Waste, Bones, and Society

This Element’s central themes originate from working on a seemingly irrelevant protocol for the history of the sciences of the past, namely the 1819 instructions of Juan Manuel de Rosas (1793–1877), future governor of Buenos Aires, which were compiled for the administrators of his extensive estate in the pampas.Footnote ¹ He emphasized the need for constant observation and the recording of even small events. Every man on the estancia who was able to read and write kept pen and paper at hand to register observations to be transmitted to his superiors. In particular, workers had to keep an eye on the bones of dead livestock in order to maintain the estancia’s cleanliness. Juan Manuel de Rosas stipulated that

“Skeletons of all kinds” ended up including the mighty bones of South American fossil mammals that emerged from the earth every dry season, and which only recently had become objects of study and collection. It is not surprising, therefore, that travelers wandering the pampas in search of specimens of bones found them accumulated in small heaps, formed by the piling up of skeletons in rubbish dumps that eventually could be recycled for the production of bone-based products such as animal-bone glue or agricultural fertilizers.Footnote ³

Rosas’ insistence on gathering all bones into designated dumps created centralized heaps and, crucially, bureaucratic paper trails (notes, reports, instructions) that made those bones visible to travelers and naturalists. Upon entering the urban sphere, these administrative chains of information set in motion a structured process of notification: estate owners were apprised of the finds and, acting as intermediaries, communicated this knowledge to those invested in searches for fossils. Consequently, this flow of information ensured that the bones were rerouted from the bone‑crusher to the analytical purview of comparative anatomists. What this “move” entailed was the circulation of information through a small but efficient administrative chain. The notes kept by mayordomos and overseers were periodically consolidated by the estate administrators, who communicated notable findings to the landowners; these owners, in turn, often informed travelers or naturalists with whom they were connected. In this way, a bureaucratic protocol devised for managing rural labor and cleanliness inadvertently created a structured flow of information that was capable of diverting bones away from industrial recycling and toward scientific analysis.

This example, taken from my work on the fossil dealers from the pampas, shows how happenstance works (see Moynihan’s Element) and how the administration of waste and garbage contributed to making visible those objects previously considered merely as refuse or the raw materials for industry. As this Element argues, visibility is connected to the new economic and epistemological values of the waste and leftovers, reintegrating them into the economic circuit not only as raw materials but also as museum specimens, and providing a new source of jobs for people who would specialize in finding, restoring, assembling, selling, and studying them. These new ventures placed paleontology, archaeology, and related sciences of the past as parts of a new economic circuit of recycled materials.

American historian Susan Strasser, in her valuable work on waste and consumption, identified herself as a scholarly scavenger: “My method was that of the ragpicker,”Footnote ⁴ she wrote, recognizing that she was not the first in comparing historians’ practices with collecting discarded items; others before her had also pursued this research strategy. French poet Charles Baudelaire considered the ragpicker as an urban archivist, someone who sorted through “everything that the big city has cast off, everything it lost, everything it disdained, everything that it broke.” Or Irving Wohlfarth, who studied Paris chiffonniers, describing the historian as a researcher “wandering the archives, sifting through a detritus from the past that has been used for other purposes and recycling the figurative and literal rags he turns up to give them new meaning.”Footnote ⁵ Or the early twentieth-century artists who incorporated recycled materials into their work, including later developments such as French filmmaker Agnès Varda, who claimed that for her, artistic gleaning is about picking ideas, images, and emotions from other people and to make it into a film.Footnote ^6, Footnote ⁷

Thus, social historians and artists have used the metaphor of ragpicking and/or gleaning for their reflections about how they think, how they compose their work in bits and pieces, which John Ulrich, referring to Thomas Carlyle, described as the “paleontological articulation of the past.”Footnote ⁸ While traces of these concerns appear sporadically in the nineteenth‑ and early twentieth‑century historiography of archaeology and paleontology, they were only partially addressed and rarely examined as coherent analytical categories. By contrast, late twentieth‑ and early twenty‑first‑century archaeology has begun to theorize refuse explicitly – from Rathje’s “garbology” to more recent studies of waste and recycling.

At the same time, several strands of archaeological scholarship – ranging from work on fragmentation and ambiguity to recent studies of waste management and recycling in ancient and contemporary contexts – demonstrate that refuse has long been a productive analytical category within archaeology. My aim here is not to deny these contributions but to show that their nineteenth‑century antecedents operated within very different administrative, economic, and epistemic configurations.Footnote ⁹ This volume focuses on those nineteenth-century scientific, administrative, and economic dimensions of waste that have not been systematically analyzed as part of the historiography of the sciences of the past.

This is a significant point given the fact that leftovers, scrap, waste, disposal, debris, and so on, are the core materials of these disciplines, archaeologists in particular being literally waste material dealers. This Element is an attempt to describe the emergence of waste and the rubbish bin as objects of scientific study, and upon which the history of culture started to be tracked.

There is a vast literature dealing with “waste” creation and “disposal” (including reuse). In Australia, for instance, Betty Meehan’s 1975 thesis, Shell Bed to Shell Midden, is a biography of marine shell from collection to discard and reuse. The seminal Tucson Garbage Project, an archaeological and sociological study instituted in 1973 by William Rathje (1945–2012) in the city of Tucson, Arizona, was an effort to understand how garbage deposition patterns are or are not related to socioeconomic status. Rathje not only studied garbage, but framed it as a category of archaeological inquiry. While archaeologists have long worked on modern and ancient refuse, some syntheses in the history of waste management – as Sarah Newman (Reference Newman2023) has recently remarked – remain geographically uneven and often date the consolidation of municipal, industrial-scale systems to the late eighteenth and nineteenth centuries. In what follows, I connect those chronologies of waste management to the parallel emergence of the sciences of the past. If, by 1900, for the first time in human history, disposal became separated from production, consumption, and use, it was also the time when some past waste became national heritage in many countries and was put in museums to be preserved “forever”. Far from being excluded from economic circulation, this waste was transformed into more valued goods. Eventually, the garbage pit as an archaeological site became a mine of new materials that, perhaps ironically, had to be protected from processes of damage and decay and, instead, allowed them to illuminate nineteenth-century concerns about waste such as accumulation and disposal, recycling, soil processes, and chemistry.

Human-Made Heaps

Holocene science required and requires the collaboration of history, archaeology, and earth sciences, including pedology, to study the natural physical processes that affect archaeological sites, providing information about the potential anthropogenic origins of landscapes and geological features previously perceived as of natural origin. While historiography considers that the multidisciplinary approach is relatively new, this Element shows that the science of the prehistoric past emerged in a constellation created by the analysis of soil composition and of structures such as mounds and artificial hills, which revealed two different chronologies: one connected with when they were made, the second, when they were recognized as artificial structures and a monument from the past created by human waste.

