Plaster Cast Publishing in 19th Century Paleontology

580954 HOS0010.1177/0073275315580954History of ScienceRieppel earch-article2015 HOS Article Plaster cast publishing in nineteenth-century paleontology History of Science 2015, Vol. 53(4) 456–491 © The Author(s) 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0073275315580954 hos.sagepub.com Lukas Rieppel Department of History, Brown University, USA Abstract This article uses the example of Hesperornis regalis, an ancient toothed bird discovered in Kansas during the 1870s, to discuss a practice that became extremely widespread in late-nineteenth-century paleontology: the use of plaster cast replicas to circulate especially noteworthy discoveries. Building upon a growing literature at the intersection of book history and the history of science, I argue that paleontologists developed plaster casts as a compromise medium that combined some features of print with others from original natural history specimens. For example, because plaster casts were mechanical reproductions, they could be fabricated in fairly large numbers at a relatively modest price. This allowed them to circulate more readily than original fossils. At the same time, however, paleontologists treated casts as a material trace of prehistory, much like an original fossil itself. As a result, they were seen as especially authoritative and trustworthy objects of knowledge. Insofar as they combined features of print and prehistory, I argue that casts functioned as a genuine mode of publication in late-nineteenth-century paleontology, but one that foregrounded its own materiality. Keywords plaster cast, paleontology, circulation, indexicality, book history, missing link Introduction “Specimens are like the types in a printing office.” George Brown Goode, Principles of Museum Administration, York: Coultas & Volans 1895, p. 37. Corresponding author: Lukas Rieppel, Brown University, Box N, 79 Brown Street, Providence, RI 02912, USA. Email: [email protected] Downloaded from hos.sagepub.com by guest on December 1, 2015 457 Rieppel Hesperornis regalis is an extraordinarily significant but surprisingly little known fossil bird. First discovered by the Yale University paleontologist Othniel Charles Marsh in western Kansas during the early-1870s, Hesperornis quickly gained widespread notoriety among learned naturalists across Europe and North America. The specimen’s outsized paleontological importance was due to a strange set of anatomical features it combined within the frame of a single animal body. Although Hesperornis resembled a grebe or a loon in its outward appearance, a more careful examination revealed far closer affinities to Struthioniform birds, an ancient lineage that includes the ostrich and moa. Even more surprising was the fact that Hesperornis had a row of sharp teeth embedded within a single groove on both the top and the bottom of its beak. This, in combination with a number of other discordant anatomical traits, placed it in an ambiguous taxonomic position between reptiles and birds. Coming to widespread attention only a few years after the celebrated discovery of Archaeopteryx, it ranked among the period’s most recognizable “missing links,” playing a lead role in latenineteenth-century debates about the theory of common descent.1 Without discounting its significance as a transitional fossil, this essay engages the specimen from a different perspective. I will use Hesperornis as a case study through which to explore a practice that became extraordinarily widespread in nineteenth-century paleontology, the use of plaster casts as a means of distributing information about a discovery. We know from the work of Pamela Smith and others that cast-making entered the history of science much earlier, as artistic practices became bound up with the generation of vernacular knowledge about nature in sixteenth- and seventeenth-century Europe.2 However, I would like to suggest there is something distinctive about the use of casts among nineteenth-century paleontologists. The casts that Smith describes were sought-after evidence of artisanal mastery, prized for their uncanny ability to mimic nature in lifelike detail. In contrast, nineteenth-century paleontologists endowed casts with epistemic authority precisely because they were not recognized as the product of artistic genius. On the contrary, their value derived from the cast’s status as a mechanical reproduction, and it was this much more mundane quality that allowed it to stand in for an original fossil. It is a truism among historians of the nineteenth-century that technological developments during this period caused a dramatic reduction in the effort required to traverse geographic space. Steam-powered railroads and ocean liners made it possible to travel great distances at unprecedented speed, while the telegraph allowed messages to cross entire continents or even an ocean within the blink of an eye. At the same time, the scientific community, which had prided itself on its cosmopolitanism going back to the republic of letters, embraced a set of related technological advances that led to an explosion in the production of print. As this happened, men of science increasingly came to rely on newly established genres of publication, especially the periodical, as a preferred medium for communication.3 The scientific journal therefore became the primary site in which the community registered, ratified, and certified new claims to natural knowledge.4 To be an object of knowledge increasingly meant having entered widespread circulation in print.5 These new technologies of circulation were especially important for the earth sciences. Paleontological knowledge about prehistory derived from material objects—fossils—that are extremely rare if not unique. This made it all but impossible for a practicing paleontologist to consult every last specimen with a bearing on some particular research question. Instead, paleontologists developed what Martin Rudwick has described as a Downloaded from hos.sagepub.com by guest on December 1, 2015 458 History of Science 53(4) “visual language” to reduce three-dimensional specimens to two-dimensional images, which circulated far more readily.6 For example, the late-eighteenth- and early-nineteenth-century savant Georges Cuvier assembled an impressive collection of two-dimensional “proxy specimens” into a veritable “paper museum” that was “far richer than even the rich collections of real specimens at the Muséum in Paris.”7 Within a few decades, it had become standard practice for paleontologists to publish exquisitely detailed lithographic plates for precisely this purpose. These usually featured a predictable composition showing several specimens seen from the top down, neatly arranged within a single rectangular border. Mimicking the experience of peering down into a museum drawer full of three-dimensional fossils, such illustrations have been likened to a “virtual specimen drawer,” providing access to rare and unusual objects among a far-flung community of practicing naturalists.8 At the same time, paleontology faced a daunting challenge in satisfying an epistemic regime in which knowledge was established via circulation. This is because the credibility of paleontological knowledge crucially rested on demonstrations of indexicality. The latter is a mode of representation that was famously distinguished from icons and signs by the American pragmatist philosopher Charles Sanders Peirce. Invoking the Latin word for a forefinger, Peirce defined indexical representations as those which point to their referent directly. In contrast, he understood icons—which he also called likenesses—to refer via similarity or verisimilitude, whereas signs were tied to their signified by mere social convention. In Peircean semiotics, topographical maps therefore functioned as iconic representations of a region’s physical geography, whereas the phrase “topographical map” was a sign for the same thing. Indexicals were further removed from our everyday notion of representation, but canonical examples included smoke (which points to a fire) and weathervanes (referencing which way the wind blows).9 In all three cases, the representation carries information about its referent, but for different reasons. Whereas signs must actively be endowed with a meaning, indexical representations have the power to refer on their own. Icons occupy a sort of middle ground, as their production may or may not involve deliberate human agency. In this way, Peirce’s schema bears a family resemblance relation to H.P. Grice’s more recent distinction between natural and non-natural meaning.10 For Grice, a representation x “naturally” means proposition p if, and only if, it entails p. In contrast, non-natural meaning is more complex and cannot be reduced to mere implication. For example, the presence of smoke warrants an inference about the existence of fire. However, things are trickier when we hear someone yell the word “fire,” because that person may be mistaken or they may even seek to deceive us. Of course, indexical representations can be manipulated to engineer erroneous beliefs. For example, a mischievous prankster might detonate a smoke bomb during a movie. However, in that case, our fear would derive from a failure of interpretation, not signification. The same is not true if someone were to yell the word “fire” in a crowded theatre. Paleontologists did not employ Peirce’s terminology explicitly. However, as I have argued in more detail elsewhere, they did value indexical over other kinds of representation.11 This situation should not come as too much of a surprise. After all, their status as a physical trace linked indexical representations to a widespread conception of scientific propriety that Lorraine Daston and Peter Galison have characterized as “mechanical Downloaded from hos.sagepub.com by guest on December 1, 2015 459 Rieppel objectivity.” During the mid- to late-nineteenth-century, the scientific community became increasingly anxious that far from providing access to nature, human subjectivity distorted our understanding of the world. The scientific community therefore came to prize representations whose production minimized or even eliminated the involvement of human consciousness. Because they were seen as entirely devoid of personal interests and psychological biases, machines embodied this new conception of epistemic virtue especially well. Consequently, the most reliable and trustworthy images came to be those that had been mechanically produced. “Wary of human mediation between nature and representation,” Daston and Galison explain, nineteenth-century scientists enlisted technologies of the index like “cameras, wax molds, and a host of other devices in a near-fanatical effort to” create representations of “birds, fossils, snowflakes, bacteria, human bodies, crystals, and flowers” that were certifiably free from the corrosive effects of “human interference.”12 While paleontologists were not uniquely enthralled by the virtues of mechanical objectivity, they did place an unusual value on the quality of indexicality. This is because they were only all too aware of the wide temporal and epistemic chasm that separated them from their objects of study. The extinct life forms that paleontologists studied inhabited our planet during a pervious geological epoch, making them inaccessible to direct observation. For this reason, they had to be approached indirectly.13 This posed a serious challenge to paleontology, opening up the possibility that its speculative theories were entirely unmoored from empirical reality. Anxious to avoid the impression they were trading in mere fantasy, paleontologists therefore took pains to stress the material basis of their discoveries, assiduously playing up the direct physical link that united extant fossils with flesh and blood creatures from the depths of time. In other words, fossils were seen as an acceptable stand-in for the actual past insofar as both were understood to hang at opposite ends of a long causal chain that was impervious to the corrupting influences of human consciousness.14 The same could not be said of a verbal description, a printed report, or a drawing, all of which were vulnerable to becoming untethered from the prehistoric world, representative of nothing so much as a naturalist’s fertile imagination. Hence, the tactile materiality of fossils was prized for its ability to anchor paleontological knowledge in physical traces from the actual past. However, their rare occurrence, extreme fragility, often prodigious size, and attendant weight all conspired to impede fossils’ ability to travel. Nineteenth-century paleontology thus found itself in the awkward position of embracing two epistemic ideals that effectively pulled in opposite directions from one another. Although knowledge claims were conventionally established via circulation, the accepted objects of paleontological knowledge—fossils—did not travel easily. This tension helps to explain why paleontology was and remains so closely tied to the museum, an institution whose exhibition hall makes objects that are recalcitrant to widespread circulation more widely accessible. However, with print achieving a near monopoly on scientific communication, paleontologists could hardly rely on demonstrations of indexicality in the museum alone. Thus, they struggled to find ways of providing a physical link between print and prehistory. One way to do so was by tying publications to individual specimens stored in a permanent repository. For example, when Marsh issued a definitive monograph on toothed birds in 1880, he made sure to advertise the fact that every fossil described and figured therein was held in storage at Yale’s Peabody Museum of Natural Downloaded from hos.sagepub.com by guest on December 1, 2015 460 History of Science 53(4) History. The implied audience could therefore not only engage in an act of “virtual witnessing” by reading the printed report, but readers could also exercise the option of consulting the physical objects described therein. Although most readers would not have been expected to go to such lengths, the fact they could do so in theory helped to endow the publication with an air of indexicality.15 Given its oft-cited indexicality, it is surprising