Abstract
Owen Gingerich’s previously unpublished autobiographical sketch of his astronomical education, from his teenage years up through graduate school at Harvard, is presented with an introduction and notes by the editors.
Keywords
Introduction
Owen Gingerich (1930–2023) left an unpublished autobiographical essay, “Ad Astra per Aspera,” which he wrote by 2005. David DeVorkin had the opportunity to read it that year while preparing for an oral history interview with Owen. 1 The essay captures Owen’s view of his life course through 1962, when he earned his Harvard Ph.D., and is a poignant personal statement, since it carried the subtitle “From the Sewers of Kansas to Harvard College Observatory.” Equally valuable for readers of the Journal for the History of Astronomy, it contains his insights on the transformation of the Astrophysical Observatory of the Smithsonian Institution by its move to Harvard in 1955, a significant shift spurred by the changing character of the profession. 2 The essay also provides a personal appreciation of those at Harvard who helped to shape Owen’s career, including Harlow Shapley, Cecilia Payne-Gaposchkin, Donald Menzel, and his thesis advisor, Chuck Whitney. And it provides a fascinating insider’s view of graduate student life in 1950s astronomy.
One might well suspect that Owen’s deep association with Harvard and his admiration for Harlow Shapley led him to title his reminiscences in a way that echoed Shapley’s 1969 memoir, Through Rugged Ways to the Stars. 3 Owen had earlier alluded to this title in his historical article about Shapley’s work, “Through Rugged ways to the Galaxies,” published in 1990. 4 However, as Owen points out, Ad astra per aspera is the state motto of Kansas, and it appears on the state’s flag and seal. And of course the phrase has roots in classical Antiquity. Moreover, Owen signed off with “Ad astra!” in one of his very first publications, an article about telescope-building that he wrote when he was a junior at Goshen College. 5
In 2005 Owen still regarded his essay as a work in progress. Two paper copies survive, with annotations by two of his colleagues, the astrophysicists Irwin Shapiro and Fred Franklin, which Owen intended to use in a subsequent revision. Sadly, that revision apparently was never completed. However, we have been able to make use of use of these annotations and have also consulted with Dr. Shapiro to clarify several details. Owen’s widow Miriam Gingerich and their son Jonathan have kindly granted permission for publication of the essay in JHA.
Corrections to Owen’s text have been limited to a few obvious slips (identified in the notes) and typographical errors (corrected without mention). The editors have also added some explanatory notes and references. The notes are all by the editors—there are none in Owen’s typescript.
AD ASTRA PER ASPERA From the Sewers of Kansas to Harvard College Observatory O. G.
6
The state motto of Kansas, as I learned shortly after moving there in 1941, is “To the stars despite difficulties.” It occurs to me that in a reminiscent account of HCO in the 1950s, I should explain how flushing the sewers in North Newton, Kansas, led to my first summer at Harvard Observatory in 1949.
I was already an embryonic amateur astronomer when my parents moved from Iowa to the Bethel College campus in central Kansas. My father had helped me construct a telescope from a mailing tube, an eyeglass lens, and a magnifier; it was good enough to see the rings of Saturn easily. I was ready for something bigger, and I was very happy to find an amateur telescope maker not far away, who eventually put his 6-inch reflector at my disposal. Besides having the telescope, he also had a full set of the red-bound Harvard Books in Astronomy, 7 and a subscription to Sky and Telescope magazine, which was edited at Harvard Observatory. In addition, the college library subscribed to Popular Astronomy, which carried monthly reports from the American Association of Variable Star Observers (AAVSO), headquartered at Harvard Observatory, so in my amateur world Harvard loomed very large indeed, and I dreamed of the day when I could join the AAVSO and make my mark observing variable stars.
World War II produced a major manpower shortage in the US, so as a young teenager. I held a variety of after-school jobs. In 1945 I got the job of filling three underground cisterns once a week, and then releasing the water in a mighty surge to clear out the sewer pipes in our small village. This job paid $25 a year. Somehow I learned that a life membership in the AAVSO could be had for $50. Emboldened, I wrote to the AAVSO asking if I could pay the life membership in two annual installments. Leon Campbell, then the “Recorder” for the AAVSO, and his assistant Margaret Mayall were apparently so impressed by this unusual request that they said yes despite the fact that I was not yet the statutory minimum age of 16. That is how cleaning the sewers in North Newton set me on the path to Harvard.
Not long after, my father got a position at Goshen College in Indiana, so I entered college there, and on the side, I ground and polished my own 8″ mirror so that I could continue making variable star observations. By May of 1948, toward the end of my freshman year, my telescope was in operation and I began to make variable star observations once more, plus regular searches for the clusters and nebulae in Charles Messier’s famous 18th-century list. Then, in June, I stumbled onto a comet in Cygnus. Naturally I was very excited, and sent a telegram with its position to Harvard Observatory, the clearing house for Western Hemisphere comet discoveries. A few days later I hitch-hiked to Milwaukee to attend the national convention of amateur astronomers, the second meeting of the newly-founded Astronomical League. Some of the amateurs had brought along their telescopes, and in the evening were observing the comet in Cygnus, so I quickly realized that my “discovery” was already well known.
Another attraction of the program was a field trip to Yerkes Observatory to see the giant 40″ refractor. The man demonstrating the telescope was an amateur astronomer hired during the summer to take care of the visitors. Imagine that! An amateur astronomer could get employment in a professional observatory! This opened an entirely new vista to my ambition. Meanwhile, a post card came from the AAVSO thanking me for my observation of Comet Honda. I decided I had better subscribe to the Harvard Announcement Cards so as not to be caught out again.
By the spring of 1949 I decided to apply for a summer job at Harvard College. Observatory. I sent my letter of application to the AAVSO, and presently Leon Campbell replied that the chances were low, but that he had turned over my letter to the director, Harlow Shapley, then probably the most famous astronomer in America. By and by I got a terse letter from Shapley himself, saying, “Send references.” I asked the chemistry professor at Bethel College (where as a high school student I had made rather a pest of myself) to send a letter of recommendation, and he told me, “If there is a job to be had, my letter will get it for you.” There was indeed a job to be bad—never mind that it offered starvation wages—and that June found me, suitcase in hand, thumbing my way to New England.
My last ride left me off at the Harvard Business School, on the other side of the river. I had grown up in towns laid out in perfect Cartesian grids, and it was a big surprise to discover that I could not hike in a more or less straight line from the river to the Observatory, but eventually I found my way to 60 Garden Street and the winding gravel road that led up Observatory Hill, past the big frame residence and to the brick buildings of the observatory proper. I sought out Shapley’s offices on the third floor of Building D, and there found his secretary, Arville Walker, who had worked for Pickering before him. She called Virginia Nail, Shapley’s personal research assistant, and the two women directed me to a lodging on Upland Road, where I rented a room for the summer.
HCO in 1949
Eventually I became quite familiar with the layout of the Observatory, and I’ll describe this beginning with the AAVSO headquarters at the west end of the oldest structure, Building A. The Observatory originally comprised the Sears Tower with the Great Refractor flanked by two frame wings to the east and west. The west wing terminated with a two-story structure topped by a dome, then containing the 6″ Post refractor. There was an entrance at the west end, and just inside to the right was the long room where Leon Campbell and Margaret Mayall ran the AAVSO and plotted the light curves of the variable stars as the hundreds of amateur observations poured in each month.
To the left was a smaller office, I think occupied by James Baker, who was working to a large extent independently on optical design. In the cellar below, which had earlier, in the 19th century, included the clock vault, the Amateur Telescope Makers had somewhat cramped quarters for their optical shop. (Eventually I took the opportunity to make two 4½″ flats there.)
The second floor was a key place for the graduate students, the Tutorial Library, which contained non-circulating sets of all the major journals plus the basic monographs and texts. This was actually two rooms plus a corridor. There was another fairly large room, directly above the AAVSO office, and I don’t know who used it at the time, but when I came as a graduate student 2 years later, Jimmy Baker had installed an early computer there for ray-tracing work. Ed Upton was an undergraduate who worked on the computer for Baker, and as we were in the same math class, we often sat there working on our assignments. We were also intrigued by the polar telescope, which operated out of a south-facing window of this room. It included the polarizing photometer now on display in the Great Refractor alcove.
