Bengt Strömgren - Session I

Hoddeson:

Today is May 6th, 1976, and this is Lillian Hartmann Hoddeson. Gordon Baym and I are visiting Professor Bengt Stromgren in his office in the observatory in Copenhagen to discuss astrophysics in the 20th century, and particularly Professor Stromgren's role in it. This role began in 1908 in Goteborg, Sweden, where you were born to Swedish parents.

Strömgren:

Right. My father had just been appointed professor of astronomy in Copenhagen University. Before that, he had been at the University of Kiel in Germany, and working also for the editor of the ASTRONOMISCHE NACHRICHTEN. So the family was moving to Copenhagen, and I was born in Goteborg where my mother was then living with her sister.

Hoddeson:

How old were you when you actually moved —

Strömgren:

— half a year old when I came to Copenhagen. I consider myself a Dane.

Hoddeson:

And there was no problem, you felt completely at home in the Danish community, as a Dane.

Strömgren:

Yes.

Hoddeson:

Looking back, do you think that Denmark's long tradition of interest in astronomy, dating back to men like Tycho Brahe and Ole Romer, influenced your interest?

Strömgren:

I'm sure it did. Since I lived, when I was a boy, in an observatory, I saw all this and also I heard even as a boy about the great men of the past whom you mentioned.

Hoddeson:

How did it turn out that you lived in the observatory?

Strömgren:

We are now in what used to be the director's residence.

Hoddeson:

Was it a tradition for the director of the observatory to live in it?

Strömgren:

Yes. The observatory was built in 1859, and it was a classical European observatory, with the director's residence in one wing, and in the other wing the chief assistant and the assistant and a gardener. The middle section was reserved for what was then called the big refractor, a lens telescope with a 13 inch aperture, and the two meridian instruments, the transit instruments for the determination of time and the meridian sector. This meant I grew up at the observatory.

Baym:

Did you spend much time as a child looking through the telescope?

Strömgren:

I started my interest in astronomy actually, I would say, in 1920. Then I began observing with the instruments. I was taught to determine time, which was important in those days. It was just before they depended on radio. Time was determined at the observatory for Denmark.

Baym:

Who taught you to do this?

Strömgren:

One of the assistants was very kind, and then, from 1920, I also remember determining the position of nova cygni. But systematic efforts started the year after, in 1921. I used every clear night but in a very peculiar fashion, from a half hour after sundown, when it was dark enough, until 10 o'clock, because I was a school boy and I had to go to bed early. But even with the poor Danish weather, it was enough to cover this program, and in two years it was finished. The result was a catalog of right ascensions of circumpolar stars that was published in the Swedish Academy. I think you have seen the publication, from 1924.

Hoddeson:

Did you discuss this work you were doing with your father?

Strömgren:

Yes.

Hoddeson:

While you were doing it, on a regular basis?

Strömgren:

Yes. My father's efforts and interest in teaching me astronomy started earlier, with the flu epidemic in 1918, which meant that school was closed for a month. That was really the beginning, when my father taught me mathematics beyond what we were taught in school, and the basic knowledge in astronomy. Not much astrophysics. It was very good astronomy, traditional astronomy, astrometry and dynamical astronomy.

Hoddeson:

Did you meet many eminent astronomers?

Strömgren:

Yes. My father was then president of the Astronomische Gesellschaft for a number of years, beginning I think in 1920, and that meant that a number of the very well-known astronomers of the day came through Copenhagen.

Hoddeson:

Do you recall any particular examples?

Strömgren:

Well, Einstein is perhaps the example I should mention first. He lived in this house for a few days, after he had received the Nobel Prize. But I was too young to appreciate that. My brother and I didn't quite appreciate Einstein's greatness. We were too young.

Baym:

How did it turn out that you became an astronomer and your brother became a doctor?

Hoddeson:

He was a year younger?

Strömgren:

Two years younger. I was of course influenced by all I saw around me here, and since I was the oldest, my father concentrated on me. Whereas my brother was very much influenced by an uncle — that is, the husband of a sister of my mother — who was a psychiatrist. My brother spent every summer with that family, and I think that was why he chose a medical career and specialized in psychiatry. He's now head of a mental institution and also professor of psychiatry.

Hoddeson:

Your mother was also a professional person.

Strömgren:

Yes. She was a dentist, and she spent much of her time on works on the history of dentistry. She has published one work on dentistry in antiquity, and a history of dentistry in the 18th and one on dentistry in the 19th century. Her last effort was an index of all works up to a certain time concerned with the history of dentistry. I understand it wasn't an easy task, because in the early days, dentistry was included in works on surgery, so you have to go back to those sources. My mother visited all the major libraries in Europe to work on this index. That was her last effort.

Hoddeson:

Did she practice in the observatory?

Strömgren:

Well, to begin with.

Hoddeson:

She did?

Strömgren:

Right here. And later on she had her clinic, five minutes walk from here.

Hoddeson:

Interesting! Can you recall any other teachers besides your father who played a role in your early commitment to astronomy?

Strömgren:

Yes. You see, this program that I mentioned that I started, the observations of the circumpolar stars, was suggested by a man who played a great role, particularly in Germany but also in international astronomy. His name was Kustner. He was professor in Bonn. And his works — I also talked with him — influenced me a great deal. He was the type of astronomer who insisted that you should work for the future, and you can read in the preface to some of his work on positions of stars in galactic clusters that he was quite prepared to see these used fully only after hundreds of years. That may still be their most important use. Already this early work on positions of stars in galactic clusters has led to accurate determinations of proper motions, so that you can single out the members of the clusters with more certainty, and a good deal of the basic information that we have today on stellar evolution is ultimately based on knowledge of the properties of stars in galactic clusters. And so this early work from around the turn of the century was really of very great importance. He had foresight. In those days, there were other very important figures, and his opposite, in a way — and this is known also from anecdotes, stories from those days — was the Dutch astronomer Kapteyn, who was one of the great pioneers in galactic research.

Hoddeson:

Did you actually get to know him? Did he come to Copenhagen?

Strömgren:

No, he didn't.

Hoddeson:

Then you just read —

Strömgren:

That was indirectly what I knew at the time. My father knew him well, but I never met him.

Baym:

But Kustner?

Strömgren:

Yes, I went to Bonn, when I was engaged in this work, it must have been 1922. And I saw him in Copenhagen in 1925 on the occasion of a meeting of the Astronomische Gesellschaft. That was a high point, of course. I had just graduated from high school, and at this meeting, Eddington was present and Shapley and a number of others.

Hoddeson:

What kind of meetings were these?

Strömgren:

Well, you see, the Astronomische Gesellschaft in those days was the German Astronomical Society, but it was international in character. This was a time when the International Astronomical Union was new and was developing, so that for a while, both these efforts were important — the old international Astronomische Gesellschaft and the new International Astronomical Union. It's clear that about 1933, anyway, one can say that the Astronomische Gesellschaft lost its international role, and at the same time the International Astronomical Union was becoming stronger and stronger, so from that time on this was the great international organization in astronomy. But the meeting in 1925 was at a time when the Astronomische Gesellschaft was still regarded as the important international organization, and that explains why so many of the very well-known astronomers of the day came to Copenhagen.

Baym:

Where were these meetings held?

Strömgren:

Well, the opening was at the University, and we could make use of various auditoriums. In those days, astronomical meetings were not so big.

Baym:

About how big?

Strömgren:

Well, now, I think I must correct myself. We can look that up. The Astronomische Gesellschaft meeting may have been 1926. The meeting of the International Astronomical Union in 1925 was in Cambridge, England, and the number of people there was about 200. That gives you an idea what it is. In Grenoble this fall, I suppose, there will be four or five thousand people (at the IAU meeting].

Hoddeson:

Do you recall any issues that seemed most exciting at that particular meeting?

