[Footnote: Lecture delivered by Capt. W. De W. Abney, R.E., P.B.S., at
the Royal Institution, on February 25, 1887.--_Nature_.]
By Capt. W. DE W. ABNEY.
Sunlight is so intimately woven up with our physical enjoyment of life
that it is perhaps not the most uninteresting subject that can be chosen
for what is--perhaps somewhat pedantically--termed a Friday evening
"discourse." Now, no discourse ought to be be possible without a text on
which to hang one's words, and I think I found a suitable one when
walking with an artist friend from South Kensington Museum the other day.
The sun appeared like a red disk through one of those fogs which the east
wind had brought, and I happened to point it out to him. He looked, and
said, "Why is it that the sun appears so red?" Being near the railway
station, whither he was bound, I had no time to enter into the subject,
but said if he would come to the Royal Institution this evening I would
endeavor to explain the matter. I am going to redeem that promise, and to
devote at all events a portion of the time allotted to me in answering
the question why the sun appears red in a fog. I must first of all appeal
to what every one who frequents this theater is so accustomed, viz., the
spectrum. I am going not to put it in the large and splendid stripe of
the most gorgeous colors before you, with which you are so well
acquainted, but my spectrum will take a more modest form of purer colors,
some twelve inches in length.
I would ask you to notice which color is most luminous. I think that no
one will dispute that in the yellow we have the most intense luminosity,
and that it fades gradually in the red on the one side and in the violet
on the other. This, then, may be called a qualitative estimate of
relative brightnesses; but I wish now to introduce to you what was novel
last year, a quantitative method of measuring the brightness of any part.
Before doing this I must show you the diagram of the apparatus which I
shall employ in some of my experiments.
[Illustration: FIG. 1.--COLOR PHOTOMETER.]
RR are rays (Fig. I) coming from the arc light, or, if we were using
sunlight, from a heliostat, and a solar image is formed by a lens, L_{1},
on the slit, S_{1} of the collimator, C. The parallel rays produced by
the lens, L_{2}, are partially refracted and partially reflected. The
former pass through the prisms, P_{1}P_{2}, and are focused to form a
spectrum by a lens, L_{3}, on D, a movable ground glass screen. The rays
are collected by a lens, L_{4}, tilted at an angle as shown, to form a
white image of the near surface of the second prism on F.
Passing a card with a narrow slit, S_{2}, cut in it in front of the
spectrum, any color which I may require can be isolated. The consequence
is that, instead of the white patch upon the screen, I have a colored
patch, the color of which I can alter to any hue lying between the red
and the violet. Thus, then, we are able to get a real patch of very
approximately homogeneous light to work with, and it is with these
patches of color that I shall have to deal. Is there any way of measuring
the brightness of these patches? was a question asked by General Festing
and myself. After trying various plans, we hit upon the method I shall
now show you, and if any one works with it he must become fascinated with
it on account of its almost childish simplicity--a simplicity, I may
remark, which it took us some months to find out. Placing a rod before
the screen, it casts a black shadow surrounded with a colored background.
Now I may cast another shadow from a candle or an incandescence lamp, and
the two shadows are illuminated, one by the light of the colored patch
and the other by the light from an incandescence lamp which I am using
tonight. [Shown.] Now one stripe is evidently too dark. By an arrangement
which I have of altering the resistance interposed between the battery
and the lamp, I can diminish or increase the light from the lamp, first
making the shadow it illuminates too light and then too dark compared
with the other shadow, which is illuminated by the colored light.
Evidently there is some position in which the shadows are equally
luminous. When that point is reached, I can read off the current which is
passing through the lamp, and having previously standardized it for each
increment of current, I know what amount of light is given out. This
value of the incandescence lamp I can use as an ordinate to a curve, the
scale number which marks the position of the color in the spectrum being
the abscissa. This can be done for each part of the spectrum, and so a
complete curve can be constructed, which we call the illumination curve
of the spectrum of the light under consideration.
Now, when we are working in the laboratory with a steady light, we may be
at ease with this method, but when we come to working with light such as
the sun, in which there may be constant variation, owing to passing, and
may be usually imperceptible, mist, we are met with a difficulty; and in
order to avoid this, General Festing and myself substituted another
method, which I will now show you. We made the comparison light part of
the light we were measuring. Light which enters the collimating lens
partly passes through the prisms and is partly reflected from the first
surface of the prism; that we utilize, thus giving a second shadow. The
reflected rays from P_{1} fall on G, a silver on glass mirror. They are
collected by L_{5}, and form a white image of the prism also at F.
