New measurements y ielded the value 26428 .4± 2.6 Inte rnational joules per gram mass for thc heat of combustion, QB, of benzoic acid at 25° C under the conditions of the standard bomb process. The difference, 0.036 percent, between the above value and that reported in 1934 is due to five factors : (1) An error in the previous value, resulting from the effect of dissolved carbon dioxide on the determin ation of the nitric acid formed in t he bomb, (2) a change in the value used for the energy of formation of nitric acid in the bomb, (3) taking account in the present work (in the calculation of the value of QB at 25° from the observed value at 30°) of the temperature dependence of the Washburn reduction, (4) a change in the value used for the tempe rature coefficient of the heat of combustion, arising from the use of a new value tor the specific heat of benzoic acid, and (5) a smaJl difference in the results of the calorimetric measurements. The 1934 resul ts, when corrected for tile first four of the above effects, are in agreement within 0.01 percent with the results of the present measurement~. The procedure for correcting the results of calibration experiments originally calculated on the ba~is of the 1934 value of OB i5 described.With one exception, values of OB at 25° C calculated from the results of previous