Spectrum analysis : six lectures, delivered in 1868, before the Society of Apothecaries of London

Preface

In publishing the following Lectures I have endeavoured to preserve the elementary character which they naturally assumed in delivery, thinking it best to give further detail in a series of Appendices. If the book thus assumes less of the character of a complete treatise than might be desirable, it gains in value for the general reader, inasmuch as the science of Spectrum Analysis is at present in such a rapid state of growth that much of the subject is incomplete, and, therefore, necessarily unsuited to the public at large. I hope, however, that the addition of many extracts from the most important Memoirs on the subject may prove interesting to all, as it will certainlv be useful to those specially engaged in scientific inquiry, as indicating the habits of exact research and accurate observation by which alone such striking results have been attained. For the permission to reproduce exact copies of Kirclih off’s, Angstrom’s, and Huggins’ maps, together with the Tables of the positions of the dark solar and bright vx

PREFACE.

metallic lines, 1 have to thank the above-named gentlemen. These maps will render the work valuable to the student for a reference, whilst the chromolithographic plates of the spectra of the metals of the alkalies and alkaline earths, and of the spectra of the stars, nebulae, and non-mctallic elements, serve to give some idea of the peculiar beauty of the real phenomena thus represented.

Since last summer, when these Lectures were delivered, our knowledge of the constitution of the sun especially has made giant strides, and although I have been unable to introduce these newest facts into the text of the Lectures, I have still brought forward the most important of these discoveries in the Appendices to Lecture V.

As the latest news on this subject, I may mention the arrangement contrived by Mr. Huggins, by which the wonderful changes of the red solar prominences can all be viewed at once ; changes so enormously rapid that Mr. Lockyer has observed one of these red solar flames, 27,000 miles in length, disappear altogether in less than ten minutes. Mr. Lockyer has also succeeded in seeing in the flames the red (c) line of hydrogen, as well as the line in the violet, which

PREFACE.

vn he finds corresponds to the line marked 2796 on Kirchhoff’s map, and not, as was supposed, identical with Fraunhofer’s line g.

At the end of the volume will be found a tolerably complete List of Memoirs forming the literature on the subject.

My thanks are due to Mr. J. 1). Cooper and Mr. Codings for the very great care which they have bestowed upon the illustrations, and especially' upon the difficult task of reproducing Kirclihoff’s maps in four tints.

IL E. R.

Manchester, April 18(5!).

Contents

LECTURE I. — Apothecaries’ Hali,, May 2, 1868.

Pago

INTRODUCTION. — NEWTON’S DISCOVERY OP THE COMPOSITION OF WHITE LIGHT, 1675. — PROPERTIES OF SUN LIGHT. — HEATING RAYS.' — LUMI- NOUS RAYS. — CHEMICALLY ACTIVE RAYS. — THE SOLAR SPECTRUM. — POSITION OF MAXIMA. — ILLUSTRATIONS OF THESE RADIATIONS. — MEANS OF OBTAINING A PURE SPECTRUM. — FRAUNHOFER’S LINES. — PLANET AND MOON LIGHT. — STAR LIGHT 1

Introduction

alone, and require further investigation to bring them into harmony with the rest. The advance in these new fields of research is, however, so rapid that, as time rolls on, and our range of knowledge widens, new facts quickly come to support the hitherto unexplained phenomena, and thus our theory becomes more and more complete. It will be my duty, in the Lectures which I have the honour of delivering in this Hall, to endeavour to explain to you that these results, apparently as marvellous as the discovery of the elixir vitae or philosopher’s stone, are the plain and necessary deductions from exact and laborious experiment, and to show how two German philosophers, quietly working in their laboratories in Heidelberg, obtained this startling insight into the processes of creation.

