## Abstract

The rate of inversion of sucrose by strong acids has been measured at 24$\cdot $7 degrees C by a method which eliminated the mutarotation lag and the end-point uncertainty. The rate was proved to be strictly first order for at least 54 hr. and the first-order rate constant was determined with an internal consistency better than 1%. In 0$\cdot $1 M hydrochloric acid the first-order rate constant k$_{1}$ is linear in S, the number of grams of sucrose in a litre, according to the formula k$_{1}\times $ 10$^{4}$/min.$^{-1}$ = 7$\cdot $13 + 0$\cdot $97 $\times $ 10$^{-2}$S. For sucrose at 30 g./l. hydrolysed by single strong acids at molarities up to 0$\cdot $2, the first-order rate constant k$_{1}$ is related to the molarity c of strong acids by the formula k$_{1}$/min.$^{-1}$ = 6$\cdot $95 $\times $ 10$^{-3}$c $\times $ 10$^{B_{j}c_{j}}$, where B$_{j}$ is a constant determined by the anion as follows: Cl$^{-}$, 0$\cdot $28; Br$^{-}$, 0$\cdot $35; ClO$_{4}^{-}$, 0$\cdot $38; NO$_{3}^{-}$, 0$\cdot $30. In similar experiments with mixed strong acids or a strong acid with the addition of a neutral uni-univalent salt, the experimental results can be expressed by the formula k$_{1}$/min.$^{-1}$ = 6$\cdot $95 $\times $ 10$^{-3}$c$_{\text{H}}$ $\times $ 10$^{\Sigma B_{j}c_{j}}$, where c$_{\text{H}}$ denotes the molarity of strong acids and c$_{j}$ the molarity of each anion. Each B$_{j}$ has the same value as in the formula for single acids. This formula is in accordance with Bronsted's principle of specific interaction. In experiments on solutions containing bi-univalent and tri-univalent salts there is a further small negative salt effect of the multivalent cations increasing with their valency. This effect indicates that the principle of specific interaction becomes detectably inaccurate at these higher ionic strengths. Our results are compared with those of other workers. On the whole there is good agreement within the experimental accuracy of the various data. Certain discrepancies in absolute values of the rates are probably attributable to uncertainties of 0$\cdot $1 degrees C or less in the temperature.