## You are here

# Fritz Reiche - Session III

### Usage information and disclaimer

This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.

This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.

Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.

### Preferred citation

In footnotes or endnotes please cite AIP interviews like this:

Interview of Fritz Reiche by Thomas S. Kuhn on 1962 May 9,

Niels Bohr Library & Archives, American Institute of Physics,

College Park, MD USA,

www.aip.org/history-programs/niels-bohr-library/oral-histories/4841-3

For multiple citations, "AIP" is the preferred abbreviation for the location.

### Abstract

This interview was conducted as part of the Archives for the History of Quantum Physics project, which includes tapes and transcripts of oral history interviews conducted with ca. 100 atomic and quantum physicists. Subjects discuss their family backgrounds, how they became interested in physics, their educations, people who influenced them, their careers including social influences on the conditions of research, and the state of atomic, nuclear, and quantum physics during the period in which they worked. Discussions of scientific matters relate to work that was done between approximately 1900 and 1930, with an emphasis on the discovery and interpretations of quantum mechanics in the 1920s. Also prominently mentioned are: Niels Henrik David Bohr, Ludwig Edward Boltzmann, Max Born, Louis de Broglie, Byk, Compton, Albert Einstein, Arnold Eucken, Werner Heisenberg, Joe Keller, Hendrik Anthony Kramers, Krigar-Menzel, Werner Kuhn, Rudolf Walther Ladenburg, Alfred Landé, Stanislaus Loria, Max Planck, Radernacher, Rotzsayn, Clemens Schaefer, Hermann Amandus Schwarz, Otto Stern; Bad-Neuheim Conference, Naturforschertag (Salzburg, ca. 1909), Universitaet Berlin, Universitaet Breslau, and Universitaet Goettingen.

### Kuhn:

I started asking about the relationship between the mathematicians and the physicists, and then I discovered that's the wrong question. At least in many places you have to distinguish the theoretical physicists, the experimental physicists, and the mathematicians. Does this phrase a question for you? Because I want to think about this both at Berlin and then at Brelau. Who gets to give mechanics, for instance?

### Reiche:

Well I would say at that time when I came to Breslau the first time, the mechanics course — analytical mechanics I would better say - - was part of the big sequence of courses in theoretical physics. It was the first one. It was given by Clemens Schaefer. Threfore also one of his books, his first volume of theoretical physics, is devoted to mechanics and mechanics of deformable bodies. This was elaborated, as I remember, in the Planck lectures, but always the theoretical physicists gave it. Now the question could be put, "Is Clemens Schaefer a theoretical physicist or is he an experimental physicist?" He was in my opinion one of the rare men who were in a certain sense both. But,this was not the usual thing. For instance Planck, as you might know, had made only one experiment. It is even mentioned in one of his books, in a footnote. Very modest in a footnote. It's something out of thermodynamics as far as I remember.

But otherwise he never did any experiments. However, Schaefer was a man — funny as it might sound — more in the sense of what you find here as a rule, I would say. When I came over, I saw - - I knew these things already of course in Germany from reading American papers — that very often or nearly always, people are both making experiments and are making the theory. I just met for the first time Dr. (Ressler) from Cornell. Well, he is a theoretical physicist, as far as we always thought, but he is making obviously very wonderful experiments with shock tubes. And he knows both. But this was a very rare thing. If I go over the famous German theoretical physicists like Planck, Sommerfeld, Heisenberg and so on, even the younger generation, there was a very strict separation between experimental and theoretical physics.

### Kuhn:

That was true of the older generation though. Kirchoff was never an experinientalist.

### Reiche:

No, certainly not, no.... Helmholtz was both. Helmholtz was also a little more even, I would say, in the sense that he was an experimental man also, especially in optical things. This invention of the mirror to look into the eye, and so on.... So was Hertz. Hertz too. This is an exception. It was at that time, even before I was beginning to study, usual that the experimental physicists had not too much idea of theory. That was very characteristic of that time, at least in Germany.... Schaefer, at that time ... I always considered mainly as a theoretical physicist. And he knew very much of theoretical physics.... Pringsheim had the chair of theoretical physics at Breslau.... I never attended lectures by him. He has written a book which might be now old-fashioned, but I think it was one of the first books on the sun. And this probably was more, as far as I remember at least, a book on the experimental side than on the theoretical side.

But nevertheless the experiments which Lurnmer and Pringsheim did together, on heat radiation, in these he certainly was the more theoretical side of the thing. And Lummer was in no way a theoretical man, I would say. He was a typical experimental physicist. That you mentioned about the theoretical end, the physicists and the mathematicians. At that time still I think the word of Hubert was correct, that physics is much too difficult for physicists. This I would say has changed. I have the feeling. I hope at least. Not quite, but — In Breslau at that time there was a very good mathematician who was very much interested in physical things as far as they concerned let's say boundary value problems in relation to integral equations and so on.

This was Kneser, Adolph Kneser, the old father Kneser. Because he has two sons who are, I think, both alive. One is a physicist and the other is a mathematician. He was in constant touch at least with Schafer. And Schaefer also thanks him in his book for many advices and conversations. And he was really a man who was interested. Other people not so very much. Then I was later there as professor, then I met for instance Rademacher. And Rademacher, though being mainly a number man, as far as I know, was very much interested, and we have also made a little work together.