The only means of communication which we possess with the sun, planets, or far distant stars, or by which we can ascertain anything respecting their chemical constitution, is by means of the life-supporting radiation which they pour down upon the earth, producing the effects which we call light and heat. It will, therefore, be our business, in the first place, to investigate the composition of the radiations which these bodies give off, and next gradually to notice, as our field of observation enlarges, the applications to which the properties of the light thus emitted lead us. One cannot help regretting that when, as at present, the sunlight is shining so brightly, we are unable to utilize it, and illustrate by experiments made with the sunlight itself the points which we wish to explain. In this climate, however, even if we could conveniently do it, the sun shines so intermittently, and it is so doubtful if we can have it just when it is required, that we have to make

b 2

SPECTRUM ANALYSIS.

[lect. I.

use of other means, especially of this bright light of the electric arc, which, if less perfect than sunlight, is more under our complete control.

In the year 1675, Sir Isaac Newton presented to the Royal Society his memorable treatise on Optics.1 In this treatise, which contains a large number of experimental and theoretical investigations, one point especially attracts our attention : it is the discovery of the decomposition of white light. We have here (Fig. 1) a fac-simile of the drawing illustrating Newton’s experiment on this subject. He heads the first paragraph in his memoir, written in the year 1675, with the words, “lights which differ in colour differ also in refrangibility.” Newton allowed the sun to shine through a round hole (Fig. 1, f) in a shutter, and he then examined the character of this light by means of a triangular piece of glass (a b c) called a prism. He found that the white light, after passing through the prism, was bent or refracted out of its course, and split up into a coloured band (pt) which, when received on the white screen (mn), exhibited all the colours of the rainbow in regular succession, passing from red through all the shades of orange, yellow, 1 For extracts from “ Newton’s Opticks,” see Appendix A.

lect. i.] NEWTON'S DISCOVERY. 5

green and blue, to violet. Newton termed this coloured band the Solar Spectrum, and came to the conclusion that the light of the sun consists of rays of different degrees of refrangibility. He also showed that all the various portions of this coloured band, when again brought together, produce upon the eye the effect of white light. This experiment (represented in Fig. 2), Newton performed by simply allowing the light passing through the round hole (f) in the shutter to fall on a prism (a b c), producing the solar spectrum, and then, on looking at this coloured band through another prism (a b c) placed in the same direction, instead of seeing a coloured band he observed a spot of white light, thus showing that the whole of these differently coloured rays, when brought together by means of the second prism, produce on the eye the effect of white light. Here we have the intensely white electric light, and by means of these two prisms you observe that I can split the light up into its various constituent parts, and we obtain this splendid coloured band, the spectrum of the electric arc. Now I shall endeavour to show you the

G

SPECTRUM ANALYSIS.

[lect. I.

second effect wliich Newton observed. For this purpose I have only to reverse the position of one of the prisms, for if I allow this band of coloured light emitted from the first prism to pass through the second prism, placed in the opposite direction to the first, I shall again brine these coloured rays together, the second prism neutralizing altogether the effects of the first, and we obtain the bright white image of the slit.

I should like next to show you a third part of the experiment made by Newton. If I cut off, by means of this screen, a portion of the spectrum which has passed through the first prism, say the yellow or the green or the red ; you will see that another refraction through the second prism does not alter this coloured light. Here, for instance, I take the green, and bring my prism into a proper position: we shall find that I only get the same green light on the screen behind, proving that the green ray cannot again be split up by further refraction. The mode in which Newton performed this experiment is seen in Fig. 3. The green rays (g) in the spectrum (cle), when refracted through a second prism ( abc ), appear as a green band (n m) on the second screen.

Now the reason why all these different coloured lights, when they reach the eye, produce the effect of white lect. i.] COMPOSITION OF WHITE LIGHT. 7

light, is a physiological question which we cannot explain. We find that not only do all the colours of the spectrum, when they are brought upon the eye, produce the effect of white light, but that several mixtures of only a few out of all these different colours have the same power of producing upon the eye the effect of white light.