Strömgren:

Yes. I remember particularly the contributions by Eddington on the mass luminosity law and referring also to white dwarfs. And by Shapley on his research on globular clusters, that had led to a valid picture with an off-center position for the sun, whereas Kapteyn's efforts had led to a picture where the sun was rather at the center of a much smaller system. You know, today, that the reason was that he did not take interstellar absorption into account. Not accounting for where you are in the galactic frame will lead to this mistake. Now, it wasn't just an assumption. He tried to determine interstellar absorption, and his result was a low value. And we can see today also why he came to that conclusion. First of all, the determination couldn't be very accurate, and what he did was try, as one should in the beginning, to find deviations from the inverse square law. That is, if you put it in the usual way, of relations between absolute magnitude, apparent magnitude and parallax. He saw that this law proved true without an absorption term, and for that you need to know all three, absolute magnitude, and you measure the apparent magnitude and also the distance.

Now, he couldn't derive the distances from parallaxes, because when you are so close, there's no hope of detecting any absorption from this. But he did it from proper motions and by statistical methods. Even with proper motions you cannot go very far, and we know today, it so happens, that we are in a region where the absorption is somewhat less than at greater distances, and this explains it. People thinking back also have an idea that this was wishful thinking, because things become very complicated in galactic research if you have to take absorption into account. Now, the development in the thirties showed that we have interstellar reddening. It's a very strange thing that this was only firmly established in the 1930s. It had been discussed even in the previous century. But the difficulties are really great, when you think back. First, the photometry was then photographic photometry, with all the difficulties that you do not have if you have photoelectric photometry. And it was well known that in some of the work in photographic photometry there were scale errors; that your colors, even if they were the same, for bright and faint stars would come out different. And so when there were indications of reddenings, people might suspect — well, this is a scale error.

Furthermore, the theory of stellar atmospheres was in its infancy, and it was by no means certain that the stars that show the spectra of these stars also had corresponding colors. You can find articles trying to explain the fact that some of the B stars show red colors in terms of intrinsic properties of the stars. But during the thirties the situation improved. Guthnick at Babelsberg — I will come back to that because he was one of the astronomers who played a great role in my life — had developed photoelectric photometry methods, and his work was continued by collaborators such as Bottlinger and even Becker, and by the middle of the thirties their work — and also parallel work by Stebbins in the United States — had given reliable colors, so that you can compare bright and faint stars. At the same time, there had been progress in spectral classification, so that with some confidence one could classify according to luminosity all the B stars that were considered, and eliminate or segregate, I should say, the super-giants. This was another danger, that there was a systematic color difference between ordinary B stars and the super-giants, and as you go to fainter stars the proportion of these changes, and that might give rise to a false impression of reddening. But by and by, all this was put right, and interstellar reddening was established. Now, the next phase in galactic research is very much connected with the idea that you have a universal law of interstellar absorption. If you have that, then the reddening would tell you what, say, the absorption is in the visual.

Hoddeson:

When did that idea become current?

Strömgren:

Well, it came about at the same time that reddening was established. Then the hope arose, isn't it possible just to apply a factor, and you have the necessary correction for the absorption, say, in the visual or photographic region?

Baym:

Were you involved in this work?

Strömgren:

No, I only followed that at some distance.

Hoddeson:

How did you follow it? Did you read the journals?

Strömgren:

Yes, and I went to meetings. Now I'd like to come back to one important development that took place at the same time. This was Trumpler's — he was originally from Switzerland and his name is really __ — his work on galactic clusters.

Hoddeson:

We are now in the early thirties?

Strömgren:

Yes. Here the distance criterion was the diameter of the cluster. Now, galactic clusters differ in many respects. The richness of the cluster and the degree of concentration toward the center are important parameters, and he classified clusters according to these criteria. So he was reasonably confident that in each class the diameter would be about the same. And then the diameter gives an indication of the distance, and he could prove that if you, say, compute distances without allowing for absorption, you run widely off the track. And these two parallel investigations, of reddening and of the cluster distances, established the existence of interstellar absorption. We can perhaps later on come back to that, because I've spent quite a bit of my time over the last ten years on a detailed investigation of interstellar absorption, because it gives a very important clue to the properties of the interstellar medium. The particles that we think produce the absorption are traces — and when you think of the content of the heavy elements, they are more than traces — they contain perhaps most of the heavy elements.

Hoddeson:

I would like to not leave the twenties quite so fast, though.

Strömgren:

Well, I think we are just following along one aspect of it, because we started with Kapteyn's researches.

Hoddeson:

We didn't finish that, maybe we should.

Strömgren:

I remember, from the end of the thirties, a conversation with Eddington, who was in England at the time. I asked him, why did it take astronomers so long to find out that there is interstellar reddening, when after all, these are big effects when you do the proper photometry? And his answer was that astronomers are like sheep, they follow the leader. That was Kapteyn.

Hoddeson:

What was your impression when you were young of Eddington?

Strömgren:

Well, I saw him the first time in 1925, so I didn't see him when I was a boy here. But he is one of the astronomers I met who impressed me the most. He was a very kind man, and it was not difficult to talk with him.

Hoddeson:

Did he talk to you in 1925 a little bit?

Strömgren:

Yes, but my main discussions with him came later.

Hoddeson:

Here in Copenhagen?

Strömgren:

Well, also by correspondence, and at other meetings of the Astronomische Gesellschaft. I came to think of this when I was in Heidelberg yesterday, because he was at an Astronomische Gesellschaft meeting there.

Hoddeson:

Was that the major society?

Strömgren:

Well, it was the society whose meetings played an important role, because the International Astronomical Union only met every three years, and the Astronomische Gesellschaft met much more frequently. Since it was until, I'd say, beginning in 1933, international, it played a role as a place where astronomers from all over Europe met rather frequently.

Hoddeson:

While we're on the institutional background, I'd like to ask about the major journals in that period.

Strömgren:

Yes. This was in those days very simple. There was the ASTRONOMISCHE NACHRICHTEN. That had long traditions back in time. It had a tradition of publishing everything without refereeing. It was left for the readers to send criticisms, which were then published to set things right. Then there was the MONTHLY NOTICES [of the Royal Astronomical Society, London] which was a very important journal. When you think of astrophysics, it was a good deal more important than the ASTRONOMISCHE NACHRICHTEN, although some astrophysics articles were also published in the ASTRONOMISCHE NACHRICHTEN. But that tended to be more classical astronomy. That means astrometry and planetary motions. In Germany there was, in addition to the ASTRONOMISCHE NACHRICHTEN, a new journal, the ZEITSCHRIFT FUR ASTROPHYSIK, and in America, the ASTROPHYSICAL JOURNAL. Today it's only fair to say that the ASTROPHYSICAL JOURNAL is the most important journal, but that wasn't so in those days. The MONTHLY NOTICES was as important in astrophysics. In the United States was also published the ASTRONOMICAL JOURNAL, and there was a division again — somewhat similar to the division between the ASTRONOMISCHE NACHRICHTEN and the ZEITSCHRIFT FUR ASTROPHYSIK. But the classical parts of astronomy were published in the ASTRONOMICAL JOURNAL, and astrophysics in the ASTROPHYSICAL JOURNAL. Then there were —

Hoddeson:

— did you look at all of those journals, here in Copenhagen?

Strömgren:

Yes.

Hoddeson:

Even while you were still in high school?

Strömgren:

Yes. And even more systematically from, I would say, 1927 on.

Hoddeson:

When you entered the university?

Strömgren:

No, I entered the university in 1925.

Hoddeson:

I would like to ask you some questions about your education here in Copenhagen.

Strömgren:

Yes. I graduated from high school in 1925, and then I spent a year and a half studying for the comprehensive exams that you then took, after two years, in mathematics, physics, chemistry and astronomy. There was no choice. There were very definite courses, and you had to get over that hurdle. And it was a hurdle, because you had to take seven written and six oral examinations in one month, and know it all, at that time.

