Scattering from a leaf responds differently at different wavelengths to changes in leaf properties such as pigment concentrations, chemical constituents, internal structure, and leaf-surface properties. Radiation scattered by leaves and exiting the vegetation canopy toward the sensor is affected by canopy structure. The concept of canopy spectral invariants is used to decompose multiangular hyperspectral Compact High Resolution Imaging Spectroradiometer-PROBA surface reflectances over agricultural crops during peak growth season into structural and optical components. The former, called the directional area scattering factor, is determined by the canopy geometrical properties and varies with crop type. The latter is a function of the leaf scattering properties and more directly related to the leaf interior. For dense crops, the decomposition technique does not require the use of canopy radiation models, prior knowledge, or ancillary information regarding the leaf scattering properties and thus provides a powerful means to remove canopy structural influences in hyperspectral remote sensing of leaf biochemical constituents. Our results also suggest that leaf-surface characteristics can increase canopy scattering spectra. This may decrease the ability to remotely sense leaf biochemistry.