At the suggestion of Wallace J. Eckert of Columbia University, physical chemist Linus Pauling and associates at Caltech used IBM electric punch card tabulating equipment to speed up the Fourier calculations in crystal structure analysis in their researches. The first paper resulting from these applications was David E. Hughes, "The Crystal Structure of Melamine," J. Amer. Chem. Soc. 63 (1941) 1737-52.
Prior to this Leslie J. Comrie had attempted to introduce IBM Hollerith electric punched card tabulating to speed up Fourier calculations in crystal structure analysis in England, but the method did not gain acceptance.
Applications of IBM equipment in crystallographic research continued at Caltech but the method was not published until 1946: Shaffer, Philip. A., Jr.; Schomaker, Verner; and Pauling, Linus The use of punched cards in molecular structure determinations. I. Crystal structure calculations [II. Electron diffraction calculations], Journal of Chemical Physics 14 (1946) 648–658, 659–664. The offprint version of the first paper contained a 10-page supplement with 5 full-age diagrams.
"Shaffer, Schomaker, and Pauling developed methods of carrying out Fourier calculations on IBM punched-card machines, using a Type 11 electric keypunch, a Type 80 electric sorting machine, and a Type 405 alphabetic direct-subtraction tabulating machine. This paper cites work as early as 1941 performed on the structure of various less-complex organic crystals using electric tabulation methods.
"The supplement to Part I of this paper, which was included only in the offprint version, provided additional information on card design, plugboard wiring and operating procedures. 'The time factor is in all cases greatly in favor of the punched-card method relative to summation procedures used in the past. Fourier projections which by the Beevers-Lipson method required several days of calculation can now be made in 5 to 7 hours. At the same time the density of calculated points is much greater and the accuracy of the computation is assured. The machine steps in the least-squares calculations require only a few hours, as compared to one or two days with use of an adding machine, and again the accuracy of the work is assured. With the use of parameter cards and the structure-factor files the calculation of structure factors can be accomplished in about one-eighth of the time previously required.' (p. 658). Most of the detail in the technique of data processing, including information on card design, plugboard wiring, and operating procedures appears in the supplement" (Hook & Norman, Origins of Cyberspace  no. 879).
Cranswick, "Busting out of crystallography’s Sisyphean prison: from pencil and paper to structure solving at the press of a button: past, present and future of crystallographic software development, maintenance and distribution," Acta Crystallographica Section A Foundations of Crystallography A64 (2008) 65-87. (Accessed 04-20-2010).