The method we can adopt of altering the intensity of the comparison light
is by means of rotating sectors, which can be opened or closed at will,
and the two shadows thus made equally luminous. [Shown.] But although
this is an excellent plan for some purposes, we have found it better to
adopt a different method. You will recollect that the brightest part of
the spectrum is in the yellow, and that it falls off in brightness on
each side, so instead of opening and closing the sectors, they are set at
fixed intervals, and the slit is moved in front of the spectrum, just
making the shadow cast by the reflected beam too dark or too light, and
oscillating between the two till equality is discovered. The scale number
is then noted, and the curve constructed as before. It must be remembered
that, on each side of the yellow, equality can be established.
This method of securing a comparison light is very much better for sun
work than any other, as any variation in the light whose spectrum is to
be measured affects the comparison light in the same degree. Thus,
suppose I interpose an artificial cloud before the slit of the
spectroscope, having adjusted the two shadows, it will be seen that the
passage of steam in front of the slit does not alter the relative
intensities; but this result must be received with caution. [The lecturer
then proceeded to point out the contrast colors that the shadow of the
rod illuminated by white light assumed.]
I must now make a digression. It must not be assumed that every one has
the same sense of color, otherwise there would be no color blindness.
Part of the researches of General Festing and myself have been on the
subject of color blindness, and these I must briefly refer to. We test
all who come by making them match the luminosity of colors with white
light, as I have now shown you. And as a color blind person has only two
fundamental color perceptions instead of three, his matching of
luminosities is even more accurate than is that made by those whose eyes
are normal or nearly normal. It is curious to note how many people are
more or less deficient in color perception. Some have remarked that it is
impossible that they were color blind and would not believe it, and
sometimes we have been staggered at first with the remarkable manner in
which they recognized color to which they ultimately proved deficient in
perception. For instance, one gentleman when I asked him the name of a
red color patch, said it was sunset color. He then named green and blue
correctly, but when I reverted to the red patch he said green.
On testing further, he proved totally deficient in the color perception
of red, and with a brilliant red patch he matched almost a black shadow.
The diagram shows you the relative perceptions in the spectrum of this
gentleman and myself. There are others who only see three-quarters,
others half, and others a quarter the amount of red that we see, while
some see none. Others see less green and others less violet, but I have
met with no one that can see more than myself or General Festing, whose
color perceptions are almost identical. Hence we have called our curve of
illumination the "normal curve."
We have tested several eminent artists in this manner, and about one half
of the number have been proved to see only three quarters of the amount
of red which we see. It might be thought that this would vitiate their
powers of matching color, but it is not so. They paint what they see; and
although they see less red in a subject, they see the same deficiency in
their pigments; hence they are correct. If totally deficient, the case of
course would be different.
Let us carry our experiments a step further, and see what effect what is
known as a turbid medium has upon the illuminating value of different
parts of the spectrum. I have here water which has been rendered turbid
in a very simple manner. In it has been very cautiously dropped an
alcoholic solution of mastic. Now mastic is practically insoluble in
water, and directly the alcoholic solution comes in contact with the
water it separates out in very fine particles, which, from their very
fineness, remain suspended in the water. I propose now to make an
experiment with this turbid water.
I place a glass cell containing water in front of the slit, and on the
screen I throw a patch of blue light. I now change it for turbid water in
a cell. This thickness much dims the blue; with a still greater thickness
the blue has almost gone. If I measure the intensity of the light at each
operation, I shall find that it diminishes according to a certain law,
which is of the same nature as the law of absorption. For instance, if
one inch diminishes the light one half, the next will diminish it half of
that again, the next half of that again, while the fourth inch will cause
a final diminution of the total light of one sixteenth. If the first inch
allows only one quarter of the light, the next will only allow one
sixteenth, and the fourth inch will only permit 1/256 part to pass.
Let us, however, take a red patch of light and examine it in the same
way. We shall find that, when the greater thickness of the turbid medium
we used when examining the blue patch of light is placed in front of the
slit, much more of this light is allowed to pass than of the blue. If we
measure the light, we shall find that the same law holds good as before,
but that the proportion which passes is invariably greater with the red
than the blue. The question then presents itself: Is there any connection
between the amounts of the red and the blue which pass?