Accumulations of waste and remnants create heaps and mounds. German media historian Markus Krajewski has argued that there is a strong connection between this accumulation and the term “world”. Drawing on the work of German etymologist Wolfgang Pfeifer, he traces the New High German term Welt back to the Old High German weralt, which itself originates from the Indo–Germanic root al, meaning “growth” or “nourishment.” This etymological lineage underscores a semantic field in which Welt denotes not only “globe,” “earth,” and “humanity,” but – remarkably – also “toilet,” as documented in the Latin etymological dictionary by Alois Walde and later expanded by Johann Baptist Hofmann in 1954. This latter sense evokes a space where everyday human activities occur, where the remnants of sustenance accumulate and ultimately return to the earth. From this perspective, Krajewski contends that “world” emerges from waste, insofar as discarded matter provides the starting point for new cycles of formation: what appears as garbage reenters the fertile soil as humus or manure, and the assembled remnants become the material basis for renewed vegetal growth and its nourishing fruits. Waste, in this view, is not an endpoint but a generative substrate.Footnote ¹⁰

Also following Krajewski, a similar association between world‑making and accumulation appears in one of Rome’s oldest cosmological and urban concepts: the mundus. Although commonly translated as “world,” the mundus was, in ritual practice, a pit, a hollowed cavity dug at the founding of the city and filled with offerings, first fruits, and even handfuls of soil brought by settlers from their homelands. Ancient authors noted the identity of the name with the world, suggesting that the word designated both the cosmic whole and the earth-filled cavity that anchored it. The mundus thus functioned as a controlled site of accumulation: a receptacle of organic matter, ritual debris, and symbolic residues that periodically opened onto the realm of the dead. In this sense, it was a kind of ritual midden, an early recognition that deposits of refuse – material, vegetal, ancestral – could generate a place, a boundary, and a world.

Such a concept resonates strongly with the historical landscapes of pre‑modern Europe. In the Middle Ages, dumpsites formed from bulky refuse and the drainage of pit privies – or through the work of early street‑cleaning services – were a common feature of some cities. Originally located at the city gates, these accumulations were soon encircled by expanding urban settlements and relocated to the outskirts, where they often grew into substantial hills and mounds. Over time, these formations became absorbed into the urban topography, their human origin forgotten or literally buried beneath layers of soil, vegetation, and memory.Footnote ¹¹ Thus, medieval waste heaps embodied the same logic that Krajewski identified: residues generating new terrains, new forms, and ultimately new worlds.

On the other side of the Atlantic, recent research on the landscapes produced by prehistoric human occupation in Amazonia and the Lowlands of South America describes their similarities to and connections with mounds of natural origin. In the lowlands of South America, earthen mounds less than 30 cm in height, known as aterros or tesos (in Portuguese) and cerritos de indios (in Spanish), are articulated with other built structures such as elongated platforms, borrow pits, tracks, pathways, and artificial lakes. These features create archaeological complexes situated in flooded environments in the pampa from Uruguay, Argentina, and southern Brazil. The cerritos (dated between 4,700 and 200 years BP) include soils of human origin, as indicated by the presence of stone tools, faunal and botanical remains, ceramic shards, and human remains. In the Amazon basin, the study of Terra Preta de Índio or Amazonian Dark Earth, a type of fertile artificial soil with a high content of carbon, has revealed not only its composition but also that this product has a very long history that reflects the constant interaction between human occupation and pedogenetic processes. Studies on the dark earth, the Amazonian mounds devoted to agriculture and the fluvial deposits of shells (sambaquis in Portuguese), dating from the Holocene (8,000 BP)- contributed to the elucidation of an interrelated deep past that shows that the Amazonian forest was not “natural” but the result of thousands of years of human intervention, including agriculture, selection and domestication of trees and plants, as well as cooperation with other animal species.Footnote ¹²

This kind of analysis shows that pedology is a crucial tool for archaeologists trying to understand the past in areas where remains are elusive or difficult to identify with traditional fieldwork methods. In Newfoundland, for example, Canadian archaeologist Amanda Crompton, examining the transatlantic French cod fishery, is trying to understand how mariners have shaped the island’s coastlines since the sixteenth century. Aerial data captured by drone flights combined with soil chemistry analysis revealed the anthropogenic origin of today’s deserted “natural” landscape. The same techniques have been used by Chilean architect Alfredo Thiermann, who, with his team, recorded the accumulations resulting from the nitrate industry established in the Atacama Desert in the early twentieth century. The material, explored through various visual and sonic media, highlighted the alchemical transformation into financial capital, namely nitrates into saltpeter used to fertilize European soils (Figure 1). As they said,

While some “anthropo/geogenetic” processes are rooted in the past, some are still in action and visible in the European chains of hills composed of household waste, war debris, and industrial spoil. In Leipzig’s Auenwald reserve, for instance, Rosenthal Hill was formed between 1887 and 1896 by dumping 120,000 m³ of household waste to create a 20‑meter mound, later landscaped and topped by successive lookout towers, the current one built in 1975. Similar rubbish mounds – Schutteberg and Trümmerberg – were created from the tens of millions of cubic meters of rubble generated by bombing in World War II. Berlin alone has eight such hills, including Teufelsberg, which rises to 114.7 m and contains 12 million m³ of debris. While these hills have been greened and converted into parks, other accumulations continue to grow, such as the potash‑plant tailing piles known as Monte Kali or Kaliberg. The highest, in Heringen (Hessen), reaches 200 m and grows by 9 million m³ per year, with others in Philippsthal, Bokeloh, Sehnde, Wathlingen, Giesen, Zielitz, Sollstedt, and Neuhof. Composed of rock salt, clay, anhydrite, and other by‑products, they also serve as sites for guided tours and open‑air events. The stratigraphy of these human‑made hills – many contemporaneous with the rise of prehistoric archaeology – reflects industrial and urban waste accumulation since the mid‑nineteenth century, shaping new sensibilities around modern and ancient waste landscapes.

Figure 1

22°1'11.2” S 69°37'3.8” W. View of the Buena Esperanza office, the first saltpeter works bought by Henry B. Sloman from Eduardo Squire in 1892. The irregular geometry of the ripio torta is visible. The shape was regularized in later saltpeter works with the introduction of new industrial processes, which were imported from Germany from 1895 onwards. Oficina Buena Esperanza, Cantón del Toco.

Figure 1

A view of the Buena Esperanza office, the first saltpetre works bought by Henry B. Sloman from Eduardo Squire in 1892. The shape was regularised in later saltpetre works with the introduction of new industrial processes, which were imported from Germany from 1895 onwards. Oficina Buena Esperanza, Cantón del Toco. Copyright Alfredo Thiermann (courtesy Alfredo Thiermann).

© Alfredo Thiermann (courtesy Alfredo Thiermann)

In various ways and places, these examples show how diverse forms of discarded matter – organic residues, industrial by-products, and the humble debris of daily life – became entangled with the emerging sciences of the past, while the bureaucratic traces generated in managing these materials created parallel channels through which waste acquired visibility, value, and historical meaning. Through these interwoven material and documentary circuits, nineteenth-century scholars learned to read refuse not merely as detritus but as a stratified archive that linked local practices to global networks of extraction, circulation, and comparison.

The following section builds upon this foundation by turning from conceptual framing to the concrete infrastructures and economies that governed the movement of waste: it examines how excreta, bones, scraps, and their attendant administrative residues circulated between countryside and city, and how these flows – at once economic, material, and epistemic – created the conditions under which refuse first became legible as scientific evidence. Section 1, therefore, begins by examining the nineteenth‑century economic and administrative systems. Later sections return to the question of how waste, value, and scarcity shaped collecting practices and the parallel economies around natural history specimens.

