The present study investigated the validity of the two-factor solution of items selected from the Rumination Scale of the Response Style Questionnaire proposed by Treynor, Gonzalez, and Nolen-Hoeksema (2003). In the first part of this study we used samples of currently depressed (MDD), formerly depressed (FD), socially anxious (SP), and healthy control participants to examine whether the brooding and reflective pondering components differentiate participants with an anxiety disorder from participants with depression. In the second part of this study we examined whether these components of rumination were differentially related to cognitive biases in depression. Overall, the MDD group exhibited higher brooding scores than did all other groups; SP and FD groups did not differ from each other but obtained higher brooding scores than did the control participants. Only the MDD and the control groups differed on the reflective pondering factor. Importantly, brooding and reflective pondering were differentially related to cognitive biases. Specifically, the correlation between brooding/reflective pondering and memory bias was not significant when depressive symptoms were partialed out. The correlation between brooding and attentional bias for sad faces, however, remained significant even when current depressive symptoms were taken into account. In sum, our results support the formulation that rumination is composed of an adaptive reflective pondering factor and a maladaptive brooding factor.