META- AND DIMETASILICATES. 
93 
a little aluminum being replaced by iron. Barkevikite, a more basic 
species still, has its composition well indicated by the following 
expression, with some Fe replacing Al as in arfvedsonite : 
3 Fe 
Fe 
II 
.Si0 4 s 
X Si 3 0^ 
li 
Na 2 
Fe 
(A10) 2 
II 
/Si0 4x 
2 Oa< >Oa 
x Si 3 <Y 
II 
(AlO), 
Finally, in the triclinic aeniginatite, if we assume that the titanium 
(7.57 per cent TiO*) is equivalent to silica, we find the mixture 
Fe 
II 
/SiO. 
2 Fe< >Fe 
X Si 3 Or 
+ 
1 Fe; 
Fe 
II 
.Si0 4N 
Fe 
Na* 
(BO), 
with RO = AlO : FeO : : 1 : 1 almost exactly, and Si : Ti : : 7 : 1. Calculating 
with these ratios, a3iiigmatite has the following composition in com- 
parison with Forsberg's analysis :* 
Fouiid. Calculated. 
Si0 3 .. 
TiO,.. 
Fe 2 3 . 
A1 2 :J . 
FeO.. 
MnO.. 
CaO.. 
MgO.. 
Na 2 0. 
K 2 .. 
37.92 
7.57 
5.81 
3.23 
35.88 
1.00 
1.36 
.33 
6.58 
.51 
38.11 
7.26 
4.85 
3.08 
39.20 
7.50 
100. 19 
100. 00 
The variations here are plainly due to the replacements of Fe by Mn, 
Ca, and Mg, and of Na by K. 
A careful study of the best analyses in the pyroxene and amphibole 
groups will strengthen very materially the view here developed that 
the species are not true metasilicates. Although in most cases the 
approximation to inetasilicate ratios is very close, there are distinct 
variations toward orthosilicates on one side and toward trisilicates on 
the other, and it is only by assuming that we have mixed silicates to 
deal with that all the anomalies can be made to disappear. On this 
^Zeit. Kryst. Min., XVI, p. 428. 
