Rocks and rock minerals; a manual of the elements of petrology without the use of the microscope, for the geologist, engineer, miner, architect, etc., and for instruction in colleges and schools online

the closed glass tube on heating which is the best dis-
tinction from gypsum; the difference in cleavage also
aids in the discrimination.

Occurrence. Like gypsum, anhydrite forms inter-
stratified beds in sedimentary formations especially in
limestones and shales. It is also found in masses and in
geodes in such rocks and is a common associate of rock
salt and gypsum.

ROCK SALT HALITE.

Rock salt, sodium chloride, NaCl, is the only chloride which
occurs as a rock-forming constituent in such amounts as to be of
importance. It is easily recognized by its cubic crystals, perfect
cubic cleavage, solubility and saline taste. Colorless and trans-
parent to white, translucent; frequently tinted various colors by
impurities. Hardness = 2.5. Occurs in beds, sometimes of enor-
mous thickness, in the sedimentary formations, usually clays or
shales, and is frequently accompanied by gypsum and anhydrite.



CHAPTER V.
THE DETERMINATION OF ROCK MINERALS.

THE more important of the physical properties of the
rock minerals have been described in the previous chap-
ter, and in most cases the methods by which these are
to be determined have been stated under them. In the
present chapter it is proposed to present a number of
qualitative chemical tests, which can generally be made
with a few reagents and simple apparatus, and which are
of great service in mineral determination and in thus
aiding to classify rocks. This is followed by a set of
tables; one for rough approximations in the field, the other
for more complete identifications in the laboratory by
means of the properties and methods described.

Chemical Tests.

These consist mainly in observing the effect of acids
upon the mineral, whether it is dissolved or only partly
attacked or is wholly insoluble; if soluble, or partly so, in
ascertaining with certain reagents what substances have
gone into solution. These serve to detect the acid radicals
and metallic oxides which alone or in combination com-
pose the rock minerals. A few useful additional tests are
given.

A. Powdering the Sample. The first thing in testing
the solubility of minerals is to prepare a finely powdered
sample. Small chips, grains or splinters of the sub-
stance about the size of wheat grains and as pure as
possible are successively crushed and ground to a fine
powder, like flour, in a diamond, steel or agate mortar,
until a sufficient amount has been produced. It is
usually best to crush the fragments in the steel mortar,

114



THE DETERMINATION OF ROCK MINERALS 115

and unless iron is to be tested for they may be ground in
this as well. The finer the powder is ground the more
readily it will go into solution; it is therefore generally
best to grind it until it no longer feels gritty when a small
pinch is rubbed between the fingers.

B. Treatment with Acid. A small bulk of the powder
prepared as above, about equal to a pea in volume, is put
in a test tube, covered with about an inch of distilled
water and a few drops of nitric or hydrochloric acid
added. Either may be used, but if it is desired to test
for phosphorus or chlorine subsequently in the solution
the former should be employed. The test tube, if the cold
acid has no apparent effect upon the substance, may then
be gently heatea over the flame of a bunsen burner or
lamp of a suitable kind until the liquid is brought to
boiling. If the effect is slight or apparently none has
been produced more acid is added and the boiling repeated
until the substance is brought into solution or it is appa-
rent that it cannot be dissolved.

C. Carbonic Acids Carbonates. If the substance
effervesces freely and readily in cold dilute acid it indicates
that it is a carbonate and that C02 gas is being given off.
It is probably calcite. The carbonates of magnesia and
iron (dolomite, siderite, etc.) are scarcely acted upon or
but very slowly in cold acid but effervesce freely when the
liquid is heated. This serves as a convenient means of
distinction between calcite and dolomite which may be
confirmed by tests for lime and magnesia as given beyond.
The rare rock mineral cancrinite, a silicate containing
CO 2 also effervesces very slowly in hot dilute acid; the
heated solution should be examined in a good light with
a lens when a slow persistent evolution of minute bubbles
will be seen. In regarding the heated solution care must
be taken not to confuse the ebullition of steam bubbles
with the effervescence of C0 2 gas; a moment's pause
should be given to allow the former to cease.

D. Soluble Silicates. Gelatinization. If the sub-



116 ROCKS AND ROCK MINERALS

stance treated according to A wholly or partly dissolves
without effervescence it should be tested for soluble
silica. If only partly attacked the insoluble portion
should be filtered off and the filtrate used. This is con-
centrated by boiling it down in the test tube, the latter
being continuously and gently shaken to prevent crack-
ing, until the solution is greatly concentrated, if necessary
to a few drops. It is then allowed to cool and stand, and
if it becomes a jelly the presence of soluble silica is indi-
cated. If the amount of silicate which has gone into
solution is relatively large, a jelly will probably form
while the solution is being boiled down and is still hot,
otherwise the solution must be concentrated and allowed
to stand, as just stated; in the latter case care must be
taken not to confuse a thickening of the solution from
concentration in it of the salts, especially basic salts of
iron, with the true jelly of soluble silica. If the solution
is carefully carried to dryness and the residue heated for a
few moments the salts on being moistened with strong
hydrochloric acid and then warmed will go into solution
in water while the silica is left as an insoluble residue and
may be filtered off. In the filtrate the various metallic
bases, aluminum, iron, calcium, etc., which may be in
solution, can be tested for by the methods described
beyond.