The Soil We Walk On

Nineteenth-century farmers, chemists, investors, and scholars of all kinds were turning over the soil they walked on, not only the deep geological strata but the shallower organic components and nutrients of the most recent sediments, analyzing them from a chemical point of view and recording the objects and organic remains and materials they found in it.Footnote ¹⁸ The literature has already shown that the history of archaeology is related to the engineering works that, since the eighteenth century, have been uncovering buried structures, such as the cities of Pompeii and Herculaneum or the Roman ruins being unearthed in Spain. But in the nineteenth century, unearthing also took place in smaller excavations and more humble spaces. While the history of archaeology is well known to be deeply entangled with geology – stratigraphy, uniformitarianism, and other geological principles having shaped the discipline since the nineteenth century – the explicit incorporation of pedology into archaeological method emerged only in the mid‑twentieth century. By contrast, nineteenth‑century observers rarely differentiated geological processes from anthropogenic soil change produced through waste practices. Clarifying this distinction helps to situate the present argument historiographically: the period examined in this Element did not yet conceptualize “soil analysis” as a separate scientific domain, even though waste‑shaped strata were already central to the formation of archaeological sites.

Histories of archaeology have rightly emphasized the discipline’s long entanglement with geology and with techniques of recording and training; even when they are concerned with the same protagonists. What has been followed less systematically is how practices of waste disposal and anthropogenic soil change shaped the very structures archaeologists excavated, and how chemists, agronomists, and (later) soil scientists framed those deposits as data worthy of analysis. As this subsection argues, the history of archaeology is also the history of soil science, investigations into peat, and the development (and fall) of the guano trade.

The word “pedology” first appeared in German in the works by Friedrich Albert Fallou (1794 – 1877), who, in 1857, published “First Principles of Soil Science” followed by “Pedology or General and Special Soil Science” in 1862. In his work he developed his field observations of soil into a systematic approach and made the case for the recognition of soil science as a discipline, published in the Journal for German farmers.Footnote ¹⁹

It was becoming clear that in addition to physical forces, plants and animals, including humans, were heap- and soil-builders and/or destroyers. In 1844, the Swiss Léo Lesquereux (1806–1889), who would become an expert on the peat bogs of Europe and North America, published Quelques recherches sur les marais tourbeux en général devoted to restoring the Neuchatel region’s bogs due to concerns about the fate of their timber industry. Lesquereux was commissioned by the King of Prussia to study the peat swamps of Jura, and on the coasts of the North and Baltic Seas, and later Denmark, Sweden, and Norway. While the chemistry of peat bogs was hitherto unknown, peat was crucial for several industries and had been used for combustion since ancient times. In Denmark, as we will see in the next section, Japetus Steenstrup devoted a monograph on the typology of peat bogs, and in so doing described also the archaeological remains that they contained.

The formation of organic soils was a worldwide concern. When Charles Darwin published his last book in 1881 The Formation of Vegetable Mould Through the Action of Worms, with Observations on their Habits it sold thousands of copies in its first few weeks. This success was not only because of Darwin’s fame but also because his experiments on the movement of ancient and modern objects deposited in the Earth addressed the hot topics of the day: what to do with things that accumulated on the surface? How fast did things disappear? How could vegetable mould be produced?Footnote ²⁰ Darwin, for instance, analyzed the part which worms have played in the burial of ancient buildings, calculating the amount of earth brought to the surface by worms by the rate at which objects on the surface are buried and by weighing the earth brought up in a given time. Information from farmers on marl, cinders, and so on, sinking into the ground helped Darwin to make his calculations. Objects of all sorts “work themselves downwards” as farmers say. He visited Stonehenge and found some outer stones partly buried, the turf sloping up to meet them (Figure 2). Large stones sink because worms fill up any hollows with castings, then eject them beyond the perimeter, and the ground around them starts to rise. Darwin weighed castings per unit area per year, then worked out how thick a layer casting would make, compared with rates of sinking.

Figure 2

Section through the foundations of a buried Roman villa at Abinger. A A, vegetable mould; B, dark earth full of stones, 13 inches in thickness; C, black mould; D, broken mortar; E, black mould; F F, undisturbed subsoil; G, tesseræ; H concrete; I, nature unknown; W, buried wall (Figure 8 of Charles Darwin’s The formation of vegetable mould, through the action of worms, with observations on their habits. London: John Murray, 1881)

Figure 2

Section through the foundations of a buried Roman villa at Abinger. A, vegetable mould; B, dark earth full of stones, 13 inches in thickness; C, black mould; D, broken mortar; E, black mould; F F, undisturbed sub-soil; G, tesseræ; H concrete; I, nature unknown; W, buried wall (Fig. 8 of Charles Darwin’s The Formation of Vegetable Mould).

(Museo de La Plata Library)

The discussion of mould, worms, and burial rates sets up a broader nineteenth‑century obsession with accumulation and depletion. The next example – guano – shows the same problem at another scale: deposits that are simultaneously ecological archives, agricultural inputs, and objects of extraction.

The rate of soil deposition in the formation of peat bogs, mould, and other natural/artificial structures was key to several economic activities. This included the exploitation of guano deposits, such as those from the three tiny Chincha Islands off the southern coast of Peru that have served as home for seabirds, possibly for millennia. The birds fed in the rich waters packed with fish but an absence of predators, allowing their droppings to accumulate onshore to a depth of up to 60 m. The dry weather and cool ocean currents there maintained the guano’s nitrate-rich quality. However, by the late 1870s, nearly all of the Chincha Islands’ guano was gone, and the seabird habitat was ruined by the mining operations (Figure 3).

Figure 3

“Chinamen working guano. Great Heap, Chincha Islands, 1865, Chincha Islands (Peru), Guano industry, Peru,” Albumen silver print, Print Size: 8 3/4 × 6 3/4 in. (22.2 × 17.1 cm), Mat / Mount: 17 × 12 in. (43.2 × 30.5 cm), Date Issued: 1865, Place: Washington, D.C., Publisher: Philp & Solomons, “Guano’s heavy toil, Chinamen’s sweat-drenched labor yields, Peru’s rich deposit.”

Figure 3

Chinamen working guano. Great Heap, Chincha Islands, 1865, Chincha Islands (Peru), Guano industry, Peru, Albumen silver print, Print Size: 8 3 by 4 times 6 3 by 4 inches (22.2 by 17.1 c m), Mat / Mount: 17 by 12 inches (43.2 by 30.5 c m), Date Issued: 1865, Place: Washington, D.C., Publisher: Philp and Solomons, Guano’s heavy toil, Chinamen’s sweat-drenched labor yields, Peru’s rich deposit.

(Sepia Times/Universal Images Group via Getty Images)

In the early nineteenth century, farmers and chemists worldwide claimed that Chincha Islands guano was the world’s finest fertilizer and that farmers needed it to replenish the nutrients in their fields and increase their crop yield. Guano produced by cormorants, boobies, and pelicans off the coast of Peru had a remarkably high nitrogen content that promoted plant growth. Beginning in the early 1840s, it was commercially mined, transported, and sold in Germany, France, England, and the US. American farmers, especially in the mid Atlantic region, lobbied their congressmen to get more and more of the acrid stuff to replenish the soils of their tobacco and produce farms, which could yield three times the normal growth with an application. American shippers and their business partners were allowed to seek and claim islands in the Pacific and Atlantic for this trade. Within a few years, nearly 200 guano island claims were made.Footnote ²¹

Around the same time, huge deposits of nitrates and phosphates were found on land, and for the first time, fertilizers could be blended to local needs, so the mining of guano islands ended without lasting impact on the world’s agriculture or history. While the Americans exploited some 200 islands in the Pacific and the Caribbean, the North Atlantic also provided several deposits, such as those from Fogo and Funk Islands, the latter a small, barren, isolated, uninhabited island approximately 65 km northeast of Musgrave Harbor, Newfoundland, Canada. The name Funk, which means evil odor or vapor, is thought to have been given to the island because of the foul odor emanating from the nitrate and phosphate concentrations found in the guano deposited by the many millions of birds that have nested there over the centuries.

