Practical ways to achieve fluorescence rejection were developed by using the technique of Raman difference spectroscopy and a gated diode-array detector. In order to improve the signal-to-noise ratio, we had to circumvent the problem presented by the fact that the sensitivities of each element of the diode-array detector are not always uniform and linear—a problem which results in an apparent noise in the observed spectra. Therefore, in this study we tried to separately observe a Raman-plus-fluorescence spectrum ( ST) and a pure fluorescence spectrum ( SF), with a high S/N ratio, and to obtain a Raman spectrum ( SR) as their difference. With a cw Ar+ laser, ST and SF were excited at different wavelengths but recorded in the same wavelength region. This procedure was easy and quite successful for compounds, such as methyl orange and methylene blue, that have a Raman-to-fluorescence intensity ratio ( IR/ IF) of greater than 0.01. With a mode-locked and Q-switched Nd:YAG laser, ST and SF were obtained from different timings of the detector gate (6 ns), i.e., −6 to 0 ns for ST and 3 to 9 ns for SF. The temporal discrimination between Raman and fluorescence photons combined with the Raman difference technique significantly improved the 532-nm excited Raman spectrum of the tetraphenylporphyrin solution. This method is applicable to compounds with an IK/ IF of smaller than 0.01. The time-correlated single photon counting method with a time-to-amplitude converter was improved, to enable it to work with a high repetition rate, and was applied to a Rhodamine 6G solution, which was also examined by the gated Raman difference technique. Advantages and problems associated with each technique and appropriate methods for a given sample will be discussed.