Inhalation is a very old method of drug delivery, and in the 20th century it became a mainstay of respiratory care, known as aerosol therapy. Use of inhaled epinephrine for relief of asthma was reported as early as 1929, in England. An early version of a dry powder inhaler (DPI) was the Aerohalor, used to administer penicillin dust to treat respiratory infections. In the 1950s, the Wright nebulizer was the precursor of the modern hand-held jet-venturi nebulizer. In 1956, the first metered-dose inhaler (MDI) was approved for clinical use, followed by the SpinHaler DPI for cromolyn sodium in 1971. The scientific basis for aerosol therapy developed relatively late, following the 1974 Sugarloaf Conference on the scientific basis of respiratory therapy. Early data on the drug-delivery efficiency of the common aerosol delivery devices (MDI, DPI, and nebulizer) showed lung deposition of approximately 10-15% of the total, nominal dose. Despite problems with low lung deposition with all of the early devices, evidence accumulated that supported the advantages of the inhalation route over other drug-administration routes. Inhaled drugs are localized to the target organ, which generally allows for a lower dose than is necessary with systemic delivery (oral or injection), and thus fewer and less severe adverse effects. The 3 types of aerosol device (MDI, DPI, and nebulizer) can be clinically equivalent. It may be necessary to increase the number of MDI puffs to achieve results equivalent to the larger nominal dose from a nebulizer. Design and lung-deposition improvement of MDIs, DPIs, and nebulizers are exemplified by the new hydrofluoroalkane-propelled MDI formulation of beclomethasone, the metered-dose liquid-spray Respimat, and the DPI system of the Spiros. Differences among aerosol delivery devices create challenges to patient use and caregiver instruction. Potential improvements in aerosol delivery include better standardization of function and patient use, greater reliability, and reduction of drug loss.