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 Lyman Spitzer by David DeVorkin on 1991 November 27, Niels Bohr Library & Archives, American Institute of Physics, College Park, MD USA, www.aip.org/history-programs/niels-bohr-library/oral-histories/28612
For multiple citations, "AIP" is the preferred abbreviation for the location.
This interview centers around a discussion of Henry Norris Russell, and his influences as Director of the Princeton University Observatory. In this interview, Spitzer also addresses: his education at Yale University; the relationships between Russell, Sergei and Cecilia Gaposchkin, and Zdenek Kopal; and his lifetime of research on stellar evolution. Other topics and affiliations discussed include: Subrahmanyan Chandrasekher, Ray Dugan, Newton Pierce, Martin Schwarzchild, and photometric research.
I hope to gain a deeper appreciation of the flavor of the Princeton University Observatory under Russell. My goal is to gain a deeper appreciation of the flavor of the observatory atmosphere under Russell in the late 1930s and when you came back in the forties. My first question is to what extent do you feel that Russell directed the research effort of the department when you were a graduate student?
That is an interesting question. If I just think back to those days when I was a graduate student there wasn't much for him to direct. Dugan was working on eclipsing variables and I assume he kept Russell informed. I don't recall any discussions between them. [interruption] In case of Dugan, Russell may very well have pushed him into eclipsing variables or maybe Dugan did them voluntarily.
I doubt if Russell told Dugan what to do in any detailed sense. Russell steered the department by inviting people to come and work there — Rupert Wildt and Charlotte Sitterly were both at Princeton and I believe that Rupert talked to Russell fairly frequently but Rupert's field of expertise was in somewhat different areas and I doubt if Russell tried to direct him, but on the other hand Russell always had a flow of ideas on any topic. So Wildt was likely influenced by his conversations with Russell.
But this is not quite the same as the picture of Russell guiding people's research. Well, he didn't guide the Mount Wilson people either but he made a lot of suggestions which were very important to them in their choice of programs.
That has certainly been my understanding. But in the case of Dugan, who was the only one engaged in programmatic research...
That's an exaggeration because there was also Newton Pierce and John Stewart.
Yes, but to finish with Dugan: the choice of eclipsing systems that he chose to do photometric work on — was this influenced by Russell?
I can well imagine that Russell would have suggested topics that would be theoretically be interesting and Dugan from that list would have chosen those which for one reason or another were more practical to observe.
Yes. Would this be true of systems likely to show apsidal motion?
Was that a topic that was of interest to graduate students when you were here?
I don't remember it as a central issue at all. Russell talked about many different things. He didn't give apsidal motion any more emphasis than he did the principle of least squares analysis for interpreting observations. He was enthusiastic about any phase of his work or about any phase of the topics he was teaching if he himself had worked in that area.
You came here after a year at Cambridge. What degree of freedom did graduate students have to pursue course work that they wanted to take, whether they were working with Dugan or Russell, or even Stewart. What freedom did you have about your area of interest and your thesis topic?
I have little detailed recollection, but as far as I can recall I was not then engaged in any research as a graduate student. The program was quite different from what Martin and I set up after the war. We felt that a program of research for graduate students before they started their PhD thesis was an important part of their training. But my lack of recollection of doing research may have been partly because of my own status: It was during my first year here studying and going to lectures in theoretical physics, that I took my general exams in theoretical physics. From there I went directly to writing a thesis. So actually I was not formally a student in astronomy. So for that reason maybe my own experience was not typical.
You came in the fall of 1936 and by November 1937, Russell wrote to F. H. Seares saying that "Spitzer is working enthusiastically on the tracings from Adams' splendid spectrograms..." etc.
That's right. At the end of my first year I took the oral and written generals in theoretical physics. Questions that Condon asked me, and I suppose Shankland, Ladenburg. Possibly Wigner too, I don't recall. But I went to lectures by Condon and Wigner, and talks by Einstein and various visitors, and I also went to talks by Russell, I don't remember which, though he and Wildt must have given a series of lectures or a course in theoretical astrophysics. Then in the summer or spring of 1937 Russell came back from Mount Wilson and was very enthusiastic about spectra that Adams had obtained and he said that he had Adams' permission to use them for my thesis. I wasn't altogether enthusiastic.
It did not seem like much of a theoretical problem, but then I had very little contact with interpreting observations, and so perhaps it was good experience for me. In any case I went out to Pasadena during the summer to make photometer tracings of the spectra of the stars — the summer of 1937 — and then I came back and worked on my thesis during the fall. Russell was away during that period so I had very little contact with him when I was working on my thesis.
Is is correct to say that after you attended the lectures of Condon, Ladenburg, Wigner and Russell that you chose to work with Russell on your thesis?
I went to Princeton with the goal of working with Russell. I'd heard him talk in Boston when I was a student in Andover. I had started in astrophysics when I was in England, pushed into that by R. H. Fowler, somewhat to my surprise, but in any case it all sounded very interesting. But I thought that doing course work in theoretical physics might be a better training for theoretical astrophysics, rather than going into straight astronomy.
You knew that yourself at the time?
I think so, yes. But I remember telling myself that it was better to be a big frog in a little puddle than being lost in a big puddle. Physics is an enormous field and I felt that there was more chance to rise in an area like what Russell had been in, rather than trying to compete with Wigner and Dirac. [laughing]. And in retrospect, if that was my opinion, and I think it was, I think it was wise.