Between these offices and the Sears Tower was a long, single-story area that had originally contained the meridian transits. By 1948 it was an exhibition room for the public. At the ends of the room were display cases for the collection of sundials (now at the Historic Scientific Instruments Collection), and the walls were lined with illuminated astronomical transparencies.
The rotunda of the Sears Tower, and especially the four alcoves, were full of books, the overflow from the library. The telescope pier itself served to display the long light curve of one of the AAVSO’s favorite variable stars, the dwarf nova SS Cygni. Not a single outburst had been missed since the founding of the AAVSO early in the century, though there had been a near miss: one stretch during World War II had virtually no observations, and it was not until after the war that a lone observation had been reported of the missing maximum.
There were two offices in the stairwell on the south side of the Sears Tower, and these housed the Gaposchkins. 8 They were “characters,” and it seemed to some extent that they had been exiled to this rather remote part of the observatory. Cecilia was tall, gangly, quite out of it with respect to fashionable dress, and usually had a cigarette dangling from her lips. Sergei was smaller, wiry, proud of his athletic physique, and seemed quite loud and temperamental.
To the east of the Sears Tower was the more-or-less symmetrical wing that attached to the Director’s Residence. The connecting section included space for a couple of secretaries with the special typewriters used for producing some of the Annals. 9 This room led to the Residence, and since the Shapleys regularly threw parties, all the staff had a good idea about the main floor, one flight below the connecting section. I viewed these parties with some ambivalence, because they always included dancing to the tune of the record player, and I was not a dancer. I did join in for the Virginia reel that inevitably climaxed the evening, and, as I recall, Fred Whipple was often the “caller” for the reel. (Eventually, after the Shapleys had moved out and before the residence was torn down to make way for Building B, I was asked to photograph the rooms. These pictures were made into a scrapbook for a large birthday party for Shapley, and this scrapbook is now part of the Shapley papers in the University Archives.)
On the north side of the Sears Tower was a small room that had once served as the prime vertical room (for a transit aligned on an east-west line). I found it a fascinating place because it was crammed with stacks holding some of the Observatory’s oldest books: almanacs and serials going back to the 18th century. Between the original observatory structure (Building A, the Sears Tower, and the Residence) the ground sloped downward to the north to where Pickering had built the fireproof Building C for the growing photographic plate collection. A covered wooden bridge connected the prime vertical room with Building C. The focal point of Building C was the classroom and periodical reading room. The cozy classroom was just to the northeast of the stairwell. It had blackboards on two sides, and two rows of movable one-armed desk chairs, room for at most about 20 students. Beyond it, to the east, another door led into a large room that served as an extension to the library. It was divided into two parts by a long row of files that held the reprint collection. To the north was a large, sturdy, but aging oval table that held the latest journals, and there was an assortment of chairs including a bizarre one that must have been recycled from a 1930s passenger plane. On the other side of the reprint files were three stalls for graduate student desks. I associate the names of Hal Zirin and Bert Donn with these places; possibly Dick Dunn sat there, too. A few other graduate students were squirreled away among cupboards and files on the north side of the floor below. Frank Orrall, Arne Wyller, and Uco van Wijk are the names that come to mind.
The upper floor of Building C was likewise divided into a southern and northern section. The southern section provided an office and large workroom for Bart and Priscilla Bok, and an assistant (I believe it was Margaret Risley 10 ). The northern section contained two larger rooms and two tiny offices at the far end. It was still filled with a huge card file of variable star data long maintained by Annie Jump Cannon, and by a clutter of file cases and desks. This is where the graduate students Harlan Smith and Frank Kameny sat, and where I would eventually sit as a graduate student (a space I shared for some months with Hermann Bondi).
North and west of the stairwell on the main floor of Building C was the cramped editorial office of Sky and Telescope magazine, where the editors Charlie Federer and his wife Helen sat. (More of this later.) The room to the south served both as a corridor to the newer Building D and as an office for Catherine Hanley, the librarian. Like other hallways on this floor, it included a lot of overflow books from the library.
Building D, the largest edifice on Observatory Hill, was built in 1930 at the end of Shapley’s first decade as director. By then the photographic plate collection must have been bursting at the seams; in the new quarters it occupied three of the five floors on the north end of the building. The Phillips Library shared with an auditorium a large space on the southern end of the building. The serials and other observatories’ publications filled stacks in the balconies, while the monographs lined the walls of the main floor. The auditorium gave a space for colloquia and special lectures, and also for the evening meetings of the Bond Club and the Amateur Telescope Makers, two independent organizations much encouraged by Shapley. On very special occasions the chairs would be rearranged and tables set up in the form of a hollow square. These so-called “Hollow Squares” gave Shapley a chance to brief the staff on a variety of current news, to introduce visitors, and generally to build a feeling of togetherness.
Below the library was another floor of offices, including one for Fred Whipple, who would soon become Astronomy Department Chairman. The corridor included a darkroom, and on the west side a lounge where everyone occasionally gathered for coffee, gossip, and smoking. (The whole observatory more or less had a residual aroma of stale cigarette smoke.)
On the other end of the building, below the plate stacks, three women managed the circulation and distribution of Sky and Telescope. There was a storage area for back issues and other publications, and an adequate space for wrapping packages.
Finally, the upper floor of Building D held the offices for Donald Menzel and Director Harlow Shapley and their immediate staffs. I probably never saw the inside of Menzel’s office; he had a reputation of being aloof or possibly even shy, and only interested in the best students. Shapley’s office was very interesting, especially because it featured the large rotating desk originally procured by Pickering as part of the “Anonymous Gift of 1902.” 11 A central core of shelves rotated independently of the desk itself, and the desk was large enough to hold a whole series of independent projects. In one corner of his office Shapley kept the “Treasure Collection,” a splendid group of about 50 rare astronomy books that I became aware of only after they were transferred to Houghton Library. Across the hall (to the east) were several offices filled with plates of the Magellanic Clouds, and I was given a desk in one of them.
Before describing the sort of tasks assigned to me, I should answer the obvious question, where was Building B? In the yard to the west of the main buildings were several enclosures for various telescopes, some of which had already been given away to Poland and China or (such as the patrol cameras) moved to Oak Ridge Observatory. Beyond these, at the edge of the property, was a long, low building shared by the Observatory’s business office (overseen by Sibyl Chubb) and a very cluttered and filthy machine shop. This was Building B, presumably the second oldest structure on the site. I had to visit there because Miss Chubb was the one who could change the address on my subscription to the Harvard Announcement Cards.
Today it is hard to imagine the small familial atmosphere of the Observatory 50 years ago. Though I was at the bottom of the totem pole, I knew practically everybody who worked there. The Observatory community was so small, in fact, that only a handful of telephone lines served the entire establishment. A system of bells in the hallways linked the entire compound, and each person had his or her own code of longs and shorts, the signal that a call had come in on one of the Observatory’s phones! So the Observatory was seldom a quiet place for very long.
A summer assistant for Harlow Shapley
One of the first skills I learned was to master the rather tricky card catalog for the photographic plates in the plate stacks. Eventually I became all too knowledgeable about that catalog, because when there was nothing more urgent to do, I filed literally thousands of cards into it. (In mid-August I wrote to an acquaintance that I had already racked up the card-filing record, 15,000 cards.) However, it was much more exciting to use the catalog to locate plates of a given region of the sky, and one of the more memorable episodes was to search for all possible images of what we called “Luyten’s flare star,” and which is now famous as UV Ceti. Willem Luyten at the University of Minnesota borrowed Harvard plates to blink for his survey of high proper motion stars, and in the process found a comparatively nearby star that had temporarily flared to a substantially greater brightness. Dr. Shapley—as he was invariably called by the staff, never “Professor” Shapley—assigned Mrs. Nail to look into it, and since she was my immediate supervisor, I learned how to fetch the plates so that she could measure the magnitude of the star’s flares. Altogether I brought up 569 plates for examination. Several flares were found, including one in 1929 where the normally 13.8-magnitude star had brightened 16 times, 12 to 10.8. One plate was a particular bonus: by chance the telescope had moved during the exposure, so there were double images, and in one image the star was flaring and on the other it had its normal brightness. The results of the survey appear on Harvard Announcement Card 1010. 13
Mrs. Nail’s specialty was measuring variable stars, especially in the Magellanic Clouds. She used a “flyspanker,” a small section of a photographic plate that had a raster of stepped exposures of a star, and a wire frame and handle so that it indeed looked like a miniature flyswatter. With this device she could go from a standard magnitude sequence in one part of a plate over to the region with the variable stars to be measured—generally about ten of them. I would sit with the log book recording the magnitudes as she called out her estimates. Sometimes I would come up one short at the end of the row. A very stern taskmaster, she always accused me of not paying attention, and I was sure that she had simply skipped one.
