Obituary - Richard E. Marsh (1922 - 2017)

Memoir | Publications Curriculum Vitae | Videos | Slides | Articles | Obituary  


Remembering Dick Marsh  (1922 - 2017)

ACA RefleXions, Spring 2017

Richard Marsh, senior research associate in chemistry, emeritus, at Caltech, passed away on January 3, 2017, at the age of 94. Doug Rees, the Roscoe Gilkey Dickinson Professor of Chemistry and faculty director of the Molecular Observatory for Macromolecular Crystallography at Caltech, said, “Dick was a legendary one-of-a-kind crystallographer who was recognized for his mastery of the field and his rigorous standards. He trained generations of students and postdocs and was widely respected as representing the heart and soul of crystallography.” Following are the remembrances of several of our community who knew Dick well.


Harry Gray: I met Dick Marsh briefly when I first visited Caltech in 1964. But I got to know him well on a second visit in the spring of 1965. We both had deep interests in the structures of inorganic compounds, and we hit it off right away. I well recall the discussions we had on trigonal prismatic dithiolene complexes that were being investigated in my lab at Columbia. When I moved from Columbia to Caltech in 1966, I began collaborations with Dick that lasted until he left us late last year.

Dick and I did science and played tennis together for many years. The tennis game at his home on Saturday mornings was a highlight of my week during the 80s and 90s. He was a natural athlete, and he gave me fits when I played against him. I always tried to get him on my side, so I would not have to face his sneaky shots. But even when he tortured me with drop shots and wild top spin lobs, it was fun!

Doing science with Dick was special. His standards were off the charts. In doing structures of my inorganic compounds, I did not want to be ‘Marshed.’ So, on very challenging structures, I worked closely with him. One of our first joint papers, published in 1969, reported the structure of the first binuclear complex containing molecular nitrogen. That work caused quite a stir in the inorganic chemistry community. Other papers with me and members of my group were on the structures of linear chain platinum compounds. These structures were very challenging, and we could not have solved them without his help.

Dick and I talked almost every day, most especially when the X-ray gang had coffee on the 4th floor of the Beckman Institute, just down the hall from my office. More often than not, I would drop in the coffee room and ask Dick about a structure that I was confused about. Dick would always set me straight!

It is not the same on the 4th floor now. Dick was a great scientist and a treasured friend. I miss him every day. 

Jenny Glusker:
I am saddened by the news that Dick Marsh is no longer with us. He was in the lab at Caltech when I came in 1955 as a postdoctoral student to work with Bob Corey. At that time Caltech was somewhat like a monastery, and I had come from England having just obtained a doctoral degree there. How lucky I was to have Dick there in the lab! He was so generous with his time, helping anyone who needed assistance with machine usage, structure determination and publication. What he did was stress what a wonderful science subject – X-ray crystallography – we had chosen for study, and how important the results of our work were. But he also emphasized firmly that such crystallographic investigations had to be done correctly, with no short cuts and with close attention to any problems that we might detect on the way. Dick and I actually worked together on the crystal structure of a tripeptide, glycylphenylalanylglycine. Since then, through the years, we have often communicated and he helped me when I first acquired a laboratory to run and whenever I received interesting papers that needed strict review so that they could not be refused as ‘Marshed.’ I shall miss his exuberant laugh and the joy he showed whenever a difficult problem was correctly solved by any member of the crystallographic community. 

David Duchamp: Dick Marsh was a great thesis advisor, mentor and friend to me. The four years, 1961 through 1965, that I spent as Dick’s graduate student were among the most productive and the most enjoyable of my life. I arrived at Caltech in the fall of 1961 with a bachelor’s degree in chemistry and mathematics, having spent the summer working with John Burns at Oak Ridge National Laboratory. While at Oak Ridge, I learned about crystallography, and I met some of the top crystallographers at the time, including Bill Busing, Henri Levy, and George Brown, among others. When they found out I was going to Caltech, they strongly recommended that I try to work for Dick Marsh. I followed their advice, and Dick took me on as a new graduate student. This was one of the best decisions I ever made.

One of Dick’s policies was that before a student could use a computer program to calculate Fourier maps, the student had to calculate a Fourier section by hand using Beevers-Lipson strips – this was to ensure that the student understood how Fourier calculations worked. I understood how the strips worked, but I couldn’t see myself summing up all those numbers and getting the right answer – boring! So I had a discussion with Dick, and we decided that if I would write a computer program to calculate a general plane Fourier section (any orientation and any size), it would serve the same purpose and would provide a program that was useful to the group. I devised an efficient algorithm and wrote the program in machine language for the Burroughs 220 computer. Thereby I began a very worthwhile collaboration with Dick on crystallographic computer programs.

