Biomass, typically obtained from fast growing plants or over-stocked agricultural products, has been widely recognised as a replacement for traditional energy sources. The challenge is, however, to provide biomass feedstock with optimised properties best suited for downstream conversion. Willow ( Salix sp.) is one of the most common hardwood species suitable for short-rotation forestry. The goal of this research was to explore the potential of the near infrared (NIR) spectroscopy to evaluate the chemical composition of several willow clones. It was shown that near infrared spectroscopy can be an alternative technique to standard analytical methods supporting research and development of biomass production technologies. Partial least squares (PLS) regression models for quantitative prediction of wood components (lignin, cellulose, holocellulose, pentosans and extractive components soluble in hot/cold water, 1% NaOH and organic solvents) were developed. Using NIR spectroscopy, it was possible to discriminate different willow clones and to assign these into groups by means of principal components analysis (PCA).