At least I asked him to help me. That was the real reason that I came to this. I was not quite sure in this special problem about the symmetrical top whether I made the thing correctly, and whether the proof that only these functions did not go to infinity was right. In this he helped me. But he was always interested. He always came to the physics colloquia. I can say certainly that Kneser as well as Rademacher - - though they were not on too good terms with each other — were both interested. It came from their heart — they were real mathematicians and both interested in physics. There was a man Frederik (Schur), not the famous Issai Schur in Berlin, but Frederik Schur, a geometrician. And he was not at all. He was completely separated from everything that was physics.

### Kuhn:

Did physicists also go to the mathematics colloquium? Or were these on abstract topics?

### Reiche:

Ja, this was usually on abstract topics. I cannot remember, I must confess, that I was ever at a mathemaics colloquim. But Kneser very often came to our colloquium. There is no doubt about it. He was very helpful. If something like a boundary value problem came up, then one had his very efficient help on it.

### Kuhn:

How did this compare with Berlin?

### Reiche:

... Those peple whom I knew in Berlin of the older generation were very old: and very specialized. I mean take Schwarz, Herrmann Amandus Schwarz, a famous mathematician. And whether he was ever really interested in a physical problem I doubt very much. Neither was the old Schottky, father of the pnysicist. Schur; not as far as I know. I mean, at the beginning of the influence of group theory on quantum mechanics, it might be he lived still, but I don't think that he was of influence before this. I mean Schur's Lemma and all these things are mentioned in group theoretical quantum mechanics and so on, but... at least at the time which I remember in Berlin, there was no contact.

### Kuhn:

We talked about books once before. There's one book you didn't mention, and I wasn' t smart enough to ask you about, which was Riemann-Weber.

### Reiche:

We used it very extensively, ja, ja. Riemann-Weber was one of the first comprehensive things we could use. Then it was replaced by Von Mises, as far as I know. This was and is still used.... Courant-Hilbert is in my opinion more difficult to read than Riemann-Weber was, at least the edition to which I remember to have addressed myself. This was relatively easy. If you took some effort you could - -

### Kuhn:

Was this pretty uniformly used? Was that a book that was widely used? Most everybody used it usually?

### Reiche:

Ja. Ja. Very, very much in use. And later it was felt that things are missing and must be extended. The Frank - Von Mises I must confess I never used very much as far as I can remember. The Courant - Hubert, yes, some parts. But there are parts which at least for me were a little more difficult, I have the feeling. Just the beginning, just the beginning. It is unfortunate, and so it is a little discouraging in the beginning, because it might be that one doesn't feel quite interested in following all the things through: This does not mean anything of criticism of course.

### Kuhn:

Among Planck and others, was the atom now pretty generally accepted? The idea that the world is composed of atoms.

### Reiche:

Oh that.... The energeticists, ja.... I think in my time this was already completely accepted as a basis of physics. In spite of course of the fact that only a few experimental things were known which proved them, like Brown's motion and so on. Einstein's theory of Brown's motion and the work of Smoluchowski and others. They were very characteristic that something was absolutely true in these things.

### Kuhn:

Do you remember discussions about Einstein's Brownian motion paper? Was tnis easily accepted?

### Reiche:

Ja, I think so. I think so because I remember that at my time already a little book appeared by the daughter of Lorentz, who was, married to a man of the name Haas, a Dutch physicist. I'm pretty sure I had it and I read it at that time. Different modified derivations of Einstein's derivation were given about the Brownian motion, and modifications and complications of Brown's motion I think there were there. Quite a lot of things. Also by Smoluchowski. I remember myself to have heard a lecture by Smoluchowsk It was in one of these Wolfskeb lectures in Goettingen. He gave a highly interesting lecture which also is published somewhere. There was a book on this; the Planck lecture and the lecture by Srnoluchowski. There were also problems concerned with Brown's motion, not only with the general entropy concept and similar things. So I would say in my time I have at least a feeling, if my remembrance is correct, that the idea of atomism was completely accepted.

### Kuhn:

What about the aether?

### Reiche:

The aether... It is a little difficult to say exactly because after 1905, after Einstein's paper, everybody had really the feeling that there is no doubt. This has no meaning, or another meaning.... When only special relativity was discovered, I think everybody had to say so this abolishes every possibility of an aether....

### Kuhn:

Of course that doesn't happen until people take this paper really very seriously.... Now you mentioned that you had sent a reprint of your thesis to Einstein, and that he had returned with some reprints of his own, including the famous paper on special relativity.... That would I think have been 1907 or something of the sort. Had you not known the paper earlier?

### Reiche:

Probably not.... It might be that I read it, but obviously without very much understanding. You are quite right because I can only have sent my tnesis in 1907. I sent the reprint only in 1907. Ja, it might well be that I didn't read it, or that I didn't have an impression. But I cannot quite remember about this fact. I would say that it is not very likely that I didn't know it, because certainly other people with whom I was together — but now let me see. I was a little isolated in Berlin at that time, and did not give attention too much to every paper because I didn't know exactly which paper was of importance and which was not of importance. So it might well be that only in Breslau where I was in a definite circle of people who were interested, that there I knew the paper really. And I read it and was really startled and impressed.

### Kuhn:

Now with whom did you talk about it in Breslau?

### Reiche:

In Breslau there were different people, Schaefer, and also Kneser, who was at that time of corse interested very much in the structure of the theory and so on. But Schaefer was, I would say, my main theoretical adviser who inspired me to read more.

### Kuhn:

In the early part of the century there is quite a lot of work being done still in optical problems of the electromagnetic theory.... Did these seem fundamental? Did one feel that there were important things to be discovered still by solving problems in reflection of waves, phase change, that sort of thing. Or were they exercises of mathematical skill?

