A field study was carried out at Htaukkyant village in Burma to assess whether village mothers could use condensed milk tins to measure one litre of water with reasonable accuracy for the preparation of oral rehydration solutions (ORS) and whether the extent of bacterial contamination of well water was serious and if this affected the bicarbonate content of the ORS solution. Empty condensed milk tins have a fairly uniform volume around 330 ml and using three condensed milk tins full of water mothers made up one litre quite consistently. Mothers also proved capable of preparing ORS solution by dissolving one packet of oral rehydration salt (ORS) in three condensed milk tins full of water to obtain solutions which contained acceptable and safe concentrations of sodium and potassium. Contamination of well water with faecal coliforms was present. Both storing water in domestic vessels and boiling water reduced the coliform count. Storing could be a good way of reducing the risk of infection if repeated contamination from dipping in to the water could be avoided. The counts on coliforms and faecal coliforms in ORS both increased by about 1 log per day over the first and second 24 hours after the preparation with contaminated well water. Despite this the bicarbonate content of ORS remained stable. In the absence of boiled water, ORS solution can be made using the cleanest available water and using it within 24 hours.