Thus Helmholtz and Maxwell have shown that the

Fig. 4.

following mixtures of two complementary colours when brought together into the eye produce the effect of whiteness : — violet and greenish yellow ; indigo and yellow ; blue and orange ; greenish blue arid red. This I may readily illustrate to you by taking a revolving disc (Fig. 4), upon which segments of various colours have been painted. When I turn this disc, which you observe contains to begin with all the colours of the spectrum in due proportion, and then, when it is

SPECTRUM ANALYSIS.

[lect. I.

revolving, exhibit it to you by the light of burning magnesium, you will notice that the effect produced upon the eye is that of a uniformly white disc. Let me now substitute for this painted disc another one, which only contains the three colours, red, yellow, and blue. You will still perceive that when quickly revolved the disc appears white; and so I might substitute any of the above mixtures of colours, all of which produce on the eye the sensation of white light by the rapid succession of the images of the different colours, the effect of one of which not having time to disappear from the retina before that of the other comes into play. If I illuminate the rotating disc by means of a quick electric spark, you will see that all the colours of the disc become noticeable at once. This is because the period of illumination is so exceedingly short that the disc appears as if stationary. It was indeed at one time supposed that the various shades of colour in the solar spectrum were produced by an overlapping, as it were, of three distinct coloured spectra, one red, the second yellow, and the third a blue spectrum, the maxima of which are situated at different points, that of the red and blue at the extremes and that of the yellow in the middle of the visible spectrum. This theory of Brewster’s lias, however, been proved to be fallacious, for Helmholtz has shown that the green ray, for example, is not made up of blue and yellow light superposed, and we cannot separate anything else but green out of it. Hence we conclude that each particular ray has its own peculiar colour, and that light of each degree of refrangibility is monochromatic. But, on the other hand, although physically, and in the actual spectrum, there is no such thing as a superposition, or overlapping of different

. THEORY OF VISION.

spectra, yet it is very likely, nay, more than likely, that the retina is mainly sensitive to three impressions, viz. red, yellow, blue ; in fact, that there are some nerves especially sensitive to red, others to yellow, and others again to blue light, whilst the impressions of the other tints are obtained by the joint impressions produced on these three classes of nerves. This theory, indeed, was one which was proposed so long ago as the beginning of the century by our celebrated countryman Thomas Young, and quite recently it has been proved by Max Schulze that in the eyes, not indeed as yet of man, but of certain animals, there exist differences which are observable in the nerve ends situated at the back of the retina : some of these end in little red drops, some of them in yellow drops, and some of them in colourless drops. The nerves whose ends contain the little red drops are more sensitive to red colour than the others ; and so those containing yellow drops are more sensitive to the yellow colour ; and in this way we believe that the peculiar effects which we observe in the mixtures of colour may be explained.

In noticing the physical properties of the variously coloured light obtained when the sunlight or white light is decomposed into its constituent parts, we must remember, in the first place, that light is due to the undulations of the elastic medium pervading all space, to which physicists have given the name of luminiferous ether. As the undulations of the particles of water, causing the waves of the sea, differ in length and intensity of vibration, sometimes being minute and shallow like the ripples on the surface of a pond, sometimes rising and falling into the gigantic crests and valleys of the storm-ridden ocean ; so also the undulations

SPECTRUM ANALYSIS.

[lect. X.

of the ether producing light differ in amplitude and intensity, giving rise to the different effects of colour to which we have referred.