### Reiche:

Yes, I think the second is true. Certainly in a very big sense, I would say. Because all diffraction problems which came up were solved first by Huygens' principle. Then the phase was not quite correct. Then Kirchoff discovered how to do it, with the Kirchoff principle, Kirchoff-Huygens' principle as it was then called. It was already known that the boundary conditions which one uses - namely that inside the aperture the undisturbed incident wave is to be put as boundary values - - are mathematically not quite correct. In spite of this fact, there was always the feeling that nevertheless, the degree of accuracy is sufficient. As far as one had made experimental diffraction measurements, the theoretical prediction was quite nice, or very good even.

But then came the first — as far as I know — without first rigorously solved problem — the half plane - - by Sommerfeld, which by the way, I remember - I cannot say exactly when it was — I read this. I think it appeared first in a mathematical journal, as far as I remember. But went over, so to speak as reprint, into other periodicals, even into books. And then of course the question came up, is this the only one which we have and which one can rigorously solve the boundary value problem. And then came up the things like taking a sphere, a dielectric sphere or a metallic sphere or a cylinder. By the way, Schaefer was very much interested in this, especially in dielectrics.

He made also experiments with another man, Grossmann, about the experiments of diffraction of electromagnetic waves. And I think more and more such problems came up. But at that time at least there was no rigorous problem until the next rigorous problem was done by Schwarzschild about a slit. Of course with a method which is not converging, as far as I remember, too quickly. So it was always something like a reflection from edge to edge and so on.

But the distribution was absolutely mathematically rigorous.... There was the Sommerfeld solution, and the Schwarzschild solution, and a few of the other solutions, like the cylinder solution, and the sphere solution which was done mainly by me first I think, and then by Debye and different other people. I think also the parabolic cylinder has also been done already at that time, by Epstein I think.... There was a terrible big thing by a man, Moeglich, who claimed to have solved the thing for the general ellipsoid. The general eillipsoid, not the ellipsoid of revolution but with three different axes. But I do not know whether this is completely correct.

So I would say it was a mathematical sport, I would say. A kind of mathematical sport because the fame of Sormmerfeld did not allow to sleep, and one had to try to do something similar. Though of course, if you ask for the practical influence, I would say it was really not too much. Only the question of basic science. Nowadays this work is done in part in Germany. They are again going deep into diffraction problems, at least a certain group, (Franz) and (Defferman) and different other people. And here of course, in our electromagnetic group downtown in the Institute. This is put now on a really new idea by Joe Keller, who has discovered what he calls geometrical diffraction — which seems to be a contradiction. But he has enlarged and amplified the idea of geometrical optics by introducing the rays, which have phase and amplitude and so on. The introduction of the idea of diffracted rays.

So that one can do these things really with a great accuracy by his idea. But one must study carefully the recipes which he gives. And if one can compare his results with the rigorous solution, if it exists, then the Keller approximation is always excellent, I would say, in agreement with a rigorously solved problem.... There is — different places he has published this and given talks about it. He has extended so to speak the Fermat principle by the idea of diffracted rays, which are coming from the edges and are spreading out from every point of the edge which is hit by the incident wave.

And he has only to fix now in which plane lie the diffracted rays, what are the amplitudes, what phases have we to give them? This I cannot say — but I guess he made it backwards. Or in a certain sense backwards. But one had to be very good in inventing the new rules, and that's what he really has succeeded.

### Kuhn:

You talked very briefly about the Naturforscherstag. Was it in Salzburg in 1909? You spoke of hearing Einstein There.

### Reiche:

Oh ja, ja.... Once it was in the famous resort, Bad-Nauheim, a resort place for heart conditions. And there was also a big Naturforscherstag. There was this famous discussion of general relativity, that was in 1920 or '21. I am not quite sure anymore. I think it was '20. There was a big discussion under the chairmanship of Planck, in which Einstein was sitting. A lot of people stood up, especially Lenard and Gehrcke and Stark, this big anti-Einstein triplet, and asked him very, very clever questions, which were not so very easy to answer. But he answered everything in complete calm, without any anger about the thing. And this went through hours and hours.

This, by the way, was the first time that young Pauli was presented to the society, to the physical society, so to speak. Because Einstein mentioned, without knowing that Pauli was present, something about a very excellent article in the encyclopedia of mathematics and recommended everybody read it because it was very conrehensive and very clever. Then Sommerfeld took Pauli and stood him in front of the whole assembly and said, "here, this is the young man." I think he was nineteen at that time or something like that.

### Kuhn:

On the Bad-Nauheim conference. Were these questions that were put by Lenard and Stark at this period still about special relativity, or were they now concerned with general?

### Reiche:

No, this was the general idea of general relativity. The idea of the equivalence principle between an accelerated system and a system which has a gravitational field. These general things mainly were discussed, and the difficulty of understanding for the customary thinking were absolutely true. I mean a lot of things were said by these adversaries which were very impressive because everybody had the difficulties of understanding. Then Einstein stood up always and answered and said how one has to think of them from the general standpoint of general relativity. Everything is modified by the introduction of the gravitation field. I cannot remember whether very specialized, deep-going questions, except for these fundamental things, were really discussed, or could be discussed.

### Kuhn:

Now in the 1909 meeting, which you said was the place where you heard of the fluctuation paper of Einstein.

### Reiche:

Ja, there I saw Einstein for the first time.... I remember that James Frank was there. And certainly, I gues at least, I remember that also Philip Frank was there. And our whole Breslau group. Schaefer was there. This I remember exactly. Ladenburg was there. Whether Lummer was there I do not know. This I cannot remember.

### Kuhn:

Was Planck there?

