Easily constructed accessory equipment for obtaining quantitative flame spectra has been designed, and is discussed. The causes of the variations in the results obtained by the flame spectrum method have been investigated and analyzed. Variations due to manipulation seem to be of little significance except perhaps as regards control of the acetylene pressure. Plate variations are significant and are the chief limiting factor in the accuracy of the method. A comparison of chemical and spectrographic analyses shows that the differences between the results obtained by these two methods are of the same order of magnitude as the differences obtained from plate to plate when the spectrographic method alone is used. Typical analytical data obtained by chemical methods and by the spectrographic method as well are presented.