This paper is concerned with the use of an iterative technique in the identification of non-linear systems which comprise both dynamic linear elements and static nonlinear elements. In particular, it has been applied to the case in which a linear element is embedded between two non-linear elements, this being a case in which conventional methods are not usually applicable. It is assumed that test data are available for the application of an input pulse of fixed width and variable amplitude and the requirement is to identify the individual elements without reference to signals at the internal nodes. Successful results are reported for a range of simulated systems. The effect of noise is discussed and its effect measured in simulations.