Articles

After the Double Helix

Rosalind Franklin's Research on Tobacco mosaic virus
Angela N. H. Creager* and Gregory J. Morgan**  

*Department of History and Program in History of Science, 136 Dickinson Hall, Princeton University, Princeton, New Jersey 08544.

**Department of Philosophy, Spring Hill College, 4000 Dauphin Street, Mobile, Alabama 36608.

ABSTRACT

Rosalind Franklin is best known for her informative X-ray diffraction patterns of DNA that provided vital clues for James Watson and Francis Crick's double-stranded helical model. Her scientific career did not end when she left the DNA work at King's College, however. In 1953 Franklin moved to J. D. Bernal's crystallography laboratory at Birkbeck College, where she shifted her focus to the three-dimensional structure of viruses, obtaining diffraction patterns of Tobacco mosaic virus (TMV) of unprecedented detail and clarity. During the next five years, while making significant headway on the structural determination of TMV, Franklin maintained an active correspondence with both Watson and Crick, who were also studying aspects of virus structure. Developments in TMV research during the 1950s illustrate the connections in the emerging field of molecular biology between structural studies of nucleic acids and of proteins and viruses. They also reveal how the protagonists of the “race for the double helix” continued to interact personally and professionally during the years when Watson and Crick's model for the double-helical structure of DNA was debated and confirmed.

FOR MOLECULAR BIOLOGY, the 1950s was in many ways the decade of the helix: the alpha helix of proteins, the double helix of DNA, and the helical nature of Tobacco mosaic virus (TMV) were all major discoveries.1 This essay examines the interplay between structural studies of DNA and of viruses by focusing on the research of the crystallographer Rosalind Franklin. Franklin is now remembered principally for her X-ray diffraction patterns of DNA, which provided data that—still unpublished and examined without her consent—informed James Watson and Francis Crick's double-helical model.2 Neither the degree to which Watson and Crick had access to her results nor the possibility that her data contributed to, rather than corroborated, their model came to public light until the 1968 publication of Watson's The Double Helix. This confessional account, with its unflattering and inaccurate portrayal of Franklin, was published ten years after her untimely death from ovarian cancer. While Watson's book has become a classic, so, too, has Franklin become a martyr figure, a woman scientist denied credit for her key contribution to our understanding of DNA.3 Indeed, developments in the last four decades—as DNA came to occupy center stage in the life sciences and the women's movement drew attention to the pervasive sexism in elite fields, including science—have intensified this sentiment. The publication of Brenda Maddox's enlightening biography of Franklin just prior to the fiftieth anniversary of Watson and Crick's famous paper kept the vexed issue of scientific credit for the double helix at the center of media accounts of the 2003 commemorations.4

The relentless focus on the dramatic elements of the double-helix story, however, yields an incomplete and misleading understanding of Franklin's overall career and contributions—as well as of molecular biology at that time. By the late 1950s, Franklin had achieved a strong international reputation as a scientist, in large part because of her impressive contributions to structural virology. At Birkbeck College, in the laboratory of John Desmond Bernal, she used many of the same techniques she had employed with DNA to produce the finest diffraction patterns of TMV available.5 Though she abandoned her experimental work on DNA when she left King's College, London, in 1953, she remained in the same small circle of biophysicists: Watson, Crick, and Maurice Wilkins were all engaged in structural studies of viruses. Watson determined that TMV was helical in structure; Franklin and her group confirmed this insight but corrected his model by showing that TMV had forty-nine subunits per three turns of the helix. Franklin and Donald Caspar determined that the RNA in TMV was not situated in the center of the rod-shaped virus, as previously thought, but at 40 Ångströms (Å) radius. Franklin was admired for her virtuosity as an experimentalist, but she also took risks to establish her leadership in the field, publishing speculative models of the virus structure.6

Franklin's research at Birkbeck was highly collaborative.7 Her efforts from 1953 to 1958 to determine the structure of TMV involved Aaron Klug, John Finch, and Kenneth C. Holmes and drew on resources from three rival groups of biochemists—led by Gerhard Schramm in Tübingen, Heinz Fraenkel-Conrat in Berkeley, and Barry Commoner in St. Louis—that were also working on the virus. Close analysis of Franklin's research during the last five years of her life reveals how relationships among the main protagonists of the double-helix story developed after the spring of 1953—and shows the practical and conceptual connections between key research objects, especially nucleic acids and viruses, in the new field of molecular biology. Her work on virus structure makes it clear that these researchers in the 1950s were putting to rest debates about the physical and chemical nature of biological materials that had persisted for two decades. Even as the project of showing that proteins, nucleic acids, and viruses have regular molecular structures gave way to new concerns about information and sequence, a few central model systems, including viruses such as TMV and proteins such as hemoglobin and insulin, remained central to biophysical research.8

The history of molecular biology in the 1950s is hardly understudied, yet a focus on Franklin's late work provides a new perspective from which to view familiar developments.9 Two aspects of our analysis merit highlighting at the outset. First, this account offers an alternative narrative to the popular double-helix story, one that emphasizes the continuing interactions between Franklin, Watson, Crick, and their collaborators as they moved on from DNA structure to new problems. In the mid-1950s, once the hereditary role of nucleic acids had been settled, molecular biologists sought to determine the functional correlation between proteins and nucleic acids (as well as to differentiate the roles of DNA and RNA with respect to protein synthesis). Accounts of this period tend to emphasize the researchers who sought to crack this “coding” problem by viewing genes as carriers of information amenable to theoretical and computational approaches.10 By contrast, Franklin was centrally involved in a network of scientists who used structural methods to investigate the nucleic acid–protein relationship; they often worked with viruses, since their nucleic acid and protein components could be studied separately and jointly. This was not the only motivation for undertaking structural studies of viruses, but the pertinence of this approach to the coding problem and to understanding protein synthesis has been largely overlooked, despite Watson's and Crick's participation in the effort. In part, this omission persists because other groupings, such as the “RNA Tie Club,” lend themselves to such colorful historical accounts.11 However beguiling such tales may be, they miss important players and contributions and tend to reproduce patterns of exclusion that operated at the time. The RNA Tie Club actually included several scientists known for their work on biological structures—Watson, Linus Pauling, and Alexander Rich—but Franklin (who did not wear a necktie, after all) was never invited to join.

Second, we build on the insights of others in emphasizing the importance of transatlantic exchanges, both material and informational, among the first generation of molecular biologists.12 Surviving correspondence makes clear the challenges these researchers faced in the 1950s as they sought to navigate disciplinary boundaries, fend off competitors, and establish scientific priority—while at the same time trying to contribute sufficiently to scientific exchange networks to “earn” access to others' research materials and unpublished results.13 The increasingly international circulation of materials, knowledge, and researchers in the postwar decade did not diminish the relevance of differences between local cultures of science.14 British researchers, who were accustomed to accommodating institutional prerogatives in selecting their research problems (one thinks of how DNA structure had been viewed as belonging to King's College), were confronted by entrepreneurial American competitors and the massive research funds made available by the U.S. federal government.15 Franklin lacked a permanent institutional position in the United Kingdom, with its attendant supports and constraints. Yet she showed herself remarkably adept in maneuvering within the interdisciplinary and international arena and at managing relations with rivals, collaborators, and allies (often the same people in different roles over time) in order to obtain the materials and support she needed to succeed.

TMV: ONE MOLECULE OR MANY?

 

When Franklin began research on TMV in Bernal's laboratory in 1953, she was tackling an experimental subject with a long pedigree. TMV was the first virus discovered—by Dmitri Ivanovskii in 1892 and, independently, by Martinus Beijerinck in 1898; the latter made the additional claim, on the basis of its unusual traits, that it was not a bacterium at all but a contagium vivum fluidum.16 Thereafter the term “virus” was used in the literature to designate a filterable, possibly nonbacterial, submicroscopic pathogen. In 1935 Wendell Stanley announced that he had crystallized TMV as a protein, and the suggestion that this pathogen might be as chemically simple as table salt caught the imagination of scientists and journalists alike. The oversimplifications of Stanley's findings were soon made clear, thanks to the efforts of two British scientists, the plant pathologist Frederick C. Bawden and the biochemist Norman Wingate Pirie. They repeated Stanley's procedure and found that he had missed the presence of nucleic acid. They also contended, on the basis of their collaboration with the British crystallographer J. D. Bernal and the American crystallographer Isidore Fankuchen, that Stanley's “crystals” exhibited regularity only in two dimensions and so were not true three-dimensional crystals. They were more accurately described as liquid crystalline substances, or paracrystals.17

In the late 1930s Bernal and Fankuchen went on to investigate the X-ray diffraction patterns of TMV, in which the particles were oriented either in dried specimens or in liquid-crystalline gels. They found that TMV did not exhibit X-ray patterns characteristic of “an indefinite repetition of identical units in three-dimensional space” but, instead, showed regularities within the virus structure. They inferred that the virus was made up of smaller “submolecules” of dimension 22 Å × 20 Å × 20 Å, each of which was composed of two identical units. They published a full account of their findings on TMV structure in the Journal of General Physiology in 1941 as a three-part paper reporting on diffraction patterns of several plant viruses.18

Bernal and Fankuchen's pioneering X-ray diffraction analysis of the structure of TMV did not elicit much follow-up in the 1940s, no doubt on account of the difficulty of their fifty-five-page paper and the disruption of World War II. But there was also disagreement over whether TMV, once chemically isolated, retained the same structure as the infectious virus in vivo. Stanley interpreted the sedimentation behavior of purified TMV in the ultracentrifuge as evidence that the virus was a single huge molecule (estimates of its molecular weight ran as high as 50 million daltons).19 Colloidal chemists, who believed all proteins to be aggregates of smaller units, contested Stanley's claim, as did Bawden and Pirie—not because they held a colloidal view of biological material, but because they believed that the ultracentrifuged particles were artifactual aggregates unlike the native TMV that infects plants.20 The first electron micrographs of TMV, published in 1939, seemed to support Stanley's claim, showing the virus particles to be rods approximately 3,000 Å long and twenty times as long as they were wide. This result helped explain why the virus precipitated as needle-shaped paracrystals. The decline of the colloidal view of proteins also bolstered a conception of the huge virus particle as a true macromolecule rather than an aggregate. In this context, chemists such as Stanley were indifferent to evidence for viral subunits.21

Bernal and Fankuchen's finding that the virus particle was composed of smaller, regular subunits received biochemical confirmation in 1943. Gerhard Schramm, in Tübingen, found that placing TMV in alkaline solution produced homogeneous protein fragments that were roughly the same size (but not the same shape) as Bernal and Fankuchen's repeated units. Strikingly, by lowering the pH again he was able to induce these fragments (later called “A” protein) to reassemble into rods resembling normal TMV. But, as Watson later explained, Americans had little interest—or faith—in this result.22 Even so, Watson himself was one among several scientists who set out to determine the precise structure of TMV, in hope of understanding how these complex molecules infected and replicated themselves in plants.

TMV, NUCLEIC ACIDS, AND HELICES

 

When Franklin joined John Randall's large biophysics group at King's College in the spring of 1951, Maurice Wilkins was already working on DNA.23 But Wilkins was also investigating the structure of TMV, at the instigation of Gerald Oster, who had arrived from Stanley's Virus Lab in Berkeley.24 Using a polarizing microscope to probe the structure of crystalline TMV detectable in inclusion bodies of tobacco leaf-hair cells, Wilkins, Alexander R. Stokes, William E. Seeds, and Oster found a banded structure consistent with side-by-side layers of virus rods. While they could not determine the precise length of the virus particles, their observations—which were based on structures in vivo—were consistent with representations from electron microcopy and ultracentrifugation of purified particles. Thus their study undermined Bawden and Pirie's assertion that the particles of infectious TMV in plants were smaller than the rods observed in biophysical studies. They also claimed, contra Bawden and Pirie's detection of viral RNA, that their work “makes it appear even more likely than before that the crystals are pure virus protein.”25 Stokes suggested that the TMV particles might be arranged helically, but light microscopy offered little help in revealing molecular structure within individual particles.26

Watson was also working on TMV structure after his arrival in Cambridge in the fall of 1951.27 He recognized that helical diffraction theory, as set out by Crick, William Cochran, and Vladimir Vand in 1952, could explain the strange diffraction spots from TMV that had puzzled Bernal and Fankuchen, foiling their attempts to assign a unit cell in agreement with the intermolecular measurements. As Robert Olby put it, “under Crick's excellent tutorship Watson learnt a lot of crystallography.” Watson's insight was reinforced when he realized that TMV could be thought of as a small helical crystal that grows by adding material to “cozy corners”—much as suggested by F. Charles Frank's theory of crystal growth.28

Watson's reasoning that the TMV rods were helices represented a breakthrough. He needed new diffraction patterns, though, to improve on Bernal and Fankuchen's structure findings by determining the number of units per helical turn. Preparing dry paracrystalline specimens of TMV from Roy Markham at the Molteno Institute at Cambridge, and with the help of Hugh Huxley, Watson spent several months collecting photographs of TMV. As he has recounted, “The way to reveal a helix was to tilt the oriented TMV sample at several angles to the X-ray beam.”29 Using this technique, Watson obtained pictures in June 1952 with what appeared to be the critical reflection. His data supported the existence of a helical structure and confirmed Bernal and Fankuchen's general claim that the virus was composed of many equivalent subunits (though his own photographs lacked the large number of distinct reflections they had obtained). Of course, his first-hand familiarity with helical diffraction theory also enabled him to perceive the structural information contained in Franklin's “Photograph 51” of the B-form of DNA when Wilkins showed it to him in early 1953.30

Watson specified the parameters for the TMV helix: a repeat of three turns in 68 Å, with 3n + 1 protein subunits per helical repeat. The strong diffraction he saw arced on the meridian of the thirty-first layer line led him to assign a value of 10 to n, for thirty-one subunits per three turns of the TMV helix. This interpretation implied that each subunit would have a molecular weight of 35,000 daltons and that a virus particle would contain about twelve hundred such subunits.31 By analogy with Turnip yellow mosaic virus (TYMV) and T2 bacteriophage, viruses whose nucleic acid was thought to be located in the center of a protein shell, Watson suggested that the TMV ribonucleic acid formed a 35 Å–diameter core in the center of the protein helix, although his own diffraction data did not resolve its location.