Lord Rayleigh, some years ago, made a theoretical investigation of the
subject. But, as far as I am aware, no definite experimental proof of the
truth of the theory was made till it was tested last year by General
Festing and myself. His law was that for any ray, and through the same
thickness, the light transmitted varied inversely as the fourth power of
the wave length. The wave length 6,000 lies in the red, and the wave
length 4,000 in the violet. Now 6,000 is to 4,000 as 3 to 2, and the
fourth powers of these wave lengths are as 81 to 16, or as about 5 to 1.
If, then, the four inches of our turbid medium allowed three quarters of
this particular red ray to be transmitted, they would only allow (¾)^{5},
or rather less than one fourth, of the blue ray to pass.
Now, this law is not like the law of absorption for ordinary absorbing
media, such as colored glass for instance, because here we have an
increased loss of light running from the red to the blue, and it matters
not how the medium is made turbid, whether by varnish, suspended sulphur,
or what not. It holds in every case, so long as the particles which make
the medium turbid are small enough. And please to recollect that it
matters not in the least whether the medium which is rendered turbid is
solid, liquid, or air. Sulphur is yellow in mass, and mastic varnish is
nearly white, while tobacco smoke when condensed is black, and very
minute particles of water are colorless; it matters not what the color
is, the loss of light is _always_ the same. The result is simply due to
the scattering of light by fine particles, such particles being small in
dimensions compared with a wave of light. Now, in this trough is
suspended 1/1000 of a cubic inch of mastic varnish, and the water in it
measures about 100 cubic inches, or is 100,000 times more in bulk than
the varnish. Under a microscope of ordinary power it is impossible to
distinguish any particles of varnish; it looks like a homogeneous fluid,
though we know that mastic will not dissolve in water.
Now a wave length in the red is about 1/40000 of an inch, and a little
calculation will show that these particles are well within the necessary
limits. Prof. Tyndall has delighted audiences here with an exposition of
the effect of the scattering of light by small particles in the formation
of artificial skies, and it would be superfluous for me to enter more
into that. Suffice it to say that when particles are small enough to form
the artificial blue sky, they are fully small enough to obey the above
law, and that even larger particles will suffice. We may sum up by saying
that very fine particles scatter more blue light than red light, and that
consequently more red light than blue light passes through a turbid
medium, and that the rays obey the law prescribed by theory.
I will exemplify this once more by using the whole spectrum and placing
this cell, which contains hyposulphite of soda in solution in water, in
front of the slit. By dropping in hydrochloric acid, the sulphur
separates out in minute particles; and you will see that, as the
particles increase in number, the violet, blue, green, and yellow
disappear one by one and only red is left, and finally the red disappears
itself.
Now let me revert to the question why the sun is red at sunset. Those who
are lovers of landscape will have often seen on some bright summer's day
that the most beautiful effects are those in which the distance is almost
of a match to the sky. Distant hills, which when viewed close to are
green or brown, when seen some five or ten miles away appear of a
delicate and delicious, almost of a cobalt, blue color. Now, what is the
cause of this change in color? It is simply that we have a sky formed
between us and the distant ranges, the mere outline of which looms
through it. The shadows are softened so as almost to leave no trace, and
we have what artists call an atmospheric effect. If we go into another
climate, such as Egypt or among the high Alps, we usually lose this
effect. Distant mountains stand out crisp with black shadows, and the
want of atmosphere is much felt. [Photographs showing these differences
were shown.] Let us ask to what this is due. In such climates as England
there is always a certain amount of moisture present in the atmosphere,
and this moisture may be present as very minute particles of water--so
minute indeed that they will sink down in an atmosphere of normal
density--or as vapor. When present as vapor the air is much more
transparent, and it is a common expression to use, that when distant
hills look "so close" rain may be expected shortly to follow, since the
water is present in a state to precipitate in larger particles. But when
present as small particles of water the hills look very distant, owing to
what we may call the haze between us and them. In recent weeks every one
has been able to see very multiplied effects of such haze. The ends of
long streets, for instance, have been scarcely visible, though the sun
may have been shining, and at night the long vistas of gas lamps have
shown light having an increasing redness as they became more distant.