Very interesting, because I'm trying to better appreciate how a graduate student in that day, under Russell or at Princeton, or just contemplating a career, makes choices.
I think my situation was so atypical. It was true at Yale as an undergraduate. My progress through Yale was ???, and that was true at Princeton also. It was not to my credit — it just worked out that way. Because when I wrote my thesis it was officially under the physics department, but Russell was my adviser. When I had it all written it was sent to the physics department and they said: "My god, we can't give this man a degree in theoretical physics on the basis of this thesis. Its astronomy."
So they got together with Russell and asked: "What are we going to do?" My understanding was that as a compromise they decided to give me a degree in theoretical astrophysics, which they had never done, and which was not an officially recognized field. So that was done, and I don't know if it was a joint recommendation from the two departments, or whether the physics department tossed it over to Russell.
That is really interesting.
I had very few comments on my thesis from Russell, I might add. I was a little disappointed about that. But I had very little interaction with him on the dissertation. He was away for most of the year.
He was away more than normal?
Usually he was away for half of each year at most. I had quite a bit of theoretical discussion in the thesis: I had a fountain model of the atmospheres of these three supergiant stars I was working on, and I don't remember much discussion with him. He must have had some comments. At one point considered putting in a page from "Alice in Wonderland" into my thesis just to see if anybody noticed. But I didn't have the nerve. I'm not at all sure that anybody would have commented on it! [laughter]
Was that a moment of caprice, or was it something about Russell's apparent distance from your graduate training?
Well, I had seen so little of him in connection with my own work. Now, I saw a great deal of him concerning matters that he was interested in. This was not on a one-to-one basis, but I attended talks that he gave. We also had a little seminar at one point concerning needed changes in "Russell, Dugan and Stewart." Fascinating to watch, because it gave him a chance to shine in each separate department of astronomy. It was wonderful, I enjoyed it enormously. I think he enjoyed having me there because I was so enthusiastic about it. So he always talked directly to me. I was the really responsive student, and, you know, one tends to address remarks to that student who is most responsive. I was very responsive, very enthusiastic, thought he was wonderful.
Your recollection of his distance is not an unusual one among those I have asked this question.
The reason I say this is to ask: why do you think he chose you to analyze those coveted spectrograms from Adams? There were actually other graduate students who wanted to do it.
Oh really? That of course was totally unknown to me. I had no idea. Well, I think I was more trained in theory than the other students in the department.
Did Russell tell you what to do with the spectra?
I don't think he told me. He may have suggested something. Just "fascinating problems" possibly — he may have suggested doing just what I did. I don't remember any particular discussions.
Did you know Sydney Hacker?
No I did not. I think he did Alpha Bootis. Arcturus.
His thesis was very different —
He identified lines — well, that was not my idea of theoretical astrophysics. Theoretical astrophysics — well, there was not much of it then, just Russell and Chandrasekhar. I remember very clearly a talk that was given at that time by Schilt from Columbia. This was at the New York Academy of Sciences. He compared a paper that had just come out on the Eros parallax campaign summarizing the results, and a paper by Chandrasekhar and Henyey [Henrich?] on the nature of the prestellar stage.
It was beautifully speculative but had lots of physical theory and sort of elucidated what was happening when stars were forming — advanced theoretical astrophysics — whereas the Eros campaign a lot of leg work, a lot of plates measured, and when it was finally added up it did not really improve the solar parallax that much, but it had involved people from all over the world working on these astrometric problems, the old-style astronomy. And Schilt compared these two papers - the old style astronomy and the new.
And what was Schilt's comparison?
Well, it was just how exciting the Chandrasekhar-Henyey paper was but it was a minimal amount of effort, not very substantial, but at the same time it was very exciting to Schilt. At least I think that was the general tenor of his remarks. These new methods of approach to problems of the sun and stars.
Did you know that the Princeton astronomy department library, when I came here, was divided into two indexes? "Astronomy" and "Astrophysics" — and they were quite separate. The astronomers had been unwilling to concede that astrophysics was a legitimate discipline, so it was brought in as a wholly separate topic.
You mean that for all the years that Russell was here?
I guess so. But its genesis was before he became director, and he made no attempt to combine them.
Amazing and curious. Now Newton Lacy Pierce was one of the graduate students who also came with a master's degree from the University of Michigan. He had done Eros work at Dearborn observatory at Northwestern for Fox.
Yes, that was a widespread interest among many astronomers at the time.
He arrived in the fall of 1935, got his PhD here in 1937, and when you came he was already an "associate instructor."
Oh, I didn't know that such a designation existed!
Yes, and when he came here, the contrast in the type of course work he took, and most of the others, like Hacker, was significant. Hacker and others took good doses of theoretical physics and astrophysics courses, as you did. The only courses Pierce took outside of double star or photometric work was one semester with Russell on general astronomy. Then he was hired on the staff. I'm trying to figure out why he was hired and then consistently promoted.
A good question. Well, Russell was not a very good judge of people.
But one hires and promotes people for many reasons. Pierce was a very very helpful and useful person, and, as you note in a later evaluation in the 1940s, he arrived at a time when Dugan was suffering terribly from arthritis and Newton Lacy Pierce literally took over the operations of the observatory.
He lived at the observatory.
And that included organizing the offices and maintaining the library. You also wrote probably the most sympathetic review of Pierce in all those years.
