Form follows function, but it does not follow from function. Form is not derivable from the latter. To realize a desired technical function, a form must first be found that is able to realize it at all. Secondly, the question arises as to whether an envisaged form realizes the function in an appropriate way. Functions are multiply realizable—various different forms can bear the very same function. One needs to find a form of a technical artifact that realizes an envisaged function sufficiently efficient, robust, or whatever criteria might be imposed. This paper scrutinizes biomimetics as one way to find a good solution to the realization problem. Drawing on an approach from the philosophy of simulations, it reconstructs the biomimetic relation as being mediated by a theoretical model. It is shown that the robustness of the functioning system is usually reached in different ways in biological and in technological systems, which explains differences in morphogenetic mechanisms or principles found in these fields. This reconstruction helps to understand problems with robustness in synthetic biology that occur when technical design principles are implemented in a biological system. The mimetic relation between the biological and the technical realm is found to be asymmetric.