Franklin moved to Birkbeck College and began her structural determination of TMV in the spring of 1953. In addition to Bernal and Fankuchen's 1941 publication, she had access to Watson's manuscript on the evidence for a helical arrangement of subunits. Watson submitted the paper to Biochimica et Biophysica Acta on 16 April, the week before the famous Watson and Crick, Wilkins, and Franklin and Gosling DNA papers were published in Nature.32 Whether Franklin's new choice of subject was Bernal's or her own, it is hard not to be struck by the fact that she was setting herself up in competition with Watson at this juncture. Franklin's first task at Birkbeck was to install an up-to-date diffraction apparatus and camera. She spent the first several months in Bernal's laboratory (mid-March to November 1953) familiarizing herself with the literature on plant viruses while waiting for pieces of her equipment to arrive. During this time she also continued analyzing the X-ray diffraction patterns of DNA and their Patterson functions.33 As others (most recently Maddox) have noted, when Franklin left King's College Randall ordered her to leave the DNA work behind. Franklin ignored Randall's ban, however, and continued working with Raymond Gosling on his doctoral thesis and their joint publications.34

Franklin's work on TMV got off to a slow start.35 Some of the challenges were apparent from Bernal and Fankuchen's paper. Fiber-diffraction diagrams obtained from TMV were more complicated than those of DNA, and there was no simple way to index the reflections. While Franklin assembled her apparatus, Bernal asked Randall if she might borrow a camera she had used while at King's, only to be put off.36 Not all of the new obstacles were technical; Franklin was critical of how the “narrow-mindedness” of some of Bernal's left-wing associates interfered with a productive laboratory environment. On the other hand, Bernal's ideological commitment to equality between the sexes benefited women scientists, including Franklin. She found working at Birkbeck College to be a marked improvement over working at King's—“as it couldn't fail to be,” she remarked in a letter to Anne and David Sayre.37 Her professional circumstances continued to improve in early 1954. Franklin gained an important collaborator in Aaron Klug, who had arrived at Birkbeck in December 1953 to work on the structure of ribonuclease with Harry Carlisle but decided to switch to virus research after he met Franklin on the stairs of the Torrington Square house-turned-laboratory and was excited by her beautiful diffraction patterns of TMV.38

During the same year that Franklin shifted her attention from DNA to TMV, Watson also changed course. Moving to Caltech in September 1953, initially working with Max Delbrück and still supported by his fellowship from the National Foundation for Infantile Paralysis, Watson shifted from working on DNA and TMV to pursuing the structure of RNA using X-ray diffraction.39 Alex Rich, who was finishing a Ph.D. with Linus Pauling at Caltech, became his collaborator in this venture. Watson's hope was that determining the structure of RNA would clarify its role in protein synthesis, one of the major questions of the day, just as determining the structure of DNA had suggested an answer to the question of how the genetic material was replicated.40

In March 1954 Watson, Leslie Orgel, and George Gamow launched their whimsical RNA Tie Club—which would be composed of four honorary members and twenty regular members, one for each amino acid. By definition, it was an all-male enclave. Watson wrote Delbrück (who was in Germany for the spring) informing him of this “very secret society”—Delbrück was assigned tryptophan—and updating him on the progress he was making on RNA structure. He and Rich found that changing the humidity of the preparation altered the RNA fiber's length—in a way that correlated with changes in the X-ray pattern.41 They extended this investigation to RNA samples from a variety of sources: TMV, TYMV, calf liver, calf thymus, and yeast. They found that all of these RNAs produced very similar X-ray diffraction patterns and showed the humidity-induced transformation. While the biological meaning of this observation remained unclear, it is striking how closely Watson and Rich followed Wilkins's and Franklin's approach (so successful with DNA) of looking for a reversible humidity-induced change in nucleic acid structure.42

Whether RNA was in the same double-helical configuration as DNA proved hard to pin down. In March, Watson wrote Delbrück that the X-ray patterns suggested that RNA might possess a helical structure like that of DNA. This result disappointed Watson, who found the emerging picture “queer and paradoxical,” since RNA lacked the consistent complementary base ratios of DNA.43 However, in a coathored paper published in Nature that May, Watson and Rich emphasized that the X-ray patterns of RNA from various sources differed from those of DNA. In June, Watson concluded discouragingly, “Our RNA work is at a standstill. We need a cute idea or a much better X-ray photograph and neither possibility seems in the air.” Rich left Caltech for a post at the National Institutes of Health, commencing X-ray diffraction studies of TMV there.44 The world of crystallographers studying biological materials was a small one, with the same specialists engaged in structural studies of nucleic acids and viruses.

FRANKLIN: NEW DIFFRACTION PATTERNS AND COLLABORATORS

 

Franklin obtained her first X-ray diffraction photographs of TMV at the end of 1953, using a virus sample from Roy Markham. As she had done with DNA, she examined how preparing virus specimens with varying amounts of water influenced the reflections. A wet gel preparation of TMV yielded exquisite X-ray patterns: by the spring of 1954, she had obtained diagrams with more than three hundred distinct maxima, and she was calculating the cylindrical Patterson function to see if her results confirmed Watson's postulation of a helical arrangement.45 That spring she was also invited—on the basis of her earlier work on coal structure—to participate in a late summer Gordon Conference on Coal in the United States. In order to raise the necessary travel money, she contacted several scientists about giving lectures on her work on coal structure or TMV.46 Her October itinerary included lectures at Caltech and at the University of California, Berkeley, where she visited Stanley's Virus Laboratory. Both at Woods Hole early in her trip and then in California, Franklin met with Watson and brought him up to date on her new work. Following her visit to Caltech, Watson wrote Crick that he and Orgel had spent time “trying to make more sense out of TMV (promoted by Rosie's visit—very amiable!).”47 While in Berkeley, she arranged with Virus Lab biochemists C. Arthur Knight and Heinz Fraenkel-Conrat to obtain samples of their purified TMV for her continuing X-ray diffraction studies. The Berkeley preparations included a heavy-metal derivative in which a mercury atom bound each viral protein subunit at its single cysteine residue.48

By the early 1950s, an intense rivalry had emerged between the virologists in the Berkeley Virus Lab and those in Gerhard Schramm's group at the Max Planck Institute for Virus Research in Tübingen.49 Franklin navigated the fractious community of TMV biochemists with remarkable facility.50 In addition to the heavy-metal derivative of TMV she obtained from Fraenkel-Conrat as a result of her American trip, she acquired a sample of Schramm's “A” protein—presumably the TMV protein subunit—in order to determine the structure of repolymerized nucleic acid–free TMV-like helices. Barry Commoner at Washington University in St. Louis was examining a material very like the “A” protein, which he called “B8” protein, a small RNA-free protein that also polymerized into TMV-resembling rods. Franklin was interested in Commoner's findings and visited him while in the United States. They too began to collaborate, publishing a joint paper in 1955 that compared X-ray diffraction patterns of “B8” protein and TMV and argued for a structural relationship between the two proteins.51 Not all of Franklin's joint ventures were successful, however. Pirie, Stanley's early critic, proved to be an unreliable collaborator. After providing Franklin with purified TMV early in her research, he was so unhappy with her support for the identification of the virus with the 3,000 Å rods that he refused her further material. As Maddox has pointed out, Pirie became an influential adversary of Franklin's because of his influence in British governmental funding bodies.52

Late in 1954, Franklin drafted a paper on the structural features of TMV she had discerned from her X-ray diagram and calculations. After circulating the paper among several colleagues, including Watson, Crick, Max Perutz, and Pirie, she published her findings and structural model in February 1955 (see Figure 1 and Frontispiece). Her data confirmed “most satisfactorily” Watson's conclusion that the subunits were arranged in a helical structure.53 Analysis of the diffraction data with Klug indicated that there were grooves along the external surface of the TMV helix. The surface area afforded by these extensive grooves helped explain some of the biochemical properties, particularly “the surprising variety and extent of chemical modifications which it is possible to make in tobacco mosaic virus without breaking up the particle and, in some cases, without destroying its infectivity.” Franklin also calculated the distribution of density along the radius of the TMV particle—to determine how the atoms were distributed from the center out to the edge of the cylinder. Her initial calculations (based on an incorrect phasing of the data) revealed a peak of high density at 55 Å from the central axis, and she speculated that this was the nucleic acid.54 This meant that the RNA might not be located at the center of the virus—as suggested by the electron micrographs, in which it appeared like the wick of a candle—but was instead closely associated with the protein subunits.55 She acknowledged an alternative explanation: that the RNA was in the center after all, but in a hydrated and therefore less dense state.56

Figure 1. Franklin's X-ray diffraction pattern of TMV. Klug thought her diagrams were “beautiful” because they showed many more distinct intensity maxima (spots) than other researchers had obtained, which indicated both the quality of her sample preparation and her superior diffraction techniques. These maxima appear on horizontal layer lines that are perpendicular to the fiber axis. Reprinted with permission from Rosalind E. Franklin, “Structure of Tobacco Mosaic Virus,” Nature, 2005, 175:379–381, on p. 379. Copyright 1955 Macmillan Magazines Limited.

Open New Window

Using her superior X-ray pictures, Franklin was able to evaluate Watson's estimate of thirty-one subunits per three turns of the TMV helix. She found that Watson's interpretation of a meridional reflection on the thirty-first layer line was mistaken; her results led her to calculate a value for n of 12, yielding thirty-seven subunits per three turns of the helix.57 (She and members of her group would later correct this estimate to forty-nine subunits per three turns.) The unit molecular weight for the virus subunits would then be 29,000 daltons. She argued that these units were subdivided into two equivalent or near-equivalent subunits and that these smaller units in turn correlated with the protein subunits detected through chemical methods. Making use of Watson's suggestion that the pitch of the helix was sufficient to allow for a double layer of virus proteins (if they were α-helical) on each turn, Franklin offered a schematic representation of the protein subunit arrangement in TMV, as seen in Figure 2. Her description of this representation, furthermore, acknowledged helpful conversations with Crick, who himself advised against publishing such a speculative drawing.58

Figure 2. Franklin's schematic diagram of the three-dimensional structure of TMV, showing her proposed arrangement of the protein components: (a) offers a view of a short segment of the virus particle, showing subunits on six turns of the helix (the hatched lines indicate a subdivision of each subunit into two near-equivalent parts); (b) depicts the transverse section of the virus rod, showing twelve triangular subunits in one turn of the helix. Reprinted with permission from Rosalind E. Franklin, “Structure of Tobacco Mosaic Virus,” Nature, 2005, 175:379–381, on p. 381. Copyright 1955 Macmillan Magazines Limited.

Open New Window

The correspondence among Franklin, Watson, and Crick on her early work on TMV was civil, even friendly. Franklin was, it should be added, keenly aware of her competitors on virus structure, and perhaps it was the realization during the fall of 1954 that three other people were pursuing X-ray diffraction studies of TMV that spurred her to write up and publish her results after returning from the United States. It appears that her experience with DNA structure sensitized Franklin to the importance of receiving acknowledgment for her work and led her to publish more speculative ideas. For his part, Watson's response to her draft seems to have been aimed at blunting the degree to which she critiqued his earlier interpretation. As he wrote, “I was not so emphatic on the location of the RNA—I believe I was quite cautious with ‘ifs.’” Franklin annotated the letter, noting that she “only said ‘suggested’” when citing his paper. Two months later, however, he was not so ready to concede the point about RNA placement; he wrote Franklin about new electron microscopic evidence from Berkeley that indicated that “the RNA forms a central core” of the virus.59

In November 1954 Franklin received a long letter from another of her competitors on TMV structure: the biophysicist Donald Caspar. Caspar conducted his Ph.D. research on TMV at Yale University before going to Caltech as a postdoctoral fellow in December 1954.60 In contrast to Franklin's photographic apparatus, Caspar used a Geiger counter to collect fiber-diffraction data from both regular TMV and TMV bound with lead acetate. His goal, like Franklin's, was to determine the distribution of density along the radius of the cylindrical TMV particle. The crucial parameters were the phases of the diffracted X-rays; these Caspar could not measure directly. He gleaned some phase information from the shape of the intensity curves and the rest through comparison with the lead derivatives of TMV—drawing inspiration from the successes of Max Perutz, who had used the same method (heavy-atom isomorphous replacement) to investigate the structure of hemoglobin.61 This new technique enabled crystallographers to estimate which phase assignment is correct by measuring how the binding of a heavy atom perturbs the pattern of reflection intensities. After arriving at Caltech, where he collaborated with Watson, Caspar continued analyzing the diffraction data from his lead derivative of TMV, calculating a radial density profile of the helical virus. His results indicated peaks of density at 24 Å and 40 Å from the central axis of the TMV cylinder. Moreover, there was no significant density at the center of the cylinder—indicating that the TMV rods were hollow. Thus his results showed, independently of Franklin's, that the viral RNA did not form an axial core, as electron microscopic studies had suggested.62 Watson and Caspar initially took the innermost peak (24 Å) to be the RNA.

Caspar's intended postdoctoral project was to use Pauling's X-ray diffraction facilities to study the structure of spherical plant viruses, especially Southern bean mosaic virus (SBMV). As it turned out, Caspar could not delineate anything more than the diameter of this virus. He made even less progress analyzing samples of Tomato bushy stunt virus (BSV), another spherical virus obtained from Knight of Stanley's Virus Lab in Berkeley, though he did manage to grow some crystals.63 As Watson's interests were focused on RNA structure, Caspar and Watson considered the structure of TMV RNA and its relationship to the helical protein shell.64 At the time, the best estimates of the molecular weight of TMV RNA suggested that there must be more than one piece of RNA in each particle.

Watson and Caspar proposed a model of TMV with ten to twelve chains of TMV RNA bound by chemical bonds between the phosphates into pairs of chains that followed a helical grid on the inside of the particle (see Figure 3). According to this early model, as in later models, the length of the RNA determines the length of the TMV particle. Watson was fully aware of the aesthetic consequences of his model. He wrote to Franklin describing it: “The main thing in favor of the P-O-P [pyrophosphate bond] model is that it is very very pretty stereochemically. But does nature always like to be pretty?” Franklin was skeptical of their model: she replied that she thought the RNA might as well be a disordered core as far as the X-ray data were concerned. In the end, Caspar and Watson's multistranded model was never published. Nonetheless, they began to use a Spinco analytical ultracentrifuge to measure the size of TMV RNA.65 If they could determine the size of TMV RNA, they would be closer to knowing how many strands of RNA existed within each TMV, a crucial parameter in constructing an accurate model.

Figure 3. Unpublished schematic diagram of TMV by Donald L. D. Caspar and James D. Watson, depicting egg-shaped viral protein subunits arrayed helically around twelve strands of RNA that are themselves packed together to form a high-pitch helix on the inside of the protein shell. Making use of his artistic talent, Caspar drew many of his own diagrams, including this one. Image courtesy of Donald L. D. Caspar.

Open New Window

In the meantime, Caspar and Franklin corresponded about their strikingly similar results. Caspar was able to correct mistakes in Franklin's initial assignment of phase signs, which she discovered were due to an oversimplification of the effect of the groove. Franklin, in turn, was able to confirm with her independent data that Caspar had correctly assigned his phase signs.66 In April 1955 Caspar wrote Franklin that he hoped to come to England late that summer for a few months to work with her, if he could get the funds. She responded that he would be welcome, “although I am afraid you will find we have very little to offer in the way of facilities and space.”67

The reason Franklin's rather small laboratory was so crowded was that her virus research group doubled in size that year. First she hired John Finch as a research assistant. Then, in late summer 1955, the Ph.D. student Kenneth Holmes joined her group. Franklin put Holmes and Finch on parallel Ph.D. tracks. (Because Franklin did not have a faculty position at the University of London, Bernal served as their nominal advisor.) Finch began to investigate the effect of relative humidity of TMV gels, work similar to Franklin's on DNA that established the A and B forms, but he switched to work on the spherical virus TYMV with Klug after Holmes arrived. This would be the second virus that Franklin's group would study in detail.68 Holmes continued to work with TMV and would eventually write his dissertation on the comparison of diffraction patterns from different strains of the virus. Franklin was building up a full research group at Birkbeck, but to do this without a regular faculty appointment and on soft money was risky. A grant from the Agricultural Research Council (ARC) supported the laboratory from 1955 through the end of 1957, but the ARC refused to pay Franklin the full salary requested by the college. More worrisome were the obstacles she encountered in trying to use her funding to keep Klug in her group after his Nuffield Foundation fellowship ended.69

By late 1955 Franklin had managed to use Schramm's “A” protein to obtain diffraction patterns of viral-like rods lacking any nucleic acid. After a number of attempts, she managed to nurse the specimens into an oriented gel without the proteins disaggregating.70 The resulting X-ray patterns showed that repolymerized “A” protein was much closer in structure to native TMV than rods formed from Commoner's “B8” protein. In fact, it appeared to be just like the native TMV particle, but without the RNA.

In the late summer of 1955, Caspar and Watson traveled separately to Europe. Watson had already accepted an assistant professorship at Harvard, but a National Science Foundation fellowship gave him a year at the Cavendish before his move to Massachusetts. Crick wrote Franklin ahead of Watson's arrival to clarify areas of overlapping interest. He explained that Watson “is interested in TMV from the point of view of RNA structure and in particular is wondering whether the helical symmetries of the two parts (protein and RNA) may be related.” This line of research followed up his collaboration with Caspar. Watson also wished to commence work on Potato virus X (PVX), another helical virus that he had taken some diffraction patterns of three years earlier. In order to secure material, he had Crick ask Roy Markham whether the Molteno Institute would grow some for him.71 Markham, in turn, told Crick that Franklin had already requested a preparation of this particular virus. Since Watson was coming to his laboratory, Crick put himself in the position of negotiating between Watson and Franklin on the matter of PVX; so as to avoid duplication of effort, he asked Franklin about her plans regarding this virus.