The National Museum of American History in Washington, DC, has devoted an exhibition to the links between the Smithsonian and the trade in guano (See the exhibition here: https://wayback.archive-it.org/3340/20160510135136/http://americanhistory.si.edu/norie-atlas/atlas. In March 1858, the US Navy Secretary asked the Smithsonian’s first Secretary, Joseph Henry (1797–1888), to analyze samples of seabird guano from the Pacific New Nantucket and Jarvis islands. Three months later, Henry reported that “the deposits submitted to examination do not possess the peculiar characteristics of Peruvian guano … and are not equal to it in value.” The second Smithsonian Secretary, Spencer Baird (1823–1887), was also involved in the guano business through his work at the port of Woods Hole, Massachusetts. He was interested in mixing nutrient-rich meal from the abundant menhaden fish with the rapidly depleting Pacific guano to extend its supply. In 1875 he urged the Pacific Guano Company of Woods Hole to exhibit its products at the 1876 Philadelphia Centennial. Several specimens kept in Natural History collections from Europe and North America are a subsidiary result of this trade.

The preceding examples of various waste economies developing in the late nineteenth century show how different types of refuse became materials of significant economic status and commerce, and significant enough to be celebrated as such in contemporary exhibitions and collections. In the following section I show how the recognition and legibility of refuse was mirrored in the burgeoning discipline of archaeology through the definition of a new kind of evidentiary site type. The identification of mounds of shell, food refuse, and other domestic residues as kitchen‑middens provided a rich new source of data about the recent past. However, as these middens bore evidence that linked human actions to recently vanished fauna, the new site type also linked the historical period with deep time. My narrative, therefore, turns to the evidentiary frameworks that shaped nineteenth‑century debates on extinction – particularly those emerging from St. Petersburg’s zoological networks. Having traced how nineteenth-century waste economies and administrative practices made discarded matter newly visible as a resource for scientific inquiry, we now move from the new economic and material circuits to the classificatory frameworks that emerged from them.

The Short Co-Existence of Humans and Sea-Cows

In 1840 Karl Ernst von Baer published in St. Petersburg the story of the extermination of the northern sea cow, a large, slow-moving marine mammal that, until the mid-eighteenth century, had lived in the North Pacific islands between Asia and North America. It was a rather short history: the cow had become extinct by the late 1760s, just a few years after its first description in 1741 by Georg Alexander Steller, the Bavarian naturalist and physician who accompanied Vitus Bering’s Russian expedition to the coasts of Alaska and Kamchatka. After a shipwreck, which occurred in the vicinity of an uninhabited and unnamed island, but which has since sheltered the remains of the illustrious commander and bears his name, Steller had the opportunity to observe, eat, dissect, and draw this hitherto unknown animal. According to Steller, its abundance guaranteed food for all the Kamchatka adventurers who set out to hunt seals and other fur-bearing sea mammals. This proved to be a reckless claim that sealed the animal’s fate: in 1768, the last of Steller’s sea cows was killed or, at least, from that year on, no one saw them again. The islands, incorporated into the Russian administration through the Russian–American Company, were dedicated to the marine fur industry and began to be populated by administrators and hunters, many of them Aleuts, experts in the art of tracking their prey in the icy landscape of the strait named after the Danish-born seafarer Bering.

Steller died in 1746. He never returned to his port of departure, but some of his drawings and notes did reach it. In these notes, he claimed that the beast he discovered was a manatee, the Vacca marinae of the Dutch, the sea cow of the English. He thus transferred to an animal from the North Pacific the name of a mammal that lived in more southern areas, in the rather warm regions of Africa, America, and the coasts of the Indian Ocean and the southwest Pacific. No one really knew how the manatees of the different regions differed, or how they were related to each other or to other groups in the animal kingdom. Very shortly before Steller’s posthumous reciprocal encounter with his cow, Peter Artedi (1705–1735) and Linnaeus had placed them in the fish group. It was not until 1756 that Mathurin-Jacques Brisson (1723–1806), the first zoologist who created the class of cetaceans separate from fishes, would place manatees and dugongs among the quadrupeds.

When Steller’s notes were published by the Academy of Sciences in St. Petersburg in 1751 – five years before Brisson’s publication – plate XIV showed the lower and upper masticatory plates of the northern sea cow. Without a clear idea of the shape of their teeth, which were absent from European collections and from the pictures accompanying the descriptions of animals known as manatees, no one could say exactly what they ate and, therefore, to which class they belonged.

Despite the evidence presented by the Russian zoologists, the ritinas, a name between scientific and common that was used to designate them, were considered a living species until well into the nineteenth century. Thus, in Georges Cuvier’s description of the “northern manatee” published in the various editions of his Ossements fossiles (1823, 1836), no one thought that the call for an exact drawing of the animal taken from the living one could not be fulfilled even if a single depiction had been sufficient to meet all the needs of the modern anatomist. Although Cuvier was wrong about the animal’s possible survival, he was right to identify that more accurate information was needed to classify it, that is, the bones, real or drawn. With the exception of Steller’s sketch – lost in the consignment that arrived in St Petersburg – and a rather vague drawing published in Pallas’ work (Figure XX), naturalists had only one proof of the existence of this animal, namely two ribs and a dental plate, both deposited in the collections of the Zoological Museum of the Academy of Sciences in St Petersburg. The images finally appeared but the story of how that happened is extraordinary and involved several attempts to fix an image of animals that had become extinct in historical times, had been seen by very few travelers, and of which not even the bones remained.

The new evidentiary frameworks for recently vanished fauna, particularly those emerging from St. Petersburg’s zoological networks, shaped nineteenth‑century debates on extinction, required natural scientists to theorize extinction and, in so doing, move from scattered observations to broader conceptual, linguistic, and legal debates.

Extinction: A Matter of Fact

Until the nineteenth century, the word “extinction” in European languages was used to refer either to the end of a contract or to the inability of the male line of noble families to produce heirs. The transfer of extinction from the law of nations to natural history began with the assumption that there were species that had disappeared from the face of the earth forever. Historiography generally assumes that Georges Cuvier introduced extinction as a fact in his work on living and fossil elephants. However, the word “extinction” did not appear in Cuvier’s work. In the four volumes of his Recherches sur les ossemens fossiles from 1812, the adjective “éteint” (extinct) appears only four times: as a mere synonym for “espèces fossiles, perdues, détruites” and in the Discours préliminaire, where Cuvier describes the traces of extinct peoples (“Les traces presque efacées de tants de nations éteintes”) with those of the so-called revolutions of the earth.Footnote ³⁶