that paleontologists did not make widespread use of photography until well into the twentieth century.16 One reason for this is that unless they are skillfully lit to provide the illusion of three-dimensional depth, photographic representations of fossils are often confusing and difficult to interpret. Hence, paleontologists generally preferred to use woodcuts, lithographs, and other representational media that could not only be carefully shaded but also stripped of extraneous detail. Of course, it was the ability to manipulate their content in precisely these ways that made such representations suspect as genuine objects of knowledge. For this reason, paleontologists worked hard to develop alternative visualization techniques. One strategy involved zooming out and investigating trends that obtained across an entire collection of fossils. As David Sepkoski has shown, nineteenth-century paleontologists developed a range of quantitative tools for the analysis of large datasets whose results could be presented as tables, spindle diagrams, diversity curves, and, eventually, survivorship curves.17 Much like the use of summary statistics, these practices sought to satisfy the demands of mechanical objectivity by replacing individual judgment with standardized and impersonal rule-following.18 However, quantitative representations of this sort were primarily used among invertebrate paleontologists, whereas vertebrate paleontologists dealt with material that was usually too rare for statistical treatment. It is in the context of this dilemma between the competing ideals of circulation and indexicality that I seek to situate the widespread practice of issuing plaster cast reproductions of especially sought-after fossils. Plaster casts, I would like to suggest, emerged as a sort of compromise medium, combining some features of print with others from original specimens. To the extent that their production was understood as a mechanical rather than a creative process, casts were trusted as legitimate objects of paleontological knowledge. Ideally, their manufacture simply consisted of adding another link in the causal chain that connected a fossil with its prehistoric antecedent. Of course, casts were always one step removed from the original, and therefore did not carry the same epistemic weight as an actual fossil. Neither, however, were they burdened by the same material weight, nor were they nearly as rare or as fragile. For these reasons, paleontologists preferred circulating plaster casts rather than original specimens. During the nineteenth century, museums therefore began issuing three-dimensional reproductions of their most auspicious discoveries. These were usually made available for purchase, but paleontologists also developed extensive exchange networks, treating a cast much like an offprint. Indeed, it is not going too far to say that plaster cast replicas came to function as a kind of publication in their own right. In this, they resembled the use of wax models in embryology at the time.19 However, there was also a crucial difference between the two: whereas the credibility of wax models rested on their accuracy, the credibility paleontological casts rested on their indexicality. Drawing on recent work at the intersection of the history of science and book history, this essay uses the case of Hesperornis regalis to argue for a new understanding of what Downloaded from hos.sagepub.com by guest on December 1, 2015 461 Rieppel natural history specimens were, how they were understood, and what they were expected to accomplish during a period that witnessed an explosion in scientific publishing.20 Museum studies scholars and historians of science often locate natural history specimens within an “exhibitionary complex” that also included world’s fairs, department store windows, and other forms of display that saturated late-nineteenth-century culture.21 This approach has a great deal to recommend it, and natural history specimens certainly did qualify as visual spectacles.22 However, as George Browne Goode’s somewhat fanciful remark quoted above illustrates, they were also analogized to the printed word.23 Such an alternative perspective, I would like to suggest, is especially appropriate for specimens such as Hesperornis that primarily circulated as plaster casts. However, this does not mean that casts simply and unproblematically functioned as a kind of “immutable mobile.”24 Just as naturalists faced a significant challenge in stabilizing the production, consumption, and circulation of print, so too did paleontologists struggle to control their plaster cast specimens.25 Hesperornis, an ancient toothed bird Let us begin with a copy of the specimen itself. The wood box pictured in Figure 1 contains plaster cast replicas of exactly 26 bones that once belonged to the extinct bird Hesperornis regalis. According to an inventory that accompanies the box, these include several vertebrae, the pelvis, femur, tibia, several digits, and portions of both the right and left mandible, one of which has three teeth still embedded within it. The latter, an explanatory label informs us, “were numerous” and “implanted in grooves” that ran along most of the animal’s long, tapering beak.26 Othniel Charles Marsh first discovered a few scattered remains of this creature in the Cretaceous deposits of western Kansas during the summers of 1870 and 1871. However, it was not until a recent Yale graduate who accompanied him on these expeditions—T.H. Russell—found a much more complete specimen in 1872 that Marsh realized the creature’s signal importance. This is because Russell’s find included portions of the animal’s skull, clearly showing a row of prominent teeth in the rear of its beak. Hesperornis therefore held out a tantalizing promise to cement the phylogenetic connection between reptiles and birds.27 Hesperornis soon rose to become one of the nineteenth-century’s most famous transitional fossils. Often described as a “missing link,” transitional fossils then as now played an important role in arguments for and against the theory of evolution. In a polemical essay published only a few years prior to Marsh’s discovery, Thomas Henry Huxley deployed them to great effect. Evolutionists, he begrudgingly admitted, faced a genuine difficulty in explaining why plants and animals that presumably evolved from a common ancestor are “marked out by natural intervals into sundry and very distinct groups?” The answer, he went on to say, was that the anatomical gaps separating the major branches of life are themselves a product of evolution. “In previous epochs of the world’s history,” Huxley speculated, these gaps did not exist, but the “connecting forms” that once bridged them “have died out.”28 To give his argument some empirical purchase, Huxley claimed that a careful examination of the fossil record helps to close the considerable distance that Downloaded from hos.sagepub.com by guest on December 1, 2015 462 History of Science 53(4) Figure 1. Plaster cast replica of Hesperornis regalis, distributed by Ward’s Natural Science Establishment, Rochester, New York. The specimen is stored in the archives of the Peabody Museum of Natural History, New Haven, CT. separates reptiles from birds. In particular, he called upon a spectacular Archaeopteryx fossil recently discovered in Germany to demonstrate that ancient birds possessed a number of reptilian features that have been lost in their modern descendants. Extinct dinosaurs such Compsognathus longipes showed that something of the reverse holds true as well: ancient reptiles looked a great deal more similar to modern birds than extant ones do. Huxley therefore concluded the science of paleontology was well on its way to establishing the precise manner in which “birds may have been evolved from reptiles.”29 Convincing as the Archaeopteryx was, not everyone agreed that it constituted a true missing link. Describing a remarkable specimen purchased by the British Museum for several hundred pounds sterling during the early-1860s, the comparative anatomist Richard Owen consistently emphasized its bird-like over its reptilian features, arguing that the “best-determinable parts of its preserved structure declare it unequivocally to be a Bird.” As Huxley himself was reluctant to admit, part of the problem was that even the most complete and best preserved Archaeopteryx specimens available at the time were lacking a skull. This situation allowed Owen to draw on Cuvier’s “law of correlation” and infer that contrary to what we would expect of a reptile, this animal’s mouth was a “beak-like instrument” entirely “devoid of lips.”30 Downloaded from hos.sagepub.com by guest on December 1, 2015 463 Rieppel In contrast to Archaeopteryx, the Hesperornis in Marsh’s possession included a skull complete with a bird-like beak full of reptilian teeth. Given that it was an entire geological epoch younger than Archaeopteryx, this implied that the latter, too, must have had teeth (a speculation that would be borne out by subsequent discoveries in the lithographic limestone of Solnhofen, Germany). Marsh’s fossil therefore quickly came to be recognized among the most significant paleontological discoveries in a century full of noteworthy finds, so much so that Huxley himself went out of his way to visit the Peabody Museum and examine the fossil in person. During his visit, Huxley flattered Marsh with the remark that “there is nothing in any way comparable … for their scientific importance, to the series of fossils which Professor Marsh has brought together.”31 After stopping over in New Haven, Huxley delivered a series of lectures at Chickering Hall in New York. Entitled ‘Evidences of Evolution’, these lectures featured Marsh’s Hesperornis in a leading role. Directing his audience’s attention to a large-format illustration of the specimen that Marsh had supplied him (identical to the one reproduced as Figure 2), Huxley forcefully argued “the discovery of a bird that had teeth shows at once that there were ancient birds that in that particular respect approached reptiles more nearly than any existing bird does.”32 Disseminating paleontological knowledge in plaster and print Given the important role that it played as material evidence of common descent, Marsh was understandably proud of the Hesperornis. It is no surprise, then, that he wasted no time spreading word of the remarkable specimen among fellow naturalists. Immediately after having examined his initial discoveries, for example, Marsh dashed off a series of preliminary descriptions for publication in scientific journals.33 Then, having studied his collection more carefully and after commissioning the production of several lithographs, he published more synoptic articles.34 Finally, within a decade of finding his first fossils out west, Marsh completed a lavishly illustrated, leather-bound monograph whose considerable production costs were underwritten by the federal government through the United States Geological Survey. This expensive folio volume offered an exhaustive description of Marsh’s entire collection of Odontornithes, or ancient toothed birds, making a point of emphasizing their “reptilian ancestry.”35 Besides providing detailed textual descriptions, Marsh also included figures of individual bones he had collected, as well as a scale drawing of the entire, fully articulated skeleton of Hesperornis (Figure 2). Given Marsh’s speculation that Hesperornis spent its entire life on the water, only venturing out onto shore in order to breed, it is somewhat curious that he chose to depict the bird in an upright and standing position. This unusual choice notwithstanding, Marsh insisted that enough of the skeleton was at hand that it could be “confidently believed that the field of conjecture has been reduced to a minimum” in executing the restoration.36 In some ways this was a bold pronouncement to make, as Marsh’s fossils, remarkably complete though they were, did not provide definitive information of every detail. In particular, Marsh was working with a single skull whose upper cranial regions were badly crushed during fossilization. Thus, at least some measure of speculation was surely Downloaded from hos.sagepub.com by guest on December 1, 2015 464 History of Science 53(4) Figure 2. Restoration of Hesperornis regalis, plate XX in Odontornithes: A Monograph on the Extinct Toothed Birds of North America, by O.C. Marsh, 1880, US Geological Exploration of the Fortieth Parallel, Washington, DC. required to re-assemble the skeleton on paper. This is especially true given that Marsh went so far as to provide a surprisingly precise illustration of the morphology of its brain case, which is precisely the region most severely distorted during fossilization.37 The production of these monograph publications was a hugely ambitious undertaking, representing a considerable investment of time, money, and manpower. The Odontornithes volume, for example, weighed in at several hundred pages in total, comprising nearly 200 pages of descriptive text, 34 lithographic plates, and 40 woodcuts. Its execution was only possible because Marsh had assembled a whole team of dedicated assistants, many of whom occupied a series of workshops located in the basement of Yale’s Peabody Museum. By the 1880s, Marsh had succeeded in using federal funds to convert the museum into an efficient clearinghouse of paleontological knowledge, one whose complex bureaucratic structure helped him manage the collection, processing, Downloaded from hos.sagepub.com by guest on December 1, 2015 465 Rieppel Figure 3. Marsh’s team of laboratory assistants with a large brontothere skull in the basement of the Yale Peabody Museum, 1888. They are, from left to right, Erwin H. Barbour (fossil preparator), Frederick Berger (artist), Adam Hermann (fossil preparator), and Emil Crisand (lithographer). Courtesy of the Vertebrate Paleontology archives at the Peabody Museum of Natural History (VPAR.000094). and dissemination of new discoveries. Besides a large number of hands out in the field, there were clerks who managed the day-to-day work of unpacking shipments and maintaining a detailed inventory, as well as preparators to clean and remove fossils from their surrounding rock matrix. In addition, Marsh also engaged several artists to illustrate his