The present paper describes the attempt to cross-validate the three-factorial structure of the German Version of the 20-Item Toronto Alexithymia Scale in a population of 419 patients, suffering of psychogenic disorders. However, results neither revealed good internal consistencies nor was it possible to obtain an acceptable adaptation of the empirical data on the postulated model. Therefore an explorative factor analysis was performed in order to get a more consistent factor model. The hereby resulting four-factorial structure was congruent with reference to theoretical and empirical data of Alexithymia. However, additional cross-validation in other populations has to be investigated.