The potential for early detection of spider mites in cotton in California was investigated using a digital multispectral remote sensing system. Images were acquired on 26 dates during the 1998 growing season in the green (550 nm), red (660 nm), and near infrared (850 nm) wavelengths. Results indicate that through the use of computer enhancement and digital processing, spider mite damage can be detected early in the season and distinguished from other types of image anomalies, such as water stress.