We investigated the GM maize grain content of non-identity preserved (non-IP) maize samples produced in 2009 in the USA using our individual kernel detection system, involving two multiplex qualitative PCR methods coupled to microchip electrophoresis and partially real-time PCR array analysis, to clarify how many GM event maize grains were present in the samples and which GM events frequently appeared in 2009. The average percentage and standard deviation of GM maize grains on a kernel basis in five non-IP sample lots were 81.9%±2.8%, the average percentage of single GM event grains was 46.9%, and the average percentage of stacked GM event grains was 35.0%. MON88017 grains and NK603 grains were the most frequently observed as single GM event grains. The most frequent stacked GM event grains were MON88017×MON810 grains. This study shows that our method can provide information about GM maize events present in imported maize samples on a kernel basis.