In four-jet events from e + e − →Z 0 →multihadrons one can separate the three principal contributions from the triple-gluon vertex, double gluon-bremsstrahlung and the secondary quark-antiquark production, using the shape of the two-dimensional angular distributions in the generalized Nachtmann-Reiter angle θ NR ∗ and the opening angle of the secondary jets. Thus one can identify directly the contribution from the triple-gluon vertex without comparison with a specific non-QCD model. Applying this new method to events taken with the DELPHI-detector we get for the ratio of the colour factor N c to the fermionic Casimir operator C F : N c C F = 2.55 ± 0.55 ( stat. ) ± 0.4 ( fragm. + models ) ± 0.2 ( error in bias ) in agreement with the value 2.25 expected in QCD from N c =3 and C F = 4 3 .
NC, CF, and TR are the color factors for SU(3) group.
The production rates for 2-, 3-, 4- and 5-jet hadronic final states have been measured with the DELPHI detector at the e + e − storage ring LEP at centre of mass energies around 91.5 GeV. Fully corrected data are compared to O(α 2 s ) QCD matrix element calculations and the QCD scale parameter Λ MS is determined for different parametrizations of the renormalization scale ω 2 . Including all uncertainties our result is α s ( M 2 Z )=0.114±0.003[stat.]±0.004[syst.]±0.012[theor.].
Corrected jet rates.
Second systematic error is theoretical.
Relative production rates of multijet hadronic final states of Z 0 boson decays, observed in e + e − annihilation around 91 GeV centre of mass energy, are presented. The data can be well described by analytic O( α s 2 ) QCD calculations and by QCD shower model calaculations with parameters as determined at lower energies. A first judgement of Λ MS and of the renormalization scale μ 2 in O( α s 2 ) QCD results in values similar to those obtained in the continuum of e + e − annihilations. Significant scaling violations are observed when the 3-jet fractions are compared to the corresponding results from smaller centre of mass energies. They can be interpreted as being entirely due tot the energy dependence of α s , as proposed by the nonabelian nature of QCD, The possibility of an energy independent coupling constant can be excluded with a significance of 5.7 standard deviations.
Data are corrected for final acceptance and resolution of the detector. No explicit corrections for hadronisation effects are applied.