Measuring muscle mass is an important component of the nutritional assessment examination and a suggested index of this body space is the 24-h urinary excretion of creatinine. The method originated from studies in a variety of animal species in whom early workers found a parallelism between total body creatine and urinary excretion of creatinine. Assuming that nearly all creatine was within muscle tissue, that muscle creatine content remained constant and that creatinine was excreted at a uniform rate, an obvious "corollary" was that urinary creatinine was proportional to muscle mass. The so-called "creatinine equivalence" (kg muscle mass/g urinary creatinine) ranged experimentally from 17 to 22. One of the limiting factors in firmly establishing this constant and its associated variability was (and is) the lack of another totally acceptable noninvasive technique of measuring muscle mass to which the creatinine method could (or would) be compared. An improved understanding of creatine metabolism and a variety of clinical studies in recent years has tended to support the general validity of this approach. However, specific conditions have also been established in which the method becomes either inaccurate or invalid. While creatinine excretion may serve as a useful approximation of muscle mass in carefully selected subjects, there remains a need for accurate and practical indices of muscle mass for use in the individuals in whom the method cannot be reliably applied.