The use of the word “éteint” did not refer to the result of an extinction process, as per Darwin, but was intended to mean a sudden disruption of the order of nature through catastrophes that regularly created new orders. As with revolutions in history, the temporal process of the disappearance of the old was short and violent, so it is not surprising that Baer would later compare extinction to a war of extermination. For Cuvier, however, the bones of the fossil quadrupeds did not record a process, but only the fact that life was different in the deep past. The reason for the changes, therefore, had to be sought not only in the bones themselves, which only served to testify to the existence of species unknown in the present. Cuvier referred to “temps moindre” (shorter time), the observation of which would help to understand deep time. This shorter time of a few thousand years extended to the temporal limit that could be reached through historical sources, whether in the form of texts, images, or monuments. He used the writings of the ancient Greeks and Romans, as well as modern travelogues, to argue that the number and type of almost all large living quadrupeds had been identified and that no new species could or would be found. Cuvier’s observations, which have already been examined in the context of his dispute about species transformation, also make it possible to analyze the use of historical evidence in the practice of comparative anatomy.Footnote ³⁷ But what is even more important are the media that were used to record the history of animals. Cuvier examined images of animals engraved on Egyptian obelisks and the mummies of animals, such as the ibis, found during Napoleon’s Egyptian campaign. Both forms of evidence showed an astonishing similarity between the animals of the present day and those from the time of the pharaohs, which led him to conclude that three thousand years had not actually brought about any changes. While Cuvier used the sources of antiquity, for Baer these sources were not sufficient to answer the question of extinction during this period. New methods and media were required. The central problem was how to register the disappearance of a natural entity and compile observations of something that was happening in real time and on scales of difficult calibration.

New Fossils

In contrast to the number of extinct species being identified geologically in the nineteenth century, the number of historically recorded extinctions, Baer remarked, was not large. While the strata from deep geological time provided huge numbers of bones and the petrified remains of animal forms that the human eye had never seen alive, the historic sources seemed to indicate no important changes in the fauna of the last millennia. However, the invisibility of the occurrence of historic extinction was due to culture, language, and recording systems; namely, linguistic differences and the lack of a universal, systematic register of animals in the eighteenth century. Baer took the year 1766, corresponding to the publication of the twelfth edition of Linnaeus’ Systema Naturae, as the beginning of the systematic recording of extant animals. As he considered that the depictions of animals in woodcuts and engravings were a reliable recording device, he decided that the record could be extended, in a very limited amount of cases, to 300 years before the present. However, the comparison of traditional or historic animal names with the current register of extant species was never straightforward: understanding what was in an animal name involved philological and historic linguistic research to track transfers and displacements from one kind of taxon to another. Thus, accounting for current extinction confronted the zoologists with the critical study of traditions and historical sources. Baer referred to the problem he described as Namen-Übertragungen (transfers of names), that is, the transfer by travelers, hunters, seamen, soldiers, naturalists, colonists, and so on, of the popular name of an animal kind living in one particular region to a similar animal kind from another part of the world. Thus, the transfer of names from one linguistic tradition to another created apparent continuities in time and space and, often falsely, expanded the geographical distribution of a certain animal kind and prolonged its existence in another place.Footnote ³⁸ Baer believed that historians, philologists, and paleontologists should work together to show the delusional character of this apparent permanence and stability. He predicted that, thanks to the systematic register of species, the ongoing nature of extinction would become much more visible in the centuries to come.Footnote ³⁹ For historic extinctions to become visible, Baer showed that the history of life and the discipline of paleontology needed to be tied to the practices of archeology and history, the reading and understanding of several European languages, fieldwork, and ethnographic research.

While historic depictions, names, and descriptions were not entirely reliable, the new science of prehistory created a concrete, physical register of fauna beyond the records started in the eighteenth century. The excavation of prehistoric sites revealed not only human tools and remains, but also the bones and fragments of the animals hunted and eaten by prehistoric or protohistoric societies. But excavations also provided the bones of species that had been observed alive, but not scientifically described, in historic times. These are what I propose to call “new fossils” and were the remains of animals observed alive in the early years of the Linnaean system, but which became extinct before they could be fully described following the standards of the nineteenth century, that is, the comparative anatomy of the skeletons and skulls, and especially, the study of their teeth and limbs.Footnote ⁴⁰ These “new fossils” exemplified how nineteenth‑century naturalists relied simultaneously on archaeological deposits and living memories – a dual register that defines the evidentiary structure of this section.

Naturalists’ own eyes have never seen the extinct animals that lived in the geological past. Their reconstructions were and are based on fragments and the principles of comparative anatomy. This enabled them to connect teeth and feet to functions, forms, and the shape of existing species. Colors, sounds, taste, and texture were qualities later attached to those mute bones that late eighteenth-century preparators and anatomists had been mounting as hypothetical skeletons. The fossils of historically exterminated animals, on the contrary, were of a different nature: people still living had seen and named them, they still resided in the memories of fishermen, mariners, peasants, fowlers, and hunters; these animals had colors, sounds, and movements. They had been touched, traded, and eaten. It was not only in the sediments of Earth that their fragmentary remains were to be found: these animals had been depicted and described. Emerging from the records of history and ethnography, the gathering of the evidence of recent extinction demanded the combination of the skills of the naturalist with those of the man of letters. And, more importantly, it required defining how to register the vanishing of a natural entity, how to compile observations of something that was not going to be observed again.

Russian and Danish collections had gathered material evidence of these recently vanished species. Steller did not know that his sea cows, whose herds, he believed, were so abundant that they could feed the whole Kamchatka, were going to become extinct two decades after his own death in 1746.

Bering’s second voyage made known the valuable resources of the Commander Islands, stimulating the fur trade and many expeditions among the Aleutians in search of other regions inhabited by fur seals. Immediately, numerous trading companies began to develop, nearly resulting in the destruction of the fur seal herds. As an example of the dimension of the slaughtering from those years, Baer mentioned that in 1745 a single Cossack´s boat transported around 1,900 fur seals and 1,670 see otters’ furs, more than one and a half what the American company traded in a year.Footnote ⁴¹ These figures were to be found in the expedition reports, which accounted for the amount of furs as part of the requisites implemented by the regulation of the trade: to get the permission to hunt sea otters and fur seals, one-tenth of the game had to be given to the Russian State.Footnote ⁴² From these reports it became obvious that the numbers of animals was decreasing. To prevent the extermination of the herds, the fur trade passed into the control of a single company, the Russian–American company, established in 1799.Footnote ⁴³ While this move saved the herds of sea otters and fur seals, the sea cows, one of the most important sources of food and fat for the mariners and traders, disappeared. Following the reports of an expedition led by Joseph Billings, an English officer of the Russian Navy (1783–1793), the last records of the existence of the sea cow date from 1768, when the last specimen was killed. In 1838, Baer remarked that, although there were naturalists who thought the animal still lived between Kamchatka and America for longer than fifty years, there had been no news about the existence of this animal. The idea that the sea cow retired to islands further north than Bering`s was not supported. The sea cow must be considered extinct and the only way in which to obtain a specimen of it was to exhume a well-preserved skeleton from any of the shores that they used to inhabit.Footnote ⁴⁴

