ar X iv : h ep - p h / 99 02 22 5 v 1 2 F eb 1 99 9 TTP 99 - 04 Spin physics in deep inelastic scattering : Summary

The problem of our understanding of the spin structure of the nucleon has been with us since the publication of the EMC measurements of the polarised structure function of the proton in 1987. In this talk a review of the results presented in Working Group 6 at this workshop is given. In the simple quark model the spin of the proton is carried by its three valence quarks so that ∆Σ = ∆u + ∆d = 1. Here ∆q = 1 0 dx (q ↑ (x) − q ↓ (x)) where q ↑(↓) (x) are distributions for quarks with spin aligned (anti-aligned) to the proton spin and q = u, d, etc. indicates the quark flavours. The simple quark model has however proven to be inadequate long before precise measurements of the proton spin structure became available, since it predicts that the ratio of the axial vector to vector coupling constants in neutron β decay is g A = 5/3 compared to the measured value of 1.26. The parton model ascribes part of the proton spin to sea-quarks and gluons. All partons in the proton can moreover possess orbital angular momentum, which also contributes to the proton spin. Within this model, the proton spin can no longer be identified with the sum of the quark spins only, and ∆Σ can therefore not be predicted without making additional assumptions. The best-known theoretical prediction of ∆Σ is due to Ellis and Jaffe [1]. Using SU(3)-flavour symmetry with the additional assumption of vanishing of the contribution from strange quarks to the proton spin, they obtain ∆Σ ≈ 0.58. Deep inelastic scattering (DIS) with polarised charged leptons on polarised targets allows the quark distributions q ↑(↓) to be investigated. These are extracted from the structure function g 1 (x, Q 2) measured in polarised DIS using the parton model relation g 1 (x, Q 2) = 1 2 q e 2 q (q ↑ (x) − q ↓ (x)). In the early 1980s the SLAC experiments E80 and E130 [2] reported the first measurements of polarised DIS for x > 0.1. In 1988, the EMC reported measurements [3] over a range down to x = 0.015. For x > 0.1 all the data (extrapolated to x = 0 for the determination of ∆Σ) seemed to confirm the expectations of Ellis and Jaffe. However, as x decreased the EMC data fell progressively below the expectations …