Examination of Minerals associated with Serpentine. 47 
washing process is carried on at numerous points along the ser- 
pentine range in the three counties already mentioned. 
At the washings situated about six miles west of the town of 
Chester, the sand in which the chrome occurs, contains a large 
amount of bright green actinolite of which good specimens may 
be obtained. As the chrome sand itself in this locality consists 
almost entirely of small but very perfect octohedra, it might be 
supposed to be richer in chrome than the massive ore which [ had 
examined. ‘The pure sand, separated from all foreign matter, was 
analyzed by Mr. Isaac Starr in Prof. Booth’s Laboratory with the 
following results: 
SiO; 
- - - 0-619 
CreOs - - - - 60-836 
Fe:Os - - - - 38°952 
AlzOs - - - - 0-928 
NiO - - 0100 
The proportions calculated according to the formula are, FeO 
Cr2O: 89'419+ FeO Fe20s 6254-4 FeO Als Os 15784 Si03°619 
+NiO -100 = 97-970. 
The specimen being perfectly pure, proves the substitution of 
alumina for chromic oxyd. The analysis further shows that the 
pure crystallized sand is not richer in chrome than the pure mas- 
Sive ore. ‘The association of nickel with the sand is interesting, 
although present in small quantity, for the locality is some thirty 
or forty miles from that of the massive chrome containing eme- 
rald nickel. 
Magnetic Chrome Sand.—Having observed that most if not 
all chrome sands contain more or less that is magnetic, I thought 
it a question of sufficient interest to determine whether the mag- 
helic portion contained ¢hrome. For this purpose a quantity 
of the sand was extracted by a magnet, and this magnetic 
treated several times successively by the magnet to free it entirely 
from adhering chrome sand. The sand thus obtained consisted 
entirely of magnetic grains. Being then analyzed by the process 
indicated above, it yielded the following result : 
SiO; Pom Pe ~ Pe 1-25 
Cr2O2 ay i - - 41:55 
Fe. O: - - - - 62-02 
Calculated, as before, into the formula RO, R2Os, it gives, FeO 
So 61:07 FeO Fe20; 38-64+Si0; 1-25 =100:96. From 
vhich we may infer that the presence of as small a proportion 
‘@8 38 pr. ct. of magnetie iron is sufficient to render the whole 
Proportion available for the ieicotecctes of the chromates, and — 
the ; y 
that therefore a magnet will not reveal the property of a chrome _ 