Mrs. Nail showed me how they discovered variable stars by making a positive of one plate of a field, and then superimposing it onto a different negative. A careful inspection of the superimposed negative and positive stars could show any that varied. I think Dr. Shapley really wanted me to try my hand at this, but in the end the time never worked out for me to begin it.
That summer Dr. Shapley, who had never taught undergraduates before, was trying out a course on “Cosmography” 14 for the summer school, and the photographer Bob Cox taught me how to make glass lantern slides so that in case of emergency I could make the slides Dr. Shapley needed for his class. The darkroom was in the basement of the old Astronomical Laboratory, a frame building that stood near where the Harkness Common (Law School Graduate Center) now stands. It was very educational for me to work with a professional photographer, although I already knew many of the techniques since my father was an ardent amateur photographer. I spent about a third of my time there, and when Bob Cox was on vacation, I developed the glass plates that came in from Oak Ridge.
I often worked at the Observatory at night, and my office was directly across the hall from Dr. Shapley, who also sometimes worked at night. One evening, after I had made slides for him, he came over and told me about his experience making his own slides for the first AAS meeting he had attended, in Pittsburgh, while he was a graduate student at Princeton. He said that when he got to the meeting and saw how professional the other slides were, he went out to a professional photographer and spent all his lunch money having his slides remade. Shapley was enough of a showman that the story rang true.
One day Dr. Shapley called me in to see a letter from the famous Duke University parapsychologist, J.B. Rhine, who wanted to know how Jonathan Swift had predicted two satellites of Mars with pretty nearly correct periods. 15 Could I figure that out for him? I remembered that there had been a short account in an old Sky and Telescope about it, but otherwise I was baffled. The problem bugged me for 20 years, and, much too late for J.B. Rhine, I came up with a solution of sorts (JHA 1 (1970) 109–15). 16
I was interested to learn more about the telescopes at Oak Ridge, so I signed up as a volunteer on four weekends. This gave me a good chance to meet the graduate students, some of whom were spending the entire summer at the station. Bart Bok was in charge, and assigned me to a series of telescopes so that I would find out about the various instruments. During that time, at the beginning of July, there was great excitement with the discovery of a comet on a Jewett Schmidt plate taken for Bok by a graduate student from India, Vainu Bappu. When they were looking at the plate an undergraduate, Gordon Newkirk, wandered by and asked, “What’s that?” pointing to a fuzzy spot on the plate. Bok decided that the comet should bear all three names, Comet Bappu-Bok-Newkirk. Fred Whipple had discovered many comets bearing his name by a systematic survey of the patrol plates, but this is the only comet that Bok’s name is on.
At Oak Ridge I was under the direct tutelage of Henry Sawyer, a hard bitten, laconic New Englander who was the regular year-round observer. I first learned about the patrol cameras, three of them, which took unguided hour-long exposures of fields specified in a standard log form. Whoever was assigned to the 16″ MC telescope (named after its maker, Metcalf) also loaded up the patrol plates. The 16″ was a somewhat bizarre telescope in that its field was curved, which meant that a plateholder could not be used because the glass plate had to be bent into place and held under a vacuum system. The plate had to be loaded by standing into a large light-proof bag and transferring the plate from a Kodak box onto the vacuum grid. Also, as was the case with most of the telescopes, the MC was driven by a DC motor and a so-called Gerrish drive, which consisted of a pendulum that made and broke the circuit twice each second. So the silence of the summer night was broken by the click, whir, click, whir of all the telescope drives.
My big disaster came when I was assigned to the 8″ IR refractor, 17 which took Selected Areas plates for the Dutch astronomer van Rhijn. The plates were accurately timed short multiple exposures with an offset between exposures. Henry Sawyer gave me a briefing about the telescope, whose focus was very temperature dependent. Therefore, for each exposure one had to read the thermometer and adjust the focus accordingly. What he failed to make clear was that there were two scales for the focus, a coarse one and a fine one. I overlooked the coarse adjustment, and made the entire setting on the fine adjustment. As a result, the large batch of plates I took were all out of focus, an expensive waste of glass plates, not to mention the telescope time.
One night I was assigned to assist with the 24″ H telescope. 18 Again, it was an idiosyncratic telescope, in which the observers sat on the roof of the building totally exposed to the chilly night air. H presumably stood for Harvard, but that is not what the graduate students called it. This telescope was fitted with a photoelectric photometer and a large strip recorder. It was fascinating to see how it worked, but too complicated for me to solo on it. At least I could explain to the grad students how the standard cell in the chart recorder worked, something I knew as a chemistry major back home, and we all shivered there together.
The pinnacle of my instruction at Oak Ridge (which fell short of using the 61″ reflector) was being assigned for two nights in August to assist Bart Bok at the Jewett Schmidt. The Jewett was a homemade and rather clumsy Schmidt telescope, but a useful instrument for Bok’s Milky Way studies. He showed me how to insert a plate into the focal plane within the tube itself (characteristic of all Schmidt cameras), and how to guide during the long exposure. “Watch out!” he cautioned. “It’s dark in here, you’ll get sleepy and hit the fast motion button by mistake.” And this is precisely what happened to me. I quickly closed the plate shutter and repositioned the telescope. Bok really scolded me, but the developed plate showed no sign of my slight mishap.
The summers of 1950 and 1951 with Sky and Telescope
At the national meeting of the Astronomical League in Milwaukee, I had met Charlie and Helen Federer, the editors of Sky and Telescope, who had played a catalytic role in the events there. I was keen to keep in touch with this amateur movement, so in the summer of 1949 while I was working for Dr. Shapley, I asked the Federers about the possibility of going with them to Cleveland for the third national Astronomical League meeting. They agreed to take me, and I gave two papers there, one on observing the Messier objects (the embryonic outline for a later article in Sky and Telescope 19 ) and a command performance from Charlie for a paper on astronomical time. He was looking for such a topic for his magazine, but my attempt didn’t make it. However, I did break into print in the spring of 1950, after I had hitchhiked to Oshkosh, Wisconsin, to attend a regional meeting of the Astronomical League. There I gave a short paper on the photographic discovery and analysis of variable stars, based on my experience the previous summer. My brief journalistic report on the meeting appeared in the June issue. (A particularly cordial side meeting in Madison with C.M. Huffer and A.E. Whitford was very significant in setting my course toward a career in astronomy.)
As a consequence of my contacts with the Sky and Telescope editors, I was invited to return to HCO in June 1950, essentially as a summer intern with the magazine. Charlie and Helen had met at the Hayden Planetarium in New York, where they had come together in criticizing the old Sky magazine published there. Meanwhile, under the umbrella of the Bond Astronomical Club in Cambridge, Shapley had started a popular magazine, The Telescope, and had organized Donald Menzel to edit it. In the fall of 1941 the two publications merged, and the Federers moved to HCO to edit the joint publication. In addition to managing the magazine, they handled the publicity for the American Astronomical Society meetings, edited the magazine Weatherwise, and prepared various popular pamphlets for the Observatory.
Charlie was a streetwise New Yorker who had all sorts of advice about how to keep my bicycle from being stolen, and also something of a health nut—one of the few HCO workers who didn’t smoke, and someone who didn’t eat French fries because they contained too much fat. A child of the Depression, he taught his staff how to save string, and he gave me a detailed lesson in how to wrap packages—a rather good procedure that I still use. The girls in the circulation department felt this was rather much, but held their peace when he was around. Charlie assured me that he expected me to work 50 hours a week, which I did.
I sat in the rather cramped office in C34 along with Charlie and Helen. One task that fell to me was to ink the diagrams for illustrations in the magazine. I got pretty good at using a lettering guide, and I was proud of introducing a new technique, the use of a halftone overlay called Zip-a-tone, which I had learned while editing the college yearbook the previous year. (Eventually I prepared all the diagrams for Mrs. Gaposchkin’s Variable Stars and Galactic Structure. 20 )
Each month the magazine contained an observer’s guide, and I did a lot of the fact checking for this section. Also, there was a section called “American Astronomers’ Report,” which summarized the more interesting papers from the AAS meetings; authors sent copies of their abstracts to the magazine. On the basis of the abstracts I wrote up some of the reports, which was quite a stretch considering my lack of any college courses in astronomy. Each month I helped with proofreading, wherein the entire text of the magazine was read aloud against the original manuscripts.