### Reiche:

Even if I do not remember, I am pretty sure. I am pretty sure. Planck was in such things very highly conscientious. So only if he really was not able, hindered by sickness or so. Which was also very, very rare. He was a very healthy man, who all his life had only once I think a heavy pneumonia.... This is a thing which is funny. I cannot wish you to see it so very soon, but if you grow older, you remember more of what lies in your earlier times. More than what yesterday, I would say. This is a very funny event. several people to whom I talked have said they remember old things better than recent events.

### Kuhn:

Did that paper make a big impression?

### Reiche:

I must say I was very much impressed by the appearance of the second term in the fluctuation formula. Though it is of course a rather indistinct proof of'photons. I remember of course that people were opposed and tried to find another reason or tried to give to the formula another form. Of course one can put it into one term together if one manipulates a little differently. I remember to have seen such things in a collection of talks which Willy Wien gave somewhere.

I mean, if one introduces the thing so, that the energy E appears in both terms, then it is very clear that one term is proportional to E and the other is proportional to E^2. The proof is given very extensively and I think very carefully by Lorentz in his book, La Theorie Statistique et Themodymamique. I have this book still. I have always preserved it. I am looking into the book from time to time because it is really beautifully written. The proof in one of the appendixes is very extensive and very carefti and leads absolutely to the one term in which the E^2 appears.

Then this other term, which one can find again and again if one uses Wien' s law. It is shown very clearly that it behaves at least like a cloud of particles, like molecules. Of course there remains the enormous conceptual difficulty. This is cornpletely true, and therefore I think also Sommerfeld and Debye tried to do this in a different way. In the moment I am not really quite sure, I do not quite remember. Do you really think that the Compton effect was the breakthrough which convinced everybody? Because there it is still a little more. The funny thing is that the one is divided up into the scattered and different prime, and the rest is energy of the electron.

I remember that the iressiOnof the discovery of the Compton effect was enormous, enormous at that time. This I completely remember. It was '22. There was one year there when I was already in Breslau. I came there in '21, and I remember completely now that you speak of it, how great the satisfaction was, but the satisfaction was a little more expressed in the way, "So you see, Einstein was right. Einstein was right."

### Kuhn:

Now who would have said that?

### Reiche:

Well, our group, who were Einstein enthusiasts. Ladenburg for instance was very open to all these things, and always very mach impressed by all not only by Bohr, but later by Kramers and Schroedinger and so on. So he was very full of enthusiastic temperament, I would say. And he was also later the man who always pushed me to read this, read this and read this. He was in this respect a very good scientific friend. And he was very enthusiastic about this experimental proof of Einstein's idea. But the belief that it was correct was already very big, very large, at that time I would say.

### Kuhn:

How was it announced at Breslau, do you remember? Was it the Debye paper or the Compton paper, do you suppose, that was soon first?

### Reiche:

Ja, well this I cannot tell you. Probably the man who read it first in our group was probably Ladenburg, Ja.

### Kuhn:

A friend of mine noticed, in Jungk' s book,* Brighter than a Thousand Suns*, a remark about a 'chemist', Reiche, in 1941 — and I wondered, are you the chemist in that?

### Reiche:

Ja, Ja. Why they came to the idea of that, I don't know. Ja. Shall I tell you about it.... Very briefly, before we left Berlin - - I think it was even the day before, or two days before - Fritz Houtermans came to see me. And said, "Please remember if you come over, to tell the interested people the following thing. We are trying here hard, including Heisenberg, to hinder the idea of making the bomb. But the pressure from above" — the whole thing is very funny — this in parentheses — because it seems now to me that the thing just opposite. But at least he told me this — "Please say all this; that Heisenberg will not be able to withstand longer the pressure from the government to go very earnestly and seriously into the making of the bomb. And say to them, say they should accelerate, if they have already begun the thing." This was, by the way, February — no, March, '41. "They should accelerate the thing."

Well, I learned this a little by heart, and then I came over, and very shortly after we arrived the Ladenburgs took us out. We were living here somewhere in New York. And we came into a big party. There was collected Bethe, Wigner, Pauli, and all those people who are interested in these things. I remember von Neumann was also present, and different other people too. So it was a group of approximately ten or twelve people. And I told them exactly what Houtermans told me.

I saw that they listened attentively and took it. They didn't say anything but were grateful. And later I read the first report, the Smythe report. I told my wife at that time, "Well I have the impression, now I know." I didn't say then anything new, and they were already in the midst of the Manhattan Project, of course, and they knew everything.

### Kuhn:

E xcept how seriously to take this story. Did you take it literally yourself?

### Reiche:

I thought it is so. I thought really it is so that Heisenberg was strictly opposed to the whole thing. Obviously not. This was not so, I think.

### Kuhn:

It's hard to tell.

### Reiche:

It's hard to tell. I mean you cannot even say after reading Goudsmit's report. But it is the business by which I came into this book, and I think it is with a wrong first name. I think "the chemist O. Reiche." Ja, this is really true, the thing.

### Kuhn:

Ladenburg. Where did you first know him? At Breslau on your first trip?

### Reiche:

The first time when I was there, ja, my first trip. I became acquainted witn him 1908. I think I told you already, he first asked me to help him a little in a certain work about measuring intensity of X-rays.... This was an experimental work. But unfortunately my whole experimental work, as I think I also told you, was more dangerous for the Institute than help for the scientists. But nevertheless I understood how difficult it is to make good experiments. And then I became more and more acquainted with him personally. I think the main things which we did at that time and later are things which have I think no direct meaning for quantum theory, but in a certain sense at least for the measurement of these f values, which play a role in the dispersion theory.

