The pH of fruit juices and carbonated drinks are known to be low and have, therefore, been implicated in the increasing incidence of erosion. The ability of a drink to resist pH changes brought about by salivary buffering may play an important part in the erosion process. The aims of this study were to measure the initial pH of several widely available soft drinks and determine their buffering capacities. As part of a larger study, the following groups of drinks were tested: pure fruit juices, non-fruit-based carbonated drinks, fruit-based carbonated drinks, flavoured spring waters, and plain carbonated mineral waters as positive controls with still water as the negative control. The measurement of pH was carried out using a pH electrode connected to an Orion EA940 Ionanalyser. One hundred millilitres of each drink was then titrated with 1 m sodium hydroxide, added in 0·5 mL increments, until the pH reached 10. Each titration was repeated three times. The average initial pH was lowest for the non-fruit-based drinks (2·81±0·274) and highest for plain mineral water (7·4±0·1002). The buffering capacities can be ordered as follows: fruit juices>fruit-based carbonated drinks and flavoured mineral waters>non-fruit-based carbonated drinks>sparkling mineral waters>still mineral water. It is concluded that fruit juices and fruit-based carbonated beverages, with their increased buffering capacities, may induce a prolonged drop in oral pH.