0M-5-107058 - ERIC

ED 071 897

TITLE

INSTITUTION

SPONS AGENCY

BUREAU NOPUB DATECONTRACTNOTE .

DOM/KEIT RESUME

Project Physics Reader 4,Electromagnetism.Harvard 'Jail's, Cambridge,Physics.Office of Education (DREW)of Research.BR-5-1038670M-5-107058254p.; preliminary Version

SE 015 534

Light and

Mass. Harvard Project

, Washington, D.C. Bureau

EDRS PRICE MF-$0.65 HC-49.87DESCRIPTORS Electricity; *Instructional Materials; Li ,t;

Magnets;, *Physics; Radiation; *Science Materials;Secondary Grades; *Secondary School Science;*Supplementary Reading Materials

IDENTIFIERS Harvard Project Physics

ABSTRACT. . As a supplement to Project Physics Unit 4, acollection. of articles is presented.. in this reader. for studentbrowsing. The 21 articles are. included under the ,following headings:

_Letter from Thomas Jefferson; On the Method of Theoretical Physics;Systems, Feedback, Cybernetics; Velocity of Light; PopularApplications of.Polarized Light; Eye and Camera; The laser--What itis and Doe0; A .Simple Electric Circuit: Ohmss Law; The. Electronic

. Revolution; The Invention of the Electric. Light; High Fidelity; The. Future of Current Power Transmission; James Clerk Maxwell, .,

Part II; On Ole Induction of Electric Currents; The Relationship of .

Electricity and Magnetism; The Electromagnetic Field; Radiation Belts.

.Around the Earth; A .Mirror for the Brain; Scientific Imagination;Lenses and Optical Instruments; and "Baffled!." Illustrations forexplanation use. are included. The work of Harvard. roject Physics haS

...been financially supported by: the Carnegie Corporation of New York,the_ Ford Foundations, the National Science Foundation.the_Alfred P.

. Sloan Foundation, the. United States Office of Education, and Harvard.University..(CC)

An Introduction to Physics

Project Physics Reader

Light and Electromagnetism

U S DEPARTMENT OF HEALTH.EDUCATION & WELFAREOFFICE OF EDUCATION

THIS DOCUMENT HAS BEEN REPRODUCED EXACTLY AS RECEIVED FROMTHE PERSON OR ORGANIZATION ORIGINATING IT POINTS OF VIEW OR OPINIONS STATED DO NOT NECESSARILYREPRESENT OFFICIAL OFFICE OF EDUCATION POSITION OR POLICY

Project Physics Reader

An Introduction to Physics 4 Light and Electromagnetism

Preliminary Version 1967-68

An1

,

This is not a physics textbook. Rather, it is a physicsreader, a collection of some of the best articles andbook passages In physics. A few are on historic eventsin science, oth.ns contain some particularly memorabledescription of what physicists do; still others deal withphilosophy of science, or with the impact of scientificthought on the imagination of the artist.

There are old and new classics, and also some little-known publications; many have been suggested for in-clusion because some teacher or physicist rememberedan article with particular fondness. The majority ofarticles is not drawn from scientific papers of historicimportance themselves, because material from many ofthese is readily available, either as quotations in theProject Physics text or in special collections.

This collection is meant for your browsing. If you followyour own reading interests, chances are good that youwill find here many pages that convey the joy theseauthors have in their work and the excitement of theirideas. If you want to follow up on interesting excerpts,the source list at the end of the reader will guide youfor further reading.

iii

1 Letter from Thomas Jefferson I

June 1799

2 On the Method of Theoretical PhysicsAlbert Einstein

An essay-1934

3 Experiments and Calculations Relative to Physical OpticsThomas Young

A scientific paper published in 1855

4 Velocity of LightA A MichelsonA chapter from his book, Studies in Optics. published in 1927

5 Popular Applications of Polarized LightWilliam A. Shorcliff and Stanley S. BallardA chapter from the book Polarized Lip t published in 1964

6 Action at a DistanceJames Clerk MaxwellA scientific paper published in 1873.

7 The Electronic RevolutionArthur C. Clarke

1962

8 The Invention of the Electric LightMatthew Josephson1959

9 High FidelityEdgar Vilichur

Two chapters from his book Reproduction of Sound published in 1962.

5

15

29

49

69

83

91

103

The Future of Direct Current Power TransmissionN L AllenA popular article published in 1967.

11 James Clerk Maxwell, Part IIJames R NewmanA biographit.al essay published in 1955

119

123

12 Maxwell's Letters: A Collection 157

13 On the Induction of Electric CurrentsJames Clerk Maxwell

From his Treatise on Electricity and Magnetism published in 1873

14 The Relationship of Electricity and MagnetismD K C MacDonaldExcerpt from his book, Faraday, Maxwell, and Kelvin, published in 1964

15 The Electromagnetic FieldAlbert Einstein and Leopold Infeld

Excerpt from their book entitled the Evolution of Physics publishedin 1938 and 1961

16 Radiation Belts Around the EarthJames Var. Allen

1959

17 A Mirrcr for the BrainW. Grey Walter

A chapter from his book The Living Brain published in 1963.

18 Scientific ImaginationRichard P. Feynman, Robert B, Leighton, and Matthew Sands

Excerpt from The Feynman Lectures on Physics, Volume II, 1964.

165

169

177

203

213

239

Letter from Thomas Jefferson

. June 1799

Monticello June 18. 99.DEAR SIR,

I have to acknolege the reciept of your favor ofMay 14. in which you mention that you have finishedthe 6. first books of Euclid, plane trigonometry, sur-veying and algebra and ask whether I think a furtherpursuit of that branch of science would be useful toyou. There arc some propositions in the latter booksof Euclid, and some of Archimedes, which arc useful,and I have no doubt you have been made acquaintedwith them. Trigonometry, so far as this, is mostvaluable to every man, there is scarcely a day inwhich he will not resort to it for some of the purposesof common life ; the science of calculation also isindispensible as far as the extraction of the square andcube roots ; Algebra as far as the quadratic equationand the use of logarithms are often of value in ordi-nary cases ; but all beyond these is but a luxury ; adelicious luxury indeed ; but not to be indulged in byone who is to have a profession to follow for his sub-sistence. In this light I view the conic sections,

1

curves of the orders, perhaps even sphericaltrigonometry, Algebraical operations beyond the2d dimension, and fiuxions. There are otherbranches of science however worth the attention ofevery man : Astronomy, botany, chemistry, naturalphilosophy, natural history, anatomy. Not indeedto be a proficient in them; but to possess their generalprinciples and outlines, so as that we may be able toamuse and inform ourselves further in any of them aswe proceed through life and have occasion for them.Some knowlcge of them is necessary for our characteras well as comfort. The general elements of astro-nomy and of natural philosophy are best acquired atan academy where we can have the benefit of theinstruments and apparatus usually provided there:but the others may well be acquired from booksalone as far as our purposes require. I have indulgedmyself in these observations to you, because theevidence cannot be unuscful to you of a person whohas often had occasion to corridcr which of hisacquisitions in science have been really useful to himin life, and which of them have bccn merely a matterof luxury.

I am among those who think well of the humancharacter generally. I consider man as formed forsociety, and endowed by nature with those disposi-tions which fit him for society. I believe also, withCondorcct, as mentioned in your letter, that his mindis perfectible to a degree of which we cannot as yetform any conception. It is impossible for a man whotakes a survey of what is already known, not to secwhat an immensity in every branch of science yetremains to be discovered, and that too of articles towhich our faculties stem adequate. In geometry andcalculation we know a great deal. Yet there aresome desiderata. In anatomy great progress hasbeen made; but much is still to be acquired. Innatural history we possess knowlcgc ; but we wanta great deal. In chemistry we are not yet sure ofthe first elements. Our natural philosophy is in avery infantinc state; perhaps for great advances init, a further progress in chemistry is necessary.Surgery is well advanced; but prodigiously short ofwhat may be. The state of inedecine is worst thanthat of total ignorance. Could we divest ourselves of

2

every thing we suppose we know in it, we should startfrom a higher ground and with fairer prospects.From Hippocrates to Brown we have had nothingbut a succession of hypothetical systems each havingit's day of vogue, like the fashions and fancies ofcaps and gowns, and yielding in turn to the nextcaprice. Yet the human frame, which is to be thesubject of suffering and torture under these learnedmodes, does not change. We have a few inedecines,as the bark, opium, mercury, which in a few welldefined diseases arc of unquestionable virtue: butthe residuary list of the materia medial, long as it is,contains but the charlatancrics of the art ; and of thediseases of doubtful form, physicians have ever had afalse knowlege, worse than ignorance. Yet surelythe list of unequivocal diseases and remedies iscapable of enlargement ; and it is still more certainthat in the other branches of science, great fields areyet to be explored to which our faculties arc equal,and that to an extent of which we cannot fix thelimits. I join you therefore in branding as cowardlythe idea that the human mind is incapable of furtheradvgnces. This is precisely the doctrine which thepresent despots of the earth are inculcating, andtheir friei.ds here re-echoing; and applying especiallyto religion and politics; that it is not probable thatany thing better will be discovered than what wasknown to our fathers '. We are to look backwardsthen and not forwards for the improvement ofscience, and to find it amidst feudal barbarisms andthe fires of Spital-fields. But thank heaven theAmerican mind is already too much opened, to listento these impostures; and while the art of printing isleft to use, science can never be retrograde; what isonce acquired of real knowlege can never be lost.To preserve the freedom of the human mind then andfreedom of the press, every spirit should be ready todevote itself to rnartrdom; for as long as we maythink as we will, and speak as we think, the conditionof man will proceed in improvement. The genera-tion which is going off the stage has deserved well ofmankind for the struggles it has made, and for havingarrested that course of despotism which had over-whelmed the world for thousands and thousands ofyears. If there seems to be danger that the ground

3

they have gained will be lost again, that danger comesfrom the generation your contemporary, But thatthe enthusiasm which characterises youth should liftit's parracide hands against freedom and sciencewould be such a monstrous phaenomenon as,I can-not place among possible things in this age and thiscountry. Your college at least has shewn itselfincapable of it ; and if the youth of any other placehave seemed to rally under other banners it has beenfrom delusions which they will soon dissipate. Ishall be happy to hear from you from time to time,and of your progress in study, and to be useful toyou in whatever is in my power ; being with sincereesteem Dear Sir

Your friend & servtTh: Jefferson

4

On the Method of Theoretical Physics

Albert Einstein

An essay-1934.

IF YOU want to find out anything from the theoreticalphysicists about the methods they use, I advise youto stick closely to one principle : don c listen to theirwords, fix your attention on their deeds. To himwho is a discoverer in this field the products of hisimagination appear so necessary and natural that heregards theta, and would like to have them regardedby others, not as creations of thought but as givenrealities.

These words sound like an invitation to you towalk out of this lecture. You will say to yourselves,the fellow's a working physicist himself and oughttherefore to leave all questions of the structure oftheoretical science to the epistemologists.

Against such criticism I can defend myself fromthe personal point of view by assuring you that itis not at my own instance but at the kind invitationof others that I have mounted this rostrum, whichserves to commemorate a man who fought hard allhis life for the unity of knowledge. Objectively, how-ever, my enterprise can be justified on the groundthat it may, after all, be of interest to know how onewho has spent a life-time in striving with all his

5

might to clear up and rectify its fundamentals looksupon his own branch of science. The way in wl ich heregards its past and present may depend too muchon what he hopes for the future and aims at in thepresent; but that is the inevitable fate of anybodywho has occupied himself intensively with a worldof ideas. The same thing happens to him as to thehistorian, who in the same way, even though perhapsunconsciously, groups actual events around idealswhich he has formed for himself on the subject ofhuman society.

Let us now cast an eye over the development ofthe theoretical system, paying special attention tothe relations between the content of the theoryand the totality of empirical fact. We are concernedwith the eternal antithesis between the two insep-arable components of our knowledge, the empiricaland the rational, in our department.

We reverence ancient Greece as the cradle ofwestern science. Here for the first time the worldwitnessed the miracle of a logical system which pro-ceeded from step to step with such precision thatevery single one of its propositions was absolutelyindubitable I refer to Euclid's geometry. This ad-mirable triumph of reasoning gave the human intel-lect the necessary confidence in itself for its subsequentachievements. If Euclid failed to kindle your youth-ful enthusiasm, then you were not born to be asPientific thinker.

But before mankind could be ripe for a sciencewhich takes in the whole of reality, a second funda-

6

mental truth was needed, which only became commonproperty among philosophers with the advent of Kep-ler and Galileo. Pure logical thinking cannot yieldus any knowledge of the empirical world ; all knowl-edge of reality starts from experience and ends init. Propositio'is arrived at by purely logical meansare completely empty as regards reality. BecauseGalileo saw this, and particularly because he drummedit into the scientific world, he is the father of modernphysicsindeed, of modern science altogether.

If, then, experience is the alpha and the omega ofall our knowledge of reality, what is the function ofpure reason in science ?

A complete system of theoretical physics is madeup of concepts, fundamental laws which are supposedto be valid for those concepts and conclusions to bereached by logical deduction. It is these conclusionswhich nr tst correspond with our separate experienc'es ;

in any theoretical treatise their logical deductionoccupies almost the whole book.

This is exactly what happens in Euclid's geometry,except that there the fundamental laws are calledaxioms and there is no question of the conclusionshaving to correspond to any sort of experience. If,however, one regard Euclidean geometry as the sci-ence of the possible mutual relations of practicallyrigid bodies in space, that is to say, treats it a, aphysical science, without abstracting from its originalempirical content, the logical homogeneity of geometryand theoretical physics becomes complete.

We have thus assigned to pure reason and ex-

7

.

or

-.J.4 .1,,

8

perience their places in a theoretical system of physics.The structure of the system is the work of reason ; theempirical contents and their mutual relations mustfind their representation in the conclusions of thetheory. In the possibility of such a representation liethe sole value and justification of the whole system,and especially of the concepts and fundamental prin-ciples which underlie it. These latter, by the way, arefree inventions of the human intellect, which cannotbe justified either by the nature of that intellect orin any other fashion a priori.

These fundamental concepts and postulates, whichcannot be further reduced logically, form the essentialpart of a theory, which reason cannot touch. It is thegrand object of all theory to make these irreducibleelements as simple and as few in number as possible,without having to renounce the adequate representa-tion of any empirical content whatever.

The view I have just outlined of the purely fictitiouscharacter of the fundamentals of scientific theorywas by no means the prevailing one in the eighteenthor even the ninete_nth century. But it is steadilygaining ground from the fact that the distance inthought between 'he fundamental concepts and lawson one side and, on the other, the conclusions whichhave to be brought into relation with our experiencegrows larger and larger, the simpler the logical struc-ture becomesthat is to say, the smaller the numberof logically independent conceptual elements whichare found necessary to support the structure.

Newton, the first creator of a comprehensive,

N.

workable system of theoretical physics, still believedthat the basic concepts and laws of his system couldbe derived from experience. This is no doubt themeaning of his saying, hypotheses non lingo.

Actually the concepts of time .0d space appearedat that time to present no difficulties. The conceptsof mass, inertia and force, and the laws connectingthem seemed to 'ue drawn directly from experience.Once this basis is accepted, the expression for theforce of gravitation appears derivable from experi-ence, and it was reasonable to hope for the same inregard to other forces.

We can indeed see from Newton's formulation ofit that the concept of absolute space, which comprisedthat of absolute rest, made him feel uncomfortable;he realized that there seemed to be nothing in ex-perience corresponding to this last concept. He wasalso not quite comfortable about the introduction offorces operating at a distance. But the tremendouspractical success of his doctrines may well have pre-vented him and the physicists of the eighteenth andnlneteenth centuries from recognizing the fictitiouscharacter of the foundations of his system.

The natural philosophers of those days were, onthe contrary, most of them possessed with the ideathat the fundamental concepts and postulates of-hysics were not in the logical sense free inventionsof the human mind but could be deduced from ex-perience by "abstraction"that is to say by logicalmeans. A clear, recognition of the erroneousness ofthis notion really only caine with the general theory

9

10

of relativity, which showed that one could take ac-count of a wider range of empirical facts, and thattoo in a more satisfactory and complete manner, ona foundation quite different from the Newtonian.But quite apart from the question of the superiorityof one or the other, the fictitious character of funda-mental principles is perfectly evident from the factthat we can point to two essentially different prin-ciples, both of which correspond with experience toa large extent ; this proves at the same time thatevery attempt at a logical deduction of the basic con-cepts and postulates of mechanics from elementaryexperiences is doomed to failure.

If, then, it is true that this axiomatic basis of theo-retical physics cannot be extracted from experiencebut must be freely invented, can we ever hope tofind the right way? Nay more, has this right way anyexistence outside our illusions ? Can we hope to beguided in the right way by experience when thereexist theories (such as classical mechanics) which toa large extent do justice to experience, withoutgetting to the root of the matter? I answer withouthesitation that there is, in my opinion, a right way,and that we are capable of finding it. Our experiencehitherto justifies us in believing that nature is therealization of the simplest conceivable mathematicalideas. I am convinced that we can discover by meansof purely mathematical constructions the conceptsand the laws connecting them with each other, whichfurnish the key to the understanding of natural phe-nomena. Experience may suggest the appropriate

On the Method of Iheorefical Phy:,u_s

mathematical concepts, but they most certainly cannotbe deduced from it. Experience remains, of course,the sole criterion of the physical utility of a mathe-matical construction. But the creative principle residesin mathematics. In a certain sense, therefore, I holdit true that pure thought can grasp reality, as theancients dreamed.

In order to justify this confidence, I am compelledto make use of a mathematical conception. The phys-ical world is represented as a four-dimensional con-tinuum. If I assume a Riemannian" metric in it andask what are the simplest laws which such a metricsystem can satisfy, I arrive at the relativist theoryof gravitation in empty space. If in that space Iassume a vector-field or an anti. symmetrical tensor-field which can be inferred from it, and ask whatare the simplest laws which such a field can satisfy,I arrive at Clerk Maxwell's equations for empty space.

At this point we still lack a theory for those partsof space in which electrical density does not disappear.De Broglie conjectured the existence of a wave field.which served to explain certain quantum propertiesof matter. Dirac found in the spinors field-magni-tudes of a new sort, whose simplest equations enableone to a large extent to deduce the properties of theelectron. Subsequently I discovered, in conjunctionwith my colleague, that these spinors form a specialcase of a new sort of field, mathematically connectedwith the four-dimensional system, which we called"semivectors." The simplest equations to which suchsemivectors can be reduced furnish a key to the

' v

12

understanding of the existence of two sorts of ele-mentary particles, of different ponderable mass andequal but opposite electrical charge. These semivectors

are, after ordinary vectors, the simplest mathematicalfields that are possible in a metrical continuum of

four dimensions, and it looks as if they described, in

an easy manner, certain essential properties of elec-

trical particles.The important point for us to observe is that all

these constructions and the laws connecting them canbe arrived at by the principle of looking for the mathe-matically simplest concepts and the link betweenthem. In the limited nature of the mathematicallyexistent simple fields and the simple equations pos-sible between them, lies the theorist's hope of grasp-ing the real in a'l its depth.

Meanwhile the great stumbling-block for a field-theory of this kind lies in the conception of theatomic structure of matter and energy. For the theoryis fundamentally non-atomic in so far as it operatesexclusively with continuous functions of space, in

contrast to classical mechanics, whose most impor-tant element, the material point, in itself does justiceto the atomic structure of matter.

The modern quantum theory in the form associatedwith the names of de Broglie, Schrodinger, andDirac, which operates with continuous functions, hasovercome these difficulties by a bold piece of inter-pretation which was first given a clear form by MaxBorn. According to this, the spatial functions whichappear in the equations make no claim to be a mathe-

ty.

matical model of the atomic structure. Those func-tions are only supposed to determine the mathematicalprobabilities of the occurrence of such structures ifmeasurements were taken at a particular spot or in acertain state of motion. This notion is logically un-objectionable and has important successes to itscredit. Unfortunately, however, it compels one to usea continuum the number of whose dimensions is notthat ascribed to space by physics hitherto (four) butrises indefinitely with the number of the particlesconstituting the system under consideration. I cannotbut confess that I attach only a transitory importanceto this interpretation. I still believe in the possibilityof a model of realitythat is to say, of a theory whichrepresents things themselves and not merely theprobability of their occurrence.

On the other hand it seems to me certain that wemust give up the idea of a complete localization ofthe particles in a theoretical model. This seems tome to be the permanent upshot of Heisenberg'sprinciple of uncertainty. But an atomic theory in thetrue sense of the word (not merely on the basis ofan interpretation) without localization of particlesin a mathematical model, is perfectly thinkable. Forinstance, to account for the atomic character of elec-tricity, the field equations need only lead to thefollowing conclusions: A portion of space (three-dimensional) at whose boundaries electrical densitydisappears everywhere, always contains a total elec-trical charge whose size is represented by a wholenumber. In a continuum-theory atomic characteristics

13

4

14

would be satisfactorily expressed by integral lawswithout localization of the formation entity whichconstitutes the atomic structure.

Not until the atomic structure has been successfullyrepresented in such a manner would I consider thequantum-riddle solved.

e

Experiments and Calculations Relative to Physical Optics

Thomas Young

A scientific paper published in 1855.

EXPERIMENTS AND CALCULATIONS RELATIVE TO

PHYSICAL OPTICS.

From the Philosophical Transactions for 1804.

A BAKERIAN LECTURE.

READ Nov. 24, 1803.

1.Experimental Demonstration of the general Law of theInterference of Light.

IN making some experiments on the fringes of colours accom-panying shadows, I have found so simple and so demonstrativea proof of the general law of the interference of two portions oflight, which I have already endeavoured to establish, that Ithink it right to lay before the Royal Society a short statementof the facts which appear to me so decisive. The propositionon which I mean to insist, at present, is simply thisthat fringesof colours are produced by the interference of two portions oflight ; and I think it will not be denied by the most prejudiced,that the assertion is proved by the experiments I am aboutto relate, which may be repeated with great ease whenever thesun shines, and without any other apparatus than is at hand toevery one.

Exper. 1. I made a bmall hole in a window-shutter, andcovered it with a piece of thick paper, which I perforated witha fine needle. For greater convenience of observation I placeda small looking-glass without the window-shutter, in such aposition as to reflect the sun's light, in a direction nearly hori-zontal, upon the opposite wall, and to cause the cone of diverginglight to pass over a table on which were several little screens of

15

16

card-paper. I brought into the sunbeam a slip of card, aboutone-thirtieth of an inch in breadth, and observed its shadow,either on the wall or on other cards held at different distances.Besides the fringes of colour on each side of the shadow, theshadow itself n as divi,led by similar parallel fringes, of smallerdimensions, differing in number, according to the distance atwhich the shadow was observed. but leaving the middle of theshadow always white. Now these fringes were the joint effectsof the portions of light passing on each side of the slip of card,and inflected, or rather diffilieted, into the shadow. For, a littlescreen being placed a few inches from the card, so as to receiveeither edge of the shadow on its margin, all the fringes whichhad before been observed in the shadow on the wall, immediatelydisappeared, although the light inflected on the other side wasallowed to retain its course, and although this light must haveundergone any modification that the proximity of the other edgeof the slip of card might have-been capable of occasioning.When the interposed screen was more remote from the narrowcard, it was necessary to plunge it more deeply into the shadow,in order to extinguish the parallel lines ; for here the light,diffracted from the edge of the object, had entered further intothe shadow in its way towards the fringes. Nor was it for wantof a sufficient intensity of light that one of the two portions wasincapable of producing the fringes alone ; for, when they wereboth uninterrupted, the lines appeared, even if the intensity wasreduced to one-tenth or one-twentieth.

Eye?. , 2. The crested fringes described by the ingenious andaccurate Grimaldi, afford an elegant variation of the precedingexperiment, and au interesting example of a calculation groundedon it. When a shadow is formed by an object which has a rect-angular termination, besides the usual external fringes, thereare two or three alternations of colours. beginning from the linewhich bisects the angle, disposed on each side of it in curves,which are convex towards the bisecting line, and which con-verge in some degree towards it, as they become more remotefrom the angular point. These fringes are also the joint effectof the light which is inflected directly towards the shadow fromeach of the two outlines of the object; for if a screen be placed

within a few inches of the object, so as to receive only one ofthe edges of the shadow, the whole of the fringes disappear : if,on the contrary, the rectangular point of the screen be opposedto the point of the shadow, so as barely to receive the angleof the shadow on its extremity, the fringes will remain undis-turbed.

II.Comparison of Measures deduced from various Experiments.

If we now proceed to examine the dimensions of the fringes,under different circumstances, we may calculate the differencesof the lengths of the paths described by the portions of lightwhich have thus been proved to be concerned in producing thosefringes ; and we shall find that, where the lengths are equal, thelight always remains white; but that, where either the brightestlight, or the light of any given colour, disappears and reappears,a first, a second, or a third time, the differences of the lengthsof the paths of the two portions are in arithmetical progression,as nearly as we can expect experiments of this kind to agreewith each other. I shall compare, in this point of view. the;neasures deduced from several experiments of Newton, andfrom some of my own.

In the eighth and ninth observations of the third book ofNewton's Optics, some experiments are related, which, togetherwith the third observation, will furnish u, with the data neces-sary for the calculation. Two knives were placed, with theiredges meeting at a very acute angle, in a beam of the sun'slight, admitted through a small aperture, and the point of con-course of the two first dark lines bordering the shadows of therespective knives was observed at various distances. The resultsof six observations are expressed in the first three lines of thefirst Table. On the supposition that the dark line is producedby the first interference of the light reflected from the edges ofthe knives, with the light passing in a straight line between them,we may assign, by calculating the difference of the two paths,the interval for the first disappearance of the brightest light, asit is expressed in the fourth line. The second Table contains theresults of a similar calculation, from Newton's observations onthe shadow of a hair ; and the third, from some experiments of

17

18

my own, of the same nature ; the second bright line being sup-posed to correspond to a double interval, the second dark lineto a triple interval, and the succeeding lines to depend on acontinuation of the progression. The unit of all the tables isan inch.

TABLE I. Obs. 9. N.

Distance of the knives from the aperture 101Distance ofthe paperfrom theknives . II, 31, sit 32, 96. 131

Distancesbe-tween theedges of theknives, op-posite tothe point ofconcourse. .012, .020, .034, .057, .081, .087Interval ofdisappear.ance . . .0000122, .0000155, .0e.)0182, .0000167, .0000166, .0000166

tTABLE II. Obs. 3. I.

Breadth of the hairDistance of the hair from the apertureDistances of the scale fmm the apertur, .(Breadths of the shadow . . .Breadth between the second pair of bright linesInterval of disappearance, or half the difference of the

paths . . . . . .Breadth between the third pair of bright lines .Interval of disappearance, one-fourth of the difference

spa144

150. 252sto

.0000151, .00001733t

4 315

.0000130, .0000143

TABLE HI. Exper. 3.Breadth of the object . . . .434Distance of the object from the aperture 125Distance of the mall from the aperture . . . 250Distance of the second pair of dark lines from each otherInterval of disappearance, one-third of the difference . .0001011649

Erper. 4.Breadth of the wire .083Distance of the wire from the aperture 32Distance of the wall from the aperture 250(Breadth of the shadow, by three

measurements . . . .815, .826, or .827; mean, .823)Distance of the first pair of dark lines 1.165, 1.170, or 1.160; mean,.0001;11:4 5Interval of disappearance .Distance of the second pair of dark

lines . . . . . 1.402, 1.395, or 1.400; mean, 1.399Interval of disappearance . . 0000137Distance of the third pair of dark

lines . . . . . 1.594, 1.580, or 1.585; mean, 1.586.0000128Interval of disappearance .

It appears, from five of the six observationsof the first Table,in which the distance of the shadow was varied from about3 inches to 11 feet, and the breadth of the fringes was increasedin the ratio of 7 to 1, that the difference of the routes constitutingthe interval of disappearance, varied but one-eleventh at most;and that, in three out of the five, it agreed with the mean, eitherexactly, or within Th. part. Hence we are warranted in in-ferring that the interval appropriate to the extinction of thebrightest light, is either accurately or very nearly constant.

But it may be inferred, from a comparison of all the otherobservations, that when the obliquity of the refl2ction is verygreat, some circumstance takes place, which causes the i.Itcr-althus calculated to be somewhat greater : thus, in the eleventhline of the third Table, it comes out one-sixth greater than themean of the five already mentioned. On the other hand, themean of two of Newton's experiments and one of mine, is aresult about one-fourth less than the former. With respect tothe nature of this circumstance, I cannot at present form adecided opinion ; but I conjecture that it is a deviation of someof the light concerned, from the rectilineardirection assigned toit, arising either from its natural diffraction, by which the mag-nitude of the shadow is also enlarged, or from some otherunknown cause. If we imagined the shadow of the wire, andthe fringes nearest it, to be so contracted that the motion of thelight bounding the shadow might be rectilinear, we should thusmake a sufficient compensation for this deviation ; but it is dif-ficult to point out what precise track of the light would cause itto require this correction.

The mean of the three experiments which appear to havebeen least affected by this unknown deviation, gives .0000127tbr the interval appropriate to the disappearance of the brightestlight ; and it may be inferred that if they had been whollyexempted from its effects, the measure would have been some-what smaller. Now the analogous interval, deduced from theexperiments of Newton on thin plates, is .0000112, which isabout one-eighth less than the former result ; and this appearsto be a coincidence fully sufficient to authorise us to attributethese two classes of phenomena to the same cause. It is very

19

easily shown, witlespect to the colours of thin plates, thateach kind of lig' disappears and reappears where the dif-ferences of the routes of two of its portions are in arithmeticalprogression ; and we have seen that the same law may be ingeneral inferred from the phenomena of diffracted light, evenindependently of the analogy.

The distribution of the colours is also so similar in both cases,as to point innnediately to a similarity in the causes. In thethirteenth observation of the second part of the first book,Newton relates, that the interval of the glasses where the ringsappeared in red light, was to the interval where they appearedin violet light, as 14to 9 ; and, in the eleventh observationof the third book, that the distances between the fringes,under the same circumstances, were the 22d and 27th of aninch. Hence, deducting the breadth of the hair and takingthe squares, in order to find the relation of the difference ofthe routes, we have the proportion of 14 to 9.1, which scarcelydiffers from the proportion observed in the colours of the thinplate.

We may readily determine, from this general principle, theform of the crested fringes of Grimaldi, already described ; farit will appear that, under the circumstances of the experimentrelated, the points in which the differences of the lengths of thepaths described by the two portions of light are equal to a con-stant quantity, and in which, therefore, the same kinds of lightought to appear or disappear, are always found in equilateralhyperbolas, of which the axes coincide with the outlines of theshadov,, and the asymptotes nearly with the diagonal line.Such, therefore, must be the direction of the fringes ; and thisconclusion agrees perfectly with the observation. But it mustbe remarked, that the parts near the outlines of the shadow areso much shaded off, as to render the character of the curvesomewhat less decidedly marked where it approaches to itsaxis. These fringes have a slight resemblance to the hyper-bolic fringes observed by Newton ; but the analogy is onlydistant.

20

III.Application to the Supernumerary Rainbows.

The repetitions of colours sometimes observed within thecommon rainbow, and described in the Philosophical Transac-tions, by Dr. Langwith and Mr. Daval, admit also a very easyand complete explanation from the same principles, Dr. Pem-berton has attempted to point out an analogy between thesecolours and those of thin plates ; but the irregular reflectionfrom the posterior surface of the drop, to which alone he attri-butes the appearance, must be far too weak to produce any visibleeffects. In order to understand the phenomenon, we have onlyto attend to the two portions of light which are exhibited in thecommon diagrams explanatory of the rainbow, regularly reflectedfrom the posterior surface of the drop, and crossing each otherin various directions, till, at the angle of the greatest deviation,they coincide with each other, so as to produce, by the greaterintensity of this redoubled light, the common rainbow of 41degrees. Other parts of these two portions will quit the dropin directions parallel to each other ; and these would exhibit acontinued diffusion of fainter light, for 25° within the brighttermination which forms the rainbow, but for the general lawof interference, which, as in other similar cases, divides the lightinto concentric rings ; the magnitude of these rings dependingon that of the drop, according to the difference of time occupiedin the passage of the two portions, which thus proceed in paralleldirections to the spectator's eye, after having been differentlyrefracted and reflected within the drop. This difference varies,at first, nearly as the square of the angular distance from theprimitive rainbow ; and, if the first additional red be at the dis-tance of 2° from the red of the rainbow, so as to interfere a littlewith the primitive violet, the fourth additional red will be at adistance of nearly 2° more ; and the intermediate colours willoccupy a space nearly equal to the original rainbow. In orderto produce this effect, the drops must be about Tir of an inch, or.013, in diameter : it would be sufficient if they were between:,J, and ylv. The reason that such supernumerary colours arenot often seen, must be, that it does not often happen that dropsso nearly equal are found together ; but, that this may some-

21

times happen, is not in itself at all improbable : we measureeven medicines by dropping them from a phial, and it may easilybe conceived that the drops formed by natural operations maysometimes be as uniform as any that can be produced by art.How accurately this theory coincides with the observation, maybest be determined from Dr. Langwith's own words.

" August the 21st, 1722, about half an hour past five in theevening, weather temperate, wind at north-east, the appearancewas as follows :The colours of the primary rainbow were asusual, only the purple very much inclining to red, and welldefined : under this was an arcl, of green, the upper part ofwhich inclined to a bright yellow, the lower to a more duskygreen : under this were alternately two arches of reddishpurple, and two of green : under all, a faint appearance ofanother arch of purple, which vanished and returned severaltimes so quick, that we could not readily fix our eyes upon it.Thus the order of the colours was, I. Red, orange- colour, yel-low, green, light-blue, deep blue, purple. II. Light green, darkgreen, purple. III. Green, purple. IV. Green, faint vanish-ing purple. You see we had here four orders of colours, andperhaps the beginning of a fifth : for I make no question butthat what I call the purple, is a mixture of the purple of eachof the upper series with the red of the next below it, and thegreen a mixture of the intermediate colours. I send you notthis account barely upon the credit of my own eyes ; for therewas a clergyman and four ,,,her gentlemen in company, whomI desired to view the colours attentively, who all agreed thatthey appear in the manner that I have now described. Thereare two things which well deserve to be taken notice of, as theymay perhaps direct us, in some measure, to the solution of thiscurious phenomenon The first is, that the breadth of the firstseries so far exceeded that of any of the rest, that, as near ascould judge, it was equal to them all taken together. Thesecond is, that I have never observed these inner orders ofcolours in the lower parts of the rainbow, though they haveoften been incomparably more vivid than the upper parts, underwhich the colours have appeared. I have taken notice of thisso very often, that I can hardly look upon it to be accidental ;

22

upt!.1111.,-11'. i.id Csitictst3t1<,ns 1-400Itiv tc pkvstr ()t;,,

and, if it should prove true in general, it will bring the disqui-sition into a narrow compass ; for it will show that this effectdepends upon some property which the drops retain, whilst theyare in the upper part of the air, but lose as they come lower,and are more mixed with one another." Phil. Trans., Vol.XXXII. p. 243.

From a consideration of the nature of the secondary rainbow,of 54°, it may be inferred, that if any such supernumerarycolours were seen attending this rainbow, they would necessarilybe external to it, instead of internal. The circles sometimesseen encompassing the observer's shadow in a mist, are perhapsmore nearly related to the common colours of thin plates asseen by reflection.

IV.Argumentative Inference respecting the Nature of Light.

The experiment of Grimaldi, on the crested fringes withinthe shadow, together with several others of his observations,equally important, has been left unnoticed by Newton. Thosewho are attached to the Newtonian theory of light, or to thehypothesis of modern opticians, founded on views still less en-larged, would do well to endeavour to imagine any thing likean explanation of these experiments, derived from their owndoctrines ; and, if they fail in the attempt, to refrain at leastfrom idle declamation against a system which is founded on theaccuracy of its application to all these facts, and to a thousandothers of a similar nature.

From the experiments and calculations which have been pre-mised, we may be allowed to infer, that homogeneous light, atcertain equal distances in the direction of its motion, is possessedof opposite qualities, capable of neutralising or destroying eachother, and of extinguishing the light, where they happen to beunited ; that these qualities succeed each other alternately insuccessive concentric superficies, at distances whichare constantfor the same light, passing through the same medium. From theagreement of the measures, and from the similarity of the phe-nomena, we may conclude, that these intervals are the same asare concerned in the production of the colours of thin plates ;

23

but these are shown, by the experiments of Newton, to be thesmaller, the denser the medium ; and, since it may be presumedthat their number must necessarily remain unaltered in a givenquantity of light, it folhms of course, that light moves moreslowly in a denser, than in a rarer medium ; and this beinggranted, it must be allowed, that refraction is not the effect ofan attractive force directed to a denser medium. The advocatesfor the projectile hypothesis of light, must consider which linkin this chain of reasoning they may judge to be the most feeble;for, hitherto, I have advanced in this Paper no general hypo-thesis whatever. But, since we know that sound diverges inconcentric superficies, and that musical sounds consist of oppo-site qualities, capable of neutralising each other, and succeedingat certain equal intervals, which are different according to thedifference of the note, we are fully authorized to conclude, thatthere must be some strong resemblance between the nature ofsound and that of light.

I have not, in the course of these investigations, found anyreason to suppose the presence of such an inflecting medium inthe neighbourhood of dense substances as I was formerly in-clined to attribute to them ; and, upon considering the pheno-mena of the aberration of the stars, I am disposed to believethat the luminiferous ether pervades the substance of all mate-rial bodies with little or no resistance, as freely perhaps as thewind passes through a grove of trees.

The observations on the effects of diffraction and interferencemay perhaps sometimes be applied to a practical purpose, inmaking us cautious in our conclusions respecting the appear-ances of minute bodies viewed in a microscope. The shadow ofa fibre, however opaque, placed in a pencil of light admittedthrough a small aperture, is always somewhat less dark in themiddle of its breadth than in the parts on each side. A similareffect may also take place, in some degree, with respect to theimage on the retina, and impress the sense with an idea of atransparency which has no real existence : and if a small por-tion of light be really transmitted through the substance, thismay again be destroyed 1.y its interference with the diffractedlight, and produce an appearance of partial opaCity, instead Of

24

uniform semi-transparency. Thus a central dark spot and alight spot, surrounded by a darker circle, may respectively beproduced in the images of a semi-transparent and an opaquecorpuscle ; and impress us with an idea of a complication ofstructure w Nell does not exist. In order to detect the fallacy,we may make two or three fibres cross each other, and viewa number of globules contiguous to each other ; or we mayobtain a still more effectual remedy by changing the magnifyingpower ; and then, if the appearance remain constant in kind andin degree, we may be assured that it truly represents the natureof the substance to be examined. It is natural to inquirewhether or no the figures of the globules of blood, delineatedby Mr. Hewson in the Phil. Trans., Vol. LXIII. for 1773,might not in some measure have been influenced by a decep-tion of this kind : but, as far as I have hitherto been able toexamine the globules, with a lens of one-fiftieth of an inchfocus, I have found them nearly such as Mr. Hewson hasdescribed them.

V.Remarks on the Colours of Natural Bodies.Exper. 5. I have already adduced, in illustration of New-

ton's comparison of the colours of natural bodies with those ofthin plates, Dr: Wollastou's observations on the blue lightof the lower part of a candle, which appears, when viewedthrough a prism, to be divided into five portions. I have latelyobserved a similar instance, still more strongly marked, in thelight transmitted by the blue glass sold by the opticians. Thislight is separated by the prism into seven distinct portions,nearly equal in magnitude, but somewhat broader, and lessaccurately defined, towards the violet end of the spectrum.The first two are red, the third is yellowish green, the fourthgreen, the fifth blue, the sixth bluish violet, and the seventhviolet. This division agrees very nea y with that of the lightreflected by a plate of air Taiga of an inch in thickness, cor-responding to the llth series of red, and the 18th of violet.A similar plate of a metallic oxide would perhaps be aboutyr17.7a of an inch in thickness. But it must be confessed thatthere are strong reasons for believing the colouring particles of

25

natural bodies in general to be incomparably smaller than this ;and it is probable that the analogy suggested by Newton issomewhat less close than he imagined. The light reflectedby a plate of air, at any thickness nearly corresponding to the11th red, appears to the eye to be very nearly white ; but,under favourable circumstances, the 1.1th red and the neigh-bouring colours may still be distinguished. The light of somekinds of coloured glass is pure red ; that of others red with alittle green : some intercept all the light, except the extremered and the blue. In the blue light of a candle, expanded bythe prism, the portions of each colour appear to be narrower,and the intervening dark spaces wider than in the analogousspectrum flerived from the light reflected from a thin plate.The light of burning alcoh,,, appears to 'be green and violetonly. The pink dye sold in the shops, which is a preparationof the carthamus, affords a good specimen of a yellow greenlight regularly reflected, and a crimson probably produced bytransmission.

VI.Experiment on the Dark Rays of Ritter.

Exper. 6. The existence of solar rays accompanying light,more refrangible than the violet rays, and cognisable by theirchemical effects, was first ascertained by Mr. flitter; but Dr.Wollaston made the same experiment: a very short time after-wards, without having been informed of what had been doneon the Continent. These rays appear to extend beyond theviolet rays of the prismatic spectrum, through a space nearlyequal to that which is occupied by the violet. In order tocomplete the comparison of their properties with those of visiblelight, I was desirous of examining the effect of their reflectionfrom a thin plate of air, capable of producing the well-knownrings of colours. For this purpose I formed an image of therings, by means of the solar microscope, with the apparatuswhich I have described in the Journals of the Royal Institution,and I threw this image on paper dipped in a solution of nitrateof silver, placed at the distance of about nine inches from themicroscope. In the course of an hour portions of three darkrings were very distinctly visible, much smaller than the

26

brightest rings of the coloured image, and coinciding verynearly, in their dimensions, with the rings of violet light thatappeared upon the interposition of violet glass. I thought thedark rings were a little smaller than the violet rings, but thedifference was not sufficiently great to be accurately ascer-tained ; it might be as much as 71,7 or -15- of the diameters, butnot greater. It is the less surprising that the difference shouldbe so small, as the dimensions of the coloured rings do not byany means vary et the violet end of the spectrum so rapidly asat the red end. For performing this experiment with verygreat accuracy a heliostate would be necessary, since themotion of the sun causes a slight change in the place of theimage ; and leather, impregnated with the tnuriate of silver,would indicate the effect with greater delicacy. The experi-ment, however, in its present state, is sufficient to complete theanalogy of the invisible with the visible rays, and to show thatthey are equally liable to the general law which is the principalsubject of this Paper. If we had thermometers sufficientlydelicate, it is probable that we might acquire, by similar means,information still more interesting, with respect to the rays ofinvisible heat discovered by Dr. Herschel ; but at presentthere is great reason to doubt of the practicability of such anexperiment.

27

Variation within a Sphere, No. 10: The Sun.Sculptural construction of gold wire, 22 feet long, 11 feethigh, 5-1/2 feet deep. By Richard Lippold, American sculptor.The Metropolitan Museum of Art, N.Y.C.

ry

Velocity of Light

A. A. Michelson

A chapter from his book, Studies in Optics, published in 1927.

The velocity of light is one of the most important ofthe fundamental constants of Nature. Its measurementby Foucault and Fizeau gave as the result a speed greaterin air than in water, thus deciding in favor of the undu-latory and against the corpuscular theory. Again, thecomparison of the electrostatic and the electromagneticunits gives as an experimental result a value remarkablyclose to the velocity of lighta result which justifiedMaxwell in concluding that light is the propagation of anelectromagnetic disturbance. Finally, the principle ofrelativity gives the velocity of light a still greater im-portance, since one of its fundamental postulates is theconstancy of this velocity under all possible condi-tions.

The first attempt at measurement was due to Galileo.Two observers, placed at a distance of several kilometers,are provided with lanterns which can be covered or un-covered by a movable screen. The first observer uncovershis light, and the second observer answers by uncoveringhis at the instant of perceiving the light from the first.If there is an interval between the uncovering of thelantern by the first observer and his perception of thereturn signal from the second (due allowance being madefor the delay between perception and motion), the dis-tance divided by the time interval should give the velocityof propagation.

29

Needless to say, the time interval was far too smallto be appreciated by such imperfect appliances. It isnevertheless worthy of note that the principle of themethod is sound, and, with improvements that are almostintuitive, leads to the well-known method of Fizeau. Thefirst improvement would clearly be the substitution of amirror instead of the second observer. The second wouldconsist in the substitution of a series of equidistant aper-tures in a rapidly revolving screen instead of the singlescreen which covers and uncovers the light.

The first actual determination of the velocity of lightwas made in 1675 by Romer as a result of his observationof the eclipses of the first satellite of Jupiter. Theseeclipses, recurring at very nearly equal intervals, could becalculated, and Romer found that the observed and thecalculated values showed an annual discrepancy. Theeclipses were later by an interval of sixteen minutes andtwenty-six seconds' when the earth is farthest fromJupiter than when nearest to it. Romer correctly attrib-uted this difference to the time required by light to trav-erse the earth's orbit. If this be taken as 300,000,000 kilo-meters and the time interval as one thousand seconds,the resulting value for the velocity of light is 300,000kilometers per second.

Another method for the determination of the velocityof light is due to Bradley, who in 1728 announced anapparent annual deviation in the direction of the fixedstars from their mean position, to which he gave the name"aberration." A star whose direction is at right angles tothe earth's orbital motion appears displaced in the direc-

t The value originally gtv en by Romer, twenty-two minutes, is clearlytoo great.

30

tion of motion by an angle of 2e445. This displacementBradley attributed to the finite velocity of light.

With a telescope pointing in the true direction of sucha star, during the time of passage of the light from ob-jective to focus the telescope will have been displaced inconsequence of the orbital motion of the earth so that theimage of the star falls behind the crosshairs. In order toproduce coincidence, the telescope must be inclined for-ward at such an angle a that the tangent is equal 4.o theratio of the velocity v of the earth to the velocity of light,

vtan a=V '

or, since v=rD/T, where D is the diameter of the earth'sorbit and T the number of seconds in the year,

rDtan a= VT '

from which the velocity of light may be found; but, as isalso the case with the method of Romer, only to the de-gree of accuracy with which the sun's distance, ID, isknown; that is, with an order of accuracy of about 1 percent=

In 1849 Fizeau announced the result of the first ex-perimental measurement of the velocity of light. Twoastronomical telescope objectives Lz and L3 (Fig. 73) areplaced facing each other at the two stations. At the focusof the first is an intense but minute image a of the sourceof light (arc) by reflection from a plane-parallel plate N.

: The value of the velocity of light has been obtained, by experimentalmethods immediately to be described, with an order of accuracy of one inone hundred thousand, so that now the process is inverted, and this re-sult is employed to find the sun's distance.

31

The light from this image is rendered approximatelyparallel by the first objective. These parallel rays, fallingon the distant objective, are brought to a focus at the sur-face of a mirror, whence the path is retraced and animage formed which coincides with the original image a,where it is observed by the ocular E. An accurately

divided toothed wheel W is given a uniform rotation,thus interrupting the passage of the light at a. If, onreturning, the light is blocked by a tooth, it is eclipsed, toreappear at a velocity such that the next succeedinginterval occupies the place of the former, and so on.

If n is the number of teeth and N the number of turnsper second, K the number of teeth which pass during thedouble journey of the light over the distance D,

V =2NnD

K

It is easier to mark the minima than the maxima of in-tensity, and accordingly

K 2PI3

32

I

if p is the order of the eclipse. Let SK be the error com-mitted in the estimate of K (practically the error in esti-mation of equality of intensities on the descending andthe ascending branches of the intensity curve). Then

&V _8KV K

Hence it is desirable to make K as great as possible. InFizeau's experiments this number was 5 to 7, and shouldhave given a result correct to about one three-hundredth.It was, in fact, about 5 per cent too large.

A much more accurate determination was undertakenby Cornu in 1872 in which K varied from 3 to 2 1 , the re-sult as given by Cornu being 300,400, with a probable er-ror of one-tenth of r per cent. In discussing Cornu's re-sults, however, Listing showed that these tended toward asmaller value as the speed increased, and he assigns thislimit as the correct value, namely, 299,950. Perrotin,with the same apparatus, found 299,900.

Before Fizeau had concluded his experiments, anotherproject was proposed by Arago, namely, the utilizationof the revolving mirror by means of which Wheatstonehad measured the speed of propagation of an electriccurrent. Arago's chief interest in the problem lay in thepossibility of deciding the question of the relative veloci-ties in air and water as a crucial test between the undula-tory and the corpuscular theories. He pointed oat,however, the possibility of measuring the absolute ve-locity.

The plan was to compare the deviations of the lightfrom an electric spark reflected directly from the revolving

33

mirror with that which was reflected after traversing aconsiderable distance in air (or in water). The difficultyin executing such an experiment lay in the uncertainty inthe direction in which the two reflected images of thespark were to appear (which might be anywhere in 3600).This difficulty was solved by Foucault in 1862 by thefollowing ingenious device whereby the return light is

FIG. 74

always reflected in the same direction (apart from thedeviation due to the retardation which it is required tomeasure), notwithstanding the rotation of the mirror.

Following is the actual arrangement of apparatus bywhich this is effected. Light from a source S falls uponan objective L, whence it proceeds to the revolving mir-ror R, and is thence reflected to the concave mirror C(whose center is at R), where it forms a real image of thesource. It then retraces its path, forming a real imagewhich coincides with the source even when the revolvingmirror is in slow motion. Part of the light is reflected fromthe plane-parallel glass M, forming an image at a whereit is observed by the micrometer eyepiece E.

34

If now the revolving mirror is turning rapidly, thereturn image, instead of coinciding with its original posi-tion, will be deviated in the direction of rotation throughan angle double that through which the mirror turns whilethe light makes its clout ,.ansit. If this angle is a andthe distance between mix, rs is D, and the revolvingmirror makes N turns per second,

2Da= orN 'or

y=4.71VDa

In principle there is no essential difference betweenthe two methods. In the method of the toothed wheel theangle a corresponds to the passage of K teeth, and istherefore a= 2irK/ti, so that the formula previouslyfound,

V =2 NnKD

, now becomes V 411.ND the same as for thearevolving mirror. The latter method has, however, thesame advantage over the former that the method of mir-ror and scale has over the direct reading of the needle ofa galvanometer.

On the other hand, an important advantage for themethod of the toothed wheel lies in the circumstancethat the intensity of the return image is one-half of thatwhich would appear if there were no toothed wheel,

ni3whereas with the revolving mirror this fraction isrD

if the mirror has n facets), where g is the angular apertureof the concave mirror, and f is the focal length of themirror, r is the distance from slit to revolving mirror,and D is the distance between stations.

