SCIENCE 



NEW YORK, FEBRUARY 5, 1893. 



A LUMP OF SALT AND A GLASS OF WATEE.' 



With ordinary use the powers of eye, ear, smell and 

 touch fail to distinguish between the glass of pure water, and 

 that to which salt has been added. The taste alone gives 

 immediate evidence of the difference. But let us examine 

 more closely, and, first, by chemical tests. Solution of sil- 

 ver nitrate, added to the brine, gives a white, curdy precipi- 

 tate containing chlorine, a platinum wire would take up 

 enough to impart a yellow color to the Bunsen flame, indi- 

 cating sodium. Thus two constituents may be separately 

 recognized in the solution by the appropriate tests, where 

 only common salt was added. So, in general, if we wish to 

 detect a salt in solution, we depend upon properties belong- 

 ing to the basic radical and those belonging to the acid radi- 

 cal; the appropriate tests being separately applied. Such 

 properties are called " additive," since they express the sum 

 of the properties of the constituents. The special use of this 

 term may be clearer on reviewing some electrical properties. 



Two kinds of solutions are distinguished by means of the 

 electric current. Absolutely pure water seems to be a non- 

 conductor, vyhile the addition of a salt, acid, or base enables 

 the current to flow, the added body being separated into two 

 parts called ions, which appear at the two electrodes. Such 

 bodies are called electrolytes; and the quantity of electricity 

 passing through the fluid is directly proportional to the 

 quantity of electrolyte decomposed. Many organic bodies 

 are not thus decomposed, their solutions being non conduc- 

 tors. While the molecule of common salt is believed to 

 contain but two atoms, and sugar contains at least forty-five, 

 yet the former may be separated by the electrical influence 

 in a manner from which the latter is free. The forty-five 

 atoms of the sugar molecule dwell togetlier as a unit, while 

 the two atoms of common salt may part company and enter 

 into new relations, thus presenting a scene of activity and 

 complexity which we should hardly expect from its apparent 

 simplicity. 



Let a current pass through a solution of copper sulphate, 

 entering through a copper plate, and passing out at any 

 properly coated form; the copper is carried through the so- 

 lution with the current, and is deposited as an electro plate 

 coating; while the negative radical slips back to attack the 

 kathode. The quantity of basic and acid radicals thus 

 transferred, under given conditions, depends upon the con- 

 ductivity of the solution; but to compare solutions of differ- 

 ent kinds we should make the concentration proportional to 

 the chemical equivalents. In this way Ostwald has meas- 

 ured the molecular electrical conductivity of many solutions 

 of varying degrees of concentration. The following are a 

 few of his results to the nearest unit for extremely dilute solu- 

 tions — yjj^f normal. The differences are shown in small, 

 bold-face type. 



LiCl, 110 9 NaCl, 119 23 K CI, 142 



LlNOa, 105 9 



S 

 LICIO3, 97 10 



NaNOs, 114 22 



r 



NaClOs, 107 23 



KNO3, 136 



6 



KCIO3, 130 



• Abstract ol the annual address before the Washington Chemical Society, 

 delivered Jan. 28, 1892, by Ec bt. B. Warder. 



The numbers obtained for lithium salts are about 9 less than 

 for the corresponding sodium salts, and these about 23 less 

 than for the potassium salts. Comparing the horizontal 

 lines we find the numbers for chlorides about 5 higher than 

 for nitrates, and these about 7 higher than for chlorates. 



To appreciate the full meaning of these differences in the 

 numbers we may again refer to the tests of qualitative 

 analysis. A salt has no single property by which it is rec- 

 ognized, but we depend upon the several properties of basic 

 and acid radicals, which are largely independent of each 

 other. The molecular electrical conductivity is here ex- 

 pressed merely by a number; but do not be repelled by a 

 sense of vagueness. This number expresses motion, — the 

 greater the number the more activity displayed in transfer 

 of electricity. The lithium atom is less active in this way 

 than sodium; and this is true, whatever be the company in 

 which the metal is found. The activity of chlorine is greater 

 than that of the nitric radical, and this greater than the 

 chlorine radical ; but the activity of the salt must be 

 viewed as the sum of this property for the components. 

 Each number is clearly the sum of two numbers, one he- 

 longing to the basic, the other to the acid, radical. On no 

 other hypothesis can we explain the fact that when we select 

 two basic or two acid radicals the substitution of one radical 

 for the other always results in the same change of the num- 

 ber, no matter what third radical may be combined with 

 these two. In a word, the molecular electrical conductivity 

 is an additive property of salt solutions. 



If we leave water and brine in the cold both will freeze; 

 but the brine must be cooled to a lower temperature before 

 freezing begins. The differences between freezing point for 

 solutions and the solvent have been made the subject of 

 many extended researches with special forms of thermome- 

 ter. Eeadings are estimated to .01°. The result has been a. 

 flood of light upon the molecular weights of substances in 

 liquid form, together with some remarkable differences be- 

 tween salt and sugar, between brine and syrup, or between 

 the two classes of solutions which these represent. 



Take three similar barometers, introduce a drop of water 

 into the Torricellian vacuum of the first, and the niercury 

 falls; the water is partly changed to vapor, which exerts a 

 certain pressure on the mercury, and this vapor pressure may 

 be measured by the difference in level. Now put a drop of 

 brine into the second barometer, the mercury falls here alsa, 

 but to a less extent. The vapor pressure of the brine is less 

 than that of pure water. The process of evaporation or con- 

 densation in a current of air affords another means of deter- 

 mining the relative vapor pressure of various solutions. If 

 we now boil water and brine in separate vessels the pressure 

 of vapor equals that of the atmosphere; but, when this point 

 is reached, the brine is hotter than the water, — the boiling 

 point of the former is higher. Thus we have a third method 

 of comparing vapor pressures. This property of solutions, 

 in its quantitative aspect, rivals the freezing point as an ave- 

 nue to the secrets belonging to our subject, which are yet 

 only partly disclosed. As solution proceeds the denser brine 

 gradually mixes with the water above, until at last the whole 

 fluid would be practically uniform. Various salts will dif- 

 fuse at different rates. A porous membrane will transmit 