The rock-making silicates which will go into solution on
boiling with nitric or hydrochloric acid are nephelite,
sodalite, analcite, olivine, chondrodite, serpentine, anor-
thite, leucite, noselite, stilbite, heulandite and cancrinite.
All except serpentine, leucite, analcite, stilbite and
heulandite yield gelatinous silica; with these when the
liquid is boiled it turns from the milkiness caused by the
suspended material to a translucent appearance with
slimy silica suspended in it. The bases, however, have
gone into solution.

E. Insoluble Silicates. Fusion. Most rock-making sil-
icates are insoluble in acids or only partially soluble. To



THE DETERMINATION OF ROCK MINERALS 117

get them into solution so that the bases may be tested for
as in the following sections, a preliminary process of
fusion with sodium carbonate, Na2COs, must be under-
taken. For this purpose some of the powder obtained
in A is mixed with about 4-5 times its weight of dry
anhydrous sodium carbonate, placed in a platinum
crucible or spoon, and gently heated to redness over a
Bunsen lamp flame. If no crucible is at hand a coil of
platinum wire can be used instead; the mixed powder is
made into a thick paste with a little water and a quantity
taken on the coil and carefully fused before the blowpipe.
A fused bead or mass the size of a large pea may be
obtained in this way. The fusion must be conducted
until bubbling has ceased.

In this fusion the silicates are decomposed, silica is liberated
from them and takes the place of the carbonic acid in the sodium
carbonate which is thus converted into sodium silicate. The liberated
CO 2 gas is given off with bubbling and frothing of the fusion and
this effect is in itself indicative of the presence of silica in the original
substance, provided it is known not to come from combined water
by previous trial. The reaction might be illustrated in the case of
pyroxene as follows:

MgCaSi 2 O 6 + 2 Na 2 CO 3 = 2 Na 2 SiO 3 + MgO + CaO + 2 CO 2

The fused mass obtained is broken up in a diamond
mortar, placed in a test tube, and then treated with acid
until dissolved, evaporated and the silica separated and
the metallic bases brought into solution just as directed
for soluble silicates in D.

If the fusion has been made in a platinum crucible the
cake can generally be loosened and removed by boiling it
with a little water; if not, it is dissolved with water and acid
in the crucible, the latter being set in a beaker or dish.

F. . Alumina. The nitrate from the silica obtained in
D or ED combined, may be tested for alumina. It is
heated to boiling after addition of a few drops of nitric
acid, and ammonia is added in slight excess. If a white
or light-colored, flocculent, gelatinous precipitate forms,



118 ROCKS AND ROCK MINERALS

this is aluminium hydroxide. If it is reddish brown, then
it is wholly or in part ferric hydroxide, indicating the
presence of iron, which is very apt to be present in silicates,
especially colored ones, and the alumina may be masked.

If much magnesia is present in the mineral its hydroxide may
also be precipitated at this point by the ammonia, unless the solutions
are rather diluted and a considerable quantity of ammonium chloride
or nitrate has been formed by the neutralization of the acid by the
ammonia.

Scrape some of the precipitate from the filter paper,
transfer it to a clean test tube, or if it is small in amount
transfer it paper and all, and cover with about 5 cubic
centimeters of water; drop in a piece of pure caustic
potash, KOH, about the size of a pea, and boil. The
alumina, if present, will go into solution leaving the iron
hydroxide undissolved; the latter may be now filtered
off. Make the filtrate slightly acid with hydrochloric
acid, boil and then add ammonia in slight excess; alumina,
if present, will be precipitated as the white flocculent
hydroxide.

Alumina can also be detected before the blowpipe by
intensely heating the powdered mineral, moistened with
cobalt nitrate, on charcoal, when its presence is indicated
by the mass turning blue, as mentioned under topaz,
cyanite, etc., in the following tables.

G. Iron. The detection of this metal has been men-
tioned above in F. A more delicate method is to add a
few drops of potassium ferrocyanide solution to a few
drops in water of the final filtrate obtained in D or ED
combined, after boiling with a few drops of nitric acid in
case hydrochloric was originally used. The formation of
a deep Prussian blue precipitate or coloration, if the
solutions are very dilute, indicates the presence of iron.
The nitric acid converts the ferrous salt in the solution into
a ferric one. Potassium /erro-cyanide produces a Prussian
blue with ferric salts, not with ferrous, while conversely



THE DETERMINATION OF ROCK MINERALS 119

potassium /em-cyanide produces the same effect with
ferrous salts. Thus by testing portions of the original
solution of the mineral in hydrochloric acid with these two
reagents the state or states of oxidation of the iron in the
original mineral can be ascertained.

Iron