Collecting Experiments

Making Big Data Biology

Bruno J. Strasser

• Publisher: University of Chicago Press
• Year: 2019
• ISBN:
  9780226635187
• Topic: Sciences biologiques
• Subtopic: Science générale

Databases have revolutionized nearly every aspect of our lives. Information of all sorts is being collected on a massive scale, from Google to Facebook and well beyond. But as the amount of information in databases explodes, we are forced to reassess our ideas about what knowledge is, how it is produced, to whom it belongs, and who can be credited for producing it.Every scientist working today draws on databases to produce scientific knowledge. Databases have become more common than microscopes, voltmeters, and test tubes, and the increasing amount of data has led to major changes in research practices and profound reflections on the proper professional roles of data producers, collectors, curators, and analysts. Collecting Experiments traces the development and use of data collections, especially in the experimental life sciences, from the early twentieth century to the present. It shows that the current revolution is best understood as the coming together of two older ways of knowing—collecting and experimenting, the museum and the laboratory. Ultimately, Bruno J. Strasser argues that by serving as knowledge repositories, as well as indispensable tools for producing new knowledge, these databases function as digital museums for the twenty-first century.

1

Live Museums

Life science laboratories are usually home to just a few animal or plant species. Homo sapiens can be found there during the day (and sometimes at night), but the permanent residents are those species that have gained the enviable status (at least from the scientists’ perspective) of “model organisms.”1 The story of the experimental life sciences from the late nineteenth century to the present can be told from the vantage point of these select species that have served as scientists’ so-called “guinea pigs.”2 Guinea pigs proper (Cavia porcellus) are among them, and so are microbes (Escherichia coli), mice (Mus musculus), flies (Drosophila melanogaster), corn (Zea mays), and more recently worms (Caenorhabditis elegans), fish (Danio rerio), weeds (Arabidopsis thaliana), and many others.

For most of the twentieth century, experimental scientists’ narrow focus on selected species stood in sharp contrast with the broad diversity of organisms studied by naturalists. In the first half of the twentieth century, some embryologists still worked experimentally on diverse species, but overall the range of organisms found in laboratories was narrowing. By contrast, natural history museums often housed tens or hundreds of thousands of species—up to a million in the British Museum (Natural History) in London and almost as many at the American Museum of Natural History in New York. For historians and scientists, research in collections such as those housed in museums came to be viewed as distinctive to the naturalist enterprise, in contrast to the experimentalist’s focus on laboratories and model (or “standard”) organisms. Stating the opposition this way is too simplistic but is a starting place on which to build. The aim of this chapter is to capture the important role that collections of organisms have played in the development of the experimental life sciences, thus making this opposition more nuanced. This is especially true of genetics, whose practitioners were ironically keen on drawing a contrast between their experimental approach and the collection-based approach of naturalists. When we look at the histories of the two disciplines side by side, we will see that collections of living organisms, so-called “stock collections,” have been indispensable in the rise of experimental genetics. These findings help to refine our understanding of research practices and to explain the historical basis of the current use of collections in experimental research.

How should we conceptualize stock collections to understand their role in the development of scientific knowledge and of scientific communities? In many ways, they have been to the experimental life sciences what museum collections were to natural history: repositories of organisms (preserved for a possible future use), centers of standardization (defining nomenclatures), centers of distribution (providing remote researchers with specific organisms), tools for research (allowing comparative studies), centers of coordination (for complex networks of researchers), and institutions defining the norms of practices (social and epistemic). Like museums, stock collections served these many roles at the same time, co-producing communities, practices, and knowledge. As with museum collections, the single most important issue for creators of stock collections was how to develop a moral economy that would support the wide participation of researchers, their contribution of organisms to the stock collection, the sharing of information, and an obedience to community norms. The solution to this problem hinged on the subtle definition and enforcement of a boundary between private and public objects and ideas, between members of the community and outsiders, between intellectual contributions that deserved individual credit and those that should remain communal.

