An investigation into the physico-chemical and molecular mechanical properties of tertiary amines which predispose them to rapid N-dealkylation by cytochrome P450 has been undertaken. The rate of disappearance of the substrate (ke) from microsomal incubations was used as the index for determining the effect of changing the nitrogen substituent group within the series. In addition, individual routes of metabolism (N-demethylation and N-dealkylation) were quantified. Initially, untreated hepatic microsomal fractions from male and female rat, and Beagle dog were employed to investigate sex and species differences in the metabolism of the tertiary amines. The rate constants for metabolism were compared with physicochemical and molecular mechanical properties of the substrates but no significant dependence was found for metabolism on any of these parameters. The lack of significant finding was attributed to the heterogeneity of the constitutive cytochrome P450 system. Rats were pretreated with phenobarbitone and dexamethasone to obtain a predominance of specific isoenzymes CYP2B1 and CYP3A, respectively. The hydrophobicity of the substrate, as measured by the distribution coefficient (log D), was an important determinant of the rate of metabolism in both induced microsomal fractions. In addition, the strength of binding of the substrate to the enzyme (Ks) was important in the phenobarbitone-induced microsomal fraction whereas increasing rate of metabolism in the dexamethasone-induced model showed a significant dependence on increasing size of the substrate (length, area and calculated molar refractivity). Specific inhibitors of CYP2B1 (orphenadrine) and CYP3A (UK-122,802) were employed in phenobarbitone-induced microsomal fraction to further dissect out rate constants for metabolism attributable to these isoenzymes. The rate of metabolism attributable to CYP2B1 showed a dependence on log D and the binding affinity of the substrates. In addition, the electron density of the nitrogen atom was an important determinant of metabolism and this may be related to the mechanism of N-dealkylation. Optimum values of calculated molar refractivity (7. 45) and area (1320A2) were calculated for the individual routes of metabolism. (N-demethylation and N-dealkylation) attributable to CYP2B1. This implies that the CYP2B1 active site is constrained. Support for this came from the observation that N,N-bis[2-2,4-dichlorophenoxy)-ethyl]methylamine (bisDME), whose calculated molar refractivity (CMR) and area exceed the optima, was not metabolized by CYP2B1. In addition, molecular modelling studies indicated that the active site of CYP2B1 is too constrained to readily accommodate the n-alkyl chain of 2-(2,4-dichlorophenoxy)-N-pentyl-N-methylethanamine (pentylDME) when the molecule is oriented for N-depentylation. All of the tertiary amines were substrates for CYP3A. Little dependence was seen for the rate of metabolism attributable to this isoenzyme on any physico-chemical or molecular mechanical properties of the substrates. It is speculated that the series does not cover a wide enough range of lipophilicity or size for this to influence their interaction with CYP3A, which is visualized as having a large active site. This view is supported by the observations that bisDME was metabolized by CYP3A, and also that as CYP2B1 metabolism became impeded by size, eg. pentylDME, the role was taken over by CYP3A.