Three studies that address plant cell wall recalcitrance and its interaction with bioconversion processes are presented in this dissertation. First, the effects of commercial cellulase and hemicellulase mixtures on sugar productions from two perennial biomasses were measured. A low-impact, high-diversity mixture of prairie biomasses (LIHD) and switchgrass were each subjected to NaOH pretreatment, followed by hydrolysis with a commercial cellulase and [beta]glucosidase mixture supplemented with either of two hemicellulases. A novel modeling approach was presented that used "marginal effectiveness" (incremental grams sugar per incremental mg enzyme) to determine enzyme loadings required for optimal sugar productions. Results suggested that there was no need to customize cellulase loading or hemicellulase supplementation for NaOH-pretreated switchgrass and LIHD. Second, Fourier transform infrared spectroscopy (FTIR) combined with partial least squares (PLS) regression was used to estimate glucose and xylose productions from NaOH pretreatment and enzymatic hydrolysis of six plant biomasses that represent a variety of potential biofuel feedstocks. Two switchgrass cultivars, big bluestem grass, LIHD, mixed hardwood, and corn stover were subjected to four levels of NaOH pretreatment to produce 24 samples with a wide range of potential digestibility. PLS models were constructed by correlating FTIR spectra of pretreated samples to measured values of glucose and xylose conversions (g sugar per 100 g potential sugar) and yields (g sugar per 100 g TS). Third, PLS models constructed from FTIR spectra of the six raw biomasses plus four levels of pretreatment (0, 5, 10, and 20 g NaOH per 100 g TS) satisfactorily predicted solubilizations of plant cell wall constituents through pretreatment. Additionally, PLS models constructed from FTIR spectra of the six raw biomasses plus three levels of pretreatment (5, 10, and 20 g NaOH per 100 g TS) accurately predicted overall glucose and xylose conversions (g sugar per 100 g potential sugar) and yields (g sugar per 100 g TS) from pretreatment plus enzymatic hydrolysis. The ability to predict sugar yields without prior knowledge of biomass composition suggests that FTIR combined with PLS regression may be able to replace wet chemical analyses and enzymatic assays in estimating saccharification from lignocellulosic biomass.