Among the Observatory booklets the Federers oversaw was one on early astronomical photography. 21 1950 was the centennial of the first astronomical photograph, a daguerreotype of the star Vega made with the 15″ refractor, and this occasioned a photographic exhibit at the Fogg Art Museum, with a small room of early photographs from the Observatory and a larger exhibition lent by the Eastman House in Rochester, N.Y. There was also an unsuccessful attempt to repeat the original daguerreotype with the Great Refractor. Dorrit Hoffleit was the sparkplug behind this effort, and I got to assist in setting up the Fogg exhibit and in the redaguerreotyping.
As before, I attended the Hollow Squares and the occasional Shapley parties, and I spent a few weekends at Oak Ridge Observatory. Bok was not there that summer, for he was on a 19-month leave at the Boyden Station in South Africa. The station was basically in the hands of the graduate students, and one major project was an ambitious attempt to aluminize 22 the 61″ mirror, organized by graduate students Harlan Smith and Frank Kameny. Basically they proved it could be done, laying the groundwork for a more professional handling of the process in future years when a special room was built for it, attached to the 61″ dome.
That summer I poked around more in the library, for Shapley shunted the general letters of inquiry down to Sky and Telescope, and sometimes I was assigned to answer them. I was particularly pleased to find the original Messier catalog of clusters and nebulae in the 18th-century Connaissance des Temps almanacs shelved in the old prime vertical room. These led to my preparing a pair of articles for the magazine on “Messier and His Catalogue,” my first foray into historical researches, finally published in 1953. 23
One extra perk of working for Sky and Telescope was that Charlie Federer had close connections with the Boston Museum of Science, I suppose because of his earlier connection with the Hayden Planetarium in New York City. The Museum of Science was just developing its new site on the dam at the end of the Charles River, and had a museum-in-a-shed along the water. Included was a Spitz Planetarium with seats for about 50 people in the dome. The Planetarium needed weekend lecturers, and I became one of the regulars. This paid very well in student terms: $7.50 for the first lecture of the day, and $6 for each thereafter. On a Sunday afternoon it was possible to give four shows, and $25.50 was good pay. With that part-time job I financed an appreciable part of my first year in graduate school.
My work for Sky and Telescope in 1950 secured an invitation to return the following summer. By that time the die was cast: I had applied to Harvard for admission to the graduate school program in astronomy, writing that I hoped to become a science journalist, and my contacts during the summers had paved the way for my acceptance despite my having come from a small and virtually unknown college. Thus I came in 1951 as a Goshen College graduate and as a future graduate student, though the summer work was much the same.
Two things were different. I roomed with the Federers in Belmont, not necessarily a good thing, for it meant that I was continually on call and lost a good deal of independence. Toward the end of the summer I hitchhiked to Norfolk, Virginia, in a foggy attempt to observe an annular solar eclipse, and then to Chapel Hill for the meeting of the Astronomical League. When I returned, I discovered that Helen Federer had been diagnosed with breast cancer, and it fell on me to do much of the work in getting the magazine to press. I authored the lead article, on the eclipse observations, 24 and helped put together the anonymous report on the Chapel Hill meeting. I really earned my keep that month!
Astronomy graduate school
Six of us graduate students entered the Astronomy Department in the fall of 1951, a particularly large class: David Heeschen, Ed Lilley, Tom Matthews, “Mac” McCrosky, Chuck Whitney, and I. The first three already had masters degrees, so had a head start, and McCrosky had worked in the field on Whipple’s Super Schmidt meteor program. 25 I was low man on the totem in terms of math background, but had the experience of being around the Observatory for the three previous summers.
I was assigned a desk in Annie Cannon’s old room, upstairs in Building C. I shared the area with Frank Kameny, at a desk facing his. He was full of radical opinions, concealing his Jewishness, and as I found out much later when he became a prominent member of the Washington, D.C. gay community, his sexual orientation. 26 Later, when Frank finished his thesis, the space was shared with a visitor, Hermann Bondi.
In 1952, when Shapley relinquished his long directorship, he took over the rooms beyond the one where I sat, so I often saw him coming or going to his new and very tiny office. Mrs. Nail left to become a statistician with the Hood Milk Company; Shapley felt it ironical that “the world’s greatest expert on variable stars in the Magellanic Clouds” could no longer be afforded by the new Menzel regime. He hired two new (and presumably cheaper) research assistants, Ann Hearn (who later became Mrs. Thomas Matthews) and a former astronomy undergraduate, Jacqueline Sweeney (who became Mrs. Henry Kloss). Both were good friends of mine, especially because Tom Matthews was my best friend in graduate school.
Since I continued working for Sky and Telescope, I signed up for only two courses, a math course down in the Yard and Mrs. Gaposchkin’s course in astronomical spectroscopy. By this time the old Astronomical Laboratory, where I had made slides for Dr. Shapley, was torn down, and the graduate/undergraduate courses met in Byerly Hall in the Radcliffe Yard. Mrs. Gaposchkin chain-smoked as she taught, literally lighting the next cigarette from the first and spreading the ashes all over the desk, but her enthusiasm for the mysteries of actual spectra was infectious. Her teaching fellow, Harlan Smith, made numerous photographic prints in the Byerly Hall darkroom, and in that way we each got a great variety of spectra to examine. Though my undergraduate background in math was quite inadequate for what I was encountering in the Harvard math department, I was well prepared in spectroscopy, having pursued that independently at Goshen. I found her course a great delight, and I found particularly memorable my analysis of some high-dispersion Zeeman spectra from the magnet lab at MIT.
At the beginning of the second semester in February came the PhD exams. There were three of these, on consecutive Saturday mornings, on astrophysics, galactic and extragalactic structure, and on solar system and practical astronomy. The first exam was set by Menzel and Mrs. Gaposchkin, the second by Bok and Shapley, and the third by Whipple. An “MA pass” was 60 points and a “PhD pass” was 75. Once a PhD pass was obtained on a given exam, that particular exam was no longer required, but until then every graduate student who hadn’t yet passed, regardless of level of preparation, was required to take the exam. That first year Dave Heeschen and Ed Lilley got 74s on Whipple’s exam; they had done well enough, but didn’t know the constant of gravitation and Fred Whipple believed no one should have a PhD pass who didn’t know that number. I think I got an MA pass on the second and third exams. The second year I got a PhD pass on the second (having taken both of Bok’s main courses), as well as the first (astrophysics), but I flunked the third. The ruling was made that once passed was enough, so I could get my MA. I had passed the first exam largely because Mrs. Gaposchkin had included a question on the significance of a large list of specific variable stars. Dick Thomas, who had joined the faculty in the 1952–53 but who had no strong hand in the PhD exam that year, took great umbrage at my astrophysics pass, saying that variable star names weren’t astrophysics, and insisted that I retake the astrophysics exam privately the following year. I did well enough to appease him, but it upset me enough that I still didn’t make the PhD pass on the third exam, a situation of the sort that a number of students would eventually find themselves in as astronomical knowledge rapidly increased and the exams became ever more comprehensive, so that the system was finally abolished.
(This was the time of the Korean war, and most of us had student deferments. After I completed my MA, which is all that Harvard had originally accepted me for, my draft board decided my time was up. After 2 years of hassle by selective service, I was drafted, so what would have been my fifth year found me serving my time as a conscientious objector teaching at the American University in Beirut. It was there that I took the third PhD exam for the fifth time, finally passing with flying colors.)
After my first semester of graduate school, to maintain my student deferment, I had to switch to full-time status, that is, to take four courses. I continued in math, and went to Bok’s course on galactic structure, which met in the Observatory classroom. Bok had been away at the South African station during the first semester, but he immediately made his presence felt in the Department upon his return. Bok was very popular with the students because he was empathetic and interacted with them regularly, though he was hard-driving and had expressed the opinion that an occasional nervous breakdown would indicate that the general level of work was about right. He spoke several languages, all with a Dutch brogue, and he loved to tell jokes, especially at other people’s expense. Bok would come into the classroom a couple of hours before the lecture and fill the blackboards with dense notes including references. Then, during the class periods, he would tell wonderful anecdotes about Kapteyn and the other astronomers concerned with galactic structure. The anecdotes had essentially nothing to do with the required course material, and the only way to cope with that was to retire to the Tutorial Library afterward and read the original articles carefully. Because a teaching fellow copied the blackboard notes and distributed them, there seemed no reason to take notes in class, so I had a misspent youth—I should have been writing detailed notes on the anecdotes!
