Notice: We are in the process of migrating Oral History Interview metadata to this new version of our website.
During this migration, the following fields associated with interviews may be incomplete: Institutions, Additional Persons, and Subjects. Our Browse Subjects feature is also affected by this migration.
Please contact [email protected] with any feedback.
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.
In footnotes or endnotes please cite AIP interviews like this:
Interview of Bengt Strömgren by Karl Hufbauer on 1978 April 24, Niels Bohr Library & Archives, American Institute of Physics, College Park, MD USA, www.aip.org/history-programs/niels-bohr-library/oral-histories/4907
For multiple citations, "AIP" is the preferred abbreviation for the location.
Research on stellar energy in the 1930s. Eddington-Milne models controversy encourages stellar structure work. Switch from classical astronomy to astrophysics during graduate studies (Niels Bohr); contact with physicists (George Gamow, Lev Landau). Background of his review articles. Discussion of Gamow's theory; comparison of Hans Bethe and Carl Friedrich Weizsacker's work on nuclear reactions in stellar interiors; comments that energy production problem was solved by physicists.
Today is the 24th of April, 1978. I'll be interviewing Professor Bengt Strömgren in his office at the old Copenhagen Observatory. Because Drs. Hoddeson and Baym have already conducted an extensive interview with Professor Strömgren for the American Institute of Physics, I will concentrate on topics that are related to the history of research on the stellar-energy problem during the 1930s. It is my impression, Professor Strömgren, that you played a major role in work on this problem from 1936 to 1938. I am thinking of your review articles in the HANDBUCH DER ASTROPHYSIK, Vol. VII, 1936, and in ERGEBNISSE DER EXAKTEN NATURWISSENSCHAFTEN 1937. These articles provided up-to-date statements of the stellar-energy problem. I am also thinking of your contacts with Weizsäcker and Bethe, which are acknowledged in their key papers on the problem. In our discussion today I would like first to consider the background to developments of 1936-38, and then those developments. My first question, Professor Strömgren, bears on your initial contact with the stellar-energy problem. How did you learn that there was such a problem?
Actually, through the papers of Eddington and Milne. There was in the late twenties and early thirties discussion between Eddington and Milne on the question of the energy production in the sun and the stars. Milne's point was that the Eddington model, even for the sun, would not lead to sufficiently great energy production, and that therefore, one must look for quite different models, with higher temperatures. The idea Milne had was that you could start a model from the outside, integrate toward the center, and deviate from the Eddington model. In Milne's model it seemed that, as you approached the center, mass was left over, so that the central boundary condition on the mass was not fulfilled. However, Milne's idea was that perhaps under those changed conditions, the equation of state, would be radically different—that it would allow compressability that removed the discrepancy and made it possible to fulfill the conditions.
The starting point, then, was that the first calculations of the rate of the processes seemed to show that the temperatures as found on the Eddington model were not high enough for any appreciable production of energy, and the idea Milne had was that this would be remedied by this kind of model. A number of investigations were carried out by Chandrasekhar, by myself and by others, in the late twenties and early thirties, to see if Milne's ideas could be correct. And the conclusion—in fact, I think Milne's own conclusion also in the end — was that this kind of model didn't really exist, for the sun and for Main Sequence stars. I remember how I myself, when I started, had some hope that you could start the way Milne started, and end with a White Dwarf embedded in the center. But when I followed this up—two publications on the subject, one in MONTHLY NOTICES, one in ZEITSCHRIFT FÜR ASTROPHYSIK—then it turned out that the compressibility that would be necessary didn't occur. When one looks at all that has happened since, one sees that the nearest you come to the kind of model that Milne had hoped for is in the late stages of Giant stars, where you do have something approaching a neutron star at the center of the star. But the whole idea proved to be irrelevant to the Main Sequence problem.
When did you first read Eddington's INTERNAL CONSTITUTION OF THE STARS, which was published in 1926? Freundlich's review article that same year in the ERGEBNISSE DER EXAKTEN NATURWISSENSCHAFTEN? And then Russell, Dugan and Stewart in their book on astronomy, in the second volume which was mainly by Russell, had several chapters on the energy problem. Did you read these works as they came out around 1926-27?
Well, I started reading these works just a few years after they appeared. In 1925, 1926, and 1927, my main interest was in classical astronomy. It was only after I took my Master's degree—as you know from the previous interview, Niels Bohr and Hanson were my teachers in physics—that I turned to astrophysics.
Yes. Looking somewhat more closely at the 1930-31 period, I notice that four people of about the same age—Biermann in Germany, Chandrasekhar and Cowling in England, and you—all began publishing on stellar-structure problems. I would guess that the controversy between Eddington and Milne certainly encouraged this kind of work.