35

In the actual experiments of Foucault, the greatestdistance D was only 20 m (obtained by five reflectionsfrom concave mirrors), which, with a speed of five hun-dred turns per second, gives only i6o" for the angle 2awhich is to be measured. The limit of accuracy of themethod is about one second, so that under these condi-tions the results of Foucault's measurements can hardlybe expected to be accurate to one part in one hundred andsixty. Foucault's result, 298,000, is in fact too small bythis amount.'

In order to obtain a deflection 2a sufficiently large tomeasure with precision it is necessary to work with amuch larger distance. The following plan renders thispossible, ar d in a series of experiments (1878) the dis-tance D w.s about loo m and could have been made muchgreater.

The image-forming lens. in the new arrangement isplaced between the two mirrors, and (for maximum in-tensity of the return image) at a distance from the re-volving mirror equal to the focal length of the lens. Thisnecessitates a lens of long focus; for the radius of meas-urement r (from which' a is determined by the relation8 =r tan a, in which 8 is the measured displacement of the

image) is given by r =f2

, if f is the focal length of the lens;

whence r is proportional to fs., In the actual experiment,

Apart from the mere matter of convenience in limiting the distanceD to the insignificant 20 m (on account of the dimensions of the labora-tory), it may be that this was in fact limited by the relative intensity ofthe return image as compared with that of the streak of light caused bythe direct reflection from the revolving mirror, which in Foucault'sexperiments was doubtless superposed on the former. The intensity ofthe return image varies inversely as the cube of the distance, wholethat of the streak remains constant.

36

a non-achromatic lens of 25-m focus and 20-CM diameterwas employed, and with this it was found that the in-tensity of the return light was quite sufficient even whenthe revolving mirror was far removed from the principalfocus.

With so large a displacement, the inclined plane-paral-lel plate in the Foucault arrangement may be suppressed,the direct (real) image being observed. With 250 to 300turns per second, a displacement of ioo to iso mm wasobtained which could be measured with an error of lessthan one ten - thousandth.

The measurement of D presents no serious difficulty.This was accomplished by means of a steel tape whosecoefficient of stretch and of dilatation was carefully deter-mined, and whose length under standard conditions wascompared with a copy of the standard meter. The esti-mated probable error was of the order of 1: 200,000.

The measurement of the speed of rotation presentssome points of interest. The optical "beats" between therevolving mirror and an electrically maintained tuningfork were observed at the same time that the coincidenceof the deflected image with the crosshairs of the eyepiecewas maintained by hand regulation of an air blast whichactuated the turbine attached to the revolving mirror.The number of vibrations of the fork plus the number ofbeats per second gives the number of revolutions persecond in terms of the rate of the fork. This, however,cannot be relied upon except for a short interval, andit was compared before and after every measurementwith a standard fork. This fork, whose temperature co-efficient is well determined, is then compared, as follows,directly with a free pendulum.

37

For this purpose the pendulum is connected in serieswith a battery and the primary of an induction coil whosecircuit is interrupted by means of a platinum knife edgeattached to the pendulum passing through a globule ofmercury. The secondary of the induction coil sends aflash through a vacuum tube, thus illuminating the edgeof the fork and the crosshair of the observing microscope.If the fork makes an exact whole number (256) of vibra-tions during one swing of the pendulum, it appears atrest; but if there is a slight excess, the edge of the forkappears to execute a cycle of displacement at the rate of nper second. The rate of the fork is then N*n per secondof the free pendulum. This last is finally compared witha standard astronomical clock' The order of accuracyis estimated as i : 200,000.

The final result of the mean cf two such determina-tions of the velocity of light made under somewhat similarconditions but at a different time and locality is 299,895.

A determination of the velocity of light by a modifica-tion of the Foucault arrangement was completed byNewcomb in 1882. One of the essential improvementsconsisted in the use of a revolving steel prism with squaresection twice as long as wide. This permits the sendingand receiving of the light on different parts of the mirror,thus eliminating the effect of direct reflection. It shouldalso be mentioned that very accurate means were pro-vided for measuring the deflection, and finally that thespeed of the mirror was registered on a chronographthrough a system of gears connected with the revolvingmirror. Newcomb's result is 299,860.

, The average beat of such a clock may be extremely constant al-though the individual "seconds" vary considerably.

38

The original purpose of the Foucault arrangement wasthe testing of the question of the relative velocities of lightin air and in water. For this purpose a tube filled withwater and closed with plane-parallel glasses is interposed.There are then two return images of the source whichwould be superposed if the velocities were the same. Byappropriately placed diaphragms these two images maybe separated, and if there is any difference in velocitiesthis is revealed by a relative displacement in the directionof rotation. This was found greater for the beam whichhad passed through the water column, and in which,therefore, the velocity must have been less. This resultis in accordance with the undulatory theory and opposedto the corpuscular theory of light.

The experiments of Foucault do not appear to haveshown more than qualitative results, and it should be ofinterest, not only to show that the velocity of light is lessin water than in air, but that the ratio of the velocitiesis equal to the index of refraction of the liquid. Experi-ments were accordingly undertaken with water, the re: ultobtained agreeing very nearly with the index of refrac-tion. But on replacing the water by carbon disulphide,the ratio of velocities obtained was 1.75 instead of 1.64,the index of refraction. The difference is much too greatto be attributed to errors of experiment.

Lord Rayleigh found the following explanation of thediscrepancy. In the method of the toothed wheel the dis-turbances are propagated in the form of isolated groupsof wave-trains. Rayleigh finds that the velocity of agroup is not the same as that of the separate waves ex-cept in a medium without dispersion. The simplest formof group analytically considered is that produced by two

39

simple harmonic wave-trains of slightly different fre-quencies and wave-lengths. Thus, let

y= cos (nt mx)+ cos (nitmix) ,

in which n= 27IT, and m = 27/X, T being the period andX the wave-length. Let n n. = an, and m --m. =am.Then

y=2 cos Rani aMX) cos (ntmx) .

This represents a series of groups of waves such as illus-trated in Figure 75.

1FIG. 75

The velocity of the waves is the ratio V=n/m, butthe velocity of the group (e.g., the velocity of propagationof the maximum or the minimum) will be

V' = an/am ,or, since n =mV,

v,:a(V) , _av ..._vji_,_manam -rmam i \-1- yam)

or, since m= 27/X,

r=v(IT-7TO .xav

The demonstration is true, not only of this particularform of group, but (by the Fourier theorem) can be ap-plied to a group of any form.

40

It is not quite so clear that this expression applies tothe measurements made with the revolving mirror. LordRayleigh shows that in consequenCe of the Doppler effectthere is a shortening of the waves at one edge of thebeam of light reflected from the revolving mirror and alengthening at the opposite edge, and since the velocityof propagation depends on the wave-length in a dispersivemedium, there will be a rotation of the individual wave-fronts.

If co is the angular velocity of the mirror, and co= thatof the dispersional rotation,

dV dV dX(0.=dy dX dy '

where y is the distance from the axis of rotation. But

dX X X dVdy-2cov . . oh= 207

The deflection actually observed is therefore

T(2col-w,) ,

where T is the time required to travel distance 2D; or

D4 1+X_dV\Vw\ VdXf'

hence the velocity measured is

/ X dV\VH=V÷Y+V 70'

or, to small quantities of the second order,

V" =V' = group velocity .1

J. W. Gibbs (Nature, 2886) shows that the measurement is in realityexactly that of groups and not merely an approximation.

hr '4# Lt.,{ t

41

The value of (1 -F-dil ) for carbon disulphide for

X

dX

the mean wave-length of the visible spectrum is 0.93..

Accordingly,V °__ V° I r :64V' V o.93 O. 93 I' ' I

which agrees with the value found by experiment.

RECENT MEASUREMENTS OF THE VELOCITY OF LIGHT

In the expression for V, the velocity of light as de-termined by the revolving mirror, V = 471-ND/ a, there arethree quantities to be measured, namely, N, the speed ofthe mirror; D, the distance between stations; and a, theangular displacement of the mirror. As has already beenmentioned, the values of N and D may be obtained toone part in one hundred thousand or less. But a cannotbe measured to this order of accuracy. It has been pointedout by Newcombe that this difficulty may be avoided bygiving the revolving mirror a prismatic form and makingthe distance between the two stations so great that thereturn light is reflected z.t the same angle by the next fol-lowing face of the prism

The following is an oltline of a proposed attempt torealize such a project between Mount Wilson and MountSan Antonio near Pasadena, the distance being about35 km. For this, given a speed of rotation of 1,o6o turnsper second, the angular displacement of the mirror duringthe double journey would be 900; or, if the speed werehalf as great, an angle of 45° would suffice' Accordingly,

= Measures of the Velocity of Light. Nautical Almanac Office, 1882.2 It may be noted that with eight surfaces the resulting intensity will

be four times as great as with the revolving plane-parallel disk.

42

the revolving mirror may have the form of an octagon.It is, of course, very important that the angles should be

equal, at least to the orderof accuracy desired.

This has already beenattained as follows. Theoctagon, with faces pol-ished and angles approxi-mately correct, is appliedto the test angle (IV madeup of a 45° prism ce-mented to a true plane.The faces bib are made

FIG. 76 parallel by the interfer-ence fringes observed in

monochromatic light. In general, the faces a,a will not beparallel, and the angle between them is measured by thedistance and inclination of the interference bands. Thesame process is repeated for each of the eight angles, andthese are corrected by repolishing until the distance andinclination are the same for all, when the correspondingangles will also be equal. It has been found possible inthis way to produce an octagon in which the averageerror was of the order of one-millionth, that is, aboutone-tenth to one-twentieth of a second!

Another difficulty arises from the direct reflection andthe scattered light from the revolving mirror. The formermay be eliminated, as already mentioned, by slightly

= It may be noted that while a distortion may be expected when theminor is in such rapid rotation, if the substance of the mirror (glass, inthe present instance) is uniform, such distortion could only produce a veryslight curvature and hence merely a minute change of focus.

V, 14 q* 1. , 0 i L .,1* *

43

inclining the revolving mirror, but to avoid the scatteredlight it is essential that the return ray be received on adifferent surface from the outgoing.

Again, in order to avoid the difficulty in maintainingthe distant mirror perpendicular to the incident light, thereturn of the ray to the home station may be accom-

lioxe

'zkrei

IIr

77.Light path a, b, c, d, e, h, e, f, g, h, i, j

plished exactly as in the Fizeau experiment, the only pre-caution required being the very accurate focusing of thebeam on the small plane (better, concave) mirror at thefocus of the distant collimator.

Finally, it is far less expensive to make both sendingand receiving collimators silvered mirrors instead oflenses.

In Figure 77 is shown the arrangement of apparatuswhich fulfilled all these requirements.

44

Three determinations were undertaken between thehome station at the Mount Wilson Obs "rvatory andMount San Antonio 22 miles distant. The rate of theelectric tuning fork was 132.25 vibrations per second,giving four stationary images of the revolving mirrorwhen this was rotating at the rate of 529 turns per second.The fork was compared before and after every set of theobservations with a free pendulum whose rate was foundby comparison with an invar pendulum furnished andrated by the Coast and Geodetic Survey.

The result of eight measurements in 1924 gave

Va= 299,735

Another series of observations with a direct compari-son of the same electric fork with the Coast and GeodeticSurvey pendulum' was completed in the summer of 1925with a resulting value

174=- 299,690 .

A third series of measurements was made in which theelectric fork was replaced by a free fork making 528 vibra-tions per second maintained by an "audion circuit," thusinsuring a much more nearly constant rate. The resultof this measurement gave

Va= 299,704

Giving these determinations the weights r, 2, and 4,respectively, the result for the velocity in air is

V=-- 299,704

T This comparison was made by allowing the light froma very narrowslit to fall on a mirror attached to the pendulum. An image of the slit wasformed by means of a good achromatic lens, in the plane ofone edge of thefork, where it was observed by an ordinary eyepiece.

45

Table VIII shows the more reliable results of measure-ments of V with distance between stations, method used,and the weight assigned to each.

TABLE VIII

Author D Method! IS t. V

Cornu . , . 23 km Toothed wheel I 299,990Perrotin... ... 12 Toothed wheel 1 299,900M, and M,. o.6 Rev. mirror 1 299,88oNewcomb*. . 6.5 Rev. mirror 3 299,810M3 35 Rev. mirror 5 299,800

*Newcomb's value omitting all discordant observations was 298,86o.

47

4.

1

t

a

,r!

90411MMI.

ti

Lincoln Center for the Performing Arts, New York City.

48

4

b ..., Popular Applications of Polarized Light

William A. Shurcliff and Stanley S. Ballard

A chapter from the book Polarized Light published in 1964.

If there is a logical order in which the various applications ofpolarizers and polarized light should be considered, the authorshave never discovered it. The policy adopted here is to considerthe most popular and "humanistic" applications first, and themore scientific and esor .:c applications last.

POLARIZATION AND THE HUMAN EYE

The most humanistic fact about polarization of light is thatit can be detected directly by the naked eye. Nearly anyone, iftold carefully what to look for, can succeed in this. Sometimes hecan even determine the form and azimuth or polarization.

What the observer actually "sees" is a certain faint patternknown as Haidinger's brush and illustrated in Fig. 10.1. Thebrush is so faint and illdefined that it will escape notice finlessthe field of view is highly uniform: a clear blue sky makes anideal background, and a brightly illuminated sheet of whitepaper is nearly as good. The be:,t procedure for a beginner is tohold a linear polarizer in front of his eye, stare fixedly throughit toward a clear blue sky, and, after five or ten seconds, sud-denly turn the polarizer through 90°. Immediately the brush isseen. It fades away in two or three seconds, but reappears if thepolarizer is again turned through 90°. The brush itself is sym-

.

BLUE a.41...,...... U ....%NI*

YELLOW YELLOW:t:.:: 41:'. BLUE

ABOUT 3 DEGREES-9

FIG. 10-1 Approximate appearance of Haidinger's brush when thevibration direction of the beam is vertical.

metric, double-ended, and yellow in color; it is small, subtendingan angle of only about 2° or 3°. The adjacent areas appear blue,perhaps merely by contrast. The long axis of the brush is approx-imately perpendicular to the direction of electric vibration in thelinearly polarized beam, i.e., perpendicular to the transmissionaxis of the polarizer used.

Circular polarization, too, can be detected directly by eye, andeven the handedness can be determined. When an observer fac-ing a clear blue sky places a right circular polarizer in front ofhis eye, he sees the yellow brush and finds that its long axis hasan upward-to-the-right, downward-to-the-left direction, i.e., anazimuth of about +45°. This is true, of course, irrespective of theorientation of the polarizer, since a circle has no top or bottom.If he employs a left circular polarizer, he finds the brush to havea 45° orientation. In each case the pattern fades away rapidly,but can be restored to full vigor by switching to a polarizer ofopposite handedness. Instead of using a circular polarizer theobserver can use a single linear polarizer in series with a 90° re-tarder, the latter being held nearer to the eye. Turning theretarder through 90° reverses the handedness of the circularpolarization.

Some people see the brush easily; others have difficulty. A few

Pir,11,1r i

see the brush when looking innocently at the partially polarizedblue sky, i.e., without using any polarizer at all, and even with-out meaning to see the brush. Some people see the brush moredistinctly by linearly polarized light than by circularly polarizedlight, and for others the reverse is true. An observer may findthe brush to have a slightly different orientation depending onwhich eye is used.

The spectral energy distribution of the light is important. Ifthe light is rich in short-wavelength (blue) radiation, the brushis very noticeable, but if the short-wavelength radiation is elimi-nated by means of a yellow filter, the brush fails to appear. Useof a blue filter tends to accentuate the brush.

Although the phenomenon was discovered in 1844, by theAustrian mineralogist Haidinger, the cause is not yet fully under-stood. Presumably the thousands of tiny blue-light-absorbingbodies in the central (foveal) portion of the retina are dichroicand are oriented in a radial pattern, for example, a pattern suchthat the absorption axis of each body lies approximately alonga radius from the center of the fovea. Incident linearly polarizedlight will then be absorbed more strongly in some parts of thepattern than in other parts and consequently some parts willfatigue more than others. When the vibration direction of thelight is suddenly changed, the varying degrees of fatigue arerevealed as a subjective radial pattern. Presumably no suchdichroism or orientation pattern applies to longer wavelength(yellow and red) light; consequently a yellow sensation domi-nates in those regions where fatigue-to-blue has occurred.

The fact that circular polarization, also, may be detected per-haps implies that some transparent portion of the eye is weaklybirefringent and acts like a retarder, converting circularly polar-ized light to linearly or elliptically polarized light. The direc-tion of the major axis of the ellipse depends only on the direc-tion of the fast axis of the retarding layer and hence remainsfixedunless the observer tips his head.

Perhaps physicists will some day write matrices to describe theretarding layers and dichroic areas of the eye. Poets were the firstto see magic fire and jewels in the human eye; physicists will bethe first to see matrices!

51

Bees, too, can detect the vibration direction of linearly polar-ized light. The experiments of the biologist K. von Frisch duringWorld War II showed that bees "navigate" back and forth be-tween hive and source of honey by using the sun as a guide.More interesting, when the sun is obscured by a large area ofclouds the bees can still navigate successfully if they can see abit of blue sky: they can detect the azimuth of linear polariza-tion of the blue light and navigate with respect to it. One way ofdemonstrating the bee's ability to detect the azimuth of polariza-tion is to place the bee in a large box the top of which consistsof a huge sheet of linear polarizer, such as H-sheet. Each time theexperimenter turns the polarizer to a different azimuth, the beechanges his direction of attempted travel correspondingly.

Certain other animals also can detect the polarization of sky-light and navigate by it. This includes ants, beetles, and thefruit fly Drosophila. Probably many other examples will be dis-covered.

POLARIZATION OF SKY LIGHT

Blue-sky light traveling in a direction roughly at right anglesto the sun's rays is partially polarized. When an observer holdsa linear polarizer in front of his eye and gazes in a directionperpendicular to the direction of the sun, he finds that rotatingthe polarizer slowly causes the sky to change from bright to darksuccessively. The degree of polarization of sky light may reach70 or 80 percent when the air is clear and dust-free, the sun ismoderately low in the sky, and the observation direction is nearthe zenith.

The polarization is a result of the scattering of the sun's raysby the molecules in the air. Rayleigh's well-known inverse-fourth-power law relating scattering intensity to wavelength accountsfor the blue color of the scattered light, and the asymmetry as-sociated with the 90° viewing angle accounts for the polarization,as explained in Chapter 5. Some multiple scattering occurs, andthis reduces the degree of polarization somewhat; when theobserver ascends to a higher altitude, the amount of air involved

52

is reduced, multiple scattering is reduced, and the degree ofpolarization is increased. A further increase results when ayellow or red filter is used to block the short-wavelength com-ponent of the light and transmit the long-wavelength componentthe latter component is less subject to multiple scattering. (Thesituation is very different for infrared radiation of wavelengthexceeding 2 microns: much of this radiation is produced byemission from the air itself, rather than by scattering, and thisexhibits little or no polarization.)

Some persons are capable of detecting the polarization of skylight directly by eye, by virtue of the Haidinger brush phe-nomenon discussed in a preceding section; a few individualsfind the brush noticeable enough to be a nuisance. Ordinarily,of course, it escapes notice and plays little part in the affairs ofman. Its practical use by bees, ants, etc., has been indicated, andthe importance to photographers is discussed in a later section.

POLARIZATION OF LIGHT UNDER WATER

A surprising fact about the polarization found in light presentbeneath the surface of the ocean (or of a pond) is that the pre-dominant direction of electric vibration is horizontal. The oppo-site might be expected, since most of the light that enters thewater enters obliquely from above, and the most strongly re-flected component of obliquely incident light is the horizontallyvibrating component. But oceanographers and biologists, work-ing at depths of 5 to 30 feet in waters off Bermuda and in theMediterranean Sea, have found the main cause of submarinepolarization to be the scattering of the light by microscopicparticles suspended in the water. Sunlight and sky light enterthe water from above, and the average direction of illuminationis roughly vertical; consequently the polarization form of thescattered light that travels horizontally toward an underwaterobserver is partially polarized with the electric vibration direc-tion horizontal. The situation is much the samo as that discussedin Chapter 5, except that the incident light has a more steeplydownward direction and the asymmetric scattering is by micro-scopic particles instead.of molecules.

Typically, the degree of polarization is 5 to 30 percent, an

53

amount found to be important to a variety of underwater life.The water flea Daphnia tends to swim in a direction perpendicu-lar to the electric vibration direction, for reasons not yet known.When tests are conducted in a tank filled with water that is freeof suspended particles, so that the submarine illumination ispractically unpolariied, Daphnia ceases to favor any one direc-tion. But if suspended matter is added, thus restoring the polar-ization, Daphnia resumes the custom of traveling perpendicularto the vibration direction.

The arthropod Limufits (horse shoe crab) easily detects thepolarization of the underwater light and is presumed to navigatewith respect to the electric vibration direction. The same is trueof the crustacean Mysidium gracile and various other forms ofmarine life. Most tend to swim perpendicularly to the vibrationdirection; some swim parallel to it; a few swim at different rela-tive orientations depending on the time of day. For all of theseanimals, polarization is a compass that works even under water!

POLARIZING SUNGLASSES

The lenses of ordinary sunglasses employ absorbing materialsthat are isotropic, and accordingly the incident light is attenuatedby a fixed factor irrespective of polarization form. This is un-fortunate. The fact is that "glare" consists predominantly oflight having a horizontal vibration direction. Why? For thesereasons:

(a) The main source of light (sun and sky) is overhead, andconsequently the main flux of light is downward.

(b) The surfaces that are most strongly illuminated by thedownward flux are horizontal surfaces.

(c) Such surfaces are usually viewed obliquely, since a personseldom looks straight down.

(d) Most outdoor objects are of dielectric material.(e) Light reflected obliquely from a horizontal dielectric stir -

face is partially linearly polarized , ;th the dominant vibrationdirection horizontal, as explained in Chapter 4.

Polarizing sunglasses take full advantage of this fact. Thelenses are made of dichroic material (H-sheet, usually) orientedwith the transmission axis vertical, as indicated in Fig. 10-2a, so

54

(c)

FIG. 10-2 Three types of polarizing spectacles. In (a) the transmissionaxis is vertical, for eliminating glare reflected from horizontal surfaces.In (b) the axis is horizontal, for eliminating reflections from verticalwindows of trains, store-fronts (show-windows), etc. In (c) the axis di-rections are 45' and 45°, a standard arrangement used in viewing

polarization-coded stereoscopic pictures.

that almost all of the horizontal vibrations are absorbed. Thecomponent having vertical vibration direction is transmitted.Usually some isotropic absorber is included in the lenses toabsorb ultraviolet light strongly and blue and red light to amoderate extent; the sunglasses then have a greenish hue whichhas nothing to do with the polarization.

Motorists and vacationists find that polarizing sunglasses arehelpful not only in reducing the brightness of the field of viewas a whole, but also in enhancing the beauty of the scene. Be-cause specularly reflected light is absorbed preferentially, roads,trees, grassy fields, etc., appear softer and more deeply coloredthrough polarizers. Specular ly reflected light tends to veil nature'sinherent beauty; polarizing sunglasses remove the veil. -

Fishermen and boatsmen enjoy another benefit from wearingpolarizing sunglasses. They want to be able to see fish, rocks, etc.,beneath the surface of the water, yet the light from such objectsis dim and is usually lost in the "noise" of the sky light reflectedobliquely from the surface. Since the reflected light is highlypolarized with horizontal vibration direction, the polarizingsunglasses absorb this component strongly, and the visibility of

55

the underwater objects is greatly increased. The increase is great-est when the viewing direction corresponds to the polarizingangle, which, for water, is about 53° from the normal. When theviewing direction is along the normal, i.e., straight down, thereis no increase at all.

There is one interesting situation in which polarizing sun-glasses produce little increase in visibility of underwater objectseven when the angle of viewing is the polarizing angle. This situ-ation occurs when the sky is clear and blue, the sun is low inthe sky, and the pertinent portion of the sky is at 90° from thedirection of the sun. Under these circumstances the light strikingthe water is already linearly polarized at such an azimuth thatalmost none of it is reflected. There is no task left for the sun-glasses to performthere is no reflected glare to suppress. Theunderwater objects are seen with great clarity. Persons unfamiliarwith the polarization of sky light and with the dependence ofoblique reflection on polarization form are likely to ascribe theremarkable clarity to "especially clear water" rather than toabsence of reflection.

CAMERA FILTERS'

Photographers often wish to enhance the contrast between bluesky and white clouds. Thirty years ago they did this by employinga yellow filter, which absorbed most of the blue light from theclear sky but transmitted most of the white light from the clouds.Using ordinary black-and-white film, they obtained excellentcontrast by this method. Today, photographers are using colorfilm increasingly, and the use of yellow filters is no longer per-missible since it would eliminate all blue colors from the finishedphotograph.

The only known solution is to exploit the difference in polari-zation between blue sky and white clouds. Light from most por-tions of the blue sky is partially linearly polarized, as explainedin a preceding section, and light from clouds is unpolarized.Therefore a neutral color, linear polarizer mounted at the opti-mum azimuth in front of the lens will absorb a large fraction(e.g., 80 percent) of the sky light while transmitting a large frac-

56

tion (nearly half) of the light from the clouds; thus the contrastis increased by a factor of two or three. The factor is less if theair is hazy, and more if the air is extremely clear (as in Arizona)and if the camera is aimed about 90° from the direction of thesun.

The usual way of choosing the azimuth of the polarize: iscrude, but perhaps adequate. The photographer holds the polar-izer in front of his eye, finds by trial and error which azimuthmaximizes the contrast of the clouds in question, and then at-tempts to mount the polarizer on the camera without changingthe azimuth of the polarizer. One type of polarizing filter forcameras is equipped with a small "satellite" polarizer mountedat the end of a short arm and aligned permanently with the mainpolarizer. The photographer installs the main polarizer in frontof the lens, looks through the small polarizer and turns the armto whatever azimuth maximizes the contrast. Both polarizersthen have this optimum orientation. The satisfactoriness of theazimuth can be checked visually at any time. Instead of usingthese empirical methods, a scientifically minded photographercan proceed by dead reckoning, i.e., by following this well-knownrule: Mount the polarizer so that its transmission axis lies in theplane determined by camera, sun, and object photographed. (Sooriented, the polarizer performs a valuable additional se:vice: iteliminates most of the specularly reflected light from trees, roads,etc., and enhances the softness and depth of color of the scene.)

When a photographer standing on a sidewalk tries to photo-graph objects situated behind a store window, the reflection ofthe street scene from the window may threaten to spoil thephotograph. An excellent solution is to place the camera off toone side so that the window is seen obliquely at about the polar-izing angle, and mount a linear polarizer in front of the lens;the polarizer is turned so that its transmission axis is horizontal,and the polarized light reflected from the window is absorbed.The authors have a friend who has applied this same principleto a pair of special spectacles he wears while touring the countryby railroad. The lenses consist of polarizers oriented with thetransmission axis horizontal, as indicated in Fig. 10.2b; thuswhen he gazes out of the train window in oblique forward direc-

57

tion, the reflected images of passengers and newspapers are wiped

out, and the scenery appears in its pristine glory.

USE OF CIRCULAR POLARIZERS IN ELIMINATINGPERPENDICULARLY REFLECTED LIGHT

Eliminating perpendicularly reflected lights is a different prob-

lem from that of eliminating obliquely reflected light. The proc-

ess of oblique reflection at Brewster's angle causes the reflectedbeam to be linearly polarized, and accordingly a linear polarizer

can eliminate the reflected beam entirely. But the process of

normal reflection, i.e., with incident and reflected beams perpen-

dicular to the smooth glossy surface in question, produces nopolarization at all. How, then, can the specularly reflected lightbe eliminated while light originating behind the surface is trans-

mitted freely?The question is an important one to radar operators scanning

the cathode-ray-oscilloscope screens on which dim greenish spotsrepresenting airborne objects appear. The screen proper is situ-

ated in a large evacuated tube, and the greenish light emergesthrough a curved glass window at the front end of the tube.(Sometimes the window is flat; sometimes a safety plate of glass

or plastic is mounted close in front of it.) Often the operatorhas difficulty in seeing the greenish spots, not only because they

are faint, but also because they may be masked by various ex-traneous images reflected by the front surface of the window, e.g.,reflections of room lights and of people, clothing, papers, etc.,

situated near the operator. Extinguishing the room lights wouldeliminate these reflections, but would make it impossible forthe operator to read instructions or make notes. What he needsis some kind of filter that will transmit the light originating be-hind the window and absorb the light reflected approximatelyperpendicularly from it.

This need is filled by the circular polarizer. Such a device, if

mounted close in front of the window, will transmit nearly half

of the light that originates behind the window, yet will eliminate

about 99 percent of the room light that is reflected perpen-dicularly from it. The circular polarizer acts on the room light

twice: it circularly polarizes room light that is approaching the

58

window, then absorbs the reflected component. The logic behindthis requires explanation. Two key facts must be kept in mind:

(I) A beam that is reflected perpendicularly and specularly bya smooth glossy surface has the same degree of polarization asthe incident beam, since the reflection process does not intro-duce randomness of any kind.

(2) The reflection process reverses the handedness of polariza-tion, because handedness is defined with respect to the propaga-tion direction and the reflection process reverses the propagationdirection.

If the polarizer is of right-circular type, as in the arrangementshown in Fig. 10-3, room light that passes through and ap-

UNPOLARIZEDLIGHT

RETARDING LAYERSPECULARLYREFLECT I N G

GLASS PLATE

RIGHT HELIX

LEFT

RIGHT CIRCULAR POLARIZER

FIG. 10.3 Use of a circular polarizer in absorbing light reflected by asurface approximately perpendicular to the incident beam. Note thatthe reflection process reverses the handedness of circular polarization.

proaches the window is right-circularly polarized; the reflectedlight is left - circularly polarized and hence is totally absorbed bythe polarizer. In effect, the circular polarizer "codes" the light,the window reverses the coding, and the polarizer then annihi-lates the reverse-coded beam. If both faces of the window areideally fiat and smooth, if the light is incident exactly alongthe normal, and if the polarizer is truly of circular type, the

59

tion of the work.It was also found that a trial with a much larger

revolving mirror gave better definition, more light, andsteadier speed of rotation; so that it seems probable thatresults of much greater accuracy may be obtained in afuture investigation.

FINAL MEASUREMENTS

Observations with the same layout were resumed inthe summer of 1926, but with an assortment of revolvingmirrors.

The first of these was the same small octagonal glassmirror used in the preceding work. The result obtainedthis year was V=299,813. Giving this a weight 2 andthe result of preceding work weight r gives 299,799 forthe weighted mean.

The other mirrors were a steel octagon, a glass 12-sider, a steel 12-sider, and a glass 16-sider.

The final results are summarized in Table VII.

TABLE VII

Turns per Second Mirror Number Ve,ln of Light

528 Glass oct. 576 299,797528.... . Steel oct. 195 299,795352 ... Glass 12 270 299,796352.. . Steel 12 218 299,796264..... . . Glass 16 504 299,796

Weighted mean 299,796± I

46

reflected light is totally absorbed. Usually the conditions areless ideal: the rear surface of the window usually serves as sup-port for the luminescent screen and has a matte appearance; thewindow is usually curved and much of the troublesome roomlight incident on the window makes an angle of 10° or 20° ormore with the normal; and the polarizer, although circular withrespect to some wavelengths, is elliptical with respect to others.Nevertheless, the improvement provided by the polarizer is large,and the amount of faint detail that the operator can see on thescreen is greatly increased.

One precaution must be mentioned: reflections from the polar-her itself must be avoided. This is usually accomplished by tilt-ing the polarizer forward so that the only reflected images theobserver sees are images of a dark-colored floor or other darkobjects.

Television sets, also, have been equipped successfully with cir-cular polarizers. If the set is used in a brightly lit room, or is usedoutdoors, the circular polarizer performs a valuable service intrapping the specularly reflected glare and thus increasing thepicture-vsglare ratio by a factor of the order of 10.

VARIABLE-DENSITY FILTER

A pair of linear polarizers arranged in series is an almost idealdevice for controlling the transmitted intensity of light. Rotatingone polarizer through an angle 0 with respect to the other causes

used as windows of railroad cars and ocean liners. A personsitting near such a window turns a small knob to rotate onepolarizer with respect to the other and thus reduce the intensityof the transmitted light to any extent desired.

One of the authors has experimented with a variable-densityfilter employing three linear polarizers in series, in order that atransmittance range of 108 to 1 could be achieved. The deviceworked well and, as expected, obeyed a cosine-fourth, rather thana cosine-square law.

THREE-DIMENSIONAL PHOTOGRAPHY AND THE USEOF POLARIZERS FOR CODING

Millions of polarizers found their way into the motion picturetheaters of North America in 1952 and 1953 when stereoscopic(three-dimensional, or 3-D) movies achieved brief prominence.Each spectator wore a pair of polarizing spectacles called view-ers, and polarizers were mounted in front of the projectors.

A photographer who enjoys looking at 3-D still pictures in hisliving room needs no polarizers. Usually he employs a small view-ing box containing a light source and two lenses, one for eacheye; a black partition, or septum, divides the box into right andleft halves. The picture, consisting of two small transparenciesmounted about two inches apart in a side-by-side arrangement ona cardboard frame, is inserted in the box so that the right-eyetransparency lines up with the right lens and the left-eye trans-parency lines up with the left lens. (The two transparencies are,of course, slightly different because they were taken by camerassituated about two or three inches apart; the spacing used ap-proximates the spacing of the two eyes.) The side-by-side arrange-ment of the two transparencies and the presence of the septuminsure that the observer's right eye sees only the right trans-parency and the left eye sees only the left transparency. Nocross-communication, or "cross-talk," can occur. Consequentlythe observer enjoys an impressively realistic stereoscopic illusion.

When 3-D motion-picture films are projected in a theater,many complications arise. Separate projectors must be providedfor the right-eye and left-eye movie films, and the two projectorsmust be synchronized within about 0.01 second. Since there is

61

just one large screen and this is to be viewed by hundreds ofspectators, there can be no septum. Indeed, no practical geometri-cal method of preventing cross-talk is known.

Before the advent of mass-produced polarizers in the 1930's, ananalglyph system of preventing cross-talk was invented. It appliedwavelength coding to the two projected beams. The right-eyepicture was projected through a long-wavelength (red) filter, andthe left-eye picture was projected through a shorter-wavelength(green) filter. The spectator's viewers contained r.ght and leftlenses of red and green plastic, respectively, and accordingly eachlens transmitted light from the appropriate projector and ab-sorbed light from the other. Thus each eye received just thelight intended for it. The system succeeded as a short-termnovelty: stereoscopic illusions were created. But the system hadtwo major defects: chromatic "retinal rivalry" between the twoeyes, and incompatibility with the showing of colored motionpictures.

In the 1930's the problem was solved with eclat by a polariza-tion-coding system, demonstrated with great impact at the NewYork World's Fair of 1939 and improved in later years. As indi-cated in Fig. 10-4, a linear polarizer oriented with its transmis-sion axis at 45° is placed in front of the projector used for theright-eye pictures, and a polarizer at +45° is placed in front ofthe projector used for the left-eye pictures. Thus the two beamsstriking the movie screen are orthogonally coded. The lensesof the spectator's viewers consist of correspondingly orientedlinear polarizers, and so each eye receives only light that origi-nates in the appropriate projector. Superb stereoscopic illusionsresult. Since the polarizers perform well at all wavelengths in thevisual range, color movies r,n be presented as easily and faith-fully as can black-and-white movies.

The polarizers placed in front of the projectors consist, ordi-narily, of K-sheet; as explained in Chapter 3, K-sheet is highlyresistant to heat, and any polarizing filter placed close in frontof a powerful projector is bound to heat up considerably since itnecessarily ausorbs about half the light. The lenses of the 3-Dviewers are usually of HN-38 sheet; it has high major trans-mittance. k1 and small minor transmittance k2, and it is inex-

FIG. 10-4 Arrangement for projecting polaezation-coded stereoscopicmotion-picture films by means of two side-by-side projectors. Films FR

and FL containing the "right-eye pictures" and "left-eye pictures" aremounted in the right and left projectors, which are equipped withlinear polarizers PR and PL oriented at 45° and +45' respectively.The viewer contains correspondingly oriented polarizers, and accord-

ingly each eye sees only the images intended for it.

pensive, The viewers are cheap enough (about 10¢ each) thatthey can be discarded after a single use.

The polarization-coding scheme has one limitation: if thespectator tilts his head to o.te side, the polarizers in his viewersno longer line up accurately with the respective polarizers on theprojectors. Thus cross-talk occurs: the right eye sees faintly theimage meant for the left eye, and vice versa: each eye sees afaint ghost image in additicr to the main image. The spectatordoes not enjoy this. The difficulty could be avoided if the linearpolarizers wen: replaced by high-quality, achromatic circularpolarizers, but unfortunately no method it known for producingachromatic circular polarizers economically.

The effectiveness of any polarization-coding projection systemdestroyed if the screen depolat izes the light appreciably.

Screens that have a smooth aluminum coating usually conserve

63

polarization to the extent of about 99 percent, but those havinga matte white surface or a rough metallic coating produce muchdepolarization and hence much cross-talk between the twoimages. Many of the screens used in the innocent days of 1952and 1953 -were of the wrong type, and the resulting ghost imageswere a major annoyance. For that reason, and because of fre-quent lack of care in maintaining synchronism between the twoprojectors, movie-goers soon turned back to conventional 2-1)pictures. Some nostalgia remains, however. Persons who werelucky enough to see a full-color, 3-D movie showing attractiveactors filmed against a background of gorgeous scenery look for-ward to the time when well-made, well-presented 3-D movies,with their almost miraculous realism and intimacy, will animatethe theaters once again.

THE VECTOGRAPH

The type of three-dimensional photography discussed in thepreceding section is parallel-projected 3-D photography. Thetwo motion-picture films are situated side-by-side, and two pro-jectors are operated in parallel. During the late 1930's a radicallynew approach, called vectography, was developed by E. H. Land,J. Mahler, and others. In this system, the two films are arrangedin series, bonded together. Because of the permanent series ar-rangement, many problems disappear.. Only one projector isneeded, and perfect synchronism is "guaranteed at the factory."Each pair of pictures (each vectograph) is projected as a singleunit, in the same projector aperture and at the same time, andonto the same area of the same screen. If the film breaks, it canbe spliced with no concern as to preservation of synchronism.

The method can succeed of :y if means are provided for pre-serving the identity of the two coincident projected beams.Again, polarization-coding is the answer. However, because thetwo images are bonded together in series, the coding must occurwithin the images themselves. In the system used by Land andMahler each image consists of varying quantities of linearlydichroic molecules aligned in a common direction, and the direc-tions employed in the two images are mutually at right angles.Dark areas in any one image contain a high concentration of

64

dichroic molecules; light areas contain little or no dichroicmaterial; but irrespective of concentration, the alignment direc-tion is always the same. For the other image, the alignment direc-tion is always orthogonal to the first. It is to be noted that theimages contain no silver and no other isotropic absorber. Onlyaligned absorbers having high dichroic ratio are used.

A communications engineer would describe the vectograph bysaying that it provides two distinct channels. Each is assigned toone image. Each is independent of the other. Since the vecto-graph images themselves perform the polarization coding, nopolarizer is used in front of the projector; indeed, the interpo-sition of such a :liter would play havoc with the system. As be-fore, the screen must preserve the polarization and the spectator'sviewers must perform the appropriate decoding, or discrimi-nating, act. Excellent stereoscopic effects are achieved. However,the production of vectograph film is a costly undertaking involv-ing very specialized equipment, and constant attention is neededto maintain high enough dichroic ratio so that the channels aretruly independent and ghost images are avoided.

Vectograph pictures of the "still" type are easier and cheaperto make than vectograph movies. Stereo pairs of aerial photo-graphs of mountainous country, if presented in vectograph form,give a navigator (wearing an appropriate viewer) a very realisticimpression of the terrain, and a map maker can prepare an acs c-rate contour map from the vectograph with ease.

POLARIZING HEADLIGHTS

It is ironic that the main goal of Land and others in develop-ing high-quality, large-area, low-cost polarizers has never beenachieved. The polarizers are used with great success in dozensof applications, but not the application that was uppermost inthe minds of the inventors.

Their goal was to eliminate glare from automobile headlights.In an era when dual-lane highways, circumferential bypasses,and other safety engineering advances were virtually unknownand the aim and focus of automobile headlights were highlyerratic, the glare that confronted motorists at night was almost

65

unbearable, and was an important cause of accidents. As earlyas 1920 several illumination engineers recognized that the glarecould be eliminated by means of polarizersif largearea polar-izers could somehow be produced. If every headlight lens werecovered by a linear polarizer oriented with the transmissionaxis horizontal and every windshield were covered with a linearpolarizer oriented with its axis vertical, no direct light from theheadlights of Car A could pass through the windshield of on-coming Car B. Drivers in both cars could see road-markings,pedestrians, and so forth, but neither would experience anyglare from the other's headlights. Moreover, it would be per-missible for each driver to use his high beam continuously, andaccordingly his ability to see pedestrians, etc., would be greaterthan before, despite the fact that each polarizer would transmitonly about half of the light incident on it.

It was soon recognized that the analyzing polarizer should notbe made a permanent part of the windshield, but should be in-corporated in a small visor situated just in front of the driver'seyes. During the day, when headlights were not in use, the visorcould be swung out of the way. It was also recognized that careshould be taken to make sure the headlight polarizers had suffi-cient light-leak, i.e., sufficiently large k2 value, that the head-lights of oncoming cars would not fi:sappear entirely!

Land and his colleagues moved rapidly. They invented awhole series of polarizers, each superior to its predecessor. Thefirst successful type, J-sheet, employed aligned, microscopic crys-tals of the dichroic mineral herapathite; the method of manu-facture is described in Chapter 3. Then came H-sheet, whichwas better in nearly every respect and in addition was easier tomake. Finally, K-sheet appeared; it had most of the superbqualities of the earlier materials and the added virtue of beingunaffected by fairly high temperature, such as 215°F. To per-sons seeking polarizers for use in headlights, K-sheet appearedto be the pot of gold at the end of a polarized rainbow.

Concurrently, several better ways of orienting the polarizerswere proposed. One attractive scheme was to orient the head-light polarizers and the visor polarizer at the identical azimuth,namely 45°, as indicated in Fig. 10.5. Then, even a polariza-tion-conserving object in the path of the headlights would appear

66

FIG. 10.5 Automobile equipped with headlight polarizers and a visorpolarizer oriented at 45°. When two such cars approach one another,each driver is protected from the glare from the headlights of the other.

to the driver to be brightly illuminated. (This would not be thecase if his visor polarizer were crossed with his headlight polar-izers.) The 45° system disposed of the headlight glare problemadequately: if two cars A and B both equipped in this mannerapproached one another at night, each driver's visor would becrossed with the other car's headlight polarizers, and neitherdriver would experience any glare.

Using the Mueller calculus, Billings and Land compared awide variety of polarizer orientation schemes, and found severalto be particularly attractive. Perhaps the best system was onecalled " -55 °, 350.., The t-ansmission axes of the headlightpolarizers and visor polarizer are at 55° and 35° from the vertical,respectively, an arrangement that minimizes complications stem-ming from the obliquity of the portion of the windshield situatedjust in front of the driver.

Despite the successes on all technical fronts, the project boggeddown. To this day no one knows just why. Probably many littlereasons were responsible. Among these were the following:

(1) The polarizers absorbed slightly more than half of thelight incident on them, and accordingly the automobile manu-facturers felt that they would have to increase the power of thelamps themselves and perhaps use larger generators and batteriesalso.

(2) Some windshields were moderately birefringent; therefore

67

they would act like retarders, alter the polarization form of theincident light, and allow some glare to leak through.(3) Nearly every year the automobilemanufacturers increasedthe backward tilt of the windshields; such tilt tends to alter thepolarization form of light having an oblique vibration direction,and hence leads to glare-leak. .

(4) Passengers, as well as drivers, would require visors, sincepassengers also dislike glare.(5) Pedestrians might find that the glare was worse than ever.unless they too employed polarizing visors or spectacles.(6) The system would succeed only if adopted by all car manu-facturers, and therefore no one manufacturer would gain anypromotional advantage from it.(7) The first few drivers to put the system to use would getlittle benefit from it for at least a year or two, i.e., until millionsof other cars were similarly equipped.(8) It was difficult to decide when and how to force the ownersof old cars to install the necessary polarizers on their cars.(9) The patents on the only fully satisfactory polarizers wereheld by a single company.(10) To introduce the system would require formal, coordi-nated action by all States.(11) Improvements in headlight design and aiming, the in-creasing numbers of dual-lane highways, and the brighter streetlamps used in cities and suburbs led some people to believe thatthe need for a polarization-type of glare control was no longeracute.

However, persons who have actually experienced the polariza-tion method of glare removal are convinced that the drawbacksare trivial compared to the benefits.Perhaps some day the system will be tried out on a pilot scalein a small, isolated community, where all the cars could beequipped with polarizers in a few weeks. Perhaps an island ofmoderate size would make a good test ground. If the system isfound to be highly successful there, it will presumably spreadthroughout every country that teems with automobiles.

68

Action at a Distance

James Clerk Maxwell

A scientific paper published in 1873.

I HAVE no new discovery to bring before you this evening. I must askyou to go over very old ground, and to turn your attention to a questionwhich has been raised again and again ever since men began to think.

The question is that of the transmission of force. We see that two bodiesat a distance from each other exert a mutual influence on each other's motion.Does this mutual action depend on the existence of some third thing, somemedium of communication, occupying the space between the bodies, or do thebodies act on each other immediately, without the intervention of anything else?

The mode in which Faz-day was accustomed to look at phenomena of thiskind differs from that adopted by mahy other modern inquirers, and my specialaim will be to enable you to place yourselves at Faraday's point of view, and topoint out the scientific value of that conception of lines of force which, in hishands. became the key to the science of electricity.

When we observe one body acting on another at a distance, before weassume that this action is direct and immediate, we generally inquire whetherthere is any material connection between the two bodies; and if we find strings,or rods, or mechanism of any kind, capable of accounting for the observedaction between the bodies, we prefer to explain the action by means of theseintermediate connections, rather than to admit the notion of direct action at adistance.

Thus, when we ring a bell by means of a wire, the successive parts ofthe wire are first tightened and then moved, till at last the bell is rung ata distance by a process in which all the intermediate particles of the wire havetaken pxt one after the ot! 'r. We may ring a bell at a distance in otherways, as by forcing air into a long tube, at the other end of which k acylinder with a piston which is made to fly out and strike the bell. We

69

may also use a wire; but instead of pulling it, we may connect it at one enditli a voltaic battery, and at the other with an electromagnet, and thus ringthe bell by electricity.

Here are three different ways of ringing a hell. They all agree, however,in the circumstance that between the ringer and the bell there is an unbrokenline of communication, and that at every point of this line some physicalprocess goes on by which the action is transmitted from one end to the other.The process of transmission is not instantaneous, but gradual; so that there isan interval of time after the impulse has been given to one extremity of theline of communication, during which the impulse is on its way, but has notreached the other end.

It is clear, therefore, that in many cases the action between bodies at adistance may be accounted for by a series of actions between each successivepair of a series of bodies which occupy the intermediate space ;, and it is asked,by the advocates of mediate action, whether, in those cases in which we cannotperceive the intermediate agency, it is not more philosophical to admit theexistence of a medium which we cannot at present perceive, than to assert thata body can act at a place where it is not.

To a person ignorant of the properties of air, the transmission of force bymeans of that invisible medium would appear as unaccountable as any otherexample of action at a distance, and yet in this case we can explain the wholeprocess, and -'etennine the rate at which the action is passed on from oneportion to arattter of the medium.

Why then should we not admit that the familiar mode of communicatingmotion by -rushing and pulling with our bands is the type and exemplificationof all action between bodies, even in cases in which we can observe nothingbetween the bodies which appears to take part in the action?

Here for instance is a kind of attraction with which Professor Guthriehas 'mule us familiar. A disk is set in vibration, and is then brought nearlight suspended body, which immediately begins to move towards the disk, asif drawn towards it by an invisible cord. What is this cord ? Sir W. Thomsonhas pointed out that in a moving fluid the pressure is least where the velocityis greatest. The velocity of the vibratory motion of the air is greatest nearestthe disk. Hence the pressure of the air on the suspended body is less on theside nearest the disk than on the opposite side, the body yields to the greaterpressure, and moves toward the disk.

70

The disk, therefore, does not act where it is not. It sets the air next itin motion by pushing it, this motion is communicated to more and more distantportions of the air in turn, and thus the pressures on opposite &des of thesuspended. body are rendered unequal, and it moves towards the disk in conse-quence of the excess of pressure. The force is therefore a force of the oldschoola case of vis a teryo a shove from behind.

The advocates of the doctrine of action at a distance, however, have notbeen put to silence by such arguments. 'What right, say they, have we toassert that a body cannot act where it is not? Do we not see an instance ofaction at a distance in the case of a magnet, which acts on another magnet notonly at a distance, but with the most complete indifference to the nature ofthe matter which occupies the intervening space? If the action depends onsomething occupying the space between the two magnets, it cannot surely be amatter of indifference whether this space is filled with air or not, or whetherwood, glass, or copper, be placed between the magnets.

Besides this, Newton's law of gravitation, which every astronomical obser-vation only tends to establish more firmly, asserts not only that the heavenlybodies act on one another across immense intervals of space, but that twoportions of matter, the one buried a thousand miles deep in the interior ofthe earth, and the other a hundred thousand miles deep in the body of thesun, act on one another with precisely the same force as if the strata beneathwhich each is buried had been nonexistent. If any medium takes part intransmitting this action, it must surely make some difference whether the spacebetween the bodies contains nothing but this medium, or whether it is occupiedby the dense matter of the earth or of the sun.

But the advocates of direct action at a distance are not content withinstances of this kind, in which the phenomena, even at first sight, appear tofavour their doctrine. They push their operations into the enemy's camp, andmaintain that even when the action is apparently the pressure of contiguousportions of matter, the contiguity is only apparentthat a space always inter-venes between the bodies which act on each other. They assert, in short, thatso far from action at a distance being impossible, it is the only kind of actionwhich ever occurs, and that the favourite old vis a tergo of the schools has noexistence in nature, and exists only in the imagination of schoolmen.