Even though Russell always pushed his promotions, he knew that Pierce was certainly not a very hot prospect. Dugan wrote very lukewarm notations on Pierce's doctoral exams, for instance, both the first half and the second half. They thought well of him because he worked so hard, but he took an enormous amount of time to get anything done. So here is my theory: Pierce made himself extremely useful, he made himself indispensable. Is this a fair assessment?
Oh I think that is probably correct. By the time I came on the scene, his indispensability was no longer a factor. For Russell he was indeed very valuable keeping Dugan's program running after Dugan died. Pierce carried on the program, but not with any great brilliance, as far as I was concerned. We could easily drop the program, in my opinion. I enjoyed managing the library and maintaining it as well. So I would have been quite happy if he had left, but I didn't think it was likely that he would. So I tried to keep him happy. He did some useful work with his photoelectric photometer that he built — it was good in principle.
Could you expand on the photoelectric work. Two people worked with him? He was building a digital two-channel photoelectric photometer. A pulse counter. This seemed quite advanced.
Yes, but the electronics were going to be done by Blitzstein and Leavitt, who I think were from the University of Pennsylvania department of astronomy. We finally gave up the photometer, and I think gave it to the University of Pennsylvania.
What was Pierce's role?
He wanted to observe variables, they wanted to build the photometer, and may have used it finally at Penn. Newt was the person pushing it. This was technology at the forefront. A very difficult enterprise. I recall that it did not work very well while it was here.
Russell believed in the value of efficient, approximate methods of reduction and analysis in astronomy.
Oh yes, he was wonderful in devising simplified methods.
He believed this to the end of his life. But Kopal and Russell did not see eye to eye on methods of eclipsing binary reduction.
What do you know about their differences?
Not much. I knew Kopal. I met him when I was at Harvard as a postdoc. But I had little interest in the matter.
But what about the personalities on both sides? Russell's research agenda, I feel, went far beyond Princeton, and extended to Harvard at least.
Oh yes, Harvard, and Mount Wilson. I'm sure. He was a national figure.
And the research teams that he built up included Harvard staff for binaries, Princeton and Mount Wilson and staff elsewhere for spectroscopic line work — Shenstone and Dodds, Meggers and others —
Catalan was not part of Russell's team I think. He was more in a position like that of Rosseland. My feeling is that Russell fit people into certain niches to perform certain functions...
I wonder about that. It doesn't quite agree with my picture of Russell. Russell's magnificent abilities were in the field of ideas and how to solve problems in very effective ways and to get sweeping results. But that he never had any interest or ability at what I would call administration. This refers not so much to assignments given to the secretaries, but to actually choosing who does what and getting them into the right field and being responsible for them and getting them to take part in an organization. From my knowledge of him in Princeton I don't think he was very good at that.
He could push people that he sensed would relieve him of some of the less interesting phases of the work. But to say that he assembled teams suggests that he was the manage of a project and that he told people what to do. Whereas my picture of his contributions is rather more circulating among other people talking about ideas, what they ought to do: "You ought to do this" "You ought to do that" and pushing on the general scientific ideas but not really putting pegs into holes, or thinking about how they were going to relate — things that high-level scientific administrators do.
Could it be that the management of the binary work at Harvard was due to Shapley and not so much Russell?
That could be. Russell could have pushed on Shapley and then Shapley was responsible for pushing the people around.
That I could well believe.
But in the case of the line spectra, there is quite a bit of correspondence where Russell expressed frustration, trying to secure services of a physicist at Toronto, for example where there was an ultraviolet, laboratory facility, vacuum tanks, — he was trying to get people to work on iron, for instance.
That does not surprise me. He would go after a specific piece of information. But to get a team or the concept of a team organized by Russell, that does not agree with my impression of the man. But when a few people exist with a capability Russell would push in that direction to get people to do very specific things that he needs for his own work.
I see, yes. So he did not create teams for the purpose of having teams...
At least he did not think in those terms and he did not enjoy that occupation, as some people do. He considered organizing only for a particular scientific project where he wants the data and will push on anybody who can somehow work with others to provide it.
Now in the case of Kopal, Kopal simply would not follow Russell's direction or do —
Do what was needed for Russell, yes.
Not only that but there seemed to be something else I need to have clarified. Russell seemed to be annoyed with Kopal and Gaposchkin because they both were sloppy. Publishing stuff with numerical errors.
Russell dismissed Sergei Gaposchkin, but Kopal was brilliant, and Russell felt he could be valuable if he would just play ball. [talks about the Russell/Kopal relationship, and that there was no way Kopal was going to get around Russell since both Shapley and Struve sent Kopal's manuscripts to Russell for vetting. Also talks about what each thought of the other] — Spitzer was at Harvard during some of this. Any awareness of what was going on?
Well, the tension between Cecilia Payne Gaposchkin and Russell I had been made aware of in recent histories of that period — how Russell dismissed some of her ideas about hydrogen. [[Feels this happens all the time]] — its hard to give credence to the unusual ideas of other people.
But this is the late 1930s early 1940s, when you were at Harvard; Shapley, however, brought in Cecilia to manage Sergei and Kopal for the Harvard attack on eclipsing binaries, which Russell wanted.
I was not at all aware of this. I had dinner at the Gaposchkins of course. There was one event where Martin [Schwarzschild] was there, and Kopal was there. An international dinner. The three Gaposchkin children running around throwing things. General chaos.
So there was no evident tension.
None that I saw. Of course I would not have been involved. I was a youngster as a postdoctoral fellow. And I was a Russell student and was then corresponding with Russell about a lot of things.
