Experimental investigation of a speech enhancement algorithm is presented. The algorithm is designed for a spherical microphone array, and formulated in the spherical harmonics domain, based on a plane-wave model that is also formulated in the spherical harmonics domain. The proposed algorithm facilitates dereverberation by employing DOA estimation, rather that room impulse response identification. The paper presents the algorithm formulation and results from real room measurements, also compared with traditional beamforming methods.