Every one admits the presence of mist on these occasions, and this mist
must be merely a collection of intangible and very minute particles of
suspended water. In a distant landscape we have simply the same or a
smaller quantity of street mist occupying, instead of perhaps 1,000
yards, ten times that distance. Now I would ask, What effect would such a
mist have upon the light of the sun which shone through it?
It is not in the bounds of present possibility to get outside our
atmosphere and measure by the plan I have described to you the different
illuminating values of the different rays, but this we can do: First, we
can measure these values at different altitudes of the sun, and this
means measuring the effect on each ray after passing through different
thicknesses of the atmosphere, either at different times of day or at
different times of the year, about the same hour. Second, by taking the
instrument up to some such elevation as that to which Langley took his
bolometer at Mount Whitney, and so to leave the densest part of the
atmosphere below us.
[Illustration: FIG. 2.--RELATIVE LUMINOSITIES.]
Now, I have adopted both these plans. For more than a year I have taken
measurements of sunlight in my laboratory at South Kensington, and I have
also taken the instrument up to 8,000 feet high in the Alps, and made
observations _there_, and with a result which is satisfactory in that
both sets of observations show that the law which holds with artificially
turbid media is under ordinary circumstances obeyed by sunlight in
passing through our air: which is, you will remember, that more of the
red is transmitted than of the violet, the amount of each depending on
the wave length. The luminosity of the spectrum observed at the Riffel I
have used as my standard luminosity, and compared all others with it. The
result for four days you see in the diagram.
I have diagrammatically shown the amount of different colors which
penetrated on the same days, taking the Riffel as ten. It will be seen
that on December 23 we have really very little violet and less than half
the green, although we have four fifths of the red.
The next diagram before you shows the minimum loss of light which I have
observed for different air thicknesses. On the top we have the calculated
intensities of the different rays outside our atmosphere. Thus we have
that through one atmosphere, and two, three, and four. And you will see
what enormous absorption there is in the blue end at four atmospheres.
The areas of these curves, which give the total luminosity of the light,
are 761, 662, 577, 503, and 439; and if observed as astronomers observe
the absorption of light, by means of stellar observations, they would
have had the values, 761, 664, 578, 504, and 439--a very close
approximation one to the other.
Next notice in the diagram that the top of the curve gradually inclines
to go to the red end of the spectrum as you get the light transmitted
through more and more air, and I should like to show you that this is the
case in a laboratory experiment. Taking a slide with a wide and long slot
in it, a portion is occupied by a right angled prism, one of the angles
of 45° being toward the center of the slot. By sliding this prism in
front of the spectrum I can deflect outward any portion of the spectrum I
like, and by a mirror can reflect it through a second lens, forming a
patch of light on the screen overlapping the patch of light formed by the
undeflected rays. If the two patches be exactly equal, white light is
formed. Now, by placing a rod as before in front of the patch, I have two
colored stripes in a white field, and though the background remains of
the same intensity of white, the intensities of the two stripes can be
altered by moving the right angled prism through the spectrum. The two
stripes are now apparently equally luminous, and I see the point of
equality is where the edge of the right angled prism is in the green.
Placing a narrow cell filled with our turbid medium in front of the slit,
I find that the equality is disturbed, and I have to allow more of the
yellow to come into the patch formed by the blue end of the spectrum, and
consequently less of it in the red end. I again establish equality.
Placing a thicker cell in front, equality is again disturbed, and I have
to have less yellow still in the red half, and more in the blue half. I
now remove the cell, and the inequality of luminosity is still more
glaring. This shows, then, that the rays of maximum luminosity must
travel toward the red as the thickness of the turbid medium is increased.
The observations at 8,000 feet, here recorded, were taken on September
15, at noon, and of course in latitude 46° the sun could not be overhead,
but had to traverse what would be almost exactly equivalent to the
atmosphere at sea level. It is much nearer the calculated intensity for
no atmosphere intervening than it is for one atmosphere. The explanation
of this is easy. The air is denser at sea level than at 8,000 feet up,
and the lower stratum is more likely to hold small water particles or
dust in suspension than is the higher.
[Illustration: FIG. 3.--PROPORTIONS OF TRANSMITTED COLORS.]