Franklin replied immediately: “I have always intended to work on it as soon as I can get hold of some. I have asked both Roy and Pirie for it, but got no answer from either. At your suggestion I shall now write again to Roy.” She was clearly not going to cede PVX to Watson. In the fall he pursued the project anyway: “In the mornings and many evenings,” Watson recounts, “I was in the Cavendish taking X-ray photographs of RNA-containing potato virus X. It was my response to an overnight visit from Rosalind Franklin, who stayed with the Cricks. Listening to her treat Don and me as insignificant players in tobacco mosaic virus (TMV) research, I felt the need for another plant virus to call my own.” If Watson saw her attitude as disparaging, this is not the view Franklin herself conveyed in correspondence. As she wrote to Paul Kaesberg at the University of Wisconsin in a letter asking for a sample of Pea streak virus, “Jim Watson (of the DNA model) is back in Cambridge, and is also interested in these things, and between us we want to look at as many viruses as possible.”72 On the one hand, Franklin vigorously protected her own research interests regarding precious virus samples; on the other, she represented herself and Watson as cooperating in their efforts to investigate the full range of plant viruses.

Caspar met Rosalind Franklin for the first time on 12 September 1955.73 He began to work on BSV at Cambridge, and in October he obtained evidence that this virus possessed five-fold symmetry. Caspar also wanted to work on TYMV, another spherical virus. In his search for virus preparations, Caspar went to Birkbeck to rummage in the refrigerator of Bernal's assistant Harry Carlisle, where he found BSV and TYMV crystals. However, Franklin and her group were also interested in using these crystals, and she insisted that the TYMV preparation remain at Birkbeck.74 In his autobiography, Watson describes Caspar as angry that she would “put herself in competition with him.” Watson claims he served as the mediator, attempting to persuade Franklin “of the unfairness of her climbing up Don's back.” A letter Watson wrote describing the incident suggests that the “Rosy” persona later made familiar in The Double Helix was already taking shape in his perceptions: “Rosy was in Cambridge yesterday. As usual I felt exhausted by the time she departed. Don was still mad at her for her low blow with the crystals so it was left for me to save the situation with charm and diplomacy.”75 Caspar's memory of the crystals incident is strikingly different; he recalls being in a bad mood upon his return that day from Birkbeck, more because he accidentally burned a brand-new suit jacket on a Bunsen burner in Franklin's laboratory while they were talking than because of her appropriation of the crystals. He recalls an amicable division of labor, such that Franklin and Klug would work on the scarce TYMV crystals and he would continue his work on BSV.76 These negotiations over materials, territory, and credit provide evidence of a moral economy at work—albeit not without friction—among virus crystallographers.77

The scientific interactions that developed between Franklin and Caspar over TMV became congenial and highly productive.78 By comparing the radial density functions of her repolymerized “A” protein with the native TMV density functions he had obtained, they were able to conclude that the TMV RNA was neither in the center nor 24 Å from the center (as Caspar and Watson had thought) but instead lay ∼40 Å from the center. When writing Pirie of this finding Franklin made an effort to credit the others properly: “We are working on this together with Watson and Caspar.” Franklin wanted to publish the comparative analysis but first required that Caspar publish his results. According to Caspar, his procrastination in writing up the results of his dissertation led Franklin to draft his paper herself. She had finished a “hasty version” by 10 February, mailing copies to Watson and Caspar that day with requests for rewriting. Franklin's and Caspar's papers appeared as consecutive articles in Nature on 19 May 1956.79

While in Cambridge, Watson made one last effort with Alex Rich, who was also there working with Crick for six months, to decipher the X-ray pattern of RNA.80 His ambition to work out the arrangement of RNA in TMV was sidelined by Franklin's determination of its placement within the virus protein. Her finding also suggested that the structure of viral RNA was likely entirely different from that of RNA in vitro, making the relevance of the structure of the latter less clear.81 Partly in response to the rapid progress of others (especially Franklin and Caspar) on plant virus structure, Watson and Crick resumed a theoretical project they had already begun, synthesizing current results to come up with a general theory of virus structure. Their joint paper, published alongside a communication from Caspar on BSV's five-fold symmetry, appeared in Nature on 10 March 1956. They observed that “almost all small viruses are either rods or spheres” and raised the question of why this is so. “The purpose of this article is to explain this observation by means of the following simple hypothesis: a small virus contains identical sub-units, packed together in a regular manner. It has been suggested before that viruses are constructed from sub-units; but the idea has not previously been described in precise terms or put forward as a general feature of all small viruses.” The best structural evidence for Crick and Watson's theory that every virus is composed of identical subunits came from plant viruses, both helical viruses like TMV and spherical viruses such as BSV and TYMV. Symmetry considerations alone, in their view, suggested that all viruses are constructed from structural subunits that surround the nucleic acid. Relatedly, simple construction rules constrain the number of viral structures that could be built. The elegant simplicity of this idea prompted the witticism, attributed to Crick, that “any child could make a virus.”82

In the case of TMV, both biochemical and X-ray diffraction analysis had already shown the virus to be composed of identical protein subunits arranged symmetrically in a helix. For spherical viruses, Crick and Watson emphasized Caspar's finding that BSV possesses icosahedral, or 532, symmetry, meaning that the virus particle exhibits rotational symmetry along two-fold, three-fold, and five-fold axes.83 TYMV, Crick and Watson observed, had also been found to possess cubic symmetry—that is, four three-fold rotational axes. They extrapolated that other spherical viruses did as well, for any viral assembly with cubic symmetry must be “built up by the regular aggregation of small asymmetrical building blocks.” (Caspar wrote that BSV was “built up of sixty structurally identical asymmetric units,” even as preliminary biochemical evidence indicated up to three hundred chemical subunits.)84 Crick and Watson ventured that viral ribonucleic acid might also be made up of smaller identical subunits, but they did not attempt to explain how the genetic role of RNA was related to its structural role.

The goal of determining the structure of viral RNA and its relation to protein was also of central importance to Franklin and her group. TMV was the central model for this effort, for both the RNA and the protein components could be isolated, making it conceivable that the correlation of nucleic acid and protein might be determined chemically. As Franklin wrote in March 1956 in an attempt to obtain continued ARC funding: “[Our] work is concerned with what is probably the most fundamental of all questions concerning the mechanism of living processes, namely the relationship between protein and nucleic acid in the living cell. … The plant viruses consist of ribonucleic acid and protein, and provide the ideal system for the study of the in vivo structure of both ribonucleic acid and protein and of the structural relationship of the one to the other.”85

To her chagrin, her ARC funding was renewed for only one final year, despite the appeals of other virologists, including Watson.86 However, in 1957 she secured a three-year grant from the U.S. Public Health Service, at a level of £10,000 per year. As Robley Williams commented to Francis Crick, the merit of her grant hinged on the “uniqueness criterion,” satisfied by the “dearth elsewhere” of structural studies of viruses like those being conducted by Franklin and her coworkers.87

THE CIBA FOUNDATION SYMPOSIUM ON THE NATURE OF VIRUSES

 

In the three years since Watson proposed a helical structure, X-ray diffraction methods and analyses had brought the overall structure of TMV into clear view. The location of the RNA was established, the parameters of the TMV helix were known with a reasonable degree of confidence, and the number of subunits per three turns of the helix had been corrected to its current value of forty-nine using chemical and structural reasoning.88 Franklin's leadership in this area was impossible to deny, and Watson commended her contributions in a letter to Stanley:

Rosalind Franklin's x-ray analysis of TMV is becoming more and more exciting. The RNA seems now to be [at] a radius of 42 Å, not 24 Å as Caspar and I originally suspected. Also the subunit number has been revised. Using Fraenkel-Conrat's Hg++ substituted TMV, she now has good evidence that there are 49 subunits/3 turns and so there is good reason for believing that only one chemical subunit is present per crystallographic unit. If so the subunit number is ∼2150 and the TMV molecular weight around 40 million.

It is really quite fascinating how the facts are beginning to fit together.89

The time was opportune for Franklin to sum up the remarkable progress in the structural determination of TMV, and an exclusive Ciba Foundation meeting in London in March 1956 provided the occasion.

Founded in 1949, the Ciba Foundation held small, informal meetings of elite researchers from around the world in its elegant establishment at 41 Portland Place. In the view of originator Frank G. Young, an unstated goal for the March 1956 symposium on “The Biophysics and the Biochemistry of Viruses” was to “revivify” virology in England. The major centers for basic viral research were represented: Cambridge, Berkeley, Birkbeck, and Tübingen. Franklin petitioned to have Schramm invited, but the Tübingen Max Planck Institute for Virus Research was represented by Werner Schäfer instead. Maurice Wilkins was scheduled to attend, but it appears that at the last minute his place was taken by Aaron Klug.90 Of the thirty-four participants, six went on to win Nobel prizes.

The Ciba symposium on viruses proved to be a meeting of the old and the new. Among the representatives of the “new” were Crick, Watson, Franklin, Caspar, Klug, and others who were convinced that the application of new physical techniques would transform biological research and knowledge. The “old” were established virologists of an earlier generation, who were often medically trained and were accustomed to relying on immunological and biochemical techniques. The organizers of the meeting, Marinus van den Ende and Young, sought to bring these two groups together in the hope of meaningful exchange. The differing responses to the meeting's most startling news, Robley Williams's assertion that TMV nucleic acid alone was infectious, showed how much of a gap existed between the two groups of participants.91 To the surprise of Watson and Crick, some of the participants did not appreciate the gravity of Williams's finding or found it hard to accept.92 In the question session, Bawden asked for quantitative results and more experiments to control for other enhancing or inhibiting substances in the inoculum: “Unless we have answers to such questions as these, how are we to know what value to attach to the exciting statement at the end of your talk?”93 Watson later attributed the difference to the failure of older researchers to understand a key tenet of molecular biology: “They were not at home with the concept that information flows unidirectionally from nucleic acids to proteins and never backwards.”94

Delivering their talk after Williams, Franklin and her colleagues Klug and Holmes unveiled a more detailed proposal for the structure of TMV (see Figure 4).95 Unlike the earlier 1955 illustration, this model depicted the hollow core, the corrected number of protein subunits per turn, and the radial location of RNA at 40 Å. The group's results also showed the high degree of regularity in the structural repeats of the TMV protein helix. Both Watson and Crick contributed to the discussion after Franklin's paper; Watson pointed out that the X-ray diffraction evidence was just as consistent with many RNA strands as with one per TMV particle.

Figure 4. “Schematic representation of a short length of the virus particle cut in half along a plane through the particle axis, showing the helical arrangement of protein subunits (49 subunits on 3 turns of the helix), the helical groove and its accompanying helical ridge extending beyond the mean radius of the particle, and the hollow axial core.” The authors depicted the location of the RNA at 40 Å out from the particle axis, but the molecule itself was not represented. The most likely configuration, as they pointed out, was that a single RNA molecule was embedded in the protein helix along the entire length of the virus particle, but the possibility of more than one strand of RNA could not be ruled out. Image and caption from Rosalind E. Franklin, Aaron Klug, and Kenneth C. Holmes, “X-ray Diffraction Studies of the Structure and Morphology of Tobacco Mosaic Virus,” in Ciba Foundation Symposium on the Nature of Viruses, ed. G. E. W. Wolstenholme and Elaine C. P. Millar (London: Churchill, 1956), pp. 39–55, on p. 43. Reprinted by permission of Kenneth C. Holmes and Aaron Klug.

Open New Window

Crick and Watson also presented a paper at the Ciba Foundation conference, articulating general principles for the construction of spherical viruses. They built on their just-published joint paper in Nature by asking why viruses were made of identical subunits at all. Their explanation focused on the limited information that could be stored in a relatively small viral genome. The way to build a virus given this constraint, they argued, was to make a large number of identical protein subunits from the same information encoded in the viral genome and then assemble them into a structure in which each subunit occupies the same environment as any other. Two types of structures could result from such an assembly process: helical rods such as TMV and spherical structures with cubic symmetry such as BSV. Crick and Watson suggested the possibility that the “arrangement of the RNA may be practically the same in all spherical viruses … since the ways of folding a fibrous molecule so that it has cubic symmetry may be rather limited.”96 Both because of its genetic properties and because it might solve the coding problem, plant viral RNA was a key object of research. As Watson had written to Crick in February, “TMV remains the H2 of the RNA world, that is, unless PVX [Potato virus X] is pushed.”97 Just as the hydrogen molecule was an invaluable exemplar in the development of atomic theory, so the best-studied laboratory viruses still dominated research on the molecular nature of life.

EXPANDING HORIZONS: FROM SPHERICAL PLANT VIRUSES TO POLIO

 

A month later, in April 1956, many of the same biophysicists gathered in Madrid for the International Union of Crystallography symposium on “Structures on a Scale between the Atomic and Microscopic Dimensions.” A snapshot from the meeting (Figure 5) shows Franklin alongside Crick, Klug, and Caspar; afterward she traveled around southern Spain with Francis and Odile Crick, who had become good friends.98 Just a few months later, Franklin returned to the United States for a long summer trip, visiting many researchers around the country. The most useful part of her trip (if not as pleasant as her time in southern California) was her stay at the Berkeley Virus Laboratory.99 There she worked with Fraenkel-Conrat on heavy-metal derivatives of TMV. Franklin also asked Stanley for “a fresh supply of your standard TMV, as this has now become my standard preparation, and if I have to change to Cambridge or any other preparation I should have to repeat a large amount of work on the basic measurements.”100

Figure 5. Attendees at the International Union of Crystallography Symposium on “Structures on a Scale between the Atomic and Microscopic Dimensions.” From left to right: Anne Cullis, Francis Crick, Donald Caspar, Aaron Klug, Rosalind Franklin, Odile Crick, and John Kendrew. The papers from this conference were not published. In four consecutive abstracts published in the program, Caspar, Crick, and Watson considered viruses as “point crystals”; Klug considered the Fourier transforms of the cubic point groups; Franklin, Klug, Finch, and Holmes considered the structure of TMV; and Wilkins and Herbert R. Wilson considered the structure of the cell nucleus. Image courtesy of Donald L. D. Caspar.

Open New Window

Perhaps the most surprising results she saw in Berkeley were some electron micrographs taken by Russell Steere. He used a freeze-shadowing replica technique that was better able to illuminate the substructure of TYMV, the virus that her colleagues Klug and Finch were working on back in London. Franklin wrote an excited letter to Klug about the micrographs.101 If TYMV particles had 532 symmetry and consisted of shells with sixty subunits, then Franklin assumed one would see a significant fraction of rings of five “knobs.” To her surprise, she did not. Three days later she sent Klug some of the electron micrographs and a more considered opinion: “I have looked at these and others—particularly ones which show particles in a wide variety of orientations, and there seems very little doubt that the knobs lie at the vertices of a cubeoctahedron, i.e., 12 knobs. … Among hundreds of particles I have only found 2 that look remotely 5-folded and I certainly do not believe they are all like that.” She wrote a similar letter to Caspar on the same day. If TYMV resembled a cubeoctahedron—a semiregular solid formed by truncating the corners of a cube—then its symmetry differed markedly from that of BSV, the other well-studied spherical virus.102 Franklin and her coworkers were unable to reconcile this interpretation with their results.

