From 2000 to 2003 a range of Farm Scale Evaluation (FSE) trials were established in the UK to assess the effect of the release and management of herbicide tolerant (HT) crops on arable weeds and invertebrates. The FSE trials for maize were also used to investigate crop-to-crop gene flow and to develop a statistical model for the prediction of gene flow frequency that can be used to evaluate current separation distance guidelines for GM crops. Seed samples were collected from the non-GM half of 55 trial sites and 1,055 were tested for evidence of gene flow from the GM HT halves using a quantitative PCR assay specific to the HT (pat) gene. Rates of gene flow were found to decrease rapidly with increasing distance from the GM source. Gene flow was detected in 30% of the samples (40 out of 135) at 150 m from the GM source and events of GM to non-GM gene flow were detected at distances up to and including 200 m from the GM source. The quantitative data were subjected to statistical analysis and a two-step model was found to provide the best fit for the data. A dynamic whole field model predicted that a square field (150 m x 150 m in size) of grain maize would require a separation distance of 3 m for the adjacent crop to be below a 0.9% threshold (with <2% probability of exceeding the threshold). The data and models presented here are discussed in the context of necessary separation distances to achieve various possible thresholds for adventitious presence of GM in maize.