Equations for non-destructive determination of chlorophyll b : a ratios in grasses were developed from reflectance spectra of intact leaves of barley (Hordeum vulgare L.) and two barley mutants: clorina f2, which lacks chlorophyll b and clorina f104, which has a low chlorophyll b content. These plants enabled separation of effects of chlorophyll composition on reflectance spectra due to differential light absorption by chlorophylls a and b and to measure the effects of chlorophyll b on the contribution of fluorescence emitted by chlorophyll a to the reflectance spectra. Indices developed from these data were then tested on growth chamber-grown leaves from six C₃ and 17 C₄ grass species (7 NAD-ME and 10 NADP-ME subtypes). We used the chlorophyll b : a ratio because the data were less skewed than the chlorophyll a : b ratio. The best index for determination of the chlorophyll b : a ratio utilised wavelengths affected by chlorophyll absorbance: [R₆₂₆ – 0.5 (R₆₀₃ + R₆₄₇)]/[R₅₅₂– R₆₂₆]. The chlorophyll b : a ratio was significantly lower in the C₄ than C₃ grasses, but was not sufficient in itself to separate these two functional groups. However, because of differences in fluorescence characteristics, C₃ and C₄ species could be distinguished by an index based on wavelengths affected by chlorophyll fluorescence: [R_{696 to 709}/R_{545 to 567}].