By December Klug and Franklin had a paper on TYMV ready for publication that Francis Crick thought “reads very well indeed.” Crick suggested that they publish in Biochimica et Biophysica Acta rather than Nature because the article was long and technical in nature. Klug, Finch, and Franklin submitted a shorter and less technical paper to Nature on 11 January 1957. This paper presented the crystal structure of TYMV and suggested that, like BSV, TYMV possessed 532 symmetry. A month later the three submitted the longer paper to Biochimica et Biophysica Acta.103 In the penultimate section they addressed Steere's electron micrographs, suggesting that the high percentage of ammonium sulfate (50 percent by weight) in the samples he used might have introduced artifacts not present in the X-ray crystallography. Both electron micrographs taken by Kaesberg using a different technique (shadow casting) and their own work indicated that the symmetry was 532 or icosahedral.

At this time, biophysicists were intrigued by the morphological and biochemical similarities between RNA-containing spherical viruses and microsomes (now called ribosomes), the roundish particles of protein and RNA found in the cytoplasm and implicated in protein synthesis.104 As Watson recalled in 1962, “All RNA was thought to exist either as a viral component or be combined with protein in ribonucleoprotein particles.” Klug was keen to extend their studies of spherical viruses to microsomes—and in particular to determine if the ribonucleoprotein particles were composed of identical subunits. He alerted Bernal to the fact that the structural determination of microsomes was an area of intense competition: “we already have very good powder diagrams of TYMV taken on the focusing camera, from which it is possible to deduce that the particles are made up of subunits. We could do the same for microsomes if we had some material, and I think it would be quite a scoop if they turned out to have this sort of sub-structure.” Caspar, Crick, and Watson were all interested in pursuing this possibility.105 Franklin used her California connections to obtain material from Howard Schachman at Berkeley (yeast microsomes) and Jerome Vinograd at Caltech (pea seedling microsomes). Comparing X-ray powder diagrams of microsomes with those of plant viruses, Franklin's group found certain similarities. In particular, the structure of the RNA in microsomes and viruses was completely different from that of isolated RNA. Her group concluded that “the structure is essentially determined by a well-defined protein matrix in the interstices of which lies the RNA.” This seemed in contrast to DNA-protein assemblies, for which the protein “conforms to the structural configuration of the nucleic acid.”106 The observed structural arrangements did not resolve the coding problem. Even so, the rapid achievement of results with microsomes at Birkbeck attests to Franklin's ability to procure precious samples from far-flung collaborators.

In 1957 Franklin and her coworkers Klug and Finch began to study poliovirus in addition to their continuing work on TYMV. Poliovirus was the first animal virus crystallized, in work done by Frederick Schaffer and Carleton Schwerdt of the Berkeley Virus Laboratory in 1955. Crystals large enough for single-crystal X-ray diffraction took a year to grow. The Berkeley researchers gave a sample of these to Franklin in 1957.107 Some of the people working in the Birkbeck laboratories were not pleased that Franklin was working with poliovirus, believing that it posed an unacceptable risk of human infection. They convened a meeting regarding her use of poliovirus at Birkbeck, with the result that Franklin was prohibited from using the virus in her lab. The crystals were then taken from Birkbeck to the London School of Hygiene and Tropical Medicine, which had better facilities for dealing with highly infectious pathogens and was a safer place to mount the crystals. Franklin also needed a powerful X-ray tube if she was to take diffraction pictures. Klug contacted Bragg at the Royal Institution, which by then had a rotating anode X-ray tube copied from the design of the tube at Cambridge. Bragg was quite receptive to the work's being done at the Royal Institution; Klug had to get permission to work with an infectious virus, but there were no formal guidelines and so he invented his own and submitted them to an inspector.108

Franklin's attempts to mount the polio crystals in capillary tubes in early to mid-1957 were unsuccessful. Each crystal would spontaneously dissolve in the capillary before she could get any diffraction pictures. Franklin attributed this instability to an alkaline reaction occurring in the borosilicate glass of the capillary, such that salts were leaching out of the glass. She then tried using acid-treated capillaries, which delayed the dissolution of the crystals but not enough for her to get any data. In the month before her death she wrote to Bawden suggesting that Pyrex tubes might be better. Franklin also corresponded with Ronald W. Douglas, of the Department of Glass Technology at Sheffield, who recommended that she use “neutral glass.” Franklin did not live long enough to see the results of the poliovirus project, but her colleagues continued the work. With the arrival of a new batch of crystals from Berkeley, Klug discovered that quartz capillary tubes worked much better than glass ones and allowed the virus crystals to be mounted without dissolving. When crystals were at last successfully mounted in capillaries, they were transported across town, to the Davy Faraday Laboratory of the Royal Institution, in a thermos. Finch and Klug found that poliovirus, like BSV and TYMV, possesses icosahedral symmetry, showing that spherical viruses from plants and animals exhibit the same structural organization.109

In the summer of 1956, Sir Lawrence Bragg had written several biophysicists to enlist their help in preparing exhibits for the upcoming 1958 World's Fair in Brussels. As he told Crick, the organizers wanted “to make a big feature of the nucleic acid in the biology section. Wilkins will have to be brought in too and I am hoping to borrow his model or some later version of it.” He continued,

The organizers of the biology section also want to make an important show of the work on the helical and spherical viruses. I am writing to Miss Franklin but you and Watson come in here too, so it will be a case of general collaboration. I have been so deeply impressed by all the grand work you have done on these complex structures and I wish to see that it has a proper place in the Exhibition.110

As it developed, the exhibits were less a “case of general collaboration” than a matter for careful coordination, with delicate negotiations about which contributions would be jointly prepared.111 Wilkins offered to contribute, with Crick, a model of DNA; and Crick said he would take care of collagen. Their letters expose the tensions that arose when scientists from King's College and Cambridge were asked to collaborate on presenting nucleic acid structure. Franklin's name never came up in connection with the DNA models, but she and Klug contributed an exhibit of large-scale structural models of TMV and TYMV.112 Once built, this model of a segment of the TMV helix was over five feet high (see cover and Figure 6).

Figure 6. Photograph of TMV model constructed for the International Exhibit in Brussels in 1958. This large model now sits in a stairwell at the Cambridge Laboratory of Molecular Biology. The cable exposed under the egg-shaped protein subunits represents the strand of viral RNA. Copyright Gregory J. Morgan.

Open New Window

On 16 April 1958, just as the Brussels exhibition opened to international acclaim, Franklin succumbed to cancer. That summer Caspar presented a paper coauthored with Klug at the fiftieth anniversary meeting of the American Phytopathological Society in place of Franklin. They wrote a review article of X-ray diffraction studies of viruses drawing largely from Franklin's work and posthumously added her as first author. Their subsequent collaboration culminated in the Caspar-Klug theory of virus structure, a proposal that extended Crick and Watson's earlier speculation by suggesting that spherical viruses were structured like microscopic geodesic domes.113 Klug continued his work on TMV and spherical viruses and developed a general method for three-dimensional image reconstruction. The scientific recognition he received included the 1982 Nobel Prize in Chemistry. He dedicated approximately a third of his Nobel lecture to the structure and dynamics of TMV. As he commented, Franklin herself might have stood on that platform “had her life not been cut tragically short.”114

CONCLUSIONS

 

The history of structural studies of viruses provides a useful vantage point for assessing the consolidation of molecular biology in the 1950s, one that sheds new light on the better-known double-helix episode. At a personal level, the relations among Watson, Crick, and Franklin around DNA should be viewed as part of a longer historical trajectory. In fact, their interactions intensified after 1953. After leaving King's College Franklin viewed Watson and Crick not as enemies but as colleagues and, at times, competitors. Crick became a friend.115 Her relations with Watson were clearly more complicated. After 1953, Watson hoped that his new work elucidating the structure of RNA would clarify the molecular mechanisms of gene expression and protein synthesis, but this was a dead end. At the same time, it proved difficult for him to continue the work on TMV that he had started; Franklin had taken over this area, building on his early achievement but also, clearly, surpassing it. Instead, his most important subsequent contributions here resulted from his theoretical work with Crick, particularly their two important papers of 1956 on virus structure. As in the case of their earlier work on DNA, their collaboration on these papers consisted of reviewing and interpreting the results of other experimentalists, spurred in this case by Caspar's diffraction patterns of BSV. During this period Watson's interactions with Franklin were friendly, but the friction of competition also comes through in letters and reminiscences.

Franklin's work on viruses reveals a willingness to publish speculative models based on early interpretations of the data. In fact, this readiness led her initially to publish incorrect helical parameters for TMV. This attitude seems in strong contrast to her perceived approach to the DNA structure and contradicts the impression, attributable largely to Watson and Crick, that she was a talented experimentalist who did not know how to interpret her own data.116 Whether this image of cautious hesitation was never accurate or whether her willingness to stake scientific claims in her work on plant virus structure was a response to being denied credit for the helical structure of DNA cannot be ascertained. Regardless, her ambitions were undiminished by her unhappy stint at King's College. She fought hard to be recognized and remunerated on a par with other senior scientists, as evidenced by her protracted but ultimately successful struggle to win long-term funding for her internationally recognized research program at Birkbeck.117 In the end, the most direct beneficiary of Franklin's labors was Klug, who moved her research program and group to Cambridge in 1962 to join Perutz in the new Laboratory of Molecular Biology.118

By placing structural research on viruses in the foreground rather than in the background of the DNA story, we can also recover some of the key research questions that motivated molecular biologists in the mid-1950s. Significantly, the elucidation of the three-dimensional structure of TMV, including the location of its protein subunits and RNA, took place at a time when the value of the Watson-Crick double-stranded model for DNA was still insecure. As Frederic Lawrence Holmes has argued, the topological problems associated with the Watson-Crick model—namely, the fact that what Max Delbrück called an “unwindase” was needed—meant that acceptance of the model was tentative.119 The Meselson-Stahl experiment, published in 1958, gave strong support to the Watson-Crick model by demonstrating the semiconservative mode of DNA replication. But between 1952 and 1958, even though the hereditary role of DNA was accepted, there was uncertainty about the precise relationship between nucleic acid genes and their products—namely proteins—and about the role of RNA in this connection.

Amidst the more mathematical approaches to the coding problem, as it was known, many molecular biologists studied virus structure precisely because both the nucleic acid and protein components could be isolated and examined, offering a promising way to determine the underlying structural correlations. Plant viruses were especially important to this endeavor because they tended to be more chemically tractable than either bacterial or animal viruses. By 1956, biochemists had demonstrated that the RNA of TMV was infectious on its own—and fully responsible for the sequence of amino acids in the viral protein.120 As a consequence, TMV was among the most appealing model systems for breaking the genetic code in the late 1950s and early 1960s.121 Crystallographers and biochemists also turned to microsomes to understand protein synthesis, and their similarities to spherical viruses in both size and composition stimulated speculations that there might be functional commonalities.122

It is also valuable to juxtapose the structural studies of viruses against structural work on DNA because techniques, insights, and materials circulated from one realm to the other. The same fiber-diffraction techniques could be applied to discern structure in preparations of TMV and in the long oriented molecules of DNA. If Watson learned the power of model building from Linus Pauling's sensational alpha helix, he learned to see helical structures in X-ray diffraction data from his work on TMV. Franklin's varying of the humidity conditions in which fibers were examined enabled her to distinguish two forms of DNA, an A form and a wetter B form. She then employed the same method with plant virus preparations—as did Watson with RNA. The circulation of materials was a more complex matter, owing to the scarcity of chemically pure samples and the competition among researchers to acquire them. X-ray crystallographers generally depended on the generosity of biochemists to provide them with purified preparations, but the sharing of material entailed moral obligations even as it enabled individual success. One can see from their correspondence the ways in which these scientists negotiated credit and priority and sought to associate key breakthroughs with their own reputations. Watson's sense that he needed to cultivate a virus of his own is particularly telling. Franklin was savvy in navigating this exchange network, obtaining viruses from rival groups and turning potential challengers into collaborators.123 The transnational character of molecular biology as it emerged in the 1950s derived in part from the circulation of materials and information (not to mention researchers themselves) among this transcontinental network of biophysicists and biochemists studying viruses, nucleic acids, and proteins.

Given the importance of these structural studies to molecular biologists in the 1950s, why has the work on RNA, TMV, and spherical viruses been so completely overshadowed in historical memory by the earlier DNA episode? Several factors contributed to this tendency. The pertinence of structural studies of viruses (as well as microsomes) to the genetic code ended in the 1960s; thereafter TMV and other well-studied viruses began to be viewed as models of self-assembly instead.124 Along similar lines, research on protein synthesis made it clear that several specific types of RNA—messenger RNA, transfer RNA, ribosomal RNA—played distinct roles in protein synthesis, making viral RNA a less useful model for understanding the transcription and translation of genes.125 And, perhaps above all, the double helix itself became a potent icon for molecular biology. As Soraya de Chadarevian has suggested, this did not occur immediately in response to the appearance of Watson and Crick's structural model for DNA but, rather, developed gradually.126 The publication of The Double Helix and the subsequent advent of both recombinant DNA techniques in the 1970s and the biotechnology industry in the 1980s gave the structure a wider circulation and cultural valence. By contrast, DNA was not the only helix of significance in the 1950s. An examination of Franklin's virus research at close range not only enables a fuller appreciation of her scientific accomplishments but also reveals the exchange networks and international alliances from which the discipline of molecular biology crystallized.

  • Our research was supported by the National Science Foundation, through a CAREER grant, SBE 98-75012 (ANHC), and a Dissertation Research Improvement Grant, SBE 99-10891 (GJM). For access to correspondence and other historical materials we thank Donald Caspar, Aaron Klug, Jeremy Norman, Shannon Bohle at the Cold Spring Harbor Laboratory Archives, and archivists at the Bancroft Library, the Caltech Archives, the Churchill Archives Centre, the Novartis Foundation, the Royal Institution of Great Britain, the University of Maryland, Baltimore County, and the University of Melbourne Archives. We acknowledge the valuable feedback we received on versions of this essay presented at “Molecular Biology in the Twentieth Century: A Meeting to Mark the Fiftieth Anniversary of the Determination of the Structure of DNA,” organized by Frank James and held on 28–29 Apr. 2003 at the Royal Institution, London, and at a session organized by Karen-Beth Scholthof and Paul Peterson at the American Phytopathological Society Meeting on 1 Aug. 2005 in Austin, Texas. We also thank Donald Caspar, Nathaniel Comfort, John Finch, Michael Gordin, Michael Keevak, Aaron Klug, Kenneth Holmes, Bernard Lightman, Karen-Beth Scholthof, Judith Swan, Sue Tolin, Daniel Trambaiolo, Doogab Yi, and three anonymous referees for their suggestions, corrections, and criticisms. The authors alone bear responsibility for the essay, including its interpretations and any remaining errors.

  • 1 We have followed current nomenclature in italicizing the full names of virus species.

  • 2 James D. Watson and Francis H. C. Crick, “A Structure for Deoxyribose Nucleic Acid,” Nature, 1953, 171:737–738. There are many accounts of the relationship of Franklin's diffraction patterns to Watson and Crick's model; for a recent appraisal that cites others see Lynne Osman Elkin, “Rosalind Franklin and the Double Helix,” Physics Today, 2003, 56:42–48.

  • 3 James D. Watson, The Double Helix: A Personal Account of the Discovery of the Structure of DNA, ed. Gunther S. Stent (1968; New York: Norton, 1980) (hereafter cited as Watson, Double Helix). For analysis of the mythic dimensions of the representation of Franklin as a martyr, in which she is cast as the “Sylvia Plath of molecular biology,” see Brenda Maddox, “The Double Helix and the ‘Wronged Heroine,’” Nature, 2003, 421:407–408. Anne Sayre's book played a crucial role in drawing public attention to the injustices of Watson's portrayal: Anne Sayre, Rosalind Franklin and DNA (New York: Norton, 1975).

