INTRODUCTION
In the 1970s, data manipulation in fisheries science relied on paper and pencil methods aided by programmable calculators and, in some sophisticated laboratories, on huge computer systems with punch cards (see Munro, this volume). Thus, assembling large amounts of data was limited by the availability of paper copies of peer-reviewed publications, the “reprints” of lore, and grey literature. This was the environment in which Daniel Pauly found himself, struggling with how he could test his hypothesis on the relationship between gill size and the growth of fishes (Pauly, 2010; Bakun, this volume; Cheung, this volume). Testing such a hypothesis needed a large amount of empirical data, which might be available in principle, but if so, not at one's fingertips.
Inspired by Walter Fischer's work on the Food and Agriculture Organization of the United Nations (FAO) species identification sheets in the mid-1970s, Daniel believed that assembling data from already published literature was essential for a timely response to the needs of fisheries management, which, at the time, used analytical models requiring growth and mortality estimates (Munro, this volume). And index cards, he found, were ideal for recording the specific data required for assessing size at age, maximum sizes and ages, and growth and natural mortality parameter estimates, as well as temperature and other environmental variables and their sources.
The index card collection provided data for his widely used compilation of length–growth parameters (Pauly, 1978), which served as a basis for investigating the role of gills in fish growth in his doctoral thesis (Pauly, 1979) and subsequent papers, and for his now classic paper on natural mortality (Pauly, 1980).