Yes, there is nothing in your correspondence that suggests you were involved. He complained bitterly to Shapley, however. This went on through the 1940s. It drove Russell back to eclipsing binaries. At one point he said to Shapley that "I've never dropped eclipsing binaries. So when Kopal's work appears I will take it up" — and he did — he checked and rederived everything. Of course he did this sort of thing with everyone.
Yes, I know. Some referees do that. I can't claim that approach.
What is curious, however, if that he would do this, yet as you and others have said, he did not pay that much attention to your thesis.
Well, my thesis was in a field where he had not worked and had no particular views. And of course he was getting on in years.
Yet he asked you to do this.
Well, he thought it was spectra, and he thought I would do something interesting with spectra.
So you would have had almost complete freedom, once the thesis topic was decided.
Yes. There the spectra were, and I enjoyed working on it. But the analysis of experimental data was something I had never really had been exposed to. I didn't have any clear idea of what the theoretical ramifications would be involved. Since then I have indeed put more emphasis on observational astronomy than I did at that time. I am more aware of what Chandrasekhar refers to as "observational guidance" for the theorist. So in this sense, the choice of the thesis was a positive step. And as it turned out my theoretical work in that field is not of much importance — it is rarely referred to.
Talking about your thesis, you were dealing with expanding atmospheres. Mira-type variables
Mira? — really the long period supergiant variables. The biggest atmospheres known, alpha Orionis extends out beyond the orbit of Mars...
Were you considering stars expanding against gravity in an evolutionary scenario? This does not come out in your paper but I wonder if it informed your thinking.
I was not considering mass loss. And post main sequence evolution was all in the future. I was not thinking of it then. That is a matter of chemical inhomogeneity and on the mechanism for the release of stellar energy, and that was all in the future. Bethe's papers had not appeared. If I had worked out the theory of nuclear fusion at that stage the paper would have been more highly cited. [laughter]
But was any of this in your mind?
I suppose I may have thought of some of it in parts of my work but I wouldn't have known what to do with it. It would have been an entirely separate investigation to get involved in stellar evolution and stellar interiors.
This was not something Russell was compelling his students to get involved in?
No. What the relationship was between stellar atmospheres and stellar evolution was completely clouded at the time.
In April 1938 you were still here at Princeton.
Did you know at the Washington Conference on Theoretical Physics that took place then? Russell was invited but did not go. This was when Bethe became interested and started working on the problem.
If I was aware of it, I don't recall. Scientific meetings I didn't go to [laughter] in the course of forty years makes up quite a list.
By the time you went to Harvard...
It was just coming out, Bethe's work...
I'm citing a letter from Russell to Shapley, November 25 1938: you were at Harvard then. Russell says he hoped to get to Harvard for your talk on December 13th when you were giving a talk on stellar energy generation.
Yes. It was at that time that I became interested, now that we knew the origin of the energy, we could get at the lifetimes of stars. It turned out that the bright stars were all quite young.
Has the talk been preserved.
I don't know. But I have a draft of a paper I think on interstellar problems that I prepared that started off with a discussion of the finite lifetimes of the stars and the relationship between gas and dust. Various astronomers told me it was too speculative and that I should not include that, and so I omitted it. So I only published the interstellar stuff which people were quite interested in.
I was interested in exploring the relationship between nuclear processes in the stars and the composition of the interstellar medium. Baade's work later provided a confirming basis for what I was speculating on. [note added: could you identify the person[s] who advised against publishing your speculations on the finite stellar lifetimes and the mixing of nuclear products in the IM?? — what their arguments were?]
Getting back to Russell's abilities to organize teams to work on specific problems. He was the chair of the committee on line spectra of the NRC from the 1920s through 1947....
My impression was that that was not so much organizing teams to do anything as it was getting agreement by correspondence by various people on matters. For example, the fundamental spectroscopic notation which was critical. Before they did it, the notation was a mess. I think Russell was the leading intellectual spirit in that getting a scheme that was simple, made sense. With his reputation it was easy to get others to accept it and adopt it. What else did he do?
I'm quoting Russell now in April 1936: "A notable spirit of cooperation exists between workers in different institutions. An exchange of data is frequent. Special acknowledgement should be made of the generosity of many observers who have often put extensive unpublished data at the disposal of other workers [he means himself]" but without adequate experimental equipment. He is certainly talking about himself here. All of his reports talk about cooperative work. But when it came down to identifying who would work on what he would negotiate with these people.
Did he persuade them to undertake certain analyses or to measure certain lines?
I think so.
OK. Then I can well imagine that, because that would have been done in the same way he developed relationships and cooperative work at Mount Wilson. Getting at spectroscopic data that had not been measured. "We have some very interesting things here. Is there any chance that we might persuade you to do it, to work on such and so..." He had no responsibility for this particular worker, or was not in any particular line of command, recommending him to the dean or whatever, though I can well imagine that he might well have been asked for his recommendation. Sometimes people would say no, other times they would say yes. But that is quite different from organizing a formal team. To me a team means something quite different.
Yes, none of them worked for Russell necessarily.
Yes, and he did not have to worry about their health or whether they were happy working with their colleagues, or any of the problems that someone in charge of a project can't entirely ignore.
Then you won't be surprised that in the big Princeton Fund drive in the 1920s, Russell was asked for what he wanted to build, how much he wanted astronomy to expand, and he said that he wanted time for research and faculty exchanges with other institutions. That was his style.
