The focus of this paper is the emergence of the research laboratory as an organizational entity within the company structure of industrial firms. The thesis defended is that, after some groundwork by British and French firms, the managements of several of the larger German dye companies set up their own research organizations between the years 1877 and 1883. The analysis of the emergence of the industrial research laboratory in the dyestuffs industry presented here makes clear that both the older study on the subject by John J. Beer and a later paper by Georg Mseyer-Thurow contain some serious defects. Beer, like so many other authors of the 1950s who studied the ‘marriage’ between science and industry during the ‘Second Industrial Revolution’, incorrectly correlates the engagement of university-educated chemists with the rise of industrial research. The appointment of academic chemists by BASF and Hoechst at the end of the 1860s, for instance, was described as ‘the…acquisition of a research staff’. This reveals a misunderstanding of the roles of chemists within the nineteenth-century chemical industry. University-trained personnel were, in fact, working in industry as early as the start of the nineteenth century. However, they were employed as managers, works chemists and analysts, and only exceptionally in research.

Even before the success of William Perkin's mauve at the end of the 1850s, there were attempts to synthesize artificial dyes that were identical with those found in nature. Alizarin, the dye derived from the madder root, was the first to be investigated, and it was Perkin who was to file for a patent in June 1869 just one day before the German chemists Heinrich Caro, Carl Graebe and Carl Liebermann. Rivalry between the parties soon turned to negotiations and collaboration. Perkin's company retained the British trade, while the Germans, in the form of the Badische Anilin- und Soda-Fabrik (BASF) controlled the continental European and United States markets. This and similar agreements extinguished the madder trade, and subsequently artificial alizarin passed almost completely to the Germans. They achieved a monopoly by dictating the level and prices of supplies, which did much to diminish the strength of the dye-making industry in Britain. The formation in 1882–83 of the British Alizarine Company did little to redress the overall balance. This taught British dye firms a tough lesson. The same, they hoped, would not be allowed to happen again, even when the attention of the German research chemists turned to indigo.

Louis Pasteur plays a role in the creation myth of biotechnology which resembles the heroic position of his great antagonist Liebig in the story of agricultural chemistry. His intellectual development, expressed in a great book, supposedly underlay a revolution in practice. Similarly, biotechnology is conventionally traced back to Pasteur, through whose influence, it has been assumed, ancient crafts were transformed into an applicable science of microbiology. The emphasis on Pasteur's work in the history of biotechnology has served to bolster the image of progress in the technology following from periodic scientific breakthroughs. Elsewhere I have argued that biotechnology can be better seen as a boundary object, to use Star and Griesemer's terminology, between biology and engineering. As such it has been significant throughout this century, and the word has been used since 1917.

This is another paper about science and her powerful companion (technology), to use A. W. Hofmann's colourful phrase. Whereas most papers on the interaction of science and technology deal with the transfer of knowledge from academic science to industrial technology, this paper is about the contribution of an industrial researcher to academic chemistry. The boost Reppe's research gave to the study of aromaticity parallels the impact of the early synthetic dye chemistry on structural organic chemistry. This case study suggests that we cannot draw a clear distinction between ‘pure’ and ‘applied’ chemistry, in the laboratory at least.