484 F. W. Clarke — Some Nickel Ores from Oregon. 



suits ; which I give side by side with two published analyses 

 by Hood,* in order to show the variations. 



Clarke. Hood. Hood. 

 Loss at 110° C. .... 8-87 | 



Loss on ignition 6-99 j" 



Al 2 3 + Fe 2 3 1-18 1-38 1-33 



Si0 2 .... 44-73 48-21 40-55 



MgO ....'.'. 10-56 19-90 21-70 



NiO 27-57 23-88 29-66 



99-90 100-00 100-24 



Neither lime, sulphates, chromium, nor cobalt could be de- 

 tected. Like the New Caledonia garnierite, the fragments of 

 this silicate fell to pieces when immersed in water. 



Of the New Caledonia silicates many analyses have been 

 published, notably by Liversidge and Leibius, Typke, Damour, 

 Garnier and Ulrich, and they vary between widely separated 

 extremes. Not only are there the variations due to mutual 

 replacements of nickel and magnesia, with a range in the per- 

 centage of NiO from - 24 to 45*15 per cent, but there are also 

 great differences in silica and in hydration. It is therefore 

 impossible to say whether' we have to deal with one nickel salt, 

 varying only in its impurities, or with several compounds : 

 although the general similarity of the material from different 

 localities renders the former supposition the more probable. 

 According to Ulrichf the noumeaite and garnierite consist of 

 a soapstone-like base, with a hydroxide or silicate of nickel dis- 

 tributed through it in veins and patches ; while Des Cloizeaux'J: 

 regards noumeaite as a magnesian hydrosilicate impregnated 

 with nickel oxide. The latter view, however, is hardly prob- 

 able, especially when we consider the origin of the minerals ; 

 and Typke§ has cited evidence against it. The prevalent opin- 

 ion, that we have to deal with one or more definite hydrosili- 

 cates of nickel, is best sustained by careful comparative study 

 of the specimens, even though the salts may not be obtained 

 pure or positively formulated. The reciprocal variation of 

 nickel and magnesia in more than twenty published analyses, 

 excludes from further consideration the idea that the nickel is 

 present to any great extent as hydroxide. For temporary con- 

 venience we may use the well-recognized name '' genthite " 

 generically, and apply it to all the nickel silicates from the 

 above-named localities. 



Of the " country rock " surrounding the Oregon beds one 

 large, clean, fresh specimen was received. This was subjected 



* Mineral Resources of the United States for 1883. 



f This Journal, III, xi, 235. % Bull. Soc. Min., i, 29. § Chem. News, xxxiv, 193. 