Creators of stock collections from the early twentieth century to the present have relied on an unusual epistolary technology to achieve these aims: the newsletter.3 Neither a private letter nor a public journal, the newsletter was a way to address a select community of individual researchers, providing information about the content of the stock collection (a catalogue), its uses in research (research results), and the members of the community (a directory). Such newsletters were invented in the late nineteenth century for internal communication within growing corporations4 and became part of the social bond among communities of researchers working on the same organism. They helped compensate for a loss of close interpersonal relationships as research communities grew larger and helped propagate and enforce community norms and ideals. A close examination of several newsletters reveals their crucial role in the development of experimental research centered on stock collections.

The collections examined in this chapter differ from those of museums in two crucial ways. First, rather than embracing a broad range of species, they contained many variants, often mutants, of single species, which were almost always model organisms. These organisms were selected mainly for practical reasons: their anatomy, physiology, and behavior made them particularly amenable to experimental studies of some specific aspect of their biology. Thus different fields in the life sciences—embryology, physiology, or genetics—have adopted different model organisms. In the nineteenth century, the sea urchin became a favorite organism to study embryonic development because its eggs were transparent and every step following fertilization could be easily observed under a microscope. Similarly, in the mid-twentieth century, neurophysiologists focused on the Atlantic squid because electrodes could easily be inserted in its exceptionally large nerve axon, permitting experimental measurements. The qualities that made organisms well suited to genetic studies were a combination of small size, short generation time, and distinct and easily observable characteristics.

Second, unlike museum collections, those examined in this chapter are collections of live organisms. Curators of natural history museums were concerned about the preservation of their specimens. Curators of stock collections were concerned about keeping them alive and constant against natural mutations producing variation at each generation. The practices of maintaining collections of live specimens had been developed in zoos and botanical gardens, but also in marine stations, such as the Stazione Zoologica in Naples, which became a “clearing house for model organisms” and supported experimental, and often comparative, research, especially in embryology and physiology.5

Unlike embryologists and physiologists, geneticists worked with huge numbers of individuals over many generations. Typically, they performed thousands of crosses between various organisms in order to study the distribution of traits among their offspring. The probability of discovering a new trait, initially a matter of chance, could be raised by increasing the number of individual observations. Thus only small and fast-reproducing organisms could accommodate the limited space of a laboratory and the limited time span of a human researcher’s career. These important constraints mean that very few organisms have actually been studied genetically, and even fewer organisms have been used in more than one field of research. Yet those species that have had the most enduring place in the history of biology are precisely those that were suited to more than one line of investigation. The famous fruit fly began its scientific career as a genetic model and only later became a model for embryonic development. The choice of model organisms in other fields has been determined by other considerations, such as their economic importance (corn) or their evolutionary proximity to humans (mice).

The science of genetics emerged after the rediscovery of Mendel’s laws in 1900 and relied heavily on model organisms.6 To obtain material for their experiments, geneticists produced and maintained huge numbers of individuals in stock collections. Thomas Hunt Morgan’s “fly room” at Columbia University typically contained tens of thousands of individual Drosophila.7 The small size of this organism made it possible to store them all in just a few hundred milk bottles. But more important than the number of individuals, which could be increased at will thanks to the extraordinary fertility and short generation time of the fly, was the number of distinct mutants that could be produced and collected. In 1909, Morgan found his first mutant, a fly that had white eyes in place of the usual bright red ones.8 Within two decades, Morgan’s group was caring for more than six hundred different mutant strains. This diversity was particularly important for the mapping of genes along the chromosomes, the central intellectual agenda of the Morgan school of genetics, since each strain constituted a reference point for the position of a gene.9

In the 1910s and 1920s, mice, corn, and flies became widely used in genetic research. This was possible because organisms and increasingly large numbers of mutants were kept alive in stock collections and made available to individual researchers. Like museum collections, these started out as private collections intended for the exclusive use of a local group of researchers and their privileged correspondents. They functioned under rules of civility established between researchers who knew each other personally and whose community was headed by respected leaders in the field. By the 1930s, the needs of the broader community prompted several of these private collections to retrace the trajectory of museums a century earlier.10 They were converted to public “stock centers,” often funded by philanthropic institutions such as the Rockefeller Foundation or the Carnegie Institution of Washington. They were thus able to foster the sharing of organisms, research data, and social norms beyond a small initial group of researchers.