Yes, indeed. You mentioned Cowling. It's interesting that following the first paper by Milne with the suggestion that there might be centrally condensed solutions there appeared three papers that showed this was not so. The first was by myself, the second by Cowling, and the third by Henry Norris Russell. Now, as there weren't so many people working in theoretical astrophysics in those days, and as you have mentioned also Chandrasekhar, and Biermann—Biermann's great contribution was his work on the outer convective zone, and in general, on convection in stellar interiors. Through the thirties the picture regarding the role of convection became clearer. First, the important step with regard to the interior was Cowling's model of the convective core for Main Sequence stars. But Biermann's ideas on the outer convective zone gradually turned out to be correct. I recall that in the thirties, he had many discussions with Unsöld on the question, and found it hard to convince him. Admittedly, the theory for the convective transport was a rough one. The whole situation regarding the outer convective zone was only satisfied really after Wildt's discovery of the role of H- . It was only then that it became clear that the outer convective zone starts, not where hydrogen is beginning to be substantionized, but where there is a sharp increase in continuous absorption, due to the effect of H-. In other words, because of the H- effect, it takes only a small amount of ionization of hydrogen to produce the instability to convection, that means the start of the outer convective zone.
The whole discussion went on, and when you look at the literature around 1950, it's clear that the picture of the deep outer convective zone had been accepted. When Biermann made the first calculations, there was a difficulty about the deep part of the convective zone. It was clear that there is lithium in the solar atmosphere. That wasn't definitely shown in the normal solar spectral lines, although it was suggested. But there certainly was lithium in the sunspots, so there must be lithium left. And if the Biermann model of the outer convective zone was correct, the temperature at the bottom of this zone was too high, so... Now, it was only when calculations were repeated, with what we now regard as the correct atmospheric and stellar compositions, including helium, and with much less heavy elements than in the early speculations, that the temperature came out right. That isn't so high that... Now, with regard to Chandrasekhar's contributions, I recall from those early years, early thirties in Copenhagen, Chandrasekhar worked here, at the Niels Bohr Institute, and finished his pioneer work on the White Dwarfs.
I'm curious to know why you started doing theoretical astrophysics, yet did not shift all the way into theoretical astrophysics? Why did you keep up the classical astronomy side?
Well, after I got my Master's degree in 1927, I continued some work in classical astronomy. I wrote one article on parabolic orbits of comets and one on perturbation methods of elements of asteroids. Then they were published, as was my Ph.D. thesis; that was the end of my work in celestial mechanics. I was interested in the work that was going on there, but I essentially stopped doing such work myself.
When would you say that your chief interest became theoretical astrophysics?
During my studies for the Master's degree, meeting regularly at the Niels Bohr Institute, I was very much impressed with what was going on there. I had the idea that the time was ripe for applications of the new quantum mechanics to astrophysical situations. Indeed, following the publications on the Milne model, there were two publications on the hydrogen content. The program for that was suggested by the situation described by Eddington in INTERNAL CONSTITUTION OF THE STARS. He was quite clearly shows that there is discrepancy between what he called the astronomical opacity, and the theoretically calculated opacity. Now, Eddington himself still hoped at that time that revised calculations for the theoretical absorption coefficients would lead to higher values, and that the problem would be resolved. The reason he was reluctant to give up the idea of the composition that led to very high molecular weights—that is, a little higher than 2—was that it's only with these molecular weights that radiation pressure plays a conspicuous role. This was an important point to him. He thought of the stars as celestial bodies which had been arranged just so that radiation pressure plays a role. I have a letter from Eddington—maybe you would like to see it?
Yes. I would be curious to see it.
I don't have a copy with me here, but I can send it to you, [a letter] which I received when my first paper on the hydrogen content and the opacity was published. He there refers to calculations of his own that check with mine, published in the ZEITSCHRIFT FUR ASTROPHYSIK, and in the end, he refers to his reluctance to give up the model of radiation pressure. So it's quite clear from that letter that this is what held him up.
I would certainly like to see it.
Surely. I'll send you a copy.
As you mentioned just now, thanks to your new endeavors in stellar constitution and your growing reputation within astronomy, you developed new contacts with people throughout the astronomical community, especially the theoretical side of it. And I wondered if you could say something about your personal contacts with several people whose names I'll give, during the period 1930-36, before you went to America. I'll just run through them quickly and then we'll proceed more slowly. I was thinking of such men as Rosseland and Steensholt in Norway, of Vogt, Siedentopf and Biermann in Germany, Eddington whom we've just been discussing, Milne, Cowling whom you said something about, Chandrasekhar in England, and Russell in America. I want to know which of them you corresponded with, which you might have seen at various meetings, which you might have visited here or visited at their various places. We are speaking about the period 1930-1936 before you went to America. I was thinking we might start with Norway, where Rosseland had a group and Steensholt there was working on this problem of nuclear processes in stars.
