Selective serotonin reuptake inhibitors (SSRIs) are widely prescribed and widely regarded as a first-line treatment for depression. Yet, a growing body of evidence indicates that these agents are only moderately more effective than placebo in treating major depressive disorder. In recent years, it has been debated whether SSRIs offer any clinically meaningful advantage over placebos. As part of this debate, it has been argued that these agents are first-line treatments for some forms of depression but not necessarily for others. The present paper examines two hypotheses that are central to these issues. The first hypothesis is that SSRIs are more effective than placebo for some types of depression but not for others. The second is that SSRIs are more effective than psychotherapies for some types of depression than others. A review of the empirical literature reveals three main classifications of depression that are relevant to the first hypothesis: (a) more vs. less severe depression, (b) melancholic vs. non-melancholic depression, and (c) depression defined according to associated genetic factors (particularly the long vs. short allele of the serotonin transporter gene promoter). There is no strong or consistent support for (a) or (b). There is, however, emerging and consistent evidence for (c), and so the first hypothesis is tentatively supported, but only for (c). Most of the empirical evidence does not support the second hypothesis. Psychotherapies (cognitive-behavioral and interpersonal therapies) and SSRIs generally have equivalent efficacy, regardless of the severity of depression. The research literature also suggests a third hypothesis that remains to be evaluated: that SSRIs are more effective for treating anxiety disorders (and possibly other disorders) than they are for treating depression. If that hypothesis is supported by subsequent research, then the future of SSRIs may lie largely in the treatment of anxiety disorders, and in the management of particular subtypes of depression.