The best way to prove that when one body pushes another it does nottouch it, is to measure the distance between them. Here are two glass lenses,

71

72

one of which is pressed against the other by means of a weight. By meansof the electric light we may obtain on the screen an image of the place wherethe one lens presses against the other. A series of coloured rings is formed onthe screen. Thcse rings werc first observed and first explained by Newton.The particular colour of any ring depends on the distance between the surfacesof the pieces of glass. Newton formed a table of the colours corresponding todifferer t distances, so that by comparing the colour of any ring with Newton'stable, we may ascertain the distance between the surfaces at that ring. Thecolours are arranged in rings because the surfaces are spherical, and thereforethe interval between' the surfaces depends on the distance from the line joiningthe centres of the spheres. The central spot of the rings indicates the placewhere the lenses are nearest together, and each successive ring corresponds toan increase of about the 4000th part of a millimetre in the distance of thesurfaces.

The lenses are now pressed together with a force equal to the weight ofan ounce ; but there is still a measurable interval between them, even at theplace where they are nearest together. They are not in optical contact. Toprove this, I apply a greater weight. A new colour appears at the centralspot, and the diameters of all the rings increase. This shews that the surfacesare now nearcr than at first, but they are not yet in optical contact, for ifthey were, the central spot would be black. I therefore increase the weights,so as to press the lenses into optical contact.

But what we call optical contact is not real contact. Optical contact indi-cates only that the distance between the surfaces is much less than a wave-length of light. To slim that the surfaces are not in real contact, I removethe weights. The rings contract, and several of them vanish at the centre.Now it is possible to bring two pieces of glass so close together, that theywill not tend to separate at all, but adhere together so firmly, that when tornasunder the glass will break, not at the surface of contact, but at some otherplace. The glasses must then be many degrees nearer than when in mere opticalcontact.

Thus we have shewn that bodies begin to press against each other whilststill at a measurable distance, and that even when pressed together with greatforce they are not in absolute contact, but may be brought nearer still, andthat by many degrees

Why, then, say the advocates of direct action, should we continue to

maintain the doctrine, founded only on the rough experience of a pre-scientificage, that matter cannot act where it is not, instead of admitting that all thefacts from which our ancestors concluded that contact is essential to action werein reality cases of action at a distance, the distance being to small to bemeasured by their imperfect means of observation ?

If we are ever to discover the laws of nature, we must do so by obtainingthe most accurate acquaintance with the facts of nature, and not by dressingup in philosophical language the opinions of men who had no know-ledge of the facts which throw most light on the:* laws. And as for thosewho introduce retherial, or other media, to account for these actions, withoutany direct evidence of the existence of such media, or any clear understandingof how the media do their work, and Wko fill all space three and four timesover with rethers of different sorts, why die less these men talk about theirphilosophicel scruples about admitting action at a distance the better.

If the progress of science were regulated by Newton's first law of motion,it would be easy to cultivate opinions in advance of the age. We should onlyhave to compare the science of today with that of fifty years ago; and byproducing, in the geometrical sense, the line of progress, we should obtain thescience of fifty years hence.

The progress of science in Newton's time consisted in getting rid of thecelestial machinery with which generations of astronomers had encumbered theheavens, and thus "sweeping cobwebs off the sky."

Though the planets had already got rid of their crystal spheres, they werestill swimming in the vortices of Descartes. Magnets were surrounded byeffluvia, and electrified bodies by atmospheres, the properties of which resembledin no respect those of ordinary effluvia and atmospheres.

When Newton demonstrated that the force which acts on each of theheavenly bodies depends on its relative position with respect to the otherbodies, the new theory met with violent opposition from the advanced philoso-phers of the day, who described the doctrine of gravitation as a return to theexploded method of explaining everything by occult causes, attractive virtues,and the like.

Newton himself, with that wise moderation which is characteristic of all hisspeculations. answered that he made no pretence of explaining the mechanismby which the heavenly bodies act on each other. To determine the mode inwhich their mutual action depends on their relative position was a great step

73

in science, and this step Newton asserted that he had made. To explain theprocess by which this action is effected was a quite distinct step, and thisstep Newton, in his Principiu., does not attempt to make.

But so far was Newton from asserting that bodies really do act on oneanother at a distance, independently of anything between theta, that in aletter to Bentley, which has been quoted by Faraday in this place, he says:

"It is inconceivable that inanimate brute matter should, without the media-tion of something else, which is not material, operate upon and ,Ifr t othermatter without mutual contact, as it must do if gravitation, in the sense ofEpicurus, be essential and inherent in it That gravity shotdd he innate,inherent, and essential to matter, so that one body can act upon another ata distance, through a xacuum, without the mediation of anything else, by andthrough which their action and force may be conveyed from one to another, isto me so great an absurdity, that I believe no man who has in philosophicalmatters a competent faculty of thinking can ever fall into it."

Accordingly, we find in his Option/ Queries, and in his letters to Boyle,that Newton had very early made the attempt to account for gravitation bymeans of the pressure of a medium, and that the reason he did not publishthese investigations "proceeded from hence only. that he found he was notable, from experiment and observation, to give a satisfactory account of thismedium, and the manner of its operation in producing the chief phenomena ofnature*."

The doctrine of direct action at a distance cannot claim for its author thediscoverer of universal gravitation. It was first asserted by Boger Cotes, inhis preface to the Prineipia, which be edited during Newton's life. Accordingto Cotes, it is by experience that we learn that all bodies gravitate. We donot learn in any other orty that they are extended, movable, or solid. Gravi-tation, therefore, has as much right to be considered an essential property ofmatter as extension, mobility, or impenetrability.

And when the Newtonian philosophy gained ground in Europe, it was theopinion of Cotes rather than that of Newton that became most prevalent, tillat last lioscovich propounded his theory, that matter is a congeries of mathe-matical points, each endowed with the power of attracting or repelling theothers according to fixed laws. In his world, matter is unextended, and contact

Ilaclauries Account of Sesames Discoveries.

74

Actiort Distonoe

is impossible. He did not forget, however, to endow his mathematical pointswith inertia. In this some of the modern representatives of his school havethought that he "had not quite got so far as the strict modern view of'matter' as being but an expression for modes or manifestations of force'*."

But if we leave out of account for the present the development of theideas of science, and confine our attention to the extension of its boundaries, weshall see that it was most essential that Newton's method should be extendedto every branch of science to which it was applicablethat we should investi-gate the forces with which bodies act on each other in the first place, beforeattempting to explain how that force is transmitted. No men could be betterfitted to apply themselves exclusively to the first part of the problem, thanthose who considered the second part quite unnecessary.

Accordingly Cavendish, Coulomb, and Poisson, the founders of the exactsciences of electricity and magnetism, paid no regard to those old notions of"magnetic effluvia" and "electric atmospheres," which had been put forth inthe previous century, but turned their undivided attention to the determinationof the law of force, according to which electrified and magnetized bodies attractor repel each other. In this way the true laws of these actions were dis-covered, and this was done by men who never doubted that the action tookplace at a distance, without the intervention of any medium, and who wouldhave regarded the discovery of such a medium as complicating rather than asexplaining the undoubted phenomena of attraction.

We have now arrived at the great discovery by Orated of the connectionbetween electricity and magnetism. Orated found that an electric current actson a magnetic pole, but that it neither attracts it nor repels it, but causes itto move round the current. He expressed this by saying that "the electricconflict acts in a revolving manner."

The most obvious deduction from this new fact was that the action of thecurrent on the magnet is not a push-and-pull force, but a rotatory force, andaccordingly many minds were set a-speculating on vortices and streams of tetherwhirling round the current.

But Ampere, by a combination of mathematical skill with experimentalingenuity, first proved that two electric currents act on one another, and thenanalysed this action into the resultant of a system of push-and-pull forcesbetween the elementary parts of these currents.

Review of Mrs Somerville, Saturday Ream, Feb. 19, 1869.

75

The formula of Ampere, however, is of extreme complexity, as comparedwith Newton's law of gravitation, and many attempts have been made to resolveit into something of greater apparent simplicity.

I have no wish to lead you into a discussion of any of these attempts toimprove a mathematical formula. Let us turn to the independent method ofinvestigation employed by Faraday in those researches in electricity and mag-netism which have made this Institution one of the most venerable shrines ofscience.

No man ever more conscientiously and systematically laboured to improveall his power, of mind than did Faraday from the very beginning of hisscientific career. But whereas the general course of scientific method then on-sisted in the application of the ideas of mathematics and astronomy to each -fewinvestigation in turn, Faraday seems to have had no opportunity of acquiringa technical knowledge of mathematics, and his knowledge of astronomy wanmainly derived from books.

Hence, though he had a profound reepect for the great discovery of Newton,he regaread the attraction of gravitation as a sort of sacred mystery, which,as he was not an astronomer, ho had no right to gainsay or to doubt, hisduty being to believe it in the exact form in which it was delivered to him.Such a dead faith was not likely to lead him to explain new phenomena bymeans of direct attractions.

Besides this, the treatisr.# of Poisson and Ampere are of so technical aform, that to derive any a sistance from them the student must have beenhoroughly trained in rafoueinatics, and it is very doubtful if such a trainingcan be begun with advantage in :natare years.

Thus Faraday, with his penetrating intellect, his devotion to science, andhis opportunities for experiments, was debarred from following 64 course ofthought which had led to the n.chievements of the French philosophers, andwas obliged to expluin the phenomena to himself by means of a symbolismwhich he could understand, instead of adopting what had hitherto been theonly tongue of the learned.

This new symbolism consisted of those lines of force extending themselvesin every direction from electrified and magnetic bodies, which Faraday in hismind's eye saw as di musotly as the solid bodies from which they emanated.

The idea of lines of force and their exhibition by means of iron filingswas nothing new. They had been observed repeatedly, and investigated mathe

76

matically as an interesting curiosity of science. But let us hear Faradayhimself, as he introduces to his reader the method which in his hands becameso powerful*.

"It would be a voluntary and unnecessary abandonment of most valuableaid if an experimentalist, who chooses to consider magnetic power as representedby lines of magnetic force, were to deny himself the use of iron filings. Bytheir employment he may make many conditions of the power, even in com-plicated cases, visible to the eye at once, may trace the varying direction ofthe lines of force and determine the relative polarity, may observe in whichdirection the power is increasing or diminishing, and in complex systems maydetermine the neutral points, or places where there is neither polarity norpower, even when they occur in the midst of powerful magnets. By their useprobable results may be seen at once, and many a valuable suggestion gainedfor future leading experiments."

Experiment on Lines of Force.

In this experiment each filing becomes a little magnet. The poles of oppo-site names belonging to different filings attract each other and stick together,and more filings attach ther. -.elves to the exposed poles, that is, to the endsof the row of filings. In this way the instead of forming a confusedsystem of dots over thc paper, draw together, filing to filing, till long fibresof filirgs are formed, which indicate by their direction the lines of force inevery part of the field.

The mathematicians saw in this experiment nothing but a method of exhibit-ing at one view thc direction in different places of the resultant of two forces,one directed to each pole of the magnet ; a somewhat complicated result ofthe simple law of force.

But Faraday, by a series of steps as remarkable for their geometricaldefiniteness as for their speculative ingenuity, imparted to his conception of theselines of force a clearness and precision far in advance of that with which themathematicians could then invest their own formulae.

In the first place, Faraday's lines of force are not to be considered merelyas individuals, but as forming a system, drawn in space in a definite manner

Exp. Rea. 3284.

774t-

ft,

78

so that the number of the lines which pass through an area, say of one squareinch, indicates the intensity' of the force sting through the area. Thus thelines of force Income definite in number. The strength of a magnetic pole ismeasured by the number of lines which proceed from it; the electrotonic stateof a circuit in measured by the number of lines which pass through it.In the second Pplace, each individual line has a continuous existence inspace and time. When a piece of steel becomes a nn. .et, or when an electriccurrent begins to flow, the lines of force do not start into existence each inits own place, but as the strength increases new lines are developed withinthe magnet or current, and gradually grow outwards, so that the whole systemexpands from within, like Newton's rings in our former experiment. Thus everyline of force preserves its identity during the whole course of its existence,though its shape and size may be altered to any extent.I have no time to describe the methods by which every question relatingto the forces acting on magnets or o.. currents, or to the induction of currentsin conducting circuits, may be solved by the consideration of Faraday's lines offorce. In this place they can never be forgotten. B means of this newsymbolism, Faraday defined with mathematical precision the whole theory ofelectro-magnetism, in language free from mathematical technicalities, and appli-cable to the most complicated as well as the simplest cases. But Faraday didnot stop here. Re went on from the conception of geometrical lines of forceto that of physical lines of force. He observed that the motion which themagnetic or electric force tends to produce is if.variably such as to shortenthe lines of force and to allow them to spread out laterally from each other.He thus perceived in the medium a state of stress, consisting of a tension,like that of a rope, in the direction of the lines of force, combined with apressure in all directions at right angles to them.This is quite a new conception of action at a distance, reducing it to aphenomenon of the same kind as that action at a distance which is exertedby means of the tension of ropes and the pressure of rods. When the musclesof our bodies are excited by that stimulus which we are able in some unknownway to apply to them, the fibres tend to shorten themselves and at the sametime to expand laterally. A state of stress is produced in the muscle, and thelimb moves. This explanation of muscular action is by no means complete.It gives no account of the cause of the excitement of the state of stress, nordoes it even investigate those forces of cohesion which enable the muscles to

support this stress. Nevertheless, the simple fact, that it substitutes a kind ofaction which extends continuously along a material substance for one of whichwe know only a cause and an effect at a distance from each other, inducesus to accept it as a real addition to our knowledge of animal mechanics.

For similar reasons we may regard Faraday's conception of a state of stressin the eleetro-magnetic field as a method of explaining action at a distance bymeans of the continuous transmission of force, even though we do not knowhow the state of stress is produced.

But one of Faraday's most pregnant discoveries, that of the magneticrotation of polarised light, enables us to proceed a step farther. The phe-nomenon, when analysed into its simplest elements, may be described thus:Of two circularly polarised rays of light, precisely similar in configuration, butrotating in opposite directions, that ray is propagated with the greater velocitywhich rotates in the same direction as the electricity of the magnetizingcurrent.

It follows from this, as Sir W. Thomson has shewn by strict dynamicalreasoning, that the medium when under the action of magnetic force must bein a state of rotation--that is to say, that small portions of the medium,which wo may call molecular vortices, are rotating, each on its own axis, thedirection of this axis being that of the magnetic force.

Here, then, we have an explanation of the tendency of the lines of mag-netic force to spread out laterally and to shorten themselves. It arises fromthe centrifugal force of the molecular vortices.

The mode in which electromotive force acts in starting and stopping thevortices is more abstruse, though it is of course consistent with dynamicalprinciples.

We have thus found that there are several different kinds of work to bedone by the electro-magnetic medium if it exists. We have also seen thatmagnetism has an intimate relation to light, and we know that there is a theoryof light which supposes it to consist of the vibrations of a medium. How i4this luminiferous medium related to our electro-magnetic medium?

It fortunately happens that electro-magnetic measurements have been madefrom which we can calculate by dynamical principles the velocity of progagatiouof small magnetic disturbances in the supposed electro-magnetic medium.

This velocity is very great, from 288 to 314 millions of metres per ger:mid,according to different experiments. Now the velocity of light, accordit g to

79

Foucault's experiments, is 298 millions of metres per second. In fact, the differentdeterminations of either velocity differ from each other more than the estimatedvelocity of light does from the estimated velocity of propagation of small electro-magnetic disturbance. But if the luiniferous and the electro-magnetic mediaoccupy the same place, and transmit disturbances with the same velocity, whateason have we to distinguish the one from the other? By considering themas the same, we avoid at least the reproach of filling space twice over withdiffeent kinds of taller.

Besides this, the only kind of electro-magnetic disturbances which can bepropagated through a non-conducting medium is a disturbance transvee to thedirection of propagation, agreeing in this respect with what we know of thatdisturbance which we call light. Hence, for all we know, light also may be auelectro-magnetic disturbance in a non-conducting medium. If we admit this, dieelectro-magnetic theory of light will agree in every respect with the undulatorytheory, and the work of Thomas Young and Fresnel will be established on afirmer basis than ever, when joined with that of Cavendish and Coulomb bythe key-stone of the combined sciences of light and electricityFaraday's greatdiscovery of the electro-magnetic rotation of light.

The vast interplanetary and interstellar regions will no longer be regardedas waste places in the universe, which the Creator has not seen fit to fill withthe symbols of the manifold order of His kingdom. We shall find them to bealready full of this wonderful medium; so full, that no human power can removeit from the smallest portion of space, or produce the slightest flaw in itsinfinite continuity. It extends unbroken from star to star; and when a moleculeof hydrogen vibrates in the dog-star, the medium receives the impulses of thesevibrations; and after carrying them in its immmise bosom for three years,delivers them in due course, regular order, and full tale into the spectroscopeof Mr Huggins, at Tulse Hi 11.

But the medium has other functions and operations besides bearing lightfrom man to man, at from world to world, and giving evidence of the absoluteunity of the metric system of the universe. Its minute parts may have rotatoryas well as vibratory motions, and the axes of rotation form those lines ofmagnetic force which extend in unbroken continuity into regions which no eyehas seen, and which, by their action on our magnets, are telling us in languagenot yet interpreted, what is going on in the hidden underworld from minuteto minute and from century to century.

80

Aci10,, at a Dist,a,ce

And these lines must not be regarded as mere mathematical abstractions.They are the directions in which the medium is exerting a tension like thatof a rope, or rather, like that of our own muses. The tension of the mediumin the direction of the earth's magnetic force is in this country one grainweight on eight square feet. In some of Dr Joule's e.cperiments, the mediumhas exerted a tension of 200 lbs. weight per square inch.

But the medium, in virtue of the very same elasticity by which it is ableto transmit the undulations of light, is also able to act as a spring. Whenproperly wound up, it exerts a tension, different from the magnetic tension, bywhich it draws oppositely elect-ified bodies together, produces effects throughthe length of telegraph wires, and when of sufficient intensity, leads to therupture and explosion called lightning.

These are some of the already discovered properties of that which hasoften been called vacuum, or nothing at all. They enable us to resolve severalkinds of action a a distance into actions between contiguous parts of a con-tinuous substan e. Whether this resolution is of the nature of explication orcomplication, I must leave to the metaphysicians.

81

---:,-..-:-: Si

._.4 .,,, 2_,:-: .,,, .......

*.

--, ''',..NA:41..

------- ..-:-. .:;_ii.;,-;4,i-- :;.- -__;--%M-, .i.-

_ .. -- _ - -- - :::-.,---: \in_

..,. ...,..-_- r..:7.-..--,-...-.--z-.--- --- ...:.----,..,,--" 25-'4,,--. ..,,,...::...." = ---1---- ,,,4"S,,,a."`

...

.,

-

.,-,--,,------------'Y---' - ''f,"-,:.-"..:-:;-.---':"--:;'----;:-....'----0f''''''-'::'...:7'7. ,, ...-- .4"' .44 ..../..."..,,- -

..." .." ., r" ,;=470" ---- -:---;-"--"".*,.- ' ,,,,, it ,,, -

,', ,4',4 4. 4 "494,/..." .--'-'

//,' _...... ,.. 7--"-44,

'., -- -,','"-,,,;"..-<,,,1;--v;

If '')tit !II V4t0.'Ti '''''yer'4'<'.7 ....' / .;.A:

, .0, .- . A

0?)fk, ..,..,.:711 r;e:,,...i:e0j ;.; ' .

-A.,' de --"4---- :f.-.(:'5.--z---q--;-...t,----=t4-- --. - t'7.

y- ii,:- ,..,4,--;;,/, --,;-:,_--:-...-c _,-,....,-,,,

:t.%"...-;*--'9:::?-;*--t.:7L'--rf. z.,-,: .7 1":-.';- ' '1''''''':: ,,,. ,.

, .-irer;2:1-, ?"'`'-='----' '----- . ,-,Z7`." .." '''.."'''' '.

:

k`..::"-..:s..".....,.f:.

.Z...:.`'

:

'*..'''

--'

:_:.

,,-

: '.:

:,N

.

k

.

A. ...,1/4i

`:

,

tt' ; "5

Sx.;":..

1.)'

...',

.. ,

= ' A a1

'''''.k:-.4f`',.:ti, 'ti-''N:..:':":;N.4 11

i.

1; f i ' i

1 '' 1'7.--'?/,-.),f,'/,-../,/iv,"1,!'p- i:--,, (1/,f, v,

9 ,--ii-, )1,i I q4 ii .i I; 41:..." Tt i i .::' Pi,.c , - i ,

VI' '; ;:,-11.!

,,,;%;",I'',,..:;',/,c'V \ 1/4'1'1 'I'

ts ks i"N ;.... , 4` - '%,. .* 4--4,1 ,..s._- , ,. '._ \''!,:..).,-

F4')

4/4

14Xf.;9: 1.

41;41,Li ,

4.4:r" iftlA

Electric field between oppositely charged conductors.

82

-,i

A brief, ink---IcE reviei: :1 fi-e eiedrpnic J}e, )1:taelo pre.3onf,

7 The Electronic Revolution

Arthur C. Clarke

1962

The electron is the smallest thing in the universe; it wouldtake thirty thousand million, million, million, million of themto make a single ounce. Yet this utterly invisible, all butweightless object has given us powers over nature of whichour ancestors never dreamed. The electron is our most ubiqui-tous slave; without its aid, our civilization would collapse in amoment, and humanity would revert to scattered bands ofstarting, isolated savages.

We started to use the electron fifty years before we dis-covered it. The first practical application of electricity (whichis nothirg more than the ordered movement of electrons)began with the introduction of the telegraph in the 1840's.With really astonishing speed, a copper cobweb of wires andcables spread across the face of the world, and the abolition ofdistance had begun. For over q century we have taken theinstantaneous transfer of news completely for granted; it isvery hard to believe that when Lincoln was born, communi-cations were little faster than in the days of Julius Caesar.

Although the beginning of "electronics" is usually datedaround the 1920's, this represents a myopic view of tech-nology. With the hindsight of historical perspective, we cannow see that the telegraph and the'telephone are the first twolandmarks of the electronic age. After Alexander Graham Bellhad sent his voice from one room to another in 1876, societycould never be the same again. For the telephone was the first

83

t

(

84

electronic device to enter the home and to affect directly thelives of ordinary men and women, giving them the almostgodlike power of projecting their personalities and thoughtsfrom point to point with the speed of lightning.

Until the closing years of the nineteenth century, men usedand handled electricity without knowing what it was, but inthe 1890's they began to investigate its fundamental nature,by observing what happened when an electric current waspassed through gases at very low pressures. One of the first,and most dramatic, results of this work was the invention ofthe X-ray tube, which may be regarded as the ancestor of allthe millions of vacuum tubes which followed it. A cynicmight also argue that it is the only electronic device whollybeneficial to mankindthough when it was invented manyterrified spinsters, misunderstanding its powers, denouncedpoor Röntgen as a violator of privacy.

There is an important lesson to be learned from the X-raytube. If a scientist of the late Victorian era had been asked"In what way could money best be spent to further theprogress of medicine?" he would never by any stretch of theimagination have replied: "By encouraging research on theconduction of electricity through rarefied-gases." Yet that iswhat would have been the right answer, for until the dis-covery of X rays doctors and surgeons were like blind men,groping in the dark. One can never predict the outcome offundamental scientific research, or guess what remote andunexpected fields of knowledge it will illuminate.

X rays were discovered in i895 the electron itself just oneyear later. It was then realized that an electric current consistsof myriads of these submicroscopic particles, each carrying aminute negative charge. When a current flows through a solidconductor such as a piece of copper wire, we may imagine theelectrons creeping like grains of sand through the intersticesbetween the (relatively) boulder-sized copper atoms, Anyindividual electron does not move very far, or very fast, but itjostles its neighbor and so tie impulse travels down the line at

,,

AV

,

speeds of thousands of miles a second. Thus when we switchon a light, or send a Morse dash across a transatlantic cable,the response at the other end is virtually instantaneous.

But electrons can also travel without wires to guide them,when they shoot across the empty space of a vacuum tube likea hail of machine-gun bullets. Under these conditions, nolonger entangled in solid matter, they are very sensitive to thepull and tug of electric fields, and as a result can be used toamplify faint signals. You demonstrate the principle involvedevery time you hold a hose-pipe in your hand; the slightestmovement of your wrist produces a much greater effect at thefar end of the jet. Something rather similar happens to thebeam of electrons crossing the space in a vacuum tube; theycan thus multiply a millionfold the feeble impulses picked upby a radio antenna, or paint a fluorescent picture on the endof a television screen.

Until 1948, electronics was almost synonymous with thevacuum tube. The entire development of radio, talkies, radar,television, long-distance telephony, up to that date dependedupon little glass bottles containing intricate structures of wireand mica. By the late 194D's the vacuum tube had shrunkfrom an object as large as (and sometimes almost as luminousas) an electric light bulb, to a cylinder not much bigger than am...ds thumb. Then three scientists at the Bell TelephoneLaboratories invented the transistor and we moved from thePaleoelectronic to the Neoelectronic Age.

Though the transistor is so smallits heart is a piece ofcrystal about the size of a rice grainit does everything that aradio tube can do. However, it requires only a fraction of thepower and space, and is potentially much more reliable. In-deed, it is hard to see how a properly designed transistor canever wear out; think of little Vanguard I, still beeping awayup there in space, and liable to continue indefinitely untilsome exasperated astronaut scoops it up with a butterfly net.

The transistor is of such overwhelming importance becauseit (and its still smaller successors) makes practical hundreds

85

86

of electronic devices which were previously too bulky, too ex-pensive or too unreliable for everyday use. The pocket radio isa notorious example; whether we like it or not, it points theway inevitably to a day when person-to-person communica-tion is universal. Then everyone in the world will have hisindividual telephone number, perhaps given to him at birthand serving all the other needs of an increasingly complexsociety (driving license, social security, credit card, permit tohave additional children, etc.). You may not know where onEarth your friend Joe Smith may be at any particular mo-ment; but you will be able to dial him instantlyif only youcan remember whether his number is 8296765043 or8296756043.

Obviously, there are both advantages and disadvantages insuch a "personalized" communication system; the solitudewhich we all need at some time in our lives will join thevanished silences of the pre-jet age. Against this, there is lioother way in which a really well-informed and fast-reactingdemocratic society can be achieved on the original Greekplanwith direct participation of every citizen in the affairsof the state. The organization of such a society, with feedbackin both directions from the humblest citizen to the Presidentof the World, is a fascinating exercise in political planning. Asusual, it is an exercise that will not be completed by the timewe need the answers.

A really efficient and univert -ommunications system,giving high-quality reception on ail wilds between all pointson the Earth, can be achieved only with the aid of satellites.As they come into general use, providing enormous informa-tion-handling capacity cn a global basis, today's patterns ofbusiness, education, entertainment, international affairs willchange out of all recognition. Men will be able to meet faceto face (individually, or in groups) without ever leaving theirhomes, by means of closed circuit television. As a *suit ofthis, the enormous amount of commuting and traveling thatnow takes place from home to office, from ministry to United

,.

Nations, from university to conference hall will steadily de-crease. There are administrators, scientists and businessmentoday who spend about a third of their working lives eithertraveling or preparing to travel. Much of this is stimulating,but most of it is unnecessary and exhausting.

The improvement of communications will also render obso-lete the city's historic role as a meeting place for minds and acenter of social intercourse. This is just as well anyway, sincewithin another generation most of our cities will be strangledto death by their own traffic.

But though electronics will ultimately separate men fromtheir jobs, so that (thanks to remote manipulation devices)not even a brain surgeon need be within five thousand milesof his patient, it must also be recognized that few of today'sjobs will survive long into the electronic age. It is now a clichéthat we are entering the Second Industrial Revolution, whichinvolves the mechanization not of energy, but of thought.Like all clichés this is so true that we seldom stop to anall...:what it means.

It means nothing less than this: There are no routine, non-creative activities of the human mind which cannot be, carriedout by suitably designed machines. The development of com-puters to supervise industrial processes, commercial transac-tions and even military operations has demonstrated thisbeyond doubt. Yet today's computers are morons compared tothose that they themselves are now helping to design.

I would not care to predict how many of today's professionswill survive a hundred years from now. What happened to thebuggywhip makers, the crossing sweepers, the scriveners, thestonebreakers of yesteryear? (I mention the last because I canjust remember them, hammering away at piles of rock in thecountry lanes of my childhood.) Most of our present occupa-tions will follow these into oblivion, as the transistor inheritsthe earth.

For as computers become smaller, cheaper and more re-liable they will move into every field of human activity. Today

they are in the office; tomorrow they will be in the home.Indeed, some very simple-minded computers already do ourhousehold chores; the device that programs a washing ma-chine to perform a certain sequence of operations is a special-ized mechanical brain. Less specialized ones would be able tocarry out almost all the routine operations in a suitably de-signed house.

Because we have so many more pressing problems on ourhands, only the science- fiction writersthose trail-blazers ofthe futurehave given much thought to the social life of thelater electronic age. How will our descendants be educated forleisure, when the working week is only a few hours? We havealready seen, on worldwide scale, the cancerous growthsresulting from idleness and lack of usable skills. At everystreet corner in a great city you will find lounging groups ofleather-jacketed, general-purpose bioelectric computers of aperformance it will take us centuries and trillions of dollars tomatch. What is their future and ours?

More than half a century ago H. G. Wells described, inThe Time Machine, a world of decadent pleasure lovers,bereft of goals and ambitions, sustained by subterranean ma-chines. He set his fantasy eight hundred thousand years in thefuture, but we may reach a similar state of affairs within adozen generations. No one who contemplates the rising curveof technology from the Pilgrim fathers to the Apollo Projectdare deny that this is not merely possible, but probable.

For most of history, men have been producers; in a very fewa .turies, they will have to switch to the role of consumers,devoting their energies loo per cent to absorbing the astro-nomical output of the automated mines, farms and factozies.

Does this really matter, since only a tiny fraction of thehuman race has ever contributed to artistic creation, scientificdiscovery or philosophical thought, which in the long run arethe only significant activities of mankind? Archimedes andAristotle, one cae.not help thinking, would still, have left their

hadon history even if the ,5C1 ha lived in a society based on

88

i!", I.,-

robots instead of human slaves. In any culture, they would beconsumers of goods, but producers of thought.

We should not take too much comfort from this. The elec-tronic computers of today are like the subhuman primates often .pillion years ago, who could have given any visitingMartians only the faintest hints of their potentialities, whichincluded the above mentioned Archimedes and Aristotle.Evolution is swifter now; electronic intelligence is only dec-ades, not millions of years, ahead.

And thatnot transistor radios, automatic homes, globalTVis the ultimate goal of the Electronic Revolution.Wheiler we like it or not, we are on a road where there is noturning back; and waiting at 'ts end are our successors.

89

T. A. EDISON.Electric-Lamp.

No. 223,898. Patented /an. 27, 1880.

hl .ma

165414;4vi

.16

itdatwarot

Sityread (Z *(iratton/

i4 4.4491(.294teieJcof==tat f A

EDISONS PATENT on the incandescent lamp Was accompanied thickened ends of filament (c). platinum wire* id). damps (hi,by this draw inc. The labeled parts are the carbon filament (o), leadins wires is), rapper wires le), tube to vacuum pomp Ins).

90

The Invention of the Electric Light

Matthew Josephson

1959

41can hire mathematicians, butmathematicians can't hire mel"By such declarations in the time

of his success and world-wide fameThomas Alva Edison helped to paint hisown portrait as an authentic. Americanfolk hero: the unlettered tinkerer andtrial-and-error inventor who achievedhis results by persistence and a na-tive knack for things. He is said, for ex-ample, to have tried more than 1,800kinds of material ("paper and cloth,thread, fishline, fiber, celluloid, box-wood, coconut-shells, spruce, hickory,hay, maple shavings, rosewood, punk,cork, flax, bamboo and the hair out of ared-headed Scotehman's beard') untilhe hit upon the loop of carbonized cot-ton thread that glowed in a vacuum formore than half a day on October 21,1879 Today, in a world that relies forits artificial illumination largely on hisincandescent lamp, this invention is notregarded as an especially profound con-tribution to technology. It rates rather asa lucky contrivance of Echson's cut-and-try methodsof a piece with his stockticker. mimeograph machine, phono-graph and alkaline storage-batter inthe esteem of a public that has come toappreciate the enormous practical signif-icance of higher mathematics and ab-struse physical theory.

If Ediso IA contribution to the lightof the world consisted solely in the se-lection of a filament, this estimate of hisperson and achievements might be al-lowed to stand. But the history that isso obscured by legend tells quite an-other story. Edison's electric light wasnot merely a lamp but a system of elec-tric lighting. His invention was an idearather than a thing. It involved not onlytechnology but also sociology and eco-nomics. Edison was indisputably thefirst to recognize that electric lighting

would require that electricity be gen-erated and distributed at high voltagein order to subdivide it among a greatmany high-resistance "%loners," eachconverting current at low amperage(that is, in small volume) with greatefficiency into light.

In the 15 months between the timehe conceived his invention and the dateon which he demonstrated it to the pub-lic, Edison and his associates designedand built a new type of electric genera-tor, successfully adapted the then much-scorned parallel or "multiple-arc" circuitthat would permit individual lights tobe turned on or off separately and, lastof all, fashioned a lamp to meet thespecifications of his system. The labora-tory notebooks of those months of fran-tic labor show the Wizard of Menlo Parkendowed with all the prodigious capaci-ties attributed to him by contemporarylegend, They show in addition that thisself-taught technologist was possessed ofa profound grasp of the nature of elec-tricity and an intuitive command of itslogic and power.

It was on September 8, 1878, thatEdison was inspired to devote his talentsfrill time to the challenge of electriclighting. On that day he went to An-sonia, Conn., to visit the brass-manufac-turing plant of \Wham Wallace, co-inventor with Moses G. Farmer of thefirst practical electric dynamo in theU. S. Wallace showed Edison eight bril-liant carbon-are lights cf 500 candle-power each, powered by r dynamo ofeight horsepower. It was such asystem that Wallace and Farmer, as wellas Charles Brush of Cleveland, werethen beginning to introduce the electriclight on a commercial scale, for street-lighting and for illuminating factoriesand ships. Farmer had made the firstdemonstration of arc-lighting in this

country two years earlier, at the Cen-tennial Exposition in Philadelphia, andJohn Wanamakees store in that city wasalready illuminated with are lights.

Carbon arcs are still employed insearchlights and in theater floodlightsand projectors to produce light of highintensity The current crossing a smallgap between the electrodes creates anare. Ionization and oxidation of the car-bon in the heat of the arc generate abrilliant blue -white

In the 1870's Europe was a decadeahead of the U. S. in the technology ofarc-lighting. Stores, railway stations,streets and lighthouses in Britain andFrance were equipped with arc lights.Shedding an almost blinding glare, theyburned in open globes that emitted noxious gases, and they could be employedonly high overhead on streets or in pub-lic buildings. Since they consumed largeamounts of current, they had to be wiredin series, that is, connected one to an-other in a single continuous circuit sothat all had to be turned on or off to-gether. The multiple-arc circuit, with thelights connected as in the rungs of a lad-der between the main leads of the cir-cuit, was not adapted to such systemsand was considered prohibitive in cost.

Edison himself had experimentedwith are lights, using carbon strips asburners. He had also investigated the

arron's NOTE

The author has !rased this articleon material in his biography Edi-son, just published by the McCraw -Hill Book Company. Copyright1959 by Matthew Josephson.

91

4r.

incandescent light, as had many Inven-tors before him, But the slender rod orpencil of carbon or metal %mold Assaysbum up, sooner rather than later, uponbeing heated to incandescence by thecurrent, It u mild do so though substan-tially all of the air had been pumped outof the glass envelope in %Ouch it ssascontained Edison hail abandoned theeffort to des ote himself to .0 more prom-ising invention. the phonograph

Now at Wallace's establishment, con.fronted with the achievements of othersin the field, he regained his earlier en-thusiasm As an eyewitness recalled."Edison was enraptured.... He fairlygloated He ran from the instruments[the dynamos) to the lights, and thenagain from the lights back to the electricinstruments He sprawled over a tableand made all sorts of calculations. Ile cal-culated the posser of the instrumentsand the lights, the probable loss of passer

transmission, the amount of coal theinstrument ssould use in a day, a week,a month, a year."

To William Wallace he said challengingly. "1 beliese I can beat you makingthe electric light. I do not think you areworking in the right direction." Theyshook hands in friendly fashion. andwith a diamond.pointed stylus Edisonsigned his name and the date on a gobletprovided by his host at dinner.

From Edison's own complete and ex.phut notebooks add from the buoyantInterviews that he gale to the press atthis tune sse know what made him feelIII such fine fettle as he left Wallace'splant. "I saw for the first time everythingin practical operation," he said. "I sawthe thing had not gone so far but that Ihad a chance. The intense light had notbeen subdivided so that it could bebrought into private houses In all elec-tric lights theretofore obtained the lightwas very great, and the quanaty [oflights) very low. I came home and madeexperiments two nights in succession. Idiscos ered the necessary secret, so sim-ple that a bootblack might understandIt. . The subdivision of light is allright

The Subdivision of Light

At this time there flashed into Edison'smind the image of the urban gasiightingsystem, ssith its central gashouse and gasmains running to smaller branch pipesand leading into many chselling places atlast to gas jets that could be turned onor off .it will. During the past lialfven-t gat-lighting had reached the statureof a major industry hi the U. S. It was

92

restricted, of course, to the cities. threefourths of the U. S population still heedIII meat ireas by the dim gloss of kero.scale lamps or candles. Ruminating insolitude. Edison sought to god a clearstatement to his objective III Inc note-book, under the title "Electricity sersusGas as a General Illuminant," he %%rote.

"Object. E . to effect exact mina-tion of all done by gas, to replace light-ing by gas by lighting by electricit, Toimprove the illumination to such an ex-tent as to meet all requirements of natural, artificial and commercial condi-tions. . Edison's great effortnot tomake a large light or a blinding light.but a small light having the mildness ofgas

To a reporter for one of the leadingNew York dailies who had shadowedhim to Ansonia, Edison described is vi-sion of a central station for electric light-ing that he would create for all of NewYork City. A netscork of electric wirewould deliver current for a myriad ofsmall household lights, unlike the daz-/ling are lights made by Farmer andBrush In some way electric curreatwould be metered and sold Edison saidhe hoped to have his electric.light involtam) ready in six weeks' At MenloPark, N.J where his already famousworkshop was located, he would wire allthe residences for light and hold a "grandexhibition."

Thus from the beginning Edison riv-eted his attention not so much upon thesearch for an improved type of Meandescent filament as upon the analysis ofthe social and economic conditions forwhich his invention was intended. As heturned ssith immense energy to expand-ing the facilities at Menlo Park and se.curing the essential financing, he conmined lus studies of the gaslighting in-dustry. In parallel he projected the m.o.lionizes of the electric-lighting system heenvisioned.

Gas had its incomenience and dan-gers 'So unpleasant . . that in the newMadison Square theater every gas jet isventilated by small tubes to carry awaythe products of combustion But Wir.liek er is to replace gas must have "a gencral system of distributionthe only pos-sible means of economical dlumin.tionGathering all the back files of the gasindustry's journals and scores of volumesbearing >n gas illumination, he studiedthe operations and habits of the indus-try, its seasonal curves and the 1.1)011t ofits distribution systems. In his mind hemapped out a network of electriclightImes for an entire city, making theshrewd judgment. "Poorest district for

light, best for powerthus evening upwhole city." Ile meant that III slum dis-tricts there ssould be higher demand forsmall industrial motors. Against tablesfor the cost of converting coal to gas hecalculated the cost of converting coaland steam into electric energy. An ex.pert gas engineer, whose sen ices Edisonengaged at this time. obsened that fewmen knew more about the ssorIcTs gasbusiness than did Edison

Edison had a homo occonornrcus %sniin him, a ss ell.developed social and com-mercial sense, though he was careless ofmoney and was not an accountant of thetype exemplified by his contemporaryJohn D. Rockefeller. Before the experi-mental work on his invention was underway, he had formed a dear notion,stated in economic terms, of what .ts oh-m:I must be. This concept guided hissearch am I determined the pattern of histechnic . decisions, so that the resultwould I no scientific toy but a ,inductuseful to people everywhere. By histeal cal( illation of the capital investmentin machinery and copper for a wholesystem of light distribution he was ledto define the kind of light he sought andthe kind of generating and distributingsystem he needed

Backers of the Electric Light

In the crucial matter of financing hisinventive work Edison had the generousand imaginative aid of Grosvenor Low-rey, a patent and corporation lawyerwell established in the financial com-munity of Wall Street. Lowrey had fall-en completely under Edison's spell andregarded hum much as a collector ofpaintings regards a great artist whoseworks he believes are destined for immortality. Using his extensive coin'(ions and the favorable pressmotices thathe encouraged Edison to secure duringlate September and early October. 1878,Lowrey assembled a sponsoring syndicate of some of the most important fin-anciers of the time. The underwritersof the Edison Electric Light Company,sshich was incorporated in midOctober.included William H. Vanderbilt andJ. P. Morgan s partner Egisto Fabbri.This was an unprecedented deselop-ment in U. S. business Inventors hadbeen backed in the development of in.ventions al-eady achieved; Edison's fi-nanciers were backing him in researchthat was to lead to a hopedfor invention.In many respects the venture marks thebeginning in this country of close relaeons between finance and technology.

Their money," Edison said, "was

Thc., Invention of the Electr:c Llpht

XS

EDISON AND IIIS PHONOGRAPH were photographed in 1878 turned to the more promising phonograph. in the year that thisby Mathew Brady. Ile had worked with electric lights but had photograph was made. however, he resumed his work on Hinting.

93

0

94

MENLO PARK was depleted in Fronk Leslie's illustrated News. laboratory Is t isible :.! the far right. In Rs windows patsengers onpope, for January 10, 1880. The barnhke 'tabernacle- of Edison'. the nearby railroad could see his experimental lights burning.

tested in confidence of my ability tobring it back again" The 31-year-oldEdison was by now a wellknottn figurein Wall Street. His quadruples telegraphsystem. by %%hid) four separate messagescould be transmitted ot er a single wire,had furnished the pivotal issue in the vasteconomic oar waged bettteen WesternUnion and the rival telegraph empire ofthe robber baron Jay Could. Edison'scarbon microphone had transformed thetelephone from an instrument of limitedusefulness to an efficient system of king-range communication that was now ra(hating across the country. The shares ofgas-lighting enterprises had tun:bled onthe New York and London exchangesupon Edison's announcement, in thepress campaign instigated by Lowrey.that he was now about to displace gaswith electricity in the lighting of bonusand factories.

The alliance between Edison and hissponsors tt as nonetheless an uneasy one.The first rift appeared before the end ofOctober. when the rival !mentor Wil-ham Satsyer and his partner Allow JLmannounced that they had "beaten" Edi-son and applied for a patent on a carbon-pencil light in a nitrogen-filled glasstube. There ttas a flutter of panic in thednectoratc of the Edison Electric LightCompany. The suggestion was madethat Edison should join forces with Saw -yer and Man. Lowey passed the sug-gestion on to S. L. Griffin. a formerjunior esecut we at Western Union whomLott rev had hired to help Edison withhis business affairs.

Griffin sent hack a hasty "ian.fiden-ad" reply: "I spoke to Mr. Edison re-garding the Sawyer-Man electric light.

. I was astonished 4.: the manner inwhich Mr. Edison received the informa-tion. He tt as visibly agitated and saal itwas the old story, that is. lack of confi-dence.... No combination, no consoli-

dation for him I do nut feel at liberty torepeat all lie said. but I do feel impelledto sugges respectfully that as little besaid to him as possible math regard tothe matter."

In view of Edison's talent for candidand salty language Griffin's reticence isunderstandable. After that there was nofurther talk of consolidation with Saw-s er or any other int entor.

The Menlo Park Laboratory

In his belief that he would "get aheadof the other fellows" Edison was sus-taint,' by his unbounded confidence inhis ; nugatory. its superior equipmentand its staff. The Menlo Park laboratorywas still the only full-time industrial re-search organization in the country, initself perhaps Edison's most importantintention. During this period the physi-cal plant was greatly expanded: a sep-arate office and library, a house for too80-horsepotter steam engines, and aglass blotter's shed ttere added to theoriginal barnlike "tabernacle." Evenmore important, Edison bad collected anucleus of talented engineers and skilledvans:nem ttlio were of inestimable helpto him in %tolling out his ideas.

The selLtaught Edison thought pri-marily in concrete, visual terms. Whenlie was at ttork on the quadruples tele-graph. be had even built a model madeup of pipes and valves corresponding tothe wires and relays of his system, andwith running water replacing the elec-tric current, so that he could actually seehow it worked. But now he would haveto depend far more on theory and mathematics.

One of the happiest effects of Gros-venor Lowrey's personal influence wasthe hiring of Francis II. Upton. a youngelectrical enginee- i q had worked fora year in the Bean .aboratory of the

great physicist Hermann son Helmholtz.Edison jocularly nicknamed Upton "Cul-ture." and. according to au oft-told story,put the "green" mathematician in hisplace with one of his scientific practicaljokes. He brought out a pear-shapedglass lamp-bulb and gate it to Upton.asking him to calculate its content incubic centimeters. Upton drew theshape of the bulb exactly on paper. andderived from this an equation for :hebulb's volume. Ile was about to coin, ritethe anstter when Edison returned Indimpatiently asked for the results. Uptonsaid he would need more time. "Why,'said Edison, "I would simply take thatbulb, fill it with a liquid, and measureits volume directly!"

When Upton joined the staff late niOctober. Edison had alrezdy committedhimself to the incandescent light. This,rather than the are light, was the was toimitate the mildness of gas. But the fila-ment glowing in a vacuum had beensought in vain by numerous int clawsfor half a century. hi choosing the in-candescent light rather than the arelight he was "putting aside the technicaladvance that had brought the arc lightto the commercial stage." No one. in-ibid..% himself, had succeeded in mak-ing an incandescent lamp that wouldstork for more than a few minutes.

Edison's first efforts in 1878 were notnotably more successful Knowing thatcarbon has the highest melting point ofall the elements, he first tried strips ofcarbonized paper as "burners" and man-aged to keep them incandescent for"about eight minutes" before theyburned up in the partial vacuum of hisglass containers. Turning to the infusi-ble metals, he tried spirals of platinumwoe: they gave a brilliant light butmelted in the heat. Edison accordinglydemised a feedback thermostat devicethat 7 vitched off the current when the

heat approached the melting point. Thelamp now blinked instead of going outentirely. Nonetheless, with his ese onthe problem of financing. Edison filed apatent application on October 5 and iii.sited the press in for .1 dewoostration

As this disconragnig work proceededin the weeks that followed, Edisonturned, with Upton's help. to calculatingthe current thit would be consumed bya lighting system equipped with a cer-tain numb( i of such lamps. They as-mimed that the lights would be con-nected in parallel. so their imaginaryhouseholder could turn one light in thecircuit on or off at will, as in a gas-light-ing system. Thinking in round numbers,they assumed that these lamps, whenperfected, might have a resistance of oneohm and so stools! consume 10 amperesof torrent at 10 colts. Allowing in addi-tion for the energy losses in the distribu-tion system, they found that it wouldrequire a fabulous amount of copper tolight just a few city blocks. Such a sys-tem of los -resistance lights was clearlya commercial impossibility.

This was the gist of the objeclonswhich had greeted Edison's first an-nouncements that he would use an in-candescent bulb in : parallel circuit.Typical of the scorn heaped upon himwas the opinion expressed by a commit-tee set up by the British Parliament toinvestigate the crash of gas-lighting se-curities WA the achice of British sci-

t 11

1;

'a-

' '/

entists, the members of the committeedeclared that though these plans seemed"good enough for our transatlanticfriends: they wire "unworthy of the at-tention of practical or scientific men."From Olun's law, which governs the re-tat" hip between voltage, amperageand resistaoce IU a eirtvit,"the reportargued that if an electric light of 1.000candlepower were divided into 10

smaller lights and connected in parallel,each of the smaller lights would radiatenot one tenth but "one hundredth only ofthe original light." In this judgment suchfigures as Lord Kelvin and John Tyndallconcurred. Before the Royal Institutionin London the distinguished electricianSir William Preece declared: "Subdivi-sion of the electric light is an absolutegas Mims

Ohm's law does indeed slow that theamount of torrent (amperes) flowing ina circuit is equal to the electromotiveforce (volts) divided by the resistance(ohms) in the circuit. Edison's content.poraries reasoned that an increase in thenumber of lights in a ei.cuit would in-crease the resistance and therefore re-duce the flow of current to each. It wasthought that the only way to providethese lights with sufficient current was toreduce the resistance in the distributionsystem. In a parallel circuit this meantincreosing the thickness of the copperconductors to an Impractical degree.Such were the limits on the operation

of arc lights, with their low resistanceand huge appetite for current. Upton'scalculations showed that this condo-sum also applied to Edison's first low-resistance incandescent lamps.

Edison now confounded his collabora-tor by proposing that he make the samesort of estimates for an entirely differentkind of circuit. This time he wouldassume lights of %cry high resistance.supplied with current at high soltageand low amperage. In November andDecember Upton made calculattons onthe basis of the same number of lights.but lights with the high resistance of100 ohms each. These lights were tooperate on the low current of only oneampere. Their high resistance was to beoffset, in accord with Ohm's law, by thehigh voltage of 100 volts in the circuit,The result was astonishing: A high-resistance system would require onlyone hundredth of the weight of copperconductor needed for a low-resistancesystem. And copper was the most costlyclement involvedthe decisive eco-nomic factor.

The High-Resistance System

Here was the crux of Edison's insightat Ansonia. He had recognized therethat the subdivision of light called forlamps of high resistance which wouldconsume but little current; to balancethe electrical equation it would be neces-

isli ; t

---4.---:."t- tIA

Ma..

**':`N N.

\t. \

INTERIOR OF EDISON'S LABORATORY at Menlo Park was Illustrated Newspaper. At the time of the work on the electricalso depicted in the January 10, 1680. issue of Frank Leslie's light the laboratory had expanded into several other buildings.

95

sary to supply the current at high volt.age. This was the "necessary secret" thatwas "so simple." Today every high.school physics student learns that thepower lost in transmitting electric ener-gy varies w ith the square of the current.Thus a tenfold reduction in currentmeant a decrease of a hundredfold in theenergy wasted (or a hundredfold- de-erase in the weight of the transmis-sion line). It was a conception easilyreached by an elementary applic-a.

lion of Ohm's law, but it had not oc.curred to any of Edison's contem-poraries. Even Upton did not immesh.ately grasp the full import of Edison'sidea. As he said later: "I a tnnot imagii.ewhy I did not see the elementary factsin 1878 and 1879 more clearly than Idid. I came to Mr. Edison a trained man,with a year's experience hi Helmholtz's

laboratory.... a working knowledge ofcalculus and a mathematical turn ofmind. Yet my eyes were blind in com.parison with those of today; and . Iwant to say that I had company!"