Yes, quite right. Deal on scientific matters with individual people, but not take on any expansion beyond that.
I see this also as a manifestation of Russell as a very efficient person. I interpret this as his vision of how to do science with the greatest efficiency and effectiveness. Is this fair?
I would have felt that the more realistic thing was that he had a complete lack of interest in administration. I am sure he was aware of this. If the department expanded he would have to worry himself about who goes where, who gets an office, and who would really do the administrative work for him. He just had no appetite for that sort of thing. He was also not very good at it and did not enjoy it. This is only my reaction of course.
But it helps to flesh out what was going on from his correspondence. He was always in a tremendous hurry to get things done — things that attracted him, that he finds fun, because he uses that term a lot.
I suppose that was responsible for the nervous breakdown that he had.
Yes. Let us now turn to other people who were around Princeton whom you may have encountered. John Merrill, PhD 1931, and came back a lot.
What was he doing when he came back.
Under Russell it would be merely conjecture, but after I came here he was constructing a set of tables for use in interpreting eclipsing variables. These were entirely tables that Russell had originally outlined. Russell thought this was a fine idea and should be supported, so we supported John Merrill. I did not think it would be a terribly important contribution to astronomy coming from this department, but it would be of some value, I imagined. It became rapidly of less value as computers became more and more available. You have the choice now to take the time to learn how to use the nomographs or you can take the time to learn how to program the functions on a desk calculator.
Since for many reasons people learned how to use desk calculators, so why should they bother with nomographs and tables? But we supported John with flexible funds from within the department. We paid for the printing of the tables as Princeton Contributions. I think John felt that I had let him down because I did not recommend him for promotions or for positions, not here at Princeton, but others that were open elsewhere. John asked me for recommendations, and I could not say no, but they were not as full of praise as he would have liked.
He never got a professional position completely to his liking. He did not advance much. His work on the tables was very painstaking and accurate. He was very fussy about getting the nomographs all lined up at the printers. I'm sure Russell respected Merrill's attention to detail and he may have felt that Merrill was a great scientist.
People who returned the most to work at Princeton after they had graduated were those who had engaged in the photometric work on eclipsing binaries, Merrill, F. B. Wood.
I knew Brad Wood.
What was life like at Princeton University Observatory. There was a group that really belonged with the telescopes, who used the telescopes, or were involved with computational methods. Wood, Pierce, Merrill, all did theses with words in the title like "Photometric researches — " with the systems identified. It looked like a series.
Yes, an ongoing program published in the Contributions. Yes.
Did these people form a clique?
Well, they came individually, of course. Merrill would come here and talk with Pierce, mainly.
What was their relationship prior to the war?
I don't know.
OK — lets move on to other people not in eclipsing binary work. In 1937 William Conyers Herring finished his PhD in math or physics. Tell me about him.
Oh yes, he was very brilliant theoretical physicists, went on to Bell Labs for many years, possibly in solid state work. He was one of my friends as a graduate student. I enjoyed talking with him and would get in touch with him from time to time. I don't think he was around the observatory.
His name comes up, but his records here show no astronomical connection. But you knew him through physics.
I knew him as a graduate student, either in physics courses or through the graduate college where there was a nice little group that often had dinner together. There were mainly physics and mathematics people, but possibly Hemmendinger might have joined us.
Hemmendinger got his degree in 1939, a year after you. A Thaw fellow from 1937 to 1938, and he did work on the photometric details of the solar chromosphere.
I wonder where he got his data? Russell may have gotten that through Donald Menzel the same way he got data from Adams for me. I was always friendly with Hemmendinger — I enjoyed his company — though I have no clear recollections of him or of scientific discussions with him.
He did not go on in astronomy at all. Any idea why?
No idea. He might not have enjoyed it enough to have made the sacrifice.
What was your contact with Louis Green? His thesis was on the laboratory spectrum of iron. [Recalls Green's remarks at the 1977 IAU conference regarding Russell's insensitivity to his interests]
Louis Green. His interests were more in theoretical problems. Advanced computations, Oscillator strengths, and so forth. I think he may have been more over in the physics department when I was around.
Yes, he worked with Allan Shenstone. Green also felt that Russell was distant and not around that much. He felt that Russell gave him to Shenstone. Of course he loved Shenstone. But he thought that he was pushed into an area that did not have much of a future. It was not a very exciting thesis.
He did the iron spectrum?
Yes, but he wanted to do stellar spectroscopy.
Oh, I see. Right.
In his 1977 talk, he recalls that he went to Russell with three thesis proposals and each time Russell said that he would do the iron spectrum.
In a way I got the same treatment except that I did not come to him with proposals. I went to him and he said this is what I would do. He said it in this way: "I have a wonderful prospect for you. A tremendous spectrum! Adams out of his great generosity had agreed to let you use them and I have made a great deal of effort getting them for you..." It wasn't my idea, I hadn't thought of it, but thinking it over I thought it was not too bad. But it was essentially the same attitude towards as it was towards Louis Green.
OK — That really helps. He treated everyone the same. But Louis Green had trouble after that getting a good position.
Yes, because his thesis was a very restricted subject then.
Russell eventually got him a job.
Yes, I thought he was a very good person.
But Green did not take what had been offered a position in Boyce's laboratory at MIT. He did not want that, to work on more spectra.
