A colourimetric method for screening buffer capacity of mixed paraffin stimulated saliva in general practice is described. Mixed paraffin stimulated saliva (1.00 ml) is pipetted into a standard tube belonging to a commercial "saliva kit" (Dentobuff, Orion Diagnostics Ltd). The tube is sealed with a plastic stopper and vigorously shaken for 10 seconds. The stopper is then removed and carbon dioxide is allowed to evaporate for 2--5 minutes. The colour is compared with a scale graded from pH 3 to pH 6. The method was evaluated using 136 dental patients and dental students, and a number of dentists and nurses, and the results were compared with the generally accepted clinical method developed by Ericsson (1959). Several different nurses, dentists and dental students took part in registering the results. In four cases the results differed by more than +/- 2.0 pH units between the two methods and in another 18 by more than +/- 1.0 units. Of the 80 tests performed on no occasion was a person with a "low" salivary buffer capacity evaluated as "normal" or "good". There was no obvious systematic deviation between the two methods, provided carbon dioxide was allowed to evaporate for 2--5 minutes before reading the colour. A thorough examination revealed that at a final pH over 5.5 the method gives lower values than Ericsson's method because of retained carbonic acid, a fact which was considered to be of minor importance, clinically. The method was found to be easy to handle and rapid and the accuracy was considered adequate. It is advisable, however, to repeat the test once or twice because of biological variation and in doubtful cases if "intermediate" results are repeatedly found to send the patient to a clinic with facilities for a complete test.