Some aromatic solvents (e.g. toluene, p-xylene, styrene, and ethylbenzene) show, in the rat, striking ototoxicity characterized by an irreversible hearing loss, as measured by behavioural or electrophysiological methods, associated with damage to outer hair cells in the cochlea of the exposed animals. To broaden the range of aromatic solvents studied concerning their potential ototoxicity and to compare their ototoxicity quantitatively, 21 aromatic solvents were administered orally by gastric intubation to Sprague-Dawley rats for 5 days/week for a 2-week period. The dose used was 8.47 mmol kg(-1) body weight day(-1). The possible ototoxicity of the aromatic solvents was evaluated by morphological investigation of the cochlea. Whole-mount surface preparations of the organ of Corti were made to quantify the number of missing hair cells (cytocochleogram). Among the 21 solvents studied, eight (toluene, p-xylene, ethylbenzene, n-propylbenzene, styrene, alpha-methylstyrene, trans-beta-methylstyrene, and allylbenzene) caused histological lesions of the organ of Corti. They differed widely in their potency. The least ototoxic solvents caused outer hair cell (OHC) loss in the middle turn of the organ of Corti. The OHC loss was slight in the first row, and greater in the second and third rows. The most ototoxic solvents caused high losses in the three rows of the outer hair cells along the entire length of the basilar membrane. There were also occasional inner hair cell (ICH) losses in the most affected animals. Although no measurements were made of the chemical concentrations reached in the blood or the brain, tentative ranking of an increasing ototoxicity of the eight aromatic solvents could be proposed on the basis of the histological losses observed-alpha-methylstyrene<trans-beta-methylstyrene=toluene< or =p-xylene<n-propylbenzene<styrene=ethylbenzene<allylbenzene. There was no relationship between the degree of ototoxicity and the lipophilic properties of the ototoxic agents as expressed by the octanol/water partition coefficients. However, it seemed that some structural constraint was essential to induce ototoxicity. It seems there must be a single side-chain on the aromatic ring for ototoxicity, except with p-xylene. The other aromatic solvents with two side-chains were not ototoxic. When the saturated side-chain was branched (isopropylbenzene, isobutylbenzene, sec-butylbenzene, tert-butylbenzene), no ototoxicity was observed. The ototoxic potency increased when the length of the saturated side-chain extended from one carbon atom to two carbon atoms. Beyond that point, the ototoxic effect decreased with n-propylbenzene and disappeared with n-butylbenzene. Moreover, unsaturation of the side-chain of allylbenzene increased the ototoxicity of n-propylbenzene substantially. Branching of the unsaturated chain (alpha-methylstyrene and trans-beta-methylstyrene) decreased the ototoxicity of styrene.