466 On the Constitution of Aqueous Solutions of Acids, fyc. 



As I also find that this fact is capable of experimental proof, 

 it is necessary to inquire how it happens that the fact is true 

 while the hypothesis that has been founded upon it is falla- 

 cious. The matter is not mysterious. 



One test atom of anhydrous sulphate of magnesia measures, 

 in round numbers, 42 septems. When this combines with 

 7 atoms of water to form the crystallized salt, it condenses 

 22 septems, and retains 20 septems as its measure in com- 

 bination. These facts result from the experiments of Karsten, 

 Kopp, and Holker, cited in the Table. 



In the year 1810 Dalton published some experiments 

 on solutions of this salt, of which, as they are inaccurate, I 

 have only quoted two, the first and the last. According to 

 these, when the solution has a specific gravity of 1'30, the 

 measure of the test atom of anhydrous salt is 25 septems ; 

 but when the specific gravity is 1*05, the measure of the test 

 atom is only 18 septems. These experiments, if correct, 

 would prove three interesting points: — 1. That the salt ex- 

 pands 5 septems per test atom on being liquefied by solution. 

 2. That it possesses a considerable volume in solution. 3. That 

 the volume in solution is variable. These results are in direct 

 opposition to Dalton's recent speculations. 



Anthon's experiments show that the test atom of anhydrous 

 sulphate of magnesia occupies about 9 septems in a solution 

 of sp. gr. 1*27; that it gradually diminishes in bulk with the 

 progress of the dilution, becomes nothing in a solution of 

 sp. gr. 1*05, and in still weaker solutions effects great con- 

 densation, which in the last solution of sp. gr. 1*006, amounts 

 to no less than 66 septems, or § of the volume of the anhy- 

 drous salt. Anthon's experiments were made with the cry- 

 stallized salt; but since the measure of 7 atoms of water is 

 1 1 2*48 septems, the reduction to the anhydrous salt is easy 

 and obvious. Anthon's Table is contained in Gmelin's Hand- 

 buch der Chemie, vierte Auflage, t. ii. p. 235 *. 



Holker's experiment corroborates the experiments of An- 

 thon, as far as a single experiment can serve as evidence in a 

 case of this nature. 



These researches show that anhydrous sulphate of mag- 

 nesia, contrary to Dalton's late assumption, does occupy space 

 when in strong solutions ; that in weak solutions it occasions 

 a condensation greater than its own volume; and that, at a 

 certain intermediate point, it effects no condensation and oc- 



* It is due to Professor Gmelin to state, that I am indebted for many 

 facts and authorities quoted in this memoir to the new edition of his 

 JIandbuch der Chemie, the most comprehensive, accurate, and convenient 

 systematic work on Chemistry that ever issued from the press. 



