Immanuel Estermann

Heilbron:

Well, perhaps we could begin, if you would, with some account of how your interest in science first was aroused.

Estermann:

Well, my interest started while I was still in high school — in the early parts of high school — about the age of 12 or 13 when I happened to get hold of some textbooks which were not even intended for me yet. I was primarily then interested in chemistry. Also, I had my first science course at that age. I went to school in Berlin then, and we started with physics rather early, in the class called (Untertertia); that’s six years back from graduation. And it was chemistry a year later. And I found the textbooks very interesting and started to read and studied a great deal on my own. But as I mentioned before, I was then primarily interested in chemistry, but I was not quite so selective. I dabbled a little bit in all other fields of science. My father was an M.D. and we had a fairly large collection of scientific books at home. Not being an athletic type or otherwise greatly interested in outdoor activities, I spent probably a good deal of my time rummaging through these books and encyclopedias and other things, so I think that’s where my interest really developed. Then in my last three years of high school we had moved to Hamburg in the meantime — I happened to have an excellent chemistry teacher who probably did a great deal to channel my interest from a random into a more directed program. The last year I was actually excused from the regular class and was given permission to work on my in the laboratory.

Heilbron:

Was that a quite irregular proceeding?

Estermann:

That was a rather irregular proceeding. Now, this was during World War I when the classes were not very great; most of our students were already in the army. I was the youngest in the class, and therefore escaped the army service in 1918. I graduated from high school in 1918, so a great many irregular things happened. The school had an excellent laboratory in which we had normally student exercises of two hours every two weeks, but I was given permission to use the other hours also in the laboratory and work on my own.

Heilbron:

Was that true of school in Berlin, too?

Estermann:

No, the school in Berlin was not that well-equipped for science. We had strictly class recitations. There was no laboratory at all. This was a particularly good school. The teacher to whom I referred became principal of the school later, and he became, much later in his career, a French (???). He was later the head of the department of secondary education in Hamburg, so he was really one of the top educators.

Heilbron:

What was his name?

Estermann:

Doermer. He also re-edited a textbook that had been written by somebody else — which I did not find so convenient for my own use because it was organized in a very different fashion from the regular textbooks. It did not start with non-metallic elements, metallic elements and so forth; it started more with the general principles and I must confess it was sometimes a little bit above my head at that age. I think that textbook was more suitable for a university type course than for a high school course. Well, at that time, when I graduated from high school, I intended to study chemistry as a major, and also I did not think of an academic career at all; it didn’t come to my mind. More than a purely scientific career, I thought more in terms of a practical career, and I was hoping to get into industry, or so.

Heilbron:

As a pure chemist, not as a chemical engineer, or whatever —?

Estermann:

Well, the concept of chemical engineer didn’t even exist in those days; that still hardly exists in Germany. The German university system does not provide a curriculum for the chemical engineer. That’s, I think, an American invention and an American specialty. There are some courses in chemical technology, and I took some of these courses, but there was no curriculum — they were just minor courses. I went through the regular chemistry curriculum, and when it came time to choose a subject for a Ph.D. thesis, I was somewhat diverted by accident. That was 1921, and it was very difficult; by that time all the soldiers had returned to the universities. The universities were over-crowded, and you just had to go shopping for a thesis. You couldn’t really choose what you wanted to because you had to get a space to work in. In these days, physical chemistry was still a relatively underdeveloped field. And just by another accident in the previous year, I began, due to a lack of opportunity to go into the regular course that I wanted to go to, to take some physical chemistry. I had some time on my hands and I spent this taking some of the basic courses — both lecture and laboratory courses in physical chemistry. So I was, on paper, qualified to go into physical chemistry, which happened to be a very lucky accident really. So since I could not get a thesis in organic chemistry — I could not get a desk in the organic laboratory — I was able to get one in the physical chemical laboratory, and so I started to specialize in physical chemistry.

Heilbron:

Hamburg University had just opened at that time?

Estermann:

Hamburg University had opened in 1919, and I was probably among the first 10 Ph.D.’s in chemistry that finished at the University. I know I was 1/2 of the first two who finished in physical chemistry because the other fellow and I had our final exam on the same day.

Heilbron:

Was the intention to make it a strong university in chemistry, particularly?

Estermann:

Well, I don’t think there was any specific intention at that time. The University started not exactly from scratch, but it was thrown together in a hurry. Mainly the purpose was to provide additional educational facilities for the large numbers of returning soldiers. We got about four generations at once entering the universities, and so to relieve this, it was more or less improvised. There was a basis for it in existence. The state of Hamburg, which is now two, was one of the almost sovereign states of the Federal Germany. The state of Hamburg sponsored a considerable amount of scientific research already before the University existed, and it had a series of state laboratories which were, however, mainly organized in support of trade and industry. So a good deal of their function was to provide services for customs. The department of botanics, for instance, had to inspect shipments of grain that came in and certify to their classification.

The physical laboratory was handling such things as the testing of meters and the certification of thermometers. There was a large export business of scientific instruments from Hamburg, especially to South America, and the certificate of the Hamburg laboratory was considered to be a mark of highest quality, a Gold Seal. It was actually in many cases regarded by the overseas customers as more important that that of the Federal Testing Institute in Berlin, the Physikalisch-Technische Reichsanstalt. Now, this was all in existence. And in spite of the practical motivation for the establishment of these laboratories, the state government always allowed a considerable amount of the total effort to be expended on research, so there was a nucleus of research activities there. There was also a certain amount of educational work going on. It was of an academic type but without academic credit. There were series of courses given by the professors at these laboratories which were open to the public, and they were given in the evening but without any academic credit. I remember that I attended at least one of the series in physics while I was still in high school.

Then, of course, things changed very rapidly, and, well, as I said, in 1919 things were still quite haphazard. In 1921, when I began my thesis work, there were already a number of new people — new professors — who had been brought in from other places. And I would say that a not too strong effort was made to make the University into a real good one. It was mainly hampered by the reluctance of the city government, or state government to appropriate the necessary funds for buildings. So the old state laboratories became University institutes, but without a great deal of improvement of the physical plant, so we had to work there in very crowded conditions. I stayed later on on the faculty, and it took about 10 years before we got the first new building, and even that was a very small one. But for us it was luxury; we moved from 4 rooms into about 8 or9 rooms. But it took a long, long fighting. So, again, my direction into physical chemistry was more or less prompted by accident. But from there on, I think I had a somewhat more planned career. My thesis professor was Professor Volmer who later went to Berlin and then to Russia and became a rather important person in the East German scientific set-up.

Heilbron:

He collaborated with Stern, didn’t he?