Baym:

How many classmates did you have?

Strömgren:

Well, typically for an exam like this, there would be between 10 and 20, and it was held twice a year. The classes were much bigger, because we were taught mathematics and physics together with the students at the Technical University, so there were classes of a couple of hundred people, and in chemistry the same. We were taught chemistry together with the pre-med students. Astronomy was smaller classes, 30 or 40.

Hoddeson:

Were you aware at that time of the tremendous developments that were taking place in physics?

Strömgren:

Yes. Very much so. You see, I took that exam after a year and a half. And then followed an extremely happy period.

Hoddeson:

— we are now in 1927-28, the period in which you worked on your doctoral dissertation.

Strömgren:

1927, and now I studied for the final degree. I had already published some papers in astronomy, the one I mentioned, and a few more — some on determination of cometary orbits, and some on observational data, to record transits of stars through application of the photo-cell instead of the eye. From early January, 1927 I could concentrate on my major subject, and since I had published in astronomy, I could concentrate on physics. It was in that year, from January till the time I took my Master's degree in physics and astronomy, the astronomy part being the publications —

Hoddeson:

Was it a Master's degree before the Ph.D.?

Strömgren:

No, in Denmark the final degree at the university was the Master's degree, and then the Ph.D. has no examination. It's just a thesis, which traditionally was a good deal bigger, and which then gave the right to teach at the university. It's not so different today, but the trend is certainly away from the very big thesis that was sometimes published in those days. Well, so my aim was to finish at the university and get the Master's. For that, I studied physics, and it was a wonderful year. For instance, Shrodinger's collected works, you know that volume on wave mechanics, had just appeared, and that was part of the curriculum. I read all those. And one of the happiest periods was when I read Courant and Hilbert's preparation, the MATHEMATISCHE...

Hoddeson:

MATHEMATICAL PHYSICS, yes.

Strömgren:

But in addition, we then studied some of the Sommerfeld volume on atomic theory.

Hoddeson:

You're talking about the work on spectral lines?

Strömgren:

Yes, ATOMBAU UND SPECTRALINIEN. You see, that was before quantum and wave mechanics, whereas this was then supplemented with the Schrodinger book on wave mechanics. One thing for me that was extremely important was Oscar Klein's on the interpretation of the wave picture in terms of probabilities. I learned more from that than from most of the other papers.

Hoddeson:

My impression is that this was a standard part of the curriculum in Copenhagen at that time. Is this correct?

Strömgren:

Well, when you were past this hell where there was no choice, there were many choices. But for those who studied with Niels Bohr, as I did, this was it. I studied with two professors —

Hoddeson:

Yes, I'd like to hear all about that.

Strömgren:

— Niels Bohr and H.M. Hansen. You know, Hansen had done some very fine work in spectroscopy. There is also a paper by Hansen and Mischner, which was well known in those days. He taught geometrical optics, so that was part of the curriculum. I have never regretted that I had to read that book for the examination. Well, we read, of course, Drude. Other books, just on geometrical optics, but that was —

Baym:

Did you study at the Institute? [The Institute for Theoretical Physics of the University of Copenhagen, now the Niels Bohr Institute]

Strömgren:

Yes. Actually, that year Niels Bohr didn't give many lectures, but he came regularly to the students' colloquium, once a week. In the beginning it was Niels Bohr, Hansen and Kramers. Then later it was Heisenberg, and when the time of my examination approached it was Oscar Klein.

Hoddeson:

Who were in the student seminars?

Strömgren:

Well, Christian Moller and I were contemporaries. He is a little older than I. Then there were a number of people who just left physics and became high school teachers. But some of course remained — for instance [ ] who's also older than I was. He was in the seminar group. At the same time there were other colloquia we of course could go to. There was, for instance the colloquium where Heisenberg first presented the uncertainty principle, in his form. Then we heard many lectures by Niels Bohr.

Hoddeson:

Were you very much aware of the great revolutionary importance of these ideas?

Strömgren:

Yes, I think in that year that we are talking about — this was far the most exciting, more exciting than regular astronomy. No question.

Hoddeson:

Do you feel that this year that you spent on physics, at that crucial time, played a very important role in your own career development?

Strömgren:

Yes. You see, when you look at the theory of stellar interiors and the theory of stellar atmospheres in those days — it was very much pre-quantum mechanics. It was quite clear that there would be very important applications of the new physics to astrophysics. Although I didn't immediately go in that direction, already when I prepared for the final exam I was sure that this was what I wanted to take up ultimately — simply, the application of the new physics in astrophysics. But that was not what I did immediately, when I came up for the final examination in November of 1927, and passed that.

Hoddeson:

Who examined you?

Strömgren:

Niels Bohr.

Hoddeson:

Niels Bohr examined you.

Strömgren:

Well, the written exams were handled by Oscar Klein, but it was Niels Bohr. He was also kind to the students. The month before the exam, he called me to his office — you know this story? — and he explained to me what was expected of me, and what were the most important things. Then the conversation ended with Niels Bohr saying, "And most important of all, (placing hands on forehead) mmmmm..." [This is a reference to Bohr's occasionally indistinct manner of speaking.] I didn't ask, you know. I never got to know what was most important. (laughter) Then, you see, the examination ends with a lecture, and for that I talked on Saha's theory. Then it was finished. But from then on, from 1927 to 1929, when I published the thesis and got my Danish Ph.D., my main work was in classical astronomy. My thesis was on determination of orbits of comets.

Hoddeson:

Let's spend a few minutes on your thesis. Who suggested the problem to you?

Strömgren:

It had, in a way, suggested itself, because the problems of work here was the work for what was called the Central Bureau of the International Astronomical Union. When my father was in Keil, the Central Bureau was also there. That meant that discoveries of comets, interesting asteroids and novae were sent by wire to the Center, and the Center then distributed it. They often looked critically, there were novae that turned out to be vague and that sort of thing. But at the beginning of World War I, it was clear that the Central Bureau in Keil couldn't function, so on his own initiative my father took it over and established it here. With some difficulties, he managed. Then he collaborated with Pickering.

Hoddeson:

What were the difficulties?

Strömgren:

Well, people would of course, even though Denmark was a neutral country, be quite suspicious of information that was channeled from one side of the belligerent countries to the other. But actually, it worked, and in that respect there was still the same unification. Any discovery that was made of this kind was distributed all over the scientific world. For the United States, E.C. Pickering at Harvard had established a similar service — the Harvard announcement cards — and my father and Pickering worked rather closely together on that. (end tape)

Strömgren:

In answer to your question, because of this activity, there was a lot of activity here providing the astronomical world as fact as possible with predictions of where new comets would be found in the next several weeks. That is, you computed an orbit and an ephemeris, and that was then distributed so that more observations could be obtained. It was really an exciting game. You waited for the first three observations to come in, and then you could determine an orbit. So already while I was still in high school I did some orbit calculations, and I was really fascinated by the numerical problems. This is what this was about. Today it looks very different, when you program a computer to do that.

Baym:

Did you carry out the calculations by hand?

Strömgren:

Yes, with a desk calculator. And one of the innovations in this were these nomograms for the solution of the equations. You find the final solution by trial and error, and you get very close to that solution, usually, by these diagrams.

Baym:

Is this a technique you had developed?

Strömgren:

Yes. Here is one of the basic texts on this by Paul Herd [Herget]. He bases his discussion on this. It has survived, in spite of the development of the electronic computer.

Hoddeson:

Who did you discuss your thesis with, as you worked on it?

Strömgren:

Well, really nobody. My father was an expert at making calculations of the classical types, with the table of logarithms, and there was a new trend to use the desk calculator. But in a way, this tradition of emphasizing numerical methods had developed already before my father came here — but during his time, it was I am sure quite important to the way I approached problems. And later, when I left classical astronomy and started work of the type which we were talking about — that is, applying the new physics to astrophysical problems — it was the numerical approach that I chose as the field I'd like to work in.