I also took Astronomy 140, mathematical astronomy, a standard course taught by Bart
J. Bok and Ivan King. (Ivan King had been with Bok in South Africa, and was just finishing his thesis.) A group of us met in Byerly Hall for lectures, but we generally worked on the problem sets up at the Observatory. Central to the course was solving one each of the three kinds of binary star orbits: visual, spectrographic, and eclipsing. Today these are handled quite differently by computers, but then we worked graphically with planimeters or with elaborate tables (for the eclipsing binaries) found in the Princeton Observatory Publications. Finally the big blockbuster assignment was working out the trigonometric expansions for Delaunay’s lunar theory. 27 I remember working hard on this assignment with Mac McCrosky at his house.
Because of my full-time enrollment, I was obliged to give up my work at Sky and Telescope, so I was given a research assistantship under Mrs. Gaposchkin’s direction. Now, essential to the quantitative study of spectrograms was being able to make intensity tracings of them. The Observatory had earlier managed this by a Moll microphotometer, a rather clumsy Dutch device that worked photographically. However, Dr. Menzel had cut some sort of deal (as a result of his Air Force consulting, I believe) for a major electronics company in Iowa to provide a photomultiplier, amplifier, and chart recorder, which had been installed on a measuring engine with an appropriate screw drive to make a modern microdensitometer. It was my task to tame this device and make it work. I spent many hours on this project, being paid out of an Air Force contract. Eventually, the following year, I was able to demonstrate that there was so much scattered light in the optical set-up that it simply couldn’t make satisfactory scans.
In addition to the courses I actually took, I audited two others. In the fall Fred Whipple gave an omnibus course related to his interest in comets, meteors, the interplanetary medium, and the earth’s upper atmosphere. He lectured on his dirty snowball model of comets, which had just been published, and about the distinction between cometic meteors and asteroidal fireballs. 28 I was greatly impressed when he showed an illustration of the physical state of the earth’s upper atmosphere in which each point involved the firing of a captured V-2 rocket, at a cost of $25,000 per point. The other course, in the spring, was team-taught on aspects of observational cosmology. Shapley rarely lectured in the graduate school, but was on deck for this course. He assigned to Frank Orrall the problem of figuring out why the globular clusters of the Andromeda galaxy seemed systematically fainter than those of the Milky Way. It did not occur to either one of them that the distance scale might be involved. That news broke to the astronomical world just a few months later at the Rome IAU meeting where Walter Baade announced the revision in the distance scale. 29
Approximately every other week there would be a colloquium in the Phillips Library. At the front of the room was a large exhibit case with the vertical mirror across the back. Bok always sat in the front row and would use the mirror to scan the audience, taking roll of the graduate students. One could miss a colloquium with impunity, but missing a second time was sure to bring a warning notice from Bok about the importance of attending all the colloquia.
A very important astronomical discovery had taken place in the Harvard physics department in the spring of 1951, before we had come to the graduate school. This was Purcell and Ewens’ detection of the 21-cm radio radiation from neutral hydrogen, the direct result of a long visit from Henk van de Hulst, who had predicted the line. 30 Professor Bok had been on leave during that year, but on his return he saw an opportunity to open a promising new field of investigation. In this he was fortunate to have Heeschen and Lilley as graduate students, as they had a background in electronics and were keen to get into the new field of radio astronomy. Also, Tom Matthews, with a background in galactic structure in his MA from Toronto, was interested in this line of research. Accordingly, Bok, those graduate students, and an undergraduate, Campbell Wade, began building a 25-foot radio telescope at Oak Ridge.
In the summer of 1952 Bok enlisted the graduate students in a special summer school program, with well-prepared lectures by various graduate students twice a week, Heeschen and Lilley, for example, reported on the current status of radio stars. I researched and spoke on studies of galactic structure in Cygnus. At the time I was working once more for Sky and Telescope, but Bok pressed me into preparing the ditto notes for the summer school series. I typed the masters on my own typewriter, which had a series of special Greek-letter keys, and I printed the notes on the Observatory’s ditto machine. Getting these out very promptly after each session produced a serious conflict of interest with the Federers. I found myself having two masters, not a very pleasant situation. The graduate students all worked very hard that summer, and Fred Whipple, the department chairman, was known to have disparaged Bok’s series, saying that young minds should lie fallow during the summer. This brought about the famous graduate student rejoinder, that they wanted to apply for Whipple fallowships.
In my second year I took Donald Menzel’s standard Astronomy 200 quantum mechanics and theoretical spectroscopy course, heavily based on Menzel’s own work and on the textbook by Condon and Shortley. 31 It was, essentially, a required graduate course for students hoping to pass the astrophysics PhD exam. I found the course extraordinarily baffling, with mathematics very difficult to relate to Mrs. Gaposchkin’s world of observed spectra. Though I couldn’t understand what was going on, I was fascinated by matrices and got quite good at normalizing them. I finished the 3-hour final exam in just 2 hours, and got a 100% score.
That fall I also took Bok’s course on stellar motions and galactic dynamics, which alternated with his stellar distributions class. Bok put it as a challenge for anyone who wanted an automatic A on a course paper to do a tau-upsilon solution 32 for solar motion and statistical parallaxes for a chosen spectral type. Today this would be pretty trivial with modern computers, but then it was a heroic chore with a Marchant calculator. Bill Tifft (an undergraduate) and I look up the challenge and got our As.
My interest in observational spectroscopy made it natural for me to be assigned to Mrs. Gaposchkin as advisor. Though I had done well in the spectroscopy courses, my initial encounter with math at Harvard was a major disaster—my background from Goshen was quite inadequate. I got in over my head, and I had done so poorly that I was essentially on probation and unqualified for a teaching fellowship.
1952: Menzel becomes acting director
By this time the Observatory was undergoing major changes, for Shapley had come to the end of his directorship. In September of 1952 Menzel was appointed acting director, and he immediately began a campaign of cleaning up the Observatory and repainting the somewhat dingy rooms. He had little patience with history, and set to work to throw out what he considered a useless residue of clutter, including Annie Cannon’s card index of variable stars, many fading framed pictures including most of the Trouvelot lithographs, and perhaps 100,000 of the glass plates in the photographic collection that showed no images or hopelessly blurred or fogged images. This latter action created a storm of criticism, but he stuck to his goal, which was largely carried out by Sergei and Cecilia Gaposchkin. 33 Plans were begun at once to replace the old frame building A that housed the Tutorial Library and the AAVSO.
Menzel had got the acting directorship because he had about 6 month’s seniority over
Bart Bok, who remained the overwhelming favorite of the graduate students. Or at least that is what we graduate students were told. The information was leaked to us that Bok had been earlier offered the directorship of the University of Michigan Observatory, and Shapley had urged him to stay at Harvard because he was the obvious successor when Shapley retired. The graduate students were quite exercised by the matter, and the great majority of us signed a petition to the Harvard Corporation to make Bok the permanent director. We were basically unaware of the heavy politicking at very high levels, and animosities that went back to the 1930s when Shapley was a major competitor to James Bryant Conant for the presidency of Harvard itself. Conant considered Shapley and his leftist involvements a thorn in his side, and he and his lieutenant Paul Buck believed that the way forward in the ever-more-expensive sciences was to gain government support. Both Shapley and Bok were very suspicious of the entanglement with government regulations if they followed this course. Menzel and Whipple, with wartime government connections, saw federal support as essential to get the Observatory out of a poverty mode under which it had long been operating. 34
In addition to the animosities at high levels in the Harvard administration, there were considerable tensions within the Observatory over Shapley’s behavior with respect to the previous summer’s news from Baade concerning the revision of the distance scale. 35 When Shapley had decided to leave Mt. Wilson in 1921 for the position at Harvard, he knew that HCO’s southern station offered an unrivaled opportunity to study the Magellanic Clouds. The smaller apertures of the Harvard instruments were compensated by the relative closeness of the Clouds, compared to the Mt. Wilson 100″ and the Andromeda Galaxy. Shapley was sure that some major discovery awaited the investigator of 36 the Magellanic Clouds. He had gambled his career on this when he left Mt. Wilson for Harvard. There was indeed a discovery to be made, but the problem was that he didn’t make it. This was finding cluster-type stars in the Clouds, which involved looking fainter than his earlier assumptions suggested. Shapley felt so faked out that he could scarcely help trying to get into the game himself. The other Council 37 members (or at least Gaposchkin, Menzel, and Whipple) felt humiliated by Shapley’s actions, and wrote an extremely apologetic letter to Baade. We graduate students knew essentially nothing of this, and continued to support Bok. We only knew that Shapley had been relegated to an extremely tiny office and that his chief assistant had been sent packing.
