Civil aircraft engines are required to operate reliably in a variety of inclement weather situations. Ingestion of airborne water has been found to cause problems for certain engine types, particularly during aircraft descent at low engine power settings. The effects of water ingestion under these engine operating conditions are not well understood. It is believed that the engine pe1formance is largely dete1mined by the behaviour of liquid water in the core compression system. A programme of experimental and theoretical research has therefore been undertaken to investigate the effects of water ingestion on an axial flow compressor operating at low speed, in isolation from other engine components. Experimental studies have been conducted using a low-speed, four-stage axial flow laboratory compressor, modified to facilitate injection of water through spray nozzles fitted at the inlet. In the process, new methods were developed for the reliable measurement of static pressure in a wet environment. Compressor pe1formance changes were quantified by measming the overall total-to-static pressure-rise characteristic and the torque requirement. Initiation of water ingestion was shown to move the compressor operating point on the characteristic to a new position at reduced pressure rise and reduced air mass flowrate, and to increase significantly the torque requirement. In some cases, it was shown that initiation of water ingestion would lead to rotating stall. Experiments were conducted to investigate the effects of water-to-air mass flowrate ratio, droplet size and spray coverage. The spray coverage is the fraction of the compressor inlet area over which the water is distributed. This was determined principally by the divergence (or spray) angle of the water emerging from each nozzle. The results suggest that droplet size is relatively unimportant in determining the performance with water ingestion. The spray coverage is, however, important, with small spray coverage resulting in little or no loss of compressor surge margin and large spray coverage resulting in significant loss. The pe1formance of the first compressor stage is c1itical in explaining these differences in behaviour. Theoretical studies were conducted with the aim of predicting the spatial disttibution of liquid water inside a compressor. New models were developed for key aspects of liquid water behaviour and incorporated into a computer program. Droplet velocity (or momentum) is often found to play an important role in determining the water behaviour. Theories are suggested to explain the experimentally observed effects of water ingestion, drawing on insight gained from both the expe1imental and theoretical results. Compressor modifications that may reduce the adverse effects of water ingestion are then suggested. These comprise, firstly, the removal of water from the casing immediately upstream of the compressor and, secondly, the application of a "non-stick" coating to the blades of the first rotor blade row.