  • 4 Brenda Maddox, Rosalind Franklin: The Dark Lady of DNA (New York: Harper Collins, 2002) (hereafter cited as Maddox, Rosalind Franklin). A few examples of the media accounts are Tara Pepper, “Genes, Girls, and Gall,” Newsweek, 5 Aug. 2002, p. 54; Jim Holt, “Photo Finish: Rosalind Franklin and the Great DNA Race,” New Yorker, 28 Oct. 2002, p. 102; Bernadine Healy, “Let's Remember Rosy,” U.S. News and World Report, 24 Feb. 2003, p. 47; and Denise Grady, “A Revolution at Fifty: Fifty Years Later, Rosalind Franklin's X-ray Fuels Debate,” New York Times, 25 Feb. 2003, p. 2. The PBS program Nova produced a segment on Franklin that was broadcast on 22 Apr. 2003. For an analysis of the commemorations see Pnina G. Abir-Am, “DNA at Fifty: Institutional and Biographical Perspectives,” Minerva, 2004, 42:191–213.

  • 5 Franklin's publications on TMV are cited in the course of this essay; for an overview of her contributions see Kenneth C. Holmes, “Rosalind Franklin and the Tobacco Mosaic Virus,” in DNA 50: The Secret of Life, ed. Miriam Balaban (London: Faircount, 2003), pp. 200–208.

  • 6 This risk taking stands in contrast to the image of Franklin as such a cautious experimentalist that she resisted structural speculation; see Watson, Double Helix, pp. 45, 95–96.

  • 7 This, too, contradicts the impression—drawn from her work on DNA at King's College and based largely on the portrayal offered in The Double Helix—of Franklin as a solitary investigator. (Even there, she worked closely with Raymond Gosling, if not with Wilkins.)

  • 8 On the importance of hemoglobin and insulin to biochemistry and biophysics see Soraya de Chadarevian, “Sequences, Conformation, Information: Biochemists and Molecular Biologists in the 1950s,” Journal of the History of Biology, 1996, 29:361–386; de Chadarevian, “Following Molecules: Hemoglobin between the Clinic and the Laboratory,” in Molecularizing Biology and Medicine: New Practices and Alliances, 1910s–1970s, ed. de Chadarevian and Harmke Kamminga (Amsterdam: Harwood, 1998), pp. 171–201; and de Chadarevian, Designs for Life: Molecular Biology after World War II (Cambridge: Cambridge Univ. Press, 2002) (hereafter cited as de Chadarevian, Designs for Life). Secondary literature on virus research is cited throughout this essay.

  • 9 For an insightful synoptic account that cites the secondary literature up to the mid-1990s see Michel Morange, A History of Molecular Biology, trans. Matthew Cobb (Cambridge, Mass.: Harvard Univ. Press, 1998).

  • 10 On the coding problem see Lily E. Kay, Who Wrote the Book of Life? A History of the Genetic Code (Stanford, Calif.: Stanford Univ. Press, 2000); and Horace Freeland Judson, The Eighth Day of Creation (New York: Simon & Schuster, 1979), Chs. 5–8. Both Kay and Judson make it clear that researchers employing computational and theoretical methods (largely members of the RNA Tie Club) were not successful in actually cracking the code; this was accomplished by Heinrich Matthei and Marshall Nirenberg using biochemical methods in the early 1960s.

  • 11 The RNA Tie Club was part joke and part scientific network. Watson, George Gamow, and Leslie Orgel launched this clique of scientific correspondents to encourage work to resolve the structure of RNA and to explicate its role in forming proteins, specifically by providing a forum for speculative ideas and untested theories. Other founding members included Crick, Gunther Stent, and Alexander Rich. See Judson, Eighth Day of Creation, pp. 264–265; and de Chadarevian, Designs for Life, pp. 186–198. On structural models of protein synthesis (notably the template theory of Linus Pauling) see Bruno Strasser, “A World in One Dimension: Linus Pauling, Francis Crick, and the Central Dogma of Molecular Biology,” History and Philosophy of the Life Sciences, 2006, 28:491–512.

  • 12 Pnina G. Abir-Am, “From Multidisciplinary Collaboration to Transnational Objectivity: International Spaces as Constitutive of Molecular Biology, 1930–1970,” in Denationalizing Science: The Contexts of International Scientific Practice, ed. Elisabeth Crawford, Terry Shinn, and Sverker Sörlin (Dordrecht: Kluwer, 1992), pp. 153–186; de Chadarevian, Designs for Life; Jean-Paul Gaudillière, Inventer la biomédecine: La France, l'Amérique et la production des savoirs du vivant (1945–1965) (Paris: Découverte, 2002); and Bruno J. Strasser, La fabrique d'une nouvelle science: La biologie moléculaire à l'âge atomique (1945–1964) (Florence: Olschki, 2006).

  • 13 These were transactions, if not monetary ones: scientists exchanged materials and results in return for credit or in the expectation of reciprocity. This understanding of the circulation of scientific information and objects in terms of “gift exchange” draws on economic anthropology and sociology; for an excellent discussion and references see Warwick Anderson, “The Possession of Kuru: Medical Science and Biocolonial Exchange,” Comparative Studies in Society and History, 2000, 42:713–744, esp. pp. 714–716.

  • 14 For a nice example of differences in expectations between French and American cultures of molecular biology see Jean-Paul Gaudillière, “Paris–New York Roundtrip: Transatlantic Crossings and the Reconstruction of the Biological Sciences in Post-war France,” Studies in History and Philosophy of Biological and Biomedical Sciences, 2002, 33:389–417, esp. pp. 406–408. On this general issue see Soraya de Chadarevian and Bruno Strasser, “Molecular Biology in Postwar Europe: Towards a ‘Global’ Picture,” ibid., pp. 361–365.

  • 15 For an insightful account of how this played out at Cambridge see de Chadarevian, Designs for Life, esp. Ch. 10. On the perception of King's College's prerogative regarding the DNA structure problem see Watson, Double Helix, pp. 13–14.

  • 16 Dmitri Ivanovskii, “Über die Mosaikkrankheit der Tabakspflanze,” Bulletin de l'Académie Impériale des Sciences de St. Pétersbourg, Sér. 3, 1892, 35:67–70, trans. James Johnson and rpt. as “Concerning the Mosaic Disease of the Tobacco Plant,” Phytopathological Classics, 1942, 7:27–30; and M. W. Beijerinck, “Über ein Contagium vivum fluidum als Ursache der Fleckenkrankheit der Tabaksblätter,” Verhandelingen der Koninklijke Akademie van Wetenschappen te Amsterdam, Afdeeling Natuurkunde, 1898, 6:3–21, trans. Johnson and rpt. as “Concerning a Contagium vivum fluidum as a Cause of the Spot-Disease of Tobacco Leaves,” Phytopatholog. Classics, 1942, 7:33–52. For more on the significance of early research on TMV see Ton van Helvoort, “What Is a Virus? The Case of Tobacco Mosaic Disease,” Studies in History and Philosophy of Science, 1991, 22:557–588; Angela N. H. Creager, The Life of a Virus: Tobacco Mosaic Virus as an Experimental Model, 1930–1965 (Chicago: Univ. Chicago Press, 2002) (hereafter cited as Creager, Life of a Virus), Ch. 2; and Karen-Beth G. Scholthof, John G. Shaw, and Milton Zaitlin, eds., Tobacco Mosaic Virus: One Hundred Years of Contributions to Virology (St. Paul, Minn.: American Phytopathological Society Press, 1999).

  • 17 For Stanley's announcement see W. M. Stanley, “Isolation of a Crystalline Protein Possessing the Properties of Tobacco-Mosaic Virus,” Science, 1935, 81:644–645. On the reception of Stanley's paper as a scientific sensation see Lily E. Kay, “W. M. Stanley's Crystallization of the Tobacco Mosaic Virus, 1930–1940,” Isis, 1986, 77:450–472; and Creager, Life of a Virus, Ch. 3. For the clarification see F. C. Bawden, N. W. Pirie, J. D. Bernal, and I. Fankuchen, “Liquid Crystalline Substances from Virus-Infected Plants,” Nature, 1936, 138:1051–1052. These authors illustrated the spontaneous birefringence of TMV, which indicated the presence of highly elongated particles, with a photograph showing the pattern left by a goldfish swimming in a solution of the virus.

  • 18 Bernal and Fankuchen prepared both “wet” and “dry” gels of TMV; the wetter preparations, particularly one in which the long particles were oriented in capillary tubes, gave the clearest diffraction patterns, with hundreds of distinct spots. The analysis of X-ray diagrams from these materials is based not on crystallography proper but on fiber-diffraction methods. See J. D. Bernal and I. Fankuchen, “Structure Types of Protein ‘Crystals’ from Virus-Infected Plants,” Nature, 1937, 139:923–924, on p. 923; and Bernal and Fankuchen, “X-ray and Crystallographic Studies of Plant Virus Preparations, I: Introduction and Preparation of Specimens; II: Modes of Aggregation of the Virus Particles; III,” Journal of General Physiology, 1941, 25:111–146 (Pts. I and II), 147–165 (Pt. III). For a discussion of Bernal and Fankuchen's diffraction analysis of TMV see Robert C. Olby, The Path to the Double Helix: The Discovery of DNA (1974; New York: Dover, 1994), pp. 164–165, 259–263.

  • 19 The Svedberg used sedimentation studies to estimate a molecular weight for TMV of 17 million daltons in 1937; by 1940 Stanley had revised this figure to 50 million daltons, on the basis of the assumption that the virus was cylindrical rather than spherical in shape. See Inga-Britta Eriksson-Quensel and Theodor Svedberg, “Sedimentation and Electrophoresis of the Tobacco-Mosaic Virus Protein,” Journal of the American Chemical Society, 1936, 58:1863–1867; W. M. Stanley, “The Biochemistry of Viruses,” Annual Review of Biochemistry, 1940, 9:545–570; and Creager, Life of a Virus, Ch. 4.

  • 20 See F. C. Bawden and N. W. Pirie, “Contribution to Aggregation of Purified Tobacco Mosaic Virus,” Nature, 1938, 142:842–843. On these debates see Creager, Life of a Virus, Ch. 4.

  • 21 The first electron micrographs of TMV were published in G. A. Kausche, E. Pfankuch, and H. Ruska, “Die Sichtbarmachung von pflanzlichen Virus im Übermikroskop,” Naturwissenschaften, 1939, 27:292–299. The development of the RCA electron microscope led Stanley to collaborate on micrographs of his TMV preparation; see W. M. Stanley and Thomas F. Anderson, “A Study of Purified Viruses with the Electron Microscope,” Journal of Biological Chemistry, 1941, 139:325–338. Measurements from their micrographs led Stanley to assign a length of 2,800 Å and a width of 150 Å. Bawden and Pirie interpreted the long rods visualized in electron micrographs of TMV as artifactual aggregates, believing that the biologically active virus particles were much smaller and possibly even spherical. See F. C. Bawden, “Virus Diseases of Plants,” Journal of the Royal Society of Arts, 1946, 94:136–168, esp. p. 166. On electron microscopy see Nicolas Rasmussen, Picture Control: The Electron Microscope and the Transformation of Biology in America, 1940–1960 (Stanford, Calif.: Stanford Univ. Press, 1997). On the decline of the colloidal view of proteins see Joseph Fruton, “From Colloids to Macromolecules,” in Molecules and Life: Historical Essays on the Interplay of Chemistry and Biology (New York: Wiley-Interscience, 1972), pp. 131–148.

  • 22 Gerhard Schramm, “Über die Spaltung des Tabakmosaikvirus in niedermolekulare Proteine und die Rückbildung hochmolekularen Proteins aus den Spaltstücken,” Naturwissenschaften, 1943, 31:94–96. Schramm's techniques are discussed further below. According to Watson, “There already existed biochemical evidence for protein building blocks. Experiments of the German Gerhard Schramm, first published in 1944, reported that TMV particles in mild alkali fell apart into free RNA and a large number of similar, if not identical, protein molecules. Virtually no one outside Germany, however, thought that Schramm's story was right. This was because of the war. It was inconceivable to most people that the German beasts would have permitted the extensive experiments underlying his claims to be routinely carried out during the last years of a war they were so badly losing. It was all too easy to imagine that the work had direct Nazi support and that his experiments were incorrectly analyzed”: Watson, Double Helix, p. 68. On Stanley's skepticism about Schramm's result see Creager, Life of a Virus, pp. 249–253.

  • 23 On the institutionalization of biophysics in the postwar United Kingdom, including Randall's laboratory at King's College, see de Chadarevian, Designs for Life, Ch. 3. Maddox sheds new light on why Franklin's arrival to work on DNA created misunderstandings and friction with Wilkins in Rosalind Franklin, pp. 114–116, 128–129, 149–150.

  • 24 M. H. F. Wilkins, “The Molecular Configuration of Nucleic Acids,” in Nobel Lectures in Physiology or Medicine (Amsterdam: Elsevier for the Nobel Foundation, 1964), Vol. 3, pp. 754–782, on p. 755; and Olby, Path to the Double Helix (cit. n. 18), p. 331. Another first-hand account of this collaboration can be found in Gerald Oster to Wendell Stanley, 26 Apr. 1949, Wendell M. Stanley Papers, Bancroft Library, University of California, Berkeley, 78/18c (hereafter cited as Stanley Papers), carton 11, folder Oster.

  • 25 M. H. F. Wilkins, A. R. Stokes, W. E. Seeds, and G. E. Oster, “Tobacco Mosaic Virus Crystals and Three-Dimensional Microscopic Vision,” Nature, 1950, 166:127–129, on p. 127. They estimated the length of virus rods at 2,800 Å. In his autobiography, Wilkins credits Oster with inspiring him to pursue the DNA structure using X-ray diffraction; see Maurice Wilkins, The Third Man of the Double Helix (Oxford: Oxford Univ. Press, 2003), p. 116.

  • 26 In addition, Robley Williams at the Berkeley Virus Lab strongly challenged the claim of Wilkins and his collaborators that the pattern of banded striations in polarizing light demonstrated a “zig-zag” orientation of virus particles. See correspondence between Robley Williams and M. H. F. Wilkins, Nov. 1952, Feb. 1953, Robley C. Williams Papers, Bancroft Library, University of California, Berkeley (hereafter cited as Williams Papers), 73/7c, carton 5, folder W. On Stokes's helical interpretation see Wilkins, Third Man of the Double Helix, p. 116.

  • 27 Watson went to Europe on a Merck National Research Council Fellowship, which was cut short in 1952 because of his decision to leave Copenhagen for Cambridge; see Watson, Double Helix, p. 66. Thereafter, Max Delbrück helped arrange a fellowship for Watson through the National Foundation for Infantile Paralysis. See Watson, Double Helix, p. 66; Olby, Path to the Double Helix (cit. n. 18), p. 378; and Victor K. McElheny, Watson and DNA: Making a Scientific Revolution (Cambridge, Mass.: Perseus, 2003), p. 46. On the role of the National Foundation for Infantile Paralysis in supporting basic virus research see Creager, Life of a Virus, Ch. 5.

  • 28 W. Cochran, F. H. C. Crick, and V. Vand, “The Structure of Synthetic Polypeptides, I: The Transform of Atoms on a Helix,” Acta Crystallographica, 1952, 5:581–586; Bernal and Fankuchen, “X-ray and Crystallographic Studies of Plant Virus Preparations” (cit. n. 18), p. 148; and Olby, Path to the Double Helix, pp. 260, 311–312, 316 (quotation). For Frank's theory see F. C. Frank, “The Influence of Dislocations on Crystal Growth,” Discussions of the Faraday Society, 1949, 5:48–54; and Frank, “Crystal Growth and Dislocations,” Advances in Physics, 1952, 1:91–109.

  • 29 Watson, Double Helix, p. 69.

  • 30 Ibid., p. 98. According to Maddox, Watson's visit to King's College, during which Wilkins indiscreetly showed him Franklin's “Photograph 51,” took place on 30 Jan. 1953; see Maddox, Rosalind Franklin, pp. 193–197.