publications. For example, he hired the two principles of the lithographic firm Berger & Crisand to produce all the illustrations of Hesperornis in the Odontornithes volume. As Figure 3 indicates, Frederick Berger was in charge of “delineating” or drawing Marsh’s copious specimens, leaving the task of transferring these onto a lithographic stone to his partner, Emil Crisand.38 But lithographs were not the only means by which information about Hesperornis was mechanically reproduced. In addition to traditional monographs, Marsh also issued casts of certain fossilized bones. To that end, the Peabody Museum engaged the services of several sculptors and architectural plasterers who produced molds with which to churn out a large number of three-dimensional replicas. Upon the completion of a new monograph, copies of the most important fossils described therein could therefore be shipped out to fellow naturalists throughout the US and Europe.39 This is precisely what happened with Hesperornis. Just as the Odontornithes monograph was nearing completion, Marsh sent its fossilized bones to Tobias Kappeler, a Swiss sculptor who had recently settled in New Haven. Kappeler charged Marsh $15 a specimen to produce a series of 30 Downloaded from hos.sagepub.com by guest on December 1, 2015 466 History of Science 53(4) identical casts that were subsequently distributed to learned naturalists all over the world, including paleontologists at the Smithsonian Museum in Washington, DC, the Museum of Comparative Zoology at Harvard, the Philadelphia Academy of Natural Sciences, the British Museum in London, Oxford University, and the Muséum d’Histoire Naturelle in Paris, as well as the royal museums in Berlin and Vienna.40 Although it certainly ranked among his most well-known fossils, Hesperornis was not the only specimen that Marsh cast in plaster. During the early-1880s, Kappeler replicated a number of additional specimens ranging from the dinosaur Stegosaurus to the flying reptile Pterodactyl and the extinct North American ungulate Dinoceras.41 As he did with Hesperornis, Marsh sent these out to museums across Europe and North America, charging a modest fee for the service. It is not clear whether Marsh ever profited financially from the sale of these objects, but the scant evidence that remains indicates he most likely sold them at or near cost.42 By the mid-1880s, then, Marsh was literally running a distribution center for plaster cast replicas out of the Peabody Museum. To facilitate that endeavor, he directed his staff to compile a series of extensive catalogues and price lists describing the various specimens offered for sale. Perusing these, an interested party might learn that it would only cost $5 to acquire a cast of the left femur of Stegosaurus, whereas the entire hind leg of Camptosaurus would set them back $15.43 In this way, Marsh was able to disseminate information about his many discoveries among a coterie of fellow naturalists much in the same way that he distributed off-prints of journal articles or copies of his monograph publications out of the museum. Plaster cast publishing in paleontology Marsh’s decision to cast parts of his growing collection in plaster was far from unique. In fact, the practice of cast-making extends back to ancient Egypt, if not before. Initially, the technique was primarily used for the creation of architectural ornament, especially in the making of stucco sculptures. By the sixteenth and seventeenth centuries, its popularity intensified alongside the growing enthusiasm for amassing impressive collections of rare, opulent, and curious objects. Owners of such collections often had especially desirable statues cast in plaster and other, more valuable media, while artisan naturalists, such as Bernard Palissy, perfected the art of creating exquisite and lifelike sculptures by taking casts directly from real plants and animals.44 Thus, by the eighteenth century, casts had become a ubiquitous component of the well-stocked Wunderkammer. This was especially true in central and northern Europe, where a thriving market for reproductions of Greek and Italian statuary began to take shape. During the eighteenth century, the growing popularity of antiquarian pursuits helped to transform casts into objects of learned contemplation as well. Increasingly seen as an authoritative record of classical antiquity, casts played an indispensable role in the creation of modern archaeology and art history by German scholars like Johann Joachim Winckelmann.45 However, their popularity did not reach its zenith until the nineteenth century, with the proliferation of new institutions dedicated to teaching industrial art and design. For reasons of cost as well as propriety, these schools and academies often preferred to rely on plaster casts over live models in courses on drawing and sculpture. As a result, the production of plaster casts was scaled up to a new Downloaded from hos.sagepub.com by guest on December 1, 2015 467 Rieppel level of efficiency among the dozens of firms that sprang up in nineteenth-century cultural centers like Paris and London. In addition to streamlining the production process, these firms distinguished themselves by access to a large stock of molds they claimed were scrupulously faithful to their originals, and it was not uncommon for one firm to take another to court over piracy and other infringements of trademark and copyright law.46 At the same time, the growth of public museums during the nineteenth century also contributed to the increasing demand for plaster cast reproductions. This was especially true in the United States, whose art museums overwhelmingly preferred replicas of European masterpieces to domestic originals of far less renown.47 Because they were understood to serve an educational mission, tasked with cultivating the rudimentary aesthetic senses of ordinary Americans, these art museums sought to put visitors in touch with the highest achievements of an older and more refined culture. As a curator at the Boston Museum of Fine Arts wrote in The Nation, the high cost of overseas voyages ensured that “casts and other art reproductions” were the sole means by which most Americans could “become familiar with the great masterpieces of European galleries.”48 It was only once wealthy philanthropists began to compete with one another over who could donate the most impressive original artworks to public museums that American galleries began to prize the quality of authenticity above all else. As a result, the earlytwentieth century saw nearly all of these institutions relegate their plaster collections to storage or even destroy them outright.49 The nineteenth-century enthusiasm for casts was not confined to the art world; it was widespread in the sciences as well. For example, physicians had long used moulages of pathological anatomy—what the historian Thomas Schnalke describes as “pictures of diseases in wax”—to train medical students.50 Similarly, anthropologists and archaeologists also made widespread use of plaster casts, and they continue to do so today. During the early-twentieth century, for instance, the firm R.F. Damon and Company made a brisk business selling reproductions of early hominid fossils, including Raymond Dart’s famous Australophithecus specimen from South Africa.51 Not long thereafter, the Wenner–Gren Foundation teamed up with the University of Pennsylvania Museum to set up a centralized distribution center for casts of specimens the community deemed were especially important for teaching and research purposes.52 Much like anthropologists and archaeologists did, paleontologists made prolific use of plaster cast reproductions. During the mid-nineteenth century, for example, a “formatore and modeler” in the science and art department of the Galleria delle Belle Arti in London named Domenico Brucciani began to take casts in the British Museum of Natural History. By the 1880s, the firm of D. Brucciani and Co. has been appointed as the sole and official dealer of that institution’s casts. As its keeper of geology explained, these reproductions were prepared “for the purpose of supplying Museums, and the Public generally, with copies of all the most striking objects in the collection.”53 Initially, the regular price lists issued by the London museum included some 230 casts, but by the 1890s they were running to over 500 objects in all. In the United States, by far the largest dealer in plaster cast reproductions was Ward’s Natural Science Establishment in Rochester, New York. Its proprietor, Henry Augustus Ward, first became involved in the business while he was enrolled as a student of Downloaded from hos.sagepub.com by guest on December 1, 2015 468 History of Science 53(4) geology and mineralogy at the prestigious École des Mines in Paris. When a stroke of bad luck left Ward without continued financial support from back home, he was forced to sell off parts of his own collection to make ends meet. Although this decision was initially made out of desperation, Ward soon realized he could do far better than merely keeping himself afloat by selling mineralogical and paleontological specimens. Before long, he was even able to finance a number of trips through Germany, England, and Russia, where he toured museums and met with eminent naturalists for the purpose of buying, selling, and exchanging natural history objects.54 By the year 1860, Ward had succeeded in making a name for himself as a tireless collector and knowledgeable geologist, which led to his being offered a teaching position at the recently founded University of Rochester.55 Before taking up the new job, however, Ward made another tour through Europe to round out his growing collection and cast a number of especially rare and spectacular fossils in plaster. When he finally returned to the United States, the molds he had acquired in Europe would go on to establish Ward as America’s leading supplier of natural history specimens.56 During the mid-1860s, Ward issued the first catalogue of casts that his firm was pleased to offer for sale. Bound in green cloth and numbering over 200 pages in length, this booklet opened with a revealing argument for why schools and colleges across the United States should avail themselves of Ward’s services. Emphasizing the widespread popularity and considerable economic importance of geology and paleontology, Ward insisted that proper training in the field required far more “than simple textbooks or oral teaching.” Rather, Ward argued that “[v]isible, tangible objects” were required to give students a full appreciation of God’s natural handiwork. However, it was not enough simply to acquire a few choice specimens and be done with it. To be of significant and lasting value, paleontological collections “should be as complete as possible” and able to provide students with “an unbroken view of ancient life.”57 However, he admitted that few if any American colleges had the “princely” sums that were required to form a collection consisting of anything but “trivial portions of the larger and, in some sense, more interesting and important forms.” It was precisely this difficulty that Ward’s catalogue was supposed to solve, making available plaster cast replicas of fossils whose “originals” were “either unique specimens or are so very rare that it is altogether impossible to obtain them.” Finally, lest anyone harbor doubts about the quality of his casts, Ward assured readers they were “intended to be exact copies … of the originals, and such care is taken in their coloring that large numbers of them will hardly be detected, as casts, when ranged with original specimens on the Cabinet shelves.”58 Although it was clearly designed to increase revenues, Ward hoped the Catalogue of Casts would find favor as more than a purely commercial document. It was also supposed to serve as a genuine educational guide. In addition to listing the price of each specimen offered for sale, the catalogue doubled as an extensive inventory of God’s creation. To that end, Ward made sure to arrange the list of available specimens according to George Cuvier’s widely adopted classification scheme. As he confided in a letter to his friend, everything had to be “done in the most scientific way.”59 In addition, Ward also included detailed descriptions of individual items offered for sale. For example, an entry on the Megatherium alone ran to several pages and described everything from a history of the Downloaded from hos.sagepub.com by guest on December 1, 2015 469 Rieppel animal’s discovery to the details about its osteology, functional anatomy, classification, and behavior. This practice too served to communicate Ward’s epistemic ambitions, his desire to gain recognition as more than the proprietor of a mail-order curio shop.60 In a provocative essay about the history of three-dimensional teaching tools, Nick Hopwood has likened the widespread circulation of embryological models in late-nineteenth-century Germany to book publishing. Tellingly, the most prominent supplier of wax models in Germany at the time—Adolf Ziegler—explicitly invoked the phrase “plastic publisher” [plastischer Verleger] to describe his own role in helping scientific “authors” share their findings.61 My suggestion is that Henry Augustus Ward can be thought of along similar lines. His position as a Professor at the University of Rochester, his mastery of the scientific literature, and his credentials as a fellow of the Geological Society (which he made sure to list on the Catalogue’s cover), all served to distinguish him from the artisans he employed. At the same time, Ward saw his role as consisting in the dissemination of natural knowledge rather than personally adding to the stock of available ideas. Not unlike Ziegler in Germany, then, Ward can be seen as a plastic publisher, except that he produced plaster casts rather than wax models. Ward engaged in a number of practices that help to solidify the conceptual link between the circulation of plaster casts and traditional book publishing. For example, in 1870, Ward began to advertise an “Academy” and “College Series” of “celebrated fossils” from European and American Museums. Deliberately designed to serve as a comprehensive, room-sized “textbook” in geology and paleontology, the College Series consisted