Extent of cross-fertilization in maize by pollen from neighboring transgenic hybrids

There is an increasing concern about the preservation of genetic identity of conventional maize (Zea mays L.) and of distance required to segregate non-genetically modified (non-GM) from GM grain production since the introduction of Bacillus thuringiensis (Bt) and other transgenic events into commercial hybrids. Field experiments were conducted at three sites in Ottawa, Canada, for 3 yr to determine (i) the extent of cross-fertilization of a maize genotype by foreign pollen of neighboring hybrids and (ii) the practical distance required to isolate conventional maize hybrids from neighboring GM maize fields. At each site, yellow-kernel Bt maize was planted in the center (27 by 27 m) of a field surrounded in all directions by the distance equivalent to 24 or 48 rows (37 m) of white-kernel maize, and a 200-m non-maize crop was maintained in all directions. Phenology and weather conditions were closely monitored during the tasseling and silking period. At maturity, a thorough examination on the cross-fertilization was conducted in the white maize population. Our results showed that the rate of cross-fertilization in maize was dependent upon the distance from the pollen source, wind direction and synchronization of silking and pollen shedding of the two genotypes involved. Up to 82% out-cross was measured in the first row adjacent to the Bt maize. The level of out-cross was <1% beyond the 37th border row (28 m) downwind and the 13th row (10 m) upwind in all site-years. An exponential decline model was fitted well (P < 0.01) to the cross-fertilization data as a function of distance from the yellow maize pollen source with R-2 up to 0.64. Our data suggested that it is possible to produce non-GM maize grains by removing the outside rows of non-GM maize plants (about 30 m) neighboring the GM maize field in concern if the acceptance level is set at less than or equal to1% out-cross. The generally recommended 200-m distance between two genotypes (inbreds, populations, hybrids, and wild relatives) appears to be appropriate for Bt or other GM maize, as well.