  • 31 J. D. Watson, “The Structure of Tobacco Mosaic Virus, I: X-ray Evidence of a Helical Arrangement of Sub-units around the Longitudinal Axis,” Biochimica et Biophysica Acta, 1954, 13:10–19. Watson took pictures of both wet and dry TMV preparations, but the table of meridional reflections he used to come up with an n of 10 was based on work with the dry specimen. Regarding the protein subunit number, contemporary—and unexpected—biochemical evidence from proteolytic digests of TMV in the Berkeley Virus Lab gave an estimate closer to three thousand; see J. Ieuan Harris and C. Arthur Knight, “Action of Carboxypeptidase on Tobacco Mosaic Virus,” Nature, 1952, 170:613–614. On these developments at Berkeley see Creager, Life of a Virus, pp. 266–270; on the various kinds of evidence for TMV subunits from crystallography, physical chemistry, and biochemistry see Donald D. L. Caspar, “The Radial Structure of Tobacco Mosaic Virus” (Ph.D. diss., Yale Univ., 1955), Introduction.

  • 32 Franklin cites Watson's unpublished paper in Annual Report, 1 Jan. 1953 to 1 Jan. 1954, Anne Sayre Collection of the American Society for Microbiology Archives at the University of Maryland, Baltimore County (hereafter cited as Sayre Collection), box 3, folder 6, p. 3. For the published version see Watson, “Structure of Tobacco Mosaic Virus.”

  • 33 Rosalind Franklin, Annual Report, 1 Jan. 1953 to 1 Jan. 1954, Sayre Collection, box 3, folder 6. It seems likely that completing the DNA work occupied most of Franklin's time. One virologist has commented that reading through the relevant sources in the plant virus literature could not have taken her more than a few weeks: Karen-Beth Scholthof, personal communication to Angela N. H. Creager, 21 May 2007.

  • 34 Maddox quotes from Randall's letter asking Franklin not only to cease working on DNA but to stop thinking about it; Franklin found this absurd: Maddox, Rosalind Franklin, pp. 212–213, 221. There were five joint publications: Rosalind E. Franklin and R. G. Gosling, “Molecular Configuration in Sodium Thymonucleate,” Nature, 1953, 171:740–741; Franklin and Gosling, “Evidence for Two-Chain Helix in Crystalline Structure of Sodium Desoxyribonucleate,” ibid., 1953, 172:156–157; and Franklin and Gosling, “The Structure of Sodium Thymonucleate Fibres, I: The Influence of Water Content; II: The Cylindrically Symmetrical Patterson Function; III: The Three-Dimensional Patterson Function,” Acta Crystallog., 1953, 6:673–677, 678–685; 1955, 8:151–156. Commentators have speculated on how long it might have taken Franklin to deduce the double-helical structure of DNA during these months had Watson and Crick not already published their model. Franklin's near recognition of the structure is argued in A. Klug, “Rosalind Franklin and the Double Helix,” Nature, 1974, 248:787–788; and Elkin, “Rosalind Franklin and the Double Helix” (cit. n. 2); and is taken into account by Maddox, Rosalind Franklin. A more skeptical assessment is offered by Horace Freeland Judson, “Reflections on the Historiography of Molecular Biology,” Minerva, 1980, 18:369–421.

  • 35 “I'm sorry you are not having much luck. I'm enclosing a specimen of TMV—which is very highly aggregated indeed. It is also as clean as or cleaner than the specimen which Watson used”: Roy Markham to Rosalind Franklin, 23 Nov. 1953, Papers of Rosalind Franklin, Churchill Archives Centre, Churchill College, Cambridge (hereafter cited as Franklin Papers), FRNK 2/33.

  • 36 Franklin to A. L. Patterson, 1 Dec. [1953], Sayre Collection, box 3, folder 1 (complications of TMV fiber diagrams); and Maddox, Rosalind Franklin, p. 229, citing J. D. Bernal to John Randall, 10 Oct. 1953, and Randall to Bernal, 4 Nov. 1953: Franklin Papers, FRNK 2/31.

  • 37 Regarding both the narrow-mindedness of Bernal's left-wing associates and the virtues of working at Birkbeck see Franklin to Anne and David Sayre, 17 Dec. [1953], Sayre Collection, box 3, folder 1. Dorothy Crowfoot Hodgkin, another preeminent woman crystallographer, had been in Bernal's laboratory ten years earlier; see Georgina Ferry, Dorothy Hodgkin: A Life (London: Granta, 1998). Bernal had done his own crystallographic training with William H. Bragg, who had also been supportive of women, most notably Kathleen Lonsdale. Marcel Mathieu played a similarly welcoming role for women crystallographers in France; see Maureen M. Julian, “Women in Crystallography,” in Women of Science: Righting the Record, ed. G. Kass-Simon and Patricia Farnes (Bloomington: Indiana Univ. Press, 1990), pp. 335–383. On the significance of socialism in opening research opportunities in radioactivity to women see the excellent article by Maria Rentetzi, “Gender, Politics, and Radioactivity Research in Interwar Vienna: The Case of the Institute for Radium Research,” Isis, 2004, 95:359–393. As in the case of X-ray crystallography, the fact that radioactivity research “involved meticulous, routine, and repetitive work” has been used to explain the relatively high numbers of women in the field (ibid., p. 360). Rentetzi instead encourages a serious consideration of the role of progressive politics, an interpretation that seems equally compelling in accounting for the women in X-ray crystallography, at least in Bernal's laboratory.

  • 38 Interviews, Gregory J. Morgan with Aaron Klug, Cambridge, 2 June 1999, 17 July 2000; and Andrew Brown, J. D. Bernal: The Sage of Science (Oxford: Oxford Univ. Press, 2005), p. 353.

  • 39 As Watson explains in his Nobel lecture, he actually took a few diffraction pictures of RNA in 1952 while he was working on TMV, but these were “very diffuse”: James D. Watson, “The Involvement of RNA in the Synthesis of Proteins,” in Nobel Lectures in Physiology or Medicine (cit. n. 24), Vol. 3, pp. 785–808, on p. 787.

  • 40 Alexander Rich and J. D. Watson, “Some Relations between DNA and RNA,” Proceedings of the National Academy of Sciences, USA, 1954, 40:759–764, esp. p. 759. While others (especially Crick) shared Rich and Watson's presumption that DNA was the genetic material and RNA was responsible for protein synthesis, this view was complicated by the fact that in most plant viruses, including TMV, no DNA was present, so the viral RNA was presumed to act as the genetic material—assuming the viral protein was not genetic. Rich and Watson state that in “these viruses the genetic material must be the RNA component or the protein component, or possibly both.” They point out that if RNA served as genetic material in some viruses, you might expect the base ratios to be complementary in those cases, but in fact the opposite result was observed: “Plant virus RNA's show great departure from the 1:1 ratio, while RNA's from sources to which we need not necessarily postulate a genetic role (e.g., microsomes, mitochondria) often provide beautiful examples of complementarity. We have no explanation for this finding” (ibid., p. 763). Writing retrospectively of his renewed interest in TMV in 1954, Watson states, “Always troublesome to me was the apparent necessity to postulate both genetic and protein synthesis roles for RNA”: Watson, “Early Speculations and Facts about RNA Templates,” in A Passion for DNA: Genes, Genomes, and Society (Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory Press, 2000), pp. 23–32, on pp. 27–28. On Watson's research at Caltech see Watson, Genes, Girls, and Gamow: After the Double Helix (New York: Knopf, 2002) (hereafter cited as Watson, Genes, Girls, and Gamow), esp. Chs. 6, 15; and Frederic Lawrence Holmes, Meselson, Stahl, and the Replication of DNA: A History of “The Most Beautiful Experiment in Biology” (New Haven, Conn.: Yale Univ. Press, 2001), Ch. 1.

  • 41 See James D. Watson to Max Delbrück, 25 Mar. 1954, Max Delbrück Papers, California Institute of Technology Archives, Pasadena (hereafter cited as Delbrück Papers), 23.23, on this result and on the “very secret society.” On the RNA Tie Club see also Watson, Genes, Girls, and Gamow, pp. 67–68. Rich and Watson's first publication also focuses on changes in the diffraction patterns from raising the relative humidity of the RNA samples: Alexander Rich and J. D. Watson, “Physical Studies on Ribonucleic Acid,” Nature, 1954, 173:995–996.

  • 42 Attention to the effects of humidity on fiber length originated with Wilkins before being adapted so successfully by Franklin, who controlled the specimen's water content using saturated salt solutions, enabling her to detect different structural conformations of the DNA. See M. H. C. Wilkins, R. G. Gosling, and W. E. Seeds, “Physical Studies of Nucleic Acid: Nucleic Acid: An Extensible Molecule?” Nature, 1951, 167:759–760; Jeremy Bernstein, “A Sorry and a Pity: Rosalind Franklin and The Double Helix,” in Experiencing Science (New York: Basic, 1978), pp. 143–162, esp. p. 153; Wilkins, Third Man of the Double Helix (cit. n. 25), pp. 122–124; and Aaron Klug, “The Discovery of the DNA Double Helix,” Journal of Molecular Biology, 2004, 335:3–26.

  • 43 Watson to Delbrück, 25 Mar. 1954, Delbrück Papers, 23.23; Rich and Watson, “Some Relations between DNA and RNA” (cit. n. 40); and Alexander Rich, “The Nucleic Acids: A Backward Glance,” Annals of the New York Academy of Sciences, 1995, 758:97–142.

  • 44 Rich and Watson, “Physical Studies on Ribonucleic Acid” (cit. n. 41); and Watson to Delbrück, 1 June 1954, Delbrück Papers, 23.23. One paper resulting from Rich's work at the NIH is Alexander Rich, J. D. Dunitz, and P. Newmark, “Abnormal Protein Associated with Tobacco Mosaic Virus: Structure of Polymerized Tobacco Plant Protein and Tobacco Mosaic Virus,” Nature, 1955, 175:1074–1075.

  • 45 Rosalind Franklin, Annual Report, 1 Jan. 1953 to 1 Jan. 1954, Sayre Collection, box 3, folder 6; Markham to Franklin, 23 Nov. 1953, Franklin Papers, FRNK 2/33 (on the virus preparation); and Franklin to Stanley, 7 May 1954, Stanley Papers, carton 8, folder Franklin, Rosalind. See similar information in Franklin to Ernest Pollard (of Yale), 13 Apr. 1954, Franklin Papers, FRNK 2/34, quoted in Maddox, Rosalind Franklin, pp. 234–235.

  • 46 Regarding the invitation and Franklin's efforts to raise sufficient funds for the trip, including an application to the Rockefeller Foundation, see Anne Sayre's notes on “REF 1954 US journey,” Sayre Collection, box 3, folder 6. To complicate her efforts, Franklin was initially denied a visa by the American consul in London “owing to a misunderstanding about whether she was to be compensated over and above her expenses” (ibid.). On the effort to schedule lectures see Franklin to Stanley, 7 May 1954, Stanley Papers, carton 8, folder Franklin, Rosalind. According to Judson, Crick helped facilitate contacts for her schedule of talks; see Judson, Eighth Day of Creation (cit. n. 10), p. 268.

  • 47 On the planned visit to the Virus Laboratory see Franklin to Stanley, 6 July 1954, Stanley Papers, carton 8, folder Franklin, Rosalind. For Watson's report see Watson to Francis Crick, 15 Oct. 1954, Francis Crick Papers, posted on National Library of Medicine, Profiles in Science: http://profiles.nlm.nih.gov/SC/B/B/J/Q/_/scbbjq.pdf; this meeting with Watson is also described in Maddox, Rosalind Franklin, pp. 240–241.

  • 48 Franklin to Stanley, 14 Oct. 1954, Stanley Papers, carton 8, folder Franklin, Rosalind.

  • 49 On the formation and early accomplishments of the German group see Hans-Jörg Rheinberger, “Virusforschung an den Kaiser-Wilhelm-Instituten für Biologie und Biochemie, 1937–1945,” in Epistemologie des Konkreten: Studien zur Geschichte der modernen Biologie (Frankfurt am Main: Suhrkamp, 2006), pp. 185–218; Christina Brandt, Metapher und Experiment: Von der Virusforschung zum genetischen Code (Göttingen: Wallstein, 2004); and Jeffrey Lewis, “From Virus Research to Molecular Biology: Tobacco Mosaic Virus in Germany, 1936–1956,” J. Hist. Biol., 2004, 37:259–301. For an account of the rivalry from the Berkeley side see Creager, Life of a Virus, pp. 251–273. The Max Planck Institute for Virus Research gained independent status in 1954; earlier in the postwar period it was a Division of Virus Research within the Institute for Biochemistry.

  • 50 When appealing for funding, however, Franklin argued that her group's reliance for material on biochemists in other countries was “a most unsatisfactory situation” and reflected the inadequacies of virus research in England in the mid-1950s; see Rosalind Franklin, “X-ray Diffraction and the Structure of Viruses,” 17 Oct. 1955, appended to “Note on the Future of the A.R.C. Research Group in Birkbeck College Crystallography Laboratory,” 9 Mar. 1956, Franklin Papers, FRNK 2/36.

  • 51 Rosalind E. Franklin and Barry Commoner, “Abnormal Protein Associated with Tobacco Mosaic Virus: X-ray Diffraction by an Abnormal Protein (B8) Associated with Tobacco Mosaic Virus,” Nature, 1955, 175:1077–1082. By “abnormal,” the authors meant that the protein was not found in uninfected tobacco leaves but was due to the presence of the virus.

  • 52 Pirie criticized Franklin severely for accepting that the 3,000 Å viral rods were not artifacts and for assuming that there was only one type of protein subunit: N. W. Pirie to Franklin, 6 Dec. 1954, Franklin Papers, FRNK 2/33. On Pirie's influence see Maddox, Rosalind Franklin, pp. 251–253.

  • 53 See Watson to Franklin, 3 Dec. 1954, and Crick to Franklin, 8 Dec. 1954: Franklin Papers, FRNK 2/33; and Rosalind E. Franklin, “Structure of Tobacco Mosaic Virus,” Nature, 1955, 175:379–381, on p. 380. In this paper she cited biochemical studies of TMV subunits from both the Berkeley Virus Lab and the Max Planck Institute for Virus Research in Tübingen, as well as work from the laboratories of William N. Takahashi (University of California, Berkeley), Commoner (Washington University, St. Louis), and Raymond Jeener (University of Brussels) on a low-molecular-weight virus-like protein from infected plants that polymerizes into rods. Franklin may have gotten the idea that the subunits were divided into two from Crick: Donald Caspar, personal communication to Morgan, 14 Nov. 2007.

  • 54 Franklin, “Structure of Tobacco Mosaic Virus,” p. 381. The helical grooving also helped account for the unusually close packing of TMV in dry gels. As Franklin explained, “there are holes between the particles in the strongly dried material, which is inevitable if the particle contour is helically grooved” (ibid., p. 380). See also Rosalind Franklin and Aaron Klug, “The Nature of the Helical Groove on the Tobacco Mosaic Virus Particle,” Biochim. Biophys. Acta, 1956, 19:403–415.

  • 55 Schramm himself used this metaphor of the candlewick; see Gerhard Schramm, “Neuere Untersuchungen über die Struktur des Tabakmosaikvirus und ihre biologische Bedeutung,” Zentralblatt für Bakteriologie, Parasitenkunde, Infektionskrankheiten und Hygiene, Sect. 2, 1956, 109:322–324, on p. 322.

  • 56 This was the one aspect of Franklin's structure that Pirie found plausible (on his objections see note 52, above): Pirie to Franklin, 6 Dec. 1954, Franklin Papers, FRNK 2/33.

  • 57 As Franklin wrote Watson in June 1954: “My measurements of the innermost reflection of each layer line agree very well with yours (except for the 31st which is definitely split …). So the dimensions you give for the outermost helix are likely to turn up again in my work, but I'm hoping the measurements over the whole photograph will tell us something about the ‘stuffing’ of the rod.” Franklin to Watson, 4 June [1954], James D. Watson Papers, Cold Spring Harbor Laboratory Archives, Cold Spring Harbor, New York (hereafter cited as Watson Papers).

