These measurements of the ratios of the specific heats of nitrogen, a t room and a t lower tem peratures, have been made, and the results have been calculated, in exactly the same way as those obtained when air and hydrogen were used. The latter were described in a former paper, (1) in which will be found a description of the apparatus and an account of the method of dealing with the experimental results. In the work which follows the same apparatus was employed, the only changes being th a t a new platinum wire (from the same reel) was used on the thermometer, and, owing to breakage, a rather finer quartz suspension had been p u t on the galvanometer. However, as the apparatus had to be dismantled for transference to another laboratory I thought it desirable to confirm some of the effects and relationships previously noted ; therefore, a full set of experiments with the four expansion vessels was made a t room tem perature. I t may be recalled th a t the method has been especially worked out, using small expansion vessels in order th a t values of the ratios of the specific heats can be readily obtained a t different tem peratures. Up to the present time work has been restricted to determinations a t temperatures below th a t of the laboratory. Several vessels are used in order to allow a correction to be made for a systematic change in the apparent value of the ratio, which depends on the size of the vessel employed.