The vasculitides in man remain a puzzling and poorly understood group of disorders in spite of considerable advances over the years in knowledge of the pathologic processes that can give rise to vasculitis in experimental animal models, the most well known of which are the Arthus phenomenon and experimental serum sickness or circulating immune complex disease. More recently, it has become apparent that cell-mediated immune reactions can be involved in the production of vasculitis in experimental animals. Nomenclature and classification remain unsatisfactory, but a useful distinction can at once be made on the basis of the size of vessel affected; in general, small vessel disease is unlikely to give rise to serious problems, whereas disastrous consequences can ensue when the large vessels are the site of disease. One of the more common and readily identifiable small vessel diseases is Henoch-Schönlein purpura in which IgA appears to have a prominent role as indicated by the presence of IgA deposits in the affected vessels, whether they be in skin, kidney, or another internal organ. IgA is produced in excess amounts, but it is not yet clear whether the IgA originates primarily from mucosal or bone marrow sites. The source and precise role of the IgA is not understood, although it may be a component of circulating immune complexes. The renal disease is unlikely to give rise to serious problems in the long term. In some ways it closely resembles that found in idiopathic IgA nephropathy. Treatment other than simple measures is rarely required, and the value of systemic corticosteroids is unproved and not without hazard. Urticarial vasculitis, sometimes with hypocomplementemia, is one disorder in which steroids may be of value. Safer than the steroids and worth trying in many small vessel vasculitides is dapsone. In another rare vasculitis, erythema elevatum diutinum, this is the drug of choice. Vasculitis associated with intermediate size complexes, as in benign hypergammaglobulinemic purpura of Waldenstrom, is indeed so benign that treatment is hardly ever required; thus differs from the vasculitis associated with either single component or the more common mixed component cryoglobulinemia in which skin and even internal involvement may be severe enough to justify plasmapheresis and cytotoxic therapy. The large vessel vasculitides often present as an obscure systemic illness without—in the early stages—any notable skin signs and only uninformative investigations. However, early diagnosis is highly desirable because progression of polyarteritis nodosa or Wegener's granulomatosis can occur fairly swiftly and result in severe and perhaps irreversible damage to various internal organs. In the absence of diagnostic histology, arteriography can in many instances reveal aneurysms indicative of polyarteritis nodosa. In the case of Wegener's granulomatosis and microscopic polyarteritis nodosa, the presence in the serum of antibodies against neutrophil cytoplasmic components appears to be of diagnostic value. It remains to be seen whether these recently described antibodies have a pathogenic role. For both polyarteritis nodosa and especially Wegener's granulomatosis, there are now several studies to indicate the value of the prompt introduction of therapy with cyclophosphamide in addition to steroids, and there have been evaluations of various dosage schedules. The nature of a group of vasculitides associated with endarteritis obliterans, perhaps with livedo reticularis, and on occasion with cerebrovascular disturbances remains to be elucidated, as does the involvement of the lupus anticoagulant that is found in some patients. Among the readily remediable causes of vasculitis are those that are druginduced and those that are caused by infections. Although very uncommon, the bacterial infections can in many instances be rapidly diagnosed by use of the appropriate stain on smears or biopsies.