The revealed-preference estimates begin with estimates of the premium organic-food buyers pay. These are estimated using prices of organic and conventional foods recorded at several local stores. Except for some information on the size of certain items, data to control for other quality differences between samples, such as differences in color, consistency, taste, and freshness were unavailable. Without such data it is impossible to separate the estimated organic premium into components attributable to the presumed absence of pesticides and to differences in other qualities. Because these data provide virtually no information that can be used to control for differences that might affect the incremental taste function τi(qi), the function is assumed to be uniformly zero.