Ilya Gavrilovich Wosnessenski (or Voznesenski), traveler–naturalist and conservator of the Academy Zoological Museum, was sent in 1839 to Alaska, by and at the cost of the Imperial Academy of Sciences of St. Petersburg, to make collections. He sailed on a Russian American Company’s ship from Kronstadt and reached Sitka (New Archangel), the capital of Russian America in Alaska, in 1840. He traveled on the American coast from Upper California (1840–1841) to Kotzebue Sound and the Bering Sea (1842–1843), the Aleutian Islands (1842–1843), the Kurile Islands (1844), and into the Sea of Okhotsk and Kamchatka (1845–1848). He returned to Sitka at the end of July 1849. He found an incomplete skull of the sea cow on Bering’s Island after spending eight years trying to find a live animal.Footnote ⁴⁵ Described by the director of the Zoological Museum of the St Petersburg Academy of Sciences, the German Russian naturalist Johan Friedrich von Brandt (1802–1879), the skull encouraged further bone-seeking. The Academy, following Baer and Brandt´s advice, called for a prize to obtain more remains of Rhytina gigas.Footnote ⁴⁶ A complete, big skull, ribs, scapulae, a mandible, and occipital bones, found in different areas of the Bering Strait and the northern seas, were shipped to Saint Petersburg by officers of the Russian–American Fur Company.Footnote ⁴⁷ A nearly complete skeleton arrived in the early 1860s, which made some European zoologists think that the sea cow was really still alive. Brandt clarified that it had been disinterred.Footnote ⁴⁸ (Figure 6)

Figure 6

Skull of Rhytina stelleri as described by Brandt (public domain)

In 1861, Alexander von Nordmann (1803–1866), professor of zoology at the Imperial Alexander University of Helsingfors in the Russian Grand Duchy of Finland, described in detail the skeleton of an immature sea cow. It was collected following instructions given in 1858 to the Finnish–Russian vice-admiral Johan Hampus Furuhjelm (1821–1909), Governor of the Russian–American colony. In December 1860, Furuhjelm reported that two Aleuts had found a sea cow skeleton on Bering Island. These bones appeared to be genuinely associated, unlike all the other existing skeletons, which were museum composites. Once in Helsinki, Nordmann took a photograph of the complete skull and sent it to his colleagues Brandt and Baltic–German zoologist Leopold von Schrenck (1826–1894), both at St. Petersburg.Footnote ⁴⁹ (Figure 7)

Figure 7

Skeleton of Rhytina stelleri as mounted by Nordmann (public domain)

However, the debate about the status of the species continued. Baltic–German geologist Karl Eduard von Eichwald (1795–1876), author of the Paleontology of Russia, suggested that it was difficult to assess the extinction of sea animals because oceans were practically unknown, and, therefore, that the claim of extinction was premature.Footnote ⁵⁰ Eichwald insisted, too, that there was always the possibility of hidden places to which poorly known animals could have moved: the ocean depths, the peaks, the poles, the jungle. A disinterred skeleton proved that an individual had been there, but it was not enough to say anything about the subsequent history of its congeners. As had happened with the acceptance of the mere idea of extinction, the possibility of migration and the existence of refuges in some remote corners of the world moderated the acceptance of total obliteration in contemporary times.

Almost at the same time, A. impennis attracted the attention of English oologist and egg dealer John Wolley (1823–1859), who had become acquainted with Steenstrup during a visit to Copenhagen in 1856. In April 1858, Wolley, accompanied by British ornithologist Alfred Newton (1829–1907), sailed to Iceland to track the elusive great auk.Footnote ⁵¹ They had become aware that, since the late 1830s, there were warnings that the great auk was to become extinct, having long disappeared from the coasts of Norway, the Faroe, and Iceland. Up to the nineteenth century, the bird and its eggs had been much used as a food supply by mariners doing the North Atlantic routes. In the 1840s there was still a constant, though very limited, supply of specimens, which kept dribbling into the market of natural history.Footnote ⁵² Old narratives and voyages to the Faroe Islands mentioned that the inhabitants of those regions eviscerated penguins, thrusting a wick into the cavity of its stomach, setting fire to it, and letting this singular apparatus burn as if it were a lamp, so very fat and oily was the bird. On a little island near the coast of Newfoundland they burned these birds, for want of other fuel, as if they were logs of wood, and in this way they cooked one individual by the help of his companion.

While in Iceland, Wolley and Newton gathered evidence on the former distribution of the great auk, the number of birds fowled and remembered by the informants, the spots where they used to breed, and where they had been hunted or seen by others. Newton divided the evidence into two classes: that which may be considered documentary and could be found in libraries and that which was merely oral, offered by living witnesses, either as traditional information handed down from one generation to another or as personal observation.Footnote ⁵³ Newton elaborated on the information-gathering process and explained that Wolley wrote much of the evidence down “in a most painstaking manner, having passed through an interpreter.” All of these explanations referred to the problematic nature of the disappearance of a species because such a claim relied on the observation of absence and the problem of transferring words from one linguistic tradition to another. It became clear that a critical analysis of the historical sources should therefore be applied to the study of animals mentioned within them. Fragmented and contradictory versions of reports were examined, cross-examined, and reexamined, but still, the inaccuracies and contradictions remained, as did the possibility of the survival of the birds on some remote islet.

The elusive and contested character of the evidence appeared over and over again, not only regarding the existence of the bird but also regarding geological events and the identification in the maps of the spots where the bird had bred. Volcanic eruptions were dated in the Icelandic historic records and, as such, incorporated into the history of the extinction of the great auk.Footnote ⁵⁴ Newton, Wolley, and the English doctoral student William T. Preyer (1841–1897) investigated in the historic annals of Iceland the volcanic events that could be correlated with the destiny of the great auk.Footnote ⁵⁵ Thanks to this combination of historical research and the study of natural events, Preyer identified all the outbreaks and eruptions off Cape Reykjanes, beginning in ad 1219. In 1240, during the eighth historic eruption, a number of small islands were formed but afterward mostly disappeared. The twelfth eruption occurred in 1830, when the sea stack on which the great auk bred disappeared beneath the waves. This event had a most important impact, as it greatly hastened the extinction of the species, compelling the birds to seek a home on skerries that were easier to access from the mainland.Footnote ⁵⁶ Newton questioned the data gathered by Preyer and his statement that in March 1830, shortly after the volcanic submergence of the skerry, a colony of birds appeared at Eldey, an almost circular sea stack, where the so-called “last pair of great auks” was supposedly killed in 1844. Newton questioned not only the formation of the Eldey colony in 1830, but also thought that Preyer had wrongly identified the location of the birds’ previous breeding localities

However, another kind of evidence helped to track the natural history of the bird in a most conclusive manner. Wolley and Newton spent some weeks at Reykjavik and then departed for a hamlet, close to the rocky islet where the bird was presumably last seen. At Cape Reykjanes, Wolley had picked up from a heap of blown sand two or three birds’ wing bones – bone-seeking having become one of their main occupationsFootnote ⁵⁷ – and compared them to those depicted in Steenstrup’s engravings. They dug in the kitchen refuse of the place, but contrary to the Danish mounds, no bird bone was found. Instead, two or three bones were retrieved from the wall of a churchyard, sticking out of the turf used to keep the stones in place. In another spot they found remains that belonged to at least eight individual birds, many of them bearing marks of a knife. In contrast to the inaccuracy of some historic sources and locally sourced information, the successful search for bones of Alca impennis had anchored the existence of the bird to some concrete locations.

Artifacts of the State

Wolley, the egg collector, died soon after his return to England, but not before he had donated his collection to the Norwich Museum and his manuscript notes on the great auk to Alfred Newton, who would champion Wolley’s calls for a register of the existing British birds. Baer had also considered a register of species as the requisite for understanding and “observing” extinction processes in real time. But, as we shall see, the mere introduction of this idea created the problem of how to assess the changes occurring in time and space.