Sloppy crystallographic work was not well tolerated by Dick. He was an editor of Acta Crystallographica, and he meticulously checked all papers that came to his desk. Early on, I learned from other members of Dick’s group, Ned Webb, B. D. Sharma, and Noel Jones, to watch for Dick’s reaction when he found mistakes in crystallographic work. One member of the group would watch Dick’s mailbox when a new issue of Acta Crystallographica was due. When Dick brought his mail to his office, we had been alerted and speculated on how long it would take Dick to find something wrong. When he did, we all knew instantly from the loud reaction coming from his office. Dick’s insistence on doing the best job possible with the data available rubbed off on me and on other students of his.


Dick Marsh at work in the Caltech X-ray lab in the 1970s.

One day when I was in search of a suitable crystal structure to work on, Harden McConnell came to Dick asking if someone could determine the structure of trimesic acid, benzene-1,3,5-tricarboxylic acid. McConnell’s group was interested in studying the excitons of crystalline trimesic acid. Dick turned the task over to me. I quickly determined that there were six independent trimesic acid molecules in the asymmetric unit of the crystal structure. This knowledge resulted in McConnell’s group abandoning their project. However, Dick and I both thought that trimesic acid must have a very interesting crystal structure. We both understood the magnitude of the task of determining its crystal structure – 90 independent carbon and oxygen atoms in the asymmetric unit, no heavy atoms, and about 12,000 independent reflections. (This was before automatic diffractometers and before direct methods.) I was eager to try to determine this complex structure, and Dick encouraged me to give it a try. Always concerned for his students, Dick promised that if it didn’t work out, he would find me a different structure to work on for my thesis.

Dick did not micromanage his students, but he kept up with what was going on and didn’t hesitate to offer advice when he felt it was needed. One day I remember he found me studying precession photographs trying to understand what the diffraction pattern of trimesic acid would look like in various orientations. Dick told me about a large optical diffraction machine in the basement that Holmes Sturdivant had constructed. He suggested that I might use this instrument to better understand the diffraction pattern of trimesic acid. As with other advice he offered, Dick’s suggestion was right on the money. Sturdivant was happy for me to use his optical diffractometer. The insight I gained from optical diffraction helped with the solution of the trimesic acid structure.

When Dick learned that Caltech was going to get an IBM 7094 computer to replace the aging Burroughs machine that we were then using, he got a group of us together to work on crystallographic software for the new computer. Dick thought it was wasteful for a crystallographer to have to run separate programs for each type of calculation – the way crystallographic computing was done at that time. He challenged us to come up with an integrated system where the user could move effortlessly from one calculation to the next. We divided up the work, with each of us working in areas that would benefit his current crystallographic project. Ned Webb worked on high-symmetry least-squares software, Carlo Gramaccioli (then a new postdoc in the group) worked on low-symmetry least-squares software, and I had the data reduction component, Fourier calculation software, and overall responsibility for integrating the system. George Reeke and Ivar Ambats also contributed software to the project. Early on we named our new system CRYRM, in light of the fact that when we hit a problem, we cried for help from Richard Marsh. Under Dick’s guidance we produced the first truly integrated crystallographic computing system.

After I completed my Ph. D. at Caltech, and moved on to work in the pharmaceutical industry at Upjohn, Dick and I continued our collaboration on crystallographic software. I rewrote the CRYRM system in Fortran in collaboration with Dick and his people at Caltech. (Of necessity to handle large problems, like trimesic acid, CRYRM was written in assembler language because the IBM 7094 had limited random access memory.) We changed the name of the Fortran version to CRYM to differentiate it from its predecessor. Dick’s group implemented CRYM on VAX machines, and I made it work on the IBM 360 – Upjohn’s research computer.

In summary, Dick Marsh had a tremendous positive influence on me. While I was a graduate student, and since, we had many discussions that were both educational and inspirational. He molded me into the scientist I became, and in the process I adopted many of Dick’s traits, such as, intolerance for sloppy scientific work, and not to be afraid to tackle tough problems.

Ton Spek: Early papers by Dick Marsh made me aware of the issue of missed higher symmetry in published papers and of the danger of ‘being Marshed.’ Over the years, Dick reported on over 1,000 such cases. Not everyone was amused when appearing in his “Some More Space Group Corrections” papers. Several other authors joined in the effort to correct the literature.

My response to the problem was the development of software, based on an excellent algorithm published by Yvon LePage, to routinely and automatically address the missed symmetry issue. With that software, I was able to find and correct one of my early structures before Dick might have found out.