Estermann:

That was during the war. During the first part of the war, the German scientists were just drafted into the army and put on regular army service, but then as the war went on, the military recognized that that was a waste of brain-power. So towards the end of the war they were all concentrated in laboratories and put on research work. And Stern and Volmer met, I think, around 1916 or ‘17 in Berlin and were working together in the institute that was then under the direction of Nernst. A good deal of my scientific education has some leads back to Nernst. Although I have never had any intimate contact with Nernst, I have seen him a few times and heard a few of his general talks at meetings and at the colloquium in Berlin, which I may have attended a few times. I happened to be in Berlin. But both Stern and Volmer had certain more direct links to Nernst, although neither of them was a student of Nernst directly. Nernst had a very strong influence on the development of physical chemistry and physics in Germany. So that goes to about the point where I had completed my formal academic work; I finished my thesis very rapidly, so that by the end of 1921 I had my Ph.D. I felt, however, that I hadn’t completed my education, and that since I had veered more or less from chemistry to physical chemistry to physics that there were big holes and gaps in it… I felt in particular that I was too weak in theoretical physics. I was then able to support myself for a year or two without requiring a job, and I decided I wanted to fill this gap.

Actually Volmer had told me that according to the rule book I wasn’t really ready for the Ph.D. yet… But he said I would probably learn much more comfortably and effectively afterwards, so that since I had a passable thesis, he would let me take the exams then; I could go and fill up the holes afterwards. And he recommended that I should go and join Stern. Stern was then professor of theoretical physics in Rostock; he had just been appointed to that post a few months before I got my Ph.D. So I arranged to work with him for awhile as an unpaid assistant. It turned out that he found that he had even money to pay me, so that was an afterthought which was not originally in the program. Well, this didn’t quite turn out the way I had intended because Stern, although he was professor of theoretical physics, was far more interested in experimenting, and he put me to work on the experimental side. Although he said that if I wanted to work on some theoretical problems he would give me some assistance, but his heart wasn’t in it. While I learned a little, I learned very little of theoretical physics during that time. So I did not accomplish what I had set out to do, but, on the other hand, I probably got far more benefit in the direction in which I had not embarked. And this is when I began the molecular beam work.

Heilbron:

Stern was completing the experiment with Gerlach at that time?

Estermann:

Yes. The Stern-Gerlach experiment was effectively completed already. Gerlach came up to Rostock for a month or so during his academic vacation just to put on some final touches and polish things up. But the Stern-Gerlach experiment was completed before Stern went to Rostock, and that was the end of his collaboration with Gerlach. That did not continue. They embarked on some other project, and this is a rather interesting one because it was an experiment that was conceived on the basis of the then known facts and the then current theories, and it turned out to be negative. It’s never been published… At that time the Bohr theory — that was 1922 — had already reached a considerable amount of refinement, especially through Sommerfeld. And according to the Bohr theory as it existed then, the hydrogen atom consisted of the nucleus with an electron running around in an orbit, so it more or less had a kind of disc-shaped symmetry. And if it had a disc-shaped symmetry, then by orienting such an atom in a magnetic field, one should expect an optical anisotropy, which would manifest itself as a bi-refringence. Now, one couldn’t make this experiment with hydrogen atoms because hydrogen atoms were then not available in any concentration under normal conditions, but Stern and Gerlach — I don’t know who of the two had the idea; I presume it was Stern, but I’m not sure of that — decided to try this experiment with sodium. Now, sodium atoms are in many respects similar to hydrogen atoms, and it was generally assumed that a sodium atom was pretty much the same as a hydrogen atom except that it has a core instead of a point nucleus.

And so they tried to find this bi-refringence of sodium vapor in a magnetic field and couldn’t find anything; the effect was not there. It was not until 1924 when the quantum mechanics developed that one understood why. According to Schrodinger’s theory, the hydrogen atom has a spherical symmetry and the magnetic moment is only due to the intrinsic magnetic moment of the electron, not to the electron motion, as it had been assumed in the Bohr theory. It turned out to be numerically the same amount of magnetic moment, but the whole concept had changed, and, of course, on that basis one should not expect to find bi-refringence in a magnetically oriented sodium vapor because only the electron spins were oriented. The rest of the atom was not influenced by the magnetic field. So this is, I think, an historically interesting event — that here an experiment was conceived that gave an unexpected result which could possibly have given a lead to the modification of theory, but it was not a strong enough experiment to really over-throw well-entrenched concepts.

Heilbron:

I’m glad you mentioned this because in Stern’s first paper describing the outline of what he had planned to do with the space quantization, he refers to the apparent lack of optical bi-refringence in gases subjected to magnetic fields as telling against the possible success of the experiment. And I’ve wanted to know just how he conceived his chances of success all the way through this experiment, whether he really believed he was going to get the effect or not.

Estermann:

No, he did not. I think from what he told me in those days — he said that he had a perfectly open mind, but he thought he had an experiment here that would give a yes or no answer. At that time, nobody really believed in the Sommerfeld space quantization, I think. Most people thought that was a rather interesting mathematical formulation, yes, but nobody thought or really believed that any such thing could exist. It is still conceptually very difficult to see, not that the space quantization exists, but how it is established. But that mechanism, at least as far as I can see, is still not explained, but the present approach to the problem is that you don’t even need to explain that. You’re not bothered by mechanisms anymore. But this was the follow-up to the Stern-Gerlach experiment. After the Stern-Gerlach experiment had a positive result, Stern was pretty confident that this bi-refringence should exist, but it did not. It was then abandoned because this was just a short-term affair — Gerlach might have been in Rostock for 4 weeks or 6 weeks; I don’t remember — it was just during the academic vacation. We worked very hard on it. Throwing certain precautions into the wind, they used a mercury arc lamp with a quartz window, and Gerlach didn’t believe in wearing protective goggles, so one morning he woke up with eyes completely swollen. It took several days before he recovered from this exposure to the ultraviolet.

Heilbron:

Did Stern have any explanation for the process in which the space quantization sets in?

Estermann:

I don’t know. I don’t think at that time, certainly. Later on, about towards the end of Hamburg there were several papers written that had to deal with this subject. This was the work which was really the foundation stone for the Rabi experiments. We’ll come to that a little later. At that time I think the Stern-Gerlach experiment was strictly an attempt to solve the puzzling theoretical prediction by an experiment which was designed to give a yes or no answer. Since the answer was a yes, the birefringence experiment was probably intended to be an experiment to supply additional support for the theory. Now, that gave a negative answer, so there was a certain contradiction between the two which was not resolved until quantum mechanics came out. But it was not published, and I don’t think that the creators of quantum mechanics even knew that this experiment had been tried; I’m quite sure they didn’t.

Heilbron:

But the experiment, at least so far as Stern and Gerlach were concerned, was definite in its result?

