PLATES. I. & II. Illustrative of Mr. E. H. Griffiths's Paper on the Influence of Temperature on the Specific Heat of Aniline. III. Illustrative of Mr. E. F. Northrup's Paper on a Method for Comparing the Values of the Specific Inductive Capacity of a Substance under Slowly and Rapidly Changing Fields. IV. Illustrative of Prof. A. Smithells's Paper on the Luminosity of Gases. V. & VI. Illustrative of Prof. A. Schuster's Paper on the Scale-Value of the late Dr. Joule's Thermometers. VII. Illustrative of Mr. J. Evershed's Paper on the Radiation of Heated Gases. ERRATUM. Page 168, line 23 from top, insert 15°. ERRATA IN VOLUME XXXVIII. Page 232, line 5 from bottom, for no electrification read comparatively 234, ,, little electrification. 10 from top, for 6×10−5 read 6×10-11; and for 4·8 per cent. read 4.8 × 10-8. Delete first paragraph of § 14. line 11 from bottom, for Even read But. THE LONDON, EDINBURGH, AND DUBLIN PHILOSOPHICAL MAGAZINE AND JOURNAL OF SCIENCE. [FIFTH SERIES.] JANUARY 1895. IN I. Further Studies on Molecular Force. By WILLIAM SUTHERLAND *. N two recent papers on the Attraction of Unlike Molecules the result has been established, that if two molecules of mass m1 attract one another with a force 31A1 m2/r, and two of mass m, with a force 32A2m22/r1, then an m, attracts an me with a force 3(1A12A2) m1m2/4; from which it follows that Am2 is resolvable into two factors Am each belonging to a molecule. Now in the "Laws of Molecular Force (Phil. Mag. xxxv., March 1893), values of M2l, which is proportional to Am2, are given for a large number of substances as determined by five methods, so that it is only necessary to take the square roots of the tabulated values of M2l to get relative values of Am. Indeed, in originally studying the values of M2l, the first step I took was to examine the values of (M2)*; and I noticed the reign of law amongst them, but apparently not over so wide a range of values as was dominated by the empirical law M2=6S+66S2, where S is the sum of certain constants characteristic of the atoms in the molecule, and called the Dynic Equivalents of the atoms. Accordingly this empirical expression was adopted in the "Laws of Molecular Force." Yet even this expression failed to apply to the simpler typical compounds; but in connexion with the values of (M2) I have discovered that there are two principles in operation amongst them, one * Communicated by the Author. Phil. Mag. S. 5. Vol. 39. No. 236. Jan. 1895. B applying to the serial compounds of carbon, and the other to the simpler compounds such as those of inorganic chemistry. The advantage of the empirical expression M2=6S+66 S2 was that it happened to lend itself to some of the transition cases which occur between the two classes. But now it is clearly our duty to set aside the empirical law, and confine our attention to (M2) proportional to Am. The following is a brief statement of the order in which these further studies will be taken, and of their results. 1. Values of (M2) in the carbon compounds, with determination of the parts contributed to them by various atoms and radicals, and proof that in the non-metallic atoms these parts are approximately proportional to the volumes of the atoms. 2. (a) and (b). Development of two methods of determining (M2) for inorganic compounds, especially compounds of the metals, tabulation of the results of the methods, with proof that valency controls the magnitude of molecular force in these compounds. (For example, if RS, is a compound of a metal R of valency n with n atoms of S, then (M2) for RS, is the sum of a value for R and n times a value for S, all divided by the square root of n.) 3. Determination of (M27) for the uncombined metals, with proof that in the main chemical families (M27) for each atom is proportional to the square root of the volume of the atom. and also to the square root of its valency. Relation between the volumes of the metallic atoms in combination and the parts contributed by them to (M27) in their compounds. 4. General summary of results, and analysis of molecular attraction into the sum of atomic attractions, with general statement of their laws. 1. Values of (M2) in the Carbon Compounds. To begin with, the law of the Dynic Equivalents remains unchanged, for the dynic equivalent of an atom being the number of CH, groups which would contribute as much to the value of M2l for a molecule as the atom does, it remains the same for (M27) as for M2l. This will appear in all the values of (M2) that follow. From Table XXV. of the Laws of Molecular Force we get the following values of (M2) for the paraffins : A*m The difference for each CH, group is 97 from C, H, to C6H14, 90 from C&H14 to CH18, and 85 from CH18 to C10H22. There appears to be a progressive diminution in these differences for CH2. It will be of advantage, therefore, to study a more extended series, such as is given on page 266 of the "Laws of Molecular Force" for the paraffins from C5H12 to CH34, the values of M2 being got, as there explained, from Bartoli and Stracciati's surface-tensions at 11° C. To complete the series we can obtain approximate values of M2 for CH, and C4H10 by the approximate relation Ml=1190 v1 Tb, given on page 247 of the "Laws of Molecular Force," dividing by 10% to reduce to the same units as are employed in table xxv. To is the boiling-point measured from absolute zero, and v, is the volume of a gramme at the boiling-point: for C3H8, T=246 and v1=1.75; and for C4H10, T=274 and v1=1.71; and thus the values of M2l to go along with the rest are 22.1 and 32-3; and for (M2) we have the series of values and differences : 89 b (M2) Diff. TABLE I. C2H. CH. C4H10. C,H12 CH14 CH16. CH18 CH20. 3.8 47 5-7 | 6.9 1 7.6 8.8 | 9-5 10.5 CH30 C15H32 C18H34 C10H22 CH24 C12H26 C13H2 16 (M2)...... 11-3 | 12-1 13-1 | 138 14-7 15.2 | 100 Diff.... .8 .8 1.0 .7 .9 Here, although the values of the differences for CH, are unsteady (doubtless on account of the difficulty of preparing the paraffins pure), there is on the whole a progressive diminution in the value of the difference: thus from C2He to C8H18 the difference is 5.7, and from C8H18 to C14H30 it is 5.2 But in the case of the esters from C3H6O2 to C10H2002, to which we proceed, there is no such diminution. The following are the mean values of (M27) for the different isomers under each formula : TABLE II. C,H,O.. CHO,. C,H,O. CH2O, CHO, CHO,. The mean difference is here 9, which is the mean difference |