Actually, I had seen Rosseland when I was a boy. I came actually to the observatory library here, and to the Niels Bohr Institute when I was still in high school, and in those years, Rosseland worked at the Niels Bohr Institute. These were the years when he worked with Klein, and worked on theory of collisions, where they demonstrated the importance of collisions of the second kind. These were the years also when he studied Eddington's work and published his important papers, both on stellar interiors and also on interstellar matter, following up some important aspects that Eddington had worked on. I was never in very close personal contact with Rosseland, but I read all his publications—and the same is true for Steensholt. There were a few people who, in this early phase, like Steensholt, tried to see what the results from nuclear reactions would lead to—this is an aside; I promised to write a chapter on this question, for the handbook on astronomy, and the second half of this year, will give me a chance to go back in my own mind, and clear up some questions to which I don't today know the answer. What were the wrong inputs that led to the wrong conclusions? In other words, that the Eddington model temperatures were not adequate.
Did you ever visit Rosseland's institute in Bergen?
No, in Oslo. No, we met in Copenhagen. He was still a regular at the Niels Bohr Institute.
Turning to the German theorists—Vogt, Siedentopf and Biermann—you've already spoken a bit about your contacts with Biermann.
Yes. Actually Biermann and I saw each other rather regularly, because I visited the Babelsberg observatory regularly in those years.
He wasn't there until about 1937.
Yes, that's right.
We're talking about the period 1930-1935(36).
Yes. After my return from America, yes. But I went to the meetings of the Astronomische Gesellschaft.
Yes. I noticed that both you and Biermann gave papers at the Göttingen meeting in 1933.
Yes. That is the main occasion.
Did you attend the Budapest meeting in 1930?
And what about the meeting in Bern in 1935?
I didn't go to that.
What about the meeting in Breslau in 1938; or the very late meeting in Danzig, 1939.
You were elected to the Vorstand in 1939.
Yes. Strange. I'd never attended a Vorstand meeting.
What about contacts with Vogt and Siedentopf? My impression is that Vogt's work had somewhat more lasting value than most of Siedentopf's, but these two were both very active in theoretical matters.
Yes. I never got to know Vogt personally, but I read his papers as they came along very carefully. Now, I saw Siedentopf rather regularly, beginning with the meetings in Göttingen—these were meetings other than the Astronomische Gesellschaft meetings. They were often kind of neighborhood meetings, both in Babelsberg and in Göttingen and in those years I attended them.
You went down to what essentially were colloquia?
Have you by chance kept your old passports, which would indicate which years you went down?
No, I'm afraid I haven't really collected much. But I was in Germany, I should think, every year, in that period.
Turning to the English group, you've spoken about your relations and contacts with Eddington, especially around 1930.
You also mentioned your contact with Chandrasekhar when he was here at the Bohr Institute.
What about with Cowling or Milne? Did you have any correspondence or visits with either of them?
Not personally with Cowling. But Milne visited Copenhagen, and I got to know him quite well. When I wrote a paper for the MONTHLY NOTICES, showing that, right away, you didn't get this model out, I sent it to Milne, and he presented it to the Royal Astronomical Society.
Yes, I noticed that he was the one who presented it.
I had many discussions with him.
He visited then around 1930, did he? Or do you recall?
Yes, I can pretty well fix that, it must have been 1932. He was invited to give a lecture in this house, which was the Director's residence. My parents lived here.
Did your father keep a guest book to the observatory? That would be another thing that would give a clue as to exactly—
—no. You may have seen that there is a guestbook at Yerkes, which was kept in the early years. It mentioned my father during World War I, when he traveled in America.
I will investigate that further. Did you ever attend a meeting of the Royal Astronomical Society?
I guess I'm thinking of this early period, rather than—
—oh, that's much earlier, because I was participating, I was allowed to be present at the IAU Meeting in Cambridge in 1925. And after that there was a meeting of the Royal Astronomical Society at which Shapley and Jeans spoke.
That must have been fun.
I have a feeling that the debates were quite heated.
Well, the Jeans-Eddington debates were.
What about your contacts in the period before you went to America with Russell? He, so far as I can ascertain, was the chief person interested in the theory of stellar interiors in America.
—yes, actually, I think I only saw him personally on one occasion. That was the Astronomische Gesellschaft meeting in Heidelberg [in 1928].
Turning from astronomy to physics, I would like to learn more about your contacts with physics and physicists, from the time that you took your exam (November 1927) until you left for the States in 1936. I'm wondering if you made a practice of attending the special meetings at the Bohr Institute?
