A series of jar tests and a full-scale assessment of iron (II) chloride dosing at Abingdon sewacre treatment works were carried out to identify factors affecting phosphorus removal. Variables considered were pH, redox potential and dissolved oxygen concentration using crude sewage. As expected, there was a strong link between phosphorus removal and iron:phosphate ratio but this trend was not experienced with suspended solids removal. The main factor affecting the efficiency of the phosphorus removal was pH, followed by dissolved oxygen (DO) and redox potential. This was assessed using a seven variable factorial design experiment. The conversion of iron (II) to iron (III) averaced close to 68.7% at optimal conditions which were: high DO (1.0-5.7 mg l(-1)), mid range redox potential (57-91 mV) and high pH (7.5-8.0). As before, the most important variables were pH and DO concentration. Full-scale observations appeared to support these findings as iron conversion and phosphorus removal was good (85.6%) despite the very low redox potential (- 13 9 mV). This could be due to the high pH at point of dose (8.2) and the relatively high DO (4.6 mg l(-1)). In poor conditions, where less iron (II) is converted to iron (III), chemical consumption costs will increase, solids removal may decrease and the system is likely to become less reliable.