Standardized local measures for preparing oral rehydration solution (ORS) in Nigeria were re-evaluated under laboratory conditions. Our results confirm those of the standardization team in respect of granulated and cube sugar. However, our mean weight of one salt measure (2.8155 +/- 0.292 g) is about 20% greater than their value. Consequently, correct use of the measures in our study gave solutions of 211-297 mmol-1 total concentration and 60-80 mmol-1, Na+ as against their values of 173-251 mmol 1-1 and 45-70 mmol-1, respectively. This discrepancy is most likely due to differences in salt type. Analysis of home-made solutions prepared by 40 illiterate mothers showed that 60% of them made accurately composed solutions. All the rest made hypertonic solutions. Salt type, spoon size and levelling technique are all possible causes of their error. The tendency to err only on the side of greater rather than lower salt concentration may be culture based or simply due to natural maternal instinct. To combat this trend, health education programmes in Nigeria should emphasize the danger in feeding a hypernatremic solution to a dehydrated child.