During the cotton harvest, other parts of the cotton plants in addition to the cotton fibers are collected. This plant debris causes a number of problems in cotton processing, including decreased yarn quality and increased number of yarn breakages during spinning. Because these debris break down in size physically after harvest, human visual identification of and discrimination among these contaminants are difficult or impossible, but identification of the contaminant type is desirable for investigations of process breakdowns and efficiency monitoring of cleaning processes. To achieve contaminant identification and discrimination, a number of projects that explore spectral comparison in general and develop solutions to the specific problem of cotton contaminant discrimination is reported. Chemometric methods for contaminant class discrimination have been developed for and applied to the FT-IR attenuated total reflection (ATR) spectra of cotton contaminants. Novel voting scheme algorithms were developed to improve spectral identification by library searching for a USDA spectral library of cotton contaminants. Improvements in contaminant identification were also realized via partial least squares discriminant analysis (PLS-DA). Quantitative analysis of cotton contaminant mixtures was achieved with the use of partial least squares (PLS) regression and a novel error correction algorithm that was developed. This work also reports the development of a mixture generator algorithm to generate sets of mixtures representative of mixture spaces of arbitrary dimensions. Finally, the spectral differences caused by the use of different FT-IR spectrometers and ATR accessories were investigated by measuring spectra of a polyethylene terephthalate film with the use of several different spectrometers and accessories. The spectra were compared before and after corrections for depth of penetration and anomalous dispersion effects. The results show that these correction methods do not always achieve the goal of increased spectral similarity.