ABSTRACT
This third method of simulating plant morphogenesis by computer (after the geometric method and the logical and algorithmic method) was first conceived and developed in the early 1970s by a French agricultural engineer, Philippe de Reffye. Unlike the earlier methods, it evolved from a purely pragmatic need. There were a number of more or less fortuitous, yet linked, causes behind its emergence. Contrary to the cases of the earlier authors, these reasons were no longer solely personal, speculative, rhetorical or aesthetic; instead, they were based on a combination of technical, institutional and political motivations. Since this work was no longer merely speculative, it involved the efforts of a number of researchers, although de Reffye was undeniably the main instigator for much of it. After I have reviewed de Reffye’s principal pioneering suggestions, therefore, I will assess the work of his laboratory’s followers. We will then be able to see how the fields of agronomy and forestry felt the need to go beyond the mathematical model stage and enter the stage of computer simulation. We will see, above all, that the strength of simulation was based on three things: 1) the decision to proceed by what I have called fragmented modelling; 2) the possibility of creating an integrative computer simulation – based on these fragments once they have been reconstituted by computer – that can be calibrated in the field; and 3) the possibility of taking the spatial and temporal heterogeneity of plant-part growth rules into consideration and thus bypassing the limitations of traditional biometric models.
