A two-resource model of terminal investment.

The most widely known theoretical basis for the hypothesis of terminal investment is the classic model by George C. Williams (1966). Although this model predicts that reproductive effort (i.e. the proportion of available resources devoted to reproduction) increases with decreasing reproductive value, it implies that reproductive allocation in absolute terms should remain stable. This contrasts with the empirical evidence on terminal investment reported to date: the vast majority of positive case studies report an increase in some aspect of reproductive allocation in absolute terms. Also, a substantial number of studies have failed to record terminal investment, despite expectations. Here, I present a simple conceptual model which explains such results. I argue that to explain terminal investment, an organism's reproductive capacity must not be considered as a common pool of resources (often described by the term 'reproductive value'), but as a set of different resources which are not easily convertible to each other, and should be exhausted in balance. Thus, if one resource accidentally decreases, in response, the others must be expended at higher rate. To test this model, each reproductive allocation should be measured in a more specific currency (or currencies) than traditional 'reproductive effort'. The model is consistent with both the positive and the negative case reports on terminal investment.