This study was undertaken to develop techniques for measuring absolute rates of sterol synthesis in extrahepatic tissues in vitro and to estimate the magnitude of the errors inherent in the use of various 14C-labeled substrates for such measurements. Initial studies showed that significant errors were introduced when rates of synthesis were estimated using [3H]water since about 20 nmol of water were bound to each mg of tissue cholesterol isolated as the digitonide. This source of error could be eliminated by subtracting apparent incorporation rates obtained at 0 degrees C from those obtained at 37 degrees C or by regenerating and drying the free sterol. In a second set of experiments, the H/C incorporation ratio in cholesterol was determined in the liver by measuring the absolute rates of hydrogen and acetyl CoA flux into sterols. The ratio of 0.69 +/- 0.03 was found to be independent of the rate of hepatic cholesterol synthesis, the rate of hepatic acetyl CoA generation, or the source of the acetyl CoA. In a third set of studies, rates of incorporation of [3H]water or 14C-labeled acetate, octanoate, and glucose into digitonin-precipitable sterols were simultaneously measured in nine different extrahepatic tissues. Assuming that the H/C ratio measured in the liver also applied to these tissues, the [3H]water incorporation rates were multipled by the reciprocal of the H/C ratio to give the absolute rates of sterol synthesis in each tissue. When these were compared to the incorporation rates determined with the 14C-labeled substrates the magnitude of the errors in the rates of sterol synthesis obtained with these substrates in each tissue could be assessed. Only [14C]octanoate gave synthesis rates approaching 100% of those obtained with [3H]water and this occurred only in the intestine and kidney; in the other extrahepatic tissues this substrate gave rates of 6--66+ of the absolute rates. Rates of [14C]acetate incorporation in sterols varied from 4 to 62% of the [3H]water incorporation rates while those obtained with [14C]glucose were only 2--88% of the true rates. These studies document the large and highly variable errors inherent in estimating rates of sterol synthesis in extrahepatic tissues using 14C-labeled substrates under in vitro conditions.