The giant panda (Ailuropoda melanoleuca) subsists on bamboo, but lacks gastrointestinal modifications for fiber digestion. Pandas display dietary selectivity for certain plant parts; however, cues for selection are unknown, and few have examined panda feeding behavior and bamboo nutrient composition simultaneously. Bamboos are clonal plants, with seasonal cycles of nutrient accumulation related to peak photosynthesis and recruitment. These cycles can be destabilized by aboveground disturbance; however, effects of disturbance on nutritional quality and sustainability of bamboo for primary consumers have not been studied. I examined 4 factors that may contribute to bamboo nutritional quality or sustainable harvest: season, ramet maturity and age, and disturbance. I conducted a 3-year study with Phyllostachys aureosulcata (PLLAU), P. glauca (PLLGL), and P. rubromarginata (PLLRU), at the Shelby County AgriCenter, Memphis TN, applying 3 disturbance treatments in a replicated split-plot design: control, 3% and 20% annual removal of biomass. Bamboo was sampled 8 times/year and divided into shoot, leaf, and culm (central stem) for analysis of crude protein (CP), neutral- and acid-detergent fiber (NDF and ADF), ash, lipid, minerals, and acid-insoluble ash (AIA). I quantified bamboo-stand production, maturity, recruitment and mortality. Biomass removed from PLLAU was used in diet-selection trials with 2 giant pandas, observing consumption of plant parts and bamboo from different disturbance treatments, and sampling bamboo for allelochemical and starch analysis. Disturbance effects were evident only for branch and culm dry mass in PLLRU. In Year 2, drought conditions and peak roosting blackbird populations confounded results. Defoliation and guano deposition preceded changes to soil chemistry, bamboo composition, and stand structure; recruitment increased for PLLRU. After accounting for confounding factors, less-mature bamboo had greater concentrations of most nutrients, but less NDF, ADF, and K. Two-year-old ramets had greater CP, NDF, lipid, P, K, S, and Cu, but less Ca and Fe, than 1-year-old ramets. I confirmed seasonal plant-part selection by pandas, but found no correlation between leaf consumption and AIA concentration. Culm starch analyses were insufficient to establish a clear pattern. I recommend further investigation of non-structural carbohydrates in bamboo, and palatability testing of undisturbed versus disturbed bamboo, which I was unable to examine.