1. 1. When protein precipitates were incinerated with NaOH or Na 2CO 3 in crucibles or wide-mouthed test tubes, losses of iodine, in some cases exceeding 50% of 131I added to the precipitates, occurred from volatilization. The loss occurred during the decomposition of the organic matter while the temperature was rising from 400 to 600. Continuation of the heating at 600° did not increase the loss, nor did any loss occur when inorganic 131I was heated with alkali to 600 in the absence of organic matter. 2. 2. The loss of iodine during ineineration could be kept within 2 or 3% by using KOH instead of Na 2CO 3 or NaOH as alkali, and by using incineration vessels of a shape that minimized circulation of gases. Such vessels are tlie 15 by 125 mm test tubes used by BARKER 3, or 50 ml test tubes witli necks narrowed to 8 mm bore. 3. 3. A second loss by volatilization of iodine can occur if tlie dry ash is dissolved in excess acid. The loss appears to occur with the stormy evolution of CO 2. It is avoided if the ash is first extracted with water (vilkki 7) and the acid, necessary for subsequent reaction of Ce(IV) and As(III), is added to the solution instead of to the dry ash. 4. 4. Carbon particles left in the ash by incomplete incineration are difficult to remove from acid solution of the ash by centrifugation. They exert a marked accelerating effect on the reduction of Ce(IV) by As(III), and can cause gross plus errors in estimation of iodine from the rate of reduction. Occurrence of carbon in the ash can be prevented bv marlow's 13 device of adding kclo 3 to the alkali incineration mixture, or by opening the muffle oven for short intervals during incineration, after temperature has reached 600°, to permit access of air to the ash. 5. 5. VILLKI'S 7 demonstration, that inorganic iodine can be separated from serum proteins as completely by 12-hour dialysis as by precipitation and washing of the proteins, is confirmed. However, in the writers' experience, precipitation and washing proved to be more rapid and convenient for routine analyses. 6. 6. A procedure is described for determining the presence of iodine contaminating the reagents. It depends on the fact that a slight amount of mercuric ion inhibits the accelerating effect of iodine on tlie reduction of Ce(IV) by As(III). 7. 7. For determining iodine in the ash extract by means of the catalytic effect of I − on the reduction of Ce(IV) by As(III), conditions of reaction time and of cerie, arsenite, and chloride concentrations have been ascertained such that extinction readings for minimal to maximal iodine values cover the extinction range, 0.70 to 0.15, where percentage error of reading is minimal. 8. 8. A procedure for PBI determination is outlined which is believed to minimize the danger of aberrant results.