The interactive effects of lactic acid bacteria (LAB) and yeasts were identified through isolation of sourdough sponges. Fourteen sourdough samples were collected from different bakeries in Isparta. Carnobacterium divergens (Lactobacillus divergens) (6.1%), Lactobacillus brevis (15.1%), Lactobacillus amylophilus (6.1%), Lactobacillus sake (6.1%), Lactobacillus acetotolerans (6.1%), Lactobacillus plantarum (3.0%), Pediococcus pentosaceus (6.1%) and P. acidilactici (6.1%), Tetragenococcus halophilus (Pediococcus halophilus) (3.0%) were isolated from sourdoughs as LAB while Saccharomyces cerevisiae (27.0%), S. delbrueckii (2.7%), Torulopsis holmii (10.8%) and T. unisporus (2.7%) were also isolated from sourdough sponges as yeasts. Seven kinds of bread were made with the inoculation 1.5% S. cerevisiae and/or 1.5% lactobacilli (Lb. amylophilus, Lb. brevis, Lb. plantarum, Lb. sake and Lb. acetotolerans). Six kinds of bread were produced with the mixture of the above five different LAB and S. cerevisiae, and the seventh bread made with S. cerevisiae (commercial culture) alone as control bread. The sample prepared with 1.5% Lb. amylophilus and 1.5% S. cerevisiae showed the best results on theological properties of bread while that with 1.5% mixed LAB and 1.5% S. cerevisiae was the poorest. LAB used in sourdough breads increased shelf life and delayed staling. In sensory analyses, 1.5% Lb. sake and 1.5% S. cerevisiae application were preferred well by the panel. (C) 2004 Elsevier Ltd. All rights reserved.