Estermann:

Yes; it was definite in its result, but since, well, they had other interests; and a negative result is always something that you don’t like to publish — or you hesitate to publish it much more than a positive one, because after all it is logically impossible to prove that something does not exist. I say it’s extremely improbable, but you always have the reservation: “Maybe if I’d been a little more clever, I could have observed it.” There is another point in Stern’s career which is also not so well-known in which a similar thing happened. That is connected with isotope separation. There’s a paper by Volmer and Stern — Stern and Volmer, that’s the order in which they published because it was alphabetical. It is an attempt to decide whether the observed deviation from the factor of 16 in the atomic masses of hydrogen and oxygen could be explained by isotopes — that means by the fact that there is a heavier hydrogen isotope. Now, this experiment was made by Stern and Volmer also with a negative result. And, again, it took 10 or 15 years before one could understand why the result was negative. That was strictly an experimental fact, but not a silly one. The reason was that in order to be very careful they used electrolytic hydrogen. And it so happens, as was found by Urey and Brickwedde later on, that the electrolytic process involves already a separation of isotopes, and that electrolytic hydrogen has a much smaller deuterium content than chemically produced hydrogen. So if Stern and Volmer had used ordinary hydrogen for this experiment, I think it’s almost certain that they would have found the deuterium already then. But just trying to be particularly careful and using the purest obtainable hydrogen, namely electrolytic hydrogen, thwarted that experiment.

Heilbron:

That’s a very nice story.

Estermann:

Yes. Now, Stern always told me that he was convinced that they had an effect, not on the hydrogen end, but on the oxygen end. He was convinced that they had observed a separation of oxygen isotopes and the presence of a heavy oxygen isotope. But it was just at the limit of error, and he couldn’t convince Volmer that it was a real effect, so they never said anything about it. But I think this is one of Stern’s strong points — he has always been able to grasp enough of the state of the theoretical knowledge to think of an experiment that would give a yes or no answer, a simple one. Well, in the isotope experiment they did not succeed for strictly experimental reasons, but that just emphasizes the point that I made that one has to be very careful to draw conclusions from a negative experiment. In this case, if they had drawn the conclusion that no isotopes of hydrogen or oxygen exist, it would have been a wrong, conclusion.

It was one of these cases where you had a negative result that was well-substantiated by experimental evidence but was still due to an experimental artifact, you might say. Maybe they had the same feeling with regard to the bi-refringence — that there was still something wrong with the experiment that accounted for why they didn’t find it. In any case, to the best of my knowledge, that work has never been published. Now, if I may digress a little from work at the laboratory and go to my own education. I wanted to learn theoretical physics, and I did not get too much assistance from Stern, so I did a considerable amount of reading on my own. I think the one thing which had. probably more influence than anything else in further kindling my interest was Sommerfeld’s book, Atombau und Spektrallinien, which had just appeared in the 3rd edition at about that time. This made a tremendous impression on me; I still consider this one of the best scientific books that I have ever read. It was much better than the next edition. The 4th edition came at a time when things began again to be messed up. I think the 3rd edition was written in ‘21.

Heilbron:

‘22.

Estermann:

It was published in ‘22, yes. And at that time there were, of course, many things that were not clear in the Bohr-Sommerfeld theory. Nevertheless, it was written after a period in which successes were so much more prominent than failures, that the spirit was very optimistic. One hoped that the short-comings — the impossibility to explain, in particular, the anomalous Zeeman effect which was then really the stumbling block — would be overcome somehow. One tried a number of things, and one was quite optimistic. While the short-comings could not be explained there, the whole spirit of the look was so good. Now, the next edition which appeared I think in 1923 or ’24 — but before the discovery of the spin — was already full of doubts. I mean, the difficulties were found more pronounced in that book than in the previous edition.

Heilbron:

And that reflected the general feeling of the physics community?

Estermann:

Well, I confess that’s hard to tell, but that affected me in that way. It was a somewhat more pessimistic viewpoint whether one would be able to understand and, untangle this mess. Another interesting fact that probably has been pointed out to you by others, but which I think is very remarkable is the Landé g formula. You see, that was already developed purely on the basis of analysis of experimental results, and it is a rather complicated affair. I think it’s a very important achievement. Of course, it can be derived from the spin very easily, but before you had the spin, it was really something very phenomenal. So it was a big riddle — why this formula applied and how one could get such a complicated formula. I think it’s about the same type of an achievement as the original Balmer formula where, again, you had no theory to go by and you arrived at a somewhat unusual series to represent the frequencies of the hydrogen lines… So I thought this was one of those mysteries that had come up there, and one hoped one would find a theory to explain it. But I think in ‘23, ‘24 the 4th edition of Sommerfeld’s book came out, and it reacted on me, at least, as throwing too much cold water.

Heilbron:

Well, I want to ask you about some of these events that occurred in ‘23, but before I do, there’s another point in the Stern-Gerlach experiment which, in view of what is said in Sommerfeld’s book, seems to be a possible source of trouble. And that is that they observed only two traces using the silver. So far as I can see from anything that Sommerfeld says, one should expect three, or at least cannot rule out the possibility of an undeviated beam.

Estermann:

Yes. This is quite right, but this was also cleared up with the quantum mechanics, yes, yes.

Heilbron:

True. But what did Stern think?

Estermann:

Well, Stern said, “Well, we don’t know why the zero line is not there.” I mean that puzzled him still; that was a puzzling thing that was not cleared up then. I haven’t looked at this paper for a long time, I don’t remember whether he refers to the absence of the zero line or not. The Sommerfeld theory predicted the possibility of three positions, the minus 1, plus 1, and then the zero. And the zero just was not observed, but this was something that went through the whole spectroscopy then. And it’s quite obvious because spectral lines also should not have been doublets then. This was only rectified by the spin 1/2. The half quantum numbers started to come out already then; I think the non-integer or half quantum numbers cropped up already in…

Heilbron:

Well, Landé introduced them in ‘21, I think.

Estermann:

Yes. Landé and Pauli and all — they were still groping with it. Pauli almost discovered the spin already at that time because he found that the statistics of the atoms had another factor of two there. That was in connection with his paper on para-magnetism where he mentions, that there is an unexplained factor of two in the statistical weight of the particular states. These were states in which the moment was an electronic moment, and where because of the positive and negative spin you would get a statistical weight two. And I think one thing which has maintained its validity through the whole development of the theory of quantization is that whether the quantum numbers were integers or not, the spacing was always integral. And that, I think, was already recognized before the spin was discovered.

Heilbron:

Yes. Pauli was very explicit on this.

Estermann:

Yes, but that was one of those mysteries then — where this factor of two comes from. Of course, that mitigated against the doublet in the theory, so there was a considerable amount of discussion of why the zero position, which means the position at right angles to the field, was not observable.

Heilbron:

Now, there’s a very curious thing about that, and that is that in the first paper written by Stern before he had results to offer he says that for an atom characterized by an azimuthal quantum number of n there should be 2n traces. That is what was to me something of a mystery — there’s no explanation, no reference. In view of what you say now, had he at the time he published this paper already observed sufficient…?

Estermann:

I don’t think so; I don’t think so, and I really can’t explain why he said that because it should have been, obviously, 2n + 1. Yes. Well, it may not have been; one ought to read Sommerfeld’s book again because I have a strong, well, a faint, recollection that there was a long argument about this question of whether it should be 2n or 2n + 1. And I think the evidence came really from the Zeeman effect because in the normal Zeeman effect you get 2n + 1. But in all these others — in the anomalous effects and the doublet structures — you had the feeling that it should only be 2n. But that was because of the half-integer number, because if you have half numbers, then you have only 2n, then 2n + 1 is then really 2n. I think it would be 2n on the assumption that the quantum numbers are integers, but it is 2n + 1 if you take the half-quantum numbers into account. So there was a considerable amount, I know, in the colloquium about all this; that was one of the things one talked about, but one didn’t make much headway because the evidence was confusing — partly contradictory. The normal Zeeman effects gave the 2n + 1 and the anomalous effects gave 2n on the basis of integer quantum numbers.

