In this paper experimental data obtained in evaluating the performance of an angle diversity system are presented and discussed. Data were collected at 8 Gc over a 170-miie tropospheric scatter system using large antennas equipped with multiple feeds. Tests were conducted during the period June, 1962, through February, 1963. The following are discussed. 1) The relative performance of a dual angle diversity system compared to a dual space diversity system. 2) The relative performance of the experimental system, operating in

-fold angle diversity, compared to a theoretical

-fold diversity system. 3) Variations in the method of obtaining angle diversity in terms of operating one or more transmitters at a single frequency, or multiple transmitters (and receivers) at different, but closely spaced frequencies. 4) The general characteristics of tropospheric scatter propagation at 8 Gc. The analysis of the experimental results shows that, with

-fold angle diversity operation, diversity improvement comparable to that obtained by the use of

-space diversity channels (which would require

antennas) is possible. It is shown that successful operation is achieved with a single, narrow, transmitting beam illuminating the troposphere, and a group of equally narrow receiving beams, all operating at the same frequency, receiving energy from this region of the troposphere. Significantly, the power loss in those receiving beams that are not directly intercepting the narrow transmitter beam is small. This is attributed to "beam broadening." In addition to the measured performance, results of some theoretical studies showing expected performance under angle diversity conditions are included.