Hoddeson:

What were the circumstances that enabled you to get your degree so quickly — at 21, wasn't it?

Strömgren:

Well, yes — I think it was familiarity with the whole field for many years before I really started. There were two of us who took these two examinations that you take after two years, or a year and a half — there was the mathematician Ferguson. After that, since I didn't have to worry about the Master's thesis, I could concentrate on the new developments in physics.

Hoddeson:

Did you also learn some nuclear physics at that time?

Strömgren:

Not much.

Hoddeson:

That came later. Stromgren. You see, when you look at the state of the art in astrophysics — and we know very well what the state of the art is because Eddington published INTERNAL CONSTITUTION OF THE STARS, which for some years was the Bible on stellar interiors — what could be assumed in nuclear physics was really very little.

Baym:

Were you acquainted with the Agnes Clerk book as a child?

Strömgren:

No. Not at all. I read this in connection with writing the articles in 1950. You know, in the Hynek volume.^[1] That's when I read it. So it didn't influence me.

Hoddeson:

I see. Well, what about the writings of (George Ellery] Hale? When did you first study those?

Strömgren:

About that time. Yes.

Hoddeson:

In the fifties?

Strömgren:

Yes — no — but already in the — But you know, I didn't appreciate Hale enough, in those days.

Hoddeson:

Was he generally recognized in those days as the towering figure that we consider him as today?

Strömgren:

In America, yes; but not so much in Europe. I could have learned much earlier very important things, but — these things happen — my father just didn't like him. (laughter) He told stories — well, this is really off the record, but during his trip to America, he visited Pickering, and he got along so well with Pickering; he visited Frost at Yerkes, and they got along very well indeed. And then he came to Mount Wilson, and had a talk with Hale, who was just explaining how atrocious the Germans were, and was raging and saying, "We want no Kings and Kaisers," which was a bit peculiar, since after all, England was on the right side — Well, these stories left an impression of Hale, maybe, which was very wrong.

Hoddeson:

Do you know just about when your father went to America?

Strömgren:

Yes, this was in 1917, during the war.

Baym:

You stayed in Copenhagen?

Strömgren:

Oh yes. This was a time I remember so well, during the war, when my brother and I were here alone, and my mother and my father were crossing the Atlantic, during the war.

Hoddeson:

Did your father do a lot of traveling?

Strömgren:

Yes. He visited the States twice, which in those days was unusual, and he traveled a great deal to France and Germany, particularly Germany.

Hoddeson:

And whom would he visit there?

Strömgren:

He had many friends and colleagues before the war, in France; and he stayed at some of the observatories in Paris, and even in one of the provincial observatories in Toulouse. He told stories about provincial life in France. I remember that. After the war — in the period between the wars, he had very close relations with German astronomers.

Hoddeson:

With whom for example?

Strömgren:

There were, right after the war, two centers. One was Potsdam, the old Astrophysical Observatory, and the director there was then Luddendorf — a brother of the general, but very different from his brother, a very gifted man. The director in the other center, that was the New Berlin Observatory in Old Babelsberg which had been, just before the war — that was Guthnick, whom I mentioned before, and even when I was in high school, I visited both these places with my father, and then was left behind, and spent a week in one place and a week in the other.

Hoddeson:

Oh, that must have been very exciting.

Strömgren:

It was. I mean, these were great observatories. Yes. Compared to what you had here. And the development that was most important for my future work was that when I spent a week in Babelsberg, Guthnick, the director, taught me photoelecric photometry. As I mentioned, he was not the only one. During the war years, he was excused from military service for health reasons, so he could concentrate on photoelectric observations, both of magnitude and colors, and he investigated, among other problems, stars with very small amplitudes, where this technique was bringing quite new results. He and Stebbins and also another German astronomer, Hans Rosenberg, who was a wealthy man, independently wealthy, and had no positions, but just did this for fun, as happened in those days, they developed photoelectric photometry, and it was, in those days, a very primitive affair. They used a refractor that wasn't much bigger than the one here, and —

Hoddeson:

— the one here then, or the one here today?

Strömgren:

Well, the 13-inch that I mentioned, the so-called big refractor. And the photomultiplier — it was certainly not a photomultiplier, it was a simple photocell, with an amplification, because it was gas filled, that gave it an amplification of the primary color by a factor of 50, maybe up to a hundred. So the counts were very small, and they were then measured with an electrometer that was hanging off the telescope, in a carbon mounting, and what you did was to get the star shine on the photo cell, and then watch the electrometer. But that method gave 1 percent accuracy, there was linearity, and it was way beyond anything of the photographic — but it was limited to bright stars.

Baym:

Had you yourself begun to use this technique?

Strömgren:

Well, you see, this was when I was in high school. And what I did in 1925 was then to try to use this technique to record transits of stars. For that, it was necessary to introduce a DC amplifier. Those were just appearing, in those days, and the photocell was then coupled to a DC amplifier. The electrometer wouldn't do for recording that, so there was amplification, and the input to the amplifier was, the first tube was the so-called electrometer tube, very well insulated — so that you could have an input resistor of 1010 ohms, and that made it possible to record stars of third, fourth, fifth magnitude. Of course, you introduced a delay – 1010 ohms and a capacity of 10-11 farads, that's about a tenth of a second, so in the paper I published, I worried about that delay, and showed that under certain circumstances, when you recorded the variations of the carbon resistor — the star passes a grating, made photographically, and you focus — one could expect systematic errors from that delay, which you have to have, because otherwise, the voltage that you've got would be too small to measure. Well, I spent a lot of time on that, and I didn't leave it until, oh, about 1933, when my father had the [ ] instead of B. That was described in a paper published in 1933. Instead of recording individual transits over these edges, I introduced the periodic grating, and the switching between two photomultipliers, with a period that was adjusted to the speed of the star — this of course I built. And it's obvious that you have on one photocell the lag time, and the other, and the rate should be such that when they are equal, then you have that switch, just the moment when it comes out. And this meant that instead of concentrating on intervals of 102, you just integrated. I even tried then to do pulse counting, but never succeeded. You can imagine, with the amplification of 100, instead of what we have today, 106, you had to try to record very small voltages, of a millionth, and the only hope was that each photoelectron event takes place in a very short time. I never managed.

Baym:

Did you do your own electronics?

Strömgren:

Yes. Oh, I loved that. It was so different from electronics today. I made many DC amplifiers, and I — almost invented a negative feedback amplifier.

Baym:

Where did you learn the techniques?

Strömgren:

There was a man who was hired, an engineer who was hired to modify our time service — let's see, what was done first, as I explained, you determined time with a transit instrument. Then came the radio signals, and this was recorded. You had earphones. You made marks on a chromograph. But he was hired to do this automatically, wire an amplifier and amplify the signals so that you could actually count them, which was successful, and we worked together on this problem. And he showed me how to design DC amplifiers.

Baym:

You were how old then?

Strömgren:

This was in 1925, so I was 17. I never paid any attention to that. You see, it wasn't a difficulty for me, because it was an axiom, if we were interested in the same things — well, that was that. I don't know what impression I made on people, just having no inhibitions as a younger man should. I even remember, in that connection, when I was in Berlin, I must have been 18 or 19, I called up a man who was very famous, Manfred [ ]. He was a big man in amplifiers in those days. He explained the problems. He was rather kind, but a bit astonished. — (laughter)

Baym:

Did you visit him?

Strömgren:

Yes, and I brought back with me a device that he had developed. These were the bright filament amplifiers, but they were fitted together three tubes in one big clasp, that was his big contribution. And I also visited the engineers at Siemans to get the newest type of electrometer tube, one with high resistance. We were already testing. But, you see, in the period we are speaking about, after my examination, from 1927 to 1929, I was not much concerned with astrophysics.