With Upton's figures before him Edi.son was convinced that a row and strate-gic invention lay surely within his grasp.It was clear what kind of distributingsystem he wanted And he knew whatform of incandescent burner would serve,,is purpose. To offer the necessary re.sistance to the passage of current it musthave a small cross section and so wouldhave a small radiating surface.

By January. 1879. Edison was testinghis first highresistance lamp. It had aspiral of very fine platinum wire set ina globe that contained as high a vacuumas could be achieved with an ordinaryair pump. The results were encourag.

ing; these lamps lasted "an hour or two."He then attacked the dual problem ofgetting a higher vacuum and improvinghis incandescing element. After anothertrial with carbon. he returned to metals.platinum, iridium, boron. chromium,molybdenum, osmiumvirtually everyinfusible metal. He thought of tungsten,but could not work it with existing tools.Discouraged by the problem, Edisontried niirogen in his globe and then re.sumed his cfforts to obtain a highervacuum. Hearing of the new and effi.dent Sprengel vacuum pump, whichused mercury to trap and expel air, hesent Upton to borrow one from the near-by College of New Jersey (now Prince.ton University), When Upton retuiredwith the pump late that night. Edisonkept him and the other men on the staffup the rest of the night trying it out

h`tit .

"VI

GENERATOR which Edison deaeloped for the needs of electriclighting appears at right in this engraving from Sur NTIVIC Astral.

96

1: .1../ hi I

le,

4

\-41E 3 OP.

CAN for October 18. 1819 tat that time this magazine appearedweekly). The generator was called the long.waisted Mary Ann."

At this stage Edison made a usefulfinding: "I have discovered," he noted,that many metals which have gas with.

in their pores have a lower melting pointthan when free of such gas." With theaid of the Sprengel pump he devised amethod of expelling these occludedgases, by heating the element while theair was being exhausted from the bulb.The platinum wire within the bulbthereupon became extremely hard andcould endure far higher temperatures.Edison later said that at this stage he"had made the first real steps toward themodern incandescent lamp."

Meanwhile the spirits of his financialsponsors had begun to droop. Their bill-hant inventor, far from having achievedanything tangible, was hinting plainlythat he needed more money. The firstBrush are lights were ablaze over lowerBroadway. and more were being installed elsewhere with impressive effect.Edison's backers began to have seriousdeubts as to whether he had pursued theright course To shore Pp their moraleLowrey arranged to have Edison givethem a private demonstration.

In April, as one of Edison's associatesrecalled it, They came to Menlo Parkon a late afternoon train from NewYork. It was already dark when theywere conducted into the machine shopwhere we had several platinum lamps in-stalled in series." The "boss" showed hisvisitors pieces of platinum coil he wasusing in the lamps, pointed out thearrangement of the lights and describedthe type of generator he hoped to build.Then. the room having grown quitedark, he told "Honest John" Kruesi to"turn on the juice slowly."

'Today, I can still sec those lampsrising to a cherry.red ... and hear Mr.Edison saying 'A little more juice andthe lamps began to glow. 'A little more.'...and then one emits a light like astar, after which there is an eruptionand a puff, and the machine shop h inPaul darkness. ... The operation wasrepeated two or three times, with aboutthe same results."

The platinum coils still consumed alot of power for the light they gave, andthey were costly and shortlived. Thetemporary WallaceFarmer dynamosheated up badly. and were not powerfulenough to enable Edison to connect hislamps in parallel. Edison admitted thatthe system was not yet "practical."

It was a gloomy gathering that brokeup on that raw April evening. All ofLowrey's abounding faith would be nec-essary to rally. the spirits and funds ofEdison's despondent backers. Some

The lrivsmtion of tree Electiir

..r.r,* ..-r*""'

. // ie .1' 1

: I I I V. .VACUUM PUMP used to remove air from lamp bulb. (top center) was of a new type aboutwhich Edison had read in a scientific Journal. The man Is holding a vessel of mercury.

rumors of the disappointing demonstranon leaked out; the price of Edison stockfell sharply, "fide that of gaslightingsecurities "After that dcmonstration," Edison's associate relates, "wehad a general house deaning at the labofanny, and the metallic lamps werestored away."

Edison now rallied his staff to effortson a much broader area of the front"under siege." He followed three mainlines of investigation. One group hedetailed to the task of developing thedynamo to supply the constantvoltage

current required by his high-resistancesystem. He set another group to pullingdown a still higher vacuum in the glassbulbs. The third team, tnder his watchful eye. carried out the series of experi-ments in which 1,600 different materialswere tested for their worth as ineandescent elements.

The "Long.Waisted Mary Ann"

To subdivide the electric current fornumerous small lights in parallel Edisonneeded a dynamo which would produce

a higher soltage than any dynamo in ex.istence, and which would maintain thatvoltage constant unier varying demandsfor current from the system. Existingdynamos were designed around the fal-lacious notion, held by most electricalexperts, that the internal resistance ofthe dynamo must be equal to the ex.temal resistance of the circuit. Throughstudy of battery circuits they had provedthat a dynamo could attain a maximumefficiency of only 50 per cent. In 1877 acommittee of scientists appointed by theFranklin Institute in Philadelphia hadbeen impressed to discover that the mostsuccessful European dynamo, designedby amobe Theophile Gramme, con.vetted into electricity 38 to 41 per centof the mechanical coPigy supplied toit, The efficiency of the Brush dynamowas even tosser: 31 percent. These ma.chines and their theoretically successful

contemporaries all produced current ata relatively, low voltage.

Edison had concluded. however, thathe must produce a dynamo of reducedinternal resistance capable of generatingcuricnt at a high voltage. Such a ma.chine would not only meet the needs ofhis lighting system but would also con-vert mechanical energy to electrical en.erg with far greater efficiency. As hisassociate Francis jehl recalled. Edisoasaid that The did not intend to build upa system of distribution in which the ex-ternal resistance would be equal to theinternal resistance. He said he was justabout going to do the opposite: heseat. "el a large external resistance anda low internal resistance. He said hewanted to sell the energy outside thestation and not waste it in the dynamoand the conductors. where it brought noprofits." jehl, who carried out the tests

01.0

=IMMO

SERIES CIRCUIT (top) requires that a number of electric lights (circles) be turn d onor off at the same time by atingle snitch (break in circuit). Parallel circuit (bottom),which was adopted by Edison, makes it possible to tarn lights on or off one at a line.

98

EDISON'S LIGHT.

The Great Inventor's Triumph in

Electric Illumination.

A SCRAP OF PAPER.

It Makes a Light, Without Gas or

Flame, Cheaper Than Oil,

TRANSFORMED I THE FURNACE,

Complete Details of the PerfectedCarbon Lamp.

FIFTEEN MONTHS OF TOIL.

tory of His Tireless Experiments with

Burners and Generators.

SUCCESS IN A COTTON THREAD.

The Wizard's Byplay. with Roddy Pain

and Gold "Tailings."

HISTORY OF ELECTRIC LIGHTING.

Ths aw oprmat of pumas siluSition atLlisonl keg looted rot electrie kith. eneosaced toSlgo siva en Sew UV'S In at MestoPisk, otsaud, occasion that p.m wilt be illominetwl withWe new light. las ,,clod pane MUM% in theMeat roomer's worksheet throeihont the cleillsedvoid *duffels tad people generally Sr. susoustyswelling the malt. From ihe beonsied Of his ex.moms ha eleeri luaus( to As present timelir. Edison has kept his laboratory guardedlyclued. sad no solnotheilee moon (mein thatPublished in the lissom tome Months ago fellatio;to hie trot assent) or Any of us Important steps ofSo provost lase ben nude Com of pro-Mate the haunter and Outlet* sooMoorl carhie ova putout.. The laaA144 sow.howooff.otablog to swami to its marts felt sad stoma)..t,OOla of MO wort from lie Menthes to lie um-Oftles.

11011212 seraLIMA'. gentle light, Intuitable lb It 111671110/46".

Is prodsced lna ollttlo sloe.* of papor-4 tiny Orlyor taper that Mr* would Stew wey. Throsgh

FIRST NEWSPAPER ACCOUNT of Edison's brilliant success appeared in TheNew York Herald he December 21, 1879.

of resistance, also remarked that the artof constructing dynamos was then asmysterious as air navigation. All elec.trical testing was In the emhryonic stage."There Here no Instruments for ineasurmg volt: and amperes directly: it waslike a carpenter without his foot rule."

Upton himself bail his difficulties inthis hitherto Imexplored field; "I re.member distinctly when Mr. Edison gaveinc the pro: :am of placing a motor incircuit, in multiple arc, with a fixed re.sistance, and 1... could find no priorsolution. There was nothing f could findbearing on the (effect of thel counter-elcet ro-motive force of the armature ...and the resistance of the armature on thework given out by the armature, It was awonderful experience to have problemsgiven me by him based on enormous ex-perience in practical work and applyingto new lines of progress?'

The problem of a constant-voltagedynamo was attacked with the usualEMsoman 6Ian Seeking to visualizeevery possible structural innovation forhis dynamt armature, he had his menlay out nume, ms wooden dummies onthe floor and w.nd wire around them,spumng them on in their task by layingwagers as to who would finish first,

After Edison hat, decided upon theform of winding and type of electromag-nets to be used, Upton made drawingsand tables from which the real armatureswere wound and attached to the com-mutator. Edison eventually worked outan armature made of thin sheets of ironinterleaved with insulating sheets ofmica; this armature developed fewereddy currents and so produced less heatthan the solid armature eons then used.When the new cores were testrun, itwas Upton who made the mathematicalcalculations from these tests and drewup the final blueprints.

The self-effacing Upton can be givenprincipal credit for interpreting Ed:-sons ideas and translating them intomathematical form. A careful student ofcontemporary electrical knowledge, heseems to have been conversant with, andto have guided himself by, the design ofa German dynamo, made by the Siemensworks, that employed an auxiliary sourceof current to excite its field magnets.

The new Menlo Park dynamo com-prised many admirable features for thatperiod. With its great masses of iron andlarge, heavy wires, it stood in bold con-trast to its contemporary competitors.Owing to the two upright columns of itsfield electromagnets, it was nicknamed"Edison's long-waisted Mary Ann:"

When the dynamo was run at the cor-rect speed, the voltage between its .1M1a.

ture brushes was approximately 110,and remained fairly constant, falling butslightly %shell increasing amounts of cur.rent were taken out of the machineEdison and Upton also contrn ed a sun-ple but ingenious dynamometer bywhich the torque of a -drive belt wasused to measure the work output of thesteam engine that powered the dynamo.When Erusi completed the first oper-ating machine. Upton carefully checkedthe results To his astonishment andquite as Edison had "guessed"the newdynamo, tested at full load, showed 90.per-cent efficiency in converting steampower into electrical energy.

Edison was as jubilant as a small boy.As was usual with him. the world wassoon told alt about his "Faradic ma.chine." It was described and depicted inSCIENTIFIC ANIEHICAN for October 18,1879, in an article written by Upton.

Once more there was scoffing atEdison's "absurd claims." The hectoringof Edison by some of the leading U. S.electrical experts, among them HenryMorton of the Stevens Institute of Technology, nosy seems traceable to theirignorance. Reading Morton's predictionsof failure, Edison grimly promised thatonce he had it all running "sure-fire,"he would erect at Menlo Park a littlestatue to his critic which would be eter-nally illuminated by an Edison lamp.

As a matter of fact, this allegedlyignorant "mechanic" was to be foundreading scientific journals and institu-tional proceedings at all hours of the dayand night. It, was thus that he hadlearned about the Sprengel vacuumpump. Th:: device enabled him toachieve an increasingly greater vacuumand to test a broad variety of metals, rareearths and carbon compounds underhitherto unexplored conditions.

The globe itself was also much im-proved, by the inventor's own design,after he had brought to Menlo Park anartistic German glass blower namedLudwig Pochm. Edison one day drew asketch of a one-piece, allglass globewhose joint was completely sealed, andlate in April, 1879, Boehm, workin g skill-fully with hand and mouth, fashionedit in the small glass blower's shed in backof the laboratory.

"There never has been a vacuum pro-duced in this country that approachedanywhere near the vacuum which isnecessary for me," Edison wrote in hisnotebook. After months of effort he couldsay exultantly: "We succeeded in mak-ing a pump by which we obtained avacuum of onemillionth part of anatmosphere?'

In the late summer of 1879 he realized

with growing excitement that a key posiLion had been won. He hid a dynamosupplying constant high voltage, and atight glass globe containing a high vaenum In his mind's eye he saw what mightbe clone with an extremely fine, highlyresistant incandescing substan .e underthese conditions Ills state of tension isreflected in the laboratory notebooks bysuch exclamations as I Classbusted by Boeh mi" All that remained forlint was to discover a filament thatwould endure.

The Carbon Filament

In late August or early Septemberabout a year after he first took up hissearchhe muted back to experiment.ing with carbon. this time for good. Therods of carbon he had tned earlier hadbeen impossible to handle, as he nowunderstood, because carbon in its porousstate has a marked propensity for ab-sorbing gases. But once he had a trulyhigh vacuum and a method for expellingoccluded gases he saw that he mightachieve better results with carbon thanwith platinum.

In a shed in back of the laboratorythere sus a line of kerosene lamps alwaysburning, and a laborer engages, in scrap-ing the lampblack from the glass chim-neys to make carbon cake. But lampblackcarbon by itself was not durable enoughto be made into fine lamp filaments. Edi-son and Upton had arrived at the conclu-sion that, given a 100volt multmlearecircuit, the resist a nee of the lamps shouldbe raised to about 200 ohms, this meantthat the filament could be no thickerthan a 64th of an inch.

Through the summer months Edisonand his staff worked at the tantalizingtask of making fine reeds of lampblackcarbon mixed with tar. His assistantskept kneading away at this puttylikesubstance for hours. It seemed impos-sible to make threads out of it; as anassistant complained one day, the stuffcrumbled.

"How long did you knead it?" Edisonasked.

"More than an hour.""Well just keep on for a few hours

more and it will come out all right."Before long they were able to make

filaments as thin as seven thousandthsof an inch. Edison then systematicallyinvestigated the relations between theelectrical resistance, shape and heatradiation of the filaments. On October7, 1879, he entered in his notebooka report on 24 hours of work: "A spiralmade of burnt lampblack was even bet.ttr than the Wallace (soft carbon) mix.

ture." This was indeed promising: thethreads lasted an hour or two beforethey burned out. But it was not yetgood enough

As he felt himself approaching thegoal Edison drove his co-workers harderthan ever. They held watches over cur-rent tests around the clock, one mangetting a feu hours' sleep while anotherremained awake. One of the laboratoryassistants invented what was called a"corpse.reviver," a sort of noise machinethat would be set going with horribleeffect to waken anyone who overslept.Upton said that Edison "could neverunderstand the limitations of thestrength of other men because his ownmental and physical endurance seemedto be without limit."

The laboratory notebooks for October,1879, show Edison's mood of anticipationpervading the whole staff. He pushed onwith hundreds of trials of fine filaments,so attenuated that no one could conceive

how they could stand up under heatFinally he tried various methods oftreating cotton threads, hoping that theirfibrous texture might give strength tothe filament even after they had beencarbonized. Before heating them in thefurnace he packed them with powderedcarbon in an earthenware cruciblesealed with fire clay. After many (alluresin the effort to clamp the delicate fila-ment to platinum leadin wires, Edisonlearned to mold them together withlampblack and then fuse the joint be.tween them in the act of carbonization.

Then, as Edison later related, it wasnecessary to take the filament to theglass blower's shed in order to seal itwithin a globe: "With the utmost pre-caution Batchelor took up the preciouscarbon, and I marched after him, as ifguarding a mighty treasure. To our con-sternation, just as we reached the glassblower's bench, the wretched carbonbroke. We turned back to the mainlaboratory and set to work again. It was

late :n the afternoon before we producedanother carbon, which was broken bya )eweler's scu.wdriver falling againstit. But we turned back again and beforenightfall the carbon was completed andInserted in the lamp. The bulb was ex-hausted of air and sealed, the currentturned on, and the sight we had so longdesired to see met our eyes."

"Ordinary Thread"

The entries in the laboratory note-books, although bare and impersonal,nonetheless convey the drama and senseof triumphant resolution pervading thelaboratory that night. "October 21No. 9 ordinary thread Coats Co. cordNo. 29, came up to one-half candle andwas put on 18 cells battery permanentlyat 1:30 A.M.... No. 9 on from 1:30AM. till 3 P.M.-1331 hours and was thenraised to 3 gas jets for one hour thencracked glass and busted."

As the light went out the weary men

EARLY EXPERIMENTAL LAMP is de.pitted in one of Edison's notebooks. This FRANCIS R. UPTON made invaluable calculations for Edison's system. An electrical en.lamp bad a filament of platintm. It oohed. pincer who had studied with Hermann von Helmholte. he was named "Culture" by Edison.

waiting there jumped from their chairsand shouted with joy. Edison, one ofthem recalled, remained quiet and thensaid: "If It can burn that number of hoursI know I can make it burn a hundred."Yet all the workers at Menlo ParkErh-r n, Upton, Kruesi, Boehm and the restwere completely astonished at their suc-cess. Thy had become accustomed tolabonng without hope. "They neverdreamed," as one contemporary accountput it, "that their long months ofhard work could be ended thus abruptly,and almost by accident. The suddennessof it takes their breath away."

For once Edison tried to be discreetand keep his momentous discoveries asecret until he could improve upon hislamp filament. At length, after experi-menting with various cellulose fibers,he found that paper, in the form of toughBristol cardboard, proved most endur-ing when carbonized. Edison was ex-ultant when this filament burned for 170hours, and swore that he would perfecthis lamp so that it would withstand 400to 1,000 hours of incandescence beforeany news of it was published.

On November 1, 1879, he executed apatent application for a carbonfilamentlamp. Its most significant passage wasthe declaration: "The object of the in-vention is to produce electric lamps giv-ing light by incandescence, which lampsshall have high resistance, so as to allowthe practical subdivision of the electriclight....The invention consists in alight-giving body of carbon wire ... tooffer great resistance to the passage ofthe electric current, and at the sametime present but a slight surface fromwhich radiation can take place." Thespecifications called for a distinctiveone-piece allglass container, lead-inwires of platinum that passed throughthe glass base and were fused to thecarbon filament, and joints that weresealed by fusing the glass.

Here were the essential features ofthe basic Edison carbonfilament lamp,in the form that was to be known to theworld during the next half century. Itwas not the "first" electric light, noresen the first incandescent electriclamp. It was, however, the first practicaland economical electric light for uni-versal domestic use.

Edison had spent more than $42,000on his experimentsfar more than hehad been advanced by his backers. Nowhe asked for more money so that hemight complete a pilot lightandpowerstation at Menlo Park. But the directorswere still uncertain about the future ofthe invention. Was it "only a laboratorytoy," as one of them charged? Would

it not need a good deal of work before itbecame marketable? Grosvenor Lowreystoutly defended his protege, He got noresults until he prematurely, and overEdison's objections, made the secret ofthe electric lamp public.

Rumors had been spreading for sev-eral weeks. New Jersey neighbors toldof bnlhant lights blazing all night atMenlo Park, and railroad passengers be-tween New York and Philadelphia alsoSaw the bright lights with astonishmentfrom their train windows. In Wall Streetthere was a flurry of speculation inEdison stock; the price rose briefly to43,500 a share.

Then came a front-page story in TheNew York Herold on Sunday, December21, 1879. There followed an exclusivearticle about the inventor's struggles forthe past 14 months, told to the world,con amore, by Marshall Fox, who hadwritten much of Edison before. The de-tailed treatment of such an adventurein applied science as a feature story wassomething of an inn, ation. Also some-what unusual in the journalism of thetime was its relative accuracy of detail,owing to help provided by Upton, whoalso supplied drawings for the Herald'sSunday supplement. The writer did hisbest to explain how this light was pro-duced from a "tiny strip of paper thata breath would blow away"; why thepaper filament did not burn up but be-came as hard as granite; and how thelight-without-flame could be ignitedwithout a matchwhen an electric cur-rent passed through it, giving a "bright,beautiful light, like the mellow sunsetof an Italian autumn."

In the week following Christmas hun-dreds of visitors made their way to theNew Jersey hamlet. Edison hurried withhis preparations for an announced NewYear's Eve display as best he could, butwas forced to use his whole staff of 80persons to handle the crowds. He could

'do no more than put on an improvisedexhibition, with only one dynamo anda few dozen lights.

The closing nights of the year 1879fumed into a spontaneous festival thatreached its climax on New Year's Eve,when a mob of 3,000 sight-seers floodedthe place. The visitors never seemeZ totire of turning those lights on and off.

The inventor promised the sight-seersthat this was but a token of what wasin store. He was awaiting the completionof a new generator, he said, and intendedto illuminate the surroundings of MenloPark, for a square mile, with 800 lights.After that he would light up the dark-ness of the neighboring towns, and eventhe cities of Newark and New York.

The Invention of the Electf ic Ugh

101

'11

High Fidelity

Edgar Villchur

Two chapters from his book Reproduction of Sound published in 1962.

TT MIGHT APPEAR that following a dis-1 cession of the nature of sound, thelogical subject to consider would bethe criteria for reproducing this soundwith "high fidelity" to the original. Oneother element, however, should be cov-ered firstthe way in which we hear.

Perception of SoundWe have already seen, in examining

units of measurement for pitch andpowerthe octave and the decibelthatour perception of sound does not neces-sarily correspond directly to the objec-tive reality. The illusion is consistent,however, so that a given sound alwayshas the same effect on a normal ear.

An important element in the percep-tion of sound was discovered by Fletcherand Munson in 1933. These investigatorsdemonstrated that our impression ofloudness did not depend solely on theamplitude of the sound wave, but onother things as well. Specifically, theyshowed that sound in the lower treblerange of the frequency spectrumthe3500-cps regionappeared to be muchlouder than sound of the same amplitude

at any other part of the spectrum. Thus,if the frequency scale was swept by atone which continuously rose in fre-quency but kept exactly the same ampli-tude, the loudness, or apparent ampli-tude, would increase to a maximum atabout 3500 cps and then fall off again.

This fact does not have much practicalinterest for the person listening to re-produced music, except as it describesthe relative nuisance value of differenttypes of noise. No matter how lop-sidedour interpretation of acoustic reality, wemake the same interpretation in the con-cert hall as in our living room, and thecraftsmen who designed musical instru-ments (who worked to satisfy their ears,not sound-level meters) perceived soundin the same way.

Fletcher and Munson made a seconddiscovery, however, that does bear di-rectly on the reproduction of sound.They found that the effect describedabove took place in varying degree, de-pending on the over-all level of thesound. For very high amplitude soundthe drop in loudness with frequencybelow 3500 cps hardly occurred at all,

103

while for very soft sound the effect wasmaximum. Above 3500 cps the effect re-mained constant, within 2 or 3 db, nomatter what the over-all sound level.

The well-known "equal loudness con-tours," also referred to as the Fletcher-Munson curves, are reproduced in Fig.2-1. Each curve plots the sound ampli-tude required to produce the same per-ceived loudness at different frequenciesof the scale. It can be seen that normalhearing losses in the bass end becomeprogressively greater as the over-allsound level is decreased.

This means that if an orchestra playsa musical passage at the sound level rep-resented by 90 db, and if this music isreproduced at the 60 db level, we willhear the bass with less relative loudnessthan we would have heard it at the con-cert itself. If you follow the 90- and60-db curves, shown superimposed inFig. 2-2, you will see that there is ap-proximately a 14 db perceived loss at50 cpsit takes 14 db more of actualamplitude, in the lower curse, to pro-duce the same relative loudness at 50cps as it does in the upper curve.

In order to re-create the original bal-ance of perceived frequencies at low vol-

Fig. 2.1. The F:etcher-Munson equal k,uel..Mess contours. Forach curve, theheight at any pointrepresents the soundamplitude required toproduce the samesubjective loudnessas at 1000 cps. (AfterFletcher and Munson)

104

a3

0

0

ume levels, it has become customary tointroduce bass boost which is related tothe setting of the volume control, eitherautomatically or otherwise.

A volume control tied to automaticbass boost is called a loudness control.(Some loudness controls also boost thetreble Spectrum appreciably at low vol-ume settings. There is no justificationfor this in the Fletcher-Munson curves.)

High Fidelity to What?

The assumption will be made herethat the purpose of high fidelity equip-ment is to reproduce as closely as possi-ble the experience of the concert hall,not to transcend or improve it.

I remember an exhibition at NewYork's Museum of Modern Art, duringthe late thirties, of "high fidelity" repro-ductions of water color paintings. Life-size reproductions were hung side byside with the originals, and it was oftendifficult or impossible to tell them apart.There was no question in anyone's mindabout how to judge the quality of theseprints. The only criterion was accuracy.The public that visited the exhibit wasused to looking at paintings, and wasable to make an immediate comparison

FEELING20

INt20

onminnummumplawbsaraniateiLISIMUNIM

IIINT:\16211111110111iiiilgenmL11111111M1145111111,111NERNIWA

11111111111111MWIllrliinallal111111111111111181:91.116111:21111/41111111111111111:1111MEMICiall

100

110

20

100 200 500 1000 2000 5000FREQUENCY IN CYC ES PER SECOND

10,000 20.000

; 11111111111111111111111110

.1.01111110111111112111111111111MUME

20DB

IOW

FREQUENCY IN CYCLES PER ....^nNO

between the copy and the original. No.one thought of the prints as entities inthemselves, with qualities independentof the qualities of the originals.

This point of view does not alwayshold in the field of high fidelity musicalreproduction. Only a minority of today'shigh fidelity public are concert-goers.Many have never attended a live con-cert; they know the sound of the orches-tra or of individual musical instrumentsonly as it is reported by amplifiers andloudspeakers. They may know what theylike in reproduced sound, but they haveno way of evaluating the realism ofreproduction.

This partly explains why so much vari-ation is tolerated in audio equipment.The same record may sound very dif-ferent when played through differentbrands of equipment, each brand equallyacceptable in the market place. Theevaluation of high fidelity componentsis popularly thought of as an entirelysubjective matter, like comparing thetone of one violin to that of anotherrather than like holding a facsimile up toits original.

For similar reasons high fidelity dem-onstrations such as the annual Hi-Fishows can get away with a lot of soundthat is startling but essentially non-mu-sical. Some of the "reproduced" soundthat greets the show visitor is necessarilyunfamiliar because it has no live coun-terpart. A harmonica blown up in vol-ume to the dimensions of a theatre organis a new and different instrument. A

Fig. 2-2. The 60 and90 db Fletcher-Mun-son curves superim-posed. The shadedarea represents thedifference in normalhearing loss from onesound level to theother.

crooner whispering into a microphonean inch away invents a new sound; hisunamplified voice is never heard in pub-lic. A combination of Bongo drum,chimes and electric guitar creates a tuttiwhich one may like or dislike, but forwhich there is no equivalent in one'smemory to serve as a live standard.

Such sound can only be accepted asa self-sufficient entity, like an old calen-dar chromo. Any resemblance to livemusic or to painting is purely coinci-dental, and the science and/or art ofreproduction is not really involved.

High fidelity has undoubtedly in-creased rather than decreased the ranksof music lovers, and there are probablymore people than ever who are unim-pressed with gimmick sound. Many de-signers and manufacturers in the fieldwork only for naturalness of reproduc-tion. The designer of integrity avoid,like the plague those exaggerations thatsometimes attract the noviceover-em-phasized bass for "depth," over-em-phasized mid-range for "presence,"over-emphasized treble for "brilliance."These distortions are more properlycalled, respectively, boominess, nasalityor "honkiness," and harshness.

Many demonstrations are not, fortu-nately, of the gimmick type, and usemusical material played at musical lev-els. There have also been concerts stagedwith live musicians, in which direct com-parisons of reproduced sound to thesound of the live instruments could bemade, in the same way that direct com-

High Fidelity

105

parisons of prints to original paintingswere made at the Museum of ModemArt. The live vs. recorded public concertis one method of giving direction toequipment designers and perspective tohigh fidelity consumers. Although trans-ferring concert hall atmosphere to thehome has special problems of its own,success in creating an identity of soundin the concert hall itself solves the majorpart of the problem. Even more vital tomaintaining balance and perspective inthe high fidelity world is live concertattendance.

We are now prepared to discuss thetechnical standards of quality that maybe applied to a sound reproducing sys-tem. There will be no dividing lines pro-posed, at which low fidelity becomesmedium, high, or super.

Frequency ResponseThe frequency response of a sound re-

producing system, or of one of its com-ponents, describes its relative handlingof parts of the input signal which differin frequency. "Handling" may refer toelectrical amplification, as in an amplifier, to conversion of mechanical to elec-trical energy, as in a pickup, or to con-version from electrical to acousticalenergy, as in a loudspeaker.

There are two aspects of frequencyresponse: the range of frequencies han-dled, and the uniformity with which theunit or system responds to different fre-.quencics. Knowledge of the first of theseis useless without knowledge of the sec-ond. Let us therefore pass over the ques-tion of range for the moment, and deter-mine what uniformity will be requiredfor the range we finally decide on.

Uniformity of ResponseAlthough the trained ear can usually

perceive a change of sound level of adb or less in test signals, the averageobserver is probably less sensitive to achange of sound level in a particularfrequency range of a musical passage.

106

Reproduction which remains constantover its frequency range within one ortwo db would thus probably be ade-quate for perfect apparent fidelity, otherthings being equal.

This standard can be met in amplifierswithout much difficulty, even at highpower levels. The best pickups are alsoable to conform, but loudspeakers arelaggard in this respect.

The results of non-uniform reproduc-tion are several. Undue volume in a par-ticular section of the sound spectrumcan produce stridency or boominess asopposed to natural musical sound. Moreparticularly, the existence of sharp peaksin the response c,..rve, usually repre-senting a resonant condition, mean thathangover or ringing may he presentthe speaker cone or section of cone willcontinue to vibrate after the signal hasstopped. This is perceived as a "rain-barrel" effect, a muddying up of thesound and impairment of the distinct-ness of the different instrumental voices.Such an effect is also indicated when thelistener is unable to distinguish clearlythe pitch of low-frequency tones.

Another important effect of peakedfrequency response is the exaggerationof unwanted noise components suchas turntable nimble or record surfacescratch. This effect was not given its duerecognition in the earlier days of highfidelity, when the existence of rumbleand surface noise was proudly displayedas evidence of extended frequency range.

The amount of surface noise in agood quality modern LP record and theamount of rumble from a good recordplayer are such that there will not bemuch significant noise produced in asystem with uniform frequency response,even though the frequency range be ex-tended to the limits of the present stateof the art. In a comparison test con-ducted recently between two tweeters,the one which was able to reproducealmost an octave more of treble (into theinaudible region) showed a dramatic

decrease of surface noise, due to its ex-treme evenness of response. There wasno seloctive reproduction of discrete fre-quency regions, and the switch to thesuperior speaker produced a fuller, morenatural treble simultaneously with thereduction in surface noise.

A similar situation exists with regardto turntable rumble. A peaked systemwhose response falls off rapidly below60 cps may exhibit more turntablerumble than a smooth system whosefull response extends an octave lower.

Tell-tale evidence of the existence ofpeaked reproduction in the bass may begathered from listening to the reproduc-tion of speech. The male speaking voiceordinarily contains no sound compo-nents whose frequency is below 100 cps,and the reproducing system should giveno hint (by a boomy, resonant qualityin the voice) that it is also capable ofspeaking in the tones of the double bass.

Range of ResponseIt is generally agreed among acoustics

authorities that the range of 40 to 15,000cps is sufficient for perfect or near-per-fect apparent fidelity in the reproductionof orchestral music. The phrase "near-perfect" is meant to imply that whensuch a range has been achieved the de-signer should direct his attention to inac-curacies of reproduction more gross thanare associated with the frequency limita-tions indicated.

For the pipe organ enthusiast, how-ever, there is significant intelligence(significant, that is, from the point ofview of the emotional impact of themusic) down to 32 cps or lower. 32.7cps is three octaves below middle C rela-tive to A-440, and is the lowest note ofthe average pipe organ, although manylarger organs reach down an octavelower. These low organ tones are distin-guished by the fact that they containa strong fundamental component. Thelowest tones of the piano, on the otherhand, contain no fundamental energy

that significantly affects the quality ofthe sound. Even though the lowest keyon the piano strikes 27.5 cps, responsedown to this frequency is not requiredfor the reproduction of piano music.

Probably no characteristic of audiocomponents is so freely booted aboutby advertising copywriters as frequencyrange. Any numerical range of frequen-cies listed is totally meaningless unlessaccompanied by a description of thedecibel tolerance above or below refer-ence that is being used, and, for a loud-speaker, by a description of off-axis re-sponse as well. A 3-in. speaker madefor portable radios will "respond" whenstimulated by a 30-cps signalperhapsby having its cone tear loose and fly outinto the airand almost any speaker,even a woofer, will make some kind ofsound when stimulated by a high-pow-e7,!d 15,000-cps signal. A frequency re-sponse rating must mean somethingmore than that a signal of given fre-quency makes a speaker move audibly,or that it makes an amplifier show anelectrical output of some sort at its ter-minals. It must mean that within a statedfrequency range, and, for power devices,within a stated range of power, the fun-damental output of a given device isuniform to a stated degree.

Treble Dispersion

The on-axis response of a loudspeakermay be very deceiving, because thehigher frequencies tend to be directedin a beam which continually narrows asthe frequency is raised. Good sound dis-persion must therefore be a qualifyingfactor for any treble response curve.

A speaker which has relatively uni-form treble output both on-axis and off-axis (over a reasonably large solid angleperhaps 45 degrees in any directionfrom the axis) will reproduce music witha "spaciousness" that does not existwhen there is more concentrated beam-ing of the treble. Furthermore, severelyattenuated off -axis response in the treble

High ro

107

means that the total sound power radi-ated at treble frequencies is considerablyless than that implied by the on-axisresponse curve. It is this total radiatedpower, rather than the on-axis pressure,that determines whether a speaker willsound dull, natural, or over-bright in anormally reverberant room.

Transient RasponstoTransient response refers to the ac-

curacy of reproduction of the waveenvelope, and is concerned with thereproduction of attack and decay char-acteristics of the sound. We have seenthat uniform frequency response predictsthe absence of ringing; if the steady-state frequency response curve does nothave peaks, the reproduced sound willdie away just as in the original.

Consider, for example, the tone repre-sented in (A) of Fig. 2-3. Perfect re-production would produce an identicalwave form, differing perhaps only inamplitude, while poor transient responsewould be indicated by the hangover thatis apparent in (B). The continuation ofthe reproduced signal after the original

END OfMUSICAL TONE

END Of REPRO.DUCED TONE

SHOWINGHANGOVER

A

Fig. 24. Poor transient response.

has ended may be compared to a colorsmear on a reproduced painting.

Attack time involves the reproductionof frequencies higher than the funda-mental. Although a percussive tone mayhave a low fundamental pitch, the fre-quency components associated with itssteep attack characteristic may be veryhigh. Natural reproduction of a drum

108

beat through a two-way speaker sys-tem may thus be accomplished by the"woofer" handling the fundamental toneand its proper decay, while the "tweeter"contributes the sound components thatmake up th.; sharp attack.

Harmonic and intarmodulationDistortion

Reproducing devices have a charac-teristic way of performing with less thanperfect accuracy. In addition to the fre-quencies at which they are asked tovibrate mechanically (or alternate elec-trically) they introduce new modes ofoscillation of their ownand these newfrequencies are harmonics, integral mul-tiples of the original frequency. Thisinaccuracy is called luirmonic distor-tion. It is measured as the ratio of theamplitude of the spurious harmonics tothe true signal, in per cent.

We have seen that harmonics of fun-damental frequencies are produced inany case by musical instruments. Yetsmall amounts of harmonic distortionproduce very unpleasant effects. Thesound becomes harsh, unmusical; thebass is wooden and the treble painful.

The primary reason for this is thatwith harmonic distortion comes an at-tendant evilintermodulation distortion.Intermodulation distortion can be de-scribed as the introduction of new soundcomponents, at sum and difference fre-quencies, when tones of two or more fre-quencies are passed through a non-linearsystemthat is, a system which createsharmonic distortion. These sum and dif-ference frequencies are harmonically un-related to the original musical tones.They are musically discordant, and theyserve to create raucous, unmusical soundin a degree proportional to their relativestrength. The formation of intermodula-tion products is illustrated in Fig. 2-4.

The primary importance of low dis-tortion has always been recognized byaudio authorities. It has also become in-

r

50

I1000 CPS

50 CPS

H1000 CPS

NON-DISTORTINGREPRODUCING

DEVICE

REPRODUCINGDEVICE WITHDISTORTION

...--11.

--

SOLE COMPONENTSARE 50 CPS AND

1000 CPS

COMPONENTSINCLUDE 950 CPS

AND 1050 CPS

Rg. 2.4. Intermodulation distortion as a result of harmonic distortion of the low-frequency waveform. Note that the wave envelope of the high-frequency tone is "modulated."

creasingly recognized by the high fidel-ity public in recent years, after the firstflush of excitement over reproducingregions of the frequency spectrum pre-viously untouched. Amplifier manufac-turers now feature distortion data overfrequency response data; unfortunatelyit is very rare for loudspeaker specifi-cations to make any quantitative refer-ence to distortion at all. The reason liesin the fact that while both harmonic dis-tortion and intermodulation distortion(the latter is usually greater by a factorof 3 or 4) can be kept to extremely lowvalues in high quality amplifiersa smallfraction of one per cent at rated powerthe corresponding values for loudspeak-ers are much higher. In the octave below60 cps it is a rare speaker indeed whichcan hold harmonic distortion, at anyappreciable sound level, below the 5 percent mark over the entire octave, andmany speakers produce percentages ofdistortion in this frequency region ten

times as great. But the listening resultsare not as bad as might appear at firstglance: speaker response is normallyseverely attenuated in this lower range,which helps, and there is comparativelylittle musical material of such low fre-quency to be distorted. .

When the reproducing system has aminimum of low frequency distortion,very low bass tones of high power, suchas might be produced by organ pedalpipes, not only remain pure in timbrethemselves but do not create intermodu-lation with the rest of the music; theydo not destroy the purity of the trebleby introducing false tones.

Power CapabilityThe power capability of a high-quality

reproducing system should be such asto be able to establish an intensity levelof sound in the living room equal to thelevel at a good seat in the original con-cert hall. The electrical power required

High Fidelity

109

110

of the amplifier for achieving this goaldepends upon the efficiency of thespeaker, and the sound power requiredof the speaker depends on the size andother acoustical characteristics of theroom. Concert-hall level can be estab-lished in a living room with a tinyfraction of the acoustical power of asymphony orchestra, because the lowerpower is concentrated in a much smallerarea.

"Concert-hall level" is sometimes mis-interpreted to mean the sound levelwhich would be created if the orchestrawere somehow jammed into the livingroom itself. The writer has yet to ex-perience at a live concert, even duringfortissimo passages, an assault on hisears that compares to hi-fi assaults hehas weathered. It is interesting to notethat certain hi-fi demonstrations pre-clude intelligible conversation which isnot shouted, while whispered conver-sations in a concert hall are liable toprove extremely distracting and annoy-ing to one's neighbors. It is the soundintensity level at the ear, not the powerof the orchestra, that we are trying toreproduce.

Noise Level

Any sound component not present inthe original program material, otherthan distortion products, is referred toas noise, even though it may be periodicand not conform to our strictly scientificdefinition. Hum, rumble, surface scratch,tube hiss or other circuit noise and sim-ilar disturbances tend to destroy theauditory illusion, and must be kept to aminimum.

A standard for satisfactorily low noisehas been established by the FCC for FMbroadcast stations. It is that the powerratio of the maximum signal to the noisemust always be at least 60 db; this rep-resents a ratio of one million to one.

Dynamic RangeThe dynamic range, or range of ampli-

tude of the reproduced sound from soft-est to loudest, is determined by the twofactors just discussed, noise level andpower capability.

Soft musical passages can be maskedby any of the types of noise referred to,and therefore the lowest sound levelsthat can be used must be much louderthan the noise level. The maximum soundlevels that can be used, of course, arelimited by the power capability of thesystem.

A dynamic range of 60 db, or a mil-lion to one power ratio between highestand lowest sound levels, is generallyconsidered adequate for reproductionof the largest symphony orchestra.

stereoAll of the above considerations apply

equally to monaural and to stereophonicreproducing systems. These objectiveelements of equipment fidelitylow dis-tortion, adequate frequency response,dynamic range, etc.are able, in stereo,to contribute more to the. subjective illu-sion of musical reality than in a mon-aural system.

A stereo record-reproduce systemhas in effect two parallel and completemonaural systems. The work of eachcomponent along the way is done twice.The sound is picked up by two separatemicrophones; the output of each micro-phone is recorded on a separate track ofthe tape; the record groove, althoughnot doubled, is cut in such a way as toindependently contain . the record ofeach signal channel; the pickup con-tains tw separate generating elementswhich independently sense and trans-mit each signal channel; the two signaloutputs of the pickup are sent throughindependent amplifiers and fed to twoindependent loudspeakers. There arevariations on this ideal scheme, but theabove describes the basic concept ofstereo.

The purpose of this dual-channel re-production is, in the simplest terms, to

help recreate the acoustical atmosphereof the concert hall. In the old-fashionedstereopticon each visual channel gave aslightly different perspective view of thesubject. Similarly, in stereo recording,each microphone gets a slightly differ-ent auditory perspective. It is importantto note that this auditory perspective isof the orchestra or soloists in the hall inwhich they are performing, not merelyof the musical performers in the ab-stract. This is important because a goodpart of the sound that reaches our earsat a concert does not come directly fromthe orchestra, but is reflected from thewalls and ceiling of the concert hall.

The channels of a stereo system areidentified as "right and "left." This doesnot mean that one microphone picks upthe sound of the right section of theorchestra only, and that the other micro-phone picks up the sound from the leftsection of the orchestra. It does meanthat one microphone has a right-oriented perspective of the total soundin the recording hall, and that the othermicrophone has a left-oriented perspec-tive of the total sound. When these tworecorded channels (which, like the twophotos on a stereopticon card, are verysimilar to each other) are reproducedthrough two separate loudspeakers theycreate, although not perfectly, the illu-sion of the acoustical environment andsense of space of the concert hall. Thereis an increased awareness of the phys-ical position of different instruments,but this is very much less important

than the general increase in realism andthe consequent increase of clarity, par-ticularly from the point of view of thedistinctness of the different musicalvoices.

There is an approach to stereorecording, commonly referred to as"ping-pong" stereo, which provides anexaggerated separation between theright and left channels. If only the leftside of the orchestra were playing dur-ing a particular passage, there would bepractically no sound from the right re-cording channel. The left-right orienta-tion of the different instruments is theprimary goal in this case, rather thanreproduction of the original acousticalenvironment. The degree to which one'sattention is directed to the physicalposition of the instruments in "ping-pong" stereo is often much greater thanthat at the live concert itself.

The greatest benefit of good stereorecording and reproduction is that itfrees us, to a greater extent than waspossible previously, from the acousticalenvironment of the listening room, andtransports us to some extent to theacoustical environment of the hall inwhich the recording was made. Thenormal living room does not provide theproper acoustical atmosphere for amusical concert, particularly of a largeorchestra. Musical instrument designersworked in terms of the tonal qualitiesthat would be produced in the type ofconcert hall with which they werefamiliar.

111

THE SOUND REPRODUCING SYSTEM

MHE PHONOGRAPH iS a classic example1 of an invention that cannot be cred-

ited wholly to one man. In 1877 Edisondirected his assistant, John Kruesi, toconstruct the first complete record-reproduce system, but sound recorderswere sold on a commercial basis as earlyas 1880, and Thomas Young's "A Courseof Lectures on Natural Philosophy de-scribed and illustrated a crude but prac-tical sound recorder in 1807.

Young's recorder consisted of a sharpmetal stylus held by spring tensionagainst a revolving cylinder, the cylindercoated with wax and turned by a gov-ernor-controlled gravity motor. Whena vibrating body such as a tuning forkwas held against the stylus, a wavyline was cut into the wax. This line rep-resented the wave form of the vibra-tions, and it could be studied and ana-lyzed at leisure. The recorder was amechanical draftsman, that could sensevery small motions and record pressurechanges that took place within a periodof a very small fraction of a second.

By 1858 Leon Scott de Martinvillehad constructed the "phonautograph"

112

(self-writer of sound) illustrated inFig. 3-1. The sound wave form wasscratched by a hog-bristle stylus on thesurface of a cylinder coated with lamp-black, but the big advance over Young'smachine was the fact that the phonauto-graph could record directly from theair. The force of the acoustical vibrations

Fig. 3-1. The phonautograph of Lion Scoff deMartinvillea commerdal sound recorder ofthe eighteen sixties. (Courtesy SmithsonianInstitution)

was concentrated by a horn onto a dia-phragm, and the stylus was attached tothe diaphragm, so that the recordingneedle did not have to actually touchthe vibrating source of sound. This de-vice, which corresponds in function tothe modern oscilloscope, was a catalogueitem of the Paris firm of Koenig, and wassold as a measuring instrument to acous-tical laboratories.

The phonautograph which is at theSmithsonian Institution at Washingtonwould undoubtedly reproduce music ifa proper record were placed on its re-volving cylinder. The theoretical possi-bility of playback was understood then,too, but the lampblack records were use-less for playback, as their grooves werenot rigid enough to direct the vibrationsof a playback needle. About half a yearbefore Edison got his brainstorm CharlesCros conceived a method for bringingthe groove sinuosities back to life as

sound. The lampblack recording was tobe photo-engraved on a metal cylinder,and running a needle through the hardgroove would then cause the needle tovibrate from side to side, in the sametime pattern as the hog bristle stylusthat first inscribed the line.

For reasons which may be related tonineteenth century differences in tradi-tion between the scholar and the indus-trial engineer, Cros didn't even constructa working model, but merely filed a com-plete, sealed description of his systemwith the Academie des Sciences. On theother hand, less than a month afterEdison first conceived of a reproducingphonograph the country was readingabout a working unit in newspaper head-lines. There was a great stir of excite-ment over this amazing tonal imitator,(see Fig. 3-2) with public demonstra-tions, lectures before august scientificbodies, and a visit to the White House.

4

Fig. 3.2. Edison with his tin-foil phonograph. (Photograph by Brady courtesy Smithsonian Insti-tution)

High Fq1el,tv

113

ft/

The excitement soon died down, asthe Edison machine was an impracticaltoy, with neither permanent records norusable fidelity. The recorded groove wasindented into a semi-hard material, tinfoil; it was only able to retain its shapepartially, and that for very few play-ings. Subsequent technical improve-ments, however, made the phonographa popular device by the turn of the cen-tury. It is curious that our modern re-cording system, in which the record is amechanical copy of the original master,is more closely related to Cros' systemthan to Edison's. Emil Berliner, thefather of the moulded or cast record,began his research work by successfullycarrying out Cros' proposals.

The Mechanical or "Acoustic"Phonograph

It would be useful to consider the de-sign of the non-electric phonograph, asillustrated in (A) of Fig. 3-3. A betterinsight can thereby be gained into the

A

B

Fig. 33. (A) Thy mechanical phonograph. (6)Th. *tunic phonograph.

114

function of the various components ofa modern electronic system.

The wave forms frozen intc the recordgroove control the vibrations of the play-back stylus when the groove is draggedpast the stylus by a revolving turntable.These stylus vibrations, although theycontain a fairly large amount of me-chanical energy, engage practically noair, like the revolutions of a bladelesselectric fan. The needle is thereforeattached to a diaphragm, which vibratesin sympathy with the stylus and has amuch larger surface area in contact withthe air of the room.

But even the reproducing diaphragmdoesn't get a sufficient bite of the airfor practical purposes. Therefore the dia-phragm is placed at the narrow throatof an acoustical horn, and the actualusable sound emerges into the roomfrom the much larger mouth of the horn.The system works somewhat as thoughthe diaphragm area were really that ofthe horn's mouth.

It can be seen that all of the energyradiated by the horn is taken from themechanical vibrations of the needle, andthe forces between needle and recordgroove are necessarily great. This hasobvious implications for record wear,but perhaps more important, the de-mands for power placed on the "sound

.x" or "speaker" (old-fashioned termsfor the needle-diaphragm-head assem-bly) place a severe limitation on musi-cal fidelity. High distortion and peakedand severely limited frequency responseare to be expected.

The Phonograph AmplifierThe solution to this problero lies in

changing the function of the phono-graph pickup, from the primary genera-tor of sound power to a device whichcontrols an outside source of power. Ifthe power from the outside source ismade to oscillate in imitation of theneedle vibrations, two benefits can result:

1. The final output sound derived

from the record groove can be muchlouder.

2. The power ....tmands on thepickup itself are I, ) longer heavy. Thepickup can be designed for qualityrather than loudness; the problemsof achieving uniform, extended fre-quency response and low distortionare considerably lessened. So, inci-dentally, is the required weight onthe pickup and the grinding away ofthe record groove.

The control of an outside source ofpower to conform to given oscillationsis called amplification. The first phono-graph amplifier was pneumatic: theneedle was made to actuate an air valve,which periodically throttled a flow ofcompressed air. Most of the work of ra-diating sound power was thus performedby the air compressor, and the styluswas relieved of part of its burden.

All modern sound reproducing sys-tems use amplifiers, but unlike the firstpneumatic systems these amplifiers areelectronic. The phonograph pickup is nolonger a sound generator but an elec-tric generator. It produces small alter-nating voltages at its terminals, whosewave forms conform to those of thegroove and of the recorded sound. Thepickup has to generate very little power,because the output voltage can be ampli-fied to almost any desired degree. Theamplified electrical power must finally,of course, be converted back into soundby a loudspeaker. The two types of re-producing system, electrical and purelymechanical, are shown in Fig. 3-3.

Th Medium Sound RoproduoingSystem

The purpose of the historical approachused above has been to furnish thereader with an appreciation of the rea-son for the modem audio system beingdesigned as it is. With the electronicamplifier supplying the brute force, soto speak, the mechanical componentspickup and loudspeakercan be built

in such a way as to suppress the naturalresonant tendencies inherent in mechani-cal vibratory systems

Before discussing each of the audiocomponents in detail, it would be usefulto make a brief survey of the entire re-producing system. A complete monauralsystem is illustrated in Fig. 3-4.