  • 58 “Experience in the past has shown that it is rash to include a drawing with speculation features. It turns up for years and years, and one's reservations get lost in the process”: Crick to Franklin, 8 Dec. 1954, Franklin Papers, FRNK 2/33. His appeal for caution apparently did not deter Franklin.

  • 59 Watson to Franklin, 3 Dec. 1954 (with Franklin's annotations), 28 Feb. 1955, Franklin Papers, FRNK 2/33.

  • 60 Caspar to Franklin, 1 Nov. 1954, Franklin Papers, FRNK 2/33. Regarding Caspar's Ph.D. research and his move to Caltech see Caspar to George Beadle, 16 Dec. 1953; Caspar to Beadle, 10 June 1954; and Caspar to David Powell, 30 July 1954: Biology Division Papers, California Institute of Technology Archives, Pasadena (hereafter cited as Biology Division Papers), 21.23. According to Caspar's letter to Powell, he expected to finish his dissertation by the fall of 1954, although the Ph.D. from Yale would be dated June 1955. In the end, having collected data in 1953 and 1954, Caspar continued working on the dissertation after going to California, and the bibliography included papers from early 1955, including Franklin's paper in Nature, “Structure of Tobacco Mosaic Virus” (cit. n. 53). Caspar's work at Caltech was supported by a U.S. Public Health Service Fellowship.

  • 61 On the development of Perutz's technique see de Chadarevian, Designs for Life, pp. 125–126.

  • 62 It was some time before the X-ray diffraction data was taken as more definitive on the issue of RNA location; in Feb. 1955 Watson wrote Franklin that he had seen electron micrographs from Stanley's Virus Lab (taken by Roger Hart and Robley Williams) that “definitely establish that the RNA forms a central core of diameter somewhere between 30 Å and 50 Å”: Watson to Franklin, 28 Feb. 1955, Franklin Papers, FRNK 2/33. This followed up Watson's letter dated 3 Dec. 1954 cited in note 59, above.

  • 63 Both virologists in the 1950s and those working today use acronyms extensively to refer to the viruses, as do we, but in the case of Tomato bushy stunt virus we have departed from the currently accepted TBSV to use BSV, in accordance with the usage of researchers in the 1950s.

  • 64 See Biology Division Annual Report, 1955, Biology Division Papers, 21.23; and Watson to Franklin, 28 Feb. 1955, Franklin Papers, FRNK 2/33.

  • 65 Watson to Franklin, 9 Apr. 1955; and Franklin to Watson, 10 June 1955: Franklin Papers, FRNK 2/33. On use of the Spinco centrifuge see Watson, Genes, Girls, and Gamow, p. 130. A few pages later, Watson describes how he and Caspar visited Stanley's Virus Lab to give a joint talk and to use Robley Williams's RCA electron microscope in an unsuccessful attempt to visualize purified TMV RNA.

  • 66 Caspar to Franklin, 1 Nov. 1954; and Franklin to Caspar, 8 Nov. 1954, 28 June 1955: Franklin Papers, FRNK 2/33. Because Caspar's TMV equatorial diffraction data were centrosymmetric, the phases of the diffracted X-rays change sign in many of the valleys separating the peaks of the intensity curve. His assignment of signs was confirmed by Franklin when she discovered that in the high-resolution, low-angle diffraction data there was a small peak in the data between the first two subsidiary intensity maxima.

  • 67 Caspar to Franklin, 9 Apr. 1955; and Franklin to Caspar, 19 May 1955: Franklin Papers, FRNK 2/33.

  • 68 In contrast to TMV, one could obtain true three-dimensional crystals from spherical plant viruses such as TYMV and analyze their X-ray diffraction patterns using crystallographic methods rather than those of fiber diffraction. Finch switched projects prompted by Caspar's discovery of TYMV crystals in Harry Carlisle's refrigerator (see below). Bernal's role as advisor to Holmes and Finch is noted in Brown, J. D. Bernal (cit. n. 38), p. 356. For more on the history of spherical virus crystallography see Gregory J. Morgan, “Historical Review: Viruses, Crystals, and Geodesic Domes,” Trends in Biochemical Sciences, 2003, 28:86–90; Morgan, “Early Theories of Virus Structure,” in Conformational Proteomics of Macromolecular Architecture, ed. R. Holland Cheng and Lena Hammar (Singapore: World Scientific, 2004), pp. 3–40; Morgan, “Virus Design, 1955–1962: Science Meets Art,” Phytopathology, 2006, 96:1287–1291; and Morgan, “Why There Was a Useful Plausible Analogy between Geodesic Domes and Spherical Viruses,” Hist. Phil. Life Sci., 2006, 28:215–236.

  • 69 See Maddox, Rosalind Franklin, pp. 251–252, 256, 263. This was despite the remarkable results Franklin and Klug had obtained together. Especially impressive was their demonstration that the number of subunits per three turns of the helix varied slightly—by hundredths of a subunit—among different strains of the virus. See Rosalind E. Franklin and A. Klug, “The Splitting of Layer Lines in X-ray Fibre Diagrams of Helical Structures: Application to Tobacco Mosaic Virus,” Acta Crystallog., 1955, 8:777–780.

  • 70 Aaron Klug manuscript, “Bernal and Virus Research at Birkbeck,” p. 6, Franklin Papers, FRNK 2/37; and Rosalind E. Franklin, “Structural Resemblance between Schramm's Repolymerized A-Protein and Tobacco Mosaic Virus,” Biochim. Biophys. Acta, 1955, 18:313–314. Schramm's method involved disaggregating the TMV particles in alkali, separating the protein and nucleic acid components with electrophoresis, and then reaggregating the protein component in a mild acid. See Gerhard Schramm, “Über die Spaltung des Tabakmosaikvirus und die Wiedervereinigung der Spaltstücke zu höhermolekularen Proteinen, II: Versuche zur Wiedervereinigung der Spaltstücke,” Zeitschrift für Naturforschung, 1947, 2(B):249–257.

  • 71 Crick to Franklin, 3 June 1955, Franklin Papers, FRNK 2/33. Watson expressed an interest in PVX in Watson to Crick, 27 May 1955, Crick Papers, National Library of Medicine, Profiles in Science: http://profiles.nlm.nih.gov/SC/B/B/J/J/_/scbbjj.pdf.

  • 72 Franklin to Crick, 6 June 1955, Franklin Papers, FRNK 2/33; Watson, Genes, Girls, and Gamow, p. 180; and Franklin to Paul Kaesberg, 18 July 1955, Franklin Papers, FRNK 2/33.

  • 73 Interview, Morgan with Donald Caspar, Tallahassee, Fla., 4–5 Dec. 1998. Caspar recalls that together they attended the opening performance of the Japanese Azuma Kabuki dancers and musicians in Covent Garden.

  • 74 Watson, Genes, Girls, and Gamow, p. 183 (evidence of five-fold symmetry); and personal communication, Caspar to Morgan, 18 Apr. 2003. As Caspar more recently described the cache in Carlisle's fridge, “Finding the sparkling BSV and TYMV crystals was like finding a hoard of diamonds in a secret cavern.” Franklin agreed that Caspar could have the BSV crystals he found at Birkbeck. Personal communication, Caspar to Morgan, 14 Nov. 2007.

  • 75 Watson, Genes, Girls, and Gamow, p. 188; and Watson to Christa Mayr, undated [Nov. 1955] letter circulated with an advance copy of Watson's Genes, Girls, and Gamow (obtained courtesy of Donald Caspar).

  • 76 Personal communication, Caspar to Morgan, 18 Apr. 2003 (burnt jacket). Caspar expressed his take on the division of labor: “My memory and Aaron Klug's memory is somewhat different from Jim Watson's memory. What we had decided was that Rosalind and Aaron would work on Turnip yellow mosaic virus crystals and I would carry on with work on the BSV crystals.” Interview, Morgan with Caspar, 4–5 Dec. 1998. Aaron Klug confirms Caspar's memory: communication with Morgan, 16 Aug. 2007.

  • 77 Several historians of science have elaborated on how moral economies of science function, taking inspiration in various ways from E. P. Thompson's classic usage: “The Moral Economy of the English Crowd in the Eighteenth Century” and “The Moral Economy Reviewed,” in E. P. Thompson, Customs in Common (New York: New Press, 1991), pp. 185–258, 259–351. See, e.g., Steven Shapin, A Social History of Truth: Civility and Science in Seventeenth-Century England (Chicago: Univ. Chicago Press, 1994); Lorraine Daston, “The Moral Economy of Science,” Osiris, N.S., 1995, 10:3–24; and Robert E. Kohler, “Moral Economy, Material Culture, and Community in Drosophila Genetics,” in The Science Studies Reader, ed. Mario Biagioli (New York: Routledge, 1999), pp. 243–257. We draw predominantly on Kohler's model, given our interest in the circulation of materials and allocation of credit. Along these lines see also Anderson, “Possession of Kuru” (cit. n. 13); and Paula Findlen, “The Economy of Scientific Exchange in Early Modern Italy,” in Patronage and Institutions: Science, Technology, and Medicine at the European Court, 1500–1750, ed. Bruce T. Moran (Rochester, N.Y.: Boydell, 1991), pp. 5–24. Needless to say, our pointing to a moral economy at work does not imply that all of the participants were virtuous or exemplary; Watson and Crick's failure fully to credit Franklin's experimental data on DNA in their 1953 paper was an abrogation of the usual conventions even at the time and remains a troubling aspect of the history.

  • 78 Caspar had sent Franklin his Ph.D. dissertation in June 1955, before he went to England.

  • 79 Franklin to Pirie, 3 Feb. 1956, Franklin Papers, FRNK 2/33. Information regarding Franklin's initial drafting of Caspar's paper comes from interview, Morgan with Caspar, 4–5 Dec. 1998. See also Franklin to Watson, 10 Feb. 1956, Jeremy Norman Collection (this collection—hereafter cited as Norman Collection—was in San Francisco when Morgan used it but has since been auctioned); Watson quotes her cover letter in full in Genes, Girls, and Gamow, p. 203. The published versions of the papers are Donald L. D. Caspar, “Radial Density Distribution in the Tobacco Mosaic Virus Particle,” Nature, 1956, 177:928; and Rosalind E. Franklin, “Location of the Ribonucleic Acid in the Tobacco Mosaic Virus Particle,” ibid., pp. 928–930.

  • 80 Watson, Genes, Girls, and Gamow, pp. 189–190. On Rich's stay in Cambridge and his work there with Crick on the triple-helical structure of collagen see Alexander Rich, “Fifty Years with Double-Stranded RNA,” Scientist, 2006, 20:34–39.

  • 81 Watson, Genes, Girls, and Gamow, pp. 202–203.

  • 82 F. H. C. Crick and J. D. Watson, “Structure of Small Viruses,” Nature, 1956, 177:473–475, on p. 473; D. L. D. Caspar, “Structure of Bushy Stunt Virus,” ibid., pp. 475–476; and Crick, as quoted by Aaron Klug in his historical introduction to “Session I: Particle Structure,” at “Symposium on Tobacco Mosaic Virus: Pioneering Research for a Century,” sponsored by the Royal Society of Edinburgh in association with the Royal Society of London, 7 Aug. 1998, Edinburgh. Klug dates the remark to the late 1950s: personal communication to Creager, 17 Apr. 2001. Caspar dates Crick's remark to around 1955; see D. L. D. Caspar, “Movement and Self-Control in Protein Assemblies: Quasi-Equivalence Revisited,” Biophysical Journal, 1980, 32:103–138.

  • 83 A few familiar objects possess the same kind of symmetry. E.g., a soccer ball constructed from twelve black pentagons and twenty white hexagons has 532 symmetry: five-fold symmetries though the centers of the pentagons, three-fold symmetries through the centers of the hexagons, and two-fold symmetries through the edges between the hexagons. Dorothy Hodgkin had earlier speculated that BSV has cubic symmetry but had not drawn general conclusions; see Dorothy Crowfoot Hodgkin, “X-ray Analysis and Protein Structure,” Cold Spring Harbor Symposia on Quantitative Biology, 1950, 14:65–78. For more detail on Caspar's work see Morgan, “Early Theories of Virus Structure” (cit. n. 68).

  • 84 Crick and Watson, “Structure of Small Viruses” (cit. n. 82), p. 474; and Caspar, “Structure of Bushy Stunt Virus” (cit. n. 82), p. 476. These results are not inconsistent if each structural subunit consists of multiple chemical subunits.

  • 85 Franklin, “Note on the Future of the A.R.C. Research Group in Birkbeck College Crystallography Laboratory,” 9 Mar. 1956, Franklin Papers, FRNK 2/36.

  • 86 In the letter quoted above concerning the incident of Caspar and the TYMV crystals, Watson stated: “This morning was spent persuading Victor Rothschild that Rosy should be supported in spite of her continued insults to the administrative heads of the Agriculture Research Council. This weekend I must write a long report to be submitted to ARC, otherwise she will hopelessly flounder without adequate financial support.” Watson to Mayr, undated [Nov. 1955] letter circulated with an advance copy of Watson's Genes, Girls, and Gamow. In his autobiography, Watson describes how he spoke on behalf of Franklin's needs to Rothschild earlier that summer as well and reprints a letter he wrote to Franklin with advice on how to secure funds for a much-needed upgrade to her crystallography equipment: Watson, Genes, Girls, and Gamow, pp. 154–156.

  • 87 Williams to Crick, 20 Sept. 1957, Williams Papers, carton 4, folder C. On the grant from the Public Health Service see Maddox, Rosalind Franklin, pp. 290–292. Williams provided advice to Franklin on her application. See Franklin to Williams, 19 Oct. 1956; Williams to Franklin, 24 Oct. 1956; and Franklin to Williams, 6 Dec. 1956: Williams Papers, carton 4, folder F. Franklin immediately used the American funds to pay for a research assistant at the Molteno Institute to grow and purify more TYMV for crystals, which were in short supply: Franklin to Kenneth Smith, 17 July 1957, and Smith to Franklin, 16 Oct. 1957, Norman Collection. Franklin had requested a letter from Watson supporting her application to the U.S. Public Health Service; her letter to Watson indicates how instrumental Williams was in helping her apply successfully and how crucial the American grant was to enable her and Klug to continue working together. She also mentioned that, since her return from the United States, “I've spent all my time being ill”: Franklin to Watson, 14 Nov. [1956], Watson Papers.

  • 88 Franklin and Holmes looked for the influence of the mercury atom on the outer parts of the zero layer line and were able to infer that n = 16 was the correct solution to Watson's equation (i.e., 3n + 1 = 49): Kenneth C. Holmes, personal communication to Morgan, 24 Dec. 2006. See Rosalind E. Franklin and Kenneth C. Holmes, “Tobacco Mosaic Virus: Application of the Method of Isomorphous Replacement to the Determination of the Helical Parameters and Radial Density Distribution,” Acta Crystallog., 1958, 11:213–220. On the state of knowledge in 1956 see Franklin to Pirie, 3 Feb. 1956, Franklin Papers, FRNK 2/33. One missing piece of information was that the handedness of the TMV helix was not revealed by diffraction data.

  • 89 Watson to Stanley, 27 Jan. 1956, Stanley Papers, carton 13, folder Wa Misc. There are reasons to be cautious about taking this letter at face value, even though we find Watson's assent to Franklin's standing in the field to be notable. Stanley was a major power broker in the field, and Franklin was collaborating closely with members of his lab. At the same time, Watson did not find Stanley's Virus Lab very scientifically stimulating; see Watson to Crick, 11 Dec. 1954, Crick Papers, National Library of Medicine, Profiles in Science: http://profiles.nlm.nih.gov/SC/B/B/J/N/_/scbbjn.pdf. Moreover, he had been recommended for an opening there but failed to receive an offer; see Watson to Gunther S. Stent, 9 Dec. 1954, Gunther S. Stent Papers, Bancroft Library, University of California, Berkeley, 99/149z, box 15, folder Watson, J. D.