Yes. I went there as much as I could. And I was present at a vast number of the colloquia and the smaller meetings, to which Bohr had invited me. This, to begin with, didn't really lead to any publications of mine, since, as you know, my first papers were on models. But it greatly influenced my later work, in that what I tried to do was to absorb all I could on the new quantum mechanics, and the possibilities of calculations that would be involved. All this was related to [ ]'s questions of ionization and opacity, whereas [ ] nuclear [ ]....
I was wondering, among the visitors who came to the Bohr Institute, I find several people who were interested, either at that time or later times, in problems of stellar energy. For instance, Fowler wrote a letter to Bohr in 1930, prior to his visit, saying that he was excited about Milne's work, and that he thought this work would finally produce temperatures high enough for nuclear reactions.
Yes. And you know, Fowler also published a very general paper on the types of solutions, and came to the same conclusion having been reached by Cowling, Russell, and myself.
No, I didn't.
It's mentioned in my article in the HANDBUCH.
Did you and Fowler speak about astrophysical problems when he visited in this period?
Yes. Yes. But we were on questions of stellar atmospheres. Actually, this was the period when he worked with Milne on what could be called the one-point representation of stellar atmospheres, where you represent, as far as ionization and [ ] the whole atmosphere by one representative point where you try to assign the correct temperature and pressure. My discussions with Fowler were in the form of questions, whether he thought it would be possible and worthwhile to compute model stellar atmospheres, that would give the temperatures and densities, and then use these for more accurate predictions. In other words, it was a question that later on one referred to as "coarse analysis" and fine analysis that came up in our discussions. He was a very kind man. He listened to me and said it would be worthwhile to do that.
Let us turn to some physicists who were involved with astrophysics at this time—Gamow and Landau, the two Russians.
Gamow had actually collaborated or assisted Atkinson and Houtermans with their papers.
And must have had some kind of residual interest in this kind of astronomical problem.
Did you talk about these problems at the Bohr Institute?
We discussed them, but never in any very deep way. The period when we did discuss them was the period prior to, shall we say, Weizsäcker-Bethe, when the input from nuclear physics was still very modest.
In your paper for the ZEITSCHRIFT FÜR ASTROPHYSIK, the first paper you did for that journal, you acknowledged assistance from Landau.
Yes. Landau was a frequent visitor, and he was deeply interested in these questions, but had his own views that differed radically from those of other people. You know, there are publications by—
—oh yes, I've seen them—
—And his idea was that you have a very high density core, in the sun, and that release of gravitational energy therefore plays a role. For this reason he was inclined to disregard all of Eddington's work.
Did he have this skepticism toward Eddington in this early period, in the early thirties, as you recall?
When he was in Cambridge he may have had some close contacts with Fowler, which would have—
—that I don't know—
—which would have increased this skepticism toward Eddington.
It's possible that there are, in the Archives of the Niels Bohr Institute, some letters that he might have written on the subject. But I think that his own contributions came at a time when I was in America.
Yes. Do you remember any contacts with Weizsäcker in this early period?
Well, I saw him even before he came to the Niels Bohr Institute. He was the son of the Ambassador, and was interested in natural sciences as a boy, when he visited the observatory.
Do you remember that visit?
Yes. I showed him the observatory. We didn't talk any astrophysics, but—
I remember from your interview that after he went back to Germany, you visited Berlin to look at photometric techniques and so on.
Did you visit Weizsäcker at all on those visits?
No. My discussions with him were when he visited the Niels Bohr Institute, and then once after I had returned from America, when I was giving lectures in Berlin.
We'll come to that in a little bit. One final question about physics in the period prior to your departure for America, and that is, did you feel conversant with the development of nuclear physics, insofar as it was represented in discussions at the Niels Bohr Institute?
Yes. Also there I tried to learn as much as I could. I particularly remember the period when Niels Bohr lectured many times on [ ].
Now we want to turn to the period 1936-38, and I would first like to hear how it was that you happened to write so many review articles, maybe one should call them monographs, in this period. First there were two for the HANDBUCH DER ASTROPHYSIK, then there's one for the HANDBUCH DER EXPERIMENTALPHYSIK, and finally there's one for the ERGEBNISSE DER EXAKTEN NATURWISSENSCHAFTEN, so, four major review articles, which seem to have taken much of your time in this period say early 1935 up to 1937?
Start with Volume VII—
—of the HANDBUCH DER ASTROPHYSIK.
Yes. Actually, the editors asked me to write these, because they were in a slightly difficult position. Milne had written the first articles. They were excellent. But then there was the Milne-Eddington controversy, and what made it more difficult was that Milne had included his own ideas on the pulsation theory, which had been proved to be wrong. So he was not inclined to bring it up to date in this Ergänzungsband. Since the editors knew that I had been working on this field, they asked me to write them and I was glad to do so.
