Passing the El Greco test.

In recent years, the increasing use of modelling to capture biological complexity has revealed two complementary approaches. One is to build ever more comprehensive models in the hope that they will lead to useful predictions. The other approach is to get to the essence of the underlying processes by simplifying and clarifying complex problems through various levels of abstraction. Unfortunately, these two approaches are sometimes confounded, as is highlighted by a recent commentary by Alvarez-Buylla et al. [HFSP Journal, 1(2), 99-103 (2007)]. Their commentary raises several concerns about a paper we published on the development of inflorescences, the branching structures that bear flowers. Alvarez-Buylla et al. claim that the model we propose is too simplistic and propose an alternative "toy model," which they believe is more realistic and can account for the data just as effectively. However, they do not run simulations based on their model. We show here that, depending on how this toy model is interpreted, it either does not deliver what Alvarez-Buylla et al. claim, or is a special case of the model we proposed, or is so vague that its consequences are unclear. Our analysis demonstrates the importance of modelling in simplifying, clarifying and following through the consequences of particular hypotheses. Without this it is all too easy to construct elaborate models on shaky foundations.