeated on
each side of the middle, or by one being on one side,
and the other on the other.
Landscapes, where the clear colours on the sky are
reflected from water in the foreground, afford approxi-
mate examples on the former variety; while landscapes,
where those clear colours are matched by the deep reds,
greens, or other hues, and dark shades of the fore-
ground, do the same for the latter.
71. RULE IX. Colours and groups of colours should
occupy appropriate breadths. There should be some
degree of concordance between the breadths of the
spaces occupied by the different colours in the com-
position, and the breadths which the eye is capable of
viewing distinctly and with satisfaction, at the distance
from which (for the most part) it is to be viewed,
without changing its direction more slowly or more
quickly than is usual in contemplating an object.
It is obvious that if the parts which are differently
coloured are much too small, they become blended
together in the eye into some indistinct average colour,
in which the harmony of the different colours is for
the most part lost: while if they are much too large,
so as to require the eye to move over any considerable
angle in passing from one to another, the beauty of
49
the composition suffers from the inability of the eye
to take in readily the different parts; whence a want
of excitement while it dwells too long on the parts
that are dark, and a want of relief while it dwells on
those that are bright.
Hence in general the breadth may be greater for
colours which do not differ very widely from the medium
gray, than for colours extremely bright or extremely
dark, or having some very powerful hue.
A similar consideration applies to the breadths of
the spaces occupied by groups of kindred colours,
which have on the whole the same sort of effect on
the eye.
72. RULE X. Variety of colour and treatment may
conduce to beauty. Variety of colour conduces to the
beauty of a composition, if the conditions of harmony
are not contravened by its introduction. It is easy to
see that increasing the number of different colours,
and thereby increasing also the number of different
kinds of gradations and contrasts, must greatly in-
crease the difficulty of assorting them all so as to
satisfy the eye with their harmony; but if this is
effected equally well, a composition with many colours
will always excel a similar one with few in beauty and
continuing charm.
To illustrate this in the simplest manner, compare
the effect of a pattern of equidistant stripes of gray
upon black, with that of a pattern in which the
equidistant stripes upon black are alternately blue and
yellow (or any other two complementary colours), or
else red, green, and blue (or any other three colours
which will neutralize each other). In each case the
three colour-sensations may be equally excited and
relieved; but in the first there is no variety of colour,
because the three sensations are everywhere combined
in a uniform gray: while in the second there are two
different positive colours, in the third three. The
beauty arising from variety is thus made very appa-
rent.
D
50
73. Conditions restrictive of rules. The application
of rules of harmony must of course be restricted by
the exigencies and proprieties of the subject of design;
its nature, use, importance, value, size, intended dis-
tance from the eye, and accompaniments; for all such
circumstances must obviously affect our choice, both
of colour and of treatment. But besides this, it is
frequently required to produce some special effect,
congenial with the circumstances of the case; and
for this end it is useful to study the mental effects
which seem to attend different classes of colour and
modes of treatment. These will be recounted in the
next chapter.
74. Necessary compromise of principles. In attempt-
ing to fulfil the laws of harmony in any case, it will
be found that a selection or compromise must be made
between them. Full conformity with any one of them
is usually impossible, much more with all. Did no
other difficulty arise, the designer is sure to be ham-
pered more or less by the imperfection of the best
material means available for producing any desired
effect.
But a very moderate compliance with sound prin-
ciples produces a sensibly good effect; and if the eye
is pleased with some good points in a composition, it
will generally be indulgent towards inevitable defi-
ciencies.
It should be remembered, moreover, that while an
acquaintance with the principles of harmony, and with
the true relations of colours to each other, is most
useful, both in designing and in judging of colour-
compositions, it does not supersede the advantage of
good natural taste and sensibility to colour, or of the
skill and facility which can only be acquired by care-
ful practice.
51
CHAPTER VII
MENTAL EFFECTS OF COLOURS AND OF
COLOUR-COMPOSITIONS.
75. Impressiveness of colour.
The deepest colours
impress the mind most with a certain idea of solemnity
and magnificence, which is often intended to be con-
veyed by such words as strength, intensity, richness,
fulness, power. This is the more remarkable because
their real brightness and real strength of hue is never
great, and may be, as in the deepest black, nothing at
all. It would seem that the eye intuitively compares
the strength of the predominant colour-sensation with
the whole brightness of the colour; and the more nearly
these approach to equality, the more impressive the
colour becomes in comparison with others of the like
hue.
Of all positive colours the deepest blue, purple,
crimson, and red, have most of this quality; but the
deepest green, seagreen-green, azure, and even the
deepest orange, yellow-green, and yellow, as in certain
rich golden browns and olive-greens, are not quite
devoid of it.*

* In observing this and other mental effects of colours, it is well to
compare all the colours with the medium gray, viewing them against a
background of that colour which is the mean or average of all colours.
52
76. Liveliness of colour. Clear colours are also
particularly striking, but in a different way. They
are best distinguished by their lively effect. Both
these and the brighter of the deep colours are often
said to be "vivid," and not without reason, since they
are apt to look brighter than they really are.
As the clearest possible colours are the perfect com-
plementaries of the deepest, it follows that yellow,
light green, seagreen, and the like, next to white,
afford the best examples of this class; yet pink,
cerise, and even the best light red and light blue are
not without a measure of it.
There is one remarkable difference between these
two classes of striking colours; the deep colours are
invariably so, and from their nature always impress
the mind as such; whilst the clear colours lose all
their excellence in the presence of some brighter
colour. Thus what appears the full white in a com-
position equally illuminated, seems obscured at once
when a sunbeam lights up another part which before
was far less brilliant.

