Transgenic Brassica napus can be easily crossed with wild Brassica rapa. The spread of the transgene to wild species has aroused the general concern about its effect on ecological and agricultural systems. This paper was designated, by means of population genetics, to study the fate of a transgene escape from B. napus to B. rapa. Three models were proposed to survey the change in gene frequency during successive backcross processes by considering selection pressures against aneuploids, against herbicide-susceptible individuals, and by considering A-C intergenomic recombination and the effect of genetic drift. The transmission rate of an A-chromosome gene through an individual to the next generation was 50%, irrespective of the chromosome number; while that of a C-chromosome transgene varied from 8.7% to 39.9%, depending on the chromosome number of the individual used in the backcross. Without spraying herbicide, the frequency of an A-chromosome gene was 50% in the BC(1) generation, and decreased by 50% with the advance of each backcross generation; that of a C-chromosome gene was around 39.9% in BC(1), 7.7% in BC(2), 1.2% in BC(3) and 0.1% in the BC(4) generation. Under the selection pressure against herbicide-susceptible individuals, the frequency of a transgene reached a stable value of about 5.5% within six generations of successive backcrossings. The effect of genetic drift and intergenomic exchange on gene transmission rate was discussed. It is suggested that the transgene integrated on a C-chromosome (or better on a cytoplasm genome) is safer than that on an A-chromosome. The transgenic cultivars should be cultivated rotationally by year(s) with other non-transgenic varieties in order to reduce the transfer of the transgene to wild B. rapa species.