Hoddeson:

Did your father also learn some of the new physics?

Strömgren:

Actually, he wrote some very nice popular books, and in one of them, he explains Eddington's theory, but he never tried to see what was happening in the new physics. No. And this he had in common with many astronomers of his age.

Hoddeson:

Did you start noticing strong differences between you and your father in your approach to astronomy at that time?

Strömgren:

That came in the period which comes next, after I had my Ph.D. In 1929. This was the time to choose a direction in which to go. It didn't take so long till I got quite absorbed in problems of stellar interiors. You may remember that this was a time when Gaunt published — this is TRANSACTIONS of the Royal Society — a paper on the calculation of the opacities. That interested me enormously, and here again, my familiarity with numerical problems from classical astronomy helped me a great deal. And it was quite clear, then — now we are in 1930-31 that the opacity question was one of the major questions in the theory of stellar interiors. This was quite clearly explained in Eddington's INTERNAL CONSTITUTION OF THE STARS, but there is a major discrepancy. Following the semi-classical approach, he had computed the opacity, and then he used it, with his assumptions of internal composition, and he came out with the wrong luminosity for the sun, off by a factor of, he estimated, of ten. He still wanted to attribute this to errors in the calculation of the opacity, because he saw that if you changed the chemical composition, and say, "Well, it isn't dominated by the heavy elements, there's a very considerable fraction of hydrogen," then you can make it fit. He knew that, but he didn't want to accept it, and one reason was his strong attachment to the role of radiation pressure.

And if you choose the other, [a high hydrogen fraction) that is not so. Then radiation pressures never, except in very exceptional, very massive stars, comes in. Whereas, the other, he put it this way in his popular writings: "As you increase the mass, radiation pressure comes into play, and what happens is the star." Now, he didn't want to give up that idea. So he stuck with the old composition. And there was at that time no firm evidence on hydrogen in the atmosphere. It was quite unknown what was the ratio of hydrogen to heavy elements. And if there were suggestions that after all, the hydrogen lines are so strong, and there must be more hydrogen, then people were saying, "Well, hydrogen is floating through dust —" (interruption)

Strömgren:

You see, in that situation, it was clear that the recalculation of the opacity, using the new physics, was important. And Gaunt made an important contribution to that, in that he had determined, to a very good approximation, the absorption coefficient in the continuous absorption for each of the continua, when you assume that an electron is moving [well in a box]. And this was in all the work at the time a basic assumption, and a very good one, because it was known that the degree of ionization of the matter in stellar interiors was high, and that most of the ions had very few electrons attached to them, so it appeared, to a very good approximation, was [ ]. Now, actually, even before Gaunt, in a publication that hadn't been generally noted, Sugiura had given the complete expression, without approximation. That was just one part of the story. To get the opacity, you had to combine your knowledge of the absorption coefficient electron in each of the states, with the distribution, the ionization distribution, and the Boltzmann distribution, and then add the continua, and finally strike the proper medium, which was known then to be the Rosseland — that's the way we got the stellar opacity. Now, there were conflicting results. There were, at the same time, in the literature results that differ from Gaunt's, and the first numerical problem I tackled was to verify Gaunt.

Hoddeson:

What were some of the opposing views?

Strömgren:

Well, there was a paper by Oppenheimer.

Hoddeson:

Really?

Strömgren:

It was wrong, but — we won't get into that — I never discussed that with Robert, but it was wrong. But anyway, that was the first work. And then — it's really such a simple problem, but it hadn't been solved then. Saha had guessed towards what the equation of ionization would be, and later, he had the right one — you can find that in R.H. Fowler's STATISTICAL MECHANICS, the second edition.

Hoddeson:

You mentioned Saha? Did he have a very large impact on you at that time?

Strömgren:

Yes. Actually, much of the general discussion in those days was dominated by the qualitative work of Saha. And the follow-up was the work by Fowler and Milne, and by Cecilia Payne — Cecilia Payne-Gaposchkin now. This culminated, in a way, with the publication in 1935 of Cecilia Payne's book on stellar atmospheres — all pre-wave mechanics, quantum mechanics.

Hoddeson:

Did you know all these people?

Strömgren:

I particularly knew Milne and R.H. Fowler. In England, in those days, Eddington was the great man. Their work was extremely interesting to me. I read all their papers. R.H. Fowler came quite regularly to Copenhagen. Milne also visited, but Fowler was a regular visitor at the Niels Bohr Institute. Not so much discussing these problems, but other problems, that were being discussed in those years.

Hoddeson:

When he came to Copenhagen, would you talk to him about astrophysical questions?

Strömgren:

Yes. You can find in "The Rise of Astrophysics" a reference to discussion with Fowler. You see, this has to do with the state of the art.

Hoddeson:

We're now looking at the paper published in the ANNALS of the New York Academy of Sciences, 1972.

Strömgren:

Yes. I think it's referred to in there. This is —

Hoddeson:

— page 252. I see. Yes.

Strömgren:

You see, the state of the art was that you chose one representative point in stellar atmospheres, and what Fowler and Milne had done was to take for each stellar atmosphere one temperature and pressure, and compute, much better than Saha did, the ionization equilibrium. And then they would follow for fixed pressure the variation of the lines, qualitatively with temperature, and particularly with emphasis on the maxima, and the location of the maximum would depend on the pressures, so that by comparison with observations, they got some idea of the pressures. And this was very important, because this was how it became clear that pressures were a good deal lower than had been imagined before. Now, that was just the beginning of the story, because they didn't quite realize then that clearly, what they were determining this way was the electron pressure, the [ ] and not the total pressure, and that for instance, in the sun, you have the ratio of 10,000 to 1, because of the very low degree of ionization of hydrogen, and the fact — which wasn't at all known then — that the fraction of elements contributing electrons is so small. It's smaller by a factor of 10,000. But you see, this paragraph refers to a discussion with R.H. Fowler. I was very enthusiastic about the idea, instead of choosing one representative point, including the whole model atmosphere integration, using the hydrostatic equilibrium equation for the pressure. I remember that there was an attempt in that direction described in a publication that wasn't widely read. It was an article by Emden in the ENCYCLOPADIE DER MATHEMATISCHE WISSENSCHAFT.^[2] You might like to look that up — it's in the middle of the twenties — and in those days, the ENCYCLOPADIE was an important publication. You remember, Pauli's theory was in that. And — well, in stellar interiors, the ENCYCLOPADIE didn't do so well. The time wasn't ripe for that. He was the one who wrote GOSKUGEL, — and by the way, you know, he is an uncle of Martin Schwarzschild's. Martin was Karl Schwarzschild's son and Emden was his uncle, and his father died before he could discuss it with him. He told me that Emden was a very tacit man, he never discussed any science with him. But that ENCYCLOPADIE article influenced me, in thinking that it could be done. This is what I discussed with Fowler.

Hoddeson:

Did you know Otto Struve in those days?

Strömgren:

Yes. Not personally. We got to know each other personally in 1936.

Hoddeson:

At Chicago?

Strömgren:

At Chicago. He — at the beginning of 1936 — wrote me a letter, inviting me to join the department of astronomy, which was being expanded, during the years 1935 and 1936. First Kuiper and I then Chandrasekhar joined the department. It already consisted of among others, of course — Otto Struve, and William Morgan, the spectroscopist, in K classification.

Hoddeson:

I have lots of questions to ask you about Chicago, but I think, I'd like to ask you one or two more questions about the period before you went to the United States.

Baym:

Where were you between 1929 and 1931?

Strömgren:

Here in Copenhagen.

Baym:

Did you have any position?

Strömgren:

Yes, I was an assistant in the observatory. And I had a fellowship, which in those days, was enormous — it was 1800 crowns per year, so I was well off. And then in 1932, I also became a lecturer. So when I left Copenhagen in 1936, I was an assistant and a lecturer — something of the same structure that you have today, without tenure.

