Comments on the paper "A statistical assessment of differences and equivalences between genetically modified and reference plant varieties" by van der Voet et al. 2011.

van der Voet et al. (2011) describe statistical methodology that the European Food Safety Authority expects an applicant to adopt when making a GM crop regulatory submission. Key to their proposed methodology is the inclusion of reference varieties in the experimental design to provide a measure of natural variation amongst commercially grown crops. While taking proper account of natural variation amongst commercial varieties in the safety assessment of GM plants makes good sense, the methodology described by the authors is shown here to be fundamentally flawed and consequently cannot be considered fit for purpose in its current form.