The so-called diet and patch models were foundational for evolutionary studies of foraging ecology. The diet model analyzed an organism’s choice between at least two different food or prey items. The patch model asked how much time a forager should invest in exploiting a resource patch that offered diminishing returns before moving on to search for another patch. Concerning sources of inspiration, however, some pioneers of optimal foraging theory explicitly denied an input from economics, others explicitly credited marginalist economics, while (...) the majority remained tacit on this issue. An inquiry into the historical sources reveal an interesting patter. As far as diet is concerned, at least two models existed. One was the, probably homegrown, so-called contingency model of diet choice, the other was imported from marginal utility theory of economics. The import could not be operationalized in a way allowing predictions about feeding behavior. It was, therefore, used as a framework for interpreting data on food choices and gaining insights about the relative value of food items. The contingency model was hypothetico-deductive instead. It allowed predictions of behavior that could be tested. It was, therefore, the more popular among the peers and the pioneering publications have become citation classics, whereas the publications importing marginal utility theory int foraging ecology are largely forgotten. The situation is different for the patch model. At face value, only one homegrown model exists that was discovered independently at least twice. A closer look reveals obscure links suggesting knowledge dispersal from marginal productivity theory of economics towards optimal foraging or mating theory, respectively, along routes of private and unpublished communications. These communications were between alma mater and student in the case of Eric Charnov and between father and son, in the case of Geoffrey Parker. This study suggests two concusions: 1. models that are operationalized in a way allowing hypothetico-deductive research are more popular among researchers than others requiring a more inductive an interpretive approach. 2. The sciences are interlinked far more closely and deeply than mere citations would suggest. Sometimes a knowledge transfer that is crucial for a discovery can occur as an informal and private communication. (shrink)