Abstract
Evidence is presented concerning errors in the direct determination of human serum cholesterol with a ferric-chloride color reagent. It is shown that in physiologic concentrations the vitamins A, D, E, K, and niotinic acid cause no error, but that hemoglobin, anticoagulants, and impurities in acetic acid can cause errors. The control of these errors is discussed. Errors caused by steroids, added salts, and high bilirubin content of serum are discussed.