Four operating parameters of supercritical carbon dioxide extraction were optimized by the response surface method. The central composite rotatable design was used to maximize extraction recovery of essential oil (specifically diosgenin) from Tribulus terrestris. The obtained data were adequately fitted into two second-order polynomial models for essential oil and diosgenin. The linear terms of pressure, temperature, and extraction dynamic time, the quadratics of pressure and extraction dynamic time, and the interactions of the following two variables: (i) pressure and temperature, (ii) pressure and extraction dynamic time, and (iii) temperature and extraction dynamic time had significant effects on diosgenin recovery.