Hole fragmentation in deep-inelastic processes

For deep inelastic eN and P N processes we argue that there are three fragmentation regions instead of two when w is very large. (Comments and Addenda to Phys. Rev.) Work supported by the U. S. Atomic Energy Commission. I' Applications of the parton model to the problem of hadron final states in deep inelastic electroproduction have led to the concept of parton fragmentation 1,2,3,4,5 and to the proposal that the ideas of short-range correlation in rapidity,. successfully used in ordinary hadron-hadron collisions, 6 can be taken over directly to the deep inelastic processes. Hence one again envisages three regions in longitudinal phase space (which we describe in terms of the rapidity variable y): a target fragmentation region (of length- 2), a photon (or W) fragmentation region (of length-log Q2), and, if w>> 1, a central region (of length- log w). This follows from rather general considerations. 7 In the parton-fragmentation picture, one concentrates on the region of leading hadrons whose longitudinal fraction 2 = plab /v is, sayLO. (i.e., ymax- y<, 2). In that region one argues that production of hadrons of type i is proportional to the probability fj(w) of the incident current (virtual y or W) of having struck a parton of type j and to the probability Gji(z) that the parton "fragments " into hadron i plus anything:- = dNi dY