The relationship between hunting, game, and preservation/ conservation movements is not new in the literature.Footnote ⁵⁸ Less attention, however, has been paid to the registers of hunting activities and the accounting of cattle. While accounting for domesticated herds was a practice as old as the domestication of animals,Footnote ⁵⁹ accounting for the numbers of wild animals required new procedures and strategies. While there were examples of methods for counting cattle and registers of game, both relied on the organization and administration of landed properties and state resources.

As mentioned before, from the late eighteenth century, Baer worked with the records of the Russian American colonies, which indicated the number of traded furs. By extrapolating the number of animals the furs represented, he analyzed the decrease or the status of the animal herds over the years. The register of the fur trade was also a means to transform the animals into artefacta produced by the work of the State in the sense defined by Kant for sheep, cattle, and domestic fowl:

The records of the fur trade and the issuing of hunting and fishing permits have to be understood in the frame of the new policies that characterized the administration of the resources of the State at the end of the eighteenth and in the early nineteenth centuries. From the Danish Kingdom to the Spanish and Russian empires, whether in the North or South Atlantic or on the Pacific coasts, the exploitation of some animals was regulated by the State: to fish or to hunt whales, sea otters, cod, or fur seals required gaining permits, the payment of taxes, and the submission of reports. Other measures emerged from specific circumstances, such as the potential decrease in the revenues of the fur trade or the intervention of the Danish State to counterbalance the effect of the Icelandic famine that started in 1783 after the eruption of a submarine volcano and subsequent earthquakes. As fisheries were ruined, livestock died, and the population declined, Danish commissions tallied the numbers of people and animals affected by this natural disaster, producing reports that were later going to be read by British and other European naturalists traveling to Iceland.Footnote ⁶¹ When Baer referred to “artificial precautions” he meant also these bureaucratic means that transformed animals into artifacts of the State, and gave them legal protection. Administration, paperwork, and policies governed these animals, and provided the conditions to regulate exploitation and to avoid extermination. As Baer showed, animals such as the sea cow and the great auk could be annihilated if left “on their own.” Extermination, from this point of view, was a collateral effect of the exploitation of the regulated species.

The methods for counting hunted animals and the study of prehistoric remains and written sources are related only to the study of phenomena that had already occurred. The present, however, remained elusive. On the eve of taking the human census of the British Empire of 1861, Alfred Newton proposed the idea of a somewhat similar enumeration as regards the fauna of England. However, assessing the number of living animals in England seemed even more difficult than accounting for people in the British Empire. With this in mind, he communicated to the Linnaean Society some remarks bearing upon the question of a general zoological census and called the attention of ornithologists to the particular application of a project that Wolley had proposed before passing away: a census of the birds of England taken with the involvement of nearly all the ornithologists in the country. Newton wished to make a few observations on the subject; however, he remarked: “I am not at present prepared to offer any suggestions as to the method to be employed.”Footnote ⁶²

Among Wolley’s papers, there were only a few memoranda or suggestions bearing upon the subject. Newton acknowledged that Wolley first acquainted him with his idea in a letter dated June 7, 1856, in which he stated that, at the meeting of Scandinavian naturalists, he was going to “give some account of the British birds, of which so little is known on the Continent beyond the bare list.” He wished to begin “by naming the birds which are commonest in England and most characteristic of our bird-fauna”. Above all, he wanted “to be able to represent by numbers the relative abundance of each species: throughout Great Britain put a Sparrow at 1,000,000, and an Osprey at 1, what will be the intermediate figures?” Newton was unable to supply such data, and consequently Wolley’s design was abandoned. However, they discussed the matter many times and some ideas were proposed. Wolley, who had studied medicine, ran a detailed account of his commercial affairs as a dealer in birds’eggs. Moreover, as a member of two ancient, landed gentry families – the Wolleys, owners of Allen Hill, and the Hurts, owners of Alderswasley and WirksworthFootnote ⁶³ – he was aware of the methods for registering (dead) animals for the administration of game on their lands. Newton alluded to the “Game Books,” which were commonly kept on different manors and proposed that the inspection of these registers would furnish a sufficiently accurate return to assess in what districts it would be possible or impossible to increase the stock.Footnote ⁶⁴

By the 1860s, there were two kinds of game books: printed and manuscript. The printed game books were published each year and consisted of printed tables, which enabled “the Lover of Field Sports to keep a Register of Sporting Occurrences from year to year, in the manner of a Journal.” They were similar in plan to the older practice of the nobility and gentry for keeping at the residences, in manuscript books, a record of the shooting done in their estates. These books, in folio or quarto size and bound in vellum, showed the amount of game killed each day. The manufacturing of printed “game-books” for personal use started around 1815. They were printed in a small volume adapted to the pocket and game bag of the sportsman, or in a book, large enough to contain the “sporting annals” of many years. Normally they were offered and designed by stationers, account book manufacturers, and printers. The printed game books were also marketed as serviceable to the game keeper, in keeping their account of the disposal of game through presents, and distribution amongst acquaintances, the local brewer, butcher, baker, and sundry farmers. In the columns was entered the produce of each day’s sport, every head of game that was bagged, under their several names and descriptions. Some were printed on both sides with headings “Game killed, from/ Gentlemen’s names/Total brought forward&” printed horizontally; and with columns printed horizontally for the animal kind. Other printed game books had vertical columns that included “where killed,” “when,” “killing and missing shots,” “observations,” blank columns, the number of head killed, guns, and shots, total each day and week. Horizontal ruling divided each page into six days of the week and a total.Footnote ⁶⁵ Invented and standardized in the nineteenth century, the game books recorded the number of heads hunted in connection with the place and meteorological conditions. The records were restricted to the lands where the hunter had hunting privileges or the gamekeeper had to keep control of the disposal of game. In this way, private property and bookkeeping organized the data available for the study of animals. Some game books also presented a summary of all the game shots annually over a period of several decades. The game books showed that some birds were uncommon or difficult to hunt and, since the columns were organized by animal (bird) kind, they also indicated population changes of the recorded species. Even if the game was artificially stocked, the game books – as protocols used in the private administration of the gentry’s landscapes and hunting rights – provided a format to record the occurrence of animals and to do so over time. Wolley’s own account of the great auk on Iceland had followed this way of organizing information in terms of pieces hunted and memories of the fowlers. Several accounts on the extinction of the great auk referred to the fowling regulations established by the governors of Iceland or the colonial administrations in Canada. Depending on the scarcity or abundance of birds, the governors gave or declined permission to fowl. Thus, restocking and the regulation of hunting privileges were practices linked to the administration of supplies and hunting.

The brothers Alfred and Edward Newton, however, incorporated the organization of data of the game books – including meteorological observations – into their wider observations of birds at their family estate of Elveden, Suffolk, adding new categories such as movements, nesting, and their general habits. A. Newton insisted on two things: on the experimental character of his first trials, and on the difficulties of recording a continuous series of observations and the keeping of a regular journal for any length of time.Footnote ⁶⁶ With the system they adopted, they recorded the bird species, their sex, singing, abundance, pairing, arrival date, and number in the flocks (Figure 8).