77. Warmth of colour. The three colour-sensations
produce different effects on the mind, or, more pro-
bably, produce the same effect in different degrees.
Red is usually described as exciting, warm, advancing;
green as refreshing, soft, pleasant; and blue as quiet,
cool, retiring. In truth, perhaps, red has more of an
exhilarating or exciting effect (more of what is usually
called warmth of colour) than green, and green more
than blue.*
To compare the effects of different colours in this
respect, they should of course be reduced to an equal-
ity in brightness and strength; for the eye, as in other
cases, seems to estimate the effect in proportion to the
whole brightness of the colour. Thus, white, though
*It is worthy of notice that this may arise from the fact that red light
has the slowest vibrations, and blue the quickest; since a similar differ-
ence may be observed in musical notes of slow and quick vibrations; bass
notes being more exciting, and appearing nearer at hand, than treble
notes.
53
containing the sensation of red in a higher degree
than the colour of the strongest red pigment, does not
seem to equal it in warmth.
If the cube of colours be divided into two equal
parts by a plane passing through the corners of black,
green, white, and pink, the colours on the red and
yellow side of that plane may be regarded as warm,
the more so the nearer to red; while those on the blue
and seagreen side will be cool (that is, less warm than
the average), and the more so the nearer to blue; and
the colours in or near the dividing plane have a medium,
effect, and may be considered indifferent as to warmth
or coolness.
The combination of this warmth of hue with depth
of colour gives an attractive richness of which cool
deep colours are destitute; and its combination with
clearness of colour produces a gayness in the light
clear reds and yellows, very different from the modest
retiring liveliness of the light clear blues and sea-
greens.

78. Effects of colours in colour-compositions. Com-
positions partake of the character of their predominant
colours. When a variety of colours of like character
are brought together, the general result is to make
that character more marked: and the effect is further
enhanced when the colours are so arranged that they
mutually improve each other.
Thus a good composition of the deep primaries with
black and with some of their deep intermediates, or of
the clear secondaries with white and with some of their
clear intermediates, is more impressive, or more lively,
than any single colour could be; and the effect is more
abiding, because the variety leads the eye to contem-
plate it longer with pleasure. So also a combination
of red, orange, yellow, and other warm hues, dark and
light, produces a stronger effect in that direction than
any single colour; and the same may be said of a com-
bination of all sorts of cool retiring colours.
So also compositions of dull and nearly neutral
54
colours are uninteresting, and if the colours are dark,
even gloomy and sad, unless redeemed in some degree
by the manner in which they are put together.
On the other hand, colours widely differing in cha-
racter may form compositions extremely beautiful,
presenting admirable examples of the harmony of
colours, though neither very lively nor very impress-
ive, very warm nor very cool.
79. Effects of modes of composition. The character
of a composition is also affected by the number of
its colours and by the manner in which they are put
together.
If the colours are few, the composition in that respect
is more simple; and simplicity itself has a charm
where other causes of beauty are present; but without
them is merè poverty. If the colours are many, there
is danger lest the advantage of variety be lost in
complication, confusion, and superfluity.
Sharp contrasts tend to give strength and vigour ;
easy gradations lead to gentle and soft effects. The
danger of the former is to degenerate into harshness,
of the latter into tameness and weakness.
The one
may be considered as analogous to rectilineal figures,
having angular points or cross lines, the other to
figures formed by flowing curves.
80. Imaginary mental effects. A feeling of con-
gruity or incongruity which attends the use of colours
in certain cases seems to deserve notice here, though
perhaps it mainly rises from a mere association of
ideas.
When green is used in apparel, for instance, the
effect is generally felt to be unpleasant, the preference
being given to colours in which red and blue predomi-
nate, or at least are combined in equal strength with
green. It is true that greens generally are poorer
colours than reds and blues of the same brightness;
but this difference is insufficient to account for the
objection. It more probably arises from the felt in-
55
congruity of clothing a human being in a colour
which is so general in vegetation and so uncommon
in animated nature; and it may be noticed that
greens which differ most from those of plants (either
in depth or clearness or in tendency to seagreen) are
most preferred in dress.
There are many ideas connected with colours by
association, especially by imaginative persons, which
have nothing to do with any mental emotion really
excited by the colour itself. Thus one says that red
on black is "an atrocious combination, suggestive of
the executioner," while another may just as reasonably
ascribe the like effect to a dingy blue, the colour of a
butcher's shirt. One, again, who has his mind im-
pressed with the splendour of dark gold will call the
same colour magnificent, which another who thinks of
bile, considers a sickly yellow.
It is quite common for those who have imbibed
prejudices in favour of certain artificial combinations
of colours to be disgusted when they see something
new or unexpected, though such as have learnt to
judge according to natural taste, and the real effect
on the eye, may be delighted with the same.
Colours of any strong hue, again, have been incon-
siderately condemned in general by some persons, as
unpleasantly garish, or unbecomingly obtrusive; and
colours nearly neutral have obtained an undue prefer-
ence with many. The cause of this may be traced in
great measure to the unskilful manner in which
colours of strong hue have been applied, in ignorance
of the principles of harmony; but it may be confi-
dently asserted of every colour, without exception,
that there are circumstances in which it would not
only be beautiful, but the most beautiful of all colours,
as conducing most to the good effect of the whole
composition.

