INTRODUCTION

Among the many different meanings of the category of a model currently applied by historians and philosophers of science, one refers to the fact that scientists often have to invest a considerable amount of work in order to match an accepted fundamental theory, abstract theoretical principles, a general conceptual scheme etc., with new empirical objects. If a smooth assimilation of the particular and concrete object to the general and abstract intellectual framework is impossible scientists build ‘models’ linking the particular to the general.

In my paper I study a paradigmatic achievement of model building in nineteenth-century chemistry when chemists attempted to extend an accepted conceptual framework to a new experimental domain. In the late seventeenth century and the early eighteenth, chemists created a conceptual framework that encompassed concepts like chemical compound, constitution, affinity, chemical decomposition and recomposition, chemical reaction etc. This scheme shaped the identification and classification of substances as well as experiments investigating their transformations. In the eighteenth century, chemists applied this network of concepts nearly exclusively to inorganic substances. Ordering a particular inorganic substance such as an acid, an alkali or a salt into this network required a series of experiments but was normally seen as unproblematic. However, chemists faced many problems when they began to apply this conceptual network to experiments performed with organic substances.