Hoddeson:

How were the job opportunities at that time, generally?

Strömgren:

Very limited. Actually, the philosophy was that those who took their degree in mathematics, physics, chemistry or astronomy mostly prepared themselves for jobs as high school teachers, and the whole curriculum was worked out so that they would be competent. The standards were high in the Danish high schools, so this was a very good education. In astronomy, there were other opportunities. For instance, you could join the Geodetic Institute. But the main job was in high school, and to discourage people from choosing the scientific career they had determined then that the salaries of the assistants were just a shade lower than you got as a salary when you entered high school teaching. It wasn't looked down upon, but it certainly wasn't encouraged.

Hoddeson:

Was it considered not useful?

Strömgren:

No, I think it — these were the years of the Depression, and it was a matter of strict economy. For instance, it took a very considerable effort to convince the Minister of Education that there should be a lectureship in astronomy. It was turned down first, and then the university applied again and they succeeded. But the budgets of the university were really very low in those days.

Hoddeson:

Do you think the Depression greatly inhibited the development of astronomy in that period at all?

Strömgren:

Well, there was very little in the way of a development in astronomy at that time. The number of positions in the whole country was very small. There was no equipment to speak of. It was all antiquated, and what was done was that the activities were in the theoretical field, as a consequence. My father's work was in dynamical astronomy, and in the period we are now talking about my work was in theoretical astrophysics. We really didn't have very much to go on at all in that period. The Niels Bohr Institute was the exception, and —

Hoddeson:

Why?

Strömgren:

Well, it was Niels Bohr's personality, that meant everything. And he got support from various sides — Carlsberg Foundation, Rockefeller Foundation — and could maintain an activity at a very high level, in those fields, despite the Depression.

Hoddeson:

This is also the period, 1931, you published a book with your father.

Strömgren:

Yes, a textbook on astronomy.^[3] This had long been —

Hoddeson:

— here it is —

Strömgren:

You have the second edition, haven't you?

Hoddeson:

Yes.

Strömgren:

This had long been the desire of the students. They took notes from my father's lectures, but there was only an ancient textbook. This is the one that's referred to, in Norwegian.^[4] But what we did was to rewrite that, and my share was stellar astronomy and astrophysics, which in the first edition was a fairly modest contribution, but in the second, as you can see, it's a good deal more.

Hoddeson:

Did your father write the rest of the book?

Strömgren:

Yes. Well, we discussed every paragraph together. He rewrote the old Geelmuyden work,^[5] and incorporated his own lecture notes, and that's how that was written. Then, the other was just really written from scratch.

Hoddeson:

So there was a very specific audience that you were aiming the book at?

Strömgren:

Yes. Since, you know, at that time, there was still this first part examination, where there was really little choice, all the students in mathematics and physics had to take astronomy. So they needed that textbook.

Baym:

Had you studied the Geelmuyden text yourself?

Strömgren:

Well, yes, as a boy I had read it, sure.

Hoddeson:

We were discussing the textbook that you wrote with your father.

Strömgren:

Yes. It was of great value to me to write these parts on astrophysics and stellar astronomy. It meant that I really read rather completely all the literature there was on this subject, up to 1930. Obviously, I read much more than went into the book, but that was very useful, the undertaking to get familiar with practically everything that had been written. In those days it was still possible. Very strange.

Hoddeson:

Were you working with spectra then?

Strömgren:

No. There were no facilities here, and my familiarity with spectra was just what you could see when you visited other observatories. And that was rather limited. I had seen some of the work that was done at [ ] but it wasn't until I came to the States that I really could get familiar with observational material of that kind. But returning to the textbook, after the Danish had been published, there was a German edition.

Hoddeson:

Much longer.

Strömgren:

Yes, that was considerably enlarged, and came at a somewhat later time. A number of things had happened in the meantime. Then there was a second Danish edition which we prepared during the war. This is the book you have here, of course, published in 1945. There was also a second German edition, but Springer wanted to edit that, and the manuscript was complete and the illustration, but it was all destroyed during the war. So it never appeared. So many years elapsed before Springer was financially ready to publish it, that in the meantime astronomy had changed so much that it meant rewriting the book, and I really couldn't find the time for that, although they pressed for it. So that never came out.

Hoddeson:

Did you use your book in your own teaching?

Strömgren:

Yes. You see, my father retired in 1940, and from 1940 to 1950 I was responsible for the teaching. I taught both the elementary, the intermediate and the advanced courses.

Hoddeson:

Did you like teaching?

Strömgren:

Yes, I liked it very much. Even this elementary teaching, where I used the textbook, it was all right. We combined the teaching with exercises. They had to do problems which were of a very standard nature.

Hoddeson:

How large were the classes you taught at the time [1940s]?

Strömgren:

Well, the elementary classes were 30 or 40 people. And the advanced were never more than three or four. A number of very able people went through there, took their major in astronomy, and wrote theses of course. As I explained the Master's thesis isn't very extensive, and doesn't have to be publishable.

Hoddeson:

Did any of these people go on to make contributions to astrophysics?

Strömgren:

Yes, a few. But it was very limited —

Hoddeson:

Any particular example?

Strömgren:

Well, Professor Rud Kjobing in Ouchs was my student, and when I resigned from my position here in 1957, Professor Reiz from Sweden became professor of astronomy. He wasn't my student in the same way, but for a couple of years during the war, he came regularly and worked with me, research study. And of the younger people, for instance, Dr. Gyldenkaerne who is now — he was director for a few years, he is now out at Roskiid. He was a student and we worked together.

Hoddeson:

When you say you worked together —

Strömgren:

We published together.

Hoddeson:

I see. Do you like to work with collaborators?

Strömgren:

Well, up to a point. I haven't done too much of that. Most of my publications are alone. But in recent years, I've depended very much on it, because the work that's based on observations at LaSilla (European Southern Observatory] is mostly by younger people. I've gone there and observed, but I just can't do it anymore. It's quite hard. These people do what I did when I started this observation work. You have periods when you work 100 hours a week, for weeks at a time.

Baym:

When did you become interested in the problem of the formation of H II regions around stars?

Strömgren:

This happened during my stay at Yerkes from 1936 to 1938, when Otto Struve had developed a new technique for observing very faint light in the Balmer lines. Up to then, it was known that a number of diffuse nebulae, like the Orion Nebula, emitted an emission spectrum of strong in Chile Balmer lines. But Struve found that there were large regions of hundreds of square degrees that had a faint glow. He used a very interesting device — it was built on a slope of McDonald Observatory that pointed toward the Pole, at an altitude — I mean the slope was just right — pointed toward the Pole, and it had a focal length of 100 feet. For this reason it was called the Centipede. (laughter) But it was a way of recording very faint spectra of extended areas, and his efforts were successful. In a number of areas he discovered this glow, and it was of course challenged. He tried to explain why in some regions around the galactic equator you have this, and in others you don't. So the first time around, I worked out, a very simple model, where you had a luminous star, a high temperature star, embedded in a uniform medium of hydrogen, and the most striking result of the analysis was a very sharp transition. Qualitatively you can understand that at once — first, the ionization is very high. But once you reach about 1 percent neutral, then the ion absorption comes into play, and there's no UV radiation. So that this of course came out of the calculation — the transition from complete ionization to no ionization was a very narrow zone.

Baym:

You hadn't anticipated that it would be so sharp?

Strömgren:

Well, not before I — not that sharp. Not before I did the actual calculation. But then it was of course quite understandable.

Baym:

Was this a problem other people were working on?

