We present a new type of flow analysis, based on a particle-pair correlation function, in which there is no need for an event-by-event determination of the reaction plane. Consequently, the need to correct for dispersion in an estimated reaction plane does not arise. Our method also offers the option to avoid any influence from particle misidentification. Using this method, streamer chamber data for collisions of Ar+ KCl and Ari-BaI, at 1.2 GeV/nucleon are compared with predictions of a nuclear transport model.Many intermediate-energy heavy ion experiments have been directed toward the goal of inferring properties of the nuclear equation of state (EOS) [I]. In parallel with this effort, theoretical work in the area of nuclear transport models has focused on the task of identifying the most appropriate experimental observables for probing the EOS and on the related task of establishing a quantitative connection between such observables and the EOS [2]. Many factors, both theoretical and experimental, have contributed to the current lack of a Consensus on Data [3,4] from the Diogene and Plastic Ball detectors Support this assumption for rapidities other than the midrapidity region where the "squeeze-out" [5] effect can result in a more complex distribution. In the present study, we restrict our analysis to forward rapidities (see below). The maximum azimuthal anisotropy, as defined by Welke et al. [ 6 ] , is even a relatively coarse characterization of the compresl + h R=-sional potential energy at maximum density (in other 1-h ' words, a characterization of the EOS as relatively "hard" or "soft"). One such factor, for example, arises from the fact that detector inefficiencies and distortions can be difficult to simulate and quantify (particularly in the case of a 4n-detector), and this leads to systematic uncertainties in measurements of collective flow. This paper presents a new form of collective flow analysis for two data sets from the Bevalac streamer chamber. The most noteworthy feature of this new method is that it is designed to minimize the type of systematic uncertainty mentioned above; more specifically, the influences of particle misidentification and dispersion of the reaction plane can be removed.For a nonzero impact parameter, the beam direction ( z ) and the line joining the Centers of the nuclei determine the reaction plane, i.e., the X -2 plane. The azimuthal angle of a fragment in this coordinate system is We assume that the distribution function of 4 in an interval of rapidity centered on y , can be described by an expression of the form The method proposed by Welke et al. [6] for determining R in an experiment involves estimating 4 in Eqs. (1) and (2) using the relation 4=+obs-+R, where +obs is the observed azimuth of a fragment, and +R is the estimated azimuth of the reaction plane as deter.mined from the observed fragments in the final state. This method requires that the resulting R be corrected upward, to allow for the fact that 4R is distributed about + = O with a finite dispersion. Eac...