of more than 600 specimens ranging over the entire history of life on the earth. Intended to form “a complete Cabinet for the Geological Museum of our Colleges,” the series was organized according to the latest taxonomic standards and could be purchased for the price of $1000. Besides the specimens themselves, the cabinet also included printed labels that referred back to the detailed descriptions and figures in the printed catalogue. Ward therefore sought to provide colleges with an immersive educational experience, one that gave students any number of avenues from which to access the extinct life of our planet.62 Perhaps the most striking connection between Ward’s business and that of a traditional book publisher, however, was his desire to secure an exclusive copyright on the models offered for sale. To that end, the early-1860s saw Ward seek legal counsel to help prevent pirates from infringing his intellectual property. In response to one such request, the law firm of Munn and Company informed Ward that although American copyright law did not extend to three-dimensional models, an “appropriate trade-mark affixed to each cast produced by you can be patented if you wish, and by doing so you will be able to secure the trade.”63 There is no evidence that Ward ever acted on this advice, but it is noteworthy that Ward’s 1866 Catalogue did emphasize that casts would only be sold “on the express condition that the individual or Institution which purchases them will not copy them, nor allow them to be copied.”64 In this respect, Ward’s business practice resembled that of other plaster cast publishers. For example, the firm of R.F. Damon and Co. claimed to hold a “copyright” on its casts, agreeing to pay “royalties” to the owners of original specimens.65 As we shall see, the transfer of intellectual property rights would go on to form an important sticking point in Ward’s negotiation with naturalists in the decades that followed. Downloaded from hos.sagepub.com by guest on December 1, 2015 470 History of Science 53(4) Expanding into a popular audience By the time Marsh discovered Hesperornis in the American west, plaster cast publishing had become something of a mainstay in North American paleontology. As early as 1873, Ward began pressuring Marsh to make some of his fossils available for the Catalogue, arguing that doing so would greatly augment his renown, adding especially to his “European reputation.”66 As we have seen, Marsh took this suggestion to heart, but not in the way Ward likely intended. Rather than distributing casts of his specimens out of Rochester, Marsh preferred to set up his own plaster cast publishing operation, hiring the Swiss-born Thobias Kappeler to do the work in New Haven. This must have come as a disappointment to Ward, but it was consistent with other developments at Marsh’s museum, which also began issuing its own print publication around the same time. Despite having his initial advances rebuffed, Ward continued to plead for permission to reproduce some of Marsh’s most notable specimens.67 Despite his best efforts, however, it would be another 10 years before the latter finally agreed. Marsh must have known working with Ward would dramatically increase the size of his audience. Whereas Ward’s casts were displayed everywhere from schools and small colleges to large, philanthropic museums and World’s Fairs, Marsh’s distribution network cantered on a fairly small coterie of fellow naturalists. For example, rather than issue a printed catalogue of all the available casts, Marsh circulated hand-written price lists and sold his specimens via personal correspondence. The decision to begin dealing with Ward may therefore have been motivated, at least in part, by a growing desire on Marsh’s behalf to reach a wider and more popular audience.68 Negotiations between the two parties began in earnest during the mid-1880s, but were slowed by Marsh’s several anxieties. These anxieties will be familiar to anyone who has signed a contract with a traditional book publisher. Chief among them was Marsh’s desire to maximize his control over the products bearing his name, protecting what might be described as Marsh’s “brand” among fellow naturalists. For example, Marsh made Ward promise to distribute only “first class specimens” whose quality was “equal to those now made by Mr. Kappeler.” At the same time, Marsh was keen to ensure his specimens were affordable and would circulate widely. To that end, he insisted that Ward agree to charge no more for a specimen than it cost to have them produced in New Haven. Finally, Marsh was extremely sensitive about issues of priority, insisting that Ward agree to hold off on distributing anything “until the volume describing the specimen is published.”69 Marsh’s anxiety about issuing reproductions of specimens whose description had not formally been circulated in print is deeply revealing. As is well-known, American paleontology at the time witnessed a bitter feud between Marsh and his rival, Edward Drinker Cope.70 Both naturalists competed to name more species from the fossils being discovered in the American West, and much of the controversy between them cantered on priority disputes, with both men accusing each other of taxonomic malpractice. Marsh especially denounced Cope’s frequent use of newspaper articles, telegrams, and printed pamphlets when he did not have time to register new additions to biological nomenclature in scientific journals and monographs.71 Against the backdrop of these heated debates about which media were appropriate for establishing taxonomic priority, Marsh feared that plaster cast reproductions of a new specimen might compete directly with Downloaded from hos.sagepub.com by guest on December 1, 2015 471 Rieppel more traditional publications. Even worse, plaster casts might actually trump the authority of a printed report because of its status as an acceptable stand-in for the original fossil. The possibility that rival naturalists might gain access to casts of his specimens from Ward therefore posed a particularly dire threat, allowing them to author their own print publications before Marsh had a chance to do so himself.72 The degree to which plaster casts could function as genuine publications in paleontology at the time is reinforced by the fact that issues of intellectual property quickly emerged as another sticking point in the negotiations with Ward. Although he mostly assented to Marsh’s several demands, Ward was unwilling to leave both the mold and a series of five casts at the Peabody Museum in exchange for the privilege of replicating a specimen. Part of the disagreement hinged on Ward’s judgment that Marsh was asking for too much payment up front, saying that he preferred to pay royalties based on the proceeds from sales instead. Beyond that, however, Marsh’s desire to keep the molds in New Haven would also have prevented Ward from issuing additional series of specimens. In other words, under Marsh’s proposed terms, Ward would have been unable to publish a second “edition” of casts without first going back to secure the original molds. The only way to avoid doing so would have involved making a second-generation copy (i.e. making a mold from a cast rather than the original), which is precisely what the stricture against selling casts of an inferior quality was designed to prevent.73 With all these disagreements to be ironed out, it is no surprise that negotiations between Marsh and Ward dragged on over several years. Predictably, just as the two parties were about to conclude a deal, concerns over intellectual property flared up again. Ward, whose business only made financial sense if it was conducted “on a large scale,” insisted on being granted an exclusive right to distribute casts of Marsh’s specimens, allowing him to publicize the fact that “we alone (save any reserves which you make for yourself) can furnish them.”74 It was thus only once Marsh had agreed to cease the practice of distributing casts out of the Peabody Museum that a contract was finally signed. Once this step had been taken, however, Ward wasted no time in printing a new pamphlet to announce the availability of Marsh’s extraordinary fossils to his regular customers.75 Despite Ward’s stated desire to distribute a large series of Marsh’s fossils, the latter only granted permission to cast two specimens from his extensive collection. In many respects, these were polar opposites of one another. The first—Hesperornis—was relatively small and did not constitute a particularly compelling visual spectacle. Given its intermediate position between reptiles and birds, however, it was of immense scientific importance. The second, Dinoceras mirabile, arguably had less far-reaching implications about the history of life on the earth, but its reputation already eclipsed that of Hesperornis among a non-specialist audience. An extinct ungulate from the Eocene period that also hailed from the American interior, its widespread appeal was largely due to the animal’s extraordinary size and outlandish appearance. A stunningly strangelooking creature, the Dinoceras resembled a large rhino or tapir, except for a number of bony protuberances growing out of the top of its head and the large saber-like canines extending down past its lower jaw (Figure 4). Given these unusual features, the Dinoceras quickly became famous among fellow naturalists and the broader public alike.76 Downloaded from hos.sagepub.com by guest on December 1, 2015 472 History of Science 53(4) Figure 4. A cast of the Dinoceras mirabile (left), the skeleton of an Irish Elk (middle), and a cast of Hadrosaurus foulki (right) on display at the Field Columbian Museum, Chicago, 1898. Courtesy of the Field Museum of Natural History. Digital Identifier: CSGEO3813. Dinoceras had achieved a level of popular notoriety even before Ward began distributing casts of the specimen out of Rochester. However, its inclusion in Ward’s 1891 catalogue truly cemented the animal’s status as one of the most recognizable paleontological specimens of the time, and the Dinoceras quickly became a mainstay of museum exhibits across North America and Europe. Now that it was available for purchase through Ward’s Natural Science Establishment, an interested party no longer had to be on the Peabody Museum’s own mailing list to acquire a copy. Rather, anyone willing to part with $20 could send in for an authoritative replica of its skull. Even more affordable were individual vertebrae, which went for as little as $2 apiece. However, as readers of the catalogue would no doubt have noticed, Ward was especially keen to sell casts of the Downloaded from hos.sagepub.com by guest on December 1, 2015 473 Rieppel whole specimen outright. Thus, for $350, it was possible to purchase its entire skeleton, fully articulated and mounted on a hardwood pedestal. Measuring 5 feet 9 inches in height and 10 feet 8 inches in length, the latter was billed as an imposing centerpiece for any natural history exhibit.77 In contrast to the Dinoceras, the Hesperornis could not be mounted as a fully articulated, freestanding display. As a matter of fact, Ward did not even have casts of the entire specimen available for purchase. Most notable, perhaps, was the absence of a complete skull. Portions of both the right and left mandible were present, clearly showing the prominent row of teeth embedded in the animal’s beak. Its braincase, occiput, and other hind portions of the cranium were conspicuously absent though. This suggests that while the Dinoceras was sold as a spectacular centerpiece, Hesperornis was intended for more careful and close-up inspection. It is significant, for example, that while he did not reproduce the animal’s entire skull, Marsh made sure to include the pelvis and pelvic arch in the cast, elements of the skeleton which demonstrated its taxonomic proximity to more distinctly reptilian forms. This decision suggests the specimen’s importance lay in its obvious implications for the hotly contested theory of common descent, more so than its ability to impress museum-goers with its sheer size and visual strangeness. Most likely, museums that purchased the entire specimen of Hesperornis for $30 were expected to lay its bones out on a flat surface. This mode of presentation would have invited a detail-oriented examination over a holistic appreciation, helping to showcase the specimen’s considerable epistemic status. At the same time, the exhibition of individual bones presented a significant challenge, as lay visitors without a background in paleontology or comparative anatomy could not be expected to make intuitive sense out of such a display. For this reason, Ward also supplied a detailed explanatory label to draw the attention of viewers to the specimen’s more noteworthy features. This label, which had been composed by Marsh, informed readers that although Hesperornis resembled diving shore-birds in several respects, the “finer points of its anatomy” clearly indicated that it was closely related to modern Struthioformes, meaning that it was “essentially a carnivorous swimming Ostrich.”78 In addition, the purchase of a full specimen also included a large poster depicting Marsh’s imaginative, two-dimensional reconstruction of the Cretaceous bird, the same illustration previously published in the Odontornithes volume and used by Huxley in his popular lectures (see Figure 2). Measuring 32 by 42 inches and showing the specimen in life size, this poster would have given spectators imaginative access to the specimen’s functional anatomy. By looking back and forth between the cast and the large-format wall drawing, viewers could have manipulated the objects laid out before them, fitting the bones together and articulating the Hesperornis in the mind’s eye.79 Establishing indexicality In light of the care that he took in promoting the specimen, readers may be surprised that Marsh never chose to articulate Hesperornis into an impressive centerpiece like he did with the Dinoceras. In fact, Marsh was generally reluctant to assemble any of his prehistoric fossils (or casts thereof) into a freestanding skeleton. His treatment of Hesperornis was therefore in line with Marsh’s usual exhibition practice, whereas Dinoceras served Downloaded from hos.sagepub.com by guest on December 1, 2015 474 