Strömgren:

Well, there was a very interesting paper by Eddington on diffuse matter in space, but it didn't really discuss these problems. At that time, the belief that was generally held was that the density of interstellar gas was very low. For a long time the only lines, the interstellar absorption lines, were calcium and sodium. And they were thought of as perhaps exceptional. They are the ions that have resonance in the high intensity part of the spectra of certain flare stars. And therefore radiation pressure might be important in driving them out. Well, the explanation is of course that, for the same reason, they're the only ones that we're observing, of stellar absorption. But in the thirties, many great efforts were made at Mt. Wilson at adding to the number of absorption lines, and Adams and Dunham contributed a good deal, so that a number of lines were identified. But still, this all didn't lead to the conclusion that there was much mass, and it hadn't been assumed that there was much hydrogen. Now, the analysis of the ionized regions immediately gave the number, and it was close to what we knew, in those days, was the upper limit to interstellar matter, from Roth's analysis of the oscillations of stars at right angles to the plane of the galaxy. You can find the mass per unit area.

And the stars contribute some of that, much of it in fact, at least one half. And the result sets an upper limit to how much interstellar matter there could be. And the interval between what was established in the ionized regions, and what was the maximum was not that large. So right away it was clear how much hydrogen was there. And since Struve's results referred to very large volumes, there were substantial amounts. Now, the analysis of the ionization limits suggested that there would be neutral hydrogen everywhere. And that of course led to the belief that there was indeed sufficient matter there for star formation to take place. It is I think interesting to remember that, in the twenties and early thirties, there were no compelling reasons to assume that we have star formation in the present phase of the galaxy. The estimated age of the galaxy was then, according to all indications, three billion; the age of the earth, three billion; and the Hubble Constant led to an age of three billion. On the other hand, when you look at the estimates of the lifetimes of the sun, how long it could live on nuclear energy, with no specific knowledge, most people thought, well, it must be some process that converts 1 percent of the mass. That was reasonable, and you have it all available, because there is some mixing in the stars, and that gives 1011 (years]. So that for the solar type star evolution was nothing.

And then, there were the very luminous B stars, what about those, that consumed their energy more rapidly? But even there, the evidence wasn't too compelling, and one reason was that we really couldn't understand the giants and globular clusters. They were as bright, and yet, from their distribution in space, it looked very much like they were the oldest. It wasn't known that they just spend the last bit — the last fraction of their life at high luminosity. And it was really only when the nature of stars on the main sequence was understood, after Bethe, and after the correct models had been evolved —

Hoddeson:

In the very late thirties?

Strömgren:

Yes. And also after it became clear that mixing wasn't too important — it would only be say, for a sun, 10 percent would be available in the main sequence phase — that it became clear that evolution would give the lifetimes of the same order, at least, or maybe would be shorter than the lifetime of the galaxy. The revision of the ages came in the fifties, with the step by step revision of the Hubble Constant. But it's quite clear that, particularly in the twenties, when people like Jeans would say, "Well, why should it be limited to 1 percent? The whole mass can be converted into energy." He gave, in his popular brooks, strong arguments in favor of lifetimes of 1013 years. So to return to interstellar matter, there wasn't the strong compelling reason to assume that we must have some. People weren't so worried about the lack of interstellar matter. But that all changed in the late thirties. We knew, then. The final evidence came of course when the 21 centimeter line was observed. Then there was no doubt about it. And the doubts, in connection with how much interstellar matter is there, were then, in the next period, all connected with molecular hydrogen. People were saying — "All right, we find the atomic hydrogen, but perhaps most of it is molecular." And the lack of correlation between reddening and 21 centimeter emission was a contributing factor. The escape was of course — and that has proved to be right — that most of the molecular hydrogen is indeed in the regions with high reddening. The particulars are evidently active in forming molecules, and they also protect them from the dissociating radiation. So that the Copernicus [satellite] results strongly suggest that whenever you have most of hydrogen in molecular form, it's when there is also a dense particle medium around.

Hoddeson:

What attracted you to America and to Chicago in particular in 1936?

Strömgren:

Well, I must say, I'd always hoped to go to America. That was the thing to do. And Niels Bohr had talked to me about getting a Rockefeller Fellowship, and I'm sure he could have gotten it for me, but I wasn't quite ready. One thing and another had to be finished, the textbook, and so —

Hoddeson:

You got married too, in the meantime.

Strömgren:

Well, that wouldn't have prevented it. I was married in 1931. I can tell you in that connection, Martin Schwarzschild and I were talking about the good old days, when you saw to it that you had a job before you married. Yes. My Ph.D. ring is dated the 12th of December, and my engagement ring the 14th of December. So I did that in the good old way. Our oldest girl was born in 1933, and it wouldn't have been very easy to go to the States then. Our youngest was born in 1935, and the offer to join the department came early in 1936. Struve wanted me to come for three years, but then in the end it was a compromise. It was a little difficult for me to be away for three years, so it was a year and a half.

Hoddeson:

Then you definitely considered it to be a temporary experience?

Strömgren:

Yes. Very definitely, yes. And this of course was one of the most important opportunities I had in my whole career — that I joined the University of Chicago and the Yerkes at a very exciting period, both in the life of the university, and particularly in the life of the department.

Baym:

Were you living in the city?

Strömgren:

The first six months I was teaching on the campus, and we were living in Dorchester.

Hoddeson:

What were your expectations of astronomy in the United States, before you came?

Strömgren:

Well, most of all, I wanted to have close contacts with the efforts in observational astronomy, which were ever so much more important in the thirties in the States than in Europe. The efforts that have now culminated in ESO [European Southern Observatory Association] and in three great national projects — the Anglo-Australian, the French-Canadian, and the German-Spanish — were non-existent, and it was only in the late thirties — say, the Observatoire de Haute of Provence was founded, and didn't come into operation until after the war. So there was very little in Europe, and I felt, particularly in confrontation of not only the results of stellar interior calculations, but particularly when I went there, the results of stellar atmosphere models, with the observational results, was what I would concern myself with.

Baym:

This was a very different situation from that in physics where Europe was so much stronger.

Strömgren:

Yes. And — why, I remember so well, when Niels Bohr returned from rather an extended stay in California — this must have been the end of the twenties or early in the thirties — he was not much impressed with theoretical physics in the States. He mentioned Robert Oppenheimer. But then of course came the influx of people, after 1933 and even before that, in Chicago, when I came, there was already strong physics. Mullikan was there, and Compton and Allison in nuclear physics.

Baym:

Did you have contact with these people?

Strömgren:

Yes. Actually when I was teaching on the campus, I saw them quite regularly, as I saw the mathematicians.

Hoddeson:

We're very much interested in some of the foreign — trained astronomers who passed through Chicago in this period. We began to talk about that earlier.

Strömgren:

Yes.

Hoddeson:

But we didn't finish, and some of the men I have listed here include Hans Rosenberg, whom you mentioned, and Gerald Kuiper.

Strömgren:

He had come from the Netherlands, and as you know, a number of astronomers from the Netherlands left for the States. There were too many of them to find positions in their homeland, and they contributed very much to astronomy.

Hoddeson:

Was that particularly so in the Netherlands?

Strömgren:

Yes. My impression is in those years, the years of the Depression, there was an over-production of astronomers, and also to some extent physicists, but they were absorbed more so than the astronomers — they could work with physics in industry. But a number of astronomers left, among them Kuiper who was first on the West Coast, and then — well, he was both at Harvard and the West Coast, and then he was offered this position by Struve.

Hoddeson:

Let's see. There was also Pol Swings.

Strömgren:

Yes. He was not a permanent staff member, but he did get to spend more time than he had expected, because of the developments during the war. And he collaborated closely with Struve on problems in spectroscopy, particularly analysis of molecular lines, and if we return to the subject of what happened in the theory of stellar atmospheres, we will come to one important contribution, due to Struve and Swings. Their work on molecular lines gave at least a rough estimate of the abundance, relative to hydrogen, of oxygen.

Hoddeson:

In what period did they do this?

Strömgren:

Well, this was in the late thirties and early forties. Do you have any other names of those —?