For, however small the particles may be, they will have a greater
tendency to sink in a rare air than in a denser one, and less water vapor
can be held per cubic foot. Looking, then, from my laboratory at South
Kensington, we have to look through a proportionately larger quantity of
suspended particles than we have at a high altitude when the air
thicknesses are the same. And consequently the absorption is
proportionately greater at sea level that at 8,000 feet high. This leads
us to the fact that the real intensity of illumination of the different
rays outside the atmosphere is greater than it is calculated from
observations near sea level. Prof. Langley, in this theater, in a
remarkable and interesting lecture, in which he described his journey up
Mount Whitney to about 12,000 feet, told us that the sun was really blue
outside our atmosphere, and at first blush the amount of extra blue which
he deduced to be present in it would, he thought, make it so. But though
he surmised the result from experiments made with rotating disks of
colored paper, he did not, I think, try the method of using pure colors,
and consequently, I believe, slightly exaggerated the blueness which
would result.
I have taken Prof. Langley's calculations of the increase of intensity
for the different rays, which I may say do not quite agree with mine, and
I have prepared a mask which I can place in the spectrum, giving the
different proportions of each ray as calculated by him, and this when
placed in front of the spectrum will show you that the real color of
sunlight outside the atmosphere, as calculated by Langley, can scarcely
be called bluish. Alongside I place a patch of light which is very
closely the color of sunlight on a July day at noon in England. This
comparison will enable you to gauge the blueness, and you will see that
it is not very blue, and, in fact, not bluer perceptibly than that we
have at the Riffel, the color of the sunlight at which place I show in a
similar way. I have also prepared some screens to show you the value of
sunlight after passing through five and ten atmospheres. On an ordinary
clear day you will see what a yellowness there is in the color. It seems
that after a certain amount of blue is present in white light, the
addition of more makes but little difference in the tint. But these last
patches show that the light which passes through the atmosphere when it
is feebly charged with particles does not induce the red of the sun as
seen through a fog. It only requires more suspended particles in any
thickness to induce it.
In observations made at the Riffel, and at 14,000 feet, I have found that
it is possible to see far into the ultra-violet, and to distinguish and
measure lines in the sun's spectrum which can ordinarily only be seen by
the aid of a fluorescent eye piece or by means of photography.
Circumstantial evidence tends to show that the burning of the skin, which
always takes place in these high altitudes in sunlight, is due to the
great increase in the ultra-violet rays. It may be remarked that the same
kind of burning is effected by the electric arc light, which is known to
be very rich in these rays.
Again, to use a homely phrase, "You cannot eat your cake and have it."
You cannot have a large quantity of blue rays present in your direct
sunlight and have a luminous blue sky. The latter must always be light
scattered from the former. Now, in the high Alps you have, on clear day,
a deep blue-black sky, very different indeed from the blue sky of Italy
or of England; and as it is the sky which is the chief agent in lighting
up the shadows, not only in those regions do we have dark shadows on
account of no intervening--what I will call--mist, but because the sky
itself is so little luminous. In an artistic point of view this is
important. The warmth of an English landscape in sunlight is due to the
highest lights being yellowish, and to the shadows being bluish from the
sky light illuminating them. In the high Alps the high lights are colder,
being bluer, and the shadows are dark, and chiefly illuminated by
reflected direct sunlight. Those who have traveled abroad will know what
the effect is. A painting in the Alps, at any high elevation, is rarely
pleasing, although it may be true to nature. It looks cold, and somewhat
harsh and blue.
In London we are often favored with easterly winds, and these, unpleasant
in other ways, are also destructive of that portion of the sunlight which
is the most chemically active on living organisms. The sunlight
composition of a July day may, by the prevalence of an easterly wind, be
reduced to that of a November day, as I have proved by actual
measurement. In this case it is not the water particles which act as
scatterers, but the carbon particles from the smoke.
Knowing, then, the cause of the change in the color of sunlight, we can
make an artificial sunset, in which we have an imitation light passing
through increasing thicknesses of air largely charged with water
particles. [The image of a circular diaphragm placed in front of the
electric light was thrown on the screen in imitation of the sun, and a
cell containing hyposulphite of soda placed in the beam. Hydrochloric
acid was then added; as the fine particles of sulphur were formed, the
disk of light assumed a yellow tint, and as the decomposition of the
hyposulphite progressed, it assumed an orange and finally a deep red
tint.] With this experiment I terminate my lecture, hoping that in some
degree I have answered the question I propounded at the outset--why the
sun is red when seen through a fog.
* * * * *