Are odors diagnostic? In this age of polymerase chain reactions, in situ hybridization, and immunohistochemical staining, is there any room left for the nose in diagnosing disease? Long ago, and perhaps far away, smell was crucial to describing an illness. Infectious diseases were known by their characteristics odors--scrofula as smelling like stale beer; typhoid, like freshly baked brown bread; rubella, like plucked feathers; and diphtheria, as "sweetish." Anosmics might be banned from medical school. Perhaps we have left the descriptions behind along with these illnesses we rarely encounter today. After all, how many young physicians, residents, or medical students have ever seen a case of diphtheria or even rubella, and how many fewer have ever plucked a chicken? We have learned that pellagra (that "must appear" diagnosis in our differential by rote, but not by example, for photosensitive dermatoses) should smell like sour bread and that the exotic favus should smell "mousy" (Table 1). What does Candida smell like--a "heavy sweetness"? Darier's disease in poor control--"organic"? Pseudomonal infections--"foul and biting"? And are not our patients with noninfected eczematous dermatitis distinct for lacking any peculiar odor, do they not actually smell "dry"? We cannot blame the abandonment of our olfactory skills on the younger generation, for how many of us could describe those odors we smell every day? Would we be able to detect a subtle change in the odor of our patient with psoriasis, a change perhaps signifying superinfection?