For me, Shapley continued to be a great hero and a principal reason for my entering astronomy. I decided it would be memorable to take a research course with him. No doubt I would have counted nebulae or worked on Cepheid periods, but it probably would have resulted in a joint paper. But my advisor, Mrs. Gaposchkin, would have none of this. “He’s just looking for a way to get free labor,” she ruled rather scornfully, and advised me to sign up with a reading course with Menzel instead. Menzel was full of the idea of hypergeometric series at that time, for he had had some success in using them to represent the hydrogen wave functions. Another graduate student, George Conant, and I were put to work investigating hypergeometric series, but it was a frustrating task that led nowhere. It would have been better for me to have been a slave for Shapley. (Over the years my relationships with Menzel considerably improved, and he was primarily responsible for my early election to the American Philosophical Society!)
One other hurdle remained in that second year for completing my MA: the language exam. This had provided great difficulties for some of the graduate students who were good at math or mechanics but not at languages. The procedure was that a 38 student would be given 50 pages of a foreign text or journal article, and 48 hours to study it. Then he would be asked to give a pretty careful oral translation of some passage chosen by the examiner. Bok gave me the French exam, asking me to work on an article on novae. My single year of French in college was enough to get me through the exam. (Later Mrs. Gaposchkin gave me the German exam with a selection from Unsöld. If I had flunked a couple of times, the tries would have done wonders for my German reading, but I passed the first time.)
In the spring of 1953 I got my MA, and my parents came to Cambridge for the Harvard Commencement. Since I had originally been accepted just for an MA, my Indiana draft board figured I was through, and moved to draft me. I mention this because the next 2 years of graduate school were filled with uncertainty about how long I could remain in Cambridge, and I found myself treading water with respect to a thesis. The other part of the problem was that I was interested in observational spectroscopy, and Harvard, apart from Mrs. Gaposchkin, was a terrible place for that since there was no observational facility for obtaining high-dispersion spectra. I was therefore at the mercy of whatever material Mrs. G could beg or borrow from Victoria or McDonald Observatory.
I did have a remarkable opportunity in the summer of 1953, when Tom Matthews and I became the Harvard representatives at a 4-week NSF astronomy summer school at the University of Michigan. This involved direct interactions with Walter Baade, and an opportunity to meet other participants such as Allan Sandage, Don Osterbrock, Vera Rubin, Geoff and Margaret Burbidge, and others who have remained good friends. I later wrote up this experience in some detail in Physics Today. 39
1953-55: Teaching and marking time
In my third year of graduate school I was deemed qualified to become a teaching fellow, so I interviewed for a position in the Nat Sci 3 course offered by a young professor, I. Bernard Cohen. He occupied a large office in Widener Library that had formerly belonged to the pioneering historian of science, George Sarton. Cohen asked me several questions about the history of physics, some of which I could not answer, which turned out to be OK because they didn’t have answers. His class was very large, 350 students filling the large B lecture hall in Allston Burr Lecture Hall. There were lots of teaching fellows, and Cohen wanted one from astronomy in order to gain entree to the Observatory for some observing nights. Sections met twice a week, so being a TF was a time-consuming task.
The other activity outside the Observatory was courting my girl friend, Miriam “Sense” Sensenig, a former classmate at Goshen College who during my second year in graduate school had come to Boston with a fellow RN and who had got a job at a research ward in Peter Bent Brigham Hospital. The following summer she had got a job as nurse at a summer camp in Colorado Springs, so after the NSF Summer School at Michigan I arranged to drive a car out to Colorado to attend the AAS meeting at Boulder and then to go with her to Los Angeles. We visited Baade on Palomar and Sandage at Mt. Wilson, and then I hitchhiked back to Cambridge in time for the Harvard fall term. During the year Sense and I went out together many times, and sometimes when she was on night duty I would go in the back door and up to her ward to visit. I told her that I hoped to go on a honeymoon that summer to the total solar eclipse in Michigan, and if she wanted to see an eclipse, she should act interested. By mid spring she agreed.
During this year (1953–54) the new Building A was constructed, and Donald Menzel moved into the new Director’s Office at the west end. He was officially appointed Director in February. Menzel had managed to throw out the AAVSO, which had been born at HCO; this move took place in part because he wanted to capture a large endowment of about $100,000 from the Stuart Wyeth bequest, the so-called Pickering Memorial Fund, which had been given directly to the Observatory though the intention seemed to have been to support the AAVSO. I felt particularly dismayed by this turn of events, since of course I had originally come to HCO through the AAVSO. The astronomy graduate students were especially exercised by this event, and most of them signed a petition against it. The Sky and Telescope staff laid low during this episode, but it was clear their days at the Observatory were also numbered. These two episodes have been recorded at greater length by Lief Robinson in the special 150th 40 HCO anniversary issue of the JHA (21(1990), 89–103).
Also at this time the residence, which adjoined the 15″ refractor’s tower, was razed. I was asked to document the rooms with my camera, and Shapley’s assistants (Ann Hearn and Jacqui Sweeney) made a scrapbook of colored prints for him; the book is now with the Shapley papers in the Harvard Archives. In addition, Menzel was keen to modernize the Great Refractor, and as a start set his newly appointed engineers, Hector Ingrao and G.R. Miczaika, to work to replace the old Clark sector drive with a new electronically-controlled drive. I think it was only very much later that I learned that the end of the telescope tube had been smashed in an unfortunate accident while the telescope was being de-assembled,
That fall I signed up for Astronomy 243, a theoretical solar astrophysics course with Dick Thomas, a galling experience because he was one of the most disorganized teachers I have ever had. Nevertheless, in the spring I signed up for his stellar atmospheres course, in which I got an A, although I can’t remember any details of either course.
While I vividly remember exactly who entered the astronomy graduate school in 1951 when I came, the members of the next classes are only fuzzily impressed on my mind. They were smaller in number, 41 and not everyone finished PhDs (for example, Pierre Fisher, George Conant, Mary Connally, who married Allan Sandage, or Mary Jane Stewart, who married Campbell Wade). By the end of the fourth year, two of my classmates finished, first Dave Heeschen, who had already come with an MA, and then Chuck Whitney. Ed Lilley, Dick McCrosky, and Tom Matthews received their degrees the following year, while I was in Beirut. Meanwhile a large group of students in addition to Heeschen, Lilley, and Matthews had been attracted to Bok’s radio astronomy program. These included Kochu Menon, Campbell Wade and Bob Davis (who had been Harvard undergraduates), May Kassim, Nanilou Dieter, and Bill Howard.
Part way through my third year, my draft board had withdrawn my student deferment, on the grounds that I had received the MA degree I had originally been admitted for, and this led to a convoluted appeals procedure that left me in a very uncertain position. This made it difficult to get launched into a thesis topic, especially because I was keen on spectroscopy and at Harvard it was essentially impossible to get adequate material. Mrs. Gaposchkin secured some spectra from Lawrence Aller, but again, with the fiasco of the microdensitometer, it was impossible to produce satisfactory tracings. I learned how to measure and calculate a curve of growth, 42 but the thesis seemed to be going nowhere. The uncertainty with the draft board continued in a new phase in the fourth year, but it became increasingly apparent that I would be drafted at the end of that year, so I threw myself into the teaching with Professor Cohen. Cohen commissioned me to write some astronomical sections for his textbook, 43 and, unusually, invited me to give at least one of the main class lectures. Otherwise I just marked time.
