548 Description and Analysis of a large mass of [Nov 



Metallic iron, 87.' 00 



Silex, ..... 11. 50 



Alumina and Loss, 1 . 50 



100. 00 

 With traces of Arsenic. 



The Scoriaceous part 



Metallic iron, 77. 00 



Silica, 17. 



Aluminum, . . 1 . 50 



Cobalt, 3. 20 



Nickel, 1. 



Chromium, 50 



Arsenic and Selenium, Traces. 



It seems at first sight to be treating the subject loosely to give only 

 these approximate quantities, but it was only after long and repeated 

 and most careful work that I could be satisfied of what I have above an- 

 nounced, and that it was wholly impossible to take any one analysis as 

 representing the average constituent parts of the specimen; but I do not 

 regret my labour, for it enables us to explain how it is that chemist after 

 chemist in Europe, and these men of the first talent, have successively 

 differed in their results, or have found new products, such as the Chro- 

 mium, in the same specimen in which others had failed to detect it. It 

 is evident to me that they obtained assays from different parts of the 

 specimen* and have thus differed, as I again and again found I did from 

 myself, to my no small surprise and perplexity. 



And philosophically considered this is what (so to speak) should 

 really occur, for if we admit these meteorites to be revolving round us 

 as their primary, and thus to be, for us, a sort of satellites, we might 

 imagine that if the earth, when it too was an incandescent asteroid had 

 fallen, like our specimen, in upon some huge siderial primary, and had 

 been there " examined and reported upon" that a chip from about the 



* And indeed this is a matter almost of course. The small specimens brought 

 from foreign countries and the minute fragments obtained from great museums as 

 special favours must all have been very imperfect averages of the whole of any large 

 mass. 