History of Science 53(4) as an exception to the rule. Indeed, of all the extinct animals that Marsh collected over the course of a long and fruitful career, Dinoceras was the only one he ever assembled into the kind of display that has become the hallmark of modern paleontology. In what follows, I will interrogate Marsh’s reticence on this score, using it to suggest that Marsh’s reluctance to mount his specimens into a freestanding display stemmed from a fear that doing so would undermine their indexicality, casting doubt on their status as genuine objects of reliable knowledge. A major problem when assembling specimens for display was the paucity of available material. For example, even though Dinoceras ranked among the most complete fossils in Marsh’s collection, important parts of the skeleton were nonetheless missing. The task of constructing a freestanding mount therefore required much more than simply arranging its bones in the right position. It also involved filling in missing portions with sculptural reconstructions. Adopting C.S. Peirce’s terminology, this meant it approached an icon more so than a mechanically produced indexical representation. As J.H. Emerton, an artisan who often worked for Marsh during the 1880s, explained to the Boston Society of Natural History, “few parts of the restoration could be cast directly from the bones, the rest having been modeled in clay.”80 As a result, most of the Dinoceras molds that were eventually given to Ward had actually been taken from Emerton’s models, not the original fossils. Another problem was that the Dinoceras fossils in Marsh’s collection hailed from a number of distinct individuals at different stages of their life history. This situation all but ensured they would fail to agree with respect to their size and proportions. To solve this problem, Marsh directed Emerton to use the type specimen—consisting of a nearly complete skull, pelvis, several lumbar vertebrae, and a neck vertebra as well as two rib bones—as a kind of standard against which to calibrate the rest. In Emerton’s own estimation, this method worked to produce a reasonably satisfactory result, leaving only a few small but crucial uncertainties. For example, because no fully complete specimen had ever been found, Marsh did not know exactly how many vertebrae made up the animal’s backbone. Reluctant simply to take a stab in the dark, he instructed Emerton to include the same number of vertebrae one might find in extant horses, tapirs, and elephants. Doing so was justified, Emerton explained, by the fact that Marsh had earlier determined these were the animals “to which the Dinoceras appears nearly related.” In the same vein, an extant hippopotamus served as a guide in sculpting the lumbar vertebrae’s lateral processes, which were also missing from Marsh’s collection.81 Although Emerton’s Dinoceras quickly became a ubiquitous and widely recognized paleontological display object, its status as an icon threatened the specimen’s credibility. As a result, it came under some intense criticism. Writing in an 1890 issue of The American Naturalist, one of Marsh’s former laboratory assistants named Erwin Barbour sounded the loudest charge of paleontological fraud. Tellingly, Barbour’s accusation did not center on whether specimens such as the Dinoceras provided an accurate representation of the deep past. Rather, he claimed that Marsh routinely directed his laboratory staff to give their paleontological models the look and feel of an original fossil. This, he alleged, made it all but impossible for viewers to discern which aspects of a display were authentic and which ones were sculpted, where nature ended and the imagination began. Downloaded from hos.sagepub.com by guest on December 1, 2015 475 Rieppel Figure 5. An allegedly adulterated cast of a dinosaur femur circulated by Marsh, from Barbour’s ‘Notes on the Paleontological Laboratory of the United States Geological Survey under Professor Marsh’, The American Naturalist, 24(280), 1890, p. 392. According to Barbour, the assistants in Marsh’s laboratory were never asked “how nearly they can approximate the truth,” but rather “‘How closely can you imitate the color and texture in that missing part?’ which being translated is, How cunningly can you deceive?” The results, he complained, left the public “in doubt as to what is real, [and] what [is] conjectural.” According to Barbour, not even the trained eye of an expert paleontologist could be trusted “to distinguish between the rusty, frost-cracked, weatherbeaten, moss and lichen effects, craftily wrought in plaster, and the conditions wrought by time on the specimens themselves.” Hence, Barbour insisted, specimens such as the Dinoceras “must sooner or later stand as monuments of reproach to the man who has so far deceived the world.”82 The errors that Marsh perpetuated, Barbour alleged, were made all the worse by the fact that he subsequently circulated them in the form of plaster cast reproductions (Figure 5). In his “zeal to out-rival all others in the startling size of his fossils,” Marsh allegedly sent “out casts of heroic stature” even though the original fossils were relatively small in size. Thus, Barbour advised, recipients of a large dinosaur femur “exceeding eight feet in height” were well advised to “saw off a two-foot back-log of the same, and then it will stand as high” as the original. Given their status as widely accepted and authoritative Downloaded from hos.sagepub.com by guest on December 1, 2015 476 History of Science 53(4) Figure 6. Right ilium (the uppermost bone of the pelvis) of the plant-eating, ornithopod dinosaur Camptosaurus in the Vertebrate Paleontology collections of the Yale Peabody Museum, collected by William Harlow Reed at Como Bluff, Wyoming, 1880. Photograph by Lukas Rieppel. stand-ins for the actual past, these casts led several members of the paleontological community unwittingly to incorporate Marsh’s cunning deception into their own textbooks. This only spread Marsh’s falsehoods further through the paleontological ranks. Of course, one might expect such frauds to be uncovered as soon as an independent third party examined the original fossil. Unfortunately, Barbour complained, these too had been altered, rendering them notoriously “plump with plaster,” and therefore incapable of restoring the community’s trust.83 We need not take Barbour’s accusations at face value to derive historical insight from his controversial account. It is clear the disgruntled former assistant suffered a personal animus against Marsh, making his decision to publish in Edward Drinker Cope’s journal The American Naturalist all the more suspect.84 Still, it is telling that among all the lines of attack available to him, Babour chose to accuse Marsh of having used plaster to contaminate the scientific record. Because extinct creatures like Hesperornis and Dinoceras were not accessible to direct observation or experimental manipulation, the physical connection that linked their fossilized bones to flesh and blood creatures was crucial to ground paleontologists’ speculative reconstructions within the actual past. Original fossils, and their plaster cast replicas, imparted paleontological knowledge with an ontological authority. Accusations of tampering with that material link therefore threatened to undermine nothing less than the very foundations upon which paleontological knowledge was built. This danger was all the more acute in the case of Mesozoic fossils like Hesperornis—those that hailed from the Jurassic, Triassic, or Cretaceous periods in Geological history—whose great age tended to make them brittle, fragile, and at best incompletely preserved. As Figure 6 illustrates, these objects often had to be built up out of countless small fragments, with missing portions filled in using plaster. Thus, even if Marsh had restored his fossils to instruct and not, as Barbour had charged, to deceive, the mere fact that his assistants did intervene in their material constitution posed an existential threat to their credibility. Downloaded from hos.sagepub.com by guest on December 1, 2015 477 Rieppel Figure 7. Map of a Stegosaurus quarry, Como Bluff, Wyoming, by Fred Brown, 1885. Traced and published in Charles Gilmore, Osteology of the Armored Dinosauria, (Washington: Government Printing Office, 1914). For the original map, see VPdoc201, and for the original tracing see VPdoc0199, Vertebrate Paleontology Archives, Yale Peabody Museum. Marsh did not have to be guilty of fraud for Barbour’s accusations to carry a considerable bite. Because the quality of indexicality did not easily scale up from an individual bone to an entire, freestanding display, this was especially true for fully articulated skeletons like that of the Dinoceras. Even remarkably complete fossils were usually found as a jumble of disarticulated pieces scattered over the space of a quarry. As such, the location of each bone in the quarry rarely mapped onto their position within the frame of a once-living animal. The task of articulating a skeleton therefore required a great deal of imaginative and inferential work, undermining the trustworthiness of the end result. As an example, consider the unusually complete Stegosaurus quarry depicted in Figure 7. It was initially uncovered by one of Marsh’s collectors in Wyoming during the mid-1880s. Anyone familiar with this creature’s functional anatomy will recognize that most of the specimens in this quarry had been broken up during the long fossilization process. By far the most difficult part of reconstructing the Stegosaurus involved figuring out what to do with a series of the large, bony plates that were consistently found near the rest of a skeleton. For example, an early article announcing the dinosaur’s initial discovery described it as having had a long body “protected by large bony dermal plates, somewhat like those of the Atlantochelys (Protostega).”85 The latter was a large, extinct sea turtle with a hard carapace, so Marsh’s decision to compare it with Stegosaurus Downloaded from hos.sagepub.com by guest on December 1, 2015 478 History of Science 53(4) signals that he initially imagined its plates lying flat against the rest of its body (Figure 8). A few years thereafter, he revised this conjecture, explaining that while some of the dermal plates likely laid flat on either side of the animal’s backbone, those on its tail may have “stood erect over the caudal vertebrae.” Moreover, he also announced the discovery of several long and sharp spikes, which he imagined to form impressive claw-like structures on its forelimbs and elsewhere along its body (Figure 9).86 Figure 8. An early and imaginative reconstruction of Stegosaurus as a bi-pedal dinosaur with an uncertain placement of bony plates and sharp spikes. The figure was executed by Frank Bond under Wilbur Knight’s direction in 1899 and later published in Charles Gilmore, Osteology of the Armored Dinosauria, (Washington: Government Printing Office, 1914). It was only as new fossils were uncovered during the mid- to late-1880s that Marsh changed his mind about the functional anatomy of Stegosaurus. An especially important development occurred during the summer of 1887, when Marsh’s collector Fred Brown opened a new quarry several meters southwest of the one depicted in Figure 7. As he dug down into the hillside, Brown turned up a beautifully preserved section of caudal or tail vertebrae with several sharp spines articulated off to the side. Here, then, was a fossil whose indexical features finally extended beyond each individual bone. Rather than having to infer the relative position of each fragment from its internal features alone, Marsh finally had a material trace of the creature’s functional anatomy in the form of a partially preserved “death pose.” The discovery of this new quarry therefore induced Marsh to Downloaded from hos.sagepub.com by guest on December 1, 2015 479 Rieppel Figure 9. Another early and imaginative reconstruction of Stegosaurus as a bi-pedal dinosaur with an uncertain placement of bony plates and sharp spikes, from “Dinosaurs,” in The Scientific American, 29 November 1884, p. 343. publish a new description that placed its boney spikes at the tip of its tail rather than on its front limbs. Instead of comparing it to a turtle covered by a hard and flat shell, Marsh now likened the strange creature to a sting-ray, describing it as an animal whose huge barbs were “arranged in pairs along the top of the distal portion of the tail, which was elongate and flexible, thus giving effective defense.” To back up this theory, Marsh went so far as to produce a lithographic plate that showed the articulated section of Stegosaurus tail just as it had been found in the new quarry (Figure 10).87 Such difficulties were not confined to the case of Stegosaurus. According to Marsh, similar problems would plague anyone who speculated about the biology of long extinct creatures based upon fragmentary and incomplete physical evidence. Thus, he strongly counseled against a plan to mount the plaster cast replica of an American dinosaur named Hadrosaurus foulkii (a copy of which is depicted in Figure 4) as part of the 1876 Centennial Exposition in Philadelphia. As he wrote in a letter to one of the exposition’s organizers, “I do not believe it possible, at present, to make a restoration … that would Downloaded from hos.sagepub.com by guest on December 1, 2015 480 History of Science 53(4) Figure 10. Illustration of the relative placement of fossil vertebrae and tail spikes in Marsh’s Stegosaurus quarry, printed in O.C. Marsh, Dinosaurs of North America, (Washington: Government Printing Office, 1896). be of real value to science or to the public.” Fully mounted specimens, he added, were especially dangerous because they captured people’s imaginations so well that any errors they perpetuated would be especially “difficult to eradicate from the public mind.”88 Marsh’s reluctance to assemble his fossil discoveries into a freestanding display stemmed from a conviction that natural history derived its legitimacy from direct, firsthand observation rather than arm-chair theorizing. Because there was no way to observe long extinct creatures in the flesh, Marsh preferred to leave them in a state that was accessible to direct observation, namely, as scattered