In his paper, PEARCE presents a method for determining the stationary probabilities of a G/Mt/1 system, the service mechanism being such that it operates regardless of whether or not customers are present. This note is an attempt to correct his results. [ABSTRACT FROM AUTHOR]