Dick did most of this corrective work by hand and by visual inspection of ORTEP illustrations. One of his papers made me send him an automatically-prepared list of structures deposited in the Cambridge Crystallographic Database that might be candidates for a space group correction. He immediately started to work on that list. The result was a joint paper with more corrections along with the message that there is software available to detect and address the problem before publication. Interestingly, Dick found out that one of the structures he had corrected previously needed again an update to an even higher symmetry. According to ‘hear-say,’ Dick enjoyed very much to have been ‘Marshed’ as well.

The permanent impact of his work in this context is that structure analysts and referees are now well aware of the missed symmetry issue, and if not, warned with an ALERT as part of the now standard IUCr checkCIF procedures. Dick also advocated the archiving of the observed reflection data along with the published structure report. The latter is now also becoming the standard. Suggestions for space group corrections without access to the primary reflection data might be invalid. It was an interesting experience to learn from and work together on one paper with Dick. Howard Einspahr: I was a postdoc with Dick Marsh from September 1970 to August 1972. It was a wonderful experience in every respect. He was an excellent mentor, tolerating my naive efforts at drafting papers, guiding me to improvements in incomplete projects, and importantly, finding me new and exciting projects to develop. He was an open-access advisor, that is, his door was always open … literally, when he was in, his door was open. In discussing my work, it was always a two-way conversation that he guided carefully by asking questions until I understood. I have said before in another context that as a mentor, it was not just what he said, it was often what he didn’t say.

Dick’s lab was continuously filled with interesting grad students, undergrads, postdocs, sabbatical scientists, and visitors, including prior students. I shared lab space with Jim ‘Lou’ Sherfinski and Tom Kistenmacher, and Frank Fronczek was a contemporary and friend, but there were upwards of 25 people who came through the lab and stayed for varying periods of time. Lou, Tom, Frank and I are still in frequent contact.

I said Dick’s door was always open. You knew this because you would hear him frequently. His powers of concentration were unrivaled, but Dick periodically broke that concentration to clear his mind and start afresh. Two times of the day were particularly important for breaks, mid-morning coffee and lunch. Coffee involved a two-block walk to the cafeteria where we would meet crystallographers of all stripes, notably those of the Schaefer, Samson, and Dickerson labs, but actually anyone who wanted to join us that day. Time for coffee for us would usually be signaled by a hearty, “Who, she cried,” from Dick’s office that echoed throughout the basement (or ‘bowels’) of Noyes. At that signal, we would pause, stop, bookmark whatever we were doing, to meet in the hall and walk over. Dick also had a fine sense of when coffee break was over, and after about 30 minutes, when there was an appropriate break, he would strike the table with his forefingers in a tattoo, a sort of shave-and-a-haircut thing, or a ba-doom-siss, and folks began to drift back to their labs.

Lunches were a different matter altogether. These were off-campus in one or another nearby restaurant, were typically in Dick’s yellow convertible Mustang, and the destination was chosen always by consensus. However, when the destination was likely to be Mijare’s, the group expanded to include a wider circle, and guacamole would be on the table waiting for us. I will never forget the Mijare’s enchilada ranchera, which became my only order. Mijare’s is still in business, now in two convenient locations, by the way.

After I left, I visited the lab a couple of times, always rewarding, and saw Dick at many ACA meetings. I realize now I always thought of Dick Marsh as immortal. I wish I’d had more time with him. 

Frank Fronczek: When I was beginning to learn crystallography at Caltech in about 1970, an older graduate student referred to Dick Marsh as, “The Guardian of Goodness and Purity in Crystallography.” How prophetic those words turned out to be, for Dick had a profound influence on the quality of crystallography practiced by everyone who ever spent time at Caltech, but also on an untold number of scientists worldwide. Some years later, Dick and I were touring a botanical garden together at an ACA meeting, and we noted that a new word had been coined; he had been Verbed. He recognized that some who used the term ‘Marshed’ did so in a pejorative sense, but those who care about Goodness and Purity in Crystallography have always used it with great admiration. For his intent was only to help others get it right, and he always afforded authors kindness and respect when he pointed out their errors. Once as a student, I proudly announced that I had solved a structure with Z'=2 in Cc. Dick gently suggested that I go look for an inversion center, and of course it was there. Later, when I did publish a P-1 structure in P1 (oops), he contacted me to allow me to publish a correction myself. This was always his approach, always with kindness, never with malice.

