PIP: Oral rehydration solutions (ORS) with low and high sodium or potassium contents were compared in 84 infants aged 3-15 months with dehydration caused by acute watery diarrhea. In randomized, double-blind trials, the ORS contained (in mmol/liter): glucose, 110; HCO3, 30; K, 10; Na, 60 (low) or 90 (high), or, in a second trial, the high Na formula plus 20 vs. 35 mEq/liter of K and extra oral water, given in the ratio of 2 bottles (240 ml) of ORS followed by 1 bottle of plain water (the 2:1 regimen). The balance technique was used to measure net absorption. Of the 84 infants, only 1 required intravenous therapy. During the first 6 hours of therapy, mean sodium absorption was significantly lower in the low Na group, and hyponatremia was more common. A few high Na group infants receiving no extra oral water had transient mild asymptomatic hypernatremia. Mean K absorption from 0-6 hours was significantly higher in the high K group, and they had no hypokalemia after oral rehydration; in contrast, 33% of the low K infants had hypokalemia. High Na ORS in the 2:1 regimen caused no electrolyte abnormalites, yielded better sodium absorption, and was usable in all age groups and for all diarrheas. High K ORS corrected K deficits better than the low K ORS. An increase in potassium concentration above that in the currently recommended ORS formula should be considered.