First of all the disc record must berevolved by a motor and turntable. Thechief operational requirements of thispart of the system are that it revolve atthe correct speed, that the speed be con-stant, and that JA.raneous vibrationsdo not communicate themselves to thepickup.

The first of these requirements is forthe purpose of keeping the reproducedmusic at the same absolute pitch atwhich it was recorded: too fast a turn-table speed will make the pitch sharp,and too low a speed will make it flat.The second condition listed, constantspeed, is required in order to avoid pitchvariations, or "wow." The third require-ment, lack of extraneous vibrations,keeps low-frequency noise called "rum-ble" out of the final sound.

The groove variations are sensed bythe needle, or stylus, which in high-quality systems is jewel tipped; it isusually diamond. The needle must havean unmarred, smooth surfaced, hard tip,normally of spherical shape.

The pickup is an electric generator(usually either of the piezo-electric,variable reluctance, or moving-coil type)whose function is to translate the me-chanical vibrations of the needle intoelectrical oscillations of the same waveform. It must do this with minimumdistortion of the wave form, and mustnot allow resonances of its own to in-fluence its output voltage significantly.It is also an advantage for the pickupto impose as little work as possible onthe needle. The greater the force re-quired for the groove to displace theneedle from side to side, the greater thevertical bearing force will have to be to

115

116

TAPE MLCHANISM

PICKUP

STYLUSTONE ARM

MOTOR

TURNTABLE

to ofTUNER

I" "I

(limn 00--- ii06._

1

CONTROL UNIT POWER AMPLIFIER

fig. 3.4. Diagram of a complete monaural sound reproducing system.

maintain proper and constant stylus-groove contact, and the greater thewear of both record and needle.

The tone arm holds the pickup inplace over the groove, and must pro-vide sufficient freedom of motion sothat the pressure of the groove wallsalone can make the needle move acrossthe record, following the recorded spiral.It must also be free enough to followwarp and eccentricity of the disc easily.The tone arm must hold the pickup ap-proximately tangent to the groove beingplayed, must provide the proper verticalforce for the pickup, and must not allowits own resonant behavior to influencethe system.

The electrical output of one type ofpickup, the piezo-electric, is usually feddirectly to the amplifier. It is of theorder of 1/2 volt or more, and is a fairlyaccurate replica of the recorded sound.This is so because the characteristic fre-quency response of the pickup is moreor less the inverse image of the frequencycharacteristics "built in" to the record.(This last subject wili be taken u1, indetail later.)

The reluctance end moving-coil pick-ups, however, produce a much smalleramount of electrical energy. The output

et:0

SPEAKER

SYSTEM

voltage of these pickups (which areclassed together as magnetic types) maybe as low as a few thousandths of avolt. Furthermore the characteristic fre-quency response of the magnetic pickupdoes not compensate for the way inwhich the frequency characteristics ofthe recorded sound has been doctored.Therefore the pickup output must bepassed enough a preamplifier before itenters the amplifier proper.

The preamplifier is normally com-bined with the main amplifier controlsections (volume and tone controls).Its functions are to increase the outputvoltage of the pickup, and to compensateaccurately for the frequency character-istics of the record so that the sound isnot deficient in bass and heavy in thetreble. Since different record companieshave made records with different char-acteristics the preamplifier may allowthe operator to choose between severaltypes of frequency compensation. Theneed for such control, which is calledvariable record equalization, has disap-peared with modem records, which arestandardized on the RIAA recordingcharacteristic.

The control section of the amplifierallows the operator to regulate the vol-

ume, and, in most cases, to either ac-centuate or attenuate ("boost" or "cut")the bass and treble portions of the repro-duced sound independently. The pri-mary function of tone control is to com-pensate for deficiencies in asscciatedequipment or program material, and tocompensate for acoustical conditions ofthe room in which the music is heard.When the control section and phono-graph preamplifier are combined on onechassis, the entire unit is commonly re-ferred to as a preamplifier.

The power amplifier receives the elec-trical signal as it is finally shaped, andreleases another signal, ideally identicalin all respects except power. The poweramplification may be tens of millionsof times, from a fraction of a micro-watt (one millionth of a watt) to dozensof watts.

Although the demands on the ampli-fier are very great, and although itappears to be the most complicated ofthe system components, it is the leastimperfect of these components. The per-centars of harmonic and intermodula-tion distortion, the irregularities of fre-quency response, and the extraneousnoise introduced by an amplifier builtaccording to the best current designpractice, and without regard for cost,

STEREO TAPE MECHANISM

STEREOPICKUP

STYLUSTONE ARM

MOTOR

TURNTABLE

I 0 0 IMULTIPLEX FM TUNER

34. A stereo reprodudfly system.

are such that they are not limiting fac-tors in the fidelity of the reproducedsound.

The final component of the soundsystem is the loudspeaker system, whichconsists of the speaker mechanism itselfand the speaker enclosure. The loud-speaker converts the alternating elec-trical output of the amplifier into me-chanical vibrations of a cone or dia-phragm. But the cone vibrating by itselfcannot, for reasons that will be discussedfurther on, produce adequate bass en-ergy. It must be mounted in an enclo-sure or baffle of some sort, which givesthe vibrating surface the "bite" of airthat it needs to radiate low-frequencysound.

The speaker and its enclosure, like theamplifier, should introduce as little dis-tortion and frequency irregularity intothe signal as possible. Typical speakerdeficiencies are irregular frequency re-sponse, poor transient response (hang-over), and harmonic and intermodula-tion distortion.

Two other components are shown inFig. 3-4. The tuner is a device whichconverts AM or FM radio signals toaudio signals that can be handled by theaudio amplifier; the tape transportmechanism, with its asscciated pream-

Minn 00CONTROL UNIT POWER AMPLIFIER

(inn 00CONTROL UNIT

SPEAKER

SYSTEMS

POWER AMPLIFIER

High Fltielity

1

plifier, provides a signal of the samenature as that coming from the tuneror phonograph pickup.

Fig. 3-5 shows the basic elements ofa stereo reproducing system. The stereotape mechanism has two heads whichindependently reproduce each channelthat is recorded in parallel on the tape;the stereo pickup provides two separateoutput signals from the two channelsrecorded in the groove (the turntableand pickup arm do not have to be dupli-

118

cated); the stereo tuner receives the"multiplex" FM stereo signal and sepa-rates it into two separate channels,which it feeds independently to each ofthe control units. Each control unit andeach power output is shown duplicated.The two control units and power ampli-fiers may be separate, or they may becombined on one chassis, or all fourunits may be combined on one chassis,but in any case they must provide inde-pendent amplification for ea,:h channel.

The Future of Direct Current Power Transmission

N. L. Allen

A popular article published in 1967.

The history of technology provides many examples ofunexpected turns of fortune, and electrical technology isno exception. It frequently happens that a principle ortechnique, originally the basis of a well-establishedsystem, is superseded by a device making a significantadvance, only to reappear in a different guise as the 'lastword' in the state of the art. An obvious example is thecrystal of the early radio receiver. This was supersededby the thcrmionic valve, but it has now developed intothe more sophisticated form of the transistor. Not manyyears before the era of the crystal receiver, an appreciableproportion of electrical energy was generated, trans-mitted, and used to the form of direct current. At thattime, generation and consumption usually took place inthe same locality, distribution was simple, and thequantities of energy transmitted were small by modernstandards. However, serious limitations appeared as itbecame necessary to distribute electrical energy morewidely, and direct current as the distributing mediumgave way to alternating current.

In many countries, the economic advantages of beingable to concentrate power generation in large stationshave led to the adoption of a comprehensive network ofpower lines that interconnect generating plant and theareas where the power is used. As the length of a powerline increases, the current passed, for minimum powerloss, decreases: the economic operating voltage for trans-mission of a given power therefore increases. The trans-mission of larger quantities of energy at high voltagesand low currents is greatly facilitated by the ease withwhich alternating current can be transformed to thevoltage most appropriate for the power lines. In thereceiving areas of the system, the voltage can equallyeasily be transformed to lower values suitable for distri-bution, and a system of far greater flexibility can be setup than is the case with direct current. Further, it is

difficult to switch and, particularly, to interrupt directcurrent. The interruption of an alternating current bycircuit breakers is relatively easy because the currentpasses through zero twice in every cycle.

This combination of circumstances made alternatingcurrent the natural choice as power systems increased insize. The main links operated initially at 132 kilovolts,but the need for increased power during the post-waryears has led to the adoption of 275 kilovolts and, morerecently, 400 kilovolts as the operating voltages of theprincipal links in Britain. The power is distributed locallyat lower voltages. During this period, the remainingdirect current distribution sy.tems have been reduced oreliminated.

Transmission over long distances

What, then, is the place of direct current? There iscertainly no good reason for turning away completelyfrom alternating current distribution. But there havealways been some situations in power distribution prac-tice M which direct current has distinct advantages overalternating current, and it is worth while consideringwhat these situations are.

One bask factor in power system design is the need tofind the simplest and most efficient means of transferringpower from one point to another. Figure t(a) shows thebask three-phase alternating current system and figuret(b) a favoured direct current system, which has positiveand negative polarities on the two lines, and is linked byconvertors to alternating current for generation at oneend and distribution at the other. In both cases, themaximum voltage to earth is E, but for alternatingcurrent, it is the root-mean-square value E/Y2 thatdetermines the power transmitted. This is 3E/A CO3 P / 1/a,where /A is tilt current in each conductor, lagging behindthe voltage it phase by q, degrees. In the direct currentsystem, the p. wcr transmitted by each line is LID, wherelo is the current. For transmission of equal power by thetwo systems, therefore, it can be shown that eachalternating current line has 4/(3 cost tp) times the crosssectional area of the corresponding direct current line, afactor which is always greater than 1.33. Moreover, the

119

alternating current system requires three cables ratherthan two, so that the amount of copper required is2/cost e) times that in the direct current system, a factorwhich is always greater than 2.

Direct current, then, reduces the cost of the cables.This may appear trivial compared with the other capitalcosts in electrical systems, but over great distances, as inthe United States and the Soviet Union, the saving incable, and in the means of supporting the cable, becomesa very significant factor that can outweigh the cost ofproviding the convertor stations at each end of thesystem.

Great distances bring further problems in alternatingcurrent transmission that do not occur with directcurrent. These problems arise from the relationship be-tween the wavelength of the oscillation and the dimen-sions of the system. The quarter-wavelength of a 50cycles per second wave in air is about goo miles, and thetransmission of energy through a conductor can be re-garded as due to an influx of energy along its length fromthe electromagnetic field that surrounds it. Over shortdistances, this field is very nearly the same at all points,since electromagnetic energy is conveyed with the

(a)

(b)

Figure 1 Simplified distribution systems: (a) alternatingcurrent, (b) direct current,

rvelocity of light. But at distances greater than goo miles,the fact that the velocity of light is finite results insignificant differences, at any instant, in the phase of thecurrent at the two ends.

This situation leads to difficulties where two parts of apower circuit, joined by a long alternating current link,are out of phase and where a loop is formed throughanother part of the network of different length. Largecirculating currents will be set tip unless some form ofcompensation is applied. A direct current link obviatesthese diffiCulties; as a corollary, it may be noted alsothat if a direct current line is used to link two alternatingcurrent systems, they need not be synchronized witheach other.

120

Transmission over short distances

For long-distance transmission, overhead lines, supportedby towers, are used. The virtues of direct current arcmost clearly shown when the current is carried by under-ground or underwater cables. Here, the central core ofthe cable, which is at the transmission voltage, is sur-rounded by an insulant, the exterior of which is at earthpotential. This constitutes a coaxial capacitor, and thecapacitance per mile of a cable rated at 200 kV istypically about 0.3 microfarads. In ass alternating currentcircuit, this capacitance is charged and discharged,through the inductance and resistant" of the cable itself,once every half-cycle. Additional generating capacity isneeded to supply this charging current. In the examplequoted, at 200 kV, the charging current requires about5000 kilovolt-amperes per mile of cable [1, 2); at 40o kVthe figure is about 15 000 kilovolt-amperes per mile. Forappreciable lengths of cable, the losses become such thatthe charging currents must be supplied at intermediatepoints. At 200 kV, these points are about 25 miles apartfor 50-cycle alternating current; at 400 kV, only 15miles. Thus, alternating current transmission becomesimpracticable in cables over long distances. Further, thecost of the generating capacity needed to supply thecharging current is significant. Taking a rough figure ofZs() per kilowatt of installed capacity at the generatingstation, this extra cost is £250 000 per mile for a 200kilovolt cable. By contrast, with direct current in thesteady state, there is no charging current. It may well beworthwhile, therefore, to accept the cost of convertingto direct current to avoid having to provide this charg-ing current. Direct current is also advantageous in thatthere are no dielectric losses due to reversal of the elec-tric stress in the insulant.

The balance between the two systems

To summarize, direct current has significant advantagesfor the transmission of bulk power over great distancesby overhead lines, and over short or long distances bycable. In addition to the technical advantages alreadyexamined, direct current may be valuable in linking twoalternating current systems that need not then besynchronized. Alternatively, a very large alternatingcurrent system may be divided by direct current linksinto two or more smaller systems: this is a possible futuredevelopment as power systems continue to increase insize. It is necessary, however, to examine some disad-vantages of direct current, and some relevant non-technical factors, to demonstrate the balance affectingthe final choice of system.

The most obvious drawback to the use of directcurrent is the need for conversion at each end of the linkin order to integrate it with established alternatingcurrent systems. The technical details are outlined later,but it may he mentioned here that the cost of the con-version equipment is about twice that of the alternatingcurrent equipment required for the termination of apower lint of corresponding size and output [3). Thesecosts must be set against the savings inherent in thedirect current system. There is therefore, a limit to thelength of a line, below which the capital outlay on adirect current system is higher than that of an alternatingcurrent system. Estimates of the critical length for a longoverhead line naturally vary, depending mainly on thepower to be transmitted and the voltage to be employed,but figures of more than 300 miles have frequently been

The Future of Direct Current Power Transmission

quoted [4]. This approach is unlikely to be favoured,therefore, in the British Isles, but such systems are beingdeveloped in the United States and in the Soviet Union.For underground or submarine cables, where dielectriclosses and charging currents are so important, the'critical length' is reduced to about 3o miles, and it is inshort submarine links and in urban transmission linesthat direct current finds its second important application.Indeed, where large amounts of power have to be intro-duced into large cities, legal and social considerationsmay predominate over technical and economic factors.It is frequently extremely difficult to obtain permissionto erect overhead lines in urban areas, and the distur-bance to local amenities caused by the towers for high-tension cables may not be justifiable, Undergroundcables become necessary, and it is preferable to usedirect current for distances greater than about 3omiles.

In choosing between the systems, the fact that therecan be no direct current transformer and that there is nosatisfactory circuit breaker ensures that alternatingcurrent maintains its general superiority for distributionpurposes. The use of direct current is thus confined tothe bulk transmission of high power between discreteparts of a system or between two separate systems.

121

.,

P

,

3:...

*

Electric generators at Fontana Darn, North Carolina.

122

l

James Clerk Maxwell, Part II

James R. Newman

A biographical essay published in 1955.

In February, 1858, Maxwell wrote a letter to his aunt, MissCay, beginning, "This comes to tell you that I am going to havea wife." "Don't be afraid," he added, "she is not mathematical,but there are other things besides that, and she certainly won'tstop mathematics." His engagement to Katherine Mary Dewar,daughter of the principal of Marischal College, was formallyannounced the same month, and in June they were married.

123

124

Their union became very close: they enjoyed doing thingstogether horseback riding, reading aloud to each other,traveling and he even found useful tasks for her in hisexperimental work. The marriage was childless, but this veryfact increased the couple's dependency and devotion. Maxwellregarded the marriage tie in an "almost mystical manner."The published letters to his wife overflow with religiosity.*

The Aberdeen appointment terminated in 1860 when thetwo colleges, King's and Marischal, were fused into a newuniversity and Maxwell's chair in physics at Marischal waseliminated. He was not long at liberty. In the summer of thesame year he became profes6or of natural philosophy atKing's College, London, a post he retained until 1865. Theteaching schedule at King's was long and arduous; in theevenings there were lectures to be given to "artisans" as partof his regular duties. Living in London offered him the oppor-tunity to see something of varaday, with whom, up to this time,Maxwell had had only corr.'spondence, to make the acquaint-ance of other scientific man and to renew old friendships. Hewas no solitary. "Work is good, and reading is good, butfriends are better," he wrote to his friend Litchfield.

Yet despite academic duties and social distractions, the fiveyears in London were the most productive of his life. Thepaper "On the Theory of Three Primary Colors," the twoarticles in the Philosophical Magazine on "Physical Lines ofForce" and the culminating electrical memoir "A DynamicalTheory of the Electromagnetic Field," the Bakerian lecture"On the Viscosity or Internal Friction of Air and other Gases,"and the celebrated paper "On the Dynamical Theory of Gases,"all belong to this period. He also performed important experi-mental work during these years. At his house in Kensington,

He did not write in this vein to others and it is a little puzzling why he foundit necessary in corresponding with her to quote Scriptures, to express the fervent hope that the Lord would protect her from evil, and that she would gether eyes off "things seen and temporal and be refreshed with things eternal."

J ,%

in a large garret, he measured the viscosity of gases and ob-tained practical confirmation of the theoretical work I havedescribed. (For example, he found that the viscosity of air at12 millimeters of mercury measured the same as at normalpressure of 760 millimeters, thus proving that viscosity is in-dependent of density.) To maintain the necessary temperature,a fire had to be kept up in the midst of very hot weather andkettles kept boiling to produce steam, which would be allowedto flow into the room. Mrs. Maxwell acted as stoker. Anotherinvestigation dealt with the ratio of the electromagnetic to theelectrostatic unit of electricity and led to one of Maxwell'sgreatest discoveries. But I must postpone explaining this work,even though to do so means abandoning the strict chronologyof events in Maxwell's life, until I have sketched the develop.ment of his ideas on electricity.

To gain an appreciation of Maxwell's stupendous contribu-tion to this branch of science it is necessary first to describevery briefly the position of electrical theory when he embarkedon his studies.

In the eighteenth century, Charles Augustin de Coulombestablished the fundamental facts of electrostatic attractionand repulsion. He showed that an inverse-square law, resem-bling that of gravitational forces, applied to electric charges:attraction or repulsion between charged bodies is directlyproportional to the product of the charges and inversely pro-portional to the square of the distance between them.* (Thesame discoveries, and others going beyond them, were madeearlier by the brilliant English recluse Henry Cavendish, buthis researches remained unpublished until 1879.) The nextmajor advance was that of Hans Oersted, who in 1819 foundthat the flow of electric current through a wire parallel to amagnetic needle makes the needle swing to a position at right

F = k '7 ' where F equals the force; k, a constant; q and q', the charges;r2

r, the separating distance.

Pr,

125

angles to the current. In other words, a current produces, amagnetic field.

A compleentay series of advances was made early in thesame century by the French physicist and mathematicianAndre Ampere, whom Maxwell called the Newton of electric-ity. The accolade was not undeserved, but there is a specialreason for Maxwell's conferring it: Ampere was the first toexplain the relationship of electric currents in terms of me-chanical action,* an approach later perfected by Maxwellhimself. By experiment Ampere learned that a coil of wirecarrying an electric current behaves like a magnet, and hesuggested that a magnet owes its property to tiny electricalcurrents inside the steel molecules. Thus a great conceptuallink was forged; for magnetism was shown to be not distinctfrom electricity, but rather a name we give to some of theeffects of moving electric currents.

The crown of these fundamental researches was the im-mortal work of Michael Faraday. He found that an electriccurrent flowing in one circuit can cause ("induce") a currentto flow in another circuit; that there is a magnetic field betweentwo currents; that a current can also be induced to flow in awire by use of a magnet in other words, as a symmetriccounterpart to the phenomena discovered by Oersted andAmpere, that changes in a magnetic field produce an electriccurrent.

Faraday's explanation of these phenomena is of centralimportance to understanding Maxwell's work. He imaginedlines of force running through space as the instrumentality ofelectric and magnetic actions.

These lines, it should be emphasized, were not conceived asmere mathematical makeshifts, but as entities possessing phys-ical properties. The lines spread out in every direction froman electric charge or magnetic pole; every electric line of force

* He showed how to calculate the mechanical forces between circuits carryingcurrents, from an assumed law of force between each pair of elements of thecircuit.

126

starts from a positive charge and ends on a negative charge;the more powerful the source, the more lines emanate from it.Along the lines there is tension, a kind of pull, so that eachline behaves like an elastic thread trying to shorten itself; linesof force repel each other sideways; the ends of a line of force.representing charges. can move freely over the surface of aconductor but are anchored on an insulator.

This hypothetical system, for which Faraday was convincedhe had found experimental evidence, was the starting point ofMaxwell's studies. He believed in it he sought to develop it.

However, it must not be supposed that everyone acceptedFaraday's hypothesis. In fact, the majority of electricians Iuse the term in its older sense regarded lines of force as aconcept much inferior to that of "action at a distance." Theylikened electricity to gravitation. They imagined a charge (ormass) situated at one point in space mysteriously influencinga charge (or mass) at another point, with no linkage or con-nection of any kind, however tenuous, bridging the distancebetween the charges (or masses). Where Faraday sought toassimilate the behavior of electricity to that of a mechanicalsystem, in which all parts are joined by levers, pulleys, ropesand so on, the others held electricity to be a special case, towhich mechanical analogies were inapplicable. Maxwell ad-mirably summarized the cleavage between the two views:"Faraday, in his mind's eye, saw lines of force traversing allspace, where the mathematicians saw centres' of force attract-ing at a distance; Faraday saw a medium where they saw noth-ing but distance; Faraday sought the seat of the phenomena inreal actions going on in the medium, they were satisfied thatthey had found it in a power of action at a distance impressedon the electric fluids."

Maxwell's first electrical paper "On Faraday's Lines ofForce" was delivered at Cambridge in 1855, within a fewmonths after he had finished reading Faraday's ExperimentalResearches. What he tried to do was imagine a physical modelembodying Faraday's lines, whose behavior, like that of any

127

machine, could be reduced to formulae and numbers. He didnot suggest that the model represented the actual state of things;on the other hand, he had no confidence in what mathematicalmanipulations alone would reveal about the actual state ofthings. It was important, he said, so to balance the method ofinvestigation that the mind at every step is permitted "to layhold of a clear physical conception, without being committedto any theory founded on the physical science from which thatconception is borrowed."* Such a method will neither lead

The opening paragraph of the paper is worth giving in full. "The presentstate of electrical science seems peculiarly unfavorable to speculation. The lawsof the distribution of electricity on the surface of conductors have been analyt-ically deduced from experiment; some parts of the mathematical theory ofmagnetism are established, while in other parts the experimental data are want-ing; the theory of the conduction of galvanism and that of the mutual attrac-tion of conductors have been reduced to mathematical formulae, but have notfallen into relation with the other parts of the science. No electrical theory cannow be put forth, unless it shows the connection not only between electricity atrest and current electricity, but between the attractions and inductive effects ofelectricity in both states. Such a theory must accurately satisfy those laws themathematical form of which is known, and must afford the means of calculat-ing the effects in the limiting cases where the known foi-mulae are inapplicable.In order therefore to appreciate the requirements of the science, the studentmust make himself familiar with a considerable body of most intricate mathe-matics, the mere attention of which in the memory materially interferes withfurther progress. The first process therefore in the effectual study of the science,must be one of simplification and reduction of the results of previous investiga-tion to a form in which the mind can grasp them. The results of this simplifica-tion may take the form of a purely mathematical formula or of a phys;eal hypoth-esis. In the first case we entirely lose sight of the phenomena to b'. explained;and though we may trace out the consequences of given laws, we can never ob-tain more extended views of the connections of the subject. If, on the otherhand, we adopt a physical hypothesis, we see the phenomena only through amedium, and are liable to that blindness to facts and rashness in assumptionwhich a partial explanation encourages. We must therefore discover some meth-od of investigation which allows the mind at every step to lay hold of a clearphysical conception, without being committed to any theory founded on thephysical science from which that conception is borrowed, so that it is neitherdrawn aside from the subject in pursuit of analytical subtleties, nor carried be-yond the truth by a favorite hypothesis. In order to obtain physical ideas with-out adopting a physical theory we must make ourselves familiar with the exist-ence of physical analogies. By a physical analogy I mean that partial similaritybetween the laws of one science and those of another which makes each of themillustrate the other. Thus all the mathematical sciences are founded on rela-tions between physical laws and laws of numbers, so that the aim of exact sci-ence is to reduce the problems of nature to the determination of quantities byoperations with numbers."

128

(, Pli;s7

into a blind alley of abstractions, nor permit the investigatorto be "carried beyond the truth by a favorite hypothesis."

Analogies are, of course, the lifeblood of scientific specula-tion. Maxwell gives a number of examples, among them theilluminating suggestion of William Thomson comparing theformulae of the motion of heat with those of attractions (suchas gravitation and electricity) varying inversely as the squareof the distance. To be sure, the quantities entering into heatformulae temperatu flow rri heat, conductivity aredistinct from a quantity such as force entering into the formu-lae of inverse-square attraction. Yet the mathematical laws ofboth classes of phenomena are identical in form. "We haveonly to substitute source of heat for center of attraction, flowof heat for accelerating effect of attraction at any point, andtemperature for potential, and the solution of a problem inattractions is transformed into that of a problem of heat. "*

Maxwell proposed a hydrodynamical analogy to bring be-fore the mind in "convenient and manageable form those math-ematical ideas which are necessary to the study of the phe-nomenon of electricity."t The analogy was combined withFaraday's lines of force, so that they were converted fromlines into "tubes of flow" carrying an incompressible fluidsuch as water. He was then able to show that the steady flow ofparticles of this fluid would give rise to tensions and pressurescorresponding to electrical effects. The fluid moving through asystem of such tubes represented electricity in motion; theform and diameter of the tubes gave information as to strengthand direction of fluid (electric) flow. The velocity of the fluidwas the equivalent of electrical force; differences of fluid pres-sure were analogous to differences of electrical pressure orpotential. Since the tubes were flexible and elastic, and ar-

"On Faraday's Linea of Force," Transactions of the Cambridge PhilosophicalSociety, vol. X, part 1, included in The Scientific Papers of James Clerk Max-well, op. cit.

t Ibid.

129

ranged so as to form surfaces each tube being in contactwith its neighbors pressure transmitted from tube to tubefurnished an analogy to electrical induction.

One of Faraday's key concepts deals with the effect on spaceof lines of magnetic force. A wire introduced into ordinaryspace remains inert; but if magnetic lines of force are intro-duced into the space, a current flows through the wire. Faradayexplained this by saying that the introduction of the magnetthrew the space into an "electro-tonic state." This conceptcould not be fitted into the hydrodynamical analogy; Maxwelladmitted that while he could handle Faraday's conjecturemathematically, the electro-tonic state at any point of space be-ing defined "as a quantity determinate in magnitude and direc-tion," his representation involved no physical theory "it isonly a kind of artificial notation."*

It was a wonderful paper, and Faraday, to whom Maxwellsent a copy, appreciated how much it advanced the "interestsof philosophical truth." "I was at first almost frightened," hewrote Maxwell, "when I saw such mathematical force made tobear upon the subject, and then wondered to see that the sub-ject stood it so well. "t Other students, however, thought thesubject stood it not at all well. Electricity was mysteriousenough without adding tubes and surfaces and incompressiblefluids. But Maxwell, who had good training in being consid-ered queer, went on with the task of extending Faraday's ideas.

The second great memoir, On Physical Lines of Force, aseries of three papers published in the Philosophical Magazinein 1861 and 1862, was an attempt to describe a more elaboratemechanism that would not only account for electrostatic effectsbut also explain magnetic attraction and Faraday's concept of

For a discussion of Maxwell's use of physical analogy. see Joseph Turner,"Maxwell on the Method of Physical Analogy," The British Journal for thePhilosophy of Science vol. VI, no. 23, November, 1955.

t Campbell and Garnt.t, op. at., p. 519.

electromagnetic induction. Again, Maxwell used a concrete,mechanical image to exhibit and develop his theory.* For, ashe said, "scientific truth should be regarded as equally scien,tic whether it appears in the robust form and vivid colouringof a physical illu -ion or in the tenuity and paleness of asymbolic expressicn.

In the new model a magnetic field is produced by the rota-tion in space of what Maxwell called "molecular vortices."These may be thought of as slender cylinders (Maxwell him-self had a disconcerting way of modifying the image as hewent along) that rotate round the lines of magnetic force. Thelines, traced by the pattern of iron filings about a magnet, arethe axes of rotation of the cylinders: the velocity of rotationdepends on the intensity of the magnetic force. Two median-ical effects are associated with the cylinders: tension in thedirection of the lines of force, and pressure, exerted in the"equatorial" direction (i.e., lateral pressure), arising fromthe centrifugal force produced by the rotating cylinders. Com-bined, these effects mechanically reproduce magnetic phenoena: magnetism is a force exerted both along the axis andoutward from the axis.

It may now be asked how this curious arrangement fitted inwith the known facts that an electric current produces a mag-netic field, and changing magnetic forces produce an electriccurrent. Step by step Maxwell answers this question.

The first point to clarify is the structure of a uniform mag-netic field. Maxwell supposed this to consist of a portion ofspace filled with cylinders rotating at the same velocity and inthe same direction "about axes nearly parallel." But immedi-ately a puzzle confronted him. Since the cylinders are in con-tact, how can they possibly rotate in the same direction? For

*As Turner (op. cit.) points out Maxwell employed two analogies. Onebridged a stationary field and a solid under stress. The other is between elec.tricity and fluid motion; "with its suggestion that Ampere's laws be modified tosatisfy the equation of continuity."

131

Model of an electromagnetic field used by Maxwell visualized "Molecularvortices" rotating in space. In this illustration the vortices are slendercylinders seen from the end. (Maxwell gave the cylinders a hexagonal crosssection to simplify the geometry of the model.) Between the vortices aresmall "idle wheels." If a row of the idle wheels is moved from A toward B,they cause the adjacent vortices to rotate in the opposite direction. (ScientificAmerican)

as everyone knows, a revolving wheel or cylinder causes itsneighbor to revolve in the opposite direction; thus one wouldexpect the rotation of the cylinders to alternate in directionfrom one to the next. Maxwell hit upon a pretty idea. He sup-posed the cylinders to be separated by rows of small spheres,like layers of ball bearings, which acted as gears (in Max-well's words, "idle wheels"). This arrangement, resembling adevice envisaged a century earlier by John Bernoulli, the

132

j-if 0 . 0. :I. '''..L. Vt..' I! P.If t H

younger, fulfilled the requirement. The spheres rotate in anopposite sense to that of each of the cylinders with which theyare in contact, and so the cylinders all rotate in the same direc-tion.

And now, as just reward for his ingenuity, Maxwell foundthat the spheres could be made to serve another, even morevaluable, purpose. Think of them as particles of electricity.Then by purely mechanical reasoning it can be shown thattheir motions in the machine of which they are a part serve toexplain many electrical phenomena.

Consider these examples. In an unchanging magnetic fieldthe cylinders all rotate at the same constant rate; thus theymaintain a constant magnetic field. The little rotating sphereskeep their position; there is no flow of particles, hence no elec-tric current, a result that tallies with the properties of a uni-form magnetic field. Now suppose a change in the magneticforce. This means a change in the velocity of rotation of thecylinders. As each cylinder is speeded up, it transmits thechange in velocity to its neighbors. But since a cylinder nowrotates at a slightly different speed from its neighbor, thespheres between them are torn from their positions by a kindof shearing action. In other words, they begin to move fromtheir centers of rotation, in addition to rotating. This motion oftranslation is an electric current; again, a result that tallieswith the properties of a changing magnetic field.

Observe now how the model begins to live a life of its own.It was designed, as J. J. Thomson has pointed out,* to exhibitFaraday's great discovery that magnetic changes produce elec-tric currents. It suggested to Maxwell the no less striking con-verse phenomenon that changes in electric force might producemagnetism.t Assume the spheres and cylinders are at rest. If' Sir J. .1. Thompson. "James Clerk Maxwell," in James Clerk Maxwell, ACommemoration Volume, op. cit.

t Ampi.re, of course, had already demonstrated that currents in wires producedaccompanying magnetic fields.

133

i

134

a force is applied to the spheres of electricity, they begin torotate, causing the cylinders of magnetism with which they arein contact to rotate in the opposite direction. The rotation of thecylinders indicates a magnetic force. Moreover, the modelholds up even as to details. Take a single illustration. Mag-netism acts at right angles to the direction of flow of current. Ifyou will examine the diagram of Maxwell's model, you willsee that the cylinders will rotate in the direction perpendicularto the motion of the spheres, thus bearing out the observationthat a magnetic force acts at right angles to the flow of a cur-rent.

"I do not bring it forward," Maxwell wrote of his system,"as a mode of connection existing in Nature. . . . It is, how-ever, a mode of connection which is mechanically conceivableand easily investigated, and it serves to bring out the actualmechanical connection between the known electromagneticphenomena.* Certain aspects of these "mechanical connec-tions" have already been discussed rotations, pressures,tensions which account for the reciprocal relations betweencurrents and magnetic forces.t The connections also serve toexplain the repulsion between two parallel wires carrying cur-rents in opposite directions, an effect produced by the centrifu-gal pressures of the revolving cylinders acting on the electricalparticles between them. The induction of currents is similarlyelucidated: this phenomenon, says Maxwell, is simply "part ofthe process of communicating the rotary velocity of the vor-tices [cylinders] from one part of the field to another." Inother words, whenever electricity (Maxwell's particles) "yieldsto an electromotive force," induced currents result. His dia-gram and the accompanying text make this beautifully clear.

Maxwell was not done with his model. It had helped in the

"On Physical Lines of Force," op. cit.

t The model explained, for example, why a current of electricity generated heat:for as the particles (or spheres) move from one cylinder to another, "theyexperience resistance, and generate irregular motions, which constitute heat."

interpretation of magnetism, the behavior of electric currents,the phenomenon of induction; it had yet to pass the supremetest: that is, to supply a mechanical explanation of the originof electromagnetic waves. To orient ourselves in this matter wemust examine briefly the question of condensers and insulators.

An electric condenser is a device for storing electricity. Inits simplest form it consists of two conducting plates separatedby an insulating material, or dielectric as it is called. Theplates can be charged, after which the charges attract eachother through the dielectric and are thus said to be "bound" inplace. Faraday in his experiments had come upon a curiousfact. He found that two condensers of the same size, fed by thesame electric source and with insulation of equal thickness,differed markedly in their capacity to take or to hold a chargeif the insulating material (dielectric) was different. This wasdifficult to understand if all dielectrics were equally imperme-able to an electric current. Moreover, if it were true, as Max-well already was beginning to suspect, that light itself is anelectrical phenomenon, how could light pass through certaindielectrics, empty space among them? With the help of hismodel, Maxwell advanced a bold hypothesis. Conductors, hesaid, pass a current when the electrical particles they containare acted upon by an electric force. Under such an impulsion,the little particles move more or less freely from cylinder tocylinder, and the current flows as long as the force persists.Not so in a dielectric. The particles are present but an easypassage from cylinder to cylinder is impossible. This fact maybe taken as the characteristic attribute of a dielectric, havingto do with its physical structure. Yet it was known that "local-ized electric phenomena do occur in dielectrics." Maxwell sug-gested that these phenomena also are currents but of aspecial kind. When an electric force acts on a dielectric, theparticles of electricity are "displaced," but not entirely tornloose; that is, they behave like a ship riding at anchor in astorm. The medium in which they are located, the magnetic

135

cylinders, is elastic; under p' ,sure the particles move, a lim-ited distance, until the force pushing them is balanced by thestresses due to the elastic reaction of the medium. Thus a stateof equilibrium is attained. But as soon as the impelling forceceases to act, the particles snap back to their original positions.The net effect of these mechanical actions is twofold. First, theinitial displacement of the electric particles constitutes a cur-rent that passes through the dielectric. A current of this type iscalled a displacement current to distinguish it from currentsthat flow through conductors and are therefore known as con-duction currents.* Wherever there is an electric force, saidMaxwell, there is displacement; wherever there is displacement, there is a current.

Second, whenever the pressure displacing the particles is re-leased, and they snap back, they overshoot and oscillate brieflyabout their fixed positions. The oscillation is transmittedthrough the magnetic medium (the insulator) as a wave. Thiswave is the return phase of the displacement current.t (Max-well suggested this disturbance on analogy to the displacementof an elastic solid under stress.)

Maxwell next arrived at an epoch-making conclusion. Thevelocity of the displacement wave, or current, depends on theelectrical properties of the medium in which it moves. More-over, this velocity, as he showed, was "within the limits ofexperimental error, the same as that of light." Hence, he in-

The contrast between displacement currents and currents through conductorswas vividly expressed by Henri Poincare. A displacement current, he said, isan elastic reaction like the compression of a spring: it can only be effected bypressure against resistance. Equilibrium is reached when resistance balancespressure. When the pressure is removed the spring regains its original form. Aconduction current, on the other hand, is like a viscous reaction such as is en.countered in moving a body immersed in water. It can be effected only by pres.sure; the resistance depends on velocity; the motion continues as long as themotive force acts, and equilibrium will never be established. "The body does notreturn to the starting point, and the energy expended in moving it cannot berestored, having been completely transformed into heat through the viscosity ofthe water." (Maxwell's Theory and Wireless Telegraphy, New York, 1904.)

t If the electric force applied to the insulator is varied continually, it will pro.duce a continually varying displacement wave: in other words, a continuingcurrent.

136

.4I 00..

O.:. .0

: .4°.

0: +

). IA.

I +,.. +

!

Electromagnetic wave as visualized by Maxwell is a moving disturbancewhich tends to separate positive (plus sign) and negative (dot) charges. Inthe drawing at the top, magnetic lines of force (arrows) lie at right anglesto the direction in which the disturbance is moving. The drawing at the bottorn depicts the two components of the electromagnetic wave. The electricalcomponent is shown in black, the magnetic component in color. (ScientificAmerican)

ferred, "the elasticity of the magnetic medium in air is thesame as that of the luminiferous medium, if these two coex-istent, coextensive and equally elastic media are not ratherone medium."

More must be said as to how Maxwell actually arrived atthis conclusion. In the 1850s the German physicists WilhelmWeber and Friedrich Kohlrausch had investigated au impor-tant relationship, namely, the ratios of electrostatic to electro-dynamic action. The electrostatic unit of charge was definedas the repulsion between two (like) unit charges at unit dis-tance apart. The electrodynamic unit was defined as the repul-sion between two definite lengths of wire carrying currents"which may be specified by the amount of charge which travelspast any point in unit time." In order to compare the repulsionbetween static charges with that between moving charges, afactor of proportionality must be introduced, since the unitsare different for static and dynamic phenomena. That is, onemust determine how many positive units of electricity flowingin one wire, and negative units flowing in the other, are re-quired to produce between the wires a repulsion quantitativelyequal to that between two static units. The factor turns out tobe a velocity; for since the length of the wires is fixed, and thenumber of units of electricity passing a given point in a giventime can be measured, what the investigator must consider isthe dimension length divided by time or velocity. Weber andKohlrausch had found that the velocity of propagation of anelectric disturbance along a perfectly conducting wire is closeto 3 x 10" centimeters per second. This was an astonishingcoincidence, for the figure was about the same as the velocityof light as it had been determined a few years earlier by theFrench physicist Hippolyte Fizeau.

Kirchhoff remarked the coincidence, but did not pursue it;Maxwell did. In 1860 he attacked the problem experimentally,using an ingenious torsion balance to compare the repulsionbetween two static charges and two wires carrying currents.The Weber-Kohlrausch results were roughly confirmed. Also,at about the same time (he said, in fact, that the pencil andpaper work was done before seeing Weber's memoir), he cal-culated the velocity of displacement currents in empty spaceor in any other dielectric. The resulting values tallied closely.

138

In other words, currents in a wire, displacement currents in adielectric, and light in empty space ( which of course is adielectric) all traveled with the same velocity. With this evi-dence at hand, which he communicated in a letter to Faradayin 1861, Maxwell did not hesitate to assert the identity of thetwo phenomena electrical disturbances and light. "We canscarcely avoid the inference," he said, "that light consists inthe transverse undulations of the same medium which is thecause of electric and magnetic phenomena."

"On Physical Lines of Force," despite its cogwheels andother gross mechanical adjuncts, may be regarded as the mostbrilliant of Maxwell's electrical papers. If it did not claim togive a picture of the true state of things, it was at least enor-mously enlightening as to how electricity and magnetism couldinteract in a purely mechanical relationship. Maxwell himselfsummarized the achievements of the theory as follows. It ex-plained magnetic forces as the effect of the centrifugal forceof the cylinders; induction as the effect of the forces called intoplay when -there is a change of angular velocity of the cylin-ders; electromotive force as an effect produced by stress on theconnecting mechanism; electric displacement as a result of theelastic yielding of the mechanism; electromagnetic waves asan accompaniment of displacement. The paper is one of therare examples of scientific literature in which one may glimpsethe play of imagination, the actual exercise of inductive power,the cultivation of nascent ideas.

None of the basic concepts unfolded in this memoir wasdiscarded in the more mathematical writings that followed.But Maxwell now had to outgrow his model. In "A DynamicalTheory of the Electromagnetic Field," published in 1864,*Maxwell, in Sir Edmund Whittaker's words, displayed thearchitecture of his system "stripped of the scaffolding by aidof which it had first been erected."t The particles and cylinders

Royal Society Transactions, vol. CLV.

t History of the Theories of /tether and Electricity: The Classical Theories,London. 1951.

139

are gone; in their place is the field "the space in the neigh-borhood of the electric or magnetic bodies" and the aether,a special kind of "matter in motion by which the observedelectromagnetic phenomena are produced." The matter com-posing the aether has marvelous properties. It is very fine andcapable of permeating bodies; it fills space, is elastic and isthe vehicle of "the undulations of light and heat." Yet for allits refinements and subtleties, the medium is no less a mechan-ical rig than the cylinders and spheres of its predecessor. Itcan move, transmit motions, undergo elastic deformations,store potential ( mechanical) energy and release it when thedeforming pressures are removed. Though susceptible to theaction of electric currents and magnets, it is nonetheless amechanism that, as Maxwell said, "must be subject to the gen-eral laws of Dynamics, and we ought to be able to work out allthe consequences of its motion, provided we know the form ofthe relation between the motions of the parts." In the precedingpaper Maxwell already had devised a set of equations thatdescribed the possible mechanical basis of electrical and mag-netic phenomena, and, in particular, how certain changes inelectric and magnetic forces could produce electrical waves.He now elaborated the hypothesis of displacement currents andobtained the expressions that are in substance the famous Max-wellian equations of the electromagnetic field.

In their most finished form the equations appear in theTreatise on Electricity and Magnetism (1873), the culmina-tion of Maxwell's researches, which he wrote at Glenlair in theyears following his resignation from King's College. Thiscelebrated work deals with every branch of electric and mag-netic science and presents the results of twenty years of thoughtand experiment. Maxwell remained faithful to Faraday, whosepoint of view is emphasized throughout the Treatise. Charac-terizing his own part as that of an "advocate," Maxwell makesno attempt to describe the hypotheses propounded by Weber,Gauss, Riemann, Carl and Franz Neumann, or Ludwig Lorenz,

140

1

the foremost cultivators of the theory of action at a distance.The task Maxwell set himself was, first, to formulate mathe-

matically electromagnetic phenomena as observed experi-mentally, and, second, to show that these mathematicalrelationships could be deduced from the fundamental scienceof dynamics; or to put it another way, that the laws of elec-tricity in motion could be derived from the laws applicable toany system of moving bodies. As always, Maxwell was verycautious in expressing himself about the nature of electricity.It behaves, he said, like an incompressible fluid; "whereverthere is electric force there is electric displacement." These,as J. J. Thomson observed, are the only definite statementsabout electricity to be found in the treatise, which led Hertzto say that Maxwell's theory is Maxwell's equations, andcaused Helmholtz to comment that "he would be puzzled toexplain what an electric charge was on Maxwell's theory be-yond being the recipient of a symbol."

What are the Maxwellian equations? I cannot hope to givean easy answer to this question, but at the cost of deliberateoversimplification I must try summarily to explain them, forthey are the heart of the theory.

Maxwell based the equations on four principles: (1) thatan electric force acting on a conductor produces a currentproportional to the force; (2) that an electric force acting ona dielectric produces displacement proportional to the force;(3) that a current produces a magnetic force (i.e., a movingelectric charge is surrounded by a magnetic field) at rightangles to the current's lines of flow and proportional to itsintensity; (4) that a changing magnetic force (or field) pro-duces a current proportional to the intensity of the force. Thethird and fourth principles exhibit a striking symmetry. Thethird is Faraday's law of electromagnetic induction, accordingto ithich "the rate of alteration in the number of lines of mag-netic induction passing through a circuit is equal to the workdone in taking unit electric charge round the circuit." Max-

141

',7

voitswe.--001.---00"Issms....1.0000000sesulansollainirr--

;001100111111ssausassorajuninisalallr4111111111:1761

-0110--011001111111111111111111111111111111111imi

111- II elnliteraiirimes-

0111011110"61111111111113rISIU0smilanlIsijaSHItielis

life -ion

0 g* II* s a es isuali 0 °A rI I

lirsliate tiniigni :I * :a re011 41:Ii ill OFviii

MalOttrassasliessarsomanaglawatristAltitleitrea$111.--01"--$010ewssors

essimmiii-liami--4-11-*-1114-1.

0.10111111401111111111311313::::1"Inallenal1:110:1404441Ot****A6*..11nossesssuml:111:11:1"111.4111411

061* NO

140 1131 1 I I .0.1 lo s a aa ii a II a um 1 is Arlig ifillnapliiii Sili

latOPIVIVISICAVIII114111113. *I 01:311: es* lin !!!!!!!! 111:11: !!! **4 : ;11!!!! Ter ....at 4,..

VeAVIA I I: : : $ :: : 1 if OA ios:1181" o: :*::::11;4

....1:stall:tell.......... LI

0.00.0$*--0 0 .

O.st

:. 4,* . 0, a0. 44 11114 : 4 "1404

A. NI: *IIA 471***

Willattiftrgetrige44reaittt4tstSz* it/ / I r4 A"WV*:

,.. .*iv. I,- 0 1, . .4 74 0 ..._.it_

+:: W. WA,.". ***** ..4.. 0 o. V .... V

...01:Aal ........6 # ******gi:

*****gm* _.* *

tih/./44 1ILL 1 t N

lf,711iiI i ti sti$10 iitix.p.

w*.441II 14111L'ite 1like4.:°:** 44. A 4

k nv 4

*ill ..4 j4

.t.,r.,..A.L._.,stimai arirlittj:

142

Lines of force appear in Electricity and Magnetism. LEFT: "Uniform magneticfield disturbed by an electric current in a straight conductor." ABOVE: "Twocircular currents." (Scientific American)

well's complementary law, the fourth principle, is that "therate of alteration in the number of lines of electric force pass-ing through a circuit is equal to the work done in taking a unitmagnetic pole round it."

On this foundation two sets of symmetrical equations can beerected. One set expresses the continuous nature of electricand magnetic fields; the second set tells how changes in onefield produce changes in the other. In these formulations themechanical aspects of the theory are retained, perfect conti-nuity is preserved by treating electricity as if it were an in-compressible fluid, and wave phenomena are deduced as theconsequences of displacement in a dielectric.

How does the concept of the field enter the theory? We have

143

followed Maxwell as he stripped his model of its particles andcylinders and reduced it to an aetherial medium. In theTreatise, while not abandoning the medium altogether, he robsit of almost all its attributes other than form. The matter of themedium, as Poincare says, is left only with purely geometricproperties, the atoms dwindle to mathematical points, subjectto the laws of dynamics alone. The grin is left but the cat isgone. It is a perfect example of mathematical abstraction.*

The aether is a thing that wiggles when it is prodded,but does nothing on its own. An electromagnetic field con-sists of two kinds of energy, electrostatic or potential en-ergy, and electrodynamic or kinetic energy. The aether, like

Einstein made an interesting comment about Maxwell's equatior. and Ins useof the concept of the field. "He showed that the whole of what was then knownabout light and electromagnetic phenomena was expressed in his well-knowndouble system of differential equations, in which the electric and the magneticfields appear as the dependent variables. Maxwell did, indeed, try to explain,or justify, these equations by intellectual constructions. But he made use ofseveral such constructions at the same time and took none of them really seri-ously, so that the equations alone appe---41 as the essential thing and thestrength of the fields as the ultimate e..tit, 5, not to be reduced to anythingelse. By the turn of the century the concep.ion of tit, electromagnetic field asan ultimate entity had been generally accepted and serious thinkers had aban-doned the belief in the justification, or the possibility, of a mechanical explana-tion of Clerk Maxwell's equations. Before long they were, on the contrary,actually trying to explain material points and their inertia on field theory lineswith the help of Maxwell's theory, an attempt which did not, however, meetwith complete success. Neglecting the important individual results which ClerkMaxwell's life work produced in important departments of physics, and con-centrating on the changes wrought by him in our conception of the nature ofphysical reality, we may say this: before Clerk Maxwell people conceived ofphysical realityinsofar as it is supposed to represent events in nature asmaterial points, whose changes consist exclusively of motions, which are sub-ject to partial differential equations. After Maxwell they conceived physicalreality as r" resented by continuous fields, not mechanically explicable, whichare subject .o partial differential equations. This change in the conception ofreality is the most profound and fruitful one that has come to physics sinceNewton; but it has. at the same time to be admitted that the program has byno means been completely carried out yet."

