J. D. Dana’s Mineralogical Contributions. 253 
We add a remark with regard to the rhombohedral character of 
the twins. 
In the twins, like fig. 4, the series of planes in each sextant 
are closely related ; thus, as has been observed, the planes J, 22, 
i, 12, respectively have nearly the same inclinations on #2 as 23, 
3) 34, 3 have on the opposite 7. From J the planes narrow down- 
ward, and from 29 they narrow upward and so alternating around. 
In fig. 3, there is a corresponding alternation, though of less ex- 
tent. Comparing it with the simple crystal, it looks like an 
Inversion of the alternate sectors; but as the compound form 
contains only three simple crystals, an actual alternate inversion 
lsimpossible. As the plane J is a fundamental plane, the oc- 
curring one (see fig. 2) should have a supremacy in the twin, and 
with it, the series to which it belongs: and as this series in the 
simple form diminishes from J to the opposite side (the right in 
fig. 2), this would imply a reverse enlargement of the next or (4) 
_ Series, and by this alternation the rhombohedral character would 
esult. The fact, moreover, that the compound is dimorphous 
and that the other form is rhombohedral, with the same angles 
hearly as the twin of Leadhillite (as shown by Brooke and Miller) 
May suggest further reason why the twin should take the alter- 
hating or rhombohedral character. 
_ These views are especially interesting as bearing on the sub- 
Ject of dimorphism, and illustrating the passage of a trimetric 
form to a rhombohedral. 
2. On the so-called Silico-Titanates and Silico- Tantalates. 
In a former number of this Journal it was shown that Sphene 
Was a true silicate of the form (#)? Si?, or what is equivalent 
(8) 8#, in which 
: # (or R? 0?) = TiO? + CaO. 
It is consequently trimorphous with Andalusite and Kyanite. 
In this volume, page 130, the author also observes that the 
formula of Keilhauite, on the same principle may be 
(R?, ®) Si%, 
(R®, Br, Si) Si. 
The Special formula of this last afforded by the analyses, is 
(YoR* + 3,2 +N) Si, or 6R* Si+ 3%r Si+ Ni Si=(R, F, R) Si. 
The analysis by H. Rose of T'scheffkinite (Pogg. Ixii, 591) 
‘ppears to lead to the same general formula with that of Keil- 
ae, OF (Re, %) Si. 
Schorlomite afforded Whitney the formula— 
Oa? Si+ Fe Si4 Ca Ti? 
Making the ‘ti a base as above, the oxygen ratio for the bases and 
Silica is 6 ; 11. But if the silica as obtained be a little too high, 
and that of Wéhlerite 