Who were the editors?
Luddendorf and Eberhard.
And you had had some contacts with Luddendorf.
Yes. You see, these people were very close friends of my father's. I had stayed at the Eberhard house and visited the Luddendorfs. I had also stayed with another friend of my father's, Guthnick in Babelsberg, so I had close personal contacts with them too. It all seemed very natural that these contacts would result in my being asked to do it, especially since there was this background of Milne not being willing to do it. Actually, when I prepared these reviews, I spent nearly all my time for seven months, reading. This was possible in those days. I read everything, and tried to digest it. Of course, my trouble was that my articles were much longer than the editors had asked for. But that was unavoidable. They accepted it. They'd been exposed to this before, so they had made a simple rule that you were only paid for what you had been contracted to write, but you were free to write more. I think I figured once, I worked for five cents an hour!
What about the long review of experimental astrophysics for the HANDBUCH DER EXPERIMENTALPHYSIK?
Yes. That came about in the following way. The editor was Hans Rosenberg, who was one of the pioneers of photo-electric photometry, together with Guthnick and Stebbins. He was to be the editor but he was one of the Jewish scientists who left fairly early. He actually left to go to Yerkes. He had a modest position and he left so early that he was allowed to take a good deal with him. Now, that meant that he couldn't continue the job. He first suggested that my father be the editor. My father said he couldn't do this. It was mostly astrophysics and his field was purely classical astronomy. So in the end, I was the editor, and I wrote the two articles that Rosenberg himself had planned to write. Actually, in this case, I spent the whole winter on the preparations. Again, I think I read everything—one could read it then—and that meant more, because in photometry, in spectroscopy, there simply was a large number of pages to read. But I've never regretted that I had the chance and could do that. Now, these were the articles that had been finished when I left for the States. I got the proofs there, and they are the proofs that I referred to in the letter to Struve. Actually, as you mentioned, I was interested in photo-electric photometry from the very beginning, when Guthnick developed it at Babelsberg—I worked along with him, to see how it all—
When was that, what year?
In the early twenties. And my connections with this led to part of the work in classical astronomy on the photoelectric registration of transits. This was the natural continuation of my work. My first paper was on observations with a transit instrument where I had discovered certain types of personal error, and this led me to the field of trying to [ ] photoelectric— But, to come to the papers on photometry, this obviously was a field I had been interested in. In later years, as you know, it's become connected with my theoretical work, so the combination of the two became important, in certain chemical compositions for the atmospheres, and the ages of stars.
Then finally there was the article for the ERGEBNISSE...
Yes, that—there's an interesting background to that, that's referred to in the other interview. Namely, one of those who came frequently to the Bohr Institute was Hund, and we discussed questions of stellar matters with him, and in the end it was agreed that he would write an article for the ERGEBNISSE on the physics of stellar-interior matter, and I would write the corresponding astrophysics review article, and this is the one that you refer to. That was actually written during the summer months in 1937.
But you'd agreed to do it before you left?
Yes. I'd actually gathered the material and prepared for it. But in the end, I wrote it in three weeks, during temperatures of 100.
Turning from general background, I have found that the article in the ERGEBNISSE DER EXAKTEN NATURWISSENSCHAFTEN was very important for the physicists, in providing them with the current—
Yes. I found that, in the thirties, this is where they got acquainted with stellar interiors, rather than through Eddington's book. For instance, a footnote by Tolman, shows how physicists got to know about the problem. There was also a limitation—it was in German. But in those years, even in America, obviously German was studied.
Yes, I'm sure that all good physicists, especially in the theoretical area, kept up with German.
Yes, it was so necessary, German, in those years, when quantum mechanics was developing.
Now, I want to turn more specifically to the nuclear developments. In June of 1936, Weizsäcker attended a session on nuclear structure at the Bohr Institute, and from his correspondence with Bethe, I see that he talked with Teller about his plans to investigate nuclear reactions in stellar interiors at that time. I'm wondering if you recall—this is of course expecting a lot—but do you recall any conversations with him on the matter in June of 1936?
No, I really can't. I have a recollection that this went on continously, in those years, either through meetings, or correspondence. This was just the last few months before I left in September, 1936.
You mentioned in your letter to Struve—dealing with Weizsäcker's proofs, that you had reviewed a manuscript in the fall at Yerkes. I'm wondering if—
—that was at the colloquium.
Was this a manuscript of Weizsäcker's article that he gave you?
Yes. He sent me correspondence, in this case.
So you essentially had this and presented it at a colloquium. Do you remember the response in the colloquium? I gather that Struve was not at this time terribly interested in such problems. Kuiper?
Kuiper was closer to it, because of his work on galactic clusters, and the location of their Main Sequences. That was his contact with the problems.
