In trees, floral induction (FI) occurs in some shoot apical meristems (SAM) only. It is assumed to be triggered by endogenous signais, either activating or inhibiting, that could originate from leaves and fruit, respectively. To test this assumption, we developed a model named SigFlow, to simulate signal transport in 3D structures and quantify the combined impact of antagonistic signais and the distances at which they act on SAM FI. Signal transport was considered an 'attenuation' parameter, whereas SAM Tate was determined by probabifity functions depending on signal amounts. Mode' behavior was assessed on simple structures before being calibrated and validated on a unique experimental data set of 3D digitised apple trees ('Golden Delicious') manipulated for leaf and fruit removal at different scales of tree organization. The estimation of the model parameters and the comparison of two signal combination fonctions led us to formulate new assumptions on the mechanisms involved: i) the activating signal from leaves could be transported at shorter distances than the inhibiting one from fruit (roughly 50 cm vs. 1 m) and ii) SAM are more sensitive to inhibiting signal than activating one. Moreover, SigFlow showed its ability, by modifying either model parameter values or input tree architectures, to simulate contrasting FI patterns consistent with those observed in apple cultivars.