  • 90 On the symposium as an effort to revivify virology in England see G. E. W. Wolstenholme to Franklin, 21 June 1955, Franklin Papers, FRNK 2/34; and Frank Macfarlane Burnet to Lady Burnet, 13 Mar. 1956, Frank Macfarlane Burnet Papers, University of Melbourne Archives, 2/18. For Franklin's attempt to include Schramm see Franklin to Wolstenholme, 28 June 1955, Franklin Papers, FRNK 2/34. The Ciba/Novartis scrapbook (held at the Ciba Foundation, London) has a printed list of participants in which Wilkins's name is crossed out and Klug's is written in by hand.

  • 91 “It now begins to appear (Fraenkel-Conrat, 1956) that infection can be obtained from solutions in which essentially no full-length TMV particles, either native or reconstituted, are present. The active solutions are believed to be pure RNA, and are infectious when rubbed upon tobacco plants in sufficiently high concentrations”: Robley C. Williams, “Structure and Substructure of Viruses as Seen under the Electron Microscope,” in Ciba Foundation Symposium on the Nature of Viruses, ed. G. E. W. Wolstenholme and Elaine C. P. Millar (Boston: Little, Brown, 1957), pp. 19–33, on p. 31. The observation by Heinz Fraenkel-Conrat appeared in “The Role of the Nucleic Acid in the Reconstitution of Active Tobacco Mosaic Virus,” J. Amer. Chem. Soc., 1956, 78:882–883; but Alfred Gierer and Gerhard Schramm made the same discovery independently and gave it greater prominence in “Infectivity of Ribonucleic Acid from Tobacco Mosaic Virus,” Nature, 1956, 177:702–703.

  • 92 By the same token, Watson has described a hoax he played on Williams, falsifying a telegram from Stanley conveying the news that the TMV protein was infectious; Watson claims that Williams downplayed his new results with nucleic acid in his Ciba Foundation symposium talk as a result: Watson, Genes, Girls, and Gamow, p. 217.

  • 93 F. C. Bawden, in discussion following Williams, “Structure and Substructure of Viruses as Seen under the Electron Microscope,” in Ciba Foundation Symposium on the Nature of Viruses, ed. Wolstenholme and Millar (cit. n. 91); the remark appears on p. 35. Pirie argued in the same discussion that Williams should not restrict his search for TMV particles to those 2,000–3,000 Å long, as he might be missing smaller infective units (ibid., p. 36).

  • 94 Watson, Passion for DNA (cit. n. 40), p. 28. This tenet was codified by Crick as the “Central Dogma.” See Judson, Eighth Day of Creation (cit. n. 10), pp. 332–336; and Strasser, “World in One Dimension” (cit. n. 11).

  • 95 Rosalind E. Franklin, A. Klug, and K. C. Holmes, “X-ray Diffraction Studies of the Structure and Morphology of Tobacco Mosaic Virus,” in Ciba Foundation Symposium on the Nature of Viruses, ed. Wolstenholme and Millar (cit. n. 91), pp. 39–52.

  • 96 F. H. C. Crick and J. D. Watson, “Virus Structure: General Principles,” in Ciba Foundation Symposium on the Nature of Viruses, ed. Wolstenholme and Millar, pp. 5–13, on p. 12.

  • 97 Watson to Crick, 10 Feb. 1955, Crick Papers, National Library of Medicine, Profiles in Science: http://profiles.nlm.nih.gov/SC/B/B/J/L/_/scbbjl.pdf.

  • 98 Maddox, Rosalind Franklin, p. 268.

  • 99 See Franklin to Klug, 21 June 1956, Norman Collection; and Maddox, Rosalind Franklin, pp. 277–280. Whereas Franklin's hosts at Caltech had her out for dinner and even took her on a camping trip in the mountains, which she relished, she found her collaborators at the Virus Lab to be less sociable.

  • 100 Franklin to Stanley, 4 July 1956, Stanley Papers, carton 8, folder Franklin, Rosalind. Franklin's thank-you note to Stanley, written on 30 Aug. (and in the same folder of the Stanley Papers), makes it clear that she stayed in Berkeley for three weeks.

  • 101 “The most important thing here is that I've recently seen some electron micrographs of TYM by Steere's freezing-shadowing replica technique which show a magnificently clear surface structure. This only happened today, and seems very exciting, so excuse the muddle. The prominent feature is an array of six knobs on each particle around a central one. … The inter-knob distance is ∼1/4 inter-particle distance, which is consistent with your 60 Å, but the thing does not look 5 folded. … If it is not 5-folded, the question arises, was the five foldedness in the RNA. … Believe it or not, he [Robley Williams] says he had a slide of this with him at Ciba but did not show it in case the effect was due to amm. sulphate!!” Franklin to Klug, 27 July 1956, Norman Collection.

  • 102 Franklin to Klug, 30 July 1956, Norman Collection. A cubeoctahedron has 432 symmetry, not the 532 symmetry of an icosahedron such as BSV. A week later, Franklin was still thinking about Steere's electron micrographs, but now from the point of view that the photographs pertained to the determination of the unit cell of the crystal: Franklin to Klug, 5 Aug. 1956, Norman Collection.

  • 103 Crick to Klug, 14 Dec. 1956, Norman Collection; A. Klug, J. T. Finch, and Rosalind E. Franklin, “Structure of Turnip Yellow Mosaic Virus,” Nature, 1957, 179:683–684; and Klug, Finch, and Franklin, “The Structure of Turnip Yellow Mosaic Virus: X-ray Diffraction Studies,” Biochim. Biophys. Acta, 1957, 25:242–252. For more on the symmetry of TYMV see A. Klug and J. T. Finch, “The Symmetries of the Protein and Nucleic Acid in Turnip Yellow Mosaic Virus: X-Ray Diffraction Studies,” J. Molec. Biol., 1960, 2:201–215.

  • 104 On the history of research on cytoplasmic particles see Nicolas Rasmussen, “Mitochondrial Structure and the Practice of Cell Biology in the 1950s,” J. Hist. Biol., 1995, 28:381–429; Hans-Jörg Rheinberger, “Comparing Experimental Systems: Protein Synthesis in Microbes and in Animal Tissue at Cambridge (Ernest F. Gale) and at the Massachusetts General Hospital (Paul C. Zamecnik), 1945–1960,” ibid., 1996, 29:387–416; Rheinberger, “Cytoplasmic Particles in Brussels (Jean Brachet, Hubert Chartrenne, Raymond Jeener) and at Rockefeller (Albert Claude), 1935–1955,” Hist. Phil. Life Sci., 1997, 19:47–67; and Rheinberger, Toward a History of Epistemic Things: Synthesizing Proteins in the Test Tube (Stanford, Calif.: Stanford Univ. Press, 1997).

  • 105 Watson, “Involvement of RNA in the Synthesis of Proteins” (cit. n. 39), p. 787; Klug to Bernal, 31 May 1956, Norman Collection (emphasis in original); and Franklin to Klug, 17 July 1956, Norman Collection (interest of Caspar, Crick, and Watson). On the analogy between spherical viruses and microsomes see Crick and Watson, “Virus Structure” (cit. n. 96), p. 12.

  • 106 Franklin to Klug, 27 July 1956, Norman Collection (regarding Franklin's California connections for obtaining material); and Rosalind E. Franklin, A. Klug, J. T. Finch, and K. C. Holmes, “On the Structure of Some Ribonucleoprotein Particles,” Discussions Faraday Soc., 1958, 25:197–198.

  • 107 F. L. Schaffer and C. E. Schwerdt, “Crystallization of Purified MEF-1 Poliomyelitis Virus Particles,” Proc. Nat. Acad. Sci., USA, 1955, 41:1020–1023. For a discussion of their work see Creager, Life of a Virus, Ch. 5. Polio crystals were discussed at the International Poliomyelitis Conference in Geneva in July 1957, where Franklin heard Schwerdt give a talk on the crystals he had grown: Franklin notebook entry, 10 July 1957, Franklin Papers, FRNK 3/14.

  • 108 Interview, Morgan with John Finch, Cambridge, 18 July 2000.

  • 109 For correspondence regarding the attempts to mount polio crystals see Franklin to Caspar, 16 Mar. 1958; Franklin to Bawden, 20 Mar. 1958; and Franklin to R. W. Douglas, 24 Mar. 1958: Franklin Papers, FRNK 2/33. The results were published in J. T. Finch and A. Klug, “Structure of Poliomyelitis Virus,” Nature, 1959, 183:1709–1714.

  • 110 W. L. Bragg to Crick, 26 June 1956, William Lawrence Bragg Papers, Royal Institution, London (hereafter cited as Bragg Papers), 83P/1. Bragg was appointed by the president of the Royal Society (Cyril Hinshelwood) to serve as the U.K. representative on the Scientific Committee for the Brussels exhibition: Bragg Papers, 83G/1, 2, 3. Bragg also wrote Frederick Sanger about his interest in including his work on insulin chains and to Perutz about his wish to include X-ray crystallographic work on proteins and Huxley's work on muscle: Bragg Papers, 83P/3, 84A/1.

  • 111 E.g., Crick noted in his response to Bragg's letter: “Miss Franklin will cover viruses and Wilkins will, I think, be responsible for DNA. I should be quite happy to look after collagen, but I think that to collaborate with King's on this would cause unnecessary friction.” Crick to Bragg, 8 Dec. 1956, Bragg Papers, 83P/37.

  • 112 See Bragg to Franklin, 26 June 1956, Bragg Papers, 85B/164; Franklin to Bragg, 23 July 1956, Bragg Papers, 85B/165; and copies of correspondence in the same box between Franklin and D. C. Phillips. This TMV model, whose design and construction were overseen by Klug, ended up at the Cambridge MRC Laboratory.

  • 113 Rosalind E. Franklin, D. L. D. Caspar, and A. Klug, “The Structure of Viruses as Determined by X-ray Diffraction,” in Plant Pathology: Problems and Progress, 1908–1958, ed. C. S. Holton et al. (Madison: Univ. Wisconsin Press, 1959), pp. 447 461; and Caspar and Klug, “Physical Principles in the Construction of Regular Viruses,” Cold Spring Harbor Symp. Quant. Biol., 1962, 27:1–24. On the development of this theory see Morgan, “Early Theories of Virus Structure” (cit. n. 68).

  • 114 Aaron Klug, “From Macromolecules to Biological Assemblies,” in Nobel Lectures: Chemistry, 1981–1990, ed. Tore Frängsmyr and Bo G. Malmström (Singapore: World Scientific, 1992), pp. 77–109, on p. 79; also quoted in Maddox, Rosalind Franklin, p. 325. Klug was the sole recipient of that year's prize in chemistry; he felt Franklin would have been recognized earlier had she lived.

  • 115 As Maddox recounts and others have noted, Franklin went to the Cricks' home in Oct.–Nov. 1956 to recover from a second surgery rather than continuing to stay with her family: Maddox, Rosalind Franklin, p. 289.

  • 116 The characterization of Franklin as so cautious an experimentalist that she distrusted both model building and theoretical intuition has been prominent in the historiography since Watson's Double Helix. Needless to say, the depiction draws on gender stereotypes by portraying a woman scientist as attentive to detail, patient, unoriginal, and intellectually timid. In his account of Franklin, Robert Olby points to J. D. Bernal's influence on her. Bernal criticized Linus Pauling's “deductive” model building, arguing that scientists should rely on inductive methods such as “deriving chain types from Patterson sections”: Olby, Path to the Double Helix (cit. n. 18), p. 374. This observation explains Franklin's choice of approach and her alleged antihelical stance in 1952, but it does not seem fully to account for the repeated claims that she lacked or distrusted intuition. At the end of her biography, Maddox offers an astute analysis of how the image of Franklin as meticulous and unimaginative gained traction; this characterization was offered by Crick as well as by Watson, and the depiction works to excuse both of them for using her data by suggesting that she did not seem to know how to interpret it herself. See Francis Crick, “How to Live with a Golden Helix,” Sciences, 1979, 19(7):6–9. Maddox points to Franklin's work on coal and on viruses as evidence to the contrary; we feel the evidence is even stronger for TMV than Maddox suggests. See Maddox, “Epilogue: Life after Death,” in Rosalind Franklin, pp. 311–328.

  • 117 Franklin's combination of institutional marginality and scientific achievement is reminiscent of that of her contemporary Barbara McClintock. For the definitive account of the institutional obstacles that women scientists faced in the mid-twentieth century United States see Margaret W. Rossiter, Women Scientists in America: Before Affirmative Action, 1940–1972 (Baltimore: Johns Hopkins Univ. Press, 1995). On McClintock see Evelyn Fox Keller, A Feeling for the Organism: The Life and Work of Barbara McClintock (San Francisco: Freeman, 1983); and Nathaniel C. Comfort, The Tangled Field: Barbara McClintock's Search for the Patterns of Genetic Control (Cambridge, Mass.: Harvard Univ. Press, 2001).

  • 118 Negotiations for this move predated Franklin's death and were motivated by Bernal's impending retirement. In addition to Klug, Finch, Holmes, and Reuben Leberman moved from Birkbeck to Cambridge; see de Chadarevian, Designs for Life, p. 252.

  • 119 Holmes, Meselson, Stahl, and the Replication of DNA (cit. n. 40), Ch. 1. On the gradual acceptance of Watson and Crick's model see also de Chadarevian, Designs for Life, esp. Ch. 6; and Robert Olby, “Quiet Debut for the Double Helix,” Nature, 2003, 421:402–405.

  • 120 Gierer and Schramm, “Infectivity of Ribonucleic Acid from Tobacco Mosaic Virus” (cit. n. 91). The TMV “reconstitution” experiments in Berkeley provided an elegant demonstration of the hereditary role of the RNA, since hybrids composed of RNA from one strain and protein from another always gave rise, once infected into a host, to the “parent” strain from which the nucleic acid had been derived. See Heinz Fraenkel-Conrat and Beatrice A. Singer, “Virus Reconstitution, II: Combination of Protein and Nucleic Acid from Different Strains,” Biochim. Biophys. Acta, 1957, 24:540–548; and Creager, Life of a Virus, Ch. 7.

  • 121 See Francis H. C. Crick, “Nucleic Acids,” Scientific American, 1957, 197(3):188–200; Kay, Who Wrote the Book of Life? (cit. n. 10), pp. 179–192; and Creager, Life of a Virus, Ch. 7. The other promising viral model was bacteriophage; Sydney Brenner and his colleagues at Cambridge used mutagens to introduce changes into the genes of phage T4 and follow their effects. See de Chadarevian, Designs for Life, pp. 195–198.

  • 122 See Watson, “Involvement of RNA in the Synthesis of Proteins” (cit. n. 39).

  • 123 Caspar would seem to be a good example of this, although he cautions against idealizing Franklin. He emphasizes that she was strongly focused on achieving her scientific goals and asserted herself as necessary, whereas his own attitude was less proprietary: Caspar, personal communication to Morgan, 14 Nov. 2007. In the end, we see her single-mindedness as consistent with her ability to make effective use of collaboration.

  • 124 P. Jonathan G. Butler and Aaron Klug, “The Assembly of a Virus,” Sci. Amer., 1978, 239(5):62–69; and Morgan, “Early Theories of Virus Structure” (cit. n. 68).

  • 125 See Judson, Eighth Day of Creation (cit. n. 10), Chs. 5–8; and Rheinberger, Toward a History of Epistemic Things (cit. n. 104), esp. Chs. 10, 12, 13.

  • 126 De Chadarevian, Designs for Life, Chs. 6, 8. On the popular uses of the double helix see Dorothy Nelkin and M. Susan Lindee, The DNA Mystique: The Gene as a Cultural Icon (New York: Freeman, 1995).

© 2008 by The History of Science Society. All rights reserved. 0018-9902/2008/9902-0001$10.00