In the method of Rosalki and Foo (Clin Chem 1984;30:1182-6) bone and liver isoenzymes of alkaline phosphatase (EC 3.1.3.1) are quantified by using wheat-germ agglutinin (WGA). I suggest standardizing the procedure by using a WGA concentration that precipitates half of the alkaline phosphatase activity of serum pooled from an equal number of healthy women and men. By applying knowledge of the precipitation pattern in serum samples containing predominantly or exclusively bone or liver sources of alkaline phosphatase, I obtained results for the isoenzymes in healthy subjects that agreed with those by the heat-inactivation methods, as reported earlier in the literature. I then assessed the utility of the standardized procedure in a clinical study of prevention of postmenopausal bone loss. In patients receiving hormone replacement therapy, which is known to decrease bone turnover, the decrease in total alkaline phosphatase activity in serum was entirely ascribable to decreases in the bone isoenzyme activity, probably reflecting reduced bone formation, whereas the activity concentration of liver alkaline phosphatase remained unchanged.