and disarticulated fragments of fossilized bone. For this reason, The Peabody Museum exhibited its Hesperornis exactly as it had been found, with its bones still encased in the original slab of rock matrix (see Figure 11).89 In this way, the fossil preserved not just the indexicality of each bone, but also that of their relationship to one another, adding further epistemic weight to the specimen’s status as a credible object of paleontological knowledge. In contrast, Marsh chose to cast each bone of the Hesperornis individually. This was not because it would have been impossible to cast the entire assemblage as one piece. In fact, it was more difficult to remove certain parts of the fossil from their rock matrix, make a mold of them, and then reinsert the bones back into the original slab. (Evidence of this process can still be discerned in Figure 11.) Rather, doing so helped to overcome the limitations of the original specimen as a display object, promoting its circulation by maximizing the cast’s appeal among a broad audience. For one thing, it left the choice of how to arrange or lay out each bone to whomever had purchased the cast. This situation Downloaded from hos.sagepub.com by guest on December 1, 2015 481 Rieppel Figure 11. The original Hesperornis regalis fossil as it was displayed at Yale’s Peabody Museum during the late-nineteenth century. From P.D. Gingerich, ‘Evolutionary Significance of the Mesozoic Toothed Birds’, Smithsonian Contributions to Paleobiology, 27, 1976, pp. 23–33. would have allowed different museums to emphasize different aspects of the bird’s anatomy, or to provide viewers with a point-by-point comparison between the bones of Hesperornis and those of its extant relatives. At the same time, casting individual bones produced a much lighter specimen, reducing the cost of shipping it all over the world. Finally, casts of individual bones could easily be passed around in a classroom setting, encouraging students to inspect parts of the skeleton in person as their teacher lectured on the finer points of comparative anatomy. For all these reasons, plaster cast replicas clearly served a subtly different function than original fossils. Whereas the latter were prepared to maximize their indexical features, plaster casts were a compromise medium, sacrificing a measure of their indexicality to render a more flexible specimen that could circulate among a much wider and more diverse range of audiences. The cast’s status as a compromise medium helps to smooth out an apparent contradiction between Marsh’s reluctance to distribute fully mounted skeletons of the Hesperornis with his willingness to circulate large-format wall charts that showed how its bones were likely articulated in life (Figure 2). Whereas Marsh was willing to sacrifice some measure of a specimen’s indexicality to promote its circulation, plaster cast replicas were nonetheless understood as a highly indexical medium. This situation manifested itself as a widespread anxiety about allowing imaginative or inferential elements to adulterate their production. However, the same was not true for drawings, lithographic plates, verbal descriptions, and printed reports. Because the latter would not have been mistaken for an indexical representation, their credibility rested on Marsh’s authority as a learned naturalist alone. Hence, the production of a speculative illustration only threatened to undermine Marsh’s individual reputation as a learned naturalist should it turn out to be Downloaded from hos.sagepub.com by guest on December 1, 2015 482 History of Science 53(4) wrong. It did not threaten to destabilize the entire discipline of paleontology, calling its material foundations into question. As a result, plaster casts were held to a much higher standard of proof than other genres of publication. Conclusion In this article, I have tried to suggest that during the nineteenth century, plaster casts functioned as a technology of circulation. For this reason, I have resisted the temptation to locate Hesperornis within a larger “exhibitionary complex,” likening it to a visual spectacle such as one might have encountered at a world’s fair or in a department store window. Instead, I have chosen to emphasize the status of plaster cast reproductions as a genuine mode of publication. Much like print, plaster casts were mechanical reproductions designed to promote the dissemination of information. In addition, the two also shared a number of institutional hallmarks in common. For example, Henry Augustus Ward acted very much like a traditional book publisher in that he sought to secure exclusive intellectual property rights over the casts in his catalogue. However, that does not mean the two were identical. For one thing, their status as a mechanical reproduction not only allowed casts to be fabricated in large numbers and at a relatively low cost, but it was also seen as a crucial way to preserve a specimen’s indexical qualities. In the final analysis, then, plaster casts ought to be understood as a compromise medium, combining some features of print with others from original specimens. By way of a conclusion, it is worth emphasizing a further difference between plaster cast reproductions and traditional print publications; namely, that casts did not easily lend themselves to the communication of an author’s intellectual claims. Casts were primarily used to circulate the material discoveries made by naturalists such as Marsh, whereas print was more suited to publicizing their epistemic contributions to the body of paleontological knowledge. Of course, that does not mean that casts failed to embody the theoretical commitments of those who commissioned and circulated them. As I have tried to demonstrate at the start of this essay, plaster cast replicas of the Hesperornis were clearly designed as material proof for the theory of common descent, backing up Marsh and Huxley’s claim that dinosaurs constituted a “missing link” between reptiles and birds. However, as I have also been at pains to suggest, the medium of plaster was especially prized for its ability to circulate objects whose material link to prehistory remained largely in tact, more so than that of a more traditional medium like lithography or the printed word. Moreover, the emphasis that was placed on a plaster replica’s indexical power did not mean the latter failed to bestow credit upon its “author.” Far from it, as can be seen from the printed labels that accompanied each Hesperornis cast distributed by Ward’s Natural Science Establishment. Besides providing a range of information about the bird’s functional anatomy and evolutionary history, these labels primarily served to inform viewers of the specimen’s provenance, announcing that its fossil bones had been discovered by Marsh and could be seen on display at Yale’s Peabody Museum of Natural History. Similarly, on the very first page of the catalogue offering Hesperornis for sale, Ward made sure to remind potential customers they were “indebted to PROFESSOR O.C. MARSH” for his “great liberality” in circulating casts of the famous specimen.90 Downloaded from hos.sagepub.com by guest on December 1, 2015 483 Rieppel To conclude, it is worth noting that the widespread renown a naturalist such as Marsh could derive from the circulation of plaster casts illustrates something important about the credit economy of nineteenth-century natural history. Rooted in the institution of the Wunderkammer and the practices of antiquarianism, paleontology—and natural history more broadly—was governed by what might be described as an ethos of accumulation.91 By the nineteenth century, however, it no longer sufficed to amass an especially complete and impressive collection of specimens. By this time, the esteem of the scientific community primarily accrued to those who shared their discoveries with their peers. Because they could be mechanically reproduced, casts excelled at doing precisely that. Not only did they circulate widely, but their status as indexical representations also ensured they would be recognized as trustworthy objects of paleontological knowledge. Acknowledgements Numerous friends and colleagues have helped in the research and writing of this essay. Daniel Brinkman at Yale’s Peabody Museum first alerted me to the existence of Marsh’s Hesperornis cast, and he went out of his way to send me material related to its history. For helpful discussions and feedback on drafts, I would also like to thank Tal Arbel, Henry Cowles, Alistair Sponsel, Laura Stark, Lydia Pyne, David Sepkoski, Hal Cook, and Joan Richards, as well as two anonymous reviewers. Finally, thanks to Barbara Narendra for giving me access to archival material at Yale’s Peabody Museum and to Bhart–Anjan Bhullar for showing me the original Hesperornis skull. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/ or publication of this article: The production of this essay was supported by a Richard B. Solomon Faculty Research Award from Brown University. Notes 1. For more on the evolutionary connection between reptiles and birds, see Alan Feduccia, The Origin and Evolution of Birds (New Haven: Yale University Press, 1999); and Jacques Gauthier and Lawrence F. Gall (eds.) New Perspectives on the Origin and Early Evolution of Birds (New Haven: Yale University Press, 2001). For a history of these debates, see Adrian Desmond, The Hot-blooded Dinosaurs: A Revolution in Paleontology (London: Blond and Briggs, 1975); and Adrian Desmond, Archetypes and Ancestors: Paleontology in Victorian London, 1850–1875 (Chicago: University Of Chicago Press, 1982). 2. See Pamela Smith, The Body of the Artisan: Art and Experience in the Scientific Revolution (Chicago: University of Chicago Press, 2004); Lorraine Daston and Katharine Park, Wonders and the Order of Nature, 1150–1750 (Cambridge: Zone Books, 1998), pp. 276–290; and Hanna Rose Shell, ‘Casting Life, Recasting Experience: Bernard Palissy’s Occupation between Maker and Nature’, Configurations 12, 2004, pp. 1–40. 3. On serialization, see Graham Law and Robert L. Patten, ‘The Serial Revolution’, in David McKitterick (ed.) The Cambridge History of the Book in Britain, Vol. 6: 1830–1914, (Cambridge: Cambridge University Press, 2009), pp. 144–171; and Nick Hopwood, Simon Downloaded from hos.sagepub.com by guest on December 1, 2015 484 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. History of Science 53(4) Schaffer, and Jim Secord, ‘Seriality and Scientific Objects in the Nineteenth Century’, History of Science, 48(3/4), 2010, pp. 251–85. On the importance of seriality for paleontology in particular, see Marianne Sommer, ‘Seriality in the Making: The Osborn–Knight Restorations of Evolutionary History’, History of Science, 48(3/4), 2010, pp. 461–82; and Gowan Dawson, ‘Paleontology in Parts: Richard Owen, William John Broderip, and the Serialization of Science in Early Victorian Britain’, Isis 103, 4, 2012, pp. 637–67. Alex Csiszar, ‘Seriality and the Search for Order: Scientific Print and Its Problems During the Late Nineteenth Century’, History of Science, 48(3/4), 2010, pp. 399–434; and G.N. Cantor et al. (eds.) Science in the Nineteenth-century Periodical: Reading the Magazine of Nature (Cambridge: Cambridge University Press, 2004). A number of historians have recently endeavored to put circulation at the center of scientific practice. See, for example, James Secord, ‘Knowledge in Transit’, Isis, 95(4), December 2004, pp. 654–72; and Kapil Raj, Relocating Modern Science: Circulation and the Construction of Knowledge in South Asia and Europe, 1650–1900 (Houndsmills: Palgrave Macmillan, 2007). Martin J.S. Rudwick, ‘The Emergence of a Visual Language for Geological Science 1760– 1840’, History of Science, 14(3), 1976, pp. 149–95. Martin J.S. Rudwick, ‘Georges Cuvier’s Paper Museum of Fossil Bones’, Archives of Natural History, 27(1), 2000, pp. 51–68. David Sepkoski, ‘Towards “A Natural History of Data”: Evolving Practices and Epistemologies of Data in Paleontology, 1800–2000’, Journal of the History of Biology, 46(3), 2013, pp. 401–44. For more on Peirce’s semiotics, see Albert Atkin, Peirce’s Theory of Signs, Stanford Encyclopedia of Philosophy, (summer 2013 edition), Edward N. Zalta (ed.), < http://plato. stanford.edu/archives/win2010/entries/peirce-semiotics/> (accessed 19 April 2015); and Charles Sanders Peirce, ‘On the Nature of Signs’, in James Hoopes (ed.) Peirce on Signs (Chapel Hill: The University of North Carolina Press, 1991). H.P. Grice, ‘Meaning’, The Philosophical Review, 66(3), July 1957, pp. 377–88. See Lukas Rieppel, ‘Bringing Dinosaurs Back to Life: Exhibiting Prehistory at the American Museum of Natural History’, Isis, 103(3), 2012, pp. 460–90. Lorraine Daston and Peter Galison, Objectivity (New York: Zone Books, 2007), pp. 120–121. For a pessimistic take on the possibility of paleontological knowledge about prehistory, see Derek Turner, Making Prehistory: Historical Science and the Scientific Realism Debate (Cambridge: Cambridge University Press, 2007). For a more optimistic take, see Adrian Currie, ‘Rock, Bone & Ruin: An Optimist’s Guide to the Historical Sciences’, PhD Thesis, Australian National University, 2014. Of course, this view necessitated discounting the often very creative work required to prepare fossils for study and display by removing them from their original rock matrix. For more on the important role of fossil preparators as the “invisible technicians” of paleontology, see Caitlin Donahue Wylie, ‘“The Artist’s Piece Is Already in the Stone”: Constructing Creativity in Paleontology Laboratories’, Social Studies of Science, 45 (1), 2015, pp. 31–55. Othniel Charles Marsh, Odontornithes: A Monograph on the Extinct Toothed Birds of North America, Professional Papers of the Engineer Department, US Army, no. 18 (Washington: Government Printing Office, 1880). For an influential account of how textual descriptions served as a “literary technology” that allowed readers to engage in an act of “virtual witnessing” in early modern natural philosophy, see Steven Shapin and Simon Schaffer, Leviathan and the Air-Pump: Hobbes, Boyle, and the Experimental Life (Princeton, NJ: Princeton University Press, 1985). See André Bazin, What Is Cinema? (Berkeley: University of California Press, 1967); Peter Wollen, ‘The Semiology of the Cinema’, in Signs and Meaning in the Cinema (Bloomington: Downloaded from hos.sagepub.com by guest on December 1, 2015 485 Rieppel 17. 