Pretreatment is one of the biggest challenges in utilizing lignocellulosic feedstocks to meet the mandatory requirements for biofuels around the world. Earlier researchers evaluated extrusion and ozone pretreatment separately and found that sugar recovery can be improved significantly from 15-20 to 40-75 % for different feedstocks. To further improve sugar recoveries, extrusion-ozone sequential pretreatment was explored. Accordingly, optimal extruded switchgrass (176 °C, 155 rpm, 20 % moisture, and 8 mm) and big bluestem (180 °C, 155 rpm, 20 % moisture, and 8 mm) at 25-75 % moisture content were exposed to an ozone flow rate of 37-365 mg/h for 2.5 to 10 min. Pretreated samples were then subjected to enzymatic hydrolysis to determine sugar recovery. Statistical analyses confirmed significant effects of the independent variables and their interactions on sugar recoveries for both feedstocks. Maximum glucose, xylose, and total sugar recovery of 66.4, 82.3, and 70.4 %, respectively, were obtained when a low-moisture (25 %) extruded switchgrass was ozonated for 2.5 min at a flow rate of 37 mg/h. Respectively, this represents increases of 3.42, 5.01, and 3.42 times that of the control. When big bluestem at 25 % moisture was extruded and then ozonated for 2.5 min at a flow rate of 365 mg/h, resulting glucose, xylose, and total sugar recoveries of 90.8, 92.2, and 87.5 %, respectively, were obtained. These represent increases of 4.5, 2.7, and 3.9 times than that of the control. It is also noteworthy that furfural and hydroxymethyl furfural were not detected in any of the pretreatments, and only low levels (0.14-0.18 g/l) of acetic acid were measured. The results show that sequential pretreatment using extrusion and ozone is an efficient way to improve sugar recovery from herbaceous biomass feedstocks.