Digitized images from black and white video cameras are being used to measure the area and numbers of particles in cotton. The method now provides trash grades for nearly all U. S. cotton. Small samples from each bale are compressed against a glass window and illuminated for imaging. Trash area readings are based on the relative darkness of each particle when compared to the cotton lint. Herein, we report the results of visible and near-infrared spectral reflectance from several different types of trash found in cotton. The feasibility of identifying different types of trash particles such as grass and bark is discussed.