Let us now compare the power of the ear and the eye, the one to receive the vibrations of the air called sound, and the other the vibrations of the ether termed lio-ht.

o

The human ear has the power of distinguishing variations in sounds differing widely in wave-length. Thus, for instance, the musical note, the deepest in the bass of all those that can be heard by the human ear, is produced by the regular succession of impulses occurring about 16 times in each second, whereas the highest note which is perceptible to the ear is caused by about 4,000 vibrations per second. Hence the range of audible notes extends over about 11 octaves. Now let us examine what is the range to which the eye is sensitive. It is found that we can observe, ordinarily, in the solar spectrum, from the position indicated by a fixed dark line in the red portion termed a (see Frontispiece of the Solar Spectrum) to a line h in the violet or most refrangible portion of the visible spectrum. Now the difference in the length of the vibration from a to h is but slight ; it is indeed not one octave. The length of the undulation of each of these waves is excessively small. By certain methods of exact measurement, physicists have been enabled to determine with accuracy the length of these various waves of light ; and they have come to the conclusion that the wave causing this red ray which is only just visible to the eye has a length of the TWffoVouth part of an inch, whereas at the line h in the violet the length of the wave is TusVwo-iffh part : hence the difference is only from 155 to 271, and you see that the wave-length at the one point is not

LKCT. I.]

HEATING RAYS.

twice as great as it is at the other, and we are correct in stating that we can hear about 11 octaves, but that we cannot see a single octave.

Nevertheless, although they do not produce on the retina the impression which we call light, there are rays extending both beyond the visible red and beyond the visible blue. By certain devices we can make ourselves aware of the existence of these invisible rays, which play a most important part in the nature of the solar light.

If we observe the effects which are produced by the different rays constituting the visible portion of the solar spectrum, we find to begin with that those rays which mainly produce heating effects are situated at the red end of the solar spectrum. We learn that the maximum heating effect is produced at a point beyond that at which we can see any red light. The maximum of the luminous rays as affecting the eye exists in the yellow. Passing on from the red towards the violet portion, we find, by means of a thenno-pile and a delicate galvanometer, that the quantity of heating effect produced in the yellow and green portions of the spectrum gradually diminishes, and sinks down to a very insignificant amount in the violet part of the band. In the blue and violet portion of the spectrum, so slightly endowed with heating power, we have, however, to notice the existence of a new and striking peculiarity, that of producing chemical action ; that is, of causing the combination and decomposition of certain chemical substances, as for instance the decomposition and blackening of silver salts, upon which the art of photography is based.

It is to Sir W. Herschel, in the year 1800, that we are indebted for the first notice of the fact of the heating

SPECTRUM ANALYSIS.

[lect. I.

rays existing especially in the red portion of the spectrum, whilst Ritter and Scheele at the early part of this century observed the peculiar power possessed by the blue, violet, and ultra-violet rays to blacken silver salts.

In Fig. 5 we have a graphical representation of the varying intensity of the heating, luminous, and chemically active rays in the various parts of the solar spectrum. The figure exhibits three curves, a, b, and c, showing the distribution of these three actions produced by the rays of the solar spectrum, whose position is given in the upper part of the diagram.

Fio. 5.

This curve a (Fig. 5) represents to you the intensity of the heating power of the spectrum. You observe how far it extends, a long way beyond the visible red, which ends a little to the left of the line A. In fact it has been noticed, by those who have made careful measurements, that half the rays of heat reaching the earth from the sun are invisible, and I shall hope to show you directly the effect of these invisible heating rays.

In the part shaded with vertical lines only, there is no perceptible luminous or chemical activity ; in that shaded with horizontal lines there is nothing but chemical action.

LECT. T.]

CALORESCENCE.

From the beginning to the end of the luminous spectrum shaded with oblique lines, two, at least, of these three forms of action exist simultaneously. The intensity of each at any point of the spectrum is measured by the vertical line drawn from the point on the base line to meet its proper curve.

Whilst noticing these peculiar properties of the different rays, we must carefully remember that there really is no difference in kind between those rays which are called heating rays, those which are called light rays, and those which are called chemically active rays. These differ one from the other in exactly the same way that the visible yellow rays differ from the green rays, or as the green rays differ from the blue ; only in wave-length and intensity of vibration. In any particular portion of the spectrum we cannot separate the rays of light and leave the rays of heat behind ; we cannot, for instance, separate out the yellow rays of light from the yellow portion of the spectrum, and leave behind any rays of heat of the same degree of reffangibility. But, as I shall show you, we can separate from the whole radiation the luminous rays, and with them the heating effect of those luminous rays, and still leave the dark or invisible rays of heat of lower refrangibility.

