During the start-up phase of an enhanced biological phosphorus removal (EBPR) plant, the amount of eliminated phosphorus during wastewater treatment and the subsequent release during anaerobic sludge digestion was investigated. Different approaches were used to determine the mechanisms of enhanced phosphorus removal. From a comparison of the EBPR plant with a control, a strong correlation between the potassium, the magnesium and the phosphorus content of the sludge and the results gained from phosphorus fractionations we conclude that the major part of the eliminated phosphorus was stored in form of polyphosphate. During digestion of excess and a mixture of excess and primary sludge a complete release of the stored polyphosphate was found. The release of phosphorus was accompanied by a release of potassium and magnesium ions, from which only potassium remains in soluble form. Therefore, the soluble potassium concentration seems to be a good measure for the amount of phosphate released. Only a part of the released phosphate remains in soluble form. When digesting excess and mixed sludge this accounts for approximately 40% of the total phosphorus brought into the digester. The difference between the measured soluble phosphate concentration and the amount of released phosphorus was fixed, mainly due to chemical precipitation. It was found that a fixation in the form of magnesium ammonium phosphate (struvite) was likely to occur under the conditions of anaerobic sludge digestion. The amount of phosphate precipitation as struvite could be estimated using theoretical calculations at approximately 20% of the total phosphorus in the digester. Calcium dosing experiments show that calcium-phosphate precipitation plays only a minor role in phosphate fixation.