16 On some Clwrnical-uigcnclcs of lulecfridty. 



pairs of plates, in such a way that the platina in the gk?5 

 tube was negative. The. process was continued for four 

 days, when the water was found alkaline. It gave by eva- 

 poration and exposure to a heat of about -lOO' Fahrenheit, 

 soda mixed with a white powder insoluble in acids, the 

 whole weight of which was -^f^ of a grain. The glass tube 

 carefully cleaned and dried weighed 84 grains -j-Vr-* '^^^^ 

 ditTerence between the loss of weight of the lube and the 

 weiglit of the products in the water may be ea^rily explained; 

 some minute detached particles of annanthus were prc-ent, 

 and the soda must have contained water, a substance Vvhlch 

 it is probably perfectly free from in glass. 



Having obtained such results with regard to the disen- 

 gaffement of the saline parts of bodies insoluble in water, 

 I made a number of experiments on soluble compounds ; 

 their decomposition was always much mure rapid, and the 

 phcenomeha perfectly distinct. 



In these processes I employed the agate cups with platina 

 wires, connected by amianthus moistened in pure water; 

 the solutions w-ere introduced into the cups, and the elec- 

 trifying pow'er applied from batteries of fifty pairs of plates, 

 in the usual way. 



A diluted solution of sulphate of potash treated in this 

 manner, produced in four hours at the negative wire a weak 

 lixivium of potash, and a solution of sulphuric acid at the 

 positive wire. 



The phasnomcna were similar when sulphate of soda, ni- 

 trate of potash, nitrate of barytes, sulphate of ammonia, 

 phosphate of soda, succinate oxalate, and benzoate of am- 

 monia, and alum, were used. The acids in a certain time 

 collected in the tube containing the positive wire, and the 

 alkalies and earth? in that containing the negative wire. - 



Solutions of the muriatic salts, decomposed in the same 

 wav, unihjrmly gave oxynmriatic acid on the positive side. 



When compatible mixtures of neutrosaline solutions con- 

 taining the connnon mineral acid were used, the different 

 acids and the diffirent bases seemed to separate together in a 

 mixed state, without any respect to the orders of affinity. 



When mclallic solutions were employed, metallic crystals 



or 



