[An error in the determination of the bi-ischial diameter by the Thoms grid (author's transl)]

After recalling the principles of homothetics which are the basis of radiological pelvimetry by the Thoms method, the authors carry out a study of the pelvimetry accessory presently the most distributed in France. It turns out that the grid introduces an error that, depending upon the conditions under which it is used, can reach levels of 4% underestimation. This error of 4% of something that itself is merely a calculation has little impact since it only affects 8% of the results.