Hunt.] 172 [May 4, 
kingdom into four classes, namely: I, Metalline; II, Orydized ; III, 
Haloid, and iV, Pyricaustate (combustible or fire-making) species. These 
again are divided into orders, and in some cases into suborders, as was 
set forth on page 382 of the already cited volume. In the large and im- 
portant order of the Sriticares, the only one then considered in detail, 
there was recognized in each one of its three suborders of Protosilicates, 
Protopersilicates and Persilicates—five tribes, designated Hydrospathoid, 
Spathoid, Adamantoid, Phylloid and Colloid (or Porodic); called, in some 
cases, by other more distinctive synonyms, as Pectolitoid, Zeolitoid, 
Ophitoid and Argilloid ; in farther extension of which, we may say Am- 
phiboloid, Feldspathoid, Granatoid, Topazoid, Talcoid, Micoid, ete., for 
the other tribes. The characteristic species of these tribes werc then crit- 
ically examined as regards chemical composition and the relations of this 
to specific gravity and to hardness. These relations were shown in sepa- 
rate tables for the various tribes, and farther in three synoptical tables of 
the suborder (loc.ctt., pp. 899-401). The orderof the OxyDATEs (included, 
like the last, in the class of the OXY DACE AZ) was, at the same time, more 
briefly considered, and shown to include representatives of five similar 
tribes (p. 876). In various orders of the same class, such as CARBONATES 
and BoratEs, as likewise in the suborders of the HALOIDACE 4, suchas 
CuLorips, the soluble and sapid species were recognized as forming tribes— 
Carbosalinoid, Borosalinoid and Chlorosalinoid—contrary to the teaching 
of Mohs and his followers, who made these characters the basis of a class- 
distinction. It should be added that the species of all these various tribes 
have farther to be arranged in genera, and, to complete the system, re- 
quire a binomial Latin nomenclature. 
3. In the study of the various species of the order of Silicates, notice 
was, in every case, taken not only of the specific gravity of the species, 
but of the relations between this and its equivalent or so called molecular 
weight, as shown in what is generally known as its atomic volume, calcu- 
lated by the formula, p+d=—=v. For the purpose of thus comparing 
related species it was necessary to fix a simple unit for p. As we have 
since said, in the study of the species of Classes II and III: ‘‘ We assume 
as the unit for p a weight including that of H = 1.0, of Cl. = 385.5, or of 
O-+-2=8.0. By thus adopting a combining weight of 8.0 for oxygen, 
as a basis, we get a unit which gives a common term of comparison for 
oxyds, sulphids, chlorids, fluorids, and for intermediate compounds like 
the oxysulphids and oxyfluorids common in native species. It is, of course, 
a hypothetical unit, which, for elemental species, and for fluorids, chlorids, 
ete., corresponds with the normal vaporous species ; but for oxydized spe- 
cies is some fraction thereof, as in the cases of water-vapor, H,O, of 
spinels, and other oxyds. 
“We may readily extend this system of hypothetical units from sili- 
cates to carbonates, sulphates, phosphates, and more complex species, by 
dividing in all cases the empirical equivalent weight by twice the number 
