Dr. Percy's Chemical Examhiation of Lettsomite. 101 



of iron, was treated with hydrosulphate of ammonia. No 

 further precipitate occurred (absence of zinc). 



Second a7iali/ds. — Weiglit of the mineral, carefully freed 

 from extraneous matter by the aid of a lens, 4-25 grs. It 

 was introduced into a very small tube closed at one end: the 

 other end was then drawn out and connected by a caoutchouc 

 joint with a small U-tube filled with chloride of calcium. 

 This was siniilarly connected with a large U-tube, also filled 

 with chloride of calcium, and communicating with an exhaust- 

 ing pump ; so that, by alternately exhausting and allowing the 

 air to enter, the water expelled from the mineral by heat 

 might be collected in the small U-tube and weighed. The 

 tube containing the mineral was heated in an oil-bath at 500° F, 

 for a considerable time. A sensible quantity of water was 

 evolved, and the mineral acquired a dull green colour. The 

 small U-tube increased in weight 0'46 gr., whilst the tube 

 containing the mineral lost 0"4'5. The difference, 001, was 

 probably owing to the air having occasionally been allowed to 

 pass too rapidly from without through the large U-tube to be 

 completely desiccated. On gently heating the tube contain- 

 ing the mineral over a spirit-lamp, a further quantity of water 

 was separated. The heat was increased to dull redness for a 

 few seconds, and the water, which condensed in the upper 

 part of the tube, expelled. This further loss by heat was 0'53. 

 The total loss, therefore, by heat was 0*98, which may be esti- 

 mated as water, for the water last evolved did not redden lit- 

 mus. The error, certainly, does not exceed 1*27 per cent., 

 as proved by the following determination of sulphuric acid. 

 The mineral, deprived of water by heat, was digested in di- 

 lute niti'ic acid. Some reddish-brown matter remained un- 

 dissolved, which, after ignition, and including the ash of a 

 very small filter, weighed O'lO. Chloride of barium was added 

 to the solution. The sulphate of baryta, washed and ignited, 

 weighed 1*74<. 



The excess of baryta was separated by sulphuric acid. The 

 copper was precipitated by sulphuretted hydrogen, and de- 

 termined as in the first analysis. The oxide of copper weighed 

 1'98. It was redissolved in dilute nitric acid, and treated with 

 excess of carbonate of ammonia ; but no precipitate occurred 

 after standing many hours. 



The alumina was separated from the peroxide of iron by 

 potash. After adding hydrochloric acid to the potash solu- 

 tion of alumina, it was boiled with a little chlorate of potash 

 (Fresenius). Ammonia in slight excess was added. The 

 alumina, washed and ignited, weighed 0'47. 