Heilbron:

Do you recall any of the discussion of possible solutions tendered at these colloquia?

Estermann:

No, I don’t recall anything more positive except that this was one of the things that puzzled. My general impression is that the period up to the discovery of the spin was like playing blindman’s bluff. After the first tremendous stride forward in the Bohr-Sommerfeld theory, I think the climax was the relativity correction, which, of course also turned out to be a coincidence more or less. But at that time that, I think, was the climax, and that was when? About 1920? Jim

Heilbron:

1916… But then the war came, and I think that probably delays things.

Estermann:

That delayed things, and I think in 1920 or so there was more consolidation of that part. And then came the g formula and the Stern-Gerlach effect. The Stern-Gerlach experiment had the effect of convincing people that the Sommerfeld space quantization was not completely crazy. But things did not progress very far. Then Bohr came up with the Correspondence Principle, and I think that came after the war… So that was really, I think, the major subject of discussion in the years ‘21, ‘22, ‘23. One tried to solve all the problems by the application of the Correspondence Principle. And I think there was, again, a very strong set-back to this whole development in the Bohr-Kramers-Slater attempt.

Heilbron:

What was thought of that, do you remember? Among those with whom you discussed it.

Estermann:

Well, I can just reconstruct, but I may be influenced by the things that happened much later because my time resolution is not so good anymore. I think there was more or less a conviction that one was on the wrong track — “We can’t do it that way.” The other thing that worried people very much was the attempt to understand the phase relations, the coherence of light. On the basis of the strict Einstein light quantum theory, you have no basis for coherence, and being a naive young man, I always asked the great man, “Well, how do you explain the coherence?” Lenz was then in Hamburg also. He was a very, very clever person. Although he did not publish very much, he had a very good insight, and he said, “Well, we’d better not discuss that problem; there’s no basis for a discussion of that problem yet.” So there were a number of things that were leading up to the two major events of the ‘24 period, namely, the spin which cleaned up the whole mess of the anomalous Zeeman effects and the matter waves, which resolved some of the inadequacies of the correspondence theory and put order into the other things that were not spin-related.

For instance, the isotropy of the s states of the atoms, and also the coherence. And I cannot say that all these things were perfectly clarified, those things are still not clarified, but they were at least incorporated into a unified framework in which the number of independent problems had been reduced by a very, very large factor to a few. Well, people like Einstein and Bohr, I think, were never completely satisfied with the wave mechanical concept and all this statistical interpretation of everything, but really, I think, the real critical period was l924 when all the accumulated discrepancies with the earlier theory, with the Bohr-Sommerfeld theory, were very quickly wiped off the slate, or most of them were wiped, off the slate. On the other hand, it introduced another type of difficulty which is not so noticeable among the younger physicists today because they are brought up in it, but those of us who were brought up strictly in the classical theory found it extremely difficult to attach any meaning to the new concepts — to the wave functions and matter waves. But I think Stern was, in that respect, also quite forward looking and had the right nose for it because he was one of the few who really took the matter waves seriously enough to try to find them by an experiment.

Heilbron:

When did he hear of de Broglie’s theory? Do you know?

Estermann:

Oh, he must have heard it very soon after. In those days communications were quite good.

Heilbron:

I know; but it’s a peculiar thing that except for Einstein and Schrödinger and a few others, de Broglie isn’t much referred to until the success of the wave mechanics.

Estermann:

Yes. Well, I don’t know exactly. I remember that de Broglie was invited once to come to Hamburg and give a colloquium talk; I remember that. But I don’t know whether that was before Schrodinger or after Schrödinger; this I couldn’t tell you off hand.

Heilbron:

Did he go?

Estermann:

Oh yes, yes. I remember hearing this talk; he spoke in French. Well, then in these days we had a very strong international influx. I remember Langevin was invited; of course Bohr was very frequently in Hamburg. Pauli was very frequently in Copenhagen; even in the days when it was difficult to arrange because of the German inflation. One of the side effects of this was that they had frequently access to foreign exchange there, and during the inflation a great deal of internal trading went on. They were minute amounts, but millions in marks. Pauli came back from Copenhagen and had 10 kroners left over. He had to have some German money because he wanted to eat, so he traded with a physicist who had a surplus of marks which he didn’t want to spend immediately but wanted to preserve for some future day. So the same 10 kroners might have changed hands fifty times before they finally left the small community of physicists. In those days the only principle of survival was to spend your money as quickly as you got it. We were paid about twice a week, or sometimes three times a week because of the very rapid devaluation of the mark. And if we were paid by noon, you had to spend your money before the stores closed because the next day it would be worth only half as much as it was the day before.

Heilbron:

So one tried to change it into Pauli’s excess kroners if you wanted to keep it for any length of time.

Estermann:

Yes, or anybody’s. I had some access to pounds; I had an aunt in London, so some of us had some resources. But I remember that Pauli during these days went very often to Copenhagen, and, well, I know that the Russian Frenkel, for instance, spent a year in Hamburg, and the Dutchman, Kronig, was in Hamburg. Pauli attracted a great many people as early as the early twenties. In the later twenties, Rabi was there and a number of others too. Rabi came originally not to work with Stern, but with Pauli. (???).

Heilbron:

You were about to discuss Stern’s early interest in the material waves.

Estermann:

Yes. I think Stern took it pretty seriously. Well, Stern had a hunch before, because there’s a paper that was published before de Broglie in which he stated that the attempts to make very narrow beams were not disturbed by any diffraction effects. The diffraction on a slit is, of course, the most primitive evidence for wave nature of a phenomenon. There was another thing which goes back much further. There was an old physicist in Hamburg, before the University was set up, by the name of Walther, who, after the discovery of the X-rays, tried to see whether you could not get X-ray diffraction with the slit. It was done before Laue — long before Laue. I can’t say that it did not succeed because looking at the plates that he got afterwards one sees something that might be a fringe. And very often if you have the result then, of course, you can find the effect in the noise. If you don’t have it, then it’s very difficult to find it in the noise. So this Walther experiment must have been done around 1906, or thereabouts and was faced with the same fundamental difficulty as any early attempt to discover material waves because as long as you don’t know the wave lengths, you have no idea how to fix the dimensions of your experiment. Maybe the slits are too fine, or the slits are too wide; you don’t know.