Do you remember the colloquium at all? It was probably the first colloquium you gave at Yerkes.
No, I don't think so. No, I gave quite a few, small groups and little conferences.
Then in February, 1937, having received Weizsäcker's proofs and I guess having time then to do some research, since review articles were somewhat behind you except for the ERGEBNISSE one, you began this intensive investigation of the hydrogen and helium content of the stars.
And by March 26th, you were writing Struve that you were almost through with the three papers on these subjects, and you suggested the order they might be published in and so on. Yet it turns out that the second paper was published right away, the first paper wasn't published till a year later, and the last paper you mentioned wasn't published at all, so for as I can tell. Do you recall some of the history of that?
Yes. I think that is pretty clear. In fact, the calculations on the results that you get when you include helium were pretty much completed and are referred to in the ERGEBNISSE article, and that was then published in THE ASTROPHYSICAL JOURNAL. As you can see, the composition that I give in the letter to Struve is just the one that we prefer today. That I think, is an important aspect of that development. On the one hand, as is also clear from the letter, we all knew that Weizsäcker's arguments for the AUFBAU hypothesis, for this ratio of the elements, was strictly [ ] and what I wanted to show was that it was compatible. But the other evidence, the very weak evidence, but still there was evidence from the atmospheres of the hydrogen-helium content, that all led to this picture, that you have a composition where the heavy element content is just about what you see in the atmospheres. It's out of this work that, during the next ten-fifteen years, developed the idea of that it's all the same composition. This of course was then strongly supported by the results from the early history of the sun that have been mentioned throughout. Still, that wasn't clear till the early fifties. But it is just the beginning. From then on, it was very much in the picture, that you could have a composition where the atmosphere and the interior gave the same values.
Can you recall why your paper entitled "On the Helium and Hydrogen Content of the Interior of Stars," ASTROPHYSICAL JOURNAL, didn't actually appear until June 1938, roughly a year after you were doing this work?
This had to do with the fact that I had given the results, in the ERGEBNISSE and it wasn't so urgent.
So it's really a matter of the ASTROPHYSICAL JOURNAL's publication schedule rather than?
No. As soon as I finished it, it came quickly into the JOURNAL. But there were many things that I gave priority to, because I had actually described the results, and it wasn't that urgent.
I see. Then, during the summer of 1937, you finished your review article for the ERGEBNISSE DER EXAKTEN NATURWISSENSCHAFTEN which we've been discussing.
And in this article, you gave a fairly extensive summary of Weizsäcker's theory. Then so far as I can tell, you had rather little contact with stellar nuclear reactions until March of 1938, when you attended the Washington Theoretical Physics Conference which was devoted to stellar-energy problems.
I'm wondering, how you happened to get invited to the conference and what you recall about the conference? The reason I ask this is that Gamow, who was putting this conference together, so far as I know wasn't in terribly close contact with the Chicago group at this time.
No. It was actually via Copenhagen, so to speak. We knew each other from there, and the same was the case with Teller, so Gamow and Teller invited me to come. It fitted in so very well because I was on my way from Yerkes to Copenhagen. I was returning home. I spent three weeks visiting various places. One of them was this Washington meeting, and then afterwards I went to Harvard.
Do you have any special recollections of the Washington meeting? There were several physicists in attendance.
Yes. Actually my recollection was that no solution was in sight. But that certainly there was a hope that the input from nuclear physics was around the corner.
I gather there was a considerable discussion of Landau's theory at this time, because he had just published it in NATURE.
With the condensed core.
And there was probably also some discussion of Gamow's theory, which was that there was a resonance zone, a shell model of the stellar interiors, and that you had this shell that would gradually expand with the resonance temperature?
Yes. Actually, throughout this period, there were two main problems, as Eddington had formulated them—First, to understand the situation for the sun, second, to understand the situation for the Giants. It had been pointed out quite clearly by Eddington that if you use the standard model, then in Capella, volume for volume, the density is lower, the temperature is lower, and the energy production was higher. Then, actually, Gamow stuck to this idea. He was really the first to have the correct idea of energy production in the Giants, in the shells at least in some cases—this is before you got high level temperatures. Because he worked with Teller, and actually, what they published on this was positively wrong. And the first self-consistent model of this kind was one produced here, at Copenhagen observatory, and published by Reiz. In the war years, with travel between—we could do that—between Lund and Copenhagen, he worked on this problem. Actually he produced a self-consistent model on Gamow's idea, one that Gamow very much appreciated, because he had in a way been shown correct. Now, there was a big step from this finding to the theory of the Giants because it was just a model, and no attempt to show how it was developed from here to there—that is, from Main Sequence to Giant. That only came with Schwarzchild—Schwarzchild and Sandage, their papers—and actually, it was only possible after the computer had been produced so that you could really do it step by step.
