1. The rationale for extrapolation or 'scaling' across animal species is based on their underlying anatomical, physiological and biochemical similarities. 2. Research carried out in the 19th and early 20th century resulted in Benedict's famous 'mouse-to-elephant' graph which showed that the log of the basal metabolic rate plotted against the log of bodyweight (W) produced a straight line with a slope of 0.76. Since then it has become apparent that a number of other physiological variables (Y) exhibit a similar relationship which can be represented by the general allometric equation, Y = alpha W beta; where beta is the slope of the log-log plot and alpha is the intercept on the y axis. 3. The major pharmacokinetic parameters such as clearance and volume of distribution of many drugs are also related to W in a similar manner. 4. This empirical approach does not require a strong mathematical background and offers a relatively simple method of predicting the kinetics of anti-cancer drugs in patients from pre-clinical animal data. 5. The occurrence of major qualitative and quantitative differences in the metabolism of drugs between species is probably the single greatest complicating factor in the use of animals as predictors of drug toxicity and kinetics in patients. 6. Despite this, the allometric approach is useful for allowing the estimation of a more appropriate starting dose for some drugs in a Phase I trial, which might result in potential savings in escalation steps and maximize the chance that the dose which an individual receives has the potential for therapeutic value.