The main paper which I did there with him was about what we called total absorption and line absorption.... We made this other contribution which had certain importance, also later, as I was told nicely by Professor Spitzer once. He told me that this paper was a basis for a lot of experimental measurements in astro-physics. This was the question of the dependence of the total absorption on the product or the number of absorbing particles and the thickness or the absorbing layer. And there we were lucky really to be able to explain experiments which were done earlier by a French physicist, Gouy. We could really show why the result of Gouy came out.

So this was I think our first working together. And from this time on we always were in contact, except as I already mentioned I think last time, or at the telephone, I was not involved, and could not follow his last work; his last work of the last year, about boundary layer problem and measurement by using the interrerence method. This was a very extensive thing which has been published by (Berschader), one or his former students. I have the report, but this is a field really which is quite separated from what he did otherwise. And then came our beginning to work on the idea of dispersion theory.

### Kuhn:

When you worked together, was he the experimentalist and you the theoretician? You have told me, when there was experimental work to be done, he was surely the experimentalist.

### Reiche:

Ja, that's right. But I mean, the work about this absorption was his first. It was quite for himself that he did this. He was I think the first who showed really experimentally the absorption and the anomalous dispersion of the hydrogen atom, of the first member of the Balmer series. For this he had to use obviously a quite clever idea. Because one must have first the thing in the correct state to absorb. This was the first thing where I only saw what he did, and he gave talks about it.

But in this I was not at all involved. First was absortion. This he did alone. And the second was anomalous dispersion. He always used one interferometer which he liked very much, the (Jaumann) interferometer. This second work on dispersion was done by him and Stanislaus Loria, a Polish physicist who was at that time in Breslau. He later became, after the Poles took over Breslau, director of the Institute. But he is unfortunately not alive anymore. He was a very, very nice fellow. But all this was without me. I did not do anything.... I was there, but I was not taking part in anything of this measurement or giving a theory for the measurement or something like this.

This he did completely alone, and in part with Loria. And then he continued these things. I think also he worked out also experimentally magneto-rotation, this Faraday effect. But all this was on his line, but nothing with me. And I think the first thing which we really did, together was this little paper about the total absorbtion and we distinguished between the line absorption and the total absorption.

The light source on total absorbtion is a more or less continuous spectrum, at least broad against the absorbing line, And in the line aboorption we made the source also a line of the same shape as the absorbing lines. But in these things, he did, as far as I remember, not himself experiment. He had read the experiment results by Gouy and was not quite understanding why this came out and asked me whether I would try to put down the theoretical basis of the whole thing. And so I did, and then it came out — the formula for the total absorption which contained Bessel functions of the first and zero order. So I wrote this thing, and also given two special cases. It is always the product of the number N of absorbing particles, and the length, L, of the absorbing layer. And if this has a certain small value—small against a certain number then one result came out. The other result was the other extremum where the N times L is large.

And this second was just (???). This one saw at once. He wrote to me, "Yes, well you found what Gouy made." I didn't know exactly what the results of Gouy were. And then he told me, "Well, now I think we have to make a little paper from it." And this I think was the beginning really of our work together.... We corresponded on this. I was back to Berlin, because I think the paper is published in 1911, in the Annalen der Physik. It was at first published in a special periodical in Silesia. There was a society, and there we gave a talk about it, and they had the possibility of making a publication possible. But later we sent it to the Annalen der Physik, and there it is published. This was I think the first real collaboration which we made.

### Kuhn:

Was Ladenburg himself pretty strictly then an experimentalist?

### Reiche:

He was, as far as I understand, a very good experimental man, but he was one of the men who could make, let me say, easy theoretical work, He had ideas, that is the main point. He had good ideas, and it might be he wouldn't have been able to make big theoretical papers. Though of course I remember his thesis. He made the thesis in Munich under Roentgen about friction of spheres. And he also gave a quite nice theoretical appendix, so to speak, to it. The influence of walls on the motion of a sphere in a viscous medium. I have quite a lot of papers of him with me here, so I could show them to you or give them to you, except for those which I would like to keep.

But all these can be of course found in periodicals.... It only makes sense to give you those which have something to do really with quantum theory. But I mentioned this only because there he made - - probably with the help of some mathematician or some theoretical physicist - - quite a bigger paper, purely of theoretical basis, which was supporting the results of his experiments. But I would say he never did this again, as far as I know and as far, as I saw in the collection of his papers. But he could do little, but not minor, things, I mean little but important things, very well.

He knew to handle even the numbers, I would say, and the symbols. And so he has made this contribution. I think it is the first time that this relation between the old classical number N of dispersion electrons and the Einstein coefficients has been put down. As far as I know.

### Kuhn:

One of the things I've wondered is whether you had talked to him about that and been in on that. That was 1921 so I don't know how it relates to your return to Breslau.

### Reiche:

Ja, about this year. We always were in constant correspondence. Because this was then also the basis of this later paper in Naturwissenschaften, which of course was not a very rigorous theory. I think I emphasized this already to you, that one cannot take it as a proof of the dispersion problem. The real, the correct frequencies of the emission lines come in. But this we found so self-evident that we assumed this, without proof. And the only thing which was really used in this paper, I would say, was his new idea about the connection of the N with the transition probabilities. But nevertheless, we played a little around with it. We did not give as Kramers and Heisenberg, a correct revision of the complete dispersion formula.

### Kuhn:

This is the first attempt after the Bohr atom to do a dispersion theory that doesn't try to do it with the Bohr atom, that uses classical oscillators, if you will. This turns out to be terribly fruitful, not in the inmiediate sense of giving you a theory but in starting a track. In another sense in giving up the Bohr atom it's a great surrender. Did he feel that? Did you feel that?

