through the filter. The hest method is to precipitate the metal as 

 iodide from strongly ammoniacal solution, it is best to precipitate at a 

 temperatui'e of 100°, to use solutions not too dilute, to allow the mix- 

 ture to cool before filte»ing, and to wash out with ammoniated water. 

 Werther denies that the iodide is soluble in excess of iodide of potas- 

 sium, and only to the extent of -f-tyjny ^^ water. 



If the thallium is present in the form of teroxide, it can be precipi- 

 tated by ammonia, the solution must be cold, and the washing contiur- 

 uous, otherwise a portion of protoxide is formed. The teroxide may 

 also be reduced by sulphurous acid, and the metal precipitated as 

 iodide. 



Thallium and all its compounds communicate a brilliant green col- 

 our to the flame of a spirit or gas lamp, and if obtainable in sufficient 

 quantity will doubtless be used hereafter for pyrotechnic purposes. 



Protoxide is obtained by the action of baryta on the sulphate ; it 

 is easily soluble in water, attracts carbonic acid from the air, dissolves 

 silicic acid, and hence cannot be kept long in glass vessels ; the solu~ 

 tion is strougly alkaline. 



Peroxide is obtained by the action of ammonia on the perchloride, 

 and is insoluble in water, and of a brown colour. 



Protockloride is white, slightly soluble in water, fuses to a horny 

 mass like chloride of silver, unchanged by exposure to light. 



Apparently there are some other chlorides standing between this 

 and the following compound : — 



Terchloride is obtained by the action of chlorine on thallium under 

 water, or by the long continued action of the same on the proto» 

 chloride. The action must be continued until bichloride of platinum 

 gives no precipitate. The chlorine is then driven out by carbonic 

 acid, and the solution evaporated in vacuo. It cannot be produced 

 by the action of aqua regia on the metal, as mixtures are formed of 

 the proto and terchloride. The solution, when evaporated, deposits 

 long colourless thick prisms, very easily soluble and deliquescent. 

 The composition is TICP + 2H0, but the water cannot be deter- 

 mined by heating as a portion of protochloride is formed. 



Iodide is precipitated by iodide of potassiumi either of an orange^' 

 red colour or citron-yellow, according as the solutions are concentrated 

 and hot, or dilute and cold. From a solution of acetate of potassium 

 it is precipitated of an orange-red colour, and in small cubical crystals ; 

 they contain no water, fuse into a blackish-red liquid, and partly gub- 