Hoddeson:

Oh, I have a list. This is not a complete list. You can just look at it.^[6]

Strömgren:

Strand was a Norwegian who spent some time, and also longer than had been expected, because of the war.

Hoddeson:

Did you interact with Chandrasekhar?

Strömgren:

Chandrasekhar came at the same time, about, as I. I came in October, and if I remember right, he came in December, the same year.

Hoddeson:

Gerhard Herzberg.

Strömgren:

Yes. That was later, however. I met Herzberg at Yerkes when I went the second time, in 1947 and 1948.

Hoddeson:

And Carl Osborne?

Strömgren:

Yes. He was a Danish astronomer who got his training in Holland, as an assistant to Hertzprung, and he took up the specialty of Hertzsprung which was photographic determinations of relative positions of visual binaries. He worked on that, and on other astronomic programs, at Yerkes. Though he came to Yerkes a good deal later — that was also during my second visit, 1947-48, but he was there. He was at Sproul first.

Baym:

Had you had contact with Hertzsprung yourself?

Strömgren:

Yes. He was a fairly frequent visitor to Copenhagen. And he also visited Yerkes during that period that we are talking about. And his great contributions were then something that had happened many years ago, and actually, he was a man who had not really changed his field altogether, but nevertheless, of course, what he did in the early days had a tremendous impact.

Hoddeson:

Did you interact much with these very productive men?

Strömgren:

Well, the period that we are talking about, from 1936 to 1938, was in a way dominated through the new opportunity of reacting closely with people like Otto Struve, William Morgan, who had been there, Chandrasekhar and Kuiper.

Hoddeson:

How did you spend a typical day, if there was such a thing at all?

Strömgren:

Well, the first six months, I actually taught.

Hoddeson:

Where did you teach?

Strömgren:

On the campus, one elementary and one advanced course.

Hoddeson:

In astrophysics?

Strömgren:

Yes. I spent weekends at Yerkes. We were given a house we could occupy on weekends, and that's when, already during the first six months, I saw all these people. Struve came very often to the campus, so I saw him there too.

Hoddeson:

Now, Struve was chairman.

Strömgren:

He was the chairman, and director. Yes. He had succeeded Frost that I mentioned. Actually the history of the Yerkes is very interesting. An enormously promising staff, that was then — there was almost a complete interruption when Hale left, and there was an exodus of Hale and a few other key people.

Hoddeson:

When was this?

Strömgren:

To Mount Wilson. This must have been around 1905, when he left. The observatory was founded in 1893 — all that you'll find very well described in the articles. During Frost's time, there were some important developments. He was a spectroscopist, and worked on radial velocities which was an important new field, and on spectroscopic binaries. There was a group of people, who, for instance, continued earlier work by Hale on the sun, but it had deteriorated to quite routine work. And when Struve took over, there was very little activity. So he got Hutchins' cooperation, and was promised an expansion.

Hoddeson:

Hutchins was the president? (of Chicago University)

Strömgren:

Yes. Robert Hutchins had just come from Yale to the University of Chicago, a very young president.

Hoddeson:

I see. So he built up this whole —

Strömgren:

Struve started then, in the middle thirties, to build up the department, and the most promising development, from the point of view of the observational astronomer and the theoretical astronomer interested in collaboration with the observational astronomer, was the development of McDonald Observatory. This happened because McDonald willed a good deal of money to the University of Texas, and they had no astronomy to speak of, and therefore, Hutchins managed to persuade them that there would be an agreement, to run for 30 years, that the money would be used to build an observatory that would operated jointly by the University of Texas and the University of Chicago. And that of course meant that it was really up to the department of astronomy in Chicago to develop and use the observatory. And this was the situation, even when I came back in 1950, and it remained so during the time of my directorship. But then, a few years later, the contact was not renewed. It expired, and Texas was ambitious, and also, they had the money to build up a staff, and they could appoint good people. You know perhaps a little better the whole history of the University of Texas. There was a breakthrough around that time, and in a number of fields they became quite prominent, whereas when I visited the first time, it was a very sleepy place. This was in January of 1938.

Baym:

How often did you go down to Texas?

Strömgren:

During the first period, I was only there once, and there was nothing on the mountain yet except a dome and a few houses, no instruments to speak of. But when I came back in 1947-48, it was in full swing. Then I had an observing period with the 82-inch telescope. And in the fifties, when I was there, 1950-57, why, it was the place to go. So we had a house there and spent a month at a time, a couple of times a year. This is how it worked. I didn't quite answer your question on how I spent my days. Well —

Hoddeson:

—I'm trying to just get an impression of what working at Chicago in that period was like.

Strömgren:

Yes. The first period, the first six months, wasn't typical. I mean, I was then working on some problems of stellar atmospheres, that I discussed a good deal with students, and of course, starting two new courses took some of my time.

Hoddeson:

Did you have freedom to work on anything you wanted to?

Strömgren:

Yes. This was a principle — that you were not in any way asked to participate in particular work. You chose your own work, and that was so for every member of the department in principle. Some of the members of the department clearly chose to collaborate with senior people, but they were free. This was so throughout in principle throughout the university.

Hoddeson:

About how large was the department?

Strömgren:

Well, I think we were about 12 people.

Hoddeson:

Did that seem large to you?

Strömgren:

Oh yes — coming from the small European places. It was.

Baym:

It still is a very large size for an American astronomy department.

Strömgren:

Yes. And this meant that the teaching curriculum, and this activity — teaching to the graduate students at Yerkes started when I got there in April, 1937, and went on for a year — I think that was a breakthrough in America, because in those days, there was no such teaching on the West Coast. The Mount Wilson people weren't in the least interested in that. That would disturb their activities of observing, utilizing Mount Wilson and the coming Palomar. And it was only in the fifties when Jesse Greenstein went there that the department of astrophysics was built up. And similarly at Berkeley. There wasn't very close collaboration between Lick and the campus in Berkeley. The campus in Berkeley was small and dominated by a classical astronomer, who worried about orbits of comets and asteroids and that sort of thing. Leuschner was his name. On the East Coast there was Harvard, but they hadn't yet reached the point where they would give a broad-based education in physics.

Baym:

Did you travel much throughout the United States in this period?

Strömgren:

No. Only at the end, toward the end, on the way to the boat, we spent a few weeks, and my only travel at Yerkes was neighborhood meetings and then one trip to Texas. But that trip, at the end, was very interesting, as I said. Tellar and Gamow had organized a meeting on stellar structure in Washington; among those present was Hans Bethe. It was just before everything became clear, so it was very interesting.

Hoddeson:

This was in 1938?

Strömgren:

Yes, in April of 1938.

Baym:

Was that your first meeting with Bethe?

Strömgren:

Yes. Well, it's quite possible that we had been in the same room, in Copenhagen, but it was the first time I had really talked to him. But then I spent some time at Harvard also, talking mostly with Shapley. Shapley had made an effort to get, first, Rosseland to get to go there, and then Struve, but he had failed, so at that time, it wasn't a very strong teaching department, and I would say, what we did at Yerkes was better. That was so much in the picture, in the thirties — the rivalry between the University of Chicago and Harvard, when Hutchins time and again would say, "Why second best?"

Baym:

Had you any contact with the astronomy departments of other universities in the midwest?

Strömgren:

In the neighborhood meetings — but not so much. More with the University of Wisconsin, because Stebbins was there, and I was very interested in his work.

Hoddeson:

How about Russell at Princeton?

Strömgren:

Yes. That is the other thing to remember, when you think of the teaching in astronomy, of course. It was very small, but it was a very important place in those days, and Russell, as you know, traveled regularly to Mount Wilson, and he had close collaboration with Shapley. So his ideas influenced work in both places. It influenced Shapley's early work, because before he worked on globular clusters, he made a very interesting contribution on eclipsing binaries, inspired by Russell.