Moderate loadings of cellulase enzyme supplemented with β-glucosidase were applied to solids produced by ammonia fiber expansion (AFEX), ammonia recycle (ARP), controlled pH, dilute sulfuric acid, lime, and sulfur dioxide pretreatments to better understand factors that control glucose and xylose release following 24, 48, and 72 h of hydrolysis and define promising routes to reducing enzyme demands. Glucose removal was higher from all pretreatments than from Avicel cellulose at lower enzyme loadings, but sugar release was a bit lower for solids prepared by dilute sulfuric acid in the Sunds system and by controlled pH pretreatment than from Avicel at higher protein loadings. Inhibition by cellobiose was observed to depend on the type of substrate and pretreatment and hydrolysis times, with a corresponding impact of β-glucosidase supplementation. Furthermore, for the first time, xylobiose and higher xylooligomers were shown to inhibit enzymatic hydrolysis of pure glucan, pure xylan, and pretreated corn stover, and xylose, xylobiose, and xylotriose were shown to have progressively greater effects on hydrolysis rates. Consistent with this, addition of xylanase and β-xylosidase improved performance significantly. For a combined mass loading of cellulase and β-glucosidase of 16.1 mg/g original glucan (about 7.5 FPU/g), glucose release from pretreated solids ranged from 50% to75% of the theoretical maximum and was greater for all pretreatments at all protein loadings compared to pure Avicel cellulose except for solids from controlled pH pretreatment and from dilute acid pretreatment by the Sunds pilot unit. The fraction of xylose released from pretreated solids was always less than for glucose, with the upper limit being about 60% of the maximum for ARP and the Sunds dilute acid pretreatments at a very high protein mass loading of 116 mg/g glucan (about 60 FPU). Biotechnol. Bioeng. 2009;102: 457–467. © 2008 Wiley Periodicals, Inc.