OF ARTS AND SCIENCES. 25 



compound of ammonia and fluoride of silicon already described by 

 J. Davy, as was proved by the following analyses, which also show 

 that none of the desired substance (NH 3 ) 3 (SiF 4 ) 2 was formed ; and as 

 it was not obtained under these conditions, it is fair to suppose that it 

 cannot exist. The substances for Analyses I. and II. were prepared 

 by washing the aniline out of the crude product of the reaction with 

 ligroine. The substance for Analysis III. was further purified by 

 sublimation. 



I. 0.3414 grm. of the substance gave 0.1496 grm. of silicic dioxide 

 and 0.4154 grm. of sodic fluoride. 

 II. 0.4400 grm. gave 0.1937 grm. of silicic dioxide. 



III. 0.3770 grm. gave 0.1625 grm. of silicic dioxide. 



IV. 0.1920 grm. gave 0.0840 grm. of silicic dioxide. 



Calculated for Found. 



(NH 3 ) 4 (SiF 4 ) 2 . I. II. III. IV. 



Silicon 20.29 20.44 20.55 20.12 20.41 



Fluorine 55.07 55.03 



It was noticed during the purification of this substance that it sub- 

 limed at a much higher temperature than the trianiline disilicotetra- 

 fluoride. The reaction which takes place when trianiline disilicotetra- 

 fluoride is treated with ammonia gas is the following : 



(C H 5 NH 2 ) 3 (SiF 4 ) 2 + 4NII 3 = (NH 3 ) 4 (SiF 4 ) 2 + 3 C G H 5 NH 2 , 



as was proved by its quantitative investigation. 



2.8564 grm. of trianiline disilicotetrafluoride yielded after treat- 

 ment with ammonia 1.6240 grm. of (NH„) 4 (SiF 4 ) 2 . 



Calculated Percentage according to 



the Reaction given above. Found. 



(NH 3 ) 4 (SiF 4 ) 2 56.67 56.85 



When hydrochloric acid gas is passed over trianiline disilicotetra- 

 fluoride there is no action in the cold ; but if the substance is gently 

 warmed, complete decomposition sets in, aniline chloride sublimes along 

 the tube in needles, and the hydrochloric acid contains fluoride of 

 silicon, as was shown by passing it into water when a precipitate of 

 silicic acid was formed. 



The action of ethyliodide was also tried. At 100° there was no 

 action, but at 150° a product was formed which contained neither 

 fluorine nor silicon, fluoride of silicon being given off when the tube 

 was opened. Under these circumstances, we did not think it worth 