As soon as you have the wave lengths, then you know where to look. Of course, very few people knew about this Walther experiment; that had been completely forgotten. It was published I think in a local journal; there was a Naturwissenschaftliche Gesellschaft in Hamburg and a Naturwissenschaftliche Verein in Hamburg, and I think this Walther paper was published in the Proceedings of that Verein. So it probably wasn’t read outside of Hamburg at all. But, of course, we in Hamburg knew about it because Walther was still in the laboratory when we came. And it may be that this coincidence — that Walther tried to find a slit diffraction of X-rays — probably prompted Stern to try to find whether molecular beams wouldn’t also have wave properties. And by making the slits narrower and narrower, assuming that the wave lengths would be very small, he thought that one might find it. And that might have prompted him to put the sentence into the paper indicating that the experiments of making very narrow beams using very fine slits were not disturbed by any wave diffraction.

Heilbron:

That was before de Broglie?

Estermann:

That was before de Broglie, yes; well, not too long before, but before. So when de Broglie came, he picked the idea up again, and then of course you knew what the wave lengths were going to be, and it became immediately evident that you wouldn’t have much luck by using mechanical slits and that you have to use atomic gratings just as in the case of X-rays. He immediately set to work trying to find diffraction from crystal sources.

Heilbron:

But these were with molecular beams?

Estermann:

With beams, yes, with molecular beams. Now, I think he started at least as early as Davisson and Germer started with electron beams, but the technology was far more difficult, and the (further) results took considerably longer to show than in the case of the electron beams. But the diffraction work was started immediately after the de Broglie theory became known. It must have been very, very shortly thereafter.

Heilbron:

Was there any intent ever to do it with electrons?

Estermann:

No, no, he never did anything with electrons. As a matter of fact, Stern didn’t like to use any electronic equipment, even when it became available. It took a long, long time before one could convince him that an electronic amplifier had certain merits.

Heilbron:

What were his main objections?

Estermann:

He thought it was too complicated. No, we didn’t have any technology or any experience in working with electrons. I tried one thing with electrons then which has now become practical, but at that time it was not, for a lack of adequate technique. Namely this. One of the big problems in molecular beam technology was the (detector) end; we couldn’t detect very well. At the beginning the only thing we had was condensation, and then we would try to measure the density in the (???). I did a considerable amount of work on that which showed why that is a bad method. But it did not give much lead on how to make it good — a little, but not very much. Then, I think, the great step forward was the Langmuir-Taylor detector which was worked out when Taylor was a visiting scientist in Hamburg. But that works only for a limited number of elements or substances. Then we tried all kinds of things, and one of the things that I tried is what is now known as the cross-fire method; it means to bombard the neutral atoms with electrons, thus ionizing them, and then collect the ions. Ions are far easier to detect than the neutral particles. But I did not succeed; I did not get it to work. It’s a method which is now used in a number of places quite successfully, but it required much better vacuum technology and much better electronic technology than was available in those days. This must have been about 1923 or ‘24 when I tried this.

Heilbron:

What were the particular experimental difficulties in the diffraction experiments with the molecular beams?

Estermann:

Well, I think the major difficulty —. It’s hard to say; the first thing is that one didn’t know certain things. I think we were also, for a long time, on the wrong track because we were first using not very suitable crystal surfaces — we used sodium chloride which is the most common one — although we finally found it. But I think the main problem was again a very simple one, but one that one could not foresee, and that is this: we had always assumed that the best position of the gratings, the best orientation, would be where the flat side of the beam would be parallel to the crystallographic axis. And, therefore, all the experiments were made with that orientation, and one looked for a diffraction in the same plane. It turned out, somewhat by accident, but really by more careful observation of what happened and by trying to vary all the parameters, that the best orientation is 45 degrees off this direction. And this was then substantiated by the observation that the lattice which provides the diffraction is not the lattice of ions, but the lattice of ions of the same sign. So when we turned the crystal by degrees, we got far more intense and much better resolvable diffractions than before.

But that took several years before one got to that point. But in those days diffraction technology was quite difficult; the detection was difficult because we had to use beams of helium and hydrogen because all the other materials had far too short waves, and the detection of that had to be developed. Now, at the same time Tom Johnson was then at the Bartol Foundation and was working on diffraction. He had the same idea after de Broglie, but he wanted to work with hydrogen atoms. Now he was close to Baltimore where B. W. Wood had developed the method of making large concentrations of hydrogen atoms, so he made a beam of hydrogen atom, and he looked for diffraction. And he didn’t find it for the same reason, because he had his crystal in the wrong orientation. He told me later that when he read our paper we didn’t know each other at that time, but I met him three years later — he went back to the laboratory — it was in the evening — and he turned the crystal by 45 degrees, and before morning he had the diffraction! So this is again one of those things where one has a very reasonable assumption, but for one reason or another it does not work.

Heilbron:

Had you ever any doubts that it would appear?

Estermann:

No. I don’t think we had any doubt; otherwise we would have given up long before; because we worked at least five years before we had the first I wasn’t in it all the time, but…

Heilbron:

Well, how long would you say you worked before the knowledge of the orientation would have made all the difference? Could it have come two years earlier if you had rotated the crystal?

Estermann:

I don’t think it was that much because we really observed the maxima even with the wrong orientation. But they were so much weaker that… But it would have shortened it by a considerable amount. Also, we really went into business, so to speak, when we changed from a sodium chloride to a lithium chloride crystal; that made even more difference. Sodium chloride was a not very suitable material, but it was the easiest crystal that was available. I don’t remember how we got to lithium chloride; we just tried in despair to use something else, and we went for any kind. I think, (Kiropolis), who was then in Gottingen and is now at Cal. Tech, had developed a method of growing crystals. And, I think, through our connections with Gottingen we got some of his crystals, and among them were lithium chloride crystals. And we tried them all, and we found that there were other things in the… One thing, I think, which held us up was, again, a certain naive approach. We were thinking in terms of line gratings, and actually the phenomenon is the cross-grating effect which I think had never been observed elsewhere in physics.

Now, the electron diffraction that was observed was not cross grating. The electron diffraction was Bragg-type, that is, stacked planes. The electrons, just like X-rays, penetrate and you get the interference from different layers. But in the molecular beam experiments you have no penetration and you get the interference only from the surface lattice, and that is a 2-dimensional lattice. Nobody had ever sat down to work out the very simple theory of the, diffraction by a 2-dimensional lattice — at least we never had done it. We were always looking at the place where the one-dimensional lattice would give us results. We were not looking at the right place because we did not recognize first that it was a 2-dimensional lattice. And after we recognized that, we did not recognize that it was the lattice of ions of the same sign which had to be considered, so, we looked again at the wrong place.

By looking everywhere we finally found a little peak, and once we got the peak by the tail then by a systematic variation of all the parameters you very soon found how to get a good result, an intense signal. It is always the most difficult thing to find the first trace of the effect; once you have that then it becomes relatively easy because then you go from random variation to systematic variation, and that is a far more effective approach. Well, this is about what I can remember of those days. For my own personal development of interest the most important points would have been, I would say, the Sommerfeld book, which came into my hands at the same time that I started to work with Stern; then this period in Hamburg when things were just in a confused state. Then at least the problems were isolated and really better defined. It was first something like a very rough sea, but then later the individual problems started to become identifiable. Then there was this sudden resolution of all the difficulties when these three events came up: the spin, the wave, and the matrix, and the statistical theory which I think was accepted in Hamburg very quickly (while other people didn’t want to hear about it.)

