Measurements are presented of differential dijet cross sections in diffractive photoproduction (Q 2 < 0.01 GeV 2 ) based on 1999 and 2000 HERA data with integrated luminosity of 54 pb −1 . The event topology is given by ep → eXY , where the system X, containing at least two jets, is separated from a leading low-mass proton dissociative system Y by a large rapidity gap. The measurements are made in two kinematic ranges differing primarily in the transverse energy requirements on the two hardest jets. The dijet cross sections are compared with next-to-leading order QCD predictions based on recent diffractive parton densities obtained by H1. The next-to-leading order calculations predict larger cross sections than measured with the data. The cross section suppression of the data relative to the calculation is found to have no significant dependence on the photon four-momentum fraction entering the hard subprocess. There is a suggestion of a dependence of the suppression factor on the transverse energy of the jets.In photoproduction two basic leading order (LO) classes of photon interactions can be distinguished -direct and resolved. In the direct processes the photon interacts as a pointlike particle. In the resolved processes the photon can develop its partonic structure and acts as a composite particle. In diffractive ep scattering the proton stays intact or dissociates into a low-mass (M Y E cm ) state. The hard diffractive processes can be considered as an exchange of a vacuum quantum number object (Pomeron, IP ). A large gap in the rapidity distribution of the final state hadrons is observed. Owing to the photon structure there is an apparent resemblance between the resolved photoproduction and the hadron-hadron scattering. The factorization of the QCD calculable subprocess from the diffractive parton distribution functions (DPDF) in proton is not expected to hold in the hadron-hadron collisions [2]. The additional rescattering of spectators can fill the large rapidity gap and, therefore, to spoil the experimental signature of the diffractive event. Such a mechanism is expected to explain the difference between the measured structure function extracted from dijet production rates in pp collisions at Tevatron [3] and the theoretical predictions based on the diffractive parton densities by H1. Previous results concerning two jet production in diffractive photoproduction presented by the H1 and ZEUS collaborations can be found in [4] and [5], respectively. In [4] an overall suppression factor of 0.5 is applied to the NLO theory prediction in order to reproduce the measured cross section. In [5] a similar analysis is presented with somewhat higher E T range required on the jets. The global suppression factor of the NLO theory prediction is measured of 0.9 with the use of the same diffractive parton densities as in the case of [4]. The present measurements are based on luminosity of 54 pb −1 which is about a factor of three larger than previous H1 measurement in a similar kinematic domain [4]. * On behalf of the H1...