### Reiche:

Do you mean a surrender in the sense that the Bohr atom taken seriously and tied on with the usual methods of classical mechanics and classical electromagnetics?... It failed completely.

### Kuhn:

Was it clear that it failed completely?

### Reiche:

Ja, I think so. Because whatever you did, you never got the transition frequencies, but you always got mechanical frequencies of little disturbances around the orbit. Even if it was not quite elicit in all the treatments, they had to be the result. The anomalous absorption frequencies had to be the mechanical frequencies.... I guess that it was clear to him more than to me. Because he was so experienced.... Then it was, as far as I remember, always obvious. to us that this type of a dispersion theory with electrons going in an orbit, could never give the correct frequencies where we observe anomalous dispersion. This was probably the reason that we at once considered this as a completely obvious thing....

### Kuhn:

What did you suppose these oscillators were? You assume oscillators as frequencies of the transition probability, now was this for you a physical assumption?

### Reiche:

Ja, but think of what the language of Bohr was always. I mean his caution in all these things concerning what happens in the transitions and so on. All this was already a hint that we had not to consider the orbits really in the sense of classical mechanics and celestial mechanics. Therefore what do the transitions really mean. The transitions are equivalent to a kind of oscillator. I do not know whether we or Kramers first used this terminology of virtual oscillators.... It might be it is Kramers. If it was Kramers then we certainly at once incorporated it in our thinking.

We cannot say anything about it so far, but what happens in the transitions is like what happens in an oscillator. These virtual oscillators are the main things and not at all the little disturbances of the electron around the orbit. They have nothing to do with the real thing. It might be that we had a certain negative support by the fact that anyway these orbits are all unstable in the sense of mechanics; because of the electromagnetic radiation which they send out. But at that time one could only say, well, this is not yet clear.

### Kuhn:

Did you think maybe there really are oscillators at all these frequencies?

### Reiche:

No, I don' t think so. I don't think so. I think we took it as symbolic. Therefore, at least the terminology of virtual oscillators was very agreeable to us....

### Kuhn:

In the radiation theory, you do a classical computation for the radiation from an oscillator. Then you do the absorption in terms of the Einstein coefficient. And equate these for equilibrium. It's classical radiation but quantum mechanical absorption. Was that a bother? [Examining Whittaker's discussion of Landenburg's 1921 paper]

### Reiche:

Ja, but actually it isn't this way. If U is the average energy of the harmonic oscillator, then this gives the emission. Now, if you have an oscillator in a field of radiation, of density rho, then this gives you the absorption. And if you put them equal in a state of equilibrium, you get this. He has the thing a little upside down.

### Kuhn:

The '23 paper is not easy.

### Reiche:

Ja, I completely agree with you. It might be a little too much on so few pages. But it uses already obviously this previous paper by Ladenburg.... But you see, here it is quite clear that he turned the thing upside down. I would always say that this is the original expression of the emitted energy in terms of the average energy of the oscillator. This is the absorbed energy of the same oscillator in a field of radiation density rho. If you put them equal in a state of equilibrium, you get this formula. But here also we used - and I am pretty sure that (Ladenburg) also did — You see here in an earlier paper — this is a little earlier — it is quite the same. This is the same for the total absorbed energy. So I would always say I would reinterpret this. I would say this is the absorbed energy, classically absorbed, and this he put equal to the quantum absorbed energy, which is this. Then he used the relation between the B and the A.

### Kuhn:

I do see what you mean now. In any case it's to take the classical process and to apply it directly in this way to the quantum mechanical process. How did one feel about that? Is this straight correspondence principle?

### Reiche:

I always had a feeling, concerning these relations and the consequences, that this was a main idea which he had. That one should try. Let me interrupt, namely. The N which is written here, also here, also here — in a German letter — were the number of electrons which produced the dispersion. The experiment evaluated the strength of the absorption or the strength of the dispersion by observing how much the interference fringes were deviated from the straight line character. This,was always a measure of this N. But the funny thing was this N came out different in different lines. And, astonishingly, it was in some lines less than the number of atoms present. So that one cannot say that there is one whole electron responsible for the dispersion, let's say in hydrogen, in a certain line.

Or in a sodium line, in the D line.. But that there is only 3/4 or 1/2 of an electron present, which of course is senseless.... In hydrogen if one electron is responsible for the dispersion, then one should always expect that this nunber, the German N, must come out equal to the number of atoms. But Ladenburg found in different measurements, that this number N was smaller than the number of atoms. Therefore really it was a problem, how can one explain this? And this was explained by the idea that the electrons are not working like complete electrons, but like virtual oscillators which are characterized by a certain strength number. Coming back by the way to Whittaker, I have the feeling that it is of course correct, but it's making the thing a little more ununderstandable.

It tells you — which is true, of course, — that the emitted energy is proportional to the density of radiation, which, according to electromagnetics, it is not. Therefore I would always prefer the thing to be turned upside down and take this on this side as the correct emitted energy, which is independent of the radiation. Then compute in the same way, by pure electrodynamics,' the absorbed energy under, the influence, which must be proportional to the density of the radiation, to rho. Put them equal, and you get then this correct relation, which is the so-called Planck's relation. Which he used always extensively in all his papers. But now comes the idea of Ladenburg. 'Since we are in doubt about the meaning of this German N, and the funny thing happens that it is smaller even than one, so I would like to try the following things: I know the classical absorption, and the classical absorption is given by this formula. I know, with Einstein' s coefficients B, the quantum absorption. What comes out if I put them equal.

