FIRST STUDY OF RAPIDITY GAPS IN ese-ANNIHILATION * The SLD Collaboration

We present the first study of rapidity gaps in e+eannihilations, using 2’ decays collected by the SLD experiment at SLAC. Our measured rapidity gap spectra fall exponentially with . increasing gap size over five decades, and we observe no anomalous class of events containing large gaps. This supports the interpretation of the large-gap events measured in pH and ep collisions in terms of exchange of color-singlet objects. The presence of heavy flavors or additional jets does not affect these conclusions. Since the initial observation of hadronic jets rapidity has been used to characterize the momentum of particles in jets in a frame-invariant manner. The rapidity distribution has been studied in e+eannihilation, ep and hadron-hadron collisions, and fixed-target experiments, and is a characteristic of strong interactions that is well described by perturbative QCD combined with iterative models of jet fragmentation [l]. Hard scattering of quarks or gluons can be modeled by color fields between the outgoing partons that fragment into the observed final-state particles, typically populating the whole rapidity range. Recently, exchange of color-singlet objects in hard diffractive hadron-hadron scattering processes characterized by events containing large gaps in the particle rapidity spectrum has been discussed [2]. Subsequent studies of pp collisions at the Fermilab Tevatron collider found that roughly 1% of events comprising at least two hightransverse-energy (ET) jets contain a large rapidity region between the two highest-& jets with no particle activity [3, 4, 51. Th ese events have been interpreted in terms of color-singlet exchange between the interacting partons. Exchange of electroweak bosons is estimated to contribute only a small fraction of the observed rate of gap events and a model incorporating pomeron exchange is in agreement with the data [4, 51. Large rapidity gaps have also been observed in roughly 10% of all photoproduced dijet events in deep-inelastic scattering at the HERA ep collider [6, 71 and have also been interpreted in terms of color-singlet exchange. Models involving either vector meson dominance of the exchanged virtual photon or hard diffractive scattering via pomerons describe the data [6, 71. Color exchange processes account successfully for the properties of the majority of dijet events and may give rise to large rapidity gaps due to random fluctuations. In both the ep and pp cases the interpretation of rapidity-gap events in terms of color-singlet exchange depends upon an understanding of this color-exchange background. In the pp experiments this was estimated by extrapolation of fits to the particle multiplicity distribution in rapidity intervals into the zero-particle, or rapidity gap, region [4,5]. In