That may have been wise. How did he make the shift to Haverford, when he shifted to theory? He must have taken his courses in physics?
Yes, he took mainly strong physics. Now, he did sit in on the same conversations with Russell, Dugan and Stewart that you did about revising RDS. Do you have any recollections of that?
Not so much. I remember Russell — Russell was the central feature. None of us say very much
Lets move on to R. A. Lyttleton. He was here from 1937 to 1937 — as a special Proctor Fellowship when there was a special situation for students from Cambridge.
So we overlapped in 1937. Yes, I must have talked with him. My recollection is more when I went to Harvard when I sent to Russell my work on dissipation of planetary filaments.
In your first OHI in 1977 you did recollect some conversations with Lyttleton when you were a graduate student. Lyttleton had come up with his three-body encounter theory. Russell was excited about that.
I certainly got interested in that whole subject when I was here as a graduate student.
Right. OK that answers my question. In your paper on filaments, page 1 of your ApJ paper, you are pretty clear (in reference 5) that as Russell had stressed the mean temperature of the gases dragged off was quite high. What actually got you interested in this.
Oh I think it was the fact that Ray Lyttleton was here and he was interested in the origin of the solar system and we must have discussed it with Russell. Russell was of course very interested in it — he had a book on the subject. I forget when I carried out the calculations.
The article was received by the ApJ in April 1939. You did the work at Harvard. Your correspondence with Russell about it dates from May of 1939 where you send him the paper. Russell was very excited about this.
He wrote a Scientific American article about it. Recall's title: illeg.
He said to you: "You have done a valuable piece of work in a field where there is much need. I have long been convinced on general principles that the problem of the condensation of a filament into planets was the most improbable feature of conventional theories. And you have gone a long way toward proving it." He clearly paid attention to this paper!
I thought there was correspondence between Russell and Ray Lyttleton after this, and that there was some danger of a big controversy, and I hastily withdrew because Ray had rather fixed opinions. Or maybe I didn't withdraw.
That comes a bit later — you were at Yale by then. Your paper was published and then Lyttleton published a fascinating rebuttal, but Struve sent it to you for examination with the suggestion that you send it to Russell also. Lyttleton's argument was that material would be drawn along equipotential surfaces between the two stars and it was therefore somehow constrained. And you demolished that, Russell telling you that you were very diplomatic about it, saying "understatement unfortunately does not work in the case of Ray Lyttleton..."
Yes, quite right.
Could you expand on that?
I think I must have analyzed his mechanism, and found that it did not work. I wrote it up for the ApJ but I decided to send it to Ray first, which I did, and Ray sort of erupted. There was a question of what to publish. I may have written Ray a letter saying I would not publish. Struve had suggested that we settle the matter out of court.
Are you on good terms with Lyttleton.
I think so, but not as close touch as I have been with Fred Hoyle.
Is there any functional relationship between your thesis and your filament paper? [side issue on depositing papers in Mudd and Firestone] The technique used in the working notes to your thesis you talked there about the ability of a mass of gas to expand against gravity...
In my thesis? It was individual fountains being pushed up rather than out — the fountain model. I don't think I talked at all about the equations of fluid motion. It was individual particles — this was a different approach. If I had written my thesis after the filament paper, I might have used the fluid approach.
So the filament paper was where you started thinking about fluid behavior. The cylinder model.
Yes. The expansion of the cylinder.
So this was really outside the realm of your training under Russell.
Yes, other than the interest in pointing out that particular problem. He probably told me about his reservations earlier.
One final question in the area of solar system formation: Russell was quite excited with Lyttleton's model because it removed the difficulties with angular momentum. But one of the biggest problems with the encounter theory had been the probability of an encounter. The improbability. But Russell wrote, on more than one occasion, about the idea that the probability of encounters may have been far greater in the early universe because then everything was far closer together.
What did he have in mind? Wasn't the expansion of the universe irrelevant since galaxies do not expand? I don't know what he had in mind. Maybe he envisioned stars forming first and that they were all close together early in the expanding universe.
Russell did believe that the heavy elements pre-existed, and that stars could be older than the universe. This was a man who believed in the Mosaic story of creation, and said on many occasions that the astronomers job begins with the universe formed, and matter in it, and that astronomers did not ask questions beyond that wall. [Russell also thought of the stellar universe more than a universe of galaxies] Russell was impressed that the universe was 2.5 billion years old in 1941 which agreed with the age of the earth. He said "This is a true undesigned coincidence"—
How did he get the age of the Universe?
Hubble time. The stars' ages were greater — the stars were older than the universe, which he could conceive of apparently.
Yes but that was the same discrepancy that led Hoyle and Lyttleton to continuous creation.
Did Russell talk with you about time scales.
I suppose so, but no specifics. Probably at the time I was excited.
Lets talk about the Rosseland/Chandrasekhar/Spitzer situation. It looks as though you were offered a professorship first. Yale then matched it.
I recall that it was an associate professorship.
The document is October 2 1946: a meeting of the scientific research committee held on this date: Taylor, Russell, Smythe, Harvey, Lefshitz, Furman, Buddington:
I remember my conversations with Rosseland. I came here to talk with him to get his advice about the offer. It seemed to me that my chances of becoming director of the Princeton University Observatory would be maximized if I did not accept the offer then but went back to Yale instead. That is the one thing of which I am reasonably certain. I can easily check my files because I have all those documents.
That agrees with what you said in 1977.