Heilbron:

And Stern accepted that quickly with no —?

Estermann:

Yes, Stern accepted that very quickly. I think he was really very soon convinced of the reality of that.

Heilbron:

Well, there were other aspects of work in the Hamburg Laboratory, like the magnetic moments of protons, and so forth.

Estermann:

Well, that started much later. That started only after the diffraction was put to bed. The Hamburg Laboratory was a small laboratory, and the grown-up people were usually only working on one problem at a time; there were not enough. And some of the graduate students were then doing some clean-ups and subsidiary problems. For instance, after the Stern-Gerlach experiment and the measurement of the first atomic moments by the beam method was developed, students worked on the hydrogen atom; some of them worked on various other alkalis. There was an American, (Lewis) who was then there. He made a rather nice thesis utilizing the fact that alkali metal vapors contain diatonic, Na₂, molecules, and that these are non-magnetic. He determined the chemical equilibrium between, let’s say, sodium atoms and Na₂ molecules by a beam method because atoms are being deflected in the magnetic field, and the molecules go straight through. So by measuring the intensity ratio between the center and the deflected beams, one can get the proportion and by varying the temperature of the oven from which the mixture escapes one can then get the chemical equilibrium constant. So these were all more limited problems that were worked on by the graduate students because it has been our policy, which I still maintain, that it is unfair to give a graduate student a problem when you cannot foresee any result in any reasonable time. So these things like the diffraction or the proton moment were considered to be not suitable for thesis problems because one could not estimate whether one would ever get a result, and if so, how long it would take before one would get it.

Heilbron:

Did Stern determine almost all the activities of the laboratory?

Estermann:

Oh, yes. It was very small, you see; first it was Stern and two assistants — I was one of them. And then it finally added up to four, and there were never more than maybe six or eight others in the laboratory, if that many. There were, maybe, one or two research fellows from abroad and maybe four graduate students. I don’t think we ever had more than four graduate students at any one time.

Heilbron:

And did the funds come entirely from the University?

Estermann:

No, we had then some support from what became later the Notgemeinschaft Deutscher Wissenschaften, which is now Deutsche Forschungsgemeinschaft; I think that’s where we got some funds. But most of the funds, I think, came from the University.

Heilbron:

Were they sufficient to provide the equipment necessary, or was one condemned to roughing it?

Estermann:

Oh, we had to rough it, I think. But we were perfectly happy with that situation; that was before the days when physicists didn’t feel important if they didn’t spend millions.

Heilbron:

But you never felt the work was hampered for lack of funds?

Estermann:

No, not since we got the new building, at least. But except for the proton moment, all the other work was done before we had the new building. The new building was completed, I think, in 1929, or 1930. We had only three more years.

Heilbron:

When Stern first began to use his beam methods, apparently there was very little competition.

Estermann:

There was practically no competition at all except for competition that was created in the same school. There were very few outsiders who engaged in molecular beam work until — really until the war, practically. I think the only independent groups — not counting Rabi as an independent one, or Fraser, who did a little work here. Now, here is somebody; have you had any contact with Fraser?

Heilbron:

No.

Estermann:

Well, I don’t know how much he can contribute to something outside of molecular beams, but he was in Hamburg for about two years, I think. He started then a little work, here in Cambridge, later on. Well, I think one project or problem was investigated by Born, but even Born probably was influenced by Stern because Born was in Frankfurt at the same time when the Stern-Gerlach experiment started. But that remained a single event; even Gerlach didn’t continue.

Heilbron:

Why do you think that happened?

Estermann:

It was an extremely difficult technique; it was far too difficult for most people to bother with because there were other things where you could get results with much less effort and much less time. Stern really had the tenacity of a bull-dog when it came to this kind of experimenting.

Heilbron:

We’ve experienced some of his tenacity ourselves.

Estermann:

He could sit in the laboratory, and when an experiment didn’t want to go, he wouldn’t give up. Well, he had no other interests in life practically. He would sit until 1:00 or 2:00, or 3:00 in the morning; it didn’t matter to him at all; he wouldn't go out for dinner, he would bring an apple to the laboratory, and that was his dinner. And it was hard on the younger ones, especially those of us who were married. I think I was the only married one in the laboratory in those days.

Heilbron:

Did he expect the same attendance from his students?

Estermann:

Of course, oh sure. Now, I lived a little bit further out, so I couldn’t walk home. I was about three or four miles from the lab, so I had to quit with the last trolley, or the last sub-way later on, which meant about 1:00 or 1:30, but that gave at least a natural limit.

Heilbron:

When would activities begin in the morning?

Estermann:

Well, Stern didn’t come until noon, but we had other duties, so I would usually come around 8:30 or 9:00.

Heilbron:

It sounds arduous.

Estermann:

Now, if I see the way people work here in England, it seems arduous. Now, the diffraction experiment was then refined by the addition of the velocity selector, but I think that is all sufficiently known and there were no real problems except vacuum problems — they were really the only ones. Today one could practically make this a student laboratory experiment. In those days vacuum technology was very far behind, so that was really the major experimental problem.

Heilbron:

Weren’t there also some thermal-velocity distribution experiments made?

Estermann:

Yes, yes. Well, really, Stern’s first molecular beam experiment was the thermal velocity… He returned to that, but only very briefly. Then we went into that again because it turned out to be important to know the thermal velocity distribution for the analysis of some of the other experiments. But it became not an end in itself; it became strictly a subsidiary, supporting activity. That was not until Pittsburgh that we took that up again. Then that was not with Stern anymore; I picked it up later in connection with some of the energy exchange problems to see what the accumulation coefficient was like. I would say that had nothing to do with quantum physics in any form.

Heilbron:

What were the publishing policies of the laboratory? Did Stern just approve things that went out as such?

Estermann:

Yes. Stern very rarely put his name on the papers that were published by his more advanced graduate students, as a matter of fact. Practically all the theses were published by the student alone, just with a note somewhere acknowledging the assistance or inspiration or what-not of Stern. The papers or work that was done jointly with some of the grown-up people was published then as a joint paper.

Heilbron:

What were thought of the newer theories involving new particles such as the Dirac theory? What did people think about the neutrino, for instance?

Estermann:

Oh yes. The neutrino was, of course, invented by Pauli rather early in the game, but it was more a deus ex machina, you know. I don’t know how seriously one took that. There was one molecular beam experiment that had something to do with nuclear theory already before the proton moment was made, and that was, I think, the magnetic moment of lithium. I may be wrong; you would have to look up the papers, because that was prompted by some theory which had given the nucleus a large moment. This was before nuclear moments were really clearly discovered. I think the only effect of the Dirac theory was probably the proton moment measurement. And the only thing I can say to that is that the theoreticians were at that time so convinced that they knew what the outcome was that even people like Pauli would condescendingly come to the lab and shake his head and say, “Well, well it’s very nice that you’re doing this experiment, but you’re really doing unnecessary work because we know what answer is going to be.”