Yes, the step-by-step calculations look like they must take a long time.
It's amazing how much was done in that way.
Yes. Actually, the stellar energy integrations, they didn't use [ ] assumptions. To carry it out, that was too many steps.
As you were starting to say, soon after this meeting in Washington, you returned to Europe.
You mentioned that you had visited Berlin at some time? Do you recall when that was?
Soon after my return in 1938. I gave a colloquium at Babelsberg. I gave a general lecture at the University. That must have been the occasion when I was at the house of the Weizsäckers.
At that time you spoke with Weizsäcker most likely about these problems.
And I guess, with Biermann as well.
Yes, that's right. When I was at Babelsberg at the colloquium, it was with Biermann. The colloquium was about the development that started with the paper that is referred to in the letter to Struve which wasn't published at the time. Like others who were working on stellar model atmospheres, I was a bit frustrated. There was something wrong. This was all cleared up when Wildt discovered the role of H-. My colloquium in Babelsberg was on the solar atmosphere, when you took into account H-. Then everything falls into place. And this then led to, when I finished all that work, to the publication in the FESTSCHRIFT for my father. So there is a line from what I mentioned in the Struve letter. However, I think in a way, I was on the right track, because I mentioned Vega and Sirius, and that was free of the uncertainty that resulted from us not knowing about H-, because by the time you had that temperature, the absorption of [ ] was from hydrogen. And you could do something about it. The reason the work there didn't proceed was simply that there weren't good enough spectra. The problem was well known, but when the observational techniques are rather limited, you cannot go to the faintest lines. That means that you have saturation effects. And it's only when you proceed experimentally to the very faintest lines that you get over that. Then you can do the things that I hoped to do with sodium and potassium. But the spectra were too poor. It was, in that respect, a very interesting period. In January, 1938, I visited MacDonald Observatory. Of course it was just in beginning stage. The 22 inch wasn't there, and it came into effect much later. That was when the spectroscopic period for the atmospheres began.
Then later in 1938, in August, the International Astronomical Union met in Stockholm.
And you served as secretary for Commission # 35 on stellar composition.
And according to your published report, there was a lively discussion of the work of Weizsäcker, Gamow and Bethe. How did people know about Bethe's work? Gamow and Weizsäcker had both published several papers, but as of the August meeting, Bethe had not yet published any papers on this subject, not even the proton- proton paper.
No. I think it was just floating around, because there was correspondence. I didn't have any direct information on Bethe, but Weizsäcker must have had.
I'm wondering if you have any special recollections of that meeting? I guess it was attended by Russell, by Eddington, by Atkinson. Had you met Atkinson before, for instance?
Yes, I must have met him in Göttingen. I think so.
At that point, of course, Bethe's main papers had not yet been published, so things must have seemed somewhat—
—yes, they were still tentative. That's right. Yes.
But I did notice that Weizsäcker and Gamow were elected as members to Section 35. So there must have been some feeling that the physicists were making progress.
Yes. Yes. And actually, I think Jan Oort was responsible for that. He was the new general secretary, and had an open mind for . . .
Can you say something, while we're on Section 35 which dealt with internal constitution of stars, can you say something about your earlier contacts? I think you were a member of this section since the early thirties, but I don't know which of the meetings you attended.
No. The Stockholm meeting was the first meeting where I was a member. I wasn't a member at the time of the Paris meeting.
You didn't attend the Paris meeting?
No. No. The only meeting I'd attended before the Stockholm meeting was this Cambridge meeting in 1925.
So Stockholm was your second meeting. In April of 1939, Bethe's main paper finally appeared. Your reaction to it seems to have been far more positive than to Weizsäcker's paper. I'm thinking here of your article in the NORDISK ASTRONOMISK TIDSSKRIFT.
Yes. I actually gave a talk in the Physical Society here in Copenhagen, where I just accepted Bethe's paper as the final solution and—
That talk would have been in 1939, do you think, or already in 1940?
No, 1939. I'm pretty sure.
I'm wondering how I'll find out about that.
That is difficult. I don't think they have records for each meeting of that Society in those days.
What were the characteristics of Bethe's paper that made it seem so much more compelling say than Weizsäcker's?
Well, I felt that, first of all, he had the better physical input. And also I felt that he had done what should be done. We knew, after all, from the model work, what the temperature was about, what the density was and what the chemical composition was. Then we just had to look at, sort out all the possibilities of reactions under those conditions. And he had done that systematically—and it was just the right thing to do. And the answer came out right because he had approximately the right cross-sections. Now, until 1950, of course, it was felt that the main process was the carbon cycle. It was only in 1950—with the papers by Epstein and Hopf—that it became clear that for the sun the proton- proton process was primary. That had to do with revised cross- sections for the proton-proton process. I was working on that at CalTech in 1950, for a few months, and I received some information from Salpeter, and discussed the problem with Fowler and in a rough way had the answer. But it was only when I came to Princeton the following months that one of the graduate students, Epstein, actually did the model. That of course also had the consequence that, with the model without a convective core, because of the softer temperature dependence for the proton-proton process.