[reading a document]: "I have the honor to inform you that at a meeting of the Trustees at Princeton you were appointed Associate Professor from July 1 1946 to June 30 1949." This is from Alexander Leitch. January 10 1946.
A Three-year appointment. The minutes of the Research Board are also in there.
The plan was that Rosseland would be the director of the observatory, Chandra would get the research professorship, and that you would be hired as either a full or associate professor.
That may well be.
Were you aware of this kind of plan?
Oh I was aware of some of it....[sees a letter from Russell in the folder — December 28 1945, DeVorkin quotes Russell was writing from Bloomington to Spitzer, and he says that he has been authorized to appoint Spitzer Associate Professor, and it was to be pretty quick — they wanted him by January 1946!]. I turned them down because Yale offer me the possibility to set up a research unit in physics and astronomy, and when they set it up, I did accept it — I received a wire from Provost Furness at about Christmas time in 1945 saying that they had accepted my proposal.
You had made a proposal prior to this?
Proposal to Yale, to set up the "Astrophysical Unit"
At this point, my understanding is that Russell had every expectation that Rosseland was going to stay and be his successor.
This is right, though I think they knew that there was some chance that he might return to Norway. I can show you that in one of my letters.
When you can down here to talk with Rosseland did you know that he was going to stay here?
I think in my letter to Russell which I was just looking at, I said that "if Rosseland returned to Norway" or something like that. So it was a possibility. Rosseland must have told me it was a possibility. There is nothing explicit in the correspondence, but I think too that I knew that Chandrasekhar had accepted the President's offer.
That would have been in August 1946?
No, here is my letter of January 13th to Russell about this...
In response to the formal invitation...
I stopped in at the old house and spoke with him [Rosseland] there. We had a very pleasant half-hour discussion. Talked about Fred Hoyle. I suppose we talked about Chandrasekher too.
Here is my proposed budget for Yale. I was in the habit of submitting proposals.
I see, and you say here: "If Yale accepted my proposal for a semi-autonomous astrophysical unit overlapping the departments of physics and astronomy I would probably go to Yale, and if they did not I would probably go to Princeton." And that was what Yale did, though you did not think that they would.
And on January 19th I accepted Yale's offer.
And there was no mention of Chandra at this point?
I don't think so. He hadn't accepted by this time, though then I could not have known about it.
I believe from my chronology that he accepted the offer on August 29, 1946 and Russell writes back delighted that he accepted the position.
Oh I see, this was much later, and so I could not have known.
So the spring of 1946 was the critical period.
Yes, this is already after I returned to Yale.
Now at Yale you were looking around for a good second man.
One curious thing did come up — in an exchange of letters between you and Russell, Russell suggested Rupert Wildt.
You wrote back that you were looking for someone with more general astrophysical knowledge. Yes. Russell wrote back indicating a bit of surprise. You then replied that the "man I really had in mind was Schwarzschild." — what I am trying to uncover here is Russell's own view of astrophysics and his own view for the future. This is one of the most useful exchanges of letters where you have different views where you are helping him see where astrophysics is going — can you expand on this or correct me?
Well, I think you should not perhaps read too much into this because it wasn't in the context of comparing Rupert with Martin. I think that Martin did have wider interests and wider abilities, though Rupert was very good. When Martin said "No" we did ask Rupert! And he came then, as you know, and was at Yale for the rest of his life. But I did not really disagree with Russell as to Rupert's abilities and I did work with him very well. I also invited Rupert to join us in New York.
That helps, but I am still interested in this sense of generality which is important.
I don't think Russell's espousal of Wildt can be interpreted as meaning that Russell was in any sense unaware of what directions astrophysics was going.
OK. He certainly did not object in any way to Schwarzschild.
Did he know Martin particularly?
Oh yes, yes he did. Schwarszschild visited frequently from Columbia.
Russell must have been impressed by Martin, at least I would assume he was.
I don't know anyone who was not impressed with Martin.
There was another position that you were considering in 1944:
Pittsburgh. Yes, well — Doreen and I went to have a look around and I was not terribly interested in it but it was a possibility to explore with Yale so that they could advise me on what to do in the way of a promotion.
Russell was very very worried about you going to Pittsburgh. At least he told Shapley that it would be a tragedy if you were to go to Pittsburgh.
I'd forgotten that. Did he write me about that?
Maybe not in quite as heated terms as he did to Shapley. This fits an earlier opinion that Russell had that when Wagman, Russell really got very angry circa 1941 and he went on and on about this. Do you have any idea why Russell was so upset about Wagman?
I don't know who Wagman was other than that he was the director there. What was his field?
Well, that was appropriate. I have no letters from Russell on that subject?
1944, February 24th — you corresponded about the Pittsburgh position. It was a wonderful letter to you...recites letter. Warns against the pitfalls "You would have to repeat Struve's miraculous performance at Yerkes," and then some because the conditions at Pittsburgh were even worse than what they were at Yerkes.
I'd forgotten that.
What was he talking about?
Not really, except I could figure it out in the same way that you figured it out: I'd forgotten that Struve had such an uphill fight at first — I really wasn't interested that much in Pittsburgh.
When you came to Princeton were there any problems or fights with the "old regime" so to speak.