18. 19. 20. 21. 22. University of Indiana Press, 1969); Stanley Cavell, The World Viewed: Reflections on the Ontology of Film (New York: Viking Press, 1971); and Kendall L. Walton, ‘Transparent Pictures: On the Nature of Photographic Realism’, Critical Inquiry, 11(2), 1984, pp. 246–808. However, see also Tom Gunning, ‘What’s the Point of an Index? Or, Faking Photographs’, in Still Moving: Between Cinema and Photography (Durham: Duke University Press, 2008), pp. 23–40. David Sepkoski, ‘Towards “A Natural History of Data”: Evolving Practices and Epistemologies of Data in Paleontology, 1800–2000’, Journal of the History of Biology, 46(3), 2013, pp. 401–44. See Theodore Porter, ‘Quantification and the Accounting Ideal in Science’, Social Studies of Science, 22(4), 1992, pp. 633–51; Theodore Porter, ‘Statistics and the Politics of Objectivity’, Revue de Synthèse, 114(1), 1993, pp. 87–101; and Theodore Porter, Trust in Numbers: The Pursuit of Objectivity in Science and Public Life (Princeton: Princeton University Press, 1995). Nick Hopwood, ‘Plastic Publishing in Embryology’, in Soraya de Chadarevian and Nick Hopwood (eds.) Models: The Third Dimension of Science (Stanford: Stanford University Press, 2004). For a different take, one that treats nineteenth-century scientific models as art objects rather than publications, see Lorraine Daston, ‘The Glass Flowers’, in Lorraine Daston (ed.) Things That Talk: Object Lessons from Art and Science (New York: Zone Books, 2004), pp. 223–56. This literature is considerable and growing by the minute. See James Secord, Victorian Sensation: The Extraordinary Publication, Reception, and Secret Authorship of Vestiges of the Natural History of Creation (Chicago: University of Chicago Press, 2000); Aileen Fyfe, Steam-powered Knowledge: William Chambers and the Business of Publishing, 1820–1860 (Chicago: The University of Chicago Press, 2012); Aileen Fyfe, ‘Conscientious Workmen or Booksellers’ Hacks? The Professional Identities of Science Writers in the Mid-Nineteenth Century’, Isis, 96(2), June 2005, pp. 192–223; Jonathan R. Topham, ‘Scientific Publishing and the Reading of Science in Nineteenth-century Britain: A Historiographical Survey and Guide to Sources’, Studies in History and Philosophy of Science Part A, 31(4), December 2000, pp. 559–612; and J.R. Topham, ‘A View from the Industrial Age’, Isis, 95(3), September 2004, pp. 431–42. See also the special issue, ‘Book History and the History of Science’, The British Journal for the History of Science, Jonathan R. Topham (ed.), 33(2), 2000, pp. 155–222. See Tony Bennett, The Birth of the Museum: History, Theory, Politics (London: Routledge, 1995). For a similar approach applied to the American context, see William Leach, Land of Desire: Merchants, Power, and the Rise of a New American Culture (New York: Pantheon Books, 1993); Victoria Cain, ‘Nature Under Glass: Popular Science, Professional Illusion and the Transformation of American Natural History Museums, 1870–1940’, PhD Dissertation, Columbia University, 2007; Victoria Cain, ‘Exhibitionary Complexity: Reconsidering Museums’ Cultural Authority’, American Quarterly, 60(4), 2008, pp. 1143–51; and Victoria Cain, ‘“Attraction, Attention, and Desire”: Consumer Culture as Pedagogical Paradigm in Museums in the United States, 1900–1930’, Paedagogica Historica, 48(5), 2012, pp. 745–69. For more on the history of museological display practice in the nineteenth century, see Eilean Hooper–Greenhill, Museums and the Shaping of Knowledge (London: Routledge, 1992); Susan M. Pearce, Museums, Objects, and Collections: A Cultural Study (Leicester: Leicester University Press, 1992); Lynn K. Nyhart, ‘Science, Art, and Authenticity in Natural History Displays’, in Models: The Third Dimension of Science (Stanford: Stanford University Press, 2004); and Samuel Alberti, ‘Objects and the Museum’, Isis, 96, 2005, pp. 559–71. This is true also for plaster casts. See, for example, Ilja Nieuwland, ‘The Colossal Stranger: Andrew Carnegie and Diplodocus Intrude European Culture, 1904–1912’, Endeavour, 34(2), June Downloaded from hos.sagepub.com by guest on December 1, 2015 486 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. History of Science 53(4) 2010, pp. 61–68; and Ilja Nieuwland, ‘“The Wandering Friend”: Andrew Carnegie’s Dinosaur Invades Europe, 1902–1914’, in Joe Kember, John Plunkett, and Jill A. Sullivan (eds.) Popular Exhibitions, Science and Showmanship, 1840–1910 (London: Pickering & Chatto, 2012). See G. Brown Goode, The Principles of Museum Administration (NY: Coultas & Volans, 1895), pp. 37. A similar connection between print and museum specimens has been drawn in a recent essay by Eileen Fyfe. However, whereas Fyfe focuses on the way specimens functioned as objects of “display” in the museum’s exhibition hall and in print publications aimed at a broad, popular audience, this essay extends the analysis in the opposite direction. See Fyfe, Eileen, ‘Reading Natural History at the British Museum and the Pictorial Museum’, in Aileen Fyfe and Bernard Lightman, Science in the Marketplace: Nineteenth-century Sites and Experiences (Chicago: University of Chicago Press, 2007), pp. 196–230. For more on the notion that print and other scientific inscriptions function as an “immutable mobile,” see Bruno Latour, ‘Visualization and Cognition: Thinking with Eyes and Hands’, Knowledge and Society, 6, 1986, pp. 1–40; and Bruno Latour, Science in Action (Cambridge, Mass: Harvard University Press, 1987). For more on the difficulty of stabilizing print, see Adrian Johns, The Nature of the Book: Print and Knowledge in the Making (Chicago: University of Chicago Press, 1998). The specimen and accompanying written material are all housed in the Vertebrate Paleontology Collection of the Peabody Museum of Natural History, Yale University. For more on Marsh’s early expeditions to the American west, see C.W. Betts, ‘Searching for Dinosaur and Other Prehistoric Animal Fossils’, Harper’s New Monthly Magazine, 43(257), 1871. See also Mark Jaffe, The Gilded Dinosaur: The Fossil War Between E.D. Cope and O.C. Marsh and the Rise of American Science (New York: Crown, 2000). Thomas Henry Huxley, ‘On the Animals Which Are Most Nearly Intermediate between Birds and Reptiles’, Annals and Magazine of Natural History, 2, 1868, pp. 67. Thomas Henry Huxley, ‘On the Animals Which Are Most Nearly Intermediate between Birds and Reptiles’, Annals and Magazine of Natural History, 2, 1868, pp. 70, 75. Richard Owen, ‘On the Archaeopteryx of Von Meyer’, Philosophical Transactions of the Royal Society of London, 153, 1863, p. 46. Huxley was not alone in this view. None other than Charles Darwin went out of his way to write a personal letter in which he congratulated Marsh for having furnished “the best support to the theory of evolution, which has appeared within the last 20 years.” See Darwin to Marsh, 31 August 1880, O.C. Marsh Papers, Manuscripts and Archives, Sterling Library, Yale University. For Huxley’s statement, see W.E. Decrow, Yale and ‘The City of Elms’, (Boston: W.E. Decrow, 1882). Huxley’s lecture is reprinted in ‘Prof. Huxley in America’, The Tribune, Extra No. 36, New York, 23 September 1876. Othniel Charles Marsh, ‘Preliminary Description of Hesperornis Regalis, with Notices of Four Other New Species of Cretaceous Birds’, American Journal of Science, 3(17), 1872, pp. 360–65; Othniel Charles Marsh, ‘Discovery of a Remarkable Fossil Bird’, American Journal of Science, 13(3), 1872, pp. 56–57; Othniel Charles Marsh, ‘Notice of New Odontornithes’, The American Journal of Science and Arts, 11, 1876, pp. 509–11; and Othniel Charles Marsh, ‘Characters of the Odontornithes, with Notice of a New Allied Genus’, American Journal of Science, 14, 1877, pp. 85–87. Othniel Charles Marsh, ‘Fossil Birds from the Cretaceous of North America’, American Journal of Science, 5(27), 1873, pp. 229–31; Othniel Charles Marsh, ‘On the Odontornithes, or Birds with Teeth’, American Journal of Science, 10(59), 1875, pp. 403–8. Othniel Charles Marsh, Odontornithes: A Monograph on the Extinct Toothed Birds of North America, Professional Papers of the Engineer Department, U. S. Army, no. 18 (Washington: Government Printing Office, 1880), pp. 113, 115. Downloaded from hos.sagepub.com by guest on December 1, 2015 487 Rieppel 36. Othniel Charles Marsh, Odontornithes: A Monograph on the Extinct Toothed Birds of North America, Professional Papers of the Engineer Department, U. S. Army, no. 18 (Washington: Government Printing Office, 1880), p. 111. 37. Othniel Charles Marsh, Odontornithes: A Monograph on the Extinct Toothed Birds of North America, Professional Papers of the Engineer Department, U. S. Army, no. 18 (Washington: Government Printing Office, 1880), pp. 8–10. 38. See Charles Schuchert and Clara Mae LeVene, O.C. Marsh, Pioneer in Paleontology (New Haven: Yale University Press, 1940), pp. 290–313. Moreover, there is some evidence that the famous American artist, John Haberle, who apprenticed with Berger and Cisand, had a hand in executing these lithographs. See Zelda Edelson and Barbara L. Narendra, ‘John Haberle, A Great American Artist and his Links to the Peabody Museum’, in Discovery: The Magazine of the Yale Peabody Museum of Natural History, 20(2), 1987, pp. 25–30. 39. As Marsh reported in an annual report from 1881, for example, “Since the publication of the volume on Odontornithes a number of sets of casts of the more important bones have been distributed to various museums in this country and in Europe.” See O.C. Marsh, ‘Geological and Osteological Collections’, Yale College in 1881, Archives of the Peabody Museum of Natural History. 40. See Kappeler to Marsh, 5 April 1881, O.C. Marsh Papers, Rare Books and Manuscripts Division, Yale University Library. At times, Kappeler was paid on a daily basis as well, earning a total of $4 per day. See Kappeler to Marsh, 31 December 1874, O.C. Marsh Papers, Rare Books and Manuscripts Division, Yale University Library. For a list of the more than two dozen institutions to which Marsh initially sent replicas of the Hesperornis skeleton, see VP.DOC.0213, Sets of Casts of Hesperornis, Vertebrate Paleontology Archives, Peabody Museum of Natural History, Yale University. 41. See Kappeler to Marsh, 24 January 1880, 21 February 1881, 26 May 1882, and 24 April 1883, O.C. Marsh Papers, Rare Books and Manuscripts Division, Yale University Library. 42. At one point, for example, Kappeler countered what he considered an unusually stingy offer for employment by demanding that Marsh pay him $150 for 30 casts of a Pterodactyl, which comes out to $5 per cast. If this was more than Marsh was willing to pay, Kappeler suggested off-setting some of these costs by selling 3 of the casts to Alexander Agassiz at $10 each, thereby securing 27 casts for himself for only $120 in total. There is no record that Marsh took Kappeler up on this offer, but the fact that he had to propose such a scheme as a novel idea suggests that Marsh was not usually in the business of selling his specimens at a profit. See Kappeler to Marsh, 26 May 1882, O.C. Marsh Papers, Rare Books and Manuscripts Division, Yale University Library. 43. See VP.DOC.0246, ‘Casts of Dinosaur Remains Made by T. Kappeler’, Vertebrate Paleontology Archives, Peabody Museum of Natural History, Yale. 44. Hanna Rose Shell, ‘Casting Life, Recasting Experience: Bernard Palissy’s Occupation between Maker and Nature’, Configurations 12, 2004, pp. 1–40. 45. See Rune Frederiksen and Eckart Marchand, ‘Introduction’, as well as Charlotte Schreiter, ‘“Moulded from the best originals in Rome”’, in Rune Malone and Eckart Marchand (ed.) Plaster Casts Making, Collecting, and Displaying from Classical Antiquity to the Present (Berlin: De Gruyter, 2010). 46. For example, rival firms were often accused of using a competitor’s finished casts to produce a second-generation mold for their own use. See Peter Malone, ‘How the Smiths Made a Living’, in Rune Malone and Eckart Marchand (ed.) Plaster Casts Making, Collecting, and Displaying from Classical Antiquity to the Present (Berlin: De Gruyter, 2010), pp. 163–78. 47. Stephen Dyson, ‘Cast Collecting in the United States’, in Rune Malone and Eckart Marchand (ed.) Plaster Casts Making, Collecting, and Displaying from Classical Antiquity to the Present (Berlin: De Gruyter, 2010), pp. 557–76. Downloaded from hos.sagepub.com by guest on December 1, 2015 488 History of Science 53(4) 48. Edward Robinson, ‘The Cost of a Small Museum’, The Nation, 49(1273), 21 November 1889, pp. 405–6; quoted in Betsy Fahlman, ‘A Plaster of Paris Antiquity: Nineteenth-century Cast Collections’, Southeastern College Art Conference Review, 12(1), 1991, pp. 1–9. For more information on the use of casts in the Boston Museum of Fine Arts, see Walter Muir Whitehill, Museum of Fine Arts, Boston: A Centennial History (Cambridge, Mass: Belknap Press, 1970). 49. See Lawrence Levine, Highbrow/Lowbrow: The Emergence of Cultural Hierarchy in America (Cambridge, Mass: Harvard University Press, 1988); and Alan Wallach, ‘The American Cast Museum: An Episode in the History of the Institutional Definition of Art’, in Exhibiting Contradiction: Essays on the Art Museum in the United States (Amherst, MA: UMass Press, 1998). 50. Thomas Schnalke, ‘Casting Skin: Meanings for Doctors, Artists, and Patients’, in Soraya de Chadarevian and Nick Hopwood (ed.) Models: The Third Dimension of Science (Stanford: Stanford University Press, 2004), pp. 207–41. 51. Lydia Pyne, ‘To Russia, With Love’, The Appendix, 2(4), October 2014, <http://theappendix. net/issues/2014/10/to-russia-with-love> (accessed 19 April 2015). 52. Alan Mann and Janet Monge, ‘Reproducing Our Ancestors’, Expedition, 29(1), 1987, pp. 2–9. 53. ‘List of Casts of Fossils Reproduced from Specimens in the Department of Geology’, 1885, Central Administration Files, DF1011, Printed Forms, Vol. 1, Archives of the London Natural History Museum. 