Fluorescence (or Förster) resonance energy transfer (FRET) describes the flow of energy from a donor fluorophore to an acceptor chromophore. Among the parameters in determining FRET efficiency is the donor-acceptor energy matching, given by the overlap of the donor fluorescence spectrum and the acceptor absorption spectrum (J value). Calculations of the J value typically rely on experimentally acquired spectra. When a fluorescence spectrum is converted from the wavelength (nm) scale to the wavenumber (cm -1 ) scale, the Y-axis intensity needs to be corrected by the square of the wavelength (termed the l 2 correction), because fluorescence spectra (but not absorption spectra) are collected with a fixed wavelength bandpass. The l 2 correction causes the peak intensity of the short-wavelength (long-wavenumber) side of the fluorescence spectrum to decrease, or the peak intensity of the long-wavelength (short-wavenumber) side to increase. This issue has been known for >60 years but the impact remains little appreciated. The relatively new availability of libraries of spectral data enabled assessment here of the l 2 correction for various donor-acceptor pairs (i.e., donor fluorescence spectra and acceptor absorption spectra). The magnitude of error introduced upon omitting the l 2 correction increases with the width of the spectra. A meta-analysis of recent literature reveals trends in usage of wavenumber or wavelength scales. While either scale can be used, errors in calculating the ostensibly simple J term are not uncommon. Best methods are articulated here. The software program PhotochemCAD 3, which incorporates the l 2 correction for FRET calculations and also contains diverse spectral libraries, is freely downloadable at www.photochemcad.com.