Remarks on Tables IV. and V. Experiments 1 to 3 (3 cells) and 4 to 5 (2 cells) were performed with a view to testing the working of the various portions of the apparatus, and in order to ascertain the minimum quantity of liquid that could be used with safety. do1 The results were not satisfactory: the values of at the dt same temperature, when deduced from the different experiments in which the conditions were similar, differed considerably, occasionally by as much as 1 per cent. The cause was evident the depth of liquid was too small to give satisfactory results with this form of stirrer. Considerable difficulty was experienced in maintaining the potential balance, and it is possible that at times portions of the coils were uncovered. I therefore decided on increasing the depth of the liquid. Experiments 6 to 8.-The mass of aniline was now increased by about 60 grms., and the observations became more satisfactory. These experiments were performed with a stirring rate of 9-10 per second. I had not yet decided upon the best method of ascertaining the null point, and unfortunately did not perform any 4-cell experiments. When deducing the results, I assumed that the intersection of the 2- and 3-cell lines stood in the same relative position to the intersection of the 2- and 4-cell lines as was found in those cases where a 4-cell experiment had been performed. The null point is of course (see Plate I.) in a very different position from that found when the stirring-rate was 5·00 ; and the close agreement between the results given by these three experiments at the high rate, and experiments 13 to 15 at the same temperature but with a slow rate, is a satisfactory proof of the validity of this method of finding . N Experiments 8 to 39 call for no special comment. They were all conducted with a stirring-rate of (approximately) 5'00; Nos. 9 to 24 with the same mass of aniline as 6 to 8. From No. 25 upwards the mass of aniline was increased to 569-53 grms. On plotting the values of the capacity for heat of this mass of aniline and the calorimeter, the spot obtained from experiments 25 to 27 appeared in a somewhat abnormal position; consequently, although I had considered the experimental work as completed, I decided to repeat these experiments, and Nos. 40, 41, and 42 give the result. They agree so closely with the corresponding group as to confirm the position originally assigned to the curve at that temperature (about 16°.7 C.). The curves on Plate II. (a) give the capacity for heat of the different masses of aniline together with the calorimeter, at different temperatures. Both of the curves resulting from the larger masses show a marked change between 15° and 25° C. It is worthy of notice that the specific heat as deduced from these curves shows little or no signs of any similar change, but that it appears very markedly in the waterequivalent curve, and in such a manner as to indicate that it was approaching a minimum. Had I not been able to determine the specific heats by methods independent of the behaviour of the calorimeter, I should have concluded that the temperature coefficient of the specific heat of aniline altered considerably at the lower temperatures. This indicates the necessity for extreme caution in similar investigations. Table VI. gives a summary of the results, obtained from the values given by the curves in Plate II. (a). TABLE VI. I. gives the capacity for heat of calorimeter +358-20 grms. (in vacuo) of aniline. II. gives the capacity for heat of calorimeter +569-33 grms. (in racno) of aniline. III. is the difference between the numbers in Cols. I. & II., and is therefore the capacity for heat of 211:33 grms. (in vacuo) of aniline. IV. the specific heat of aniline at the respective temperatures. T Experiments 1 to 5 give =232-05 (Table V. supra) when J M=294.99, and 01=17°78 C. From the above table we get w, 8001 when 0,170.78. Hence S1M3+w1=232-05 w1 = 80·01 ... S1M2=152·04 hence S1=5154 (cf. with •5146 supra). Again, the early experiments with water (p. 57) gave : hence when 01=17·2 w1=79-8 =79.95 (cf. with 79.99 supra). Although I have referred to experiments 1 to 5 as unsatisfactory in themselves, the value obtained from them is (as above pointed out) in fair agreement with that deduced from the remaining experiments. The same remark applies to the absolutely independent determinations of the water-equivalent when conducted with water itself, and they afford strong corroborative testimony as to the accuracy of the conclusions. The following simple formula gives the specific-heat curve with sufficient accuracy: S1=0.5156+(01-20) ×·0004+ (1-20)2 ×·000002. (A) The following table gives the experimental results and those obtained from the above formula. I have made a careful search for records of previous determinations, but I have been unable to find any in addition to those given in Landolt and Börnstein's tables- which are as follows: Temp. •5120 Observer. 12° to 150° Petit. NOTE (Sept. 6, 1894).—Having unfortunately strained one of the leads when the tank was at a high temperature, the insulation between the coil and the steel chamber commenced to fall off on Sept. 1st. I have therefore most reluctantly been compelled to take the whole apparatus to pieces, in order to replace the ebonite insulators. On Sept. 3rd I withdrew the whole of the aniline, and on examining it I found that in colour it had darkened considerably. The experiments described in this paper were completed on Aug. 3rd and on Aug. 16th the apparatus was again set going, and was kept continually at work until Sept. 3rd. During this time electrical currents were continually passing through the coil and the stirrer must have made some millions of revolutions. The nature of the experiments that I was then engaged upon compelled close attention to the capacity for heat of the aniline, and I am convinced that even a small change in the specific heat could not have escaped my notice. It would appear, therefore, that the change indicated by this darkening was not of a nature to sensibly affect the specific heat. [An additional Note will be found among the Miscellaneous Articles in the present Number.] III. A Method for Comparing the Values of the Specific Inductive Capacity of a Substance under Slowly and Rapidly Changing Fields: Results for Paraffin and Glass. By EDWIN F. NORTHRUP*. [Plate III.] AXWELL'S electromagnetic theory of light entails the relation that the square of the index of refraction of a substance, for infinitely long waves, shall equal its specific inductive capacity. But as the refractive index has only been obtained for very short waves (excepting some experiments upon the refractive index of a few dielectrics for electric waves), it is not unexpected that this relation should not hold in the case of many substances whose specific inductive capacity has been determined in steady or very slowly alternating fields. It is to be expected, however, if the field of force under which the value of K is obtained could be made to alternate at a rate which would produce waves comparable in length to the waves with which the refractive index is determined, that the agreement of K and 2 required by the theory would be obtained. The experiments of Professor Hertz upon electrical oscillation have suggested certain methods for determining the specific inductive capacity under fields changing at a rate which approaches that required. No research, however, aiming at a direct comparison of the values of K under slowly changing and oscillating fields, which is as fully conclusive as could be desired appears to have been made. The objects held in view in the following investigation were-1st. To perfect a method which will enable the specific inductive capacity of solids (for which substances the value of K departs most from the required relation) to be determined under fields of force varying slowly or rapidly as desired; and 2nd. To employ the method on one or two substances, to learn if the agreement with the theory is closer in the latter than it is in the former case. With the method under consideration this comparison may be made with the same apparatus and under conditions similar in all essential features. Description of Apparatus. Fig. 1, Pl. III., gives in elevation the apparatus employed. Fig. 2 in plan. The dimensions of the parts are given in centimetres. The following are its essential * Communicated by the Author. the diagram in |