Heilbron:

What did Pauli say after the results of it?

Estermann:

I don’t know really. When the results were really fortified, Pauli was no longer in Hamburg; Pauli had left. But this again took a long time. I think the proton moment work was started, I think, in 1931. It didn’t take too awfully long; the first results were out within two years — the significant ones. All the rest was then refined. That’s another case where one came close to the goal rather quickly, and then getting the next decimal took an inordinate amount of labor and was never quite successful. And again it was a case where the theoreticians had made several wrong predictions which they rectified in part after the results were known. If I am permitted to be a little bit catty, I would say that the theoreticians are best at getting results when the experiments have already been performed. I know a great many other instances where that happened.

Heilbron:

Can you tell us some of the unsuccessful investigations?

Estermann:

I don’t think there were any really unsuccessful ones. Well, they were all unsuccessful attempts in the early stages, but I think they were all gradually overcome.

Heilbron:

One didn’t give up?

Estermann:

No, I don’t think that we gave up on anything of fundamental importance. Well, I gave up, for instance, on the detection method because that was something that was not really a research problem; it was a problem of design to improve technology and make experimenting easier. But, when it turned out not to make it easier, it was abandoned. I think in that respect we probably tried several things from time to time which were not pursued. There was one thing, however, in my earlier days, when I was still in Rostock, which we gave up where it became clear later why it wasn’t successful. This was on the same plane as the bi-refringence. Now, one of the early results of the Bohr theory was that an atom with unpaired electrons would have a magnetic moment, and the unit of atomic magnetic moment which was then known as the Bohr-magneton was discovered. Now, at the same time, or earlier, a man by the name of Weiss in Strasbourg made a great many experiments on magnetic susceptibility, and he found that they were multiples of a unit which is called the Weiss-magneton and which happened to be approximately one-fifth of the Bohr-magneton.

Now, this was, as we know now, I think, completely spurious. But looking the data over one found that there was a much higher probability for atoms or ions to have numbers of Weiss-magnetons which arc divisible by 5 than any other number. So 5, 10, 15, 20 and so on occurred more frequently than others like 11, or 12, or some other number. So it was rather a naive conclusion to say that these multiples of 5 are really 1, 2, 3 Bohr-magnetons. And the others which would then come out as fractions of Bohr-magnetons had something wrong with the measurements or with the definition of the material. And Stern suggested to me, since I had come in as a chemist, whether there might be something wrong with the chemistry for these measurements that were made on salts and compounds. And we tried to see whether by using more refined salts one would not squeeze these odd numbers out and get full magnetons. Well, we had some success in some cases, but the most stubborn culprit remained the nickel, which had about 12 magnetons, and we couldn’t budge it. We couldn’t bring it to either 10 or 15 — it just didn’t work. And during my Rostock days, I made a considerable number of magnetic susceptibility measurements, but, again, we never published it because it was not a positive result of any sort. The fraction of magnetons were explained also only after the spin came; that was 1922, so at this time we didn’t have any idea of this.

Heilbron:

Was there any idea of the connection between these magnetic problems and the problems of the optical spectra? Did you recognize that they were connected?

Estermann:

Oh, yes; I think that was already connected. That was probably recognized; yes, I’m quite sure it was.

Heilbron:

How did you happen to set up shop again at Carnegie Institute?

Estermann:

Well, in 1933 it became pretty obvious that the days were numbered. Now, I was away during the period from spring 1931 to November, 1932; I was in Berkeley then. And even before I came back I knew I was not going to live to a long old age in Germany, and I had decided to get out. But I was not successful in lining up a position for myself before I went back. I stopped in England on my way back and was successful in arranging at least a temporary appointment here in England, so I was determined to get out as soon as possible. I would have gotten out even earlier; I sent my family out as early as April or May 1933 — to England. I had a brother here; my brother had preceded me by several years not for political reasons, but for other reasons he went to England. So I had some kind of a mooring point. Now, Stern didn’t want to go; he thought that, well, he could survive Nazism in Germany. But he became convinced in June, or so, that it wouldn’t work either. So he turned in his resignation. But we were then right in the middle of the proton-deuteron experiment, and decided as long as we would be left alone we would continue this work. And we worked until August; then we finally quit. Several months before — whether it was June or July, I don’t know — the then President of Carnegie Tech, a certain Dr. Baker, made a trip to Germany to try to find some good scientists who might be induced to come to Carnegie. One of them was (Bell), the chemist, chemical engineer now we would say, and another, one was Stern, and the third one was myself. So he made arrangements with us to come to Carnegie. So I never started this job in England that I had already lined up for myself.

Heilbron:

There was no question in your mind that the laboratory would be permanently transported at that time?

Estermann:

Yes, I had no thought of ever going back to Germany. No, I didn’t consider that a temporary affair at all, and we actually took a considerable part of the equipment with us.

Heilbron:

How did you manage to do that?

Estermann:

Oh, I don’t think there was any problem with this. I mean, this was not commercial equipment — that was duplicated. We got authorization from Carnegie to buy the same kind of a magnet and pumps and so forth, and they were shipped to Pittsburgh by the manufacturer who duplicated the ones that we had had in Hamburg so that we could re-establish the apparatus. And the parts that were made specifically for the purpose in the local shop I think we got permission from the University authorities to take along.

Heilbron:

You say in your introduction to Atom Beam Research that the momentum of the Hamburg lab was never regained, but you give no indication why.

Estermann:

Well, I thought it was pretty obvious; the environment was not the right environment.

Heilbron:

Well, was the source of students not so good?

Estermann:

No, it was not only the source of students; there was no response on the faculty. The head of the department was very unsympathetic.

Heilbron:

Did that make a difference in the funds?

Estermann:

Yes, in part. The Dean was not a very easy person to deal with. I think the major reason was probably that the president made this whole arrangement on his own without consulting his subordinates. But on the next level, he had to fight things through, and after one year he became ill and was ill for several years until finally he resigned. So there was no support from the top after the first year anymore. Stern was something of a prima donna, as you have probably noticed. If things didn’t come his way he would retire into his (corner), and pick up his marbles and go home, so to speak; which made life even more difficult. His whole personality is not suited to an American University.

Heilbron:

Well, at Hamburg the laboratory was more or less autonomous, wasn’t it?

Estermann:

Yes. In Germany, as you probably know, each department is quite autonomous; every full professor is the director of his institute, and he has his own funds. He may have to battle with the University authorities, or with the local government for funds, or he may get funds from other sources, but once he has those, he is completely in command. Which has certain advantages, but also certain disadvantages. Originally it was set up to be the same way in Pittsburgh, but when Baker became ill, this was somewhat abrogated by the deans. This was one of the things which Stern never quite survived — in an emotional sense. Well, he never liked Pittsburgh; he never liked the whole atmosphere.

Heilbron:

Was that problem of adaptation a general one? Do you know that many European scientists who came there had the same experience?

