A theoretical investigation of the pulse technique has been carried out using typical reaction models (irreversible, reversible, and consecutive reactions) and the conversion of reactants and the yield of products were compared with those from the continuous flow technique. In the case of linear reactions with no other rate process than first order, even if they are reversible or consecutive, the results from the pulse technique are in agreement with those from the flow technique, while in the case of nonlinear reactions there is a quite remarkable disagreement. In the reversible reactions of A ⇌ 2R and A ⇌ R + S the conversion is above that from the flow technique, and in the consecutive reactions such as A + B → R, R + B → S, and A → R, 2R → S, the yield of the intermediate product, R, is larger than that from the flow technique. The results on the nth order irreversible reaction are equivalent to that from the flow technique in the case of a rectangular pulse only. Main factors peculiar to the pulse technique are discussed. The results from the pulse technique vary greatly with the inlet pulse width as well as with the different adsorption strengths of components on a catalyst, but the difference in the results is relatively small for rectangular, triangular, and error function pulses. Furthermore, a method of measurement of kinetic parameters by the pulse technique is proposed for the several reaction models. © 1968.