1 am puzzled as to what Einstein meant in saying that Maxwell's equationeliminated the notion of mechanism in explaining electromagnetic phenomena.Similar views have been expressed by many other physicists and philo.ophers.Maxwell himself would not have agreed with this position. His writings refuteit. The inference was drawn by his successors. But there is a more important

144

a universal condenser, may be conceived as storing energyin which case, being elastic, it is deformed. Since the aetherfills all space and therefore penetrates conductors as well asdielectrics, it no longer makes any difference whether we dealwith a conduction current or a displacement current in eithercase the aether is set in motion. This motion is communicatedmechanically from one part of the medium to the next and isapprehended by us as heat, or light, or mechanical force (asin the repulsion between wires) or other phenomena of mag-netism and electricity. The ruling principle of all such phe-nomena, it should be observed, is that of least action. This isthe grand overriding law of the parsimony of nature: every

point that requires clarification; namely. do the equations justify the inference?It is true that a field is not the same as a material particle. and that the motionof a particle is not the same as a change in a field. It is true also that the concept "material particle" was long held to he intuitively dear, while the concept"field" has never been so regarded. This makes it easier to say mysteriousthings about fields,.which no one would dream of saying about particles. But amore careful definition of these concepts, as physicists actually use them. raisesserious questi as to whether a field is any less suited to a "mechanistic" explanation than a system of material particles; indeed. whether a mechanisticexplanation fits either or neither case. In modern physics material particles arenot shat they once were. They are pale abstractions, quite incapable of anything so robust as a collision. But then what is a collision? One thinks of bibHard balls knocking together. as a pristine example. This, however, is a plainman's way of thinking. The modern physicist has rid his mind of such seductiveimages. (As far back as the eighteenth century, the Italian physicist Boscovichproposed the idea that the heart of an atom is not solid substance but a merecenter of immaterial force. As particles fade, the field becomes more substanfiat. Properties are now ascribed to it that make it seem more real and morepotent than a billiard hall or a boulder. Of course the field is hard to describein homely terms. Yet it is quite capable, as physicists tell us, of doing homelythings. It produces and undergoes changes now as if it were a cloud, now anengine, now an ocean. In short it has mechanical effects. By this I mean effectsof a kind produced by what used to be called material particles. Moreover, ithas mechanical properties. By this I mean properties of a kind produced bywhat we call a inaci,me. The field can do things no system of particles ormachine yet conceived can do. Since it can also do all they can do, it is a super.machine, Is there any point in saving the name? I think there is. to keep ourthinking straight. We ought to keep it to describe both fields and particles orwe ought to discard it entirely. If the word "mechanism" has any meaning in theuniverse of refined observation, it has as much meaning in relation to fields asto particles. At the same time I am quite prepared to believe that it has as littlemeaning in one case as the other; for that matter, no meaning in either.

145

action within a system is executed with the least possible ex-penditure of energy. It was of the first importance to Maxwellthat electrical phenomena should satisfy the principle, forotherwise his mechanical explanation of the phenomena wouldnot have been possible.

With these points in mind, we may examine a set of Max-well's equations in a form that describes the behavior of anelectromagnetic field under the most general conditions, i.e.. afield moving in empty space. No conductors are present, nofree charges, and the medium is a vacuum. The equations thenread

1) divE=02) div II = 0

I all3) curl E =c at

14) curl II = c aEat

The meaning of the symbols is as follows: E and 1/ representelectric and magnetic field strength; since they vary in time,and from place to place, they are functions of the space co-ordinates x, y, z (not shown) and of the time coordinate, t.C is the velocity of light and enters the equations as the rateof propagation; div (an abbreviation for divergence) and curl(an abbreviation for rotation) represent mathematical opera-tions whose physical meaning is explained below.

Divergence is essentially a measure of rate of change. Inwords, then, equation I

div E = 0

says that in a moving field the electric intensity is the same atevery point, i.e., the rate of change is zero at every point. Moreloosely, this equation extends to the field the classical principlethat electric lines of force can be neither created nor de-stroyed. Thus the equation says that the number of electriclines of force, representing the field strength, that enter any

146

, .,. f'/'

tiny volume of space must equal the number leaving it. Mak-ing use of still another analogy, if one conceives of electricityin Maxwell's idiom, as an incompressible fluid, equation 1states that as much fluid flows out of a tiny volume of space ina given time as flows in.*

For the reader interested in a little more detail, the following explanationmay be helpful. Equation 1 states that the divergence of the electric field intensity is zero at any point in space and at any instant of time. The meaning of theequation may he visualized as follows. It is customary to represent E at a giveninstant of time by a series of lines whose relative density in space is proportional to E. These lines have direction because E is a vector. Consider a pointP and a sphere surrounding 1'. Let us suppose that the intensity of the electricfield on the left hemispherical surfacie of the sphere is uniform over the surfaceand is directed at each point perpendicular to the surface.

Suppose further that some change takes place in the electric field Intensity Ein the region occupied by the sphere but such that on the right hemisphericalsurface the field E is again uniform and perpendicular to the surface but strong.er than on the left portion. We would indicate this increase in the intensity ofE by having more lines lease the sphere on the right than enter on the left.Using the number of lines as a measure of E. we count the lines entering thespherical surface and multiply this number by the area of the hemisphere. andregard this product as negative. Let us next form the analogous product of thearea and the number of lines leaving the surface, and regard this product aspositive. The algebraic sum of these two products, that is. the positive plus thenegative. is called the net electric flux through the spherical surface. This netflux is the divergence of E over the volume of the sphere. In our illustrationthe net flux of E has increased as E passes through the sphere. Hence weshould say in this case that the divergence of E through the sphere is positive.If we now divide this net flux through the sphere by the volume of the sphere.we obtain the next net flux per unit volume. We now imagine that the spherebecomes smaller and smaller and contracts to the point P. Of course the netflux per unit volume changes and approaches some limiting value. This limitingvalue, which is a mathematical abstraction, is dip E at the point P. Thus dim Eis essentially a measure of the spatial rate of change of E at the point P. Sinceequation 1 says that for electric fields div E =0 at each point /'. we may saythat the net spatial rate of change of E is zero in empty space. More looselystated, this equation says that electric field lines are neither created nor de.strayed at the point P. It is to be noted that the phrase "spatial rate of change"is intended to emphasize that the divergence is concerned with the way inwhich E changes from point to point in space at the same instant of time. Thisspatial rate must be distinguished from the rate at which some quantity. forexample, E itself in equation 4. may change during some interval of time.

147

Equation 2 .

div H = 0

makes the same assertion for magnetic lines as equation 1makes for electric lines.

Equation 3

1 DHcurl E = ,,

c at

is Maxwell's way of stating Faraday's law of induction. Theequation describes what happens in a changing magnetic field.

The right side expresses rate of change,-1H , multiplied by a?It

/very small factor, -- (the negative sign before the fraction

cis purely a matter of algebraic convenience) ; the left side ex-presses the fact that an electric field is created by a changingmagnetic field. But the equation is more than analytic; thanksto the sign curl, it actually gives a picture of the event. A simplediagram may help make this clear. Suppose the existence of amagnetic field uniform over a region of space. We draw acircle

H

surrounding a bundle of parallel lines, which represent theintensity and direction of the magnetic field. The circle lies ina plane perpendicular to the lines. If the field is changed (by

148

motion or by increase or reduction of strength), it producesan electric field that acts in a circle around the lines of mag-netic force (though it may also act in other directions). Bysumming the work done in moving unit electric charge aroundthe circle, we obtain what is called the net electromotive forcearound the circle.* If the circle were made of wire, the chang-ing magnetic lines would of course induce the flow of a cur-rent; but even without a wire and therefore no current aforce would be induced. Dividing this force by the area en-closed by the circle gives the net electromotive force (per unitarea) which "curls" around the circle. Now imagine the circlegrowing smaller and smaller and shrinking finally to the pointP. By this limiting process we obtain a limiting value of thenet electromotive force per unit area: this is curl E at P. Thusequation 3 says that the limiting value of electromotive forceper unit area equals the rate of change of H at the point P,

1multiplied by the tiny negative fraction, c .f Or, again,

more loosely stated, a changing magnetic field creates anelectric field whose electromotive force per unit area at anygiven point and instant of time equals the time rate of changeof the magnetic field at that point and instant.

Equation 4

1 DEcurl H = cDt

says that, except for the change in algebraic sign (which hasto do with the directions of the fields), the roles of E and II in

In physical terms, we obtain the net capacity of the electric field to movecurrent along the circle.

t The symbol c, which here stands for the ratio of the electrostatic to the elec.tromagnetic units of electricity, is required to translate E Ian electrostatic phenomcnon) and H (an electromagnetic phenomenon) into the same system ofunits. The equation explains how Maxwell was able to connect electrical andmagnetic phenomena with the velocity of light, for c is in fact that velocity.

149

equation 3 may be reversed. At any given point and instant themagnetomotive force (the analogue for magnetic fields ofelectromotive force) per unit of area created by a changingelectric field is equal to the time rate of change of the electric

/field multiplied by the tiny positive fraction . Now, thec

reader who has followed this discussion will perceive that theDE

' itime rate of change of E, is none other than Maxwell'sDtdisplacement current. For since the changes are taking placein the dielectric known as empty space, the only currents thatcan flow are displacement currents.* Prior to Maxwell, it wasthought that the magnetic field H could be produced only bycurrents that flowed in wires passing through the circle. If nowires were present, the law thought to be applicable wascurl H=0. It was Maxwell's great discovery, deduced me-chanically from his model and expressed mathematically inthis equation, that a time-varying electric field produces (ormust be accompanied by) a net "curled" magnetic force evenin an insulator or empty space.t

According to Maxwell's theory, the introduction of a time-varying electric force in a dielectric produces displacementwaves with the velocity of light. To put it another way, it is thesurge and ebbing of the force that produces the periodic dis-placement waves; a static charge would merely create an in-stantaneous displacement, which would be fixed, but not a

Equation 4 assumes the existence of this current and relates it quantitativelyto the magnetontotive force generated by the existent magnetic field. Physicallywe may regard the magnetic field as creating the displacement current or, conversely, regard the displacement current as creating the accompanying magneticfield and magnetomotive force.

t Maxwell called aEthe displacement current, the term "displacement" meanal

ing that the electric field intensity E was being altered or displaced as timeaEvaries, and the term "current" suggesting thatat had the properties of a cur.

rent flowing in a wire even though DEexisted in empty space.at

150

wave. Now, an electric current, as we have seen, whether in adielectric or in a conductor, is accompanied by a magneticforce; and similarly a periodic wave of electric displacementis accompanied by a periodic magnetic force. The wave frontitself, as Maxwell showed, comprises electric vibrations atright angles to the direction of propagation and a magneticforce at right angles to the electric displacement. The com-pound disturbance is therefore called an electromagnetic wave.A light wave (which is a displacement wave) is, as HenriPoincare later elaborated, "a series of alternating currents,flowing in a dielectric, in the air, or in interplanetary space,changing their direction 1,000,000,000,000,000 times a sec-ond. The enormous inductive effect of these rapid alternationsproduces other currents in the neighboring portions of thedielectric, and thus the light waves are propagated from placeto place."

The electromagnetic theory of light was testable experi-mentally, and indeed stood up remarkably well in laboratorytrials. But this was only a limited confirmation of Maxwell'ssystem, for if his reasoning was correct, there must be otherelectrical waves produced by initial disturbances of differingintensity. These waves would differ from light in wave lengthand would therefore not be visible, yet it should be possible todetect them with appropriate instruments. How to find them,not to say generate them, was now the crucial problem. Max-well did not live to see it solved. Not until ten years after hisdeath were his prophecies fulfilled and the skepticism of hismost distinguished contemporaries refuted. As late as 1888Lord Kelvin referred to Maxwell's waves as a "curious andingenious, but not wholly tenable hypothesis"; but a year laterHelmholtz's greatest pupil, Heinrich Hertz, nosed out OliverLodge in the race to demonstrate their existence. In a series ofbrilliant experiments he showed how electric waves could be"excited" ( i.e., generated) by oscillation and detected bya circular conductor provided with a small gap; and how theycould be polarized', reflected, refracted, made to form shadows

and to interfere with each other. The connection, he said, "be-tween light and electricity . . . of which there were hints andsuspicions and even predictions in the theory, is now estab-lished. . . . Optics is no longer restricted to minute aetherwaves, a small fraction of a millimetre in length; its domain isextended to waves that are measured in decimetres, metres andkilometres. And in spite of this extension, it appears merely. . . as a small appendage of the great domain of electricity.We see that this lattt: ha, become a mighty kingdom."

The Treatise, written while Maxwell was "in retirement" atGlenlair, drew only part of his energy. As a "by-work" duringthe same period he wrote a textbook on heat, which appearedin 1870, and a number of papers of considerable importanceon mathematics, color vision and topics of physics. He main-tained a heavy scientific and social correspondence, enlargedhis house, studied theology, composed stanzas of execrableverse, rode his horse, went on long walks with his dogs, visitedhis neighbors and played with their children, and made fre-quent trips to Cambridge to serve as moderator and examinerin the mathematical tripos.

In 1871 a chair in experimental physics was founded atCambridge. It is hard to realize that at the time no courses inheat, electricity and magnetism were being taught there, andno laboratory was available for the pursuit of these arcanematters. The University, as a contemporary scholar delicatelyobserved, "had lost touch with the great scientific movementsgoing on outside her walls." A committee of the faculty beganto bestir itself, a report was issued, and the lamentable factsfell under the gaze of the Duke of Devonshire, Chancellor ofthe University. He offered the money for the building andfurnishing of the famous Cavendish Laboratory. Thomson, itwas known, would not leave his post at Glasgow to take thenew chair, and Maxwell, though at first reluctant to leaveGlenlair, yielded to the urging of his friends to offer himselfas a candidate. He was promptly elected.

152

He now devoted himself to the task of designing and super-intending the erection of the laboratory. His aim was to makeit the best institution of its kind, with the latest apparatus aimthe most effective arrangements for research. He inspectedThomson's laboratory at Glasgow and Clifton's at Oxford tolearn the desirable features of both and embody them in theCavendish. He presented to the laboratory all the apparatus inhis own possession and supplemented the Duke's gift by gen-erous money contributions. With so many details to be takencare of, the structure and its appointments were not completeduntil 1874. The delay, while inevitable, was inconvenient. "Ihave no place," wrote Maxwell, "to erect my chair, but moveabout like the cuckoo, depositing my notions in the ChemicalLecture Room in the first term, in the Botannical in Lent andin the Comparative Anatomy in Easter." His "notions" werethe courses he gave, beginning in 1871, on heat, electricity andelectromagnetism, a schedule maintained throughout the ten-ure of his chair. And though the audiences were often small,some of the best students were soon attracted to his lectures,which contained much important original work. The renais-sance that followed in physical science at Cambridge was thedirect result of his influence.

Maxwell's classic Matter and Motion, "a small book on agreat subject," was published in 1876. About this time hecontributed articles on various subjects "Atom," "Aether,""Attraction," "Faraday," among others to the famous ninthedition of the Encyclopaedia Britannica. His public lecturesinclude a charming discourse "On the Telephone," which,though delivered when he was already very ill, is not only asclear as his best expositions but filled with gay, amusingasides. Speaking, for example, of "Professor Bell's inven-tion," he comments on "the perfect symmetry of the wholeapparatus the wire in the middle, the two telephones at theends of the wire, and the two gossips at the ends of the tele-phones...." A task that occupied him for five years, almost tothe very end of his life, was editing twenty packets of unpub-

153

lished scientific papers of Henry Cavendish, who was great-uncle to the Duke of Devonshire. This splendid two-volumework, published in 1879, did much to fix the reputation of animmensely gifted investigator, whose important work on elec-tricity was unknown to his contemporaries because the resultswere confided only to his manuscripts. Maxwell repeatedCavendish's experiments and showed that he had anticipatedmajor discoveries in electricity, including electrostatic capac-ity, specific inductive capacity and Ohm's law.

As Maxwell grew older, friends remarked on his "ever-in-creasing soberness" of spirit. This must not be taken to mean hewas invariably melancholy or withdrawn or that his nice senseof fun about himself no less than about others had van-ished. He continued to see his many friends, to write lightverse and parodies, to promenade with his dog Toby, who wasat Maxwell's side even in the laboratory, to play small prac-tical, but never mean, jokes, to engage in what was called"humorous mystification" by advancing preposterous scien-tific ideas in conversation while keeping a straight face. Allthings, he once remarked, are "full of jokes," though they arealso "quite full of solemn matters," and he was as likely tostress their light as their grave aspect.

But it is true he became somewhat more reticent with thepassing years, and more and more concealed his feelings andreflections beneath ar, ironical shell. The tough, rational,Scotch common-sense cord of his nature had always been inter-twined with threads of mysticism. Often plain, even blunt, inhis address, he also had an allusive way of speaking andshowed a fondness for parables. He had faith in science, yethe was at bottom skeptical as to how much could be learnedfrom science alone about nature and meaning. It was all verywell, he felt, to have "ideal aspirations"; on the other hand,"It's no use thinking of the chap ye might have been." Hiscontemporaries remember him as both modest and intellectu-ally scornful, tentative in his scientific opinions and dogmaticwhen others seemed to him tb be immoderately self-assured.

154

,

"No one knows what is meant by" so-and-so was his way ofanswering a cocksure formulation of a scientific "truth."

The most striking of Maxwell's traits was his gentleness."His tenderness for all living things was deep and instinctive;from earliest childhood he could not hurt a fly." An extraordi-nary selflessness characterized his relationship to those closeto him. When his brother-in-law came to London to undergoan operation, Maxwell gave up the ground floor of his houseto patient and nurse and left himself with a room so small thathe frequently breakfasted on his knees because there was noroom for a chair at the table. Mrs. Maxwell had a serious andprolonged illness in the last years of Maxwell's life, and heinsisted on nursing her. On one occasion it is reported that hedid not sleep in a bed for three weeks. But his work went onas usual, and he was as cheerful as if he enjoyed the ordealwhich may indeed have been the case. Nor did he give theslightest sign of being downcast or show self-pity when his ownfatal illness seized him.

In the spring of 1877 he began to be troubled with pain anda choking sensation on swallowing. For some strange reason heconsulted no one about his symptoms for almost two years,though his condition grew steadily worse. His friends at Cam-bridge observed that he was failing, that the spring had goneout of his step. When he went home to Glenlair for the sum-mer of 1879, he was so obviously weakening that he called formedical help. He was in terrible pain, "hardly able to lie stillfor a minute together, sleepless, and with no appetite for thefood which he so required." He understood thoroughly that hiscase was hopeless, yet his main concern seemed to be about thehealth of his wife. In October he was told he had only a monthto live. On November 5 he died. "No man," wrote his physi-cian, Dr. Paget, "ever met death more consciously or morecalmly." When Maxwell was buried in Parton Churchyard atGlenlair, the world had not yet caught up with his ideas. Eventoday it has not fully explored the kingdom created by hisimagination.

155

.1(

/.44rat:

1, .4

.4' 114 t ekt

?toriPfck,0

7 ,#410

James Clerk Maxwell, 1831-1879.

156

iMaxwell's Letters: A Collection

TRIN. COLL., Feb. 20, 1854.

DEAR THOMSON

Now that I hove entered the unholy estate of bachelorhood I hove begun to think of rend-ing. This is very pleasant for some time omong books of acknowledged merit wh one has not reodbut ought to. But we hove o strong tendency to return to Physicol Subjects ond severol of us herewish to ottock Electricity.

Suppose o mon to hove o populor knowledge of electricol show experiments ond o little on-tipothy to Murphy's Electricity, how ought he to proceed in reading& working so os to get a littleinsight into the subject wh may be of use in further reading?

If he wished to reod Ampere Faraday &c how should they be orronged, ond ot whot stoge &in whot order might he reod your articles in the Cambridge Journol?

If you hove in your mind ony answer to the obove questions, three of us here would be con-tent to look upon on embodiment of it in writing os advice.

I hove onother question from myself. At Ardmillon while bathing on the rocks you men-tioned that Gauss (?) hod been investigating the bending of surfaces and hod found in porticulorthat the product of the principal radii of curvature ot ony point is unchanged by bending.

I hove no meons here of finding the poper from wh you quoted so that I wd be obliged to youif you could give me some reference to it or even tell me whether he hod considered the condi-tions of bending of o finite portion of o surfoce in generol.

I have been working for some time ot the more generol problem & hove completed the theoryfor surfoces of revolution ond got severol results i- the cose of other surfoces by the considerotionof two systems of lines on the surfoce wh moy be coiled lines of bending.

These being given the effect of bending the surface is reduced to the consideration of one

indept vorble only.

These lines themselves however ore subject to certain conditions that the surfoce moy bebent ot oil, ond to odditionol conditions that these lines moy continue "lines of bending".

When the lines of bending are the lines of principal curvature the conditions are very sim-ple & are fulfilled for all surfaces of revoln.

But some of the operations are long so I am going over them by a process quite different fromthe first.

By finding what Gauss' results are I may be spared much trouble in pruning my calculations.

Finally I have heard nothing of C. J. Taylor since he went to Glasgow, and wd gladly re-ceive information about him, Ramsay & the professorship. Are there any good classical placesabout Glasgow or elsewhere, "where a man might enjoy a comfortable house".

This is a letter of questions so I go on in the same spirit to the end by enquiring after theprosperity of the College especially Nos 2 & XIII i.e. commend me to the Blackburns & Mrs.Thomson.

Yrs truly

J. C. Maxwell

158

-7

8 Paloce Garden Terrace, Kensington, W.1861 Dec. 10

DEAR THOMSON

I have not heard of you for same time except through Balfour Stewart who told me he hadseen you lotely. I hope you are now well as you are at work.,

I was not farther north than Galloway last summer and we spent all our three months vaca-tion there. Since I saw you I have been trying to develope the dynamical theory of magnetism asan affection of the whole magnetic field according to the views stated by you in the Royal Soci-ety's proceedings 1856 or Phil. Mag. 1857 vol. 1 p. 199 and elsewhere.

I suppose that the "magnetic medium" is divided into small portions or cells, the divisionsor cell-walls being composed of a single stratum of spherical particles these particles being"electricity". The substance of the cells I suppose to be highly elastic both with respec. c, com-pression and distortion and I suppose the connexion between the cells and the particles in the cellwalls to be such that there is perfect rolling without slipping between them and that they act oneach other tangentially.

I then find that if the cells are set in rotation, the medium exerts a stress equivalent to ahydrostatic pressure combined with a longitudinol tension along the lines of axes of rotation.

If there be two similor systems, the first a system of magnets, electric currents and bodiescopable of magnetic induction, and the second composed of cells and cell walls, the density ofthe cells everywhere proportionol to the capacity for mognetic induction of the correspondingpoint of the other, and the magnitude and direction of rototion of the cells proportionol to themagnetic force, then-

1. All the mechonical magnetic forces in the one system will be proportionol to forces intne other arising fro" centrifnaal farce.

2. All the electric currents in the one system will be proportional to currents of the por-ticles forming the cell walls in the other.

3. A!! the electromotive forces in the one system, whether arising from chonges of positiont..4 magnet; or currents or from motions of conductors or from chonges of ir tensity z.f mognets orcurrents will be proportional to forces urging the particles of the cell walls arising from the tan-gential action of the rotating cells when their velocity is increosing or diminishing.

4. If in a non conducting body the mutual pressure of the particles of the cell walls (whichcc,ercspands to electric tension) diminishes in given direction, the porticles will be urged in thatdirection by their mutual pressure but will be restrained by their connexion with the substance ofof the cells. They will therefore produce strain in the cells till the elasticity called forth bol-ances the tendency of the portides to move.

159

J. C. Maxwell to C. Hockin,

Glen lair, Dalbeattie, September 7th 1864

. . . I have been doing several electrical problems. I have got a theory of "electric ab-sorption," i.e. residual charge, etc., and I very much want determinations of the specific induc-tion, electric resistance, and absorption of gooddielectrics, such as glass, shell-lac, gutta-percha,ebonite, sulphur, etc.

I have also cleared the electromagnetic theory of light, from all unwarrantable assumption,so that we may safely determine the velocity of light by measuring the attraction between bodieskept at a given difference of potential, the value of which is known in electromagnetic measure.

I hope there will be resistance coils at the British Association.

164

boundary of a solid, a line as the edge of a surface, and a pointas the extremity of a line.

In like manner we may conceive the potential of a materialsystem as a function found by a certain process of integration withrespect to the masses of the bodies in the field, or we may supposethese masses themselves to have no other mathematical meaning

than the volume-integrals of irV2tY, where tii is the potential.

In electrical investigations we may use formulae in which thequantities involved are the distances of certain bodies, and theelectrifications or currents in these bodies, or we may use formulaewhich involve other quantities, each of which is continuous throughall space.

The mathematical process employed in the first method is in-tegration along lines, over surfaces, and throughout finite spaces,those employed in the second method are partial differential equa-tions and integrations throughout all space.

The method of Faraday seems to be intimately related to thesecond of these modes of treatment. He never considers bodiesas existing with nothing between them but their distance, andacting on one another according to some function of that distance.He conceives all space as a field of force, the lines of force beingin general curved, and those due to any body extending from it onall sides, their directions being modified by the presence of otherbodies. He even speaks* of the lines of force belonging to a bodyas in some sense part of itself, so that in its action on distantbodies it cannot be said to act where it is not. This, however,is not a dominant idea with Faraday. I think he would ratherhave said that the field of space is full of lines of force, whosearrangement depends on that of the bodies in the field, and thatthe mechanical and electrical action on each body is determined bythe lines which abut on it.

167h6,1?

the pr-_,perties Cf cettain antt rc,e.el2ctric Friot ; r_W1 t,: ,;irs t1C dor,. Oetstr,a's experiment with electiicc C`rit Jr.d a compOLS Si10,:(;ci that electricity

tis-n t: rel . I fc:inc. c -ne pH,..r,o,et,t in r

14 The Relationship of Electricity and Magnetism

D. K. C. MacDonald

Excerpt from his book, Faraday, Maxwell, and Kelvin, published in 1964.

We know that an electric current can produce forceson a magnet in its vicinity, or, in other words, an elec-tric current produces a magnetic "field." Faraday hadshown, moreover, that a changing magnetic field (pro-duced either by moving a magnet or by varying an elec-tric current in a coil) could induce an electric currentin a neighboring, but separate, coil of wire. Thus,through these fundamental experiments of Oersted,Ampere, and particularly Faraday, various vital factshad been discovered about how electric currents andmagnets could interact with one another and, as wehave said earlier, these discoveries were already lead-ing to exciting practical developments such as the elec-tric telegraph and the submarine cables. But, in broadterms, what James Clerk Maxwell tried to do was tobuild up a more general. picture of these interactionsbetw,:en electric and magnetic effects (or "fields")

169

without worrying so much about actual coils of wirewith electric currents in them, or about how in practiceone actually produced the magnetic fields. FollowingFaraday's general lead in concentrating on the "linesof force" or the "fields," Maxwell tried to work outdirectly and quantitatively the interaction in space ofthe electric fit..,1 on the magnetic field, and vice versa,wherever they minht -xis:. In his mind Maxwell in-vented, or designed, various semi-mechanical modelsto build up his theory, but in the end he could discardthis mental scaffo:ding and give a complete mathemati-cal description of electromagnetic behavior which holdsgood to this day.

Consider the production of a magnetic field by a cur-rent of electricity in a coil. We know that such a cur-rent always involves a movement of electric charge,so from the electrical point of view we may say thatsomething is changing all the time. One of the thingsMaxwell did was to generalize this discovery boldly,saying in essence: [I] "A Changing Electric Field WillAlways Produce a Magnetic Field."

But, on the other hand, Faraday had shown thatthe movement of a magnet could produce an electriccurrent, as we have already seen; so on the same linesthis can be generalized to say: [II] "A ChangingMagnetic Field Can Produce an Electric Field."

The ultimate result of James Clerk Maxwell's workwas, in effect, that he expressed these two basic ideasin precise, quantitative terms, and he came out finallywith what are now known as Maxwell's Equations,which, as I already have said, remain today the stand-ard method of predicting how electricity and magnetismwill behave under any given conditions. The acme ofMaxwell's work, however, was his discovery that whenapplied in free, empty space his equations took on aform which is equally descriptive of any undamped

170

,.

The Relationship of Del:tiled,/ and Magnetism

wave motion propagating itself freely from place toplace. Thus, if you drop a stone into a large pond ofwater a ripple or wave will proceed out from thatplace, and some of the energy from the falling stonewill radiate outward in the wave from the splash. Ifyou shout to somebody else some distance away, then itis a vibration or wave in the air around you which car-ries the sound to the distant person; or if you r 'gyp along, tight rope or string between two points, and then"twang" the rope, you can see a wave rt:nning alongthe rope, and this wave carries some of the energy thatyou put in the "twang." Again, if there is a violentstorm at sea, the energy from this storm gets carriedover long distances by waves in the ocean; the waveswhich smash on the rocks of Newfoundland may wellbe getting their energy from a storm a thousand milesor more out in the Atlantic Ocean. In each of these lat-ter examples the waves will be damped to some degreeor other. For example, waves traveling on the surfaceof the sea lose some energy by dragging deeper layersof water, by the very fact that water is not entirely freeto move by itself, but has a viscosity or "stickiness,"which means that the waves ultimately suffer losses byfriction.

The particularly remarkable, and unique, feature ofelectromagnetic waves is the fact that they can propa-gate themselves quite freely without damping throughempty space where no matter whatsoever is present,but it is not difficult to see from the two italicized state-ments above that a self-propelled wave motion of theelectromagnetic field might be possible.

Imagine that we have electric and magnetic fieldspresent in a small region of space, and that the fieldsare changing suitably with time. As the electric fieldchanges at some point in space it will produce a mag-netic field in the neighborhood, and if things are right

this magnetic field will then reinforce the magnetic

field in some regions, and in turn the over-all changing

magnetic field will produce again a fresh electric field

in its neighborhood. What Maxwell's equations showed

was that this process, perhaps somewhat reminiscent of

an endless game of leapfrog, could indeed be self-

maintained, with the energy constantly radiating out-

ward from where the waves started.But this was not all. Maxwell was able to predict

from this theory, moreover, the speed with which such

an electromagnetic wave should travel in space. Thisspeed was simply determited by the ratio of two meas-

urements which could be made on electric and mag-netic quantities in the laboratory, and it turned outthat the speed predicted in this way was very close to

the already known speed of light (about 300,000km /sec ::.--.- 186,000 miles/sec). Furthermore, it is also

a well-known characteristic of light that it too canpropagate through empty space, as witness the light of

day which reaches us unfailingly from the sun acrossabout a hundred million miles of empty space. So Max-

well could finally say with confidence that, physically

speaking, light must be a form of electromagnetic

radiation."Some years after Maxwell's death, Heinrich Hertz

(1857-94) was able to show experimentally, using

electrical apparatus, the direct generation and detection

of the electromagnetic waves predicted by Maxwell.These "Hertzian waves" are the great-grandfather of

the waves which carry all our radio and televisionbroadcasts today, and in fact radio waves, television

waves, light waves, X-rays, and gamma rays, are all

members of one and the same familyelectromagneticwaves. In free space they all travel with identically diesame speed, which for convenience we always refer to

as "the velocity of light." What distinguishes one type

The Relationship of Electricity and Magnetism

of wave from another is simply its rate of vibration, orthe corresponding wave length (i.e., the distance be-tween two successive "crests" or "troughs" of a wave).A typical radio wave vibrates at, or has a frequency(f) of, about a million times a second (f = 10" cy-cles/sec = 1 M c/s), and has a wave length (X) ofabout 300 meters. For those who do not mind an equa-tion, the relationship is very simple, namely fX = c,where c denotes, as always in physical science, thevelocity of light. At the other end of the scale, a gammaray might have a wave length of only about one ten-billionth part of a centimeter (X = 10-10 cm), and acorresponding frequency of vibration of about threehundred billion billion cycles/sec () = 3 X 10" c/s).

ELECTROMAGNETIC WAVES

Maxwell's electromagnetic theory also led to intensediscussion later about the fundamental nature of theelectromagnetir, waves involved. Many physicists feltthat in orde to have a wave at all there had to be"somethine to do the waving or vibrating, and they in-vented a sort of at-pervading, universal, thin soup orconsommé which they called the "aether." But whetherit is more reasonable to talk about electromagneticwaves in free space (which still worries some peoplefor the same sort of reason that "action at a distance"worried people), or whether it is better to try to thinkabout an all-permeating, vibrating "aether" is not avery burning issue today. What ml 1ters now is thatMaxwell's Equations are a generally accepted founda-tion f. - discussing electromagnetic behavior under thewidest range of possible situations, and also that Mae.-well's lead in analyzing electromagnetism by means ofthe electric and magnetic fields has led more generallyto the concept of discussing other forms of interaction

173

through some appropriate "field." Indeed, Maxwellhimself was at first very inclined to believe that gravita-tional attraction must also be propagated in this way,but he ran up against difficulties with the energy in-volved which seemed to him then insurmountable."

We have seen that, starting from the picture of "ac-tion at a distance" between charges of electricity, Max-well, following Faraday's lead, could reformulate theproblem in terms of a field acting through, and at allpoints of, space of which the charged particles are, soto speak, now just the "terminals" or "end points." Thediscovery that this electromagnetic field would vibratein free space was a great step toward identifying lightas an electromagnetic wave, since the wave phenome-non of light (interference, diffraction, etc.) had beenknown for a long time. At the same time there had al-ways been some persistent reasons for regarding lightalfernativAy as a corpuscular phenomenon, and Ein-stein was to shot:, half a century later, that Maxwell'svibrating electromagnetic aether, when coupled withPlanck's quantum theory first proposed around 1900,could also then be regarded in a more or less corpuscu-lar manner. What Planck and Einstein showed wasthat the energy in the electromagnetic field could onlyexist in certain minimum-sized bundles or "quanta"dependent in magnitude on the frequency of vibrationand the newly discovered Planck's constant. Them"bundles" of light, or more technically "quanta" of theelectromagnetic field, are generally known today asphotons. So now we can think of electromagnetic in-teractions as either conveyed by the vibrating aetheror equivzlently as conveyed by streams of photonswhich will to some extent behave like particles. In deal-ing with many kinds of interactions, including thosewhich hold an atomic nucleus together, modern physicsfinds it most valuable to be able to think in both these

174

160

_

.....elacement on thesupposition

and linearelasticities are connected

as in a "per-

tram this theattraction between two bodies

having givenquantities of

as electricityon their

surfaces, and then bycomparison with Weber's

value of the statical mea-

sure of a unit ofelectrical current I have

deduced the relation between the elasticityand density

of the cells. Thevelocity of

transverseundulations follows from this

directly and is equal to

193088 miles per second,very nearly that of light.Velocity

192,500 byaberration

of light

195, 777 by Fizeaumiles per sec.

193,118 Galbraith &Haughton's

statement ofFizeau's results

I made out theequations in the

country before I hadany suspicion of the

nearness between

the twovalues of the velocity of

propagation of magneticeffects and that of light, so that I think

I havereason to

believe that the magnetic andluminiferous media are identical

and thatWeber's

number is really, as itappears to be, one half tle velocity of light

in millimetresper second.

If there is in allmedia, in spite of the disturbing influence of gross

matter, thesome rela-

tionbetween the

velocity of light and the staticalaction of

electricity, then the"dielectric ca-

pacity", that is, thecapacity of a Leyden jar of given thickness formed of it, is

proportional to

the square of the index ofrefraction.*

Do you knowany _good

measures ofdielectric capacity of

transparentsubstances? I have

read Faraday & Harris on the subjectand I think they are likely to be

generally too small . I think

FleemingJenkin has found that of gutta

perchacaoutchouc &c. Where

can one findhis method,

and whatmethod do

you recommend?I think I see a way to work with flat plates

of differentsubstances along with

your divided

ring,electrometer.

A & E are platesconnected with each

other, and with a

source ofelectricity; Cis

connected with theground, B & D

are moveableplates

connectedwiththe two halves of the ring.

F is a dielectric.Then by a proper

placing of 0 things may

bearranged so that the

electrification of A and Eproduces no

difference of potentials in the divided ring, afterwhich the

capacity of F follows bycalculation.

ti

'We fiefat;onstlip of Etc:cutest," arid Mag-nitisni

terms without being bound to regard one picture asmore necessarily "real" than the other.

....v.v....," v. we Filifle ui 1.101UFICUIRM in the some oirecrion as me angular momentum of all thevortices, the rotation being proportional to

A the thickness of the mediumB the magnetic intensity along the axisC the index of refraction in the mediumD inversely as the square of the wave length in airF directly as the radius of the vortices

the magnetic capacity.

I have been seeking for experiments lately made which I have lost sight of showing that therotation varies faster than the inverse square of wavelength in air, so that C & D give the true law.

A & B are proved by Verdet,F is not yet capable of proofG is consistent with Verdet's result that the rotation in salts of iron is opposite to that in

diamagnetic substances. I think that molecules of iron are set in motion by the cells and revolvethe opposite way so as to produce a very great energy of rotation but an angular momentum in theopposite direction to that of the vortices.

I find that unless the diameter of the vortices is sensible, no result is likely to be obtainedby making a magnet r 'olve freely about on axis perp. to the magnetic axis. I have tried it buthave not yet got rid of the effects of terrestrial ma3netism which are very strong on a powerfulelectromagnet which I use. I think it more probable that a coil of conducting wire might be foundto have o slight shock tending to turn it about its axis when the electricity is leton or cut off,or that a piece of iron magnetized by a helix might have a similar impulse.

Yours truly

J. C. MAXWELL

161

for 11. c,f

The Electromagnetic Field

Albert Einstein and Leopold Infeld

Excerpt from their book entitled the Evolution of Physics publishedin 1938 and 1961.

DURING the second half of the nineteenth century newand revolutionary ideas were introduced into physics;they opened the way to a new philosophical view, dif-fering from the mechanical one. The results of thework of Faraday, Maxwell, and Hertz led to the devel-opment of modern physics, to the creation of new con-

DEAR STOKES,

I have been reading Jamin's note on the Theory of Reflexion and Refraction, Ann. de Ch.1860, pt. i. p. 413.

I am not yet able to satisfy myself about the conditions to be fulfilled at the surface exceptof course the condition of conservation of energy.

Jamin insists on the equality of the motion both horizontal and vertical in the two media. Ido not see the necessity for equality of motion; but I think action and reaction must to equal be-tween the media, provided the media pure and simple vibrate and nothing along with them.

If the gross matter in each medium does not vibrate, or has a different phase and amplitudefrom the ether, then there will be six relations between the four quantitievtwo portions of etherand two kinds of gross matter.

Have you written anything about the rival theories of reflexion? or can you tell me of anything you agree with or eminently differ from on that subject? I think you once told me thatthe subject was a stiff one to the best skilled in undulations.

Jamin deduces (p. 422) from his conditions of equality ofmotion in the two media for vibra-tions in the plane of incidence that the density of the medium is the same in all substances.

That is to say he gets this by pure mathematics without any experiment.

Or according to him no such vibrations could exist in the media unless they were of equaldensity.

This I think simply disproves his original assumption of the equality of the displacements inthe two media.

In fact the equality of displacements combined with the equality of energy involves theequal ity of density.

Therefore the general theory, which ought to be able to explain the case of media of une-qual density (even if there were none such) must not assume equality of displacements of contig-uous particles on each side of the surface.

162

We can represent this fact in a new way, and shall doso even though it is difficult to understand the advan-

rage of this. The small circle in our drawing representsan attracting body, sa/, the sun. Actually, our diagramshould be imagined as a model in space and not as adrawing on a plane. Our small circle, then, stands for asphere in space, say, the sun. A body, the so-calledtest body, brought somewhere within the vicinity ofthe sun will be attracted along the line connecting thecenters of the two bodies. Thus the lines in our draw-

But there is nothing in the surface of seperation of two media analogous to this gluing to-gether that I can detect.

I have now got materials for calculating the velocity of transmission of a magnetic disturb-ance through air founded on experimental evidence, without any hypothesis about the structure ofthe medium or any mechanical explanation of electricity cr magnetism.

The result is that only transverse disturbances can be propagated andthat the velocity is thatfound byWeber and Kohlrausch which is nearly that of light. This is the velocity with which suchslow disturbonces as we can make would be propagated. If the same law holds for rapid ones, thenthere is no difference between polarized light and rapid electromagnetic disturbonces in one plane.

I have written out so much of the theory as does not involve the conditions at bounding sur-faces, and will send it to the R. S. in a week.

I am trying to understand the conditions at a surface for reflexion and refraction, but theymay not be the same for the period of vibration of light and for experiments made at leisure.

We are devising methods to determine this velocity = electromagnetic electrostatic unit ofelectricity. Thomson is going to weigh an electromotive force. Jenkin and I are going to mea-sure the capacity of a conductor both ways and I have a plan of direct equilibrium between anelectromagnetic repulsion and electrostatic attraction.

Yours truly,

J. C. MAXWELL

163

The Elechomagoettc Field

a test body would behave if brought into the vicinityof the sphere for which the field is constructed.

The lines in our space model are always perpendicu-lar to the surface of the sphere. Since they divergefrom one point, they are dense near the sphere andbecome less and less so farther away. If we increase thedistance from the sphere twice or three timesohen thedensity of the lines, in our space-model, though not inthe drawing, will be four or nine times less. Thus thelines serve a double purpose. On the one hand theyshow the direction of the force acting on a bodybrought into the 'neighborhood of the sphere-sun. Onthe other hand the density of the lines of force in spaceshows how the force varies with the distance. Thedrawing of the field, correctly interpreted, representsthe direction of the gravitational force and its depend-ence on distance. One can read the law of gravitationfrom such a drawing just as well as from a dcscriptionof the action in words, or in the precise and econom-ical language of mathematics. This field representation,as we shall call if, may appear clear and interesting butthere is no reason to believe that it marks any real ad-

reasonable person, an attempt- to make our drawingsomething more than a model leads nowhere.

'We do not intend, however, to discuss the gravita-tional problem just now. It served only as an introduc-tion, simplifying the explanation_ of similar methods ofreasoning in the theory of electricity.

We shall begin with a discussion of the experimentwhich created serious difficulties in our mechanicalinterpretation. We had a current flowing through awire circuit in the form of a circle. In the middle of thecircuit was a magnetic needle. The moment the currentbegan to flow a new force appeared, acting on themagnetic pole, and perpendicular to any line connect-ing the wire and the pole. This force, if caused by acirculating charge, depended, as shiwn by Rowland'sexperiment, on the velocity of the charge. These ex-perimental facts contradicted the philosophical viewthat all forces must act on the line connecting the par-ticles and can depend only upon distance.

The exact expression for the force of a current act-ing on a magnetic pole is quite complicated, muchmore so, indeed, than the expression for gravitationalforces. We can, however, attempt to visualize the ac-tions just as we did in the case of a gravitational force.Our question is: with what force does the current actupon a magnetic pole placed somewhere in its vicin-ity? It would be rather difficult to describe this forcein words. Even a mathematical formula would becomplicated and awkward. It is best to represent allwe know about the acting forces by a drawing, orrather by a spatial model, with lines of force. Some dif-ficulty is caused by the fact that a magnetic pole existsonly in connection with another magnetic pole, form-

180

James Clerk Maxwell

From his Treatise on Electricity and Magnetism published in 1873.

528.] THE discovery by Orsted of the magnetic action of anelectric current led by a direct process of reasoning to that ofmagnetization by electric currents, and of the mechanical actionbetween electric currents. It was not, however, till 1831 thatFaraday, who had been for some time endeavouring to produceelectric currents by magnetic or electric action, discovered the con-ditions of magneto-electric induction. The method which Faradayemployed in his researches consisted in a constant appeal to ex-periment as a means of testing the truth of his ideas, and a constantcultivation of ideas under the direct influence of experiment. Inhis published researches we find these ideas expressed in languagewhich is all the better fitted for a nascent science, because it issomewhat alien from the style of physicists who have been accus-tomed to established mathematical forms of thought.

The experimental investigation by which Ampere established thelaws of the mechanical action between electric currents is one ofthe most brilliant achievements in science.

The whole, theory and experiment, seems as if it had leaped,full grown and full armed, from the brain of the ' Newton of elec-tricity.' It is perfect in form, and unassailable in accuracy, andit is summed up in a formula from which all the phenomena maybe deduced, and which must always remain the cardinal formula ofclectro-dynamics.

The method of Ampere, however, though cast into an inductiveform, does not allow us to trace the formation of the ideas whichguided it. We can scarcely believe that Ampere really discoveredthe law of action by means of the experiments which he describes.We are led to suspect, what, indeed, he tells us himself*, that he

* 77tdorie des Phenomines Eleetrodynamiques, p. 9.

165

s

The Electromagnetic Field

ing a dipole. We can, however, always imagine themagnetic needle of such length that only the force act-ing upon the pole nearer the current has to be takeninto account. The other pole is far enough away forthe force acting upon it to be negligible. To avoid am-biguity we shall say that the magnetic pole broughtnearer to the wire is the positive one.

The character of the force acting upon the positivemagnetic pole can be read from our drawing.

7

T

First we notice an arrow near the wire indicating thedirection of the current, from higher to lower potential. All other lines are lust lines of force bcloncrina rn

coverer. ivery student therefore should read Ampere's researchas a splendid example of scientific style in the statement of a dis-covery, but lie should also study Faraday for the cultivation of ascientific spirit, by means of the action and reaction which willtake place between the newly discovered facts as introduced to himby Faraday and the nascent ideas in his own mind.

It was perhaps for the advantage of science that Faraday, thoughthoroughly conscious of the fundamental forms of space, time, andforce, was not a professed mathematician. He was not temptedto enter into the many interesting researches in pure mathematicswhich his discoveries would have suggested if they bad beenexhibited in a mathematical form, and he did not feel called uponeither to force his results into a shape acceptable to the mathe-ma.ical taste of the time, or to express them in a form whichmathematicians might attack. He was thus left at leisure tolo his proper work, to coordinate his ideas with his facts, ar toexpress them in natural, untechnical language.

It is mainly with the hope of making these ideas the basis of amathematical method that I have undertaken this treatise.

529.] We are accustomed to consider the universe as made up ofparts, and mathematicians usually begin by considering a singleparticle, and then conceiving its relation to another particle, and soon. This has generally been supposed the most natural method.To conceive of a particle, however, requires a process of abstraction,since all our perceptions are related to extended bodies, so thatthe idea of the all that is in our consciousness at a given instantis perhaps as primitive an idea as that of any individual thing.Hence there may be a mathematical method in which we proceedfrom the whole to the parts instead of from the parts to the whole.For example, Euclid, in his first book, conceives a line as tracedout by a point, a surface as swept out by a line, and a solid asgenerated by a surface. But he also defines a surface as the

166

such a model is not as simple as in our previous exam-ple, where the lines of force were straight. In our nextdiagram only one line of force is drawn in order to

clarify the procedure. The force vector lies on thetangent to the line of force, as indicated. The arrow ofthe force vector and the arrows on the line of forccpoint in the same direction. Thus this is the directionin which the force acts on a magnetic pole at thispoint. A good drawing. or rather a good model, alsotells is something about the length of the force vectorat any point. This vector has to be longer where thelines are denser, i.e., near the wire, shorter where the

Tho IrornaFlot,c

we come once more to the conclusion that the forceacts in a direction perpendicular to any line connectingthe wire and the pole, for the tangent to a circle isalways perpendicular to its radius. Our entire knowl-edge of the acting forces can be summarized in theconstruction of the field. We sandwich the concept ofthe field between that of the current and that of themagnetic pole in order to represent the acting forcesin a simple way.

Every' current is associated with a magnetic field,i.e., a force always acts on a magnetic pole broughtnear the wire through which a current flews. We mayremark in passing that tnis property enables us to con-struct sensitive apparatus for detecting the existence ofa current. Once having learned how to read the charac-ter of the magnetic forces from the field model of acurrent, we shall always draw the field surroundingthe wire through which the current flows, in order torepresent *he action of the magnetic forces at anypoint in space. Our first example is the so-called sole-noid. This is, in fact, a coil of wire as shown in thedrawing. Our aim is to learn, by experiment, all wecan about the magnetic field associated with the cur-rent flowing through a solenoid and to incorpo, ate thisknowledge in the construction of a field. A drawingrepresents our result. The curved lines of force areclosed, and surround the solenoid in a way character-istic of the magnetic field of a current. Sec top p. 132.

The field of a bar magnet can be represented in thesame way as that of a current. Another drawing showsthis. The lines of force are directed from the positiveto the negative pole. The force vector always lies on

183

the tangent to the line of force and is longest near thepoles because the density of the lines is greatest at thesepoints. The force vector represents the action of themagnet on a positive magnetic pole. In this case themagnet and not the current is the "source" of the field.

Our last two drawings ..:.,,,:ld be carefully com-pared. In the first, we have the magnetic field of a cur-rent flowing through a solenoid; in the second, the fieldof a bar magnet. I et us ignore both the solenoid andthe bar and observe only the two outside fields. Weimmediately notice that they are of exactly the samecharacter; in each case the lines of force lead from onecnd of the solenoid or bar to the other.

The field representation yields its first fruit! Itwould be rather difficult to see any strong similarity

184

The Electromagnetic Field

between the current flowing through a solenoid and abar magnet if this were not revealed by our construe,lion of the field.

The concept of field can now be put to a muchmore severe test. We shall soon see whether it is any-thing more than a new representation of the actingforces. We could reason: assume, for a moment, thatthe field characterizes all actions determined by itssources in a unique way. This is only a guess. It wouldmean that if a solenoid and a bar magnet have the samefield, then all their influences must also be the same. Itwould mean that two solenoids, carrying electric cur-rents, behave like two bar magnets, that they attract orrepel each other depending, exactly as in the case ofbars, on their relative positions. It would also mean thata solenoid and a bar attract or repel each other in thesame way as two bars. Briefly speaking, it would meanthat all actions of a solenoid through which a currentflows, and of a corresponding bar magnet arc the same,since the field alone is responsible for them, and thefield in both cases is of the same character. Experimentfully confirms our guess!

How difficult it would be to find those facts withoutthe concept of field! The expression for a force actingbetween a wire through which a current flows and amagnetic pole is very complicated. In the case of twosolenoids we should have to investigate the forces withwhich two currents act upon each other. But if we dothis, with the help of the field, we immediately noticethe character of all those actions at the moment whenthe similarity between the field of a solenoid and thatof a bar magnet is seen.

We have the right to regard the field as something

185

much more than we did at first. The properties of thefield alone appear to be essential for the description ofphenomena; the differences in source do not matter.The concept of field reveals its importance by leadingto new experimental facts.

The field proved a very helpful concept. It began as

something placed between the source and the magneticneedle in order to describe the acting force. It wasthought of as an "agent" of the current, through whichall action of the current was performed. But now theagent also acts a:: an interpreter, one who translates thelaws into a simple, clear language, easily understood.

The first success of the field description suggeststhat it may be convenient to consider all actions ofcurrents, magnets and charges indirectly, i.e., with thehelp of the field as an interpreter. A field may be re-garded as something always associated with a current.It is there even in the absence of a magnetic pole to testits existence. Let us try to follow this new clew con-sistently.

The field of a charged conductor can be introducedin much the same way as the gravitational field, or thefield of a current or magnet. Again only the simplestexample! To design the field of a positively chargedsphere, we must ask what kind of forces are acting ona small positively charged test body brought near thesource of the field, the charged sphere. The fact thatwe use a positively and not a negatively charged testbody is merely a convention, indicating in which direc-tion the arrows on the line of force should be drawn.The model is analogous to that of a gravitational field(p. 1 26) because of the similarity between Coulomb's

law and Newton's. The only difference between the

186

two models is that the arrows point in opposite direc-tions. Indeed, we have repulsion of two positivecharges and attraction of two masses. However, thefield of a sphere with a negative charge will be iden-tical with a gravitational field since the small positivetesting charge will be attracted by the source of thefield.