To keep busy I was assigned to work with Bill Sinton, a recently hired instrumentalist, on a proposed eclipse expedition to Ceylon. Donald Menzel knew Ralph Damon, the CEO of TWA and a man who had enjoyed astronomy at Harvard as an undergraduate; Damon would not only support the new patrol cameras for the Observatory, but, because TWA flew all the way to Columbo, he arranged for TWA to sponsor an eclipse expedition there. I had had experience building Heathkit amplifiers and an FM tuner, so I wired the amplifiers for the eclipse photometers. Sinton was quite impressed with the tidy arrangement I produced, saying that it was in marked contrast to my Heathkits. “But,” I explained, “I followed the Heathkit instructions precisely, including the messy wiring, because I couldn’t know if the placement of the wires was essential to prevent interference in the circuits.”
The eclipse came almost immediately after the Harvard exam period, and of course I was obliged to complete the grading in Nat Sci 3. Prof. Cohen agreed that we could get out early if we teaching fellows would construct a section of the exam that could be machine graded. We did this, but I learned something important about multiple-choice exams. It was almost impossible to make hard questions without merely making them ambiguous, and it was most likely the best students who saw the ambiguities and therefore missed the questions. In any event, I was quickly ready to fly via TWA to Columbo. Sinton and I stayed for 2 1/2 days with the same propellered Constellation, while the crews and eventually all the passengers changed. Later Hal Zirin joined us in Ceylon. We found a jungle site near Sigirya, a fabulous volcanic pipe and historical stronghold. I was in charge of setting the concrete footings for the equipment. We were about ready to pour the cement when I realized that I had reversed north and south. This was the first place I had ever been where noon shadows fell to the south!
The day of the eclipse turned out to be overcast. My back was to the eclipse, because
I was assigned to make a movie film of the meters. With our filter wheel and photometers we hoped to measure the infrared radiation of the corona. When I saw that the meters were recording nothing, I turned around for the final seconds and used my Argus to photograph the diamond ring through the partial clouds as the eclipse ended. The eclipse was a failure scientifically, but a memorable tourist experience.
On the flight back I sat beside Mr. Damon going out of Columbo. At that time TWA had a contest to predict the state of commercial aviation 30 years in the future, with the entries to be microfilmed and opened in 1985. I said to Mr. Damon, “First of all, everything will be jet.”
“Oh no,” he replied. “Jets will never be commercially viable, especially for intra-continental travel.” I figured I was so out of it that it wouldn’t do any good to enter the contest. Incidentally, the prize was won by a woman historian of science who had her PhD from Harvard!
To Beirut and back
On the way back I stopped off in Beirut to check up on the American University, where Miriam and I would spend the next 3 years. During that time I learned a lot of physics, met interesting visitors such as Lee DuBridge and John Van Vleck, attended the Astrophysics Colloquium in Liège, observed Comet Arend-Roland, and finally attended the IAU meeting in Moscow on my way home in the summer of 1958. During that same time Sputnik was launched, the Smithsonian Observatory moved to Cambridge and acquired the fastest computer in New England, and Harvard Observatory began bursting at the seams with the influx of its own new staff members as well as the Smithsonian satellite-tracking personnel.
While in Beirut I had hoped to apply for an NSF graduate fellowship for college teachers, but unfortunately I was uninformed that the application deadline was moved earlier that year. By this time we had two small children, and I was concerned about an income for the next academic year, not realizing that the Astronomy Department would have taken care of us. An offer to pinch hit for a year at Wellesley solved our financial dilemma and brought us back into the Harvard circle, but deferred my graduate study yet another year.
There were two big changes at HCO when we returned. First was the enormous expansion brought about by the arrival of the Smithsonian. Second was the fact that Bok was gone, having left for Australia, and in his former office was Mrs., now Professor, Gaposchkin. “They didn’t have anyone for Department Chairman,” she ingenuously confided, “so they had to make me a professor.”
Mrs. Gaposchkin had used her brownie points to persuade Walter Baade to come for a series of lectures that fall, so twice a week I commuted in from Wellesley to listen to his class. About 40 persons attended the first lecture in the Building A classroom, including three new professors: Theodore Sterne, Tommy Gold, and J. Allen Hynek, and eventually attendance settled down to 20–25.
It was my custom to send a weekly letter to my parents, and I have preserved the carbons of these missives; I scanned through them for information about the Observatory, but my year was very much focused on our little boys, my teaching at Wellesley, and our precarious financial circumstances, which were substantially helped by the fees I collected for writing a Scientific American article 44 and a series for Canadian Audubon. The Harvard Astronomy Department helped with funds for me to attend the Rochester AAS meeting at the beginning of spring. I brought Baade to Wellesley for an informal conversation with the astronomy students (who included Andrea Kundsin [Dupree], Anne Pyne [Cowley], and Barbara Welther). In the spring a newly appointed HCO staff member, Gérard de Vaucouleurs, came to give an astronomy lecture at Wellesley, and this contact proved very interesting. He and Menzel had become interested in the forthcoming occultation of Regulus by Venus that July, and I realized that the American University of Beirut Observatory would be a particularly suitable place for observing this event. In a very short time transportation was arranged for me (via the Military Air Transport System), and I took a movie camera and film to Beirut and installed it on the 12″ refractor. My filming was only partially successful—I missed the second contact because the calculations for the duration of the event had failed to consider the rotation of the earth itself—but the measurements eventually turned out to be among the most accurate of the various stations, even better than the ones de Vaucouleurs himself had made. It took continuing work on my part during the fall and into the spring to reduce the chronograph recordings of the event. 45
Returning to graduate school, Fall 1959
Meanwhile, I received a Jewett Fellowship for the next year, and Mrs. Gaposchkin helped me organize to teach an extension school astronomy course at Clark University in Worcester. I was assigned an office on the ground floor of Building A, which I shared with Peter Usher, a new graduate student from South Africa. Early that summer we had returned briefly to my home in Goshen, Indiana, and there I learned that Leo Goldberg and William Liller were about to come to Harvard. I mentioned this freely upon my return, and shortly thereafter a very distressed graduate student, Dick Teske, pulled me aside to say that this news was secret, that he had been told very confidentially, and now the people who had told him assumed that he had spilled the beans. I found this very funny, but rather typical—that one had to go elsewhere to find out what was happening at Harvard!
It was my intention to work on the spectrum of Arcturus, because Mrs. Gaposchkin had obtained some high-dispersion plates taken at McDonald Observatory. I made many line identifications, but felt at a loss as to precisely what I should do with the material. At the end of August, as I was returning to graduate school, I attended the AAS meeting in Toronto, and asked around about potential ways to exploit the K-giant spectra I had. All the experts seemed devoid of suggestions, and I began to suspect that the reason the plates had been handed over to me was because no one else knew what to do with them either. By November it became apparent that, at the very least, I would have to undertake some appreciable theoretical work to complement any observational information that I could squeeze from the plates. This led me to consult with Chuck Whitney, who, after completing his thesis in short order had gone off to Liège as a post-doc to work on the astrophysics of pulsating variables (while I had gone to Beirut). He had since returned to the Smithsonian, where he had been pressed into service doing celestial mechanics for the satellite tracking program. Chuck was interested in getting back into stellar astrophysics, and he suggested that I should compute (by hand) a model stellar atmosphere. This I did by slide rule (in November), iterating each step for the ionization balance of the hydrogen and a representative metal.
The influx of satellite tracking staff was accommodated with space in the Gray Herbarium Building across Garden Street (now the Harvard University Press) and on a floor in the IBM Building on Cambridge Street. Chuck Whitney had his office in the IBM Building, very near the IBM 704 computer (whose configuration he had helped specify), and I soon found myself sharing his office, because the next step after hand-calculating a model atmosphere was to program it for the computer. I had to learn FORTRAN, how to keypunch cards, and how to debug programs, and these were all accomplished by being close to the computer and the programmers who also inhabited that floor. By early in December I had the hydrostatic equilibrium part of the model programmed, with newly devised logarithmic steps in optical depth, but not until the end of January was it all working properly. One of the considerable difficulties was getting the starting steps to work without the numbers getting out of range, and similarly to prevent a numerical overflow or underflow at the largest optical depths. Also, the mixture of ionizing elements was more complex than in the hand calculation, and the iteration was trickier.
At that time I officially switched from Mrs. Gaposchkin to Chuck Whitney as thesis advisor. By the beginning of March I was finally onto new ground, and had demonstrated that one published result was off by a factor of five because of the neglect of the most important term in one of the equations, which according to my family letters suitably impressed Chuck. By then I also realized that I would not finish by June, though I still had hopes of completing the work by the end of the summer. Consequently I turned down the opportunity to go in person to the Liège Astrophysics Colloquium, something I had occasion to regret later on when it took a whole additional year to complete the thesis. By mid-April I was discouraged that results from one of my routines did not match those published in the ApJ, only to discover that the published ones were wrong.
