Expectancy theory posits that anxiety sensitivity (AS) acts as a specific risk factor for the development of anxiety pathology (Reiss, 1991). Previous work suggests that AS is a risk factor for anxiety but several reports have found that AS is also related to depression. The principal aim of the present study was to determine whether anxiety sensitivity acts as a specific vulnerability factor in the pathogenesis of anxiety and depression in both a large nonclinical sample (N = 1401) as well as a patient sample (N = 53). A covariance analytic strategy indicated that AS possesses symptom specificity with respect to anxiety but is not predictive of depression when accounting for changes in anxiety symptoms. Component analyses suggest, however, that one first-order factor (phrenophobia) is likely to account for the association between AS and depression because it is non-specific (i.e. associated with unique aspects of both anxiety and depression). It is concluded that much of the general association noted between anxiety sensitivity and depression is likely to be due to covariation among symptoms of anxiety and depression.