54. Ward describes his early foray into the fossil trade in a series of letters to Chester Dewey dated 18 and 24 September 1856, 7 November 1856, as well as 2 March 1858, File Drawer 1, Catalogues, Henry Augustus Ward Papers, Department of Rare Books and Special Collections, University of Rochester Library. See also Roswell Howell Ward, Henry A. Ward: Museum Builder to America (Rochester, NY: Rochester Historical Society, 1948). For a more recent discussion of Ward’s dealings in natural history, see Sally Gregory Kohlstedt, ‘Henry A. Ward: The Merchant Naturalist and American Museum Development’, Journal of the Society for the Bibliography of Natural History, 9, 1980, pp. 647–61. 55. For more on how Ward secured the position, see Ward to Dewey, 20 May 1859, File Drawer 1, Catalogues, Henry Augustus Ward Papers, Department of Rare Books and Special Collections, University of Rochester Library. 56. Although Ward certainly dominated the market, his was far from the only business of its kind. For more on the history of nineteenth-century specimen dealers, see Mark Barrow, ‘The Specimen Dealer: Entrepreneurial Natural History in America’s Gilded Age’, Journal of the History of Biology, 33(3), 1 December 2000, pp. 493–534. 57. Henry Augustus Ward, Catalogue of Casts of Fossils from the Principal Museums of Europe and America (Rochester, NY: Benton & Andrews, 1866), pp. i–ii. Ward’s claim that students should be directly exposed to the objects of nature fit into a much larger school of thought favoring creative and hands-on learning. Rooted in the work of Johann Heinrich Pestalozzi, the enthusiasm for object-based pedagogy gained traction throughout the nineteenth century. See, for example, George Eduard Biber, Henry Pestalozzi and His Plan of Education (London: J. Souter, 1831); Auguste Pinloche, Pestalozzi and the Foundation of the Modern Elementary School (C. Scribners’s sons, 1901); John Alfred Green, The Educational Ideas of Pestalozzi (W.B. Clive, 1905); and Lawrence Arthur Cremin, The Transformation of the School: Progressivism in American Education, 1876–1957 (New York: Vintage Books, 1961). 58. Ward even quoted the noted anatomist Richard Owen as having stated that a “fossil bone and a colored plaster-cast of it are not distinguishable at first sight, scarcely by sight at all.” See Henry Augustus Ward, Catalogue of Casts of Fossils from the Principal Museums of Downloaded from hos.sagepub.com by guest on December 1, 2015 489 Rieppel 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. Europe and America (Rochester, NY: Benton & Andrews, 1866), pp. iii–vii. Such sanguine assurances aside, from time to time, the problem that plaster casts lacked authenticity did threaten to undermine Ward’s business dealings. See, for example, the letter exchange on mounting a plaster cast Megatherium for William College in the mid-1860s: Karl Gilbert to H.A. Ward, 13, 21 and 23 June 1864, File Drawer 1, Catalogues, Henry Augustus Ward Papers, Department of Rare Books and Special Collections, University of Rochester Library. Quoted in Roswell Howell Ward, Henry A. Ward: Museum Builder to America (Rochester, NY: Rochester Historical Society, 1948), p. 131. In the decades that followed, Ward would continue to use his catalogues as a means to further his own reputation and that of his business. For example, in response to the question “What is a natural science establishment,” a journalist from Scientific American quoted one of Ward’s circulars from the 1890s, which held that it “is a serious scientific institution; a great clearing house to which are brought and from which are distributed objects in all departments of nature.” See ‘America’s Cradle of Natural Science: A Curious Industry Owned by a University to Supply the Needs of Colleges’, Scientific American, 143, July 1930, p. 47. Nick Hopwood, ‘Plastic Publishing in Embryology’, in Soraya de Chadarevian and Nick Hopwood (eds.) Models: The Third Dimension of Science (Stanford: Stanford University Press, 2004). Henry Augustus Ward, Catalogue of the College Series of Casts of Fossils (Rochester, NY: Ezra R. Andrews, 1870), p. viii; and Henry Augustus Ward, Catalogue of the Academy Series of Casts of Fossils (Rochester, NY: Ezra R. Andrews, 1870). Not long thereafter, he also issued a more modest School Series priced at $175. See Ward and Howell, School Series of Ward’s Casts of Celebrated Fossils (Rochester, NY: Ward’s Natural Science Establishment, 1881). For an extensive collection of Ward’s catalogues, see File Drawer 7, Catalogues, Henry Augustus Ward Papers, Department of Rare Books and Special Collections, University of Rochester Library. Munn and Co. to H.A. Ward, Jan–Feb 1864, File Drawer 1, Henry Augustus Ward Papers, Department of Rare Books and Special Collections, University of Rochester Library. Henry Augustus Ward, Catalogue of Casts of Fossils from the Principal Museums of Europe and America (Rochester, NY: Benton & Andrews, 1866), p. vii. For a history of piracy, especially as it relates to science, see Adrian Johns, Piracy: The Intellectual Property Wars from Gutenberg to Gates (Chicago: The University of Chicago Press, 2009). Lydia Pyne, ‘To Russia, With Love’, The Appendix, 2(4), October 2014, <http://theappendix. net/issues/2014/10/to-russia-with-love> (accessed 19 April 2015). Ward to Marsh, 5 February 1873, O.C. Marsh Papers, Rare Books and Manuscripts Division, Yale University Library. See, for example, Ward to Marsh, 1 and 19 May 1, 1873 and 6 February 1884, O.C. Marsh Papers, Rare Books and Manuscripts Division, Yale University Library. For a history of popular science, see the focus section by Jonathan R. Topham (ed.), ‘Historicizing “Popular Science”’, Isis, 100(2), 2009, pp. 310–368. See also Steven Shapin, ‘Science and the Public’, in R.C. Olby et al. (eds.) Companion to the History of Modern Science (London: Routledge, 1990); Roger Cooter and Stephen Pumphrey, ‘Separate Spheres and Public Places: Reflections on the History of Science Popularization and Science in Popular Culture’, History of Science, 32(3), 1994, pp. 237–67; and Peter Bowler, Science for All: The Popularization of Science in Early Twentieth-century Britain (Chicago: University of Chicago Press, 2009). Marsh to Ward, 26 February 1884, Vertebrate Paleontology Archives, Peabody Museum of Natural History. An extensive literature has developed around the so-called “Bone Wars” between Cope and Marsh. See, for example, Henry Fairfield Osborn, Cope: Master Naturalist (Princeton, NJ: Downloaded from hos.sagepub.com by guest on December 1, 2015 490 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. History of Science 53(4) Princeton University Press, 1931); Charles Schuchert and Clara Mae LeVene, O.C. Marsh, Pioneer in Paleontology (New Haven: Yale University Press, 1940); Jane Davidson, The Bone Sharp: The Life of Edward Drinker Cope (Philadelphia: Academy of Natural Sciences of Philadelphia, 1997); David Rains Wallace, The Bonehunters’ Revenge: Dinosaurs, Greed, and the Greatest Scientific Feud of the Gilded Age (Boston: Houghton Mifflin, 1999); and Mark Jaffe, The Gilded Dinosaur: The Fossil War Between E.D. Cope and O.C. Marsh and the Rise of American Science (New York: Crown, 2000). Mark Jaffe, The Gilded Dinosaur: The Fossil War Between E.D. Cope and O.C. Marsh and the Rise of American Science (New York: Crown, 2000), pp. 90–107. See also Edward Drinker Cope, ‘On Some of Prof Marsh’s Criticisms’, The American Naturalist, 7(5), May 1873, p. 269; and Othniel Charles Marsh, ‘Reply to Prof Cope’s Explanation’, The American Naturalist, 7(12), December 1873, p. v. During the mid-1880s, Marsh pleaded with Louis Agassiz to withhold exhibiting casts of his dinosaur fossils until he had published a definitive monograph, explaining that he generally “declined to send anything but the big femur until my volume is out. Cope has already made very bad use of what he saw in my private rooms when he ‘went through’ my Museum.” See O.C. Marsh to Louis Agassiz, 9 December 1885, bAg603, O.C. Marsh Correspondence, Rare Books and Special Collections, Ernst Mayr Library, Museum of Comparative Zoology, Harvard University. See Ward to Marsh, 3 March 1884, Vertebrate Paleontology Archives, Peabody Museum of Natural History. Late-nineteenth-century paleontologists clearly felt they could distinguish between casts taken from originals and those taken from casts. In his letter of praise, Agassiz was careful only to order “a complete series of all those of your casts which were taken from original specimens; the others (casts from casts) I do not care to have.” (Emphasis in original.) See Agassiz to Ward, 21 January 1869, File Drawer 1, Catalogues, Henry Augustus Ward Papers, Department of Rare Books and Special Collections, University of Rochester Library. Ward to Marsh, 6 March and 4 June 1890, Vertebrate Paleontology Archives, Peabody Museum of Natural History. Ward’s Natural Science Establishment, Catalogue and Price List of Casts of Dinocerata and Hesperornis from Originals in Yale University Museum (Rochester, NY: Ward’s Natural Science Establishment, 1891). Charles Schuchert and Clara Mae LeVene, O.C. Marsh, Pioneer in Paleontology (New Haven: Yale University Press, 1940), pp. 295–97. Ward’s Natural Science Establishment, Catalogue and Price List of Casts of Dinocerata and Hesperornis from Originals in Yale University Museum (Rochester, NY: Ward’s Natural Science Establishment, 1891). The relationship to Struthioniform birds was important, as these creatures were recognized to bear a striking resemblance to reptiles. Indeed, modern taxonomists continue to place the ostrich and moa in a basal position on the avian tree, sister to a single lineage that includes both the Galloanserae (fowl, ducks, and their relatives) and the Neoaves (the rest of the modern birds). Copies of the explanatory label for Hesperornis casts marketed through Wards Natural Science can be found with an extant copy of the specimen itself in the Vertebrate Paleontology Archives of the Peabody Museum at Yale. Ward’s Natural Science Establishment, Catalogue and Price List of Casts of Dinocerata and Hesperornis from Originals in Yale University Museum Catalogue and Price List of Casts of Dinocerata and Hesperornis from Originals in Yale University Museum (Rochester, NY: Ward’s Natural Science Establishment, 1891). J.H. Emerton, ‘Remarks on the Restoration of the Skeleton of Dinoceras Mirabilis’, Proceedings of the Boston Society of Natural History, 23, 1888, pp. 342–43. Downloaded from hos.sagepub.com by guest on December 1, 2015 491 Rieppel 81. J.H. Emerton, ‘Remarks on the Restoration of the Skeleton of Dinoceras Mirabilis’, Proceedings of the Boston Society of Natural History, 23, 1888, pp. 342–43. 82. Erwin Barbour, ‘Notes on the Paleontological Laboratory of the United States Geological Survey under Professor Marsh’, The American Naturalist, 24(280), 1890, p. 388. 83. Erwin Barbour, ‘Notes on the Paleontological Laboratory of the United States Geological Survey under Professor Marsh’, The American Naturalist, 24(280), 1890, pp. 392, 394, 395, 397. A similar accusation about being denied access to Marsh’s collections was later echoed by Henry Fairfield Osborn. See Henry Fairfield Osborn, Cope: Master Naturalist (Princeton, NJ: Princeton University Press, 1931). 84. Writing to Louis Agassiz, Marsh complained that “Barbour is one of three or four men whom I have discharged for good cause, and hence is my enemy. Cope has taken them all up and is writing or revising their false statements against me, and using them for his own purposes, especially to make trouble wherever he can.” See O.C. Marsh to Louis Aggasiz, 28 April 1890, bAG603, O.C. Marsh Correspondence, Rare Books and Special Collections, Ernst Mayr Library, Museum of Comparative Zoology, Harvard University. 85. Othniel Charles Marsh, ‘Stegosaurus Armatus, Gen. et Sp. Nov.’, American Journal of Science and Arts, 14(84), December 1877, p. 513. 86. Othniel Charles Marsh, ‘Principle Characters of Jurassic American Dinosaurs, Pt. 3’, American Journal of Science, 19(111), March 1880, pp. 253–66; and Othniel Charles Marsh, ‘Principle Characters of Jurassic American Dinosaurs, Pt. 4’, American Journal of Science, 21(122), February 1881, pp. 167–70. 87. Othniel Charles Marsh, ‘Principal Characters of American Jurassic Dinosaurs, Pt. 9, The Skull and Dermal Armor of Stegosaurus’, American Journal of Science, 34(203), November 1887, p. 413. 88. O.C. Marsh to SF Baird, 20 December 1875, Fldr. 23, Box 29, RU 7002: Spencer Fullerton Baird Papers, Smithsonian Institution Archives. The exchange between Marsh and Baird is also reproduced in Charles Schuchert and Clara Mae LeVene, O.C. Marsh, Pioneer in Paleontology (New Haven: Yale University Press, 1940). It should be noted that despite Marsh’s best efforts, the Hadrosaurus was mounted for the 1876 Centennial Exposition in Philadelphia, after which it was transported to Washington, DC, where it remained on display at the Smithsonian Institution for several decades. For more of Marsh’s misgivings about mounting dinosaurs into freestanding displays, see Othniel Charles Marsh, ‘Restoration of Some European Dinosaurs, with Suggestions as to Their Place Among the Reptilia’, The American Journal of Science, 1(299), 1895, pp. 407–8. 89. In a newspaper account of the Peabody Museum’s public exhibits, Marsh’s collector Arthur Lakes described “cases in which are the bones of the ‘celebrated Odontornithes birds with teeth, lying on the chalk as they were exhumed from the bluffs of Kansas.’” See Arthur Lakes, ‘Yale College: A Walk Through the Misty Museum and Other Departments’, Denver Tribune, 12 February, 1879, p. 4. See also Philip D. Gingerich, ‘Evolutionary Significance of the Mesozoic Toothed Birds’, Smithsonian Contributions to Paleobiology, 27, 1976, p. 26. 90. Ward’s Natural Science Establishment, Catalogue and Price List of Casts of Dinocerata and Hesperornis from Originals in Yale University Museum Catalogue and Price List of Casts of Dinocerata and Hesperornis from Originals in Yale University Museum (Rochester, NY: Ward’s Natural Science Establishment, 1891). 91. For an account of the way these “accumulationist” practices continue to inform modern scientific practice, see Bruno Strasser, ‘Data-driven Sciences: From Wonder Cabinets to Electronic Databases’, Studies in History and Philosophy of Biological and Biomedical Sciences, 43(1), 2012, pp. 85–87; and Bruno Strasser, ‘The Experimenter’s Museum: GenBank, Natural History, and the Moral Economies of Biomedicine’, Isis, 102(1), 2011, pp. 60–96. Downloaded from hos.sagepub.com by guest on December 1, 2015