On one side I have the classical German N, and on the quantum side I have the correct number, Latin N, the real number of atoms. And so I get a relation between the German N and the Latin N by putting the two absorbed energies, classical and quantum, equal. I think that this was historically his good and. very helpful idea. He could have stayed with the B's, but he preferred also to find the relation between the German N and the, Latin N, by expressing the B's by the relation of Einstein replacing them by the A's.

### Kuhn:

Now that I see what you were driving at, it makes less difference than I had thought.

### Reiche:

Well, there's still a certain — one had to have the courage to do it, to put the classical absorbed energy with the German N equal to the quantum absorbed energy and see what comes out....

### Kuhn:

Whittaker does it here and leaves out something important. In Ladenburg's and your articles you use the *Abk lingungszeit*. You call this decay time tau and you derive, fairly straightforwardly, a formula which writes the decay time in terms of the electronic charge, the frequency, and so on. You can go through this whole , derivation without ever introducing tau, which is what's done here. Think of these as though they were continuously radiating and. continuously absorbing. Clearly then the reason for introducing the decay time is because of what you think these oscillators are like. You must think of them as taking on energy in a discrete finite process. A decay process. Then it may build up again.

### Reiche:

Ja, I see approximately your difficulty, but I am not quite clear whether I understand you correctly. It is also in a certain sense proportional to the lifetime of the higher quantum state. At least it is of the same order of magnitude. I think this has once been shown by Stern, I think, long ago. If it is an isolated atom in space, completely isolated from all external radiation, how long will it remain the higher state? And I think a good measure is tau. Which of course is gained by a purely classical calculation. I think it is so that the number is going down like e to the minus t over tau.

### Kuhn:

When this paper was done, did this seem to be the answer to the problem of dispersion?

### Reiche:

Ja, we thought. Or, at least, we thought that this formula is right. I mean, even if you are in the ground state, where the negative tems in the dispersion formula are missing, because there are no lower levels, it is not'a proof of the dispersion formula because of this business with the frequencies. I mean it's only a guess based on a relatively strong confidence and foundation, that we had to put the correct einision lines instead of any other frequencs. But concerning the German N formula, there we were pretty strongly convinced that this was correct, and that one could find really the correct factors in the dispersion formula by knowing a way to compute the different transition probabilites. Now for this one had from Bohr's theory again a certain very, very good flint. And you remember there are these big papers by Kramers, I think it was his thesis, about the Stark effect and all these things, which was very nice.

### Kuhn:

Were there numerical computations made to see how the number of dispersion electrons came out?

### Reiche:

Ja, ja, many I think. One of them was done — I tnink it was for the sodium atom — by Thomas; the same Thomas who was the discoverer of this f sum theorem. He was a young student of ours who unfortunately died of tuberculosis. In his tesis he made computations which were quite nice, of course based on the older quantum theory. There was a procedure to find good factors, amplitudes, for the orbits, and so on.... No, it was after quantum mechanics came out, I think.

Because he used, as far as I remember, a method which was first published by Fues. And Fues was an assistant to Schroedinger in Stuttgart, as far as I know.... Before the matrix mechanics this was more or less a guess. We tried to do it here, I think for the sodium atom or even for the hydrogen, but I would not (put my hand in) this. Because for the two D lines in sodium one knew the ratio at least — that it was two to one. One was two times as strong an oscillator as for the other line. One could play a little around with these things. It was possible in a certain sense for simple atonis to do it. But even in the case of hydrogen it was a pretty long computation, as far as I remember from Kramers. Kramers played with it. But they gave a very good argument with the experiments.. Not quite, it came out to be clear later, but they were very good really.

### Kuhn:

So really this did not look like one of the fundamental problems any longer? And it had earlier.... The notion of virtual oscillator really seemed to be working very well?

### Reiche:

Ja. It worked very well. But of course there was missing the proof for the dependence of the real emission frequencies. This I think was only done by Kramers and Heisenberg. We acted as if it is proved, and I think we mentioned that one has to do this. Because one saw it so very clearly, that these so-called infinities occured not at the place of any mechanical frequency or oscillation, but it was exactly at the D Lines.

This was done by — it was a very famous paper — Roschdestwensky. I think it's published in the Annalen der Physik - about the anomalous dispersion of the D lines. And he gave beautiful pictures — quite large pictures — about how the index of refraction is behaving in the immediate neighborhood, in the small region where you have the anomalous dispersion.... And this came out, as far as I remember, in the time when Ladenburg was working on these things. It was a very interesting and very good experimental paper. There one saw that there could be no doubt where the real anomalous dispersion lies. And these were exactly the fequencies of the two D lines. Yes, he has measured the 'whole thing - in thq D1 line, the D2 line, and what happens between' them.

### Kuhn:

The asymmetrical top you did in 1918?

### Reiche:

I don't think that I did, previously to this paper on the specific heat of hydrogen, anything about top.... Though it might not have been necessary to do it for the top, but only for the rotator, I nevertheless tried to do it for the top.... In the first sections I solved this problem, which in a certain sense was not quite fortunae in my opinion now. I had by this track too many quantum numbers. I put the top in a gravitational field, but this is not necessary, of course. If it is a free molecule, then you have three quantum numbers, but two of them appear always in combination of addition, n1 + n2 or nj + n3. So actually there are only two quantum numbers. If you see it from the modern standpoint, there's no doubt about it.

