Determinations of total ascorbic acid (AA) might be expected to be less affected by interference of non-AA components than direct determinations of dehydroascorbic acid (DHAA) simply because the concentration of total AA in vegetables is generally considerably greater than that of DHAA. On the other hand, they might be influenced by the action of various oxidizing agents on non-AA components if the absorption characteristics of the latter were altered.In the present study, the specificity of the 2,4-dinitrophenylhydrazine ( D N P H ) method for determining total AA in vegetables was investigated by procedures already described (3). Thus, D N P H derivatives were purified by chromatographic techniques, concentrations of total AA calculated from the fraction most closely resembling a reference coinpound prepared froin crystalline AA, and these concentrations compared with those found by the usual D N P H methods.
METHODThe D N P H method and chromatographic analysis of DN PH derivatives were carried out as described (3), except that aliquots of the same vegetable extracts used for determining D H A A were treated with indophenol ( I ) , charcoal (7.9), bromine (8,9), or hydrogen sulfide followed by bromine (8) prior to adding the D N P H reagent. In addition, the concentration of thiourea in the D N P H reagent was reduced to 0.25%, the concentration suggested by Lowry, Lopez, and Bessey ( 5 j , since preliminary studies with cabbage indicated that the same concentration of total AA was found in the presence of 1.0 and 0.25% thiourea.Standard curves were prepared from variously oxidized known solutions and were found to be essentially equivalent for the range of D H A A from 0.0015 to 0.0140 mg. per nil. Consequently, a single curve, drawn from the combined data, was used as the basis for calculating concentrations of D H A A in unknown solutions. Data for concentrations of total AA and percentage retentions were subjected to analyses ot variance, using a split plot design d i t h preliminary treatments as main effects :tnd methods of analysis as sub-effects. Duncan's multiple range test ( Z j was used to determine the significance of differences among means when significant values of F were found.