Chemical Science. 175 



water, the filtered liquid was acidified by muriatic acid and evapo- 

 rated ; it left fifteen grains of residue, which being introduced into 

 a glass tube with a piece of quicklime slightly moistened and 

 heated, gave ammonia sensible not only to test papers, but also by 

 its strong odour. Hence it results, as M. Chevallier has stated, that 

 the natural oxides of iron contain ammonia, and this fact, conjoined 

 with that of Austin, that ammonia is formed by the oxidation of iron 

 in contact with air and water, acquires a certain degree of geolo- 

 gical importance*. 



10. ATOMIC WEIGHT OF TITANIUM. (Rose.) 



M. Rose some time since endeavoured to ascertain the atomic 

 weight of titanium from the analysis of its sulphuret, but finds, as he 

 suspected, that the sulphuret often contains titanic acid, and there- 

 fore yields uncertain results. In fact, when chlorine was passed 

 over the heated sulphuret, besides the chlorides of titanium and 

 sulphur, titanic acid always appeared. 



He has, therefore, resorted to the chloride of titanium as a more 

 definite compound ; a mixture of titanic acid and charcoal is heated 

 and chlorine passed over it ; the chloride of titanium formed is recti- 

 fied from off mercury or potassium t several times to remove the 

 excess of chlorine, and is then a clear limpid fluid like water, 

 leaving no trace of chlorine when decomposed by water. If this 

 chloride and water be brought together suddenly, heat is evolved, 

 and the solution is milky ; if the chloride is left in a moist atmos- 

 phere, the action takes place without the least formation of turbid- 

 ness. After some time the titanic acid is precipitated by ammonia, 

 carefully added so as not to be in great excess, exposed to a mode- 

 rate temperature to dissipate the excess, and filtered to separate the 

 titanic acid. The above liquor is then mixed with nitric acid, and 

 the chlorine precipitated from it by a solution of nitrate of silver. 

 The titanic acid and chloride of silver are then weighed and give data 

 to determine the quantity of titanium and chlorine in the original 

 compound. From the mean of many experiments thus made, it 

 would appear that one hundred parts of the compound contain 

 Chlorine . 74.46 . . 71.461 

 Titanium . 25.54 . . 23.539 

 and as 74.46 chlorine correspond to 16.82 oxygen, that the titanic 

 acid is composed per cent, of 



Oxygen . 39.71 . I . 36.130 



Titanium . 60.29 . | . 63.870 



Dumas, some time since, endeavoured to ascertain the specific 



gravity of the vapours of the chloride of titanium, and found it to be 



6.836, that of air being 1. This would give the composition of the 



above compounds as expressed in the second column of figures. 



The cause of this difference between the results obtained is, at 



* Ann. de Chimie, xliii., 334. 

 | Potassium does not act upon the compound at boiling temperatures. 