There are only two quantum numbers in the symrietrical top. In the rotator even only one. So I needed actually only one of them. But later I took the n equal to n1 plus n2, and I replaced it simply by n. There was a certain series. First you had to take the logarithm of the series and then differentiate it with respect to a certain parameter in order to get the specific heat..,. And this is on page 670. There, where you see the framed formula is Cr, the rotational part of the specific heat divided by the gas constant R. And on the right hand side there is a certain parameter sigma, which is also defined on the previous page. And the sum of the series, as you see how in the middle of this page 670, is depending on the statistical weight. And this is the trouble. This was the only trouble.

So first I took all the quantum numbers as double, except the n equal zero, and then I got the 2n plus one as statistical weight of the quantum state of quantum number n. And with this I got I think the picture "I" on the next page, which has a maximum and a minimum and so there' s certainly no possibility of bringing this in agreement with the experiment. And this was the idea, that one tried different ideas for the statistical weights, different possible hypotheses. One of them was, for instance, that one made the zero state nonexistent. There is always a rotation. No state with angular momentum zero.

This also I think eliminated at least the maxima and minima, but it didn' t give the correct curve anyway.... Originally I thought that one had to do it in this way, tnat all the quantum numbers except the zero of course can have positive or negative values. And so I tried different curves and I think the best' still was the curve, number five. But even this was not good. Curve five is on page 684. But there was one experimental point terribly below, terribly far off. Now at that time of course one didn't know exactly whether Eucken's measurements were very good. I mean Eucken was a very good man and so one could trust. Nevertheless other people also tried to find these things. And I think among them was also Planck.

And I think Planck also got a maximum. If I am not quite wrong, it is correct that either the parahydrogen or the orthohydrogen has a maximum and only the combination made the maximum disappear. But at that time the maximum, or the maximum and minimum even, was absolutely catastrophic, and one could see this cannot be the right thing.... nevertheless, at the end of my paper I was careful. I left it open.... Later Eucken was in Breslau for a long time, and I was very much in touch with him. But I'm pretty sure that at that time, I asked him about the reliability of this deviating point. I am pretty sure. I do not have a letter or postcard, but I had some postcard, I am pretty sure now. But where is? This is a question again of the closets.... And I am pretty sure he said the point cannot be that far off, it cannot belifted to such an extent. So I am pretty sure that this was at that time really a hopeless procedure, that one Couldn't find the correct answer.

### Kuhn:

Do you think you felt that way then?

### Reiche:

There is a question of conscience. I do not know exactly. Well, I will not say too much. I do not know exactly.

### Kuhn:

You talked a little bit on the infra-red spectrum problem last time. You talked particularly about your converation with Einstein which you described just at the end there. What surprises me is the ease with which this suggestion comes from Einstein. Is it an indication that he doesn't take quantization very seriously? In Planck's second theory of course there have been lots of half quantum numbers

### Reiche:

Ja, but this I think I can say with a certain security, that Einstein never believed in Planck's second theory. I mean Planck's theory, as far as I remember, comes by the fact that the absorption is continuous.... But this Einstein never believed, quite corresponding to his radical thinking in nearly every field. And I think when I asked him this question and showed him, I remember that he said,"We must keep to the jump of the quantum numbers. I mean, this will be always h, but the absolute position of the quantum numbers, whether t is zero, 1, 2, or whether it might be" — it is only a suggestion which he made — "one half, three halves, the distance must be always the same. That I am completely certain about," he said, "but I would not say where the thing lies, whether we can shift it or not. I wouldn' t have been surprised even if he had said, "Well, try it with other position of the absolute values of the quantum numbers." But of course the next one was clear. . .

### Kuhn:

But he did not like the Planck second theory?

### Reiche:

He did not like Planck theory at all, no. This I remember that he said at different times, "no, no, in this way it cannot be."

### Kuhn:

You yourself thought that the issue was open, did you not? As to between the first version of the theory and the second?

### Reiche:

Ja, but this is due I would say to my lack of courage to decide. At that time I was very young.... I have the feeling that, due to the authority of Planck, one said this is in a certain way out of at least one difficulty of the non-classical emission. But I had always the feeling most people did not believe in Planck's second theory. I mean Miss Rotzsayn made this calculation with the second Planck theory, but this was due to Planck himself as far as I know. I think it was a thesis, and Planck had given the thesis or had suggested to do it in this way. No, most people did not follow Planck I think in this. I did it once. I remember. But this, I would say, even answers your question. I made out a paper very close to this, about paramagnetism in solids. I am not quite sure.

The whole thing was quite incorrect. I remember that Stern was very angry about me, that I did such things. He said, "Well, what do you do? Do you think of rotations in a solid?" By the way, I think later one knew that rotations of molecular groups are possible even in solids. But at that time I did it without very much criticism, and I used as part also the second Planck theory. But I would say more or less just trying out this. I do not know whether I had first tried it out with the first theory and seen that it does not give correct results. and then I tried the second theory. But I was obviously not very critical at that time about this.... Going back to your question about Einstein's idea, I think he was absolutely opposed to the second theory and said, "If at all, then in this way.

These are the only states which are really existing." And what, as he said, I still remember quite clear his words. I think he said, "Reiche, we know now" — he was always a little cautious — "I think we know now that the jump from one to the other is exactly h. But where you begin, you do not know. And why not take other beginnings?" That's what he said directly. "Try out to begin with a half h, then go to 3/2 h, and, let's see," It was a matter of a minute or not even.... So then I looked at him, of course, and said, "What shall I now do?" Because I had already the galley proofs and "Shall I leave the whole thing and put it in the waste-paper?""No" he said, "Give this what we have tried out here." So I said, "Well, I can't do that.""I give you the permission." He was very generous in all these things. And so I put this little note at the end that this solved the problem.