This article describes the shortcomings of one of the major testing tools in the foreign languages field, the Simulated Oral Proficiency Interview (SOPI), with regard to the elicitation of the function of description. In doing so, the article raises questions about the applicability of the SOPI as a surrogate for the Oral Proficiency Interview (OPI) in general and, more specifically, about SOPI testing oral proficiency at the Intermediate-High level and above. The SOPI and the OPI are not designed on the basis of the same conceptualization of description. Even though both tests base their assessment criteria on the ACTFL Proficiency Guidelines-Speaking (1999), in the SOPI both the definition of description and the description prompts themselves are problematic. Results of the research project described in this article-a comparison of SOPI and OPI Russian tests-suggest that the Russian version of the SOPI, for example, does not elicit description and therefore the SOPI test, in its current version, may be unreliable for ratings at the Advanced and Superior levels. (Contains 5 notes and 2 tables.)