Historians have sometimes considered model organisms to be akin to the physical instruments such as microscopes and spectrometers used daily by experimentalists.11 This perspective is useful because, like instruments, model organisms are made by researchers to serve special research needs. Model organisms were produced through techniques of inbreeding, i.e., the crossing of siblings. After many generations, inbred lines consisted of highly similar individuals. These standardized organisms made experimental results more reproducible, like standard scientific instruments.

Yet at least as important as the crafting of these individual model organisms was the role of organism collections in the production of knowledge. As Robert Kohler’s exemplary study of Thomas H. Morgan’s fly group has shown, the members of this group shared, with some notable exceptions, a particular communal culture of intellectual and material exchange and credit attribution.12 These norms can be linked to the very nature of Drosophila, specifically the extraordinary number of mutations it experienced, and to the particular research agenda of Morgan’s group, mapping genes, providing more problems to solve than any individual could tackle in a lifetime. When we look systematically at stock collections of microbes, corn, mice, and flies, it becomes apparent that similar systems of norms arose in various model organism communities, which raises questions about the nature of the relationship—genealogical, functional, or some other type—between the norms (or “moral economy”) in each system.13 By focusing on stock collections and the moral economies that they sustained, one based on freedom of charge and reciprocity, and on the co-production of collections and communities, this chapter provides a revised picture of the early rise of experimental life sciences, one in which knowledge production is more comparative than analytic, where contributions are more collective than individual, and where moral economies are more collaborative than individualist.

Microbes at the American Museum of Natural History

The public stock centers for mice, corn, and flies are perhaps those whose history is best known, but they were not the first. The American Type Culture Collection (ATCC), a collection of microbes founded in 1911 in New York as the Bacteriological Museum, forms an important precedent. In the 1930s, the ATCC and its European counterparts represented the largest collections of organisms in the world. Although it was not yet used in genetic research, the ATCC served some of the same purposes as stock collections in genetics, namely, to provide standardized organisms to researchers, especially microbiologists working in academic or industrial contexts. The history of the ATCC illustrates particularly well the roots of stock collections in natural history museums and their similar trajectories. Indeed, the ATCC began at the foremost natural history institution in the United States: the American Museum of Natural History (AMNH).14

In 1911, the bacteriologist and public health expert Charles-Edward Amory Winslow (1877–1957) had just accepted the position of curator of the Department of Public Health at the AMNH in New York City.15 He had obtained his MS degree from MIT in 1899 under the direction of the bacteriologist William T. Sedgwick and taught there for over a decade.16 He was already a respected researcher and public health figure, having published several studies on water supply contamination and microbiological issues in sewage treatment, collaborating with his former teacher Sedgwick, a leader in American bacteriology and founder of the first school of public health in the country at MIT.17 Winslow had also made several contributions to the biochemistry and classification of bacteria of sanitary importance, especially the widely distributed Staphylococcus, a frequent cause of human infections.18 The latter project was accomplished with collaborator Anne F. Rogers, soon to become Winslow’s wife. Their work culminated in a 1905 book in which more than five hundred strains were described and classified. Now at the AMNH, Winslow established the first public “museum of living bacteria.”19 Within a year, this collection contained 578 strains representing 374 distinct types.20 It was assembled thanks to forty-five laboratories in the United States and Canada, which had “contributed freely” from their own collections of bacteria after Winslow sent out a call to laboratories in both countries.21

The material used in this classificatory work became the basis of a collection of organisms similar to earlier natural history collections. Yet unlike the zoologist and the botanist, the bacteriologist could not examin...