Near infrared spectra was collected from leaf litter decomposed in mesh litterbags over two years from sites located in the Australian wet tropics bioregion, to determine chemical contents, model decay, for comparisons in the chemical dynamics during decomposition between locations, and subsequent studies into decomposition. Two litter bag treatments were undertaken: one used leaves collected in the litter traps and exposed at their respective sites (in situ study), and the other used leaf litter from the deciduous tree Archidendron vaillantii (F.Muell.) F.Muell. (control). The litter bag study spanned ≈ 350 d (NIR-spectral collection) with collections at four intervals. After collection from the field leaf samples were oven dried at 40°C to constant weight, and ground using a cyclone mill (Foss Cyclotec 1093 sample mill, North Ryde NSW Australia) until they passed through a 1 mm mesh. Spectra were obtained with a Fourier Transform Near Infrared Spectrophotometer Multipurpose Analyser (MPA) (Bruker Optics Inc., Clayton, Australia), using a 30-position sample wheel, on two sampling occasions in March 2008 and April 2009 (1st and 2nd year collections, respectively). Each sample was separated into two 2-cm-diameter vials, with spectral-reflectance data obtained between 780 and 2780 nm. All samples were packed into the vials with an equal pressure, by tapping down consistently on a table, to minimise errors due to inconsistent light scattering. Spectra were converted to absorbance by the logarithm of the reciprocal of reflectance (log 1/R). After collection of the NIR spectra, the analysis of the chemical composition of the leaf litter was accomplished in four general steps: selection of a representative calibration set, wet-chemical analysis of the calibration set, model development, and prediction of unknown sample concentrations. For this full method see: Parsons, S.A., Lawler, I.R., Congdon, R.A. & Williams, S.E., 2011, Rainforest litter quality and chemical controls on leaf decomposition with near infrared spectrometry, Journal of Plant Nutrition and Soil Science, 174(5), pp. 710-20. Two files are supplied here, "spectra.csv" contains the spectra (columns = wavelengths 800.44 nm to 2779.81nm, labeled “X800.44” to “X2778.81”, rows = sample names), and "sample names.csv" contains the key to the sample names including the site, time of collection, % mass remaining, and chemical data (% contents) as determined in Parsons et al. 2011. Spectra are supplied as 1st derivatives after scatter correction (Parsons et al. 2011).