and Equivalent Volume of some Mineral Species. 211 
cent. of fluorine. This last is given off by ignition as fluorid of 
silicon or boron, and the mineral then becomes soluble in hydro- 
chloric acid. Rammelsberg admits that silicic and boracie acids 
replace each other, and finds that the ratios between the prot- 
oxyds, sesquioxyds, and the two acids, vary exceedingly in the 
different varieties. Without regarding the fluorine, he divides 
the species into the five following groups. 
Ist. Magnesia tourmalines. RsSir+3# Si; the mean equivalent 
Tae deduced from six varieties is according to Rammelsberg 
AML 5 (O=8), and the mean density 3:05, giving an equivalent 
volume of 144:°6 
2nd. Magnesia-iron tourmalines. 2RsSis448Si, with a mean 
equivalent weight deduced from eight varieties of 518-81, anda 
mean density of 3:10, giving 167-3 for the equivalent volume. 
. Iron tourmalines. R;Sis+68# Si; the mean equivalent of 
Six vatiittes is 7747 mg and the mean density 3-2, giving 241 for 
the equivalent volum 
Ath. Iron- sient e tourmalines. RSi+3% Si; the mean equiv- 
want weight of six ee ae was 360°36, and the mean density 
3°08, giving a volume ‘ 
5th. Manganese Di méalines. RSit48 Si; the mean equiva- 
lent weight of four varieties was 449°3 - the mean density 
3°04, giving 148 for the equivalent volum 
A red tourmaline from Rozena of density 2-998, yields accord- 
ing to Rammelsberg the formula RSis+5RS 
If we represent R:2Os by 3rO, the icin dias of these five 
groups become as follows: 
Ist. $150, s(Raro)Ore. 
2nd. SisOn e(Rsri 2)O1 5e 
5th. $is015(R Yi 2 )O1 3. 
Multiplying by four the formulas and equivalents of the Ist 
and 5th, so as to make Siec, and then by a simple proportion 
raising the others to the same scale, we obtain the following for- 
mulas and equivalents. 
: Mean equiv. Mean D. ; Eq. vol | 
j Ist. Siz006o Be 12r36)O48 1764°6 3:05 | 578 
2nd. SjieoQOc60(Riora 0)Os0 1729-2 3°10 558 
3rd. | SizoOco(Rz4ras)Os 2} 1928-0 3°20 602 
4th. SizoOco(Rsras)Oso 1800-0 308 584 
5th. | SizoQeo(Rarss)Os2 1799°6 | 3:04 | 592 
It will be seen that the 2nd and 4th have similar formulas, and 
differ by R202 from the first, while the 3rd, rejecting as insig- 
* Pogg. Annal, 1850, Nos. 8 and 9, and this Journal, vol. xi, p. 257. 
