The purpose of this study was to apply the method of allometric analysis to a study of the comparative disposition of veterinary drugs using the Food Animal Residue Avoidance Databank (FARAD) as a source of the comparative pharmacokinetic data. An initial filtration of the FARAD data was performed in order to exclude drugs for which no pharmacokinetic data were available, in at least four species the route of administration was other than intravenous, and the matrix was different from blood, plasma or serum. This process restricted the study to a total of 44 candidate drugs. The primary pharmacokinetic parameter selected for study was half-life (t1/2). As this parameter is a composite of clearance (Cl) and volume of distribution (Vd), it was considered to be the most robust for interspecies scaling. Volume of distribution at steady state (Vdss) and clearance showed weak allometric correlations with weight across species. The relationships between body weight and elimination half-life (51/2 beta) were determined for this selected group of drugs by using the empirically determined function Y = a Wb. The function Y represents the parameter of concern (half-life), a is a coefficient typical of every drug (intercept), W is the species average body weight, and b is the scaling exponent. A total of 11 drugs (tetracycline, oxytetracycline, chlortetracycline, erythromycin, diazepam, prednisolone, cephapirin, ampicillin, gentamicin, apramycin and carbenicillin) showed statistically significant correlations and consequently are excellent candidates for interspecies extrapolation of pharmacokinetic parameters (half-life) in species of relevance to veterinary medicine. The remaining 33 drugs were divided into two groups which showed various degrees of lack of correlation. Many of the drugs that showed no allometric correlation were low hepatic extraction drugs. However, some other drugs demonstrated equivocal results which could either be due to a true lack of allometric correlation, or be inconclusive due to the lack of quality data or excessive variability due to the multi-laboratory origin of the FARAD data. The results of this study show that interspecies scaling is applicable to certain veterinary drugs. The experimental determination of the coefficients of the allometric equation for relevant pharmacokinetic parameters (clearance and volume of distribution) could be an important tool in estimating dose in species where the drug has never been studied. This could have important consequences in terms of avoiding the use of dose-titration studies in Phase I of drug development, for drugs that are experimentally 'well behaved.'