## Photoproduction of Dijets with High Transverse Momenta at HERA |

Measurements of the parton densities of the photon and proton have been performed in several processes and experiments. The quark densities in the photon have been determined in

To test predictions of perturbative QCD calculations and PDF parametrisations the paper investigates the photoproduction of dijets using the H1 detector at HERA. The transverse momentum E

Compared to the
previous H1 publication, this paper presents new measurements with increased statistical
precision and an improved understanding of systematic uncertainties.
The cross sections for dijet production are measured as a function of the
transverse energy of leading jet (E_{t,max}), the momentum fractions
of the photon (x_{gamma}) and the proton (x_{p})
participating in the hard interaction, and the angle of the dijets in
their centre-of-mass system- cos(theta*).

In order to test the theory as closely as possible, measurements are presented separately for events originating from predominantly*direct photon* (x_{gamma}>0.8) or *resolved photon*
(x_{gamma}<0.8) processes.
The region of x_{gamma}>0.8, in which the photon predominantly
interacts directly with the proton, is particularly well suited to test
the proton structure as the photon structure plays no significant role there.
The data in the region of x_{gamma}<0.8, where the photon behaves like a
hadronic object, may also provide additional constrains on the photon parton
density function.

In addition the cross sections are investigated at different jet topologies: the case where both jets are in the "backward" direction, where both jets are in the "forward" direction, and where on jet is in the "forward" and one in the "backward" direction.

The data compare well with predictions from Monte Carlo event generators based on leading order (LO) QCD and parton showers and with next-to-leading order (NLO) QCD calculations, except for large x_{p} where there is large
uncertainty from the proton parton density function. This uncertainty can be
largely reduced using this photoproduction dijet measurement in the global
QCD fit analysis.

In order to test the theory as closely as possible, measurements are presented separately for events originating from predominantly

In addition the cross sections are investigated at different jet topologies: the case where both jets are in the "backward" direction, where both jets are in the "forward" direction, and where on jet is in the "forward" and one in the "backward" direction.

The data compare well with predictions from Monte Carlo event generators based on leading order (LO) QCD and parton showers and with next-to-leading order (NLO) QCD calculations, except for large x