Though very primitive by today’s desktop computers, the IBM 704 was the largest and fastest electronic computer in New England in 1959–60. Its basic cycle time was 2 microseconds; an addition or subtraction took two cycles, and a multiplication or division six or seven cycles. It had 8000 32-bit words, and a high-speed drum for an additional 8000 words. Tiny ferrite rings wired on close-mesh screens provided the core storage. The central processor operated with hundreds of vacuum tubes, and it was quite extraordinary to watch the engineer doing routine maintenance. He would program a loop into the computer, which caused the small neon lamps on the operator’s console to glow only faintly. Then, with a mallet, he systematically tapped each tube. The unsteady, microphonic tubes would give a pulse that would stop the loop, and the neons would light up brightly on the console at the position where the program stopped. The engineer would replace the faulty tube, restart the program, and continue the process.
By May I had begun to program the non-grey 46 radiation flow through the atmosphere, using exponential integral functions to evaluate the flux as a function of depth and frequency. At first I evaluated the exponential integral function with a series expansion, but later I discovered that a Tschebyshev 47 polynomial worked significantly faster. This made the program plus its storage too large to fit into the 8000 words of storage, so I constructed it in two parts, writing the intermediate values onto the drum and picking them up for the second part of the cycle. On good days I could get perhaps three turn-arounds on the computer. Compiling the code took one pass (if there were no errors that stopped the compilation), and this produced the binary cards that were actually used to run the program. Eventually I learned enough details of the procedure so that I could sometimes correct a faulty program by hand-punching a binary card, thus saving one recompilation pass. The non-grey or wavelength-dependent part of the program, in addition to requiring a procedure to evaluate the flux at any layer, also required frequency-dependent opacity routines. For this I had to make a polynomial fitting program to approximate various opacity sources such as the H-ion. There was not enough storage to keep more than one wavelength available at a time, but I could save about a page of output at a time when I wanted to examine the optical depth or opacity details.
During this time a new building was under construction on Observatory Hill. Since the old machine shop, Building B, had long since been condemned and razed, the letter B was reassigned to the large, new building. The cornerstone was laid in April; Senator John F. Kennedy was chairman of the Visiting Committee, but too busy campaigning 48 to come.
The summer of 1960 was a long story of fighting programming bugs as I built more and more of the flux integration program. Sometimes the computer was so crowded that there was a long wait for a turn-around. Normally Chuck Whitney was around, and with his excellent physical sense he could generally sniff out when something was wrong. He helped me estimate that I needed 15 minutes of computer time per day, and arranged for me to have some Smithsonian employment in the fall. That summer I taught the Harvard summer school course in astronomy, and in the 1960–61 year I also got partial employment in the General Education program, giving part of the main lectures in Nat Sci 3 along with I.B. Cohen.
By September I wrote the final section of the program, which was to correct the initial temperature distribution so as to achieve flux constancy throughout the various layers, based on a theory by Max Krook. It proved surprisingly easy to write this part of the code in comparison to the frequency-dependent flux integrals. The program continued to be in two parts in order to fit into the IBM 704. When the logical flow of the program came to an “END” statement, the compiler assembled the code to activate the card reader. I could stand beside the computer and load the binary cards for the program into the hopper, and keep manually recycling them as the program iterated between the two sections. Unfortunately, however, Krook’s temperature correction method refused to work iteratively, so I was obliged to explore alternative correction methods.
As the new academic year began Chuck moved to 60 Garden Street, and I got my own desk at the IBM building, complete with a telephone and Monroe calculator. Chuck was offering a stellar atmospheres course that fall, and asked me to compute some samples for his class. Otherwise progress seemed slow, with innumerable bugs, and Max Krook’s “fix” for his correction method not working.
After the Christmas break Building B was finally ready. Chuck Whitney had a major new office near the southeast corner of the third floor, Luigi Jacchia and his staff had the next ones to the north, and then there was an office that I shared with another of Chuck’s students. The Smithsonian had its own library, one floor below us. The key-punch room, with about half a dozen women key-punch operators, was two floors below, and the new 7090 computer a floor below that. Some programmers remarked that the new computer room was deliberately made too small for an IBM 709, which still used vacuum tubes, so that we had to get the more powerful but more compact transistorized IBM 7090. It was downward compatible with the 704, but had 32K storage and additional instructions. By the end of January I was busy recompiling all my subroutines, with a whole new storage allocation scheme since I no longer had to use the drum and the two-stage system.
When I returned from the Christmas holidays Chuck Whitney broke the news that he thought my computations weren’t quite right. He refused to say what he thought was the matter, because he wanted me to find it myself, but he did lend me his notes. It took a little while to realize that the complexity of the calculations made the situation highly non-linear, and rather far from his linear approximations. I felt very one-up when I convinced him that my calculations were in fact correct.
Meanwhile, Donald Menzel had asked me to help with a new edition of Oppolzer’s Canon der Finsternisse, and I undertook to translate the German introduction (which turned out to make a 60-page manuscript) and to write a new preface for it. 49 My translation passed without Menzel making any editorial change, which was quite remarkable since he was known as a strict editor for the theses he had advised. I also had responsibilities to find a few new teaching fellows for our Nat Sci 3 course, where I had to make up the final exam as well as to give the final section of lectures in the first semester. As the second semester began, I agreed to give four lectures on FORTRAN programming to one of the graduate courses. I was also heavily involved in the Nat Sci 3 lectures. I wanted to use a major radioactive source from the physics department, and needed to arrange this with Gerald Holton. He really dragged his feet about it until he learned that I, and not Bernard Cohen, was giving the lectures. “That changes everything,” he replied, “I don’t like I.B. Cohen fooling around with this stuff.”
By February the 7090 was running, but it was so much faster than the printers that sometimes it took 3 hours just to get the printed results. Sometimes the machine performed erratically, and there were bugs in the systems tapes. By March I felt I had caught up to where I had been in December, and by mid-March I was exactly where I had been in October, that is, now fully convinced that Krook’s temperature correction scheme was incapable of iterating. Max Krook had been so insistent that there was something wrong with the way I was handling it that I lost a lot of time before I finally convinced him.
Eventually, on the morning of my PhD exam, he rushed in with a new variation of his method, which became the famous Avrett-Krook procedure for correcting the temperature distribution. However, by April I had found an alternative scheme, so that I could use Krook’s procedure for the initial corrections and then a less powerful but iterative procedure to finish it off. Finally I had a working tool with which I could grind out a series of non-grey atmospheres, the first machine-made models and without major mathematical simplifications.
By June I reported on my calculations at the AAS meeting in Nantucket, and most of my thesis was actually written. In the process I discovered that I had not been using Krook’s procedure correctly. But the irony was that my version actually worked better than his! He then figured out why my procedure worked better. At the end of the month Chuck again challenged some of my results, and I had to program a number of special tests that he devised. Finally I devised a test that absolutely convinced both of us that everything was OK. He also said that my thesis would be good enough that Smithsonian would pay to distribute at least 50 copies. The final production of the thesis at the beginning of August was quite an operation, with Chuck Whitney’s secretary doing the typing, Chuck handling the proofreading, one of his summer assistants running the “autostat” copying machine, and another (Dave Latham) tracing the graphs into their final form. Dave stuck it out until all five copies were in their binders. 50
Chuck was very keen to prep me for my final oral, as a good performance would reflect favorably on him. In the final week he suggested all sorts of material for me to bone up on. The oral exam took pretty long—having faced very skeptical and probing Arab college boys and having learned how to solve all the problems in the Sears and Zemansky physics text without prior preparation, I was not too fearful of the ordeal, but I was too bushed to realize that the examiners weren’t so much interested in what I knew as in what I didn’t know. It took them some time to discover that I couldn’t derive the Rosseland mean, 51 and then they were satisfied.
The next day I left Cambridge, headed for the IAU meeting in Berkeley. I had a secret role to play there. The CIA wanted to learn as much as possible about Soviet astronomy, and they proposed to use the daily IAU newspaper as a front for conducting interviews with various leading Soviet astronomers. I agreed to help as an interviewer-reporter. In any event, it paid for my trip and accommodations, and I enjoyed the opportunity to meet not only the Soviet astronomers but many others. It was my thrilling, brief role as a cold war spy!