454 MISS E. ASTON AND PKOE. BONNEY ON [Aug. 1896^ 



of the Rifielhorn, for there is some very fissile serpentine near the 

 base of the peak on the more northern side. 1 



It may be interesting to note that the huge boulders near the 

 Mattmark See, in the other branch of the Yispthal, appear to 

 belong to the same variety of serpentine. A slice cut from a specimen 

 collected by one of us from the smallest of the boulders agrees 

 very closely with those described above, and it, too, contains a 

 little awaruite. 



Proceeding now to the chemical composition of the Eiffelhorn 

 rock, we come to two specimens, generally similar under the micro- 

 scope, but in the one perhaps there was rather more of the bright- 

 coloured flaky mineral than in the other, while in it awaruite was 

 distinctly present, and a little of the pyroxene exhibited a faint 

 brown pleochroism. 



These specimens proved on analysis extremely interesting, on- 

 account of the low percentage of water compared with the amount 

 of silica and magnesia, the practical absence of lime — of which only 

 a trace could be detected, — and the presence of a considerable 

 quantity of nickel : traces of copper and arsenic were also found. 

 On first analysing the rock (from the first-named specimen), the 

 nickel was not precipitated, but the total, as will be seen below, only 

 added up to a little over 95 per cent. 



A. B. 



SiO, 4010 40-61 



A1 2 3 } 



Fe 2 3 I 8-82 8-66 



FeO J 



MgO 41-12 41-04 



CaO trace trace 



Combined water ... 490 4'67 



Moisture 0-24 0*24 



95-18 95-22 



An analysis was then made of a considerable quantity of the 

 rock from the second specimen, 5 grams being taken instead of 

 the usual 1 gram. The copper and arsenic were precipitated 

 with sulphuretted hydrogen in acid solution ; the mixed sulphides 

 were boiled with nitric acid, and then the copper was precipitated as- 

 oxide by means of soda. The arsenic was re-precipitated as sulphide, 

 and collected on a weighed filter. Both the copper and arsenic 

 were present in quantities so small that great accuracy in their 

 determination was not possible. The nickel was precipitated with 

 soda : it was weighed as oxide of nickel. Lime was most carefully 

 tested for with ammonium oxalate, but only a slight turbidity 

 appeared after long standing. 



The analysis was as follows : — 



1 See the description in Geol. Mag. 1890, p. 533. 