Caltech was a wonderful place to be in the early 1970s. Not only were we younger folks in the presence of such legendary crystallographers as J. Holmes Sturdivant, Edward W. Hughes, Jürg Waser and others, but Linus Pauling also made an occasional visit, and best of all, my lab was next door to Dick Marsh's office in the ‘bowels of Noyes’ (the basement). Dick had a quick wit and a wonderful sense of humor, which led to his unmistakable booming laughter echoing down the hall several times a day. Once, I was trying to solve the structure of a cobalt peroxo dimer with Z' >1 and quite a few potassium ions. The Patterson map was fairly complex for a student of my limited experience. I had traced the contours onto plexiglass sheets with a grease pencil and stacked them up, the 3D graphics of the day. I spent probably a week staring at the map and occasionally running off to try out what I thought might be a solution, but to no avail. One day, Dick walked in and looked over my shoulder at my graphics masterpiece. I heard, “Hmmm.” Then a minute or two later, “Oh…” then “HAHAHAHAHA,” that laugh, as he walked out of the room. He had it! And in just a couple of minutes. But he wasn't going to deprive me of the joy of figuring it out for myself. A day or so later, he suggested that I try direct methods. There were a few programs available on punch cards for doing the rote operations like assigning E values and finding Sigma2 relations, but much of it had to be done manually. I learned a lot, solved the structure, and still take pride in the result. After the fact, he and I talked about why the Patterson looked so confusing, given what the structure turned out to be. This story speaks of both his brilliance and his teaching methods. Thank you Dick Marsh.

An important part of the workday then was Coffeetime. Dick would announce it to everyone in earshot, we would take a short walk across campus, and then spend some time drinking coffee, talking about people's research problems, bug fixes in the CRYM software, and other things, even working puzzles. It will come as no surprise to anyone, given his life's work, that Dick loved puzzles, and he was very good at solving them. For awhile, he used JUMBLE puzzles from the newspaper as a way to help a non-native English speaking visiting scientist improve her English. I still cannot work a JUMBLE (or write one) without thinking of Dick Marsh.

Dick was legendary for having memorized tables of sines and cosines. In the days before computers, when a slide rule and a pencil were the only available means of doing laborious crystallographic computations, he found it easier to memorize the trig functions than to spend time looking them up. A few years ago at an ACA meeting (Cincinnati?) a session was organized to allow some seasoned crystallographers to speak on how things were done back in the pencil and slide rule days. After Dick's talk, the first question from the audience was, “Dick, what's the cosine of 26.5 degrees?” asked as I recall by Wally Cordes. His answer: “Let's see now, 0.89 4, uh…, 8?, no, nine!” Correct. Amazing.

We have lost one of our giants, The Guardian. Despite his advanced age, Dick Marsh's vitality and keen mind made it seem that he would be with us forever. And in a sense he will, in the memories of those who had the great honor of knowing him, and in the quality of the structural results we produce because of his steadying influence. 

Bernie Santarsiero: I moved to Caltech in 1980 to join Dick as a postdoctoral research fellow. I recall showing up on campus on a Monday morning, and hunting around for his office in the basement of Noyes Lab. It must have been around 9:30 am since he was leaving Noyes and heading east towards “the Greasy Spoon,” more conventionally known as Chandler, the campus cafeteria. His ritual was to have morning coffee with the crystallographers, where typically Dick Dickerson would join us; lunch at the Continental Burger; and afternoon coffee with the programmers, including Jean Westphal, his main programmer. Conversations were always casual, cordial, and entertaining. Dick was interested and knowledgeable in everything.

Unlike a conventional postdoc, where you are given a number of options to pursue, Dick encouraged me to design my own project, and I researched his past interests as well as learned more about the organometallic chemistry research going on. I often pitched ideas to him about interesting studies, only to find out that he had the same ideas decades before. The X-ray laboratory at Caltech was a teaching lab, and once a student or postdoc was trained, they had access to everything. There were more than eight cameras available to screen crystals (and powders) at different wavelengths, and two diffractometers, soon to be three, that could collect data from 20K to room temperature. The computing system was home-grown, CRYM, and if an error was suspected, Dick was asked to check the crystallographic equations while Jean would check the code. Later on, when we converted from an IBM to Digital/VAX computer, his daughter Kirby also helped with programming. The lab offered an opportunity to work with great chemists and train them in crystallography while fine-tuning my own understanding.

