148 ON THE WORKS AND CHARACTER 



rhombic character of bournonite escaped him, he having 

 taken it for quadratic. Snow he found to have the form of 

 a double six-sided p3'ramid, with a lateral angle of about 

 80° The various observations on its forms are so discrep- 

 ant, however, that it is impossible to state which are correct. 

 On page 81 he gives a crystallographical test to distinguish 

 between the chlorides of barium and of strontium. The 

 crystals of the one are rectangular, eight-sided plates; those 

 of the other librous. 



At ibis point, a handful of Smithson's manuscripts may 

 be mentioned, which escaped the lire at the institution in 

 1865. They consisted of notes on various specimens of 

 minerals and rocks belonging to his collection, and also 

 several fragments of catalogues, which seem to have been 

 begun in various years. The earliest bears the date 1796, 

 the latest 1822. These are of little or no scientitic value, 

 except in so far as they illustrate the way in which he 

 ■worked. The following are a few extracts from them : 



1:^0. 1. — Carbonate of lime containing manganese, from near 

 Aix la Ohappelle. 



It dissolved in nitric acid with effervesence like carbonate of lime. The 

 Bait obtnined from this solution by drying over a candle is quite white, but 

 on heating more it becomes brown, and then on solution in water leaves a 

 Email quantity of brown powder. Prussiate of soda and iron caused a 

 white precipitate in solution of this stone, and in it the least blue was per- 

 ceivable. Tincture of galls produced no black color with it. 



Some of the above nitrous salt melted on platinum with nitrate of pot- 

 ash gave the green color of manganese. 



Copperas put into some of this nitrous solution caused a precipitate of 

 Bulphate of lime. 



This carbonate of limo and manganese becomes brown at tho blow-pipe. 



This curbonulo of limo and manganese colored borux rod. 



No. 19. — Reduced nickel free from arsenic. 



It was made at the blow-pipe from oxido of nickel which had been fused 

 with saltpetre. It contains admixed borax. It is infusible. It probably 

 contains cobalt. 



No. 4. — Crust from the church bell of Torre del Greco^ 

 formed by tho lava in 1794. 



There is a crystal in the little group which is the most regular. The two 

 larger faces of this crystal seem to form an angle of 140° with the prism, 

 and meet together at the summit in an angle of 80°. There is a broken 

 crystal in the same group which seems to show that the four larger faces of 

 tho prism form together angles of 90°. The form of these crystals is 8- 

 sided prisms and 4-sided pyramids and are similar to III. 55. d., having 

 the four edges of tho prism slightly truncated. 



No. 7. — A small group of native gold in 24-sided crystals 

 from Vorospatak, in Transylvania. 



The matrix is evidently a quartzose stone. Shows in many parts minute 