Figure 8

(Two to be combined) Newton’s records

(Alfred Newton’s Archive, University of Cambridge Library

Recording life processes in real time confronted naturalists with the question of how to record so many different dimensions at once and how to organize these observational data in devices that could be “read” at a simple sight; some authors have called this problem “the compression of knowledge.”Footnote ⁶⁷ As a former hunter of birds, Alfred Newton was well acquainted with the devices for recording game and transferred them to his zoological observations of wild species. Thus, by the proposed method, the results of a day’s observation could be recorded in less than five minutes. The register of birds, which ran from 1849 to 1859, produced a book about eleven inches in height by about fourteen in breadth.Footnote ⁶⁸ The counting of the game not only appeared in the registers but also in the method by which this was done. To count the observed individuals, observers transferred a common gun wadding from one coat pocket to another, and “as shooting-coats have usually many pockets, the process was not difficult; and it was even possible to appropriate a pocket to each of the three species we wished to number.”Footnote ⁶⁹ Thus, game apparel, game books, and game equipment, as well as the patience of hunters, were used to account for wild animals, shaping the methods and means to organize data in a relatively small book.

Having examined how kitchen‑middens and related deposits provided crucial evidence for debates about extinction within historical time, this section turns to the global itineraries through which those methods, materials, and expectations traveled. From the Danish shores to North America, the North Pacific, and the Indian Ocean, the same practices of excavation, comparison, and collecting were redeployed across imperial and commercial networks, generating new forms of scarcity and new ambitions for museums and naturalists. As establishing extinction was recognized as an evidentiary problem I now turn to the practical response: bone‑seeking, excavation, and the mobilization of administrative and colonial networks to recover “new fossils” for museums. The global pursuit of these new fossils – and the networks that sustained it – is the subject of Section 4.

The Power of Images

The relationship between knowledge and image is part of the so-called visual turn, developing a great deal of literature after Bruno Latour’s concept of “immutable mobiles.” Latour, referring to the phenomenon of the transfer of knowledge through objects and paper (maps, drawings, or descriptions), invited us to reflect on the link between the spaces of the field and the cabinet, notes, measurements, sketches, and the characteristic and determining elements of Western science. According to him, modern science was established thanks to this capacity to produce – and accumulate – a standardized, communicable, and mobile knowledge.

This subsection will refer to the sketches of the sea cow sent by Steller to Russia. Even when the real animals were lost, they survived thanks to this capacity to reproduce them as “immutable mobiles.”Footnote ⁷⁶ These drawings, an economical and manageable way of transporting information, made it possible to tackle the zoological enterprise in another fashion.

Leonhard Stejneger (1851–1943), the Norwegian-born American zoologist in charge of the Smithsonian Institution’s collections for many years, wrote an extensive biography of Steller, devoting an appendix to “The pictures of the sea cow” that serves as the basis for this section. Stejneger reconstructed the intricate itinerary of the only pictures made of the “living animal,” tracing its copies and its life independent of the author. Apparently, the Bering expedition, for lack of space, left without a draughtsman. Steller was not a draughtsman either, and to make up for this deficiency, the skills of pilot Friedrich Plenisner were called upon to draw sea charts and, at Steller’s request, plants and animals. Plenisner drew two plates of the sea cow, images which Steller, together with the manuscript and other plates depicting a sea lion and pinnipeds, sent to St Petersburg on the annual Kamchatka supply ship in 1743. By 1746, the year of Steller’s death, the Academy retained the manuscript but the plates had been lost, which is why they do not appear in the 1751 publication. However, because other copies had been made, Plenisner’s drawings survived both their loss and the extinction of the animal.

Stejneger was able to discern that all the images of the sea cow that had been made from the second half of the eighteenth century onward came from two series of images, both made by Plenisner to serve different documents recording the expeditions to the farthest regions of the Russian Empire. The story Stejneger tells could well be that of any other series of images made: safeguard copies sent privately to colleagues and kept in personal collections, copies of copies that were reproduced beyond the author’s will or knowledge. Contrary to the idea that manuscripts – be they textual or images – are immobilized in archives, the images of Steller’s sea cow had a life of their own, were used in their “manuscript” state, were copied and moved in circuits that were uncommon but by no means nonexistent. The images, through the most diverse agents, managed to reach publication, in this case the work of the German naturalist Peter Pallas, who accessed Plenisner’s sketch and published it, posthumously and without major modifications, in his work on the fauna of Russia.

When Brandt published the first comprehensive description of the northern sea cow, he enclosed an “Icon idealis” (ideal image) of the animal and several plates showing the composition of the skull. The sketch of the outer form merely recomposed the sketch published in Pallas’ work, that is, the Plenisner/Steller drawing. Brandt’s plates were sent to the Russian–American colonies together with instructions for collecting the bones, so as to visually instruct the residents who were unable to read Latin, the language of Brandt’s work. However, the new “natives” were not only used to the bones but also to the depictions of the extinct animal, which started being seen as if they had been acquainted with it. (Figure 11) Despite this account:

Figure 11

Brandt’s Icon Idealis of Rhytina stelleri (public domain)

Nevertheless, on the basis of these testimonies, it was proposed to date the extinction of the species to 1854. Stejneger, however, pointed out that the natives, although not familiar with Steller’s text, had been instructed to collect bones for the St. Petersburg’s museum, receiving the images through Brandt’s work: “Those natives,” said Stejneger, “were far from being unaware of the oldest descriptions of the animal; a copy of the plates of the first part of Brandt’s work, Symbolae Sirenologica, was sent to the island very soon after it was published.”Footnote ⁷⁸ The observation of reality was sifted by the history of the circulation of images. These “natives” had learned to see living “sea cows,” or at least to ascribe to some animals of dubious shape the forms they had learned through the instructions of the Tsar’s museum. Stejneger, in order to unravel the “realities” created by the images mixed up in real memory, did an almost psychoanalytical work, interviewing the “natives,” confronting words with the images and the texts of the books, and discovering more than one error. In fact, books, texts, and images using Pallas’ and Brandt’s sea cow plates and photos of some skeletons already mounted in museums, went back and forth to Bering. The experience of the natives also included them, to the extent they were unable to distinguish what they had seen on paper from what appeared in the sea or had heard as stories told by the campfire.

In 1882 Stejneger had been sent to the Commander Islands under the auspices of the US Signal Service to establish observation stations. While there he studied the islands’ natural history, the fur seals, and made specimen collections, including the skeleton of a sea cow. As the problem of the fur seals and commercial sealing became an international economic and political concern, Stejneger’s studies of the seals and the sealing conditions became more involved. In 1895 he was sent to the North Pacific as an attaché of the US Fish Commission. The next year, President Cleveland appointed Stejneger to the International Fur-Seal Commission and he spent most of this time on the Pribilof and Commander Islands, as well as some time in Hakodate, Japan. He returned to the Bering Sea again in 1897 to continue his studies and investigations. His last trip to the Commander Islands was in 1922 as a representative of the US Department of Commerce. During Stejneger’s stay of about two years (1882–1883) on Bering Island, he succeeded in finding a number of skulls, ribs, vertebrae, and other bones. And, most importantly, he recorded the testimonies of his contemporaries in a wonderful interview that is revealing as to how natural history has always been a human science.Footnote ⁷⁹ One complete skeleton was found buried in the sand, but the bones were too far decayed to permit handling. From the various individual bones that were found, a fairly good skeleton was “made up” for the National Museum in Washington. This, together with the two skeletons in St. Petersburg and Helsingfors, and the two ribs in London, constituted the total amount of material pertaining to the sea cow found in the museums of the world.

This section has reviewed how the extraction and circulation of scarce remains were facilitated by networks of expertise and empire. The remains, however, did not only find a new home in museum collections. The following section follows some of those objects into the auction room, where scarcity was prized and priced accordingly, and they acquired a parallel life as luxury commodities.