The updated Vitek Yeast Biochemical Card (YBC) was compared with the API 20C by using 409 germ tube-negative yeasts and Geotrichum spp. that were either clinical or proficiency sample isolates. The API 20C was the reference standard. The 409 isolates represented nine genera and 21 species. Morphology agars were inoculated and interpreted for each isolate. The API 20C identified 406 isolates (99.3%), while the Vitek YBC identified 367 (89.7%). Both systems identified the majority of yeasts after 24 h of incubation--73.4% were identified by the API 20C and 77.4% were identified by the Vitek YBC. The Vitek 24-h reading had some incorrect identifications. These included 14 isolates of Candida tropicalis that were identified as Candida parapsilosis (91 to 97% reliability) and 3 isolates of Candida krusei that were called Blastoschizomyces capitatus (Geotrichum capitatum), Candida rugosa, and Candida zeylanoides. In total, the Vitek YBC misidentified 30 isolates, while the API 20C misidentified 3 isolates. In addition, results for 14 isolates with the Vitek YBC were listed under the category "no identification." Morphology agars were required for identification with 89 isolates (21.9%) when the API 20C was used and with 50 isolates (12.6%) when the Vitek YBC was used. Apart from the price of the Vitek instrument, the API 20C costs $1.28 more per test than the Vitek YBC. Overall, the updated Vitek YBC compares favorably with the API 20C in the identification of common yeasts such as Torulopsis glabrata, C. parapsilosis, and Cryptococcus neoformans. However, problems were encountered with the Vitek system in the identification of C. tropicalis, C. krusei, Trichosporon spp., and some Cryptococcus spp. The routine use of morphology agars with either method is recommended.