Dysthymia is highly prevalent--though underdiagnosed--occurring in at least 3% of the population. We conceptualize it as the clinical extension of adaptive traits that have developed during evolution to cope with stress and failure. A classification of dysthymias into anxious and anergic subtypes--and their putative association to bipolarity--is proposed. We further posit neurochemical and neurophysiological substrates for the two subtypes. A better recognition and understanding of dysthymic subtypes and their respective place in the affective spectrum will increase the proportion of people that may benefit from targeted treatments. It would also expand the pool of subjects that may be enrolled in genetic and pharmacogenomic research studies.