To improve bioavailability and achieve a smoother plasma-concentration profile as compared with oral administration, a matrix-dispersion-type transdermal delivery system was designed and developed for propranolol using different ratios of hydroxypropylmethylcellulose (HPMC) K4M, K15M and K100M. Formulations were evaluated for in-vitro dissolution characteristics using a Cygnus' sandwich-patch holder. Drug release followed Higuchi rather than zero-order or first-order kinetics. In-vivo evaluation was carried out on healthy volunteers (21+/-1.41 years; 60.89+/-5.35 kg) following the balanced incomplete block design. The dissolution rate constant (k) and data generated from plasma and urine (Cmax, maximum plasma concentration; t(max), time to reach peak plasma concentration; AUC, area under the curve; k(e), elimination rate constant; t1/2e, elimination half-life; k(a), absorption rate constant; t1/2a, absorption half-life) were evaluated statistically by two-way analysis of variance. Statistically excellent correlation was found between the percentage of drug absorbed and Cmax, AUC0-24 and AUC0-infinity. A highly significant difference (P < 0.001) was observed when Cmax and AUC0-infinity, generated from plasma and urine were compared, but k(e), t1/2e, k(a) and t1/2a did not differ significantly (P > 0.1). We conclude that urinary excretion data may be used as a simpler alternative to blood level data in studying the kinetics of absorption and deriving the absorption parameters.