W. L. Dudley — Curious occurrence of Vivianite. 121 



For analysis some of the mineral was pulverized and sus- 

 pended in water. The small particles of woody substance 

 floating were removed. Heavy brownish mineral matter in 

 small quantity rapidly settled to the bottom, and the water 

 containing the blue substance in suspension was carefully 

 poured off and allowed to settle during twelve hours. This 

 operation was repeated and the mineral was in a very fair state 

 of purity. It was then dried in the air at the temperature of 

 the laboratory, and finally for twelve hours over sulphuric 

 acid. Analysis gave the following result : 



Water given off at 100°C 10-59 per cent. 



" " " 230° 7*24 



Alumina 17*74 



Ferric oxide .. 9*35 



Ferrous oxide 24*58 



Lime 0-59 



Magnesia 0*43 



Phosphoric anhydride (P 2 6 ) 27*71 



Insoluble matter 1*84 



100*07 



When the water was driven off at 100°, the residue had a 

 dull green color resembling chromic oxide. After heating to 

 230° until all of the water was eliminated the color was light 

 brown. In a desiccator, over sulphuric acid, the mineral 

 gradually lost water and for several days became green. This 

 occurred more rapidly of course if the air in the desiccator was 

 exhausted. 



If in the above analysis, the lime, magnesia and insoluble 

 matter be eliminated and the percentages of the remainder 

 be calculated to 100, it is found that the mineral may be 

 very nearly represented by the formula 2(3FeO-f P 2 6 )-f-Fe 2 3 , 

 3A1,0,,(*\0.), + 1'7H 1 0, or 2Fe 1 PA + Al 6 Fe 1 P 4 M +l7H,0. 

 This seems to indicate that the ferrous iron in the mineral is 

 combined with the P 2 5 to form vivianite, Fe 3 P 2 8 + 8H 2 0, 

 and Professor F. W. Clarke arrived at the same conclusion 

 upon examining some specimens which I sent to the National 

 Museum. 



If the double molecule of vivianite, 2(Fe 3 P 2 8 -f-8H 2 0), be sub- 

 tracted from the above formula, there remains Al 6 Fe 2 P 4 22 +H 3 0, 

 which resembles an almost dehydrated double molecule of tur- 

 quois, Al 8 P 4 O 22 +10H 2 O, in which one molecule of Fe 2 3 has 

 replaced one of A1 2 3 . The mineral was so earthy and friable 

 that sections could not be cut, and therefore microscopic evi- 

 dence is impossible. 



Chemical Laboratory of Vanderbilt University, Nashville, Tenn., 

 March 21, 1890. 



