When methyl tertiary-butyl ether (MTBE) in gasoline was first introduced to reduce vehicle exhaust emissions and comply with the Clean Air Act, in the United States, a pattern of complaints emerged characterised by seven "key symptoms." Later, carefully controlled volunteer studies did not confirm the existence of the specific key symptoms, although one study of self-reported sensitive (SRS) people did suggest that a threshold at about 11-15% MTBE in gasoline may exist for SRSs in total symptom scores. Neurobehavioral and psychophysiological studies on volunteers, including SRSs, found no adverse responses associated with MTBE at likely exposure levels. MTBE is well and rapidly absorbed following oral and inhalation exposures. Cmax values for MTBE are achieved almost immediately after oral dosing and within 2 h of continuous inhalation. It is rapidly eliminated, either by exhalation as unchanged MTBE or by urinary excretion of its less volatile metabolites. Metabolism is more rapid humans than in rats, for both MTBE and tert-butyl alcohol (TBA), its more persistent primary metabolite. The other primary metabolite, formaldehyde, is detoxified at a rate very much greater than its formation from MTBE. MTBE has no specific effects on reproduction or development, or on genetic material. Neurological effects were observed only at very high concentrations. In carcinogenicity studies of MTBE, TBA, and methanol (included as an endogenous precursor of formaldehyde, without the presence of TBA), some increases in tumor incidence have been observed, but consistency of outcome was lacking and even some degree of replication was observed in only three cases, none of which had human relevance: alpha(2u)-globulin nephropathy-related renal tubule cell adenoma in male rats; Leydig-cell adenoma in male rats, but not in mice, which provide the better model of the human disease; and B-cell-derived lymphoma/leukemia of doubtful pathogenesis that arose mainly in lungs of orally dosed female rats. In addition, hepatocellular adenomas were significantly higher in female CD-1 mice and thyroid follicular-cell adenomas were increased in female B6C3F1 mice treated with TBA, but these results lack any independent confirmation, which would have been possible from a number of other studies.