Since Skeggs' classic work (1) it has been generally assumed that air segmentation of the flowing stream and attainment of a "steady state" signal are essential parts of continuous flow analysis (CFA). Recently, RGzicka and Hansen (2) in Copenhagen and Stewart et al. (3)(4)(5) in Washington, D.C., have independently performed similar experiments of injecting the sample directly into the carrier stream, and