56
CHAPTER VIII.
PECULIAR AND DEFECTIVE COLOUR-VISION.
81. Dichromism. Many persons see only two dis-
tinct hues in the prismatic spectrum, even though
they are sensible to all its rays. All that part which
extends from the beginning of the red to the end of
the green band (Sect. 7) appears to them of one hue,
beginning dark and deep, and becoming brighter and
then paler, till at the verge between the green and
blue bands the light appears neutral (the same as a
mixture of all the rays). The other hue then begins
pale and light, and becomes darker and deeper to
the end, extending over all that part of the spectrum
which constitutes the blue band. There is every
reason to believe that these two hues of dichromic
vision are the same as the complementary yellow and
blue of ordinary vision.
82. Defective sensibility to red light. Defective sen-
sibility to the less refrangible or red rays is not
uncommon in persons who can distinguish red as well
as other hues, and therefore have the ordinary
trichromic vision. The existence of the defect is
shown by a portion of the red end of the spectrum
being invisible to persons who have it. It must
affect more or less in brightness and hue the colours
of all objects, except perfect black, by diminishing
the quantity of red in their composition, just as
57
AD...

they would be affected in ordinary vision, if illumin-
ated by light that has passed through a sea-green
glass. But whatever is the colour of those objects
which reflect all the kinds of light that such persons
see, they no doubt judge it to be white, as it must be
the brightest of all their colours, and as objects white
to others must appear of that colour to them. So
objects white by daylight are commonly judged to be
white by candlelight also, though really yellow, by
reason of the deficiency of blue light.
Hence objects which in ordinary vision are red,
orange, yellow, yellow-green, or green in hue, are all
yellows to such persons, differing only in strength
of hue and brightness; and those which are seagreen-
blue or pink-blue, are all blues differing only in
strength of hue or in brightness; while sea-green and
pink are both neutral grays.
83. Dichromism with defective sensibility. This di-
chromic vision is usually attended with some (and often
with a very great) defect of sensibility for the less
refrangible kinds of light, making objects that send
out chiefly those kinds of light very dark, and modify-
ing the colours of all. Dark reds, in such cases,
are hardly distinguished from black; bright reds give
the same shades of yellow as browns and olive greens;
pink appears blue, and cerise appears gray.*
84. Effects of colour-designs in such cases. The
principles of the ocular modifications of colours, of
their harmony, and mental effects, are of course the
same in dichromic and defective, as in ordinary colour-
vision; but their application is obviously simplified
when only two hues are distinguished.
The only compositions that can appear alike in
ordinary and dichromic vision, are those containing no
other colour but whites, grays, and blacks, with
yellows and blues in all their varieties of brightness
and of strength of hue; the colours belonging to the
plane which passes through the corners of black, blue,
* Other peculiarities of colour-vision besides these are sometimes
found, but rarely.
58
white, and yellow in the colour-cube. Compositions
of green and blue, with their intermediate hues of
seagreen, will appear in dichromic vision as equally
regular compositions of yellow and blue and their
intermediates. The same may be said of compositions
of blue and red, with the intermediate crimsons,
pinks, and mauves, when there is no defect of sensi-
bility to the less refrangible rays.
Compositions on the other hand in which reds,
greens, crimsons, or seagreens prevail, however beau-
tiful to the ordinary eye, will be generally poor and
deficient in variety to persons of dichromic vision,
nor will it be possible to them to see any reason in
the assortment of the colours adopted.
As a general rule therefore to make colour-composi-
tions pleasing to persons of dichromic and defective
colour-vision as well as to others, they should contain
only the colours which are the same to each; and if
bright red, orange, yellow-green, or green, is put in,
it should only be where a yellow of corresponding
strength would have a good effect; if seagreen-blue,
pink, or mauve is put in, it should be where a blue
of corresponding strength would look well; and if
crimson or seagreen is put in, it should be where a
gray of proper darkness might be substituted without
objection; and dark red where black might take its
place.
In many cases a system of colouring may well be
adopted which would be acceptable to persons who
have these very common peculiarities of vision, with-
out displeasing others.
..
:
>55

UNIVERSITY OF MICHIGAN
3 9015 00059 0979
Filmed by Preservation 1987
DO NOT REMOVE
OR
MUTILATE CARD

Sats
મો.
૧૨