The factorial structure of the French version of the twenty-item Toronto Alexithymia Scale (TAS-20) was studied in a group of 183 healthy subjects. The TAS-20 represents the latest revised version of the TAS. His psychometric characteristics are clearly superior than the others alexithymia scales. Our subjects filled out the TAS whose 14 items are common with the TAS-20. We conducted our statistical analyses on these 14 items. A principal components analysis was carried out and yielded a two-factor solution where the items rating the ability to describe feelings and the items rating the ability to identify feelings constitute one sole factor. The second factor corresponds to the externally oriented factor. The internal consistency of the scale was also measured by the Cronbach alpha coefficient and by the correlation between each item and the total score. The value of the Cronbach alpha was 0.79 and the correlation between each item and the total score ranges from 0.19 (p < 0.05) to 0.69 (p < 0.001) with a mean of 0.52. Our results confirm a previous work on a group of 263 normal students which showed a two-factor solution of the TAS-20. The French version of the TAS-20 seems to be reliable and constitutes a useful instrument to study alexithymia in psychopathology.