Abstract. The objective of this study was to prepare and evaluate terbutaline sulphate (TBS) bi-layer tablets for once-daily administration. The bi-layer tablets consisted of an immediate-release layer and a sustained-release layer containing 5 and 10 mg TBS, respectively. The sustained-release layer was developed by using Compritol®888 ATO, Precirol® ATO 5, stearic acid, and tristearin, separately, as slowly eroding lipid matrices. A full 4×2 2 factorial design was employed for optimization of the sustainedrelease layer and to explore the effect of lipid type (X 1 ), drug-lipid ratio (X 2 ), and filler type (X 3 ) on the percentage drug released at 8, 12, and 24 h (Y 1 , Y 2 , and Y 3 ) as dependent variables. Sixteen TBS sustained-release matrices (F1-F16) were prepared by melt solid dispersion method. None of the prepared matrices achieved the targeted release profile. However, F2 that showed a relatively promising drug release was subjected to trial and error optimization for the filler composition to develop two optimized matrices (F17 and F18). F18 which consisted of drug-Compritol®888 ATO at ratio (1:6 w/w) and Avicel PH 101/dibasic calcium phosphate mixture of 2:1 (w/w) was selected as sustained-release layer. TBS bilayer tablets were evaluated for their physical properties, in vitro drug release, effect of storage on drug content, and in vivo performance in rabbits. The bi-layer tablets showed acceptable physical properties and release characteristics. In vivo absorption in rabbits revealed initial high TBS plasma levels followed by sustained levels over 24 h compared to immediate-release tablets.