Estermann:

Well, some had it and others not. It might have depended somewhat on the age. (Bell) adapted himself quite well; he was much more diplomatic; he knew how to deal with people whom he didn’t like. Whereas Stern would make it completely obvious if he didn’t like the dean. There was no question about it. Of course, that made matters somewhat more difficult. I don’t think it was so much funds, but, again, it’s difficult. You see, it took about 10 years to get the laboratory in Hamburg to the point where it was really a productive concern. Stern came to Hamburg in 1923, and it was just about at its peak when we left. It was a hard struggle there in Pittsburgh and then he was ten years older and didn’t have the energy anymore, or you might say the emotional stamina, to go through it again. He also had achieved his major successes by then, and probably became even more demanding for that reason. There was probably nobody in the physics department at Carnegie Tech who had ever heard of him before, or heard of anything of modern physics before.

Heilbron:

How benighted were things in the physics department — curriculum and so on?

Estermann:

Well, the under-graduate curriculum, I think, was not bad, but there was practically no research going on. There were no Ph.D. programs running; there were a few night-school courses leading to a Masters degree, and that was all. The first full-time day graduate students came in 1934 — the year after we came to Carnegie. The first year Stern and I worked all by ourselves; all we had was a machinist. Then we hired one other assistant, (Simpson), who is now at Argonne. And in 1934 we got three graduate students started out, but then things immediately flatted out because there was no support from the top coming anymore. Now, there were lots of disagreeable shenanigans, too. The building was unsuitable; there was a great deal of difficulty working, in it. It was not a laboratory, it was one of these big buildings which house half a dozen departments, including the electrical engineers. One thing that held us up for months was some spurious disturbances which occurred from time to time and disappeared again.

We tried to track them down and finally, after a very diligent examination of every possibility, we found out that the students had a radio-amateur club in the same place. When they went on the air, our instruments went hay-wire because their aerial was coming down right in front of our window where our lab was. That took months to locate, and then, of course, after we had found that, we quickly reached agreement that they would go off the air when we were in a critical stage of taking data. But little things like this can spoil your fun. I didn’t find, it so obnoxious, but Stern, I think, never, never felt at home there. But even I finally had to give up, so I left because I felt it was not a very desirable atmosphere. Like many of these universities there is a lot of local politics going on. Off the record — I think Berkeley made a great mistake by not inviting Stern, when he came to settle there, to join the University and be on the faculty. They could have done it even without pay because he didn’t need the money, but they almost completely ignored the fact that, he was there, and didn’t make him any offer. So I think that must have hit him pretty hard.

Heilbron:

I’ve seen him at very few meetings.

Estermann:

Well, there were some other unfortunate occurrences after Stern left. The first thing was that Gilbert Lewis died; he and Stern hit it off very well when he was there as a visiting professor before. Then there was this whole atom bomb controversy and the loyalty to his country versus —; this did not go very well with him either. And then there was a personal tragic event; his younger sister, to whom he was very much attached, died of cancer just when he had decided to move to Berkeley.

Heilbron:

He has another sister there?

Estermann:

He has another sister there, yes. But this younger sister was, oh, ten or fifteen years younger than he was and was more or less his pet, so to speak. She was a very nice girl; I knew her quite well. She was probably in her 30’s when she contracted a very, very fast developing cancer just when he had decided to move to Berkeley. So he started out with all the cards stacked against him there.

Heilbron:

Well, that makes things a lot clearer.

Estermann:

Well, of the other independent molecular beam groups, the only one of any strength that I know of was the one in Iowa, but that petered out completely; I think there’s nothing left. One or two of the fellows there went to Columbia later on and worked with Rabi for a while. Rabi started out at Columbia. I don’t think he had such an easy-going time in the first years either, but he had a different approach to things, and he slugged himself through there. But I think it required a little bit of slugging. But this you probably have more information about. Did you get some time with Rabi?

Heilbron:

We intend to, when we go back to the States.

Estermann:

Well, you might want to see Jensen. And Kopfermann might also be able to help you. Kopfermann was in Gottingen I think in those days… There is one old fellow in Vienna who is still alive and pretty lively, Przibram. Well, we have covered most everything in your letter except for the items such as the Fermi-Dirac statistics, the Dirac theory of the electron and the “hole” and the neutron and positron and neutrino. Well, I was not very much connected with this part here. The mathematics went over my head, and although I attended a number of seminar sessions in which these things were discussed, I must confess I didn’t get very much out of it. So I cannot contribute very much. I had the impression that most people, with the exception of probably Pauli and Lenz, were somewhat in the same boat. I think the discussion was mainly between those two, and the others had very little to say. Well, there’s one man from that time who is also still in Hamburg, and that is Jordan.

Heilbron:

I wanted to ask you one thing — to change the subject completely — and that is what were the mechanics of obtaining the Rockefeller Fellowships?

Estermann:

Well, just like anything else — you made an application, and you got a few people to support it, and you waited for a year. Then you were interviewed by one of their representatives. I was interviewed by Professor (Hansen), who was then the head of an office in Paris; I think Dr. Tisdale was then the head of this office, and (Hansen) was his assistant. I remember he came to Hamburg and interviewed me, and then a few months later I got the Grant.

Heilbron:

Were there a fixed number available, say, for each country?

Estermann:

This I don’t know, but the number was relatively snail. I think in physics there were never probably more than two or three from Germany at any one time.

Heilbron:

So the competition was keen.

Estermann:

Well, I’m not sure whether the competition was too keen because there were not so many physicists around either as today. At Hamburg, I think, I was the only one in physics. (Von Hoppel) was there earlier, Stuart, who is now in Mainz, was earlier there; he was immediately preceding me; we met in Berkeley. I’m not sure whether (Braider) had a Rockefeller Fellowship, or some other fellowship. But there were very few people who were anxious even to apply for one because it involved an interruption of their career and so forth. It required (???).

Heilbron:

Were they just grants, or were they for specified lengths of time, or… ?

Estermann:

They were normally for a year, and I got a four months extension. Since travel time was not included, my total absence was about 19 months. I stopped in England on the way out and on the way back and made a number of visits between the East and the West coasts. So I left Hamburg in March, ‘31, and cane back in November of 1932.

Heilbron:

Was that the first time you had visited England?

Estermann:

Not England, but the United States. No, I had visited England once before in l924. I spent about 10 days here.

Heilbron:

The language was…

Estermann:

The language didn’t bother me very much because in Hamburg the study of English was very strongly pushed. Hamburg has always considered itself more closely related to London than to Berlin. At least among those people who counted there.

Heilbron:

Well, we’d be grateful for anything more you’d care to say.

Estermann:

Well, I think this is probably all I can say now. But I would like to leave the possibility open that after you have digested this and maybe some of the other things, then if you come up with specific gaps that you would like to fill, we could have another session. Now, I’ll be back in four months and you will still be operating, I think, from what Professor Kuhn wrote me. And I hope to come up to Copenhagen sometime next summer. There is a distinct probability that I will be participating in a NATO symposium — one of these NATO summer schools, these two week affairs — sometime in July, and it’s very likely that this will be held in Denmark. This is something that is not yet completely determined, but ought to be clarified by the time I get back.