4 PROCEEDINGS OF THE NATIONAL MUSEUM vol.70 



This metallic portion was digested in a hot solution, saturated, of mercuric chlo- 

 ride, whereupon the metal went into solution. The undissolved silicate was fil- 

 tered out and weighed. The solution, after being freed from mercury, was 

 analyzed bj' the standard methods, giving the composition of the metallic 

 portion. 



The silicates w-ere thoroughly mi.ved and a 2-gram portion weighed into a 

 casserole, covered with 200 c. c. of 1: 1 hydrochloric acid and evaporated to dry- 

 ness on the steam bath. The residue was then moistened with a mixture of 

 nitric and hydrochloric acids to decompose any cohenite or schreibersite, again 

 evaporated to dryness, taken up in hydrochloric acid and filtered. The residue, 

 after thorough washing with hot water, was digested on the steam Isath with 10 

 per cent sodium carbonate solution. This was again filtered and washed with 

 hot water until free from alkali, then rinsed once with hydrochloric acid and 

 washed free from acid. The acid and alkali extracts were united, evaporated 

 to dryness, and analyzed by standard methods, giving the composition of the 

 soluble silicate portion. The residue on the filter was weighed as insoluble sil- 

 icate and analyzed by standard methods to give the composition of the insol- 

 uble silicate. Sulphur was determined by the standard method in a separate 

 portion and assigned to the soluble portion. Alkalies were determined on 

 another separate sample of the powder and relegated to the insoluble silicate. 



The results of the analyses as given above are fairly typical for 

 stones of the class. So far as relative proportion of metal and sili- 

 cates, and composition of the metal are concerned, the results are in 

 close agreement with those obtained by Whitfield on the stone of 

 Elm Creek, Kans.^ The metal content, it will be observed, is below 

 the average (1 1 .98%) of stony meteorites as given in the paper quoted. 

 The ratio of metal to nickel is about 7 to 1, and that of magnesia 

 (MgO) to ferrous oxide (FeO) in the magnesium silicates but 1 to 1 . 

 These facts are of interest in connection with generalizations that 

 have been made on the subject.^ 



EXPLANATION OF PLATES 



Plate 1 



Two views of the largest individual. Actual size about 13 by 10 by 9.5 v.iv, 

 weight, 2,250 grams. Shows primary and secondary encrustation. 



Plate 2 



Two views of second largest stone. Actual size, 1.3 by 7 by 7 cm.; weight, 1,850 



grams. 



Plate 3 



The two smaller stones. Fig. 1: Size, 7 by 10 by 6 cm.; weiglit, 1,114 grams. 

 Fig. 2: Size, 7 by 9 by 6 cm.; weight, 850 grams. 



sMem. Nai. Acad. Sci., vol. II, 191C, pp. 15-16. 

 •Mineralogical Magazine, vol. 18, Nov., 1916, pp. 26-44 



o 