The relations I had with Newton Pierce and John Stewart were a source of some concern to me. They probably had the same attitude towards Martin I would think. At our faculty meetings the four of us sat at the table and were rather frigid, to put it mildly. I think that Newton felt that he should have been made director, which does not inspire respect for his appreciation of reality. Maybe thats not right. Maybe it was Stewart who felt this way. Stewart in any case felt that he had been betrayed by Russell, and in a sense I guess perhaps he had. His retreat into social physics was sort of a result of that and he really had great ability but the circumstances were not such to encourage his growth. I think with a different director he would have had an entirely different career.
Can you be a bit more explicit about that. Does this go back to the 1920s?
I think so. He wrote a paper on ionization in stellar atmosphere I think it was, and there was some disagreement. Russell disowned the work and didn't support him, saying it was probably not right. And then later on somebody else came out with the same ideas and Russell was enthusiastic about it, and apparently never made it up with John...
This may have been about Unsold's technique for line profiles, dealing with abundances?
Unsold, yes, it was.
You would point to this as the source?
Yes, this seemed to me to be the chief source of contention, but I suspect that there may have been other problems too that sort of drove him out of astronomy.
Russell was a long-time advocate of Robert Atkinson and when Atkinson's paper on proton fusion was submitted to for the New York Academy prize around 1930 Stewart was one of the judges, and he turned it down. Also Russell's faculty evaluations of Stewart were not as positive and not as thoughtful as yours were of Stewart. You took the trouble to seek out opinions of others, a demographer at the IAS, there were others around who were doing demography and doing statistical studies of population research..
Yes, yes. But Russell probably had complete confidence in his own judgment. I was not so sure that in another fifty years that Stewart's work might not become the most important to come out of Princeton University Observatory! [laughter] I didn't really think so, but it was so different and its hard to know when something really new comes out what it is going to lead to. So I did want to get the views of some people who were in a position to comment.
I for one was impressed that you did that. It could also be a generational view of astronomy itself. Being far more interdisciplinary after the war.
discusses his GHA IVB chapter on stellar evolution and the importance of interdisciplinary work re stellar evolution post war...
Salpeter and Hoyle were here. Yes, I think though that Russell was less interested in getting other people to collaborate with him on theoretical ideas. He was more interested in getting other people to provide information that he needed for his own theories. There were not many people whose ideas could really be a positive contribution to Russell's work in his day.
He did try once or twice, with K.T. Compton, but Compton later tried to get his name off the paper.
Let me ask more about Pierce. Now, Russell never expressed annoyance with Stewart in the letters I've read, but he did about Pierce, he wrote to you.
I've forgotten this.
June 23 1947:
Just before I came back here as chairman.
Yes. You were at Mount Wilson for that summer. Russell said that there are conflicts with Pierce regarding office moves, moving people around. Pierce had blocked off a door with a desk making it impossible for Russell to get to the library when a class was going on? Now Russell tips you off about this growing personality problem.
This is entirely consistent with my view of the situation at that time, but I don't remember this particular instance.
Pierce died tragically, he was young, of a cerebral hemmorhage. I found records that Pierce spent time in the hospital before this, that he was becoming more and more irrational. He had terribly high blood pressure. Was this known?
Could be — I don't recall specifics. Have you met his widow?
Martin Schwarzschild suggested that I do not play up these negative aspects about Pierce because his wife is alive. I don't plan to, but I do feel that I should cover how Pierce made himself invaluable to Russell and why he was promoted over Wildt time and time again. Wildt never got a real position here whereas Pierce got promotions and raises during very lean times here.
Did you ever attend any of Russell's Sunday seminars?
I don't really recall. Doreen may have. She was in a wider circle than I was. I was not interested in such things. He certainly never tried to prosyletize. The only thing that I knew was that he had stored in his brain all of the verses of all the Christmas carols I ever knew, but probably of most of the Bible.
In your 1977 interview, you mentioned that your contact with Schwarzschild convinced you of the importance of numerical solutions
This was after WWII when you had extensive contact. I am curious if you knew Russell's view on numerical solutions and of approximate methods in astronomy? When you were a graduate student. December 2 1937: you were introduced at an AAS meeting by Russell where you gave your talk on expanding atmospheres. Russell as the retiring president of the AAS gave a presidential talk on the place of approximate methods in astronomy.
Oh yes! I don't remember that specific talk but I've always admired his ability to devise simple approximate methods for getting essential information. I tried to emulate that goal in my own work. I have great sympathy for that sort of thing. But I think I picked it up from him more in informal discussions than from his papers.
It is certainly the style of his eclipsing binary work, but in his stellar atmosphere work he is trying to force fitting observations together. You can see it not only in his reduction procedures, but as well in his theory where he was happy to use physical theory that others knew was wrong, but it worked for him.
His use of the Rumpf model, to deal with the valence problem...
So it had some resemblance to reality, and it was helpful. Yes. I can see that. I learned about his trait through informal contact when he talked about his work and the work of others. His comments about what they might have done, and what he hoped to do.
Did he ever express annoyance or frustration with people who insisted upon exact solutions even though they might be important.
Quite likely. That would be entirely consistent with what I remember of Russell. There again, I feel that I have been pretty much influenced by Russell and find myself trying to do very much the same things I saw Russell trying to do. Trying to understand physically what is going on, searching at first for approximate solutions. I feel I first encountered this during my Prep School years, where I met with simplified analyses to get the information that was necessary.
Notes how pleased Russell was when he knew that Spitzer was coming.
Well I was in a fortunate position because I did not compete with him at all. Of course I was much younger and I was his student, his disciple, so he was in a good position to praise my work which was designed to follow his own goals [laughter]