Dick was an amazing crystallographer. I recall working with a graduate student who had crystallized a Pt complex with a Pt–Pt bond and 4-fold symmetry. I guided the student into taking a series of photographs to determine the space group and unit-cell parameters. After a few days, we reported on our interpretation of the Weissenberg photographs, choice of tetragonal space group, and collected a complete data set on one of the diffractometers. Dick had asked us to show him the Weissenberg photographs. Following that, we generated the Patterson map, determined the coordinates of the two Pt atoms, and then completed and refined the structure. Upon reporting back to him, he asked for the coordinates of the Pt atoms. He retrieved a note on his desk, and had the coordinates of the two Pt atoms written out to three places, from the interpretation of the photographs, and they matched! He had noticed the modulation in the intensities down the c-axis yielding the Pt–Pt distance, and worked out the rest of the coordinates as a special position. This kind of insight led to his uncanny ability to look at hundreds of structures and recognize from the figures and coordinates that the space groups were assigned to ‘unnecessarily low symmetry.’ I had the office next door, and often I would hear his characteristic hoop and holler at finding another suspicious structure. Many were the result of an unrecognized inversion center, but others were far more complicated and required changing from one crystal system to another. I vividly recall one conversation with him where we were discussing a crystallographic problem, and he shouted out, “Oh good!” I asked what he meant, and he replied that he thought I was going to take the opposite side, and that would have been more difficult to argue against. He was unflappable, and he was brilliant.

Dick was an avid tennis player and golfer. He was always in good cheer. It was uncommon for him to get a ‘4’ on a hole, but when he did, he would roar with laughter. When I visited him and his wife Helen last summer, he was, as always, working on crossword and jigsaw puzzles, sippin’ on a Manhattan, and asking me, “What’s new?” He never lost his passion for crystallography, or life. 

Larry Henling: One day in 1990, Dick walked into the X-ray lab when we were collecting data on our Enraf-Nonius CAD4. He looked briefly at the reflection listing on the line printer and, without knowing anything about the compound, said, “You have a heavy atom at (x, y, z).” After the structure was refined, I ran a Patterson map for comparison. Dick's mentally calculated numbers were actually closer to the final refined coordinates than those obtained from the Patterson map. So the published paper states, “Ta atomic coordinates found by visual inspection of diffractometer intensity listing.” [CheckCIF would not approve.] But more remarkable than Dick's mastery of crystallography and computation was his generosity in sharing his knowledge. His office door was always open and he was always willing to stop and answer questions, even if he had answered the same question a few weeks or months ago. You could give him a problem structure; within a day or two you would find on your chair a typewritten sheet of paper with one or two paragraphs explaining the proper approach. Despite all the assistance he gave us, Dick seemed a bit apologetic about troubling us whenever he asked for help, usually for just a minor computer issue.

Dick was of course known for correcting erroneous space-group assignments. In fact, in December the last structure he was working on was ... well, perhaps we can let that slide. While Dick felt strongly that crystallography ought to be done correctly, he also viewed these as puzzles to be solved. And Dick enjoyed puzzles and word play of all sorts, including cryptic/diagramless crosswords. He was always amused that, “unit-cell constants were determined to be.” One newspaper headline taped to his wall was about a high-school basketball player who had “scored 69,126 points in two games.”

Coffee time was a fixture in Dick's life. It was a time to discuss crystallography, science, or anything with local and visiting scientists. As a new research fellow at Caltech in the mid-1950's, Dick would sometimes have morning meetings with Robert Corey. But if they were not finished at coffee time, Dick would just get up and walk out. In recent years, without Verner Schomaker and Bill Schaefer, the scientific rigor of coffee time decreased markedly. But we did do the Daily Jumble puzzle, which is unscrambling five- or six-letter words. Dick would do these instantaneously; if not, the wrong letters were written on the board. He could also come up with the words if you just read him all but one of the letters. The only way we could slow him down was to have someone write some of the letters in Macedonian.

Dick was known at his favorite restaurants, where he usually ate the same thing. When we went to lunch at Primo's, a local sandwich shop, he was greeted with, “Mr. Meatball-ie,” and without ordering would be served his meatball sandwich and iced tea.

Dick did not give much attention to things he didn't care about. When he had to get a new car (the old one caught fire in his driveway), we asked him what he bought. “A red car,” was his reply. After a few years of teasing, he did eventually learn what it was. But after leaving one evening, he returned to the office to call his wife Helen as he had locked his keys in his car. This was odd since he usually left the car unlocked with the key under the floor mat. I went out and retrieved his key from his car. The locked vehicle parked next to his, however, was a car and it was red.

Dick said that in college he majored in tennis and golf (he was also a good third baseman on the Caltech team). He enjoyed playing and watching both sports.

Dick's description of Lindo Patterson applies equally well to himself: “[his] high standing among crystallographers, particularly in America, goes far beyond his published scientific work. It derives to a very great extent from his character and personality, for he possessed the rare combination of a keen mind, a lively humor, and a gentle disposition.”

Editor’s Note: The Winter 2012 issue of ACA RefleXions contains a Living History of Dick Marsh that incorporates a memoir and comments from B.C. Wang and John Rose.