Holling's (1959) disc equation is the most commonly used model of a type-2 functional response. The constant-parameter version of this model makes 3 assumptions that are seldom likely to be satisfied: 1) attack rate per unit search time per unit prey density is independent of prey density; 2) handling time is independent of prey density; and 3) the relative rates of successful and unsuccessful attack are independent of prey density. Optimal foraging theory is used to suggest how each of these components of the predation process might depend on prey density. Explorations of simple models suggest that these types of adaptive variation are often consistent with type-2 responses, but generally result in shapes that differ significantly from the disc equation. The consequence of using a constant-parameter disc equation model when it is inappropriate is limited ability to predict functional responses in altered circumstances, eg in the presence of a 2nd prey. The potential ability to understand multispecies systems from their subcomponents cannot be assessed unless appropriate models of the functional response are used to describe the component systems; these models are likely to include adaptive behavior. -Author