If both electric and magnetic poles are at rest, thereis no action between them, neither attraction nor re-

187

I

st

188

pulsion. Expressing the same fact in the field languagewe can say: an electrostatic field does not influence amagnetostatic one and vice versa. The words "staticfield" mean a field that does not change with time.The magnets and charges would rest near one anotherfor an eternity if no external forces disturbed them.Electrostatic, magnetostatic and gravitational fields areall of different character. They do not mix; each pre-serves its individuality regardless of the others.

Let us return to the electric sphere which was, untilnow, at rest, and assume that it begins to move due tothe action of some external force. The charged spheremoves. In the field language this sentence reads: thefield of the electric charge changes with time. But themotion of this charged sphere is, as we already knowfrom Rowland's experiment, equivalent to a current.Further, every current is accompanied by a magneticfield. Thus the chain of our argument is:

motion of charge --> change of an electric field4,

current associated magnetic field.

We, therefore, conclude: The change of an electricfield produced by the motion of a charge is always ac-companied by a magnetic field.

Our conclusion is based on Ocrsted's experiment butit covers much more. It contains the recognitica thatthe association of an electric field, changing in time,with a magnetic field is essential for our further prgu-ment.

As long as a charge is at rest there is only an elecrro-Iptatic field. But a magnetic field appears as soon as thecharge begins to move. We can say more. The mag-

N

The Electromagnetic Field

nem field created by the motion of the charge will bestronger if the charge is greater and if it moves faster.This also is a consequence of Rowland's experiment.Once again using the field language, we can say: thefaster the electric field changes; the stronger the. ac-companying magnetic field.

We have tried here to translate familiar facts fromthe language of fluids, constructed according to theold mechanical view, into the new language of fields.We shall see later how clear, instructive, and far-reaching our new language is.

THE TWO PILLARS OF THE FIELD THEORY

"The change of an electric field is accompanied by amagnetic field." If we interchange the words "mag-netic" and "electric," our sentence reads: "The changeof a magnetic field is accompanied by an electric field."Only an experiment can decide whether or not thisstatement is true. But the idea of formulating this prob-lem is suggested by the use of the field language.

Just over a hundred years ago, Faraday performedan experiment which led to the great discovery of in-duced currents.

The demonstration is very simple. We need only asolenoid or some other circuit, a bar magnet, and oneof the many types of apparatus.for detecting the exist-ence of an electric current. To begin with, a bar mag-net is kept at rest near a solenoid which forms a closedcircuit. No current flows through the wire, for nosource is present. There is only the magnetostatic fieldof the bar magnet which does not change with time.Now, we quickly change the position of the magneteither by removing it or by bringing it nearer the sole-

189

noid, whichever we prefer. At this moment, a currentwill appear for a very short time and then vanish.

Whenever the position of the magnet is changed, thecurrent reappears, and can be detected by a sufficientlysensitive apparatus. But a currentfrom the point ofview of the field theorymeans the existence of anelectric field forcing the flow of the electric. fluidsthrough th . wire. The current, and therefore the elec-tri.c field, too, vanishes when the magnet is again atrest.

Imagine for a moment that the field language is un-known and the results of this experiment have to bedescribed, qualitatively and quantitatively, in the lan-guage of old mechanical concepts. Our experimentthen shows: by the motion of a magnetic dipole a newforce was created, moving the e:xtric fluid in the wire.The next question would be: upon what does thisforce depend? This would be very difficult to answer.We should have to investigate the dependence of theforce upon the velocity of the niagnet, upon its shape,and upon the shape of the circuit. Furthermore, thisexperiment, if interpreted in the old language, gives usno hint at all as to whether an induced current can beexcited by the motion of another circuit carrying acurrent, instead of by motion of a bar magnet.

It is quite a different matter if we use the field Ian-

190

gunge and again trust our principle that the action isdetermined by the field. We see at once that a solenoidthrough which a current flows would serve as well as abar magnet. The drawing shows two solenoids: one,small, through which a current flows, and the other,in which the induced current is detected, larger. We

Caiff619

could move the small solenoid, as we previously movedthe bar magnet, creating an induced current in thelarger solenoid. Furthermore, instead of moving thesmall solenoid, we could create and destroy a magneticfield by creating and destroying the current, that is,by opening and closing the circuit. Once again, newfacts suggested by the field theory are confirmed byexperiment!

Let us take a simpler example. We have a closed wirewithout any source of current. Somewhere in thevicinity is a magnetic field. It means nothing to uswhether the source of this magnetic fief? is anothercircuit through which an electric =cent flows, or abar magnet. Page 140 shows the closed circuit andthe magnetic lines of force. The qualitative and quanti-tative description of the induction phenomena is verysimple in term; of the field language. As marked onthe drawing, some lines of force go through the surfacebounded by the wire. We have to consider the lines offorce cutting that part of the plane which has the wire

for a rim. No electric current is present so long as thefield does not change, no matter how great its strength.But a current begins to flow through the rim-wire assoon as the number of lines passing through the surfacesurrounded by wire changes. The current is deter-mined by the change, however it may be caused, of thenumber of lines passing the surface. This change in thenumber of lines of force is the only essential conceptfor both the qualitative and the quantitative descriptions of the induced current. "The number of lineschanges" means that the density of the lines changesand this, we remember, means that the field strengthchanges.

These then are the essential points in our chain ofreasoning: change of magnetic field --> induced cur-rent --> motion of charge existence of an electricfield.

Therefore: a changing magnetic field is accompaniedby an electric field.

Thus we have found the two most imporrant pillarsof support for the theory of the electric and magneticfield. The first is the connection between the changing

192

The Electrornagnehc Field

electric field and the magnetic field. It arose fromOersted's experiment on the deflection of a magneticneedle and led to the conclusion: a changing electricfield is accompanied by a magnetic field.

The second connects the changing magnetic fieldwith the induced current and arose from Faraday'sexperiment. Both formed a basis for quantitative de-scription.

Again the electric field accompanying the changingmagnetic field appears as something real. We had toimagine, previously, the magnetic field of a currentexisting without the testing pole. Similarly, we mustclaim here that the electric field exists without the wiretesting the presence of an induced current.

In fact, our two-pillar structure could be reduced toonly one, namely, to that based on Oersted's experi-ment. The result of Faraday's experiment could be de-duced from this with the law of conservation of en-ergy. We used the two-pillar structure only for thesake of clearness and economy.

One more consequence of the field descriptionshould be mentioned. There is a circuit through whicha current flows, with for instance, a voltaic battery asthe source of the current. The connection between thewire and the source of the current is suddenly broken.There is, of course, no current now! But during thisshort interruption an intricate process takes place, aprocess which could again have been foreseen by thefield theory. Before the interruption of the currentthere was a magnetic field surrounding the wire. Thisceased to exist the moment the current was inter-rupted. Therefore, through the interruption of a cur-rent, a magnetic field disappeared. The number of lines

of force passing through the surface surrounded by thewire changed very rapidly. But such a rapid change,however it is produced, must create an induced cur-rent. What really matters is the change of the magneticfield making the induced current stronger if the changeis greater. This consequence is another test for thetheory. The disconnection of a current must be accom-panied by the appearance of a strong, momentary in-duced current. Experiment again confirms the predic-tion. Anyone who has ever disconnected a currentmust have noticed that a spark appears. This spark re-veals the strong potential differences caused by therapid change of the magnetic field.

The same process can be looked at from a differentpoint of view, that of energy. A magnetic field dis-appeared and a spark was created. A spark representsenergy, therefore, so also must the magnetic field. Touse the field concept and its language consistently, wemust regard the magnetic field as a store of energy.Only in this way shall we be able to describe the elec-tric and magnetic phenomena in accordance with thelaw of conservation of energy.

Starting as a helpful model the field became moreand more real. It helped us to understand old facts andled us to new oars. The attribution of energy to ,thefield is one step further in the development in whichthe field concept was stressed more and more, and theconcepts of substances, so essential to the mechanicalpoint of view, were more and more suppressed.

THE REALITY OF THE FIELD

The quantitative, mathematical description of thelaws of the field is summed up in what arc called Max-

194

well's equations. The facts mentioned so far led to theformulation of these equations but their content ismuch richer than we have been able to indicate. Theirsimple form conceals a depth revealed only by carefulstudy.

The formulation of these equations is the most im-portant event in physics since Newton's time, not onlybecause of their wealth of content, but also becausethey form a pattern for a new tyre of law.

The characteristic features of Maxwell's equations,appearing in all other equations of modern physics, aresummarized in one sentence. Maxwell's equations arelaws representing the structure of the field.

Why do Maxwell's equations differ in form andcharacter from the equations of classical mechanics?What does it mean that these equations describe thestructure of the field? 1-Tow is it possible that, from theresults of Oersted's and Faraday's xperiments, we canform a new type of law, which proves so important forthe further development of physics?

We have already seen, from Oersted's experiment,how a magnetic field coils itself around a changingelectric field. We have seen, from Faraday's experi-ment, how an electric field coils itself around a chang-ing magnetic field. To outline some of the characteris-tic features of Maxwell's theory, let us, for the moment,focus all our attention on one of these experiments,say, on that of Faraday. We repeat the drawing inwhich an electric current is induced by a changing mag-netic field. We already know that an induced currentappears if the number of lines of force, passing the sur-face bounded by the wire, changes. Then the current

195

will appear if the magnetic field changes or the circuit.is deformed or moved: if the number of magnetic linespassing through the surface is changed, no matter howthis change is caused. To take into account all thesevarious possibilities, to discuss their particular influ-ences, would necessarily lead to a very complicatedtheory. But can we not simplify our problem? Let ustry to eliminate from our considerations everythingwhich refers to the shape of the circuit, to its length,to the surface enclosed by the wire. Let us imaginethat the circuit in our last drawing becomes smaller andsmaller, shrinking gradually to a very small circuit en-closing a certain point in space. Then everything con-cerning shape and size is quite irrelevant. In this limit-ing process where the closed curve shrinks to a point,size and shape automatically vanish from our consid-erations and we obtain laws connecting changes ofmagnetic and electric field at an arbitrary point inspace at an arbitrary instant.

Thus, this is one of the principal steps leading toMaxwell's equations. It is again an idealized experimentperformed in imagination by repeating Faraday's ex-periment with a circuit shrinking to a point.

196

The El,;ctromagneuc Field

We should really call it half a step rather than awhole one. So far our attention has been focused onFaraday's experiment. But the other pillar of the fieldtheory, based on Oersted's experiment, must be consid-ered just as carefully and in a similar manner. In thisexperiment the magnetic lines of force coil themselvesaround the current. By shrinking the circular magneticlines of force to a point, the second half-step is per-formed and the whole step yields a connection be-tween the changes of the magnetic and electric fieldsat an arbitrary point in space and at an arbitrary instant.

But still another essential step is necessary. Accord-ing to Faraday's experiment, there must be a wire test-ing the existence of the electric field, just as there mustbe a magnetic pole, or needle, testing the existence ofa magnetic field in Oersted's experiment. But Maxwell'snew theoretical idea goes beyond these experimehtalfacts. The electric and magnetic field, or in short, theelectromagnetic field is, in Maxwell's theory, some-thing real. The electric field is produced by a changingmagnetic field, quite independently, whether or notthere is a wire to test its existence; a magnetic field isproduced by a changing electric field, whether or notthere is a magnetic pole to test its existence.

Thus two essential steps led to Maxwell's equations.The first: in considering Oersted's and Rowland's ex-periments, the circular line of the magnetic field coil-ing itself around the current and the changing electricfield, had to be shrunk to a point; in considering.Fara-day's experiment, the circular line of the electric fieldcoiling itself around the changing magnetic field had tobe shrunk to a point. The second step consists of therealization of the field as something real; the electro-

197

i

198

magnetic field once created exists, acts, and changesaccording to Maxwell's laws.

Maxwell's equations describe the structure of theelectromagnetic field. All space is the scene of theselaws and nor. as for mechanical laws, only points inwhich matter or charges are present.

We remember how it was in mechanics. By knowingthe position and velocity of a particle at one singleinstant, by knowing the acting forces, the whole futurepath of the particle could be forseen. In Maxwell'stheory, if we know the field at one instant only, wecan deduce from the equations of the theory how thewhole field will change in space and rime. Maxwell'sequations enable us to follow the history of the field,just as the mechanical equations enabled us to followthe history of material particles.

But there is still one essential difference between me-chanical laws and Maxwell's laws. A comparison ofNewton's gravitational laws and Maxwell's field lawswill emphasize some of the characteristic features ex-pressed by these equations.

With the help of Newton's laws we can deduce themotion of the earth from the force-acting between thesun and the earth. The laws connect the motion of theearth with the action of the far-off sun. The earth andthe sun, though so far apart, are both actors in the playof forces. 1

In Maxwell's theory there are no material actors.The mathematical equations of this theory express thelaws governing the electromagnetic field. They do not,as in Newton's laws, connect two widely separated

events; they do not connect the happenings here withthe conditions there. The field here and now dependson the field in the immediate neighborhood at a timejust past. The equations allow us to predict what willhappen a little further in space and a little later in time,if we know what happens here and now. They allowus to increase our knowledge of the field by small cps.We can deduce what happens here from that whichhappened far away by the summation of these verysmall steps. In Newton's theory, on the contrary, onlybig steps connecting distant events are permissible. Theexperiments of Oersted and Faraday can be regainedfrom Maxwell's theory, but . nly by the summation ofsmall steps each of which is governed by Maxwell'sequations.

A more thorough mathematical study of Maxwell'sequations shows that new and really unexpected con-clusions can be drawn and the whole theory submittedto a test on a much higher level, because the theoreticalconsequences are now of a quantitative character andarc revealed by a whole chain of logical arguments.

Let us again imagine an idealized experiment. A smallsphere with an electric charge is forced, by some ex-ternal influence, to oscillate rapidly and in a rhythmicalway, like a pendulum. With the knowledge we alreadyhave of the changes of the field, how shall we describeeverything that is going on here, in the field language?

The oscillation of the charge produces a changingelectric field. This is always accompanied by a chang-ing magnetic field. If a wire forming a closed circuit isplaced in the vicinity, then again the changing mag-netic field will be accompanied by an electric currentin the circuit. All this is merely a repetition of known

199

facts, but the study of /v1a,...,41's equations gives amuch deeper insight into the problem of the oscillatingelectric charge. By mathematical deduction from Max-well's equations we can detect the clyz racter of thefield surrounding an oscillating charge its structure

. near and far from the source and its change with time.The outcome of such deduction is the electromagneticwave. Energy radiates from the oscillating charge trav-eling with a definite speed through space; but a trans-ference of energy, the motion of a state, is character-istic of all wave phenomena.

Different types of waves have already been consid-ered. There was the longitudinal wave caused by thepulsating sphere, where the changes of density werepropagated through the medium. There was the jelly-like medium in which the transverse wave spread. Adeformation of the jelly, caused by the rotation of thesphere, moved through the medium. What kind ofchanges are now spreading in the case of an electro-magnetic wave? Just the changes of an electromagneticfield! Every change of an electric field produces a mag-ne:ic field; every change of this magnetic field pro-duces an electric field; every change of . . . , and soon. As field represents energy, all these changes spread-ing out in space, with a definite velocity, produce awave. The electric and magnetic lines of force alwayslie, as deduced from the theory, or planes perpendicu-lar to the direction of propagation. The wave pro-duced is, therefore, transverse. The original features ofthe picture of the field we formed from Oersted's andFaraday's experiments are still preserved, but we nowrecognize that it has a deeper meaning.

The electromagnetic wave spreads in empty space.

200

The Electromagnetic Field

This, again, is a consequence of the theory. If the oscil-lating charge suddenly ceases to move, then, its fieldbecomes electrostatic. But the series of waves createdby the oscillation continues to spread. The waves leadan independent existence and the history of theirchanges can be followed just as that of any other ma-terial object.

We understand that our picture of an electromag-netic wave, spreading with a certain velocity in spaceand changing in time, follows from Maxwell's equa-tions only because they describe the structure of theelectromagnetic field at any point in space and for anyinstant.

There is another very important question. Withwhat speed does the electromagnetic wave spread inempty space? The theory, with the support of somedata from simple experiments having nothing to dowith the actual propagation of waves, gives a clear an-swer: the velocity of an electromagnetic wave is equalto the velocity of light.

Oersted's and Faraday's experiments formed thebasis on which Maxwell's laws were built. All our re-sults so far have come from a careful study of theselaws, expressed in the field language. The theoreticaldiscovery of an electromagnetic wave spreading withthe speed of light is one of the greatest achievements inthe history of science.

Experiment has confirmed the prediction of theory.Fifty years ago, Hertz proved, for the first time, theexistence of electromagnetic waves and confirmed ex-perimentally that their velocity is equal to that of light.Nowadays, millions of people demonstrate that elec-tromagnetic waves are sent and received. Their ap-

201

paratus is far more complicated than that used byHertz and detects the presence of waves thousands ofmiles from their sources instead of only a few yards.

s)0r-,14'.

Tr1-11

_.] . .

16 Radiation Belts Around the Earth

James Van Allen

1959

Sfar, the most interesting and leastexpected result of mans explora-tion of the immediate vicinity of

the earth is the discovery that our planetis ringed by a regionto be exact, two re-gionsof high-energy radiation extend-ing many thot.sands of miles into space.Th.: discovery is of course troubling toastronauts; somehow the human bodywill have to be shielded from this radia-tion, even on a rapid transit through theregion. But geophysicists, astrophysi-cists, solar astronomers and cosmic-rayphysicists are enthralled by the fresh im-plications of these findings. The configu-ration of the region and the radiation itcontains bespeak a major physical phe-nomenon involving cosmic rays and solarcorpuscles in the vicinity of the earth.This enormous reservoir of charged par-ticles plays a still-unexplained role asmiddleman in the interaztion of earthand sun which is reflected in magneticstorms. in the airglow and in the beauti-ful displays of the aurora.

The story of the investigation goesback to 1952 and 1953, before any ofus could think realistically about the useof earth satellites to explore the environ-ment of the earth. Parties from Our lab-oratory at the State University of Iowaspent the summers of those years aboardCoast Guard and naval vessels. cruisingalong a 1,500nule line from the watersof Baffin Bay, near the magnetic pole inthe far northwestern corner of Green-land, southward to the North Atlanticoff the coast of Newfoundland. Alongthe way we launched a series of rocket-

carrying balloons"rockoons." (The bal-loon lifts a small rocket to an altitude of12 to 15 miles, whence the rocket car-ries a modest payload of instruments toa height of 80 to 70 miles.) Our objec-tive was to develop a profile of the cos-mic-ray intensities at high altitudes andlatitudes, and thus to learn the nature ofthe low-energy cosmic rays which atlower altitudes and latitudes are de-flected by the earth's magnetic field orabsorbed in the atmosphere.

Most of the readings radioed downfrom the rockets were in accord withplausible expectations. Two rockoonssent aloft in 1953. however, provided itswith a puzzle. Launched near New-foundland by Melvin Gottlieb and Les-he Meredith, they encountered a zoneof radiation beginning at an altitude of30 miles that was far stronger than wehad expected. At first we were u»easyabout the proper operation of our in-struments. But critical examination ofthe data convinced its that we bad un-questionably encountered somethingnew in the upper atmosphere.

Significant:), these measurements weremade in the northern auroral zone. Inthis 7011e, which forms a ring sonic 23degrees south of the north geomagneticpole, the incidence of visible aurorasreaches its maximum. Since rockets firednorth and south of the zone had revealednothing unusual, we speculated that thestrong radiation played some part in theaurora. Showers of particles from thesun, it was thought, come plunging intothe atmosphere along magnetic lines of

force and set off these displays (see "Au-rora and Airglow," by C. T. Elvey andFranklin E. Roach; SCIENTIFIC AMERI-CAN, September, 19551. But the theoryunderlying this explanation did not ex-plain satisfactorily why the aurora andthe high-intensity radiation we had de-tected should occur in the auroral zoneand not in the vicinity of the geomag-netic pole itself. Nor could it accountfor the high energies required to carrythe solar particles through the atmos-pher to such relatively low altitudes.

Tile mystery deepened when wefour d in later studies that the radiationpers sts almost con' nuously in the zoneabove 30 miles, irrespective of visibleauroral displays and other known high-altitude disturbances. More discriminat-ing detectors established that the radia-tion contains large numbers of electrons.Our original observations had detectedX-rays only; now it turned out that theX-rays had been generated by the im-part of electrons on the skin of the in-strument package (as if it ' I been the"target" in an X-ray tube) and on thesparse atoms of the upper atmosphereitself. Sydney Chapman and CordonLittle at the University of Alaska sug-gested that such a process might wellaccount for the attenuation of radio sig-nals in the lower ionosphere of the auro-ral zones.

The international Geophysical Year- nave us our first opportunity to in-

vestigate the "am soft radiation" ona more comprehensive scale. During the

203

DISTANCE I EARTH RADII I

STRUCTURE OF RADIATION BELTS revealed by contours ofradiation intensity (black lines) is shcmn schematically by shading

summer and fall of 1957 Laurence Ca-hill and I launched a number of rockoonsoff the coast of Greenland and also gotoff one successful flight in Antarctica.The latter flight established that the ra-diation exists in the southern as well asthe northern auroral zone. In February,1958. Carl Mellwain fired a series oftwo-stage rockets through visible auro-ras above Fort Churchill in Canada, anddiscovered that the radiation includes

204

(leg); dots (right) suggest distribution of particles in the Nobelts. Contour numbers give counts per second; horizontal scale

energetic protons (hydrogen nuclei) aswell as electrons.

Meanwhile all of us had been pushinga new development that greatly expand-ed the possibilities for high-altitude re-search. During the summer of 1955 thePresident and other Government author-ities were finally persuaded that itmight be feasible to place artificial satel-lites in orbit, and authorized an I. C. Y.project for this purpose. In January,

1956, a long-standing group of high-altitude experimentalists, called theRocket awl Satellite Research Panel,held a symposium to consider how thesatellites could be most fruitfully em-ployed. At that meeting our group pro-posed two projects. One was to put asatellite into an orbit nearly pole-to-poleto survey the auroral radiation in boththe north and south auroral zones. Suchorbits, however, did not appear to be

Radiation Belts Around the Earth

shows distance in earth radii (about 4.000 miles) from the centerof the earth. Particles in the inner belt may originate with the

technically feasible in the immediatefuture. For the time being we wereforced to abandon the use of a satelliteto probe farther into the auroral softradiation. We also suggested that a satel-lite orbiting over the lower latitudes ofthe earth might usefully be employed ina comprehensive survey of cosmic-rayintensities over those regions.This proj-ect was adopted, and we were author-ized to prepare suitable experimental

.

radioactive decay of neutrons !Aerated in the upper atmosphere bycosmic rays; those in the outer belt probably originate in the sun.

apparatus [see "The Artificial Satelliteas a Research Instrument," by James A.Van Allen; SCIENTIFIC AMERICAN, No-vember, 1958]. It was planned to placethis apparatus on one of the early Van-guard vehicles.

The difficulties and failures of theVanguard are now history. Sputnik Istimulated some high government offi-cials to accept a proposal that a num-ber of us had been urging for more than

a year: to use the proven Jupiter Crocket as a satellite-launching vehicle.As a result on January 31, 1958, Ex-plorer I went into orbit carrying oursimple cosmic-ray detector and a radioto broadcast its readings.

In the first reports from stations locat-ed in the U. S. the intensity of radiationincreased with altitude along the expect-ed curve, Several weeks later, however,we began to get tapes from stations in

EXPLORER IV AND PIONEER Ill gave the first detailed picture of the radiation belts.The Explorer IV satellite (Mort ellipiel monitored radiation levels for nearly too monthsat altitudes up to 1.300 miles. The Pioneer Ill lunar probe (long ellipse) provided data outto 65,000 miles. Its orbit is -boon distorted because of the earth's rotation during flight.

EXPLORER IV ORBIT covered the entire region 51 degrees north and south of the equator;the black curse shoo. a small p.m of its trace on the earths surface. More than 25 observalion station. (colored (low recorded data from several thousand of the satellite's passes.

100

10 000

,.

,0

irAIlligil

10000 20000 30000 40.000

RADIAI DISTANCE FROM CENTER Of EARTH IM US)

PIONEER III DATA gave the first confirmation of two distinct rings of particles. Countingrates on hoth the outhound (black curve) and the inbound (gray curve) legs of the flightshooed two peaks. The too curves differ because they cover different sections of the belts.

206

South America and South Africa whichgave us counting rates for much higheraltitudes, due to the eccentricity of thesatellite's orbit. These records brought usa new surprise. At high altitudes over theequatorial region the apFarent countingrate was very low; in some passes Itdropped to zero for several minutes. Yetat lower altitudes the rate had quite"reasonable" valuesfront 30 to 50counts a second. Again we were uneasyabout the trustworthiness of the mstruments. The only alternative scented tobe that cosmic rays do not stake theuppermost layers of the atmosphere overthe tropics, and we were quite unableto accept this conclusion.

Our uneasiness was increased by theincompleteness of our early data. TheExplorer 1 apparatus broadc.sst its obser-vations continuously, but its signalscould be picked up only intermittently,when the satellite came within range ofa ground station. Our original apparatus,designed and developed by George Ludwig for the Vanguard satellites, includeda magnetictape recorder which couldstore its observations for a complete orbitaround the earth and then report them ina "burst" on radio command from theground.

ny early February. working with theJet Propulsion Laboratory, we had

converted this apparatus for use its theExplorer II satellite. The first attempt toget it into orbit failed. A second rocketplaced Explorer III. carrying identicalapparatus, in orbit on March 26. Thissatellite fully confirmed the anomalousresults of Explorer I. At altitudes of 200to 300 tusks the counting rate was low.When the satellite went out to 500 to600 miles, the apparein rate ascendedravilly and then dropped almost to zero.One day, as we were puzzling over thefirst tapes from Explorer III, Mcliwainsuggested the first plausible explanationfor their peculiar readings. He had justbeen calibrattng Ins rocket instruments,and called our attention to somethingthat we all knew but had temporarilyforgotten: A sufficiently high level ofradiation can jam the counter and sendthe apparent counting rate to zero. Wehad discovered an enormously high levelof radiation, not a lack of it As Ernes'Hay, a member of our group, inaccu-rately but graphically exclaimed:"Spaceis radioactive!"

During the next two months Explorer111 produced a large number of playbackrecords, every one of which showed thesame effect. At low altitudes the count-ing rate was reasonably attributable to

DETECTOR FOR CHANNEL 4 DETECTORS FOR CHANNELS I AND 3

PLASTIC

SCINTITIATOR.

rqHOTOMULTIPLIER

FOIL

TUBE

CESIUM IODIDE

SCINTIILATOR

FOIL

PHO T OmUiTIPu ER

TUBE

IIOW POW

NPOWER

NSMITTER

ELECTRONICS, BATTERIES, HIGH POWER TRANSMITTER DETECTOR FOR CHANNELS 2 AND 5

GEIGER TUBE

NON LINEARNETWORK

1

DISCRIMINATOR

1 2 3...t28

LOW POWERTRANSMITTER

di 16

OF 16 IISCALE CHANNEL H

2048

SCALE

OF 1281 1 CHANNEL 2

CALIBRATION RELAY

1 2 3 4 2048

SHIELDED

GEIGER TUBE

`1'2

3 4, 64

SCALE OF 2048

ELECTROMETER

di 2048

CHANNEL

CHANNEL I

CHANNEL 2

CHANNEL 3

..,CATE OF 64L

CHANNEL 3

I, CHANNEL

HIGH POWERTRANSMITTER

64 COUNTS PER CYCIE

1111111111

ON !moo 10CHANNELS

EXPLORER IV INSTRUMENTS were designed to give a detailedpicture of the nature and intensity of the radiation. Plastic scint'llator counted only charged particles above certain energies; twoddlcrent scaling factors adapted it to both high and low countingrates. Cesium iodide scintillator measured the total energy input

rather than ind'vidual particles. Sh elded and unshielded Geigertubes could be compared to estimate the penetrability of the radii'.tion. Radio signals suggested by the red curves in upper drawingwere recorded by ground stations and later played through amultichannel owillograph to yield records like that shown below.

207

0

wiES

COS 7,000 5C0

"VONE MEV

TWO SETS OF CONTOURS from readings on opposite sides ofthe earth I left and center) show the northern and southern "horns"

cosmic rays. At higher altitudesthe pre-cise height depended on both latitudeand longitudethe count increased tovery high values. Up to the points atwhich the counter jammed, it showedcounting rates more than 1,000 timesthe theoretical expectation for cosmicrays. From the rate of increase and thelength of the periods of jamming wejudged that the maximum count prob-ably went to several times this level.Since the radiation appeared to resem-ble the auroral soft radiation, we wouldnot have been surprised to find it in theauroral zone or along the magnetic linesof force that connect these zones. But inthe equatorial latitudes these lines offorce lie much farther out in space thanthe altitudes attained by the satellites.

On May 1 of last year we were ableto report with confidence to the NationalAcademy of Sciences and the AmericanPhysical Society that Explorers I andIII had discovered a major new phenom-enon: a very great intensity of radiationabove altitudes of some 500 miles overthe entire region of their traverse, some34 degrees north and south of the equa-tor. At the same time we advanced theidea that the radiation consists ofcharged particlespresumably protonsand electronstrapped in the magneticfield of the earth.

We could rule out uncharged particle;and gamma and X-rays because theywould not be confined by the magneticfield, and so would be observed at loweraltitudes. The possibility that the earth's

rmES

coS I...AO 1500

of radiation, which point toward the auroral zone.: the contournumber, show radiation intensity in counts per second. The "tipped"

magnetic field might act as a trap forcharged particles was first suggested bythe Norwegian physicist Carl Stormerin a classical series of papers beginningsome 50 years ago, and there was aconsiderable body of evidence for theexistence of low-energy charged parti-cles throughout our solar system andspecifically in the vicinity of the earth.But there had been no indication thatthese particles would possess the highenergies we had detected.

From StOrmer's theoretical discussionand our own observations we evolved arough picture of the trapping mechan-ism. When a fast-moving charged parti-cle is injected into the earth's magneticfield, it describes a corkscrew-shapedtraject9ry, the center line of which liesalong a magnetic line of fore. The turnsof the helical path are ,e open overthe equator but become .ighter as theparticle reaches the stronger magneticfield toward the poles [see illustration atbottom of opposite page]. At the lowerend of its trajectory the particle goes intoa flat spiral and then winds back alonga similar path to the other hemisphere,mak.ng the transit from one hemisphereto the other in a second or so. Duringthis time its line of travel shifts slightly,so that the particle drifts slowly aroundthe earth as it corkscrews from hemi-sphere to hemisphere. An electron driftsfrom west to cast; a proton, in the op-posite direction. At each end of its paththe particle descends into regions ofhigher atmospheric density; collisions

with the atoms of atmospheric gasescause it gradually to change its trajectoryand to lose energy. After a period of daysor weeks the particle is lost into the loweratmosphere.

There was obviously an agent scien-

tific need to extend these observa-tions with equipment of greater dynamicrange and discrimination. In April of1958 we persuaded several Federalagencies to support further satelliteflights of our radiation equipment as anadjunct to the I. G. Y. program, and wereceived the enthusiastic support of theNational Academy of Sciences for thecontinuation of our work. We also per-suaded the Army Ballistic Missile Agen-cy and the Cape Canaveral Air ForceBase to try to place the satellite in anorbit more steeply inclined to the equa-tor; at an inclination of about 50 degreesto the equator It would cover a muchgreater area of earth and skim the edgesof both auroral zones.

Working night and day, we set out atonce to build new apparatus of a morediscriminating nature. We retained theGeiger tube, which we had used in pre-vious satellites, as a basic "simple-mind-ed" detector. To be ready for the highestintensities of radiation, however, weused a much smaller tube that wouldyield a lower count in a given flux ofradiation, and we hooked it into a circuitthat would scale down its count by amuch larger factor. To obtain a betteridea of the penetrability of the radiation

N

Radiation Belts Around the !Mall

drawing at right shows the essential symmetry of the radiation around the earth's magneticaxis. The structure of the radiation zone was built up from hundreds of observed points.

we shielded a similar Geiger tithe with amillimeter of lead. As a more discriminat-ing particle detector we adopted a plas-tic scintillator and photomultiplier tubeto respond to electrons with an energyof more than 650,000 electron volts andto protons of more than 10 million elec-tron volts. Finally we glued a thin cesi-um-iodide crystal to the window of :m-other photomultiplier tube; the lightemitted by the crystal when it was ir-radiated would measure the over-all in-put of energy rather than the arrival ofindividual particles. To keep out lightwhen the crystal faced the sun, weshielded it with thin, opaque nickel foil.A special amplifier gave this detector alarge dynamic range extending fromabout .1 erg per second to 100,000 ergsper second.

Explorer IV carried this apparatus in-to orbit on July 26, and sent down datafor almost two months. Magnetic tapesfrom some 25 observing stations flowedin steadily from late July to late Septem-ber; altogether we obtained some 8,600recorded passes of the satellite. A typicalpass was readable for several minutes;some of the best were readable for up to20 minutes, a large fraction of the timerequired for the satellite to make a turnaround the earth. We are still analyzingthis mass of data, but the preliminaryresults have already proved to be en-lightening.

The readings have confirmed our ear-lier estimates of the maximum levels ofradiation. Moreover, we have extended

our observations to more than 50 degreesnorth and south of the equator and havebeen able to plot the intensity of theradiation at various latitudes and longi-tudes for altitudes up to 1,300 miles.The intensity contours follow the shapeof the earth in the equatorial region, butas they approach high northern andsouthern latitudes they swing outward,then inward and sharply outward againto form "horns" reaching down toward

the earth near the auroral zones [seeillustrations at the top of these twopages). The entire picture so far is com-pletely consistent with the magnetic-trapping theory.

It ii as clear from the Jntours thatExplorers I, III and IV penetrated onlythe lower portion of the radiation belt.As early as List spring we began to makehypothetical extensions of the observedcontours out to a distance of severalthousand miles. One of these speculativediagrams showed a single, doughnut-shaped belt of radiation with a ridgearound the northern and southern edgesof its inner circumference, correspond-ing to the horns of the contours. Anothershowed two beltsan outer region witha banana-shaped cross section that ex-tended from the northern to the southernauroral zone and an inner belt over theequator with a bean-shaped cross section(see illustration on pages 40 and 41].The latter diagram seemed to fit the con-tours better. In our seminars and after-hour discussions McIlwain held out forthe two-belt theory. The rest of us tend-ed to agree with him but preferred tostay with the single "doughnut" becauseof its simplicity.

To take the question out of the realmof speculation we had to secure

measurements through the entire regionof radiation. In May, therefore, I ar-ranged to have one of our radiation de-tectors carried aboard the lunar probesplanned for the fall of 1958. On October

TRAPPED PARTICLES spiral rapidly back and forth along a corkscrewshaped pathwhose center is a magnetic line of force. At the same time they driftslowly around the earth(broken arrows). Electrons (negative) and protons (positive) drift in opposite directions.

209

,..

I1, 12 and 13 Pioneer I, the first lunarprobe, can ied our instruments nearly70,000 miles out from the earth. Thoughits readings acre spotty, they confirmedour belief that the radiation extendedoutward for 111.111y thousands of miles,unit its inasuninn intensity no more than10,000 miles above the earth.

The next attempted moon shot, Pio-neer 11, was a fizzle. Pioneer 111, how-ever, went off beautifully on December6. Although this rocket was intended toreach the vicinity of the moon, we werealmost as pleased when it failed to doso, for it gave us excellent data on boththe upward and downward legs of itsflight, cutting through the radiation re-gion for 65,000 miles in two places.

The observations on both legs showeda double peak in intensity (see illustra-tion at bottom of page 42), establi-pingthat there are indeed two belts ratherthan one. The inner belt reaches itspeak at about 2,000 miles from the earth,the outer one at about 10,000 miles.Beyond 10,000 miles the radiation in.teustty diminishes steadily; it disappearsalmost completely beyond 40,000 miles.The maximum intensity of radiation ineach belt is about 25,000 counts per sec-ond, equivalent to some 40,000 parti-

210

des per square centimeter per second.Most of us believe that this great

reservoir of particles originates lagelyin the sun. The particles are somehowinjected into the earth's magnetic field,where they are deflected into corkscrewtrajectories around lines of force andtrapped. In this theoretical scheme theradiation belts resemble a sort of leakybucket, constantly refilled from the sunand draining away into the atmosphere.A particularly large influx of solar par-ticles causes the bucket to "slop over,"mainly in the auroral zone, generatingvisible auroras, magnetic storms and re-lated disturbances. The normal leakagemay be responsible for the airglow whichfaintly illuminates the night sky and mayalso account for some of the unexplainedhigh temperatures which have been ob-served in the upper atmosphere.

This solar-origin theory, while attrac-tive, presents two problems, neither ofwhich is yet solved. In the first placethe energy of many of the particles wehave observed is far greater than the pre-slimed energy of solar corpuscles. Thekinetic energy of solar corpuscles hasnot been measured directly, but thetime-tag between a solar outburst andthe consequent magnetic disturbances

....111111MMIN

HEAD OF EXPLORER IV includes nose cone (left), instrument "payload" (cent: andprotective shell aright). Payload includes four detectors, two radio transmitters, batteriesand associated electronic circuity. The outer shell is approximately six inches in diameter.

on earth indicates that the particles areslow-moving and thus of relatively lossenergy. It may be that the earth's mag-netic field traps mil, a high.energy fracLion of the panicles Alternatively, someunknou n magnetoh, drodyna rine effectof the eartlis field may accelerate thesluggish particles to higher velocities.Sonic such process in our galaxy hasbeen suggested as responsible for thegreat energies of cosmic rays. The secondproblem in the solanorigin theory is thatit is difficult to explain how chargedparticles can get into the earth's mag-netic field in the first place. We believethat neither problem is unsolvable.

Nicholas Christofilos of the Universityof California and the Soviet physicistS. N. Vemov have suggested an entirelydifferent theory of how the radiationoriginates. They note that neutrons arereleased in large numbers in the earth'supper atmosphere by the impact of cos-mic rays. These neutrons, being umcharged, can travel through the mag-netic field without deflection. In duecourse some of them decay there intoelectrons and protons, which are trapped

Our group agrees that partide.injec-Con of this sort is going on, and at a ratewhich can be easily calculated; but wefeel for a number of reasons that it can-not be the main source of radiatiobeltparticles. If we arc right 111 supposingthat the radiation belts provide the "res-ervoir" for the aurora, the neutron hy-pothesis cannot account for more thanone 10,000th of the auroral energy out-put. Even if the association betweenthe radiation belts and the aurora turnsout to be fortuitous, preliminary indica-tions both from our work and from theRussian experience with Sputnik IIIsuggest that most of the particles in theradiation belt have much lower energiesthan those of particles that would beproduced by neutron decay. A fullknowledge of the energy distribution ofthe particles will aid greatly in clarifyingtheir origin.

Neither theory explains why thereshould be two belts rather than one. It istempting to combine the two theoriesand suppose that the inner belt orig-inates with "internal injection"Le., neu-tron-decay productsand the outer onewith "external injection" of solar cor-puscles. The two-belt configuration mayof course be a transitory phenomenon,though the data from Explorer IV amtPioneer III indicate that the separatebelts persisted in essentially the sameform for at least five months. We shouldbear in mind, 'ffinvever, that 1958 wasa year of great solar activity. Three years

-'ttikvia.1

N NA IS SA A

FOURSTAGE ROCKET launched the Pioneer Ill moon probe on outbound leg gave a continuous record of radiation out to 65,000December 6, 1958. Though the flight failed to reach the moon, its miles; the inbound leg gale data between 30,000 and 10.000

from now we may well find a muchlower overall intensity and perhaps adifferent structure altogether.

In addition to these possible long-termchanges, there may be short-term flue-tnations in the belts. While we feel surethat the infhis and leakage of particlesmust balance in the long Mil, a rosiersolar outbreak may tel.:porardy increasethe intensity of the radiation many-fold.If we were to detect such fluctuationsand were to find that they coincide withsolar outbursts on the one band andwith terrestrial magnetic disturbanceson the other, we would have a plainlead to the origin of the particles. Be-fore long we hope to launch a satellite

that will monitor radiation levels forat least a year.

Our measurements show that the max-imum radiation level as of 1958 is

equivalent to between 10 and 100 roent-gens per hour. depending on the still.undetermined proportion of protons toelectrons. Since a human being exposedfor two days to even 10 roentgens wouldhave only an even chance of survival, theradiation belts obviously present an ob-stacle to space flight. Unless some prac-beat way can be found to shield space-travelers against the effects of the radio.tion. manned space rockets can best takeoff through the radiation-free 701Ie over

the poles. A "space station" must orbitbelow 400 miles or beyond 30,000 milesfrom the earth. We are now planning asatellite flight that will test the efficacyof various methods of shielding.

The hazard to spacetravelers may notend even when they have passed theterrestrial radiation belts. According topresent knowledge the other planets ofour solar system may have magneticfields comparable to the earth's and thusmay possess radiation belts of then own.The moon, honever, probably has nobelt, because its magnetic field appearsto be feeble. Lunar probes should giveus more definite information on thispoint before long.

211

'OWif' 04ee,

1111;

'MD -swill

4

Parkes radio telescope in the Goobang Valley, New South Wales, Australia.

How does the brain work? Part of the answer lies inelectrophysiology, the study of the relation betweenelectricity and nervous stimulation.

17 A Mirror for the Brain

W. Grey Walter

A chapter from his book The Living Brain published in 1963.

THE GREEKS had no word for it. To them thebrain was merely "the thing in the head," and completelynegligible. Concerned as so many of them were about man'spossession of a mind, a soul, a spiritual endowment of thegods, it is strange they did nut articipate our much less enter-prising philosophers of some score of centuries later, and in-vent at least a pocket in the head, a sensorium, to contain it.But no, the Greeks, seeking a habitation for the mind, couldfind no better place for it than the midriff, whose rhythmicmovements seemed so closely linked with what went on inthe mind -

The Hebrews also attributed special dignity to that part ofthe body; thence Jehovah plucked man's other self. Old ideasare not always as wide of the mark as they seem. The rhythmof breathing is closely related to mental states. The Greekword for diaphragm, phren, appears in such everyday wordsas frenzy' and frantic, as well as in the discredited phrenologyand the erudite schizophrenia.

213

Above the midriff the classical philosophers found thevapours of the mind; below it, the humours of the feelings.Some of these ideas persisted in physiological thought untilthe last century and survive in the common speech of today.Hysteric refers by derivation to the womb. The four basichuman temperaments were: choleric, referring to the gallbladder; phlegmatic, related to inflammation; melancholic,black bile; and sangui=e, from the blood. This classificationof temperaments was revived by a modern physiologist, Pav-lov, to systematize his observations of learning.

As in nearly all notions that survive as long as these fossilsof language have survived, there is an element of truth, ofobservation, in them. States of mind are certainly related tothe organs and liquors designated, and may even be said ina sense to originate in them. The philosopher, William James,was responsible with Lange for a complete theory of emotionwhich invoked activity` in the viscera as the essential precursorof deep feeling. Some of the most primitive and finest phrasesin English imply this dependence of sincere or deep emotionon heart or bowels. But communication of thought is so rapidthat the Greeks overlooked the existence or need of a relaystation. Arid no doubt it is for the same reason that we allseem particularly given to the same error of over-simplificationwhen we first begin, or refuse to begin, to consider how themind works. We know what makes us happy or unhappy.Who, in the throes of sea-sickness, would think of draggingin the brain to account for his melancholy state?

More curious still is Greek negligence of the brain, con-sidering their famous oracular behest, "Know thyself." Hereindeed was speculation, the demand for a mirror, insistenceupon a mirror. But for whom, for what? Was there, among

214

the mysteries behind the altar, concealed perhaps in theMinerva myth, a suspicion of something more in the headthan a thing, and that the organ which had to do the knowingof itself must be an organ of reflection?

The brain remained for more than two thousand years inthe dark after its coming of age. When it was discovered bythe anatomist, he explored it as a substance in which mightbe found the secret dwelling of intelligence; for by that timethe mind had moved from the diaphragm to the upper story,and Shakespeare had written of the brain, "which some sup-pose the soul's frail dwelling-house." Dissection was highadventure in those days. Most people believed what anironical writer today was "astonished to learn," that "it ispossible for anger, envy, hatred, malice, jealousy, fear andpride, to be confined in the same highly perishable form ofmatter with life, intelligence, honesty, charity, patience andtruth.' The search for such prize packets of evil and virtuein the brain tissue, dead or alive, could only lead to disap-pointment. The anatomist had to be satisfied with weighingthe "grey matter"about 50 ounces for man and 5 less forwomanand making sketches of the very complicated andindeed perishable organisation of nerves and cells which hisknife revealed. He could do little more. It should enlightenus at once as to the essential character of brain activity, thatthere was no possible understanding of the mechanism ofthe brain until the key to it, the electrical key, was in ourhands.

There were some flashes of foresight, sparks in the scien-tific dark, before Galvani put his hand on the key. What gen-erated all the speculations of the day was a new notion in

215

science, the conception of physical motion which began toacquire importance with Galileo and continued with Newtonand into our own times with Rutherford and Einstein. Firstamong these imaginative flashes may be mentioned the novelproposal made by the 16th Century philosopher, Hobbes,when disputing the dualist theory of Descartes. The Frenchphilosopher contemple9c1 a non-spatial mind influencing thebody through the brain, and suggested the pineal gland asthe rendezvous for mind and matter. The proposal advancedby Hobbes, in rejecting this popular theory, was that thoughtshould be regarded as being produced by bodies in motion.Hobbes was born in the year of the Spanish Armada; theRoyal Society had received its charter seventeen years beforehe died in 1679.

The controversy about the residential status of the mindis almost as much out of date as that in which the non-existence of motion seemed to be proved by the hare andtortoise fable. But the value of Hobbes' speculation was en-during; the observation and correlation of mental and phys-ical phenomena are Loday a routine of physiological research.

More specific than the speculation of Hobbes was that ofDr. David Hartley about a century later. Hartley in 1749anticipated by two hundred years the kind of theory of mentalfunction for which evidence has been found in the last yearor two. His "Observations on Man, his Frame, his Duty andhis Expectations" is a milestone in the history of Englishthought. Hartley, a contemporary of Newton and Hume, wasa pioneer of what he termed the "doctrine of mechanism."According to this, he suggested, mental phenomena are de-rived from rhythmic movements in the brainvibrations, hecalled them; upon these is superimposed a fine structure of

216

ii

,

A Mirror for the Brain

"vibratiuncles" which give thought and personality theirsubtle shades and variations. Hartley realised quite well thevalue of the plastic and compact virtues such a system mighthave. He was also the first to develop the theory of "associa-tion of ideas" in a rigorous form, relating this to his "vibra-tiuncles" in a manner which we should now consider strictlyscientific in the sense that it is susceptible to experimentaltest. It is difficult for us to appreciate the originality of hisnotions, the gist of which is now a commonplace of electro-physiology.

Hartley wrote nearly half a century before Galvani ( 17371798) and with him we might say farewell to fancy. But topass over the famous Galvani-Volta controversy with thebald statement that the one claimed to have discovered elec-tricity in animals and the other its generation by metals, wouldbe unfair to any reader who may not know how strangelytruth came out of that maze of error.

The incident began with an experiment made by Luigi andLucia Galvani in the course of their long and patient studyof that still fresh mystery, electricity. The word had been inuse since William Gilbert coined it in the 16th century fromelektron, meaning amber, another pretty semantic shift; andHenry Cavendish had already, eight years before the inci-dent, determined the identity of its dynamic laws with thoseof gravitation. Everybody in high society was familiar withthe effects of discharges from Leyden jars upon the lifelessmuscles of executed criminals; and Louis XV had, in thewords of Silvanus Thomson, "caused an electric shock from abattery of Leyden jars to be administered to 700 Carthusianmonks joined hand to hand, with prodigious effect." But inBologna in 1790 the professor of anatomy had a notion that

217

it was atmospheric electricity which acted upon the muscletissues of animals. On a stormy evening, one version of thestory goes, he and his wife had the curious idea of testingthis point by tying a dead frog to the top of the iron balustradeof the court-yard, apparently using copper wire to hold itby the leg. They expected that, as the storm approached, thefrog would be convulsed by electric shocks. And, as theywatched the thunder cloud come near, so indeed it happened;the dead frog, hanging against the iron grill, twitched in re-peated convulsions.

Further experiments convinced the Galvani that they hadwitnessed a form of electricity derived from living processes,not merely from the atmosphere. He published a famous ac-count of his experiments on the relation of animal tissue toelectricity: De viribus Electricitatis in Motu Musculari Com-mentarius (1791). Volta seized upon this to refute the wholeof Galvani's thesis, repeating his experiment not only with-out the storm but without the frog, proving that the elec-tricity in question could be generated by copper and zincsheets. This "current electricity" as it was called, was there-fore metallic, and no nonsense about any animal variety. Soended a controversy and a friendship. So began the scienceof electrical engineering.

Eppur, the Galvani might have repeated, si muove. Fortheir discredited experiment had 'truly revealed, not indeedwhat they supposed, but something more wonderful. Whathad happened was that, swaying in the wind, the suspendedfrog had come into contact with the iron bars, between whichand the copper wire a current had been generated, activatingits muscles. The Galvani had demonstrated the electricalaspect of nervous stimulation.

218

This was an event as important to the physiologist as itscounter-event was to the physicist; it was the starting-pointof that branch of the science with which we are concernedhere, electrophysiology.

Volta's counter-demonstration led directly to the inventionof the electric battery, and economic opportunity evokedelectrical engineering from the Voltaic pile. There was nosuch incentive for research when, a generation later, theexistence of animal electricity was proved. Instead, the dis-covery was exploited by the academic dilettante and thequack. The Aristotelian doctors of the period, assuming thatwhere there is electricity there is magnetism, saw in it proofalso of Mesmer's "Propositions" which had been publishedin his "Memoire sur la Decouverte du Magnitisme Animal"in 1779, floundering deeper into mystification than Dr. Mes-mer himself, who had at least declared in his "Memoire" thathe used the term analogically, and that he "made no furtheruse of electricity or the magnet from 1776 onwards.*

There is still controversy about the origin and nature ofanimal electricity. Nobody who has handled an electric eelwill question the ability of an animal to generate a formidablevoltage; and the current is demonstrably similar in effect tothat of a mineral dry cell. On the other hand, there is no evi-dence that the electric energy in nerve cells is generated byelectro-magnetic induction or by the accumulation of staticcharge. The bio-chemist iinds a complicated substance, acetyl-choline, associated with electric changes; it would be reason-able to anticipate the presence of some such substance havinga role at least as important as that of the chemicals in a Le-clanche cell.

