SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 65 



hand, if an unknown mineral shows these bands, it is reasonably 

 certain to be zircon, for no other mineral is as yet known to contain 

 uranous uranium. 



THE GARNET GROUP 

 The red colors of garnets of the varieties pyrope, almandite, 

 spessartite, and essonite have been variously interpreted as due to 

 gold, 1 tin, 1 iron, 2 chromium, 3 manganese, 4 and vanadium. 5 Two dif- 

 ferent sets of bands seem to be superposed in the spectra of the 

 members of this group, (A) a narrow band at 620 and a broad one 

 centering at about 590 (these often coalesce) ; and (B) two broad 

 bands at about 530 and 500. In order to correlate, if possible, these 

 spectra with the amounts of the last three of the above listed ele- 

 ments, specimens were analyzed by fusing with sodium carbonate 

 and nitrate, extracting with water, comparing the color of the solu- 

 tion with that of potassium chromate of known strength, then 

 acidifying with sulfuric acid, evaporating, adding hydrogen perox- 

 ide, and titrating the vanadium with standard permanganate ° ; man- 

 ganese being determined colorimetrically in the residues (except 

 in the case of spessartite, where the average of published analyses 

 was used). The results were as follows: 



Variety 



Spectra 



A 



Cr V I Mn 



Per ct. Perct. i Per ct. 



Pyrope. . . . 

 Almandite. 

 Almandite . 

 Spessartite. 

 Essonite. . . 

 Essonite. . . 



Bohemia deep red. . 



Wrangell, Alaska deep red. . 



India violet-red. 



Amelia C. H.. Va. brown. . . . 



Ceylon brown-red 



Ceylon ' brown .... 



strong, 

 distinct 

 distinct 

 distinct 

 distinct 

 faint . 



none. . . 

 strong . 

 strong . 

 distinct, 

 none. . . 

 distinct. 



1. 12 

 0.03 

 0.02 

 0.02 

 O.02 

 0.01 



none 

 0.02 

 0.03 

 0.01 



none 

 0.01 



1 .40 

 1.45 

 1.20 

 33.65 

 0.25 



0.35 



In this table it is evident that spectrum A is connected with the 

 presence of chromium, while B is, if anything, related to the vana- 



1 " In former ages .... it was believed that gold and tin were the coloring 

 principle of garnet." Feuchtwanger, Treatise on Gems, New York, 1838, p. 18. 

 I am indebted to Dr. William S. Disbrow, of Newark, N. J., for calling my 

 attention to this reference. 



2 According to most writers ; but inspection of analyses shows no relation 

 between the color and the content of either ferrous or ferric iron. 



3 First detected by Klaproth, Beitr. Chem. Min., vol. 5, 1810, p. 171 ; men- 

 tioned as the cause of color of pyrope in many books on precious stones. 



1 Regarded as the cause of the color by various writers, and of the absorption 

 spectrum by Brun, Arch. sci. phys. nat., ser. 3, vol. 28, 1892, p. 410, and by 

 Keeley, loc. cit. 



° Uhlig, Verh. nat. Ver. preuss. Rheinl. Westfal., vol. 67, 1910, p. 307 ; Zeits. 

 Kryst. Min., vol. 53, 1913. P- 203. 



6 Cain and Hostetter, Journ. Amer. Chem. Soc, vol. 34, 1912, p. 274. 



