1818.] Scientific Intelligence. 389 



The second variety was from Zinnwald ; its colour was brown, 

 and its specific gravity 5*959. Its constituents were, 



Tungstic acid 76^- 



Lime 16-i- 



Silica 2-if 



Oxide of iron 1-^ 



Alumina, with a trace of lime, -ff- . 

 Pure alumina -A- 



r x 3 4 



98J- 



v 



If we compare these two analyses with each other, on the 

 supposition that the silica, iron, and alumina, are merely mecha- 

 nically mixed, we obtain the composition of tungstate of lime 

 from the first analysis, 



Tungstic acid 100-000 



Lime 24-437 



and from the second, 



Tungstic acid 100-000 



Lime 21-568 



These two analyses do not very well accord with each other ; 

 but the reason seems to be partly owing to the whole of the 

 lime not having been procured in a separate state in the second 

 analysis. A portion of it remained mixed with the alumina. 

 This puts it out of our power to employ the second analysis in 

 our comparison with the experiments of Berzelius. But the 

 first analysis corresponds very well with the results obtained by 

 Berzelius. According to Berzelius's analysis, the weight of an 

 atom of tungstic acid is 15-03. According to the first analysis 

 of Bucholz and Brandes, its weight is 14*834. This differs 

 considerably from 15, which results from the analysis of Berzelius, 

 and shows us the difficulty which attends the accurate determi- 

 nation of the weights of the atoms of bodies. 



V. Magnetism. 



The very little progress which the theory of magnetism has yet 

 made, and the little knowledge of the laws of the variation of 

 the compass which has yet been acquired, are known to all my 

 readers. This is probably the cause why magnetism has of late 

 years been so much neglected in this country. I am induced, 

 partly on this account, and partly in consequence of the great 

 importance of the subject, to call the attention of literary men 

 to a treatise on magnetism to be published about this time by 

 M. Hansteen, Professor of Practical Mathematics in the Nor- 

 wegian University of Christiania. Itis to be entitled " Researches 

 on the Magnetism of the Earth," and is to be divided into nine 