We know that living tissue has the capacity to concentrate

219

potassium- and distinguish it from sodium, and that neuralelectricity results from the differential permeability of aninter-face, or cell-partition, to these elements, the inside ofa cell being negatively charged, the outside positively.Whether we call this a chemical or an electrical phenomenonis rather beside the point. There would be little profit in argu-ing whether a flash-lamp is an electrical or chemical device;it is more electrical than an oil lamp, more chemical than alightning flash. We shall frequently refer to changes of po-tential as electrical rhythms, cycles of polar changes, moreexplicitly electro-chemical changes. We shall be near the truthif we keep in mind that electrical changes in living tissue,the phenomena of animal electricity, are signs of chemicalevents, and that there is no way of distinguishing one fromthe other in the animal cell or in the mineral cell. The currentof a nerve impulse is a sort of electro-chemical smoke-ringabout two inches long travelling along the nerve at a speedof as much as 300 feet per second.

The neglect and mystification which obscured Calvani'sdiscovery, more sterile than any controversy, forced electro-physiology into an academic backwater for some decades. Afew experiments weie made for example, by Biedermann,who published 2-volume treatise called Electrophysiology,and by Pabois-Reymond, who introduced Michael Fara-day's induction coil into the physiological laboratory and theterm faradisation as an alternative to galvanisation into thephysiotherapist's vocabulary. Faraday's electrical and elec-trifying research began in 1831, the date also of the foun-dation of the British Association for the Advancement ofScience; but physiology long remained a backward child ofthe family.

220

A Mirror for the Brain

Hampered though these experimenters were by lack oftrustworthy equipmentthey had to construct their owngalvanometers from first principlesthey gradually accumu-lated enough facts to show that all living tissue is sensitivein some degree to electric currents and, what is perhaps moreimportant, all living tissue generates small voltages whichchange dramatically when the tissue is injured or becomesactive.

These experiments were not concerned with the brain; theywere made on frog's legs, fish eggs, electric eels and flayedvermin. Nor could the brain be explored in that way.

Following life through creatures you dissect,You lose it in the moment you detect.

It took a war to bring the opportunity of devising a tech-nique for exploring the human brainand two more wars toperfect it. Two medical officers of the Prussian army, wan-dering through the stricken field of Sedan, had the brilliant ifghoulish notion to test the effect of the Galvanic current onthe exposed brains of some of the casualties. These pioneersof 1870, Fritsch and Hitzig, found that when certain areas atthe side of the brain were stimulated by the current, move-ments took place in the opposite side of the body.

That the brain itself produces electric currents was the dis-covery of an English physician, R. Caton, in 1875.

This growing nucleus of knowledge was elaborated andcarried further by Ferrier in experiments with the "Faradiccurrent." Toward the end of the century there was a spate ofinformation which suggested that the brain of animals pos-sessed electrical properties related to those found in nerve andmuscle. Prawdwicz-Neminslci in 1913 produced what he

221

called the "electro-cerebrogram" of a dog, and was the first toattempt to classify such observations.

The electrical changes in the brain, however, are minute.The experiments of all these workers were made on the ex-posed brains of animals. There were no means of amplifica-tion in those days, whereby the impulses reaching the exteriorof the cranium could be observed or recorded, even if theirpresence had been suspected. On the other hand, the grosserelectrical currents generated by the rhythmically contractingmuscles of the heart were perceptible without amplification.Electro-cardiography became a routine clinical aid a gen-eration before the invention of the thermionic tube made itpossible to study the electrical activity of the intact humanbrain.

From an unexpected quarter, at the turn of the century,came an entirely new development. Turn up the section onthe brain in a pre-war textbook of physiology and you willfind gleanings from clinical neuro-anatomy andPavlov. Al-most as if recapitulating the history of physiological ideas,Pavlov's work began below the midriff. He found that theprocess of digestion could not be understood without refer-ence to the nervous system, and so commenced his laboriousstudy of learning in animals.

In the gospel according to Stalin, Pavlov founded notmerely a branch of physiology as Galvani had done, but awhole new scienceSoviet physiology. His work indeed wasoriginal; it owed nothing to Galvani, lying quite outside elec-trophysiology, to which it was nevertheless eventually, thoughnot in Pavlov's day, to contribute so much in theway of under-standing.

222

For nearly two generations Pavlov's experiments were themajor source of information on brain physiology. Workers inthe English laboratories had not permitted themselves to ex-plore further than the top of the spinal cord. One took ananatomical glance at the brain, and turned away in despair.This was not accountable to any peculiar weakness of physio-logical tradition but to the exigencies of scientific methoditself. A discipline had been building up through the cen-turies which demanded that in any experiment there shouldbe only one variable and its variations should be measurableagainst a controlled background. In physiology this meantthat in any experiment there should be only one thing at atime under investigationone single function, say, of anorganand that the changes of material or function shouldbe measurable. There seemed to be no possibility of isolatingone single variable, one single mode of activity, among themyriad functions of the brain. Thus there was something likea taboo against the study of the brain. The success of Pavlovin breaking this taboo early in the century was due to hiscontrivance for isolating his experimental animals from allbut two stimuli; his fame rests on his measurement of re-sponses to the stimuli.

There was no easy way through the academic undergrowthof traditional electrophysiology to the electrical mechanismsunderlying brain functions. The Cambridge school of electro-physiology, under a succession of dexterous and original ex-perimenters beginning toward the end of the last century,developed its own techniques in special fields of research, par-ticularly in the electrical signs of activity in muscles, nervesand sense organs. At the same time, the Oxford school underthe leadership of Sherrington was beginning to unravel some

of the problems of reflex function of the spinal cord. In boththese schools the procedure adopted, to comply with thetraditional requirements of scientific method, was to dissectout or isolate the organ or part of an organ to be studied. Thiswas often carried to the extreme of isolating a single nervefibre only a few thousandths of a millimetre in diameter, soas to eliminate all but a single functional unit.

Imagine, then, how refreshing and tantalizing were thereports from Pavlov's laboratory in Leningrad to those en-gaged on the meticulous dissection of invisible nerve tendrilsand the analysis of the impulses which we induced them totransmit. After four years spent working literally in a cageand chained by the anklenot for punishment but for elec-trical screeningenlargement came when my professor ofthat date, the late Sir Joseph Barcroft, assigned me to estab-lishing a laboratory in association with a visiting pupil ofPavlov, Rosenthal. We spent a year or so on mastering thetechnique and improving it by the introduction of certainelectronic devices. The Russian results were confirmed. 'todo more than this would have required staff and equipmentfar beyond the resources of the Cambridge laboratory.

Meanwhile, another major event in the history of physiologyhad taken place. Berger, in 1928, at last brought Hartley'svibrations into the laboratory and with them a method whichseemed to hold out the promise of an investigation of elec-trical brain activity as precise as were the reflex measure-ments of Pavlov. When Pavlov visited England some timeafter we heard of this, as the English exponent of his workI had the privilege of discussing it with him on familiar terms.Among other things, I asked him if he saw any relation be-tween the two methods of observing cerebral activity, his

224

A Minot for the [hate

method and Berger's. The latter, I was even then beginningto suspect, might in some way provide a clue to how the con-ditioning of a reflex was effected in the brain. But Pavlovshowed no desire to look behind the sce: les. He was not inthe least interested in the mechanism of cerebral events; theyjust happened, and it was the happening and its consequencethat interested him, not how they happened. Soviet physiologyembalmed the body of this limited doctrine as mystically asthe body of Lenin, for the foundations of their science. Theprocess of conditioning reflexes has a specious affinity withthe Marxian syllogism. Others have found in the phenomenasufficient substantiation for a gospel of Behaviourism.

Pavlov was before his time. He would have been a greaterman, his work would have been more fertile in his lifetime,and Russian science might have been spared a labyrinthinedeviation, had the work of Berger come to acknowledgementand fruition in his day. But again there was delay; Bergerwaved the fairy wand in 1928; the transformation of Cinder-ella was a process of years.

There were reasons for this delay. For one thing, Bergerwas not a physiologist and his reports were vitiated by thevagueness and variety of his claims and the desultory natureof his technique. He was indeed a surprisingly unscientificscientist, as personal acquaintance with him later confirmed.

The first occasion on which the possibilities of clinical elec-troencephalography were discussed in England was quite aninformal one. It was in the old Central Pathological Labora-tory at the Maudsley Hospital in London, in 1929. The teamthere under Professor Golla was in some difficulty aboutelectrical apparatus; they were trying to get some records of

225

the "Berger rhythm," using amplifiers with an old galvanom-eter that fused every time they switched on the current. Gollawas anxious to use the Matthews oscillograph, then the lastword in robust accuracy, to measure peripheral and centralconduction times. I was still working at Cambridge under thewatchful eye of Adrian and Matthews and was pleased tointroduce this novelty to him and at the same time, withundergraduate superiority, put him right on a few otherpoints. When, at lunch around the laboratory table, he re-ferred to the recent publication of Berger's claims, I readilydeclared that anybody could record a wobbly line, it was astring of artefacts, even if there were anything significant init there was nothing you could measure, and so on. Gollaagreed with milder scepticism, but added: "If this new ap-paratus is as good as you say, it should be easy to find outwhether Berger's rhythm is only artefact; and if it isn't, thefrequency seems remarkably constant; surely one could meas-ure that quite accurately." And he surmised that there wouldbe variations of the rhythm in disease.

Cambridge still could not accept the brain as a proper studyfor the physiologist. The wobbly line did not convince us oranybody else at that time. Berger's "elektrenkephalograms"were almost completely disregarded. His entirely original andpainstaking work received little recognition until in May,1934, Adrian and Matthews gave the first convincing demon-stration of the "Berger rhythm" to an English audience, ameeting of the Physiological Society at Cambridge.

Meanwhile, Golla was reorganising his laboratory, and hisconfidence in the possibilities of the Berger method wasgrowing. When he invited me to join his research team asphysiologist at the Central Pathological Laboratory, my first

226

task was to visit the German laboratories, including particu-larly that of Hans Berger.

Berger, in 1935, was not regarded by his associates as inthe front rank of German psychiatrists, having rather thereputation of being a crank. He seemed to me to be a modestand dignified person, full of good humour, and as unperturbedby lack of recognition as he was later by the fame it even-tually brought: him. But he had one fatal weakness: he wascompletely ignorant of the technical and physical basis ofhis method. He knew nothing about mechanics or electricity.This handicap made it impossible for him to correct seriousshortcomings in his experiments. His method was a simpleadaptation of the electrocardiographic technique by whichthe electrical impulses generated by the heart are recorded.At first he inserted silver wires under the subject's scalp; laterhe used silver foil bound to the head with a rubber bandage.Nearly always he put one electrode over the forehead andone over the back of the head; leads wcie taken from these toan Ede lmann galvanometer, a light and sensitive "string" typeof instrument, and records were taken by an assistant photog-rapher. A potential change of one-ten-thousandth of a volta very modest sensitivity by present standardscould just bedetected by this apparatus. Each record laboriously producedwas equivalent to that of two or three seconds of modem con-tinuous pen recording. The line did show a wobble at about10 cycles per second. ( See Figure 3.) He had lately acquireda tube amplifier to drive his galvanometer, and his pride andpleasure in the sweeping excursions of line obtained by itsuse were endearing.

Berger carried the matter as far as his technical handicappermitted. He had observed that the larger and more regular

227

A

n141

k440

4444

4144

4110

41V

Arle

t410

064W

WW

W41

10,4

4110

4:1M

01,0

64/4

W/4

1010

048

00#1

0111

0414

0401

1014

1014

4110

0141

1TIIV

NW

4ore

ePtw

"tri

Pipv

reN

totti

ooir

d,N

etw

ourn

"...e

pAN

*444

i1M

liviv

IrA

ION

0041

""A

sv.

:

.A44

1dt

mlii

, 1,4

4441

1100

4YIN

Piii4

4164

110i

mum

410,

4qts

iwlii

106-

/-4

Ili I

II11

&A

.Am

igA

NA

LYS

IS,

i 11,

tIt

It.r

.r,

1t P

. 1.

I Pk

Ian

tr.

k_k_

s.V

ati!

1-91

212

SI S

ISSI

I46

7%.:1

400

%itl

IAL

4161

6116

1700

it412

;450

.:i

8t

8`i

tp

'iscy

cLcs

peR

ste

ono

Figu

re 3

. '''T

he li

ne d

id s

how

a w

obbl

e at

abo

ut 1

0 cy

cles

pc: s

econ

d?' (

a) A

trac

ing

from

one

of B

erge

r's e

arlie

st r

ecor

ds. (

b) R

ecor

d fr

oma

mod

em la

bora

tory

sho

win

g co

nsis

tenc

y of

aut

o-m

atic

ana

lysi

s ov

er 1

0- s

econ

d pe

riod

s.

A Mirror for the Brain

rhythms tended to stop when the subject opened his eyes orsolved some problem in mental arithmetic. This was con-firmed by Adrian and Matthews with leads from electrodeson Adrian's hea, tached to a Matthews amplifier and ink-writing oscillograph. This superior apparatus, and a morecareful location of electrodes, enabled them to go a step fur-ther and prove that the 10 cycles per second rhythm arises inthe visual association areas in the occiput and not, as Bergersupposed, from the whole brain.

Only some years later was it realised what an importantstep this was. Its significance could not be recognised whileso little was known about the components of the "wobblyline," the electroencephalogram or, abbreviated, EEG. Un-avoidably at the time, the significance of the salient characterof the normal EEG was overlooked; it was found, in Adrian'sphrase, "disappointingly constant" The attention of manyearly workers in electroencephalography therefore turnedfrom normal research to the study of nervous disease. In im-mediate rewards this has always been a rich field. In this in-stance, a surprising state was soon reached wherein whatmight be called the electropathology of the brain was furtheradvanced than its eleitrophysiology.

In the pathological laboratory, Golla's earlier surmise, thatthere would be variations of the rhythmic oscillation indisease, was soon verified. A technique was developed thereby which the central point of the disturbance in the tissuecould be accurately determined. For surgery, the immediateresult of perfecting this technique was important; it madepossible the location of tumours, brain injuries, or other pnys-ical damage to the brain. It was helpful in many head casesduring the war as well as in daily surgical practice.

229

occupy the attention of many EEG workers. The difficultiesThe study of epilepsy and mental disorders also began to

encountered in these subjects threw into prominent reliefthe essential complexity of the problem as compared withthose of classical physiology. The hope of isolating single func-tions had now been abandoned; those who entered this fieldwere committed to studying the brain as a whole organ andthrough it the body as a whole organism. They were thereforeforced to multiply their sources of information.

It is now the general EEG practice, not only for clinicalpurposes, but in research, to use a number of electrodes si-multaneously, indeed as many as possible and convenient.The standard make of EEG recorder has eight channels.Eight pens are simultaneously tracing lines in which therecordist, after long expel ience, can recognise the main com-ponents of a complex graph. The graphs can also be auto-matically analysed into their component frequencies. A moresatisfactory method of watching the electrical changes in allthe main areas, as in a moving picture, a much more informa-tive convention than the drawing of lines, has been devisedat the Burden Neurological Institute. This will be describedafter a simple explanation of what is meant by the rhythmiccomposition of the normal EEG; for its nature, rather than themethods of recording and analysing it, is of first importancefor understanding what follows.

If you move a pencil amply but regularly up and down ona paper that is being drawn steadily from right to left, theresult will be a regular series of curves. If at the same timethe paper is moving up and down, another series of curveswill be added to the line drawn. If the table is shaking, thevibration will be added to the line as a ripple. There will then

230

be three components integrated in the one wavy line, whichwill begin to look something like an EEG record. The linegives a coded or conventional record of the various fre-quencies and amplitudes of different physical movements. Insimilar coded or integrated fashion the EEG line reports thefrequencies and amplitudes of the electrical changes in thedifferent parts of the brain tapped by the electrodes on thescalp, their minute currents being relayed by an amplifierto the oscillograph which activates the pens.

All EEG records contain many more components than this;some may show as many as 20 or 30 at a time in significantsizes. Actually there may be tens of thousands of impulseswoven together in such a manner that only the grosser com-binations are discernible.

A compound curve is of course more easily put togetherthan taken apart. (See Figure 4.) The adequate analysis of afew inches of EEG records would require the painstaking com-putation of a mathematicianit might take him a week or so.The modern automatic analyser in use in most laboratorieswrites out the values of 24 components every 10 seconds, aswell as any averaging needed over longer periods.

The electrical changes which give rise to the alternatingcurrents of variable frequency and amplitude thus recordedarise in the cells of the brain itself; there is no question ofany other power supply. The brain must be pictured as avast aggregation of electrical cells, numerous as the stars ofthe Galaxy, some 10 thousand million of them, through whichsurge the restless tides of our electrical being relatively thou-sands of times more potent than the force of gravity. It iswhen a million or so of these cells repeatedly fire together

231

i

,.

232

that the rhythm of their discharge becomes measureable infrequency and amplitude.

What makes these million cells act togetheror indeedwhat causes a single cell to dischargeis not known. We arestill a long way from any explanation of these basic mechanicsof the brain. Future research may well early us, as it has car-

A

10 flI 9 0

e.

Figure 4. "A compound curve is more easily put together thantaken apart." (a) A compound curve in which the three componentscan be detected by visual inspection, ratios 1:2 and 2:3. (b) The threecomponents (ratios 8:9, 9:10) of this compound curve cannot bedetermined at sight. The bottom line shows their frevencies auto-matically recorded every 10 seconds. Note the accidental similarity ofthis curve to the EEC record of alpha rhythms in Figure 3 (b).

A Mirror for the Brain

ried the physicist in his attempt to understand the compo-sition of our atomic being, into vistas of ever increasingenchantment but describable only in the convention of mathe-matical language. Today, as we travel from one fresh vistato another, the propriety of the language we use, the con-vention we adopt, becomes increasingly important. Arith-metic is an adequate language for describing the height andtime of the tides, but if we want to predict their rise and fallwe have to use a different language, an algebra, with its spe-cial notation and theorems. In similar fashion, the electricalwaves and tides in the brain can be described adequately bycounting, by arithmetic; but there are many unknown quan-tities when we come to the more ambitious purposes of under-standing and predicting brain behaviourmany x's and y's;so it will have to have its algebra. The word is forbiddingto some people; but, after all, it means no more than "theputting together of broken pieces."

EEC records may be considered, then, as the bits andpieces of a mirror for the brain, itself speculum speculorum.They must be carefully sorted before even trying to fit themtogether with bits from other sources. Their informationcomes as a conventional message, coded. You may crack thecode, but that does not imply that the information will neces-sarily be of high significance. Supposing, for instance, youpick up a coded message which you think may be about a mo-mentous political secret. In the first stage of decoding it youmight ascertain that the order of frequency of the letters wasETAONI. This does not sound very useful information; butreference to the letter-frequency tables would assure you atleast that it was a message in English and possibly intelligible.Likewise, we watch the frequencies as well as the amplitudeand origin of the brain rhythms, knowing that many earnestseekers for the truth have spent lifetimes trying to decipher

233

what they thought were real messages, only to find that theirhoroscopes and alembics contained gibberish. The scientistis used to such hazards of research; it is only the ignorantand superstitious who regard him, or think he regards him-self, as a magician or priest who is right about everything allthe time.

Brain research has just about reached the stage where theletter frequencies of the code indicate intelligibility and theirgrouping significance. But there is this complication. -beordinary coded message is a sequence in time; events in thebrain are not a single sequence in timethey occur in three-dimensional space, in that one bit of space which is morecrowded with events than any other we can conceive. Wemay tap a greater number of sectors of the brain and set morepens scribbling; but the effect of this will only be to multiplythe number of code signals, to the increasing embarrassmentof the observer, unless the order and inter-relation of thesignals can be clarified and emphasised. Redundancy is al-ready a serious problem of the laboratory.

The function of a nervous system is to receive, correlate,store and generate many signals. A human brain is a mecha-nism not only far more intricate than any other but one thathas a long individual history. To study such a problem interms of frequency and amplitude as a limited function oftimein wavy linesis at the best over-simplification. And

'the redundancy is indeed enormous. Information at the rateof about 3,600 amplitudes per minute may be coming througheach of the eight channels during the average recording pe-riod of 20 minutes; so the total information in a routine recordmay be represented by more than half a million numbers; yetthe usual description of a record consists only of a few sen-tences. Only rarely does an observer use more than one-hundredth of one per cent of the available information.

234

A Mirror for tne Brain

"What's in a brain that ink may character . . . ?"For combining greater clarity with greater economy, many

elaboratons of methods have been adopted in clinic andlaboratory. They still do not overcome the fundamental em-barrassment of redundancy and the error of over-simplifica-tion, both due to the limitations of a time scale. A promisingalternative is a machine that draws a snapshot map insteadof a long history, projecting the electrical data visually on aspatial co-ordinate system which can be laid out so as to repre-sent a simple map or model of the head. This moving pano-rama of the brain rhythms does approximate to Sherrington's"enchanted loom where millions of flashing shuttles weave adissolving pattern, always a meaningful pattern though neveran abiding one." (Figure 5.)

We have called the apparatus which achieves this sort ofeffect at the Burden Institute a toposcope, by reason of itsdisplay of topographic detail. The equipment was developedby Harold Shipton, whose imaginative engineering trans-formed the early models from entertainment to education.Two of its 24 channels are for monitoring the stimuli; theothers, instead of being connected with pens, lead the elec-trical activity of the brain tapped by the electrodes for displayon the eens of small cathode-ray tubes. So instead of wavylines on a moving paper, the observer sees, to quote Sherring-ton again, "a sparkling field of rhythmic flashing points withtrains of travelling sparks hurrying hither and thither." As-sembled in the display console, 22 of the tubes give a kindof Mercator's projection of the brain. Frequency, phase andtime relations of the rhythms are shown in what at first ap-pears to be a completely bewildering variety of patterns ineach tube and in their ensemble. Then, as the practised eyegains familiarity with the scene, many details of brain ac-tivity are seen for the first time. A conventional pen machine

235

is simultaneously at the disposal of the observer, synchronisedso that, by turning a switch, a written record of the activityseen in any five of the tubes can be made. Another attachmentis a camera with which at the same time permanent snapshotrecordi of the display can be obtained. (Figure 6.)

Thus, from Berger's crude galvanometer to this elaborateapparatus requiring a whole room of its own, electroenceph-alography has progressed from a technique to a science. Itsclinical benefits, Cy-products of free research, are acknowl-edged; they can ly...1 gauged by the vast multiplication of EEGlaboratories. From Berger's lone clinic have sprung severalhundred EEG centresmore than 50 in England alone. Lit-erally millions of yards of paper have been covered with fran-tic scribblings. In every civilized country there is a speciallearned society devoted to the discussion of the records andto disputation on technique and theory. These societies arebanded together in an International Federation, which pub-lishes a quarterly Journal and organises international con-gresses.

For a science born, as it were, bastard and neglected ininfancy, this is a long way to have travelled in its first quarterof a century. If it is to provide the mirror which the brainrequires to see itself steadily and whole, there is still a longroad ahead. The following chapters give the prospect as seenfrom the present milestone, assuming that such studies areallowed to continue. Looking back, we realise that the presentscale of work as compared with previous physiological re-search is elaborate and expensive. But our annual cost of con-ducting planned investigations of a fundamental nature intoman's supreme faculties is less than half that of one mediumtank, and the money spent on brain research in all Englandis barely, one-tenth of one per cent of the cost of the nationalmental health services.

236

A Mirror for the Brain

-__±...411111111

Figure 5. ". . a moving panorama of the brain rhythms." The Toposcope Laboratory. The subject's couch andtriggered stroboscope (flicker) reflector at extreme lest beond desk of 6-channel pen recorder with remote controlpanel. The 22-channel toposcope amplifier is in the background, the display panel at right centre, camera and pro.lector at extreme right.

Figure 6. "... always a meaningful pattern though never an abiding one."Snapshots of the "sparkling field of rhythmic flashing points." Each of the tubescreens. Much form a chart of the head seen from ribose with nose at top.shows by the flashing sectors of its disc the activity of the corresponding areaof the brain. (Top Left) Resting alpha rhythms. (Top right) Theta rhythms inanger. (Bottom left) Wide response to double flashes of light. (Bottom right)Spread of response to triple flashes.

237

Scientific Imagination

Richard P. Feynman, Robert B. Leighton, and Matthew Sands

Excerpt from The Feynman Lectures on Physics, Volume II, 1964.

I have asked you to imagine these electric and magnetic fields. What do youdo? Do you know how? How do I imagine the electric and magnetic field? Whatdo I actually see? What are the demands of scientific imagination? Is it anydifferent from trying to imagine that the room is full of invisible angels? No, it isnot like imagining invisible angels. It requires a much higher degree of imaginationto understand the electromagnetic field than to understand invisible angels. Why?Because to make invisible angels understandable, all I have to do is to alter theirproperties a little bitI make them slightly visible, and then I can see the shapesof their wings, and bodies, and halos. Once I succeed in imagining a visible angel,the abstraction requiredwhich is to take almost invisible angels and imaginethem completely invisibleis relatively easy. So you say, "Professor, please giveme an approximate description of the electromagnetic waves, even though it maybe slightly inaccurate, so that I too can see them as well as I can see almost invisibleangels. Then I will modify the picture to the necessary abstraction."

I'm sorry I can't do that for you. I don't know how. I have no picture of thiselectromagnetic field that is in any sense accurate. I have known about the electro-magnetic field a long timeI was in :he same position 25 years ago that you arenow, and I have had 25 years more of experience thinking about these wigglingwaves. When I start describing the magnetic field moving through space, I speakof the E- and B fields and wave my arms and you may imagine that I can see them.

239

I'll tell you what I see. I see some kind of vague shadowy, wiggling lineshereand there is an E and B written on them somehow, and perhaps some of the lineshave arrows on theman arrow here or there which disappears when I look tooclosely at it. When I talk about the fields swishing through space, I have a terribleconfusion between the symbols I use to describe the objects and the objects them-selves. I cannot really make a picture that is even nearly like the true waves. Soif you have some difficulty in making such a picture, you should not be worriedthat your difficulty is unusual.

Our science makes terrific demands on the imagination. The degree ofimagination that is required is much more extreme than that required for some ofthe ancient ideas. The modern ideas are much harder to imagine. We use a lotof tools, though. We use mathematical equations and rules, and make a lot ofpictures. What I realize now is that when I talk about the electromagnetic field inspace, I see some kind of a superposition of all of the diagrams which I've everseen drawn about them. I don't see little bundles of field lines running about be-cause it worries me that if I ran at a different speed the bundles would disappear.I don't even always see the electric and magnetic fields because sometimes I thinkI should have made a picture with the vector potential and the scalar potential,for those were perhaps the more physically significant things that were wiggling.

Perhaps the only hope, yuu say, is to take a mathematical view. Now what isa mathematical view? From a mathematical view, there is an electric field vectorand a magnetic field vector at every point in space; that is, there are six numbersassociated with every point. Can you imagine six numbers associated with eachpoint in space? That's too hard. Can you imagine even one number associatedwith every point? I cannot! I can imagine such a thing as the temperature at everypoint in space. That seems to be understandable. There is a hotness and coldnessthat varies from place to place. But I honestly do not understand the idea of anumber at every point.

So perhaps we should put the question: Can we represent the electric field bysomething more like a temperature, say like the displacement of a piece of jello?Suppose that we were to begin by imagining that the world was filled with thinjello and that the fields represented some distortionsay a stretching or twistingof the jello. Then we could visualize the field. After we "see" what it is like wecould abstract the jello away. For many years that's what people tried to do.Maxwell, Ampere, Faraday, and others tried to understand electromagnetismthis way. (Sometimes they called the abstract jello "ether.") But it turned out thatthe attempt to imagine the electromagnetic field in that way was really standing inthe way of progress. We are unfortunately limited to abstractions, to using in-struments to detect the field, to using mathematical symbols to describe the field,etc. But nevertheless, in some sense the fields are real, because after we are allfinished fiddling around with mathematical equationswith or without makingpictures and drawings or trying to visualize the thingwe can still make the instru-ments detect the signals from Mariner II and find out about galaxies a billion milesaway, and so on.

240

Scientaw, Imagination

The whole question of imagination in science is often misunderstood by peoplein other disciplines. They try to test our imagination in the following way. Theysay, "Here is a picture of some people in a situation. What do you imagine willhappen next?" When we say, "I can't imagine," they may think we have a weakimagination. They overlook the fact that whatever we are allowed to imagine inscience must be consistent with everything else we know: that the electric fields andthe waves we talk about are not just some happy thoughts which we are free tomake as we wish, but ideas which must be consistent with all the laws of physicswe know. We can't allow ourselves to seriously imagine things which are obviouslyin contradiction to the known laws of nature. And so our kind of imagination isquite a difficult game. One has to have the imagination to think of something thathas never been seen before, never been heard of before. At the same time thethoughts are restricted in a strait jacket, so to speak, limited by the conditions thatcome from our knowledge of the way nature really is. The problem of creatingsomething which is new, but which is consistent with everything which has beenseen before, is one of extreme difficulty.

While I'm on this subject I want to talk about whether it will ever be possibleto imagine beauty that we can't see. It is an interesting question. When we lookat a rainbow, it looks beautiful to us. Everybody says, "Ooh, a rainbow." (Yousee how scientific I am. I am afraid to say something is beautiful unless I have anexperimental way of defining it.) But how would we describe a rainbow if we wereblind? We are blind when we measure the infrared reflection coefficient of sodiumchloride, or when we talk about the frequency of the waves that are coming fromso:,ie galaxy that we can't seewe make a diagram, we make a plot. For instance,for the rainbow, such a plot would be the intensity of radiation vs. wavelengthmeasured with a spectrophotometer for each direction in the sky. Generally, suchmeasurements would give a curve that was rather flat. Then some day, someonewould discover that for certain conditions of the weather, and at certain angles inthe sky, the spectrum of intensity as a function of wavelength would behavestrangely; it would have a bump. As the angle of the instrument was varied only alittle bit, the maximum of the bump would move from one wavelength to another.Then one day the physical review of the blind men might publish a technical articlewith the title "The Intensity of Radiation as a Function of Angle under CertainConditions of the Weather." In this article there might appear a graph such asthe one in Fig. 20-5. The author would perhaps remark that at the larger anglesthere was more radiation at long wavelengths, whereas for the smaller angles themaximum in the radiation came at shorter wavelengths. (From our point of view,we would say that the light at 40° is predominantly green and the light at 42° ispredominantly red.)

241

242

Wavelength

Fig. 20-5. The intensity of electro-magnetic waves as a function of wave-length for three angles (measured fromthe direction opposite the sun), observedonly with certain meteorological con-ditions.

Now do we find the graph of Fig. 2(-3 ,eautiful? It contains much more de-tail than we apprehend when we look at a rainbow, because our eyes cannot seethe exact details in the shape of a spectrum. The eye, however, finds the rainbowbeautiful. Do we have enough imagination to see in the spectral curves the samebeauty we see when we look directly at the rainbow? I don't know.. But suppose I have a graph of the reflection coefficient of a sodium chloride

crystal as a function of wavelength in the infrared, and also as a function of angle.I would have a representation of how it would look to my eyes if they could seein the infraredperhaps some glowing, shiny "green," mixed with reflections fromthe surface in a "metallic red." That would be a beautiful thing, but I don't knowwhether I can ever look at a graph of the reflection coefficient of NaCI measuredwith some instrvnent and say that it has the same beauty.

On the other hand, even if we cannot see beauty in particular measured results,we can already claim to see a certain beauty in the equations which describe generalphysical laws. For example, in the wave equation (20.9), there's something niceabout the regularity of the appearance of the x, the y, the z, and the 1. And thisnice symmetry in appearance ,f the x, y, z, and t suggests to the mind still a greaterbeauty which has to do with the four dimensions, the possibility that space hasfour-dimensional symmetry, the possibility of analyzing that and the developmentsof the special theory of relativity. So there is plenty of intellectual beauty asso-ciated with the equations.

NORMAN LEADEF ALLEN

Norman Leader Allen, British physicist, was born in1927 and received his B.Sc. from the University ofBirmingham, England, in 1948 and his Ph.D. in 1951.Allen has been a staff member of Mossochusetts ofTechnology and is now a lecturer in the Electrical andElectronic Engineering Department at the University ofLeeds. In addition to his book, Threshold Pressure forArc Discharges, he has written extensively in scientificjournals on arc discharges, cosmic rays and plasmaphysics.

STANLEY SUMNER BALLARD

Stanley S. Ballard, Professor of Physics and chairmanof the department at the University of Florida, Gaines-ville, was born in Los Angeles in 1908. He receivedhis A.B. from Pomona Colleye, and M.A. and Ph.D.from the University of California. He has taught at theUniversity of Hawaii, Tufts University, and has been aresearch physicist at the Scripps Institution of Oceanog-raphy. Ballard is a member of the Physical Science Di-vision, National Research Council, and belongs toseveral optical societies in this country and Europe,having served as president of the Optical Society ofAmerico. His specialities are spectroscopy, opticaland infrared instrumentation, and properties of renal!materials. Ballord is coauthor of Physi

ARTHUR C. CLARKE

Arthur C. Clarke, British scientist and writer, is aFellow of the Royal Astronomicol Society. DuringWorld War II he served as technicol officer in chargeof the first oircraft ground-controlled approoch project.He has won the Kai ingo Prize, given by UNESCO farthe popularization of science. The feasibility of monyof the current space developments was perceived andoutlined by Clarke in the 1930's. His science fictionnovels include Childhoods End and The City ond theStors.

ALBERT EINSTEIN

Albert Einstein, considered to be the most creatiephysical scientist since Newton, was nevertheless ahumble and sometimes rather shy man. He wos born inUlm, Germony, in 1879. He seemed to learn so slowlythat his parents feared that he might be retarded. Aftergraduating from the Polytechnic Institute in Zurich, hebecame a junior official at the Patent Office at Berne.At the oge of twenty-six, and quite unknown, he pub-

At viol. -inif '.'Vnter

lished three revolutionary papers in theoretical physicsin 1905. The first paper extended Max Planck's ideasof quantization of energy, and established the quantumtheory of radiation. For this work he received the NobelPrize for 1929. The second paper gave a mathematicaltheory of Brownian motion, yielding a calculation ofthe size of a molecule. His third paper founded thespecial theory of relativity. Einstein's later work cen-tered on the general theory of relativity. His work hada profound influence not only on physics, but also onphilosophy. An eloquent and widely beloved man,Einstein took an active part in liberal and anti-wcrmovements. Fleeing from Nazi Germany, he settledin the United States in 1933 at the Institute for Ad-vanced Study in Princeton. He died in 1955.

RICHARD PHILLIPS FEYNMAN

Richard Feynman was born in New York in 1918, andgraduated from the Mossachusetts Institute of Technologyin 1939. He received his doctorate in theoretical phys-ics from Princeton in 1942, and worked ot Los Alamosduring the Second World War. From 1945 to 1951 hetaught at Cornell, and since 1951 has been TolmanProfessor of Physics at the California Institute of Tech-nology. Professor Feynmun received the Albert EinsteinAward in 1954, and in 1965 was named a Foreign Mem-ber of the Royal Society. In 1966 he was awarded theNobel Prize in Physics, which he shared with ShinicheroTomonaga and Julian Schwinger, for work in quantumfield theory.

LEOPOLD INFELD

Leopold Infeld, a co-worker with Albert Einstein ingeneral relativity theory, was bo.n in 1898 in Poland.After studying at the Cracow ond Berlin Universities,he became o Rockefeller Fellow at Combridge wherehe worked with Max Barn in electromagnetic theory,and then o member of the Institute for Advanced Studyat Princeton. For eleven years he wos Professor ofApplied Mathematics at the University of Toronto.He then returned to Poland ond became Professor ofPhysics ot the University of Warsaw and until his deathon 16 January 1968 he was director of the TheoreticalPhysics Institute at the university. A member of thepresidium of the Polish Academy of Science, Infeldconducted research in theoreticol physics, especiallyrelotivity and quantum theories. Infeld was the authorof The New Field Theory, The World in ModernScience, Quest, Albert Einstein, and with EinsteinThe Evolution of Physics.

243

THOMAS JEFFERSON

Thomas Jefferson, third President of the United States,was born in 1743 at Shadwell in Gooch land County,Virginia. He studied Greek, Latin, and mathematicsat the College of William and Mary far two years, andlater became a lawyer. From 1768 to 1775 Jeffersonwas a member of the Virginia House of Burgesses. In1775 he was elected to the Second Continental Congress,and in 1776 he drafted the Declaration of Independence.Jefferson felt a conflicting devotion to the tranquil pur-suits of science and public service. His interests rargedover such fields as agriculture, meteorology, paleontol-ogy, ethnology, botany, and medicine. He believed inthe freedom of the scientific mind and the importance ofbasing conclusions on observations and experiment.Jefferson demanded utility of science, hence his num-erous inventions and interest in improvements and simpli-fications of ogricultural tools and techniques, and inballoons, dry docks, submarines, even the furniture inhis home (swivel chairs and music stands). Because of hisprominence as a public figure, he was influential in in-creasing and improving science education in Americo.He died on July 4, 1826, the fiftieth anniversary of theDeclarotion of Independence.

MATTHEW JOSEPHSON

Matthew Josephson, prolific writer and magazine editor,was barn in Brooklyn in 1899. He received his B.A.from Columbia University in 1920. Josephson was suc-cessively editor of the Broom, Transition, and The NewRepublic, which he left in 32. In 1948 he wr;Tiegedto the National Institute of Arts and Letters and also wasa traveling Guggenheim fellow for creotive literature.He is the author of Zola and His Time, The Robber Barons,and Portrait of the Artist As American.

ROBERT B. LEIGHTON

Robert B. leiihton, barn in Detroit, Michigan in 1919,wcs first a student and then a faculty member at Cali-fornia Institute of Technology. He is a member of theInternational Astronomical Union, the National Acad-emy of Science and the American Physics Society. Pro-fessor Leighton's work deals with the theory of solids,cosmic rays, high energy physics, and solar physics.

DAVID KEITH CHALMERS MACDONALD

David Keith Chalmers MacDonald was barn in Glasgow,Scotland, in 1920 and received his M.A. in mathe-

244

matics and natural philosophy from Edinburgh Univer-sity in 1941. After serving with the Royal Mechanicaland Electrical Engineers during World War II, he re-ceived his Ph.D. in 1946 from Edinburgh. Then heattended Oxford as a research fellow and received aPh.D. in 1949. In 1951 Dr. MacDonald went to Can-ada and started a low temperature physics researchlaboratory for the National Research Council..MacDonald was appointed to the physics department atOttawa University in 1955 and elected Fellow of theRoyal Society of London in 1960. Aside from numerousarticles in scientific journals, he was the author ofNear Zero: An Introduction to Law TemperaturePhysics and Faraday, Maxwell, and Kelvin. MacDonalddied in 1963.

JAMES CLER rs MAXWELL

See J. R. Newman's articles in Readers 3 and 4.

ALBERT ABRAHAM MICHELSON

Precision measurement in experimental physics was thelifelong passion of A. A. Michelson (1852-1931), whobecame in 1907 the first American to win a NobelPrize in one of the sciences. Born in Prussia but raisedin California and Nevada, Michelson attended theU. S. Naval Academy and was teaching there in 1879when he first improved the meth( Is of measuring thevelocity of light on earth. Atter a post-graduate educa-tion in Europe under several great professors of optics,Michelson returned to the United States where he taughtphysics at the college that became Case Institute ofTechnology, then at Clark University, and at theUniversity of Chicago. While in Europe he invented thefamous instrument called the Michelson interferometerand while in Cleveland at Case in 1887, he and E. W.Morley improved this device in on effort to measure theabsolute velocity of the Earth as it hurtles through space.The failure of the Michelson-Morley nether -drift experi-ment was an important part of the background leading toAlbert Einstein's first work on the theory of relativity.Although Michelson remained a creative experimentalistin physical optics, meteorology, astrophysics and spectro-scopy throughout his life, he died still believingexclusively in the wave model of the nature of light andin his "beloved nether." His figures for the velocity oflight, refined still further just before his death, remainthe accepted value of one cf the few "absolute" con-stants in physics today.

71,

JAMES ROY NEWMAN

James R. Newman, lawyer and mathematician, was bornin New York City in 1907. He received his A.B. fromthe College of the City of New York and LL.B. fromColumbia. Admitted to the New York bor in 1929, hepracticed there for twelve years. During World War 11he served as chief intelligence officer, U. S. Embossy,London, and in 1945 as special assistant to the SenateCommittee on Atomic Energy. From 1956-57 he wassenior editor of The New Republic, and since 1948 hodbeen a member of the board of editors for ScientificAmerican where he was responsible for the book reviewsection. At the some time he was a visiting lecturer atthe Yale Low School. J. R. Newman is the author ofWhat is Science?, Science and Sensibility, and editorof Common sense of the Exact Sciences, The World ofMathematics, and the Harper Encyclopedia of Science.He died in 1966.

MATTHEW SANDS

Matthew Sands was born in Oxford, Massachusetts, in1919 He attended Clark College, Rice Institute, andMassachusetts Institute of Technology. During WorldWar 11 he worked at the Los Alamos Scientific Laboratory.He was Professor of Physics at the California Institute ofTechnology before joining the linear accelerator groupat Stanford University. Professor Sands specializes inelectronic instrumentation for nuclear physics, cosmicrays, and high-energy physics. He served as choirmonof the Commission on College Physics.

WILLIAM ASAHEL SHURCLIFF

Born in Boston in 1909, William A. Shurcliff was educatedat Harvard, receiving his Ph.D. in physics in 1934.During the war he served as technical aide to the Officeof Scientific Research and Development, NotionalDefense Research Committee, and Manhattan project.Then he was with the Polaroid Corporation as seniorscientist and project leader. He is now a ResearchFellow at the Electron Accelerator at Harvard. Shurcliffis the author of Polarized Light: Production and Use andBombs ct Bikini. His technical interests include emissionspectroscopy, absorption spectrophotometry, atomicenergy, gamma radiation dosimeters, microscope design,and color vision.

JAMES ALFRED VAN ALLEN

James Alfred Von Allen, discoverer of the "Von Allenradiation belt," was born at Mt, Pleasant, Iowa, in

At ?1st:. anti 'ANA

1914. After his undergraduate work at lowo WesleyanCollege, he received his M.S. and in 1939 his Ph.D.from the State University of Iowa, where he is now aProfessor of Physics and Astronomy. He hos been aCornegie research fellow at the Carnegie Institution,Guggenheim fellow, and o reseorch associate at Princeton,and is the recipient of numerous honorary doctorotes.For his distinguished work in nuclear physics, cosmic raysand space probes, he hos been awarded the HickmanMedal from the American Rocket Society, the Distin-guished Civilian Service Medal of the U.S. Army, andthe Hill Award of the Institute of Aerospace Science.

EDGAR VILLCHUR

Edgar Villchur is President and Director of Research ofthe Foundation for Hearing Aid Research in Woodstock,New York. He was born in New York City in 1917 andreceived a M.S.Ed. from the City College of New York.He has taught at New York University, and was Presidentand Chief Designer of Acoustic Research, Inc., a manu-facturing company in the high fidelity field.

WILLIAM GREY WALTER

William Grey Wolter wos born in 1911 and received hisM.A. and Sc.D. (1947) from Cambridge University. Hewas a Rockefeller Fellow at the Moudsley Hospital inEngland. W. Grey Wolter is a pioneer in the use ofelectroencephologrophy for translating the minute elec-trical currents of the human brain into physical patternswhich may be studied for the information they give us onbrain processes. Wolter is the author of The Living Brain,Further Outlook, The Curve of the Snowflake and articlesto various scientific journals.

THOMAS YOUNG

Thomas Young, the versatile English physician and physi-cist, wos born in Milverton, Somersetshire, England, in1773. He become Professor of Natural Philosophy of theRoyal Institution in 1801 and was foreign secretory of theRoyal Society from 1802 to 1829. His most importantcontribution to science was his discovery of the interfer-ence of light and the measurements of the wavelengthsof various colors. Others included the definition ofenergy, study of sound and elasticity (Young's Modulus),discovery of ostigmotism (in himself), explanation ofaccommodation, and initiation of a color vision theorylater developed by Helmholtz. His interest in linguisticsled him to participate in the deciphering of Egyptianhieroglyphics. Thomas Young died in 1829.

245/246

1. LETTER FROM THOMAS JEFFERSON, JUNE,1799, from Scripts Mathematica, Volume I, 1932,pages 88-90. Permission to publish by TeachersCollege Library, Columbia University. Also inScience and the Common Understanding, by .1.Robert Oppenheimer, I953. Reprinted by per-mission of Simon and Schuster, Inc.

2. ON THE METHOD OF THEORETICAL PHYSICS,by Albert Einstein, from Essays in Science, pages12-21, Philosophical Library, New York, Copy-right (c) 1934. Reprinted with permission.

3. EXPERIMENTS AND CALCULATIONS RELATIVETO PHYSICAL OPTICS, by Thomas Young, fromMiscellaneous Works of Thomas Young, Volume 1,edited by George Peocock, John Murray, London,1855.

4. VELOCITY OF LIGHT, by A. A. Michelson, fromStudies in Optics, The University of Chicago Press,1927. Reprinted with permission of the Universityof Chicago.

5. POPULAR APPLICATIONS OF POLARIZEDLIGHT, by William A. Shurcliff and Ston ley S.Ballord, from Polarized Light (Momentum 07),Copyright 1964, D. Van Nostrond Company, Inc.,Princeton, New Jersey. Reprinted with permis-sion.

6. ACTION AT A DISTANCE, by James ClerkMaxwell, from The Scientific Papers of JamesClerk Maxwell, Volume I', edited by D. W.Niven, Cambridge University Press, Cambridge,England, 1890. Reprinted with permission.

7. THE ELECTRONIC REVOLUTION, by Arthur C.Clarke, from Voices From The Sky, pages 204-210,, Harper and Row, Publishers, New York,1965. "The Electronic Revolution" copyright (c)1962 by The New York Times Company. Re-printed by permission of the author and his agents:Scott Meredith Literary Agency, Inc., 580 FifthAvenue, New York 10036.

8. THE INVENTION OF THE ELECTRIC LIGHT,by Matthew Josephson. Reprinted with permis-sion. Copyright (c) 1959 by Scientific American,Inc. All rights reserved.

9. HIGH FIDELITY, by Edgar Villchur, from Repro-duction of Sound, Chapters 2 and 3, AR LibraryVolume 2; Copyright 1962 by Acoustic Research,Inc., Cambridge, Massachusetts. Copyright 1965by Dover Publications, Inc., New York, NewYork. Reprinted with permission of the publisher.

247

10. THE FUTURE OF DIRECT CURRENT POWERTRANSMISSION, by N. L. Allen, fromEndeavour, Volume XXVI, No. 97, January 1967.Imperial Chemical Industries Limited, London.Reprinted with permission.

11. JAMES CLLRK MAXWELL, by James R. Newman,Part II. Reprinted with permission. Copyright1955 by Scientific American, Inc. All rights re-served.

12. MAXWELL'S LETTERS: A COLLECTION

13., ON THE INDUCTION OF ELECTRIC CURRENTS,by James Clerk Maxwell, from A Treatise on Elec-tricity and Magnetism, Volume 2, 1873, TheClarendon Press, Oxford. Reprinted with permis-sion.

14. THE RELATIONSHIP OF ELECTRICITY ANDMAGNETISM, by D. K. C. MacDonald, fromFaraday, Maxwell, and Kelvin. Copyright1964 by Educational Services Incorporated. Re-printed by permission of Doubleday & Company,Inc. (Science Study Series.)

15, THE ELECTROMAGNETIC FIELD, by AlbertEinstein and Leopold Infeld, from The Evolutionof Physics, Simon and Schuster, New York, 1961.Reprinted with permission.

16. RADIATION BELTS AROUND THE EARTH, byJames A. Van Allen. Reprinted with permission.Copyright (2) 1959 by Scientific American, Inc.All rights reserved. Available separately at 20ceach as Offprint No. 248 from W. H. Freemanand Company, 660 Market Street, San Francisco,California.

17. A MIRROR FOR THE BRAIN, by W. Grey Walter,from The Living Brain, W. W. Norton and Com-pany, Inc., New York, 1963. Reprinted withpermission.

18. SCIENTIFIC IMAGINATION, by Richard P.Feynman, Robert B. Leighton, and Matthew Sands,from The Feynman Lectures on Physics, Volume II,Addison-Wesley, Reading, Massachusetts, 1964.Reprinted with permission.

FOR HARPER AND ROW ARTICLES: The use of thematerial was contributed by Harper and Row, Publishers,solely for experimental use, and it may not be repro-duced, distributed, or sold in any other form, for onypurpose without the permission of the publisher.

248

P (lure Ctl(loc

Cover: Current, 1964, by Bridget Riley. Emuisionon composition board, 58 3/8 x 58 7/8". Courtesyof The Museum of Modern Art, New York City.

6

(I) Glen J. Pearcy.(2) Jeune fille ou corsoge rouge lisant. Jean

Baptiste Comille Corot. Painting. CollectionBade, Zurich.

(3) HPP stoff photo.(4) Femme lisant. Georges Seurat. Conte crayon

drawing. Collection C. F. Stoop, London.(5) Portrait of Pierre Reverdy. Pablo Picasso.

Etching. Museum of Modern Art, N.Y.C.(6) Lecture au lit. Paul Klee. Drawing.

Klee Foundation.

P. 28 Courtesy of The Metropolitan Museum of Art, N.Y.C.

P. 48 c Lincoln Center for the Performing Arts, Inc.,1967photograph by Morris Warman.

P. 82 Courtesy of Mr. Harold M. Waage, Palmer PhysicalLaboratory, Princeton University.

P. 90 U.S. Patent Office Picture Collection.

P. 102 International Business Machines Corp.Photo by Ezra Stoller.

P. 122 Tennessee Valley Authority.

P. 156 American Institute of Physics.

P. 212 Division of Radiophysics, CommonwealthScientific and Industrial Research Organization,Australia.

P. 238 Conductron Corporation, Ann Arbor, Michigan.

249

This Reader is the property of HarvardProject Physics. Upon request it is to bereturned to Pierce Hall, 29 Oxford Street,Cambridge, Massachusetts 02138.

250