I will endeavour to prove to you, in the first place, the existence of these dark heating rays of really invisible light. We see that the maximum of these rays is placed beyond the visible red. This may be clearly exhibited with the electric light in the beautiful experiment by which Dr. Tyndall first accomplished the separation of which I have just spoken. I have for this purpose placed in the dark box (d, Fig. 6) an electric lamp (l),

SPECTRUM ANALYSIS.

[lect. 1.

which gives us a very bright light, and by means of this mirror (m) I can bring the rays to a focus at any desired point. Here is a cell (c) which Dr. Tyndall

Fig. 6.

very expressively calls a ray-filter, by which I can filter out the whole of the luminous rays, by passing them through this opaque solution of iodine in disulphide of carbon, whilst the invisible heating rays are trails-

I.ECT. l.J

CJLOltESCENCE.

mitted, and will soon render themselves evident to you. A current of cold water circulates through a double jacket on the outside of the cell, to keep the volatile disulphide cool. Now I think you will observe that no rays of light come through, but if I take a piece ol black paper, and place it in the focus of our mirror, you see the paper is ignited, owing to the presence of these dark heating rays. I may now do the same with a piece of blackened platinum (p, Fig. 6) ; you see that this also is heated to redness. I can show you this again in a variety of forms. Here is some gunpowder strewed on this paper ; you observe that it at once explodes when brought into the focus of the dark rays. Here I have some blackened gun-cotton, which instantly catches fire. I may vary the experiment by lighting a cigar ; and here you see the brilliant scintillations of charcoal burning in oxygen, having been heated up to the temperature of ignition in the focus of the dark rays. Dr. Tyndall has measured the proportional amount of the entire heating rays which, pouring forth from this incandescent carbon, has passed through this dark filter, and he has found that this consists of of the whole amount ; so that only ^ of the radiation is really visible.

Understanding then the existence in this ultra-red of a large amount of heating rays, let us pass now to consider the properties of the light which is given off* at the opposite or blue end of the spectrum, which I have called the chemically active rays. Allow me to show you an experiment to prove that it is in the blue portion of the spectrum that we have essentially these chemically active rays. In order to render the illustration more perfect, I will first make an experiment with reference simply to white light, to show you that

SPECTRUM ANALYSIS.

[lkot. I.

the brilliant light which is emitted by this burning magnesium, and is almost too dazzling for the eye to bear, contains a very large proportion of the rays which we are about to investigate.1

I have here a thin glass bulb containing a mixture of equal volumes of two gases, chlorine and hydrogen. These gases when exposed to a bright light combine together, and form hydrochloric acid gas. If I were to throw this bulb out into the sunlight, so rapid would be

Fig. 7.

the combination, and so great the consequent evolution of heat and sudden expansion, that this little bulb would instantly be shattered into a thousand fragments. Almost as sudden an effect will be produced if I simply burn a bit of magnesium wire in the neighbourhood of the bulb (Fig. 7); it explodes with a pretty loud report, the bulb is shattered, and the gases have been combined

1 See Appendix B for the measurement of the chemical intensity of magnesium light.

CHEMICALLY ACTIVE: RAYS.

by virtue of the blue rays contained in this kind of light. I will next show you that it is really the blue rays which thus act chemically. This lantern (Fig. 8) contains panes of different coloured glass, — here a white one, here a yellow one, here a red one, there a blue one. 1 am going to put another of these little bulbs filled with chlorine and hydrogen in the inside of this lantern, and then I will produce, not by magnesium wire, but by another means, a very bright blue light, a light which contains these chemically active rays in great quantity. 1 will first allow this blue light to shine upon the bulb through the red pane of glass. Here I

Fig. 8.

produce a very bright flame, by throwing some carbon disulphide into a tall cylinder full of nitric oxide gas, and igniting the mixture. Here is the bright flash, but you have noticed no explosion of the bulb, for all the

c

IS

SPECTRUM ANALYSIS.