The temperature of the sun may be estimated somewhat high in the early Bethe paper also?
That would have influenced one toward the carbon cycle.
Yes, that did. That was natural, that would come out. But Bethe certainly at the time had both reactions.
I have one final question and that is, as I've worked on this problem, I've been struck by the fact that theoretical astrophysicists posed the problem, even thought that the solution must involve nuclear processes. Yet is was the nuclear physicists or theoretical physicists in nuclear physics who eventually solved it. I'm wondering why somebody such as yourself, who knew a good deal of nuclear physics, didn't tackle the problem directly? So far as I can find, no astrophysicist, not even the ones who had some familiarity with physics, tried to tackle the problem directly. I'm wondering, in your particular instance, did you ever consider doing so?
Actually, it was so difficult to sort out what was going on in nuclear physics at the time that it took nuclear physicists who had devoted all their time to the field to sort it out. That is my impression—that we felt like amateurs.
But they must have felt the same way about astrophysics.
Yes. But you see, a meeting like the one in Washington, there it was possible for the astrophysicists to communicate these simple results: the temperature must be this, the density this, the composition this, and why. We gave convincing reasons. It was a strange situation, of course—that you could really arrive at a model, here is the temperature and the density, without using at all the input from energy production. It was simply due to this situation that, whatever the mechanism, it must be one that gives a high degree on concentration of energy production in the central region. Then there's no doubt about the model and that fixes the temperature. Once this was understood by the physicists who were ready to accept this, in spite of what, shall we say, Landau said, then that communication was easier than the other. There were so many things that were very difficult for one who wasn't a nuclear physicist to appreciate. For instance, if you look at the literature in those days, there were discussions of the possibility of the stability of atoms of mass 5.
And this seemed to some to be the necessary stepping stone, to go on, even as before Salpeter's work on triple-alpha, it was felt that you couldn't really get three alphas together. There had to be some missing link.
I think that covers all the things that I had managed to plan—
—well, I wish my archives were more complete. There isn't so very much I can tell you. If I had kept Weizsäcker's letters, it would certainly be interesting. But I will send you a copy of the Eddington letter.
I would like to see that.
C. F. von Weizsacker, PHYSIKALISCHE ZEITSCHRIFT, 39 (1938), 633-646; H. Berthe, PHYSICAL REVIEW, 55 (1939), 434-456.
Stromgren, MONTHLY NOTICES, 91 (1931), 466-472; ZEITSCHRIFT FUR ASTROPHYSIK, 2 (1931), 345-369.
Stromgren, MONTHLY NOTICES, 91 (1931), 466-472; Cowling, ibid, 472-278; Russel, ibid, 739-751.
Stromgren, ZEITSCHRIFT FUR ASTROPHYSIK, 4 (1932), 118-153; 7 (1933), 222-248.
Stromgren, ZEITSCHRIFT FUR ASTROPHYSIK, 4 (1932), 118-153.
Stromgren, MONTHLY NOTICES, 91 (1931), 466-472.
Atkinson and Houtermans, ZEITSCHRIFT FUR PHYSIK, 54 (1929), 656-665.
Stromgren, ZEITSCHRIFT FUR PHYSIK, 2 (1931), 345-369.
See Bohr Scientific Correspondence, P3, reel 23.
Stromgren to O. Struve, 18 February 1937, Struve Papers, Yerkes Observatory
Stromgren, ERGEBNISSE DER EXAKTEN NATURWISSENSCHAFTEN, 16 (1937), 465-534.
Weizsacker to Bethe, 30 September 1936, Bethe Papers, Cornell University Archives.
Stromgren to Struve, 21 February 1937, Struve Papers.
Stromgren to Struve, 26 March 1937, Struve Papers.
ASTROPHYSICAL JOURNAL , 97 (1938), 530-534.
See Chandrasekhar, Gamow, and Tuve, NATURE, 141 (1938), 982.
Landau, NATURE, 141 (1938), 333-334.
Stromgren, IAU TRANSACTIONS, (1938), 416-417.
Stromgren, NORDISK ASTRONOMISK TIDSSKRIFT, [ ] 31 (1940), 37-52.
Lectured on “Nyere Undersogelser over Grundstoffernes komiske Forekomst og Omdannelse,” on 2 October 1938, SOURCES FOR HISTORY OF QUANTUM PHYSICS, reel 35.