Although the lower respiratory tract is frequently exposed to injurious agents, the lung does possess some ability to effect repair and thus restore the damaged alveolar wall to normal; however, in some circumstances, normal repair is not possible. The result is often a markedly deranged alveolus, with improper proportions of epithelial cells (eg, relatively more cuboidal type-2-like cells), a loss of endothelial cells or a migration of endothelial cells into improper locations, and a proliferation of interstitial fibroblasts with an accompanying deposition of a collagenous extracellular matrix (ie, fibrosis). Although the development of 'fibrosis' is frequently thought to be a form of attempted 'repair' of an injured alveolar wall, this concept is not clearly established; it is possible that the expansion of fibroblastic numbers in the alveolar wall is part of the disease process itself, resulting from alveolar macrophagic activation, rather than an attempt by the macrophage to 'repair' an injured alveolar wall. Thus, it is not known if the development of fibrosis represents 'healing' and thus is beneficial (as a localized scar 'heals' a localized incision in the skin) or whether it represents part of the disease process itself. The distinction is important, as it is unclear whether therapy should be directed against the development of fibrosis per se. If fibroblastic expansion and deposition of the connective tissue products of these fibroblasts are a useful form of repair, prevention of this process may cause future loss of pulmonary function. Alternatively, if 'fibrosis' compromises pulmonary function (particularly decreased compliance), prevention of fibrosis might be beneficial. It is apparent, therefore, that what is needed is an understanding of the processes that lead to alveolar parenchymal cellular repair and how such processes might be manipulated for the benefit of the patient.