[lect. I.

chemical rays have been held back: filtered off by this red glass, they cannot pass through ; and the consequence is, there has been no action on my bulb. I will now allow another of these flashes of light to pass through the blue glass, which being of course transparent to the blue rays my little bulb will be shattered into a fine powder, as you observe. Here then we have ascertained by experiment that the blue rays act chemically, whilst the red rays produce heating effects.

This sensitive mixture of chlorine and hydrogen, which, as you have seen, explodes when the chemical activity of the light is great, may be used as a most delicate means of measuring the amount of light of a less intensity. The combination of the gases then occurs slowly, and may be rendered evident by allowing the hydrochloric acid thus formed to be absorbed by water, when the consequent diminution in bulk of the gas accurately represents the chemical action effected.

The varying intensity of the chemically active rays in different parts of the solar spectrum has been , carefully measured by means of this sensitive mixture of chlorine and hydrogen gases.1 The accompanying figure (Fig. .9) exhibits the chemical action effected by the various portions of the spectrum on the sensitive mixture for one particular zenith distance of the sun. The lines marked with the letters of the alphabet from A to w, at the bottom of the figure, represent the fixed dark lines which exist in the solar spectrum, of which I shall have much to say in the subsequent lectures. They serve as a sort of landmarks by which to ascertain the position of any given point in the spectrum. The greatest amount of chemical action is noticed between the line in the indigo marked G, 1 See Appendix C for description of method.

LECT. 1.]

CHEMICAL ACTION.

and that in the violet marked it. In the direction of the red end of the spectrum, the action becomes imperceptible about d, in the orange (the maximum of visible illumination) ; whilst towards the other end of the spectrum the action was found to extend as far as the line marked u, or to a greater distance beyond the line h in

HH I it M N NO the violet than the total length of the ordinary visible spectrum.

I may again illustrate this same fact by showing you that I can photograph with these blue rays, whereas I fail to produce the same effect with the red rays. I will coat a plate with collodion, and then darken the room, with the exception of this yellow mono-chromatic flame, produced by the volatilization of soda salts, which is incapable of acting chemically. I may, therefore, work with this light without at all affecting my photographic plate. I have now coated a plate with collodion, and sensitised it in the silver bath. I shall next expose this

c 2

SPECTRUM ANALYSIS.

[lect. I.

to the action of the light of the spectrum of the electric lamp. I would first show you that 1 have here (Fig. 10) a negative photograph, of which I am about to take; a print by the blue rays of the electric lamp. You will observe that there are two figures upon the negative, one marked V and the other marked E ; these letters being intended to signify Violet and Red. The one figure marked V I propose to place in front of my sensitised plate in the blue or violet ray, and the one marked R

1' 10. 10.

1 propose to open in the red ray, and 1 hope to be able to produce a chemical effect on that portion of my sensitised plate which has been exposed to the blue, whilst we shall get no corresponding effect on the portion exposed to the red ray. I next place my plate with its face downwards on the negative ; we now start our electric lamp, using a small spectrum in order to have the action rather more distinct. 1 then expose half my plate in

I.ECT. I.]

PHOTOGRAPHY IN BLUR RAYS.

•JL the red ra}Ts for about twenty seconds, and afterwards expose the other half, with the V upon it, for about the same length of time to the violet light. I will now develop and wash the photograph, and throw the image produced on the screen, when you will observe (Fig. 11) a very marked difference between the two portions, the one showing no picture at all, and the other giving us the perfect picture with the V upon it, rendering the difference of action of the blue and red rays visible.