Soil pH is one of the most important factors affecting the in situ preservation of archaeological artefacts of e.g. bone, shells and iron. It is therefore important to have good, reliable methods for determining the pH. However, in all methods that involve sampling of soil or water, loss of gaseous species such as carbon dioxide from the samples as well as oxidation of reduced species may lead to results that do not reflect the in situ pH value. In this paper a method for taking pH measurements directly in the soil that avoids problems with degassing and oxidation is described. The method involves the use of a solid state pH electrode (ISFET) and a specially designed steel probe for introducing the electrode into the soil. The method has been developed and used at a waterlogged site, Nydam, Denmark. The same electrode has also been used to obtain detailed pH profiles during an archaeological excavation at Amosen, Denmark. Here the major advantage is a high spatial resolution and instant results during the excavation. The method is compared to standard methods for measuring pH in soil, and it is demonstrated that such standard methods in extreme cases can give differences of up to 2 pH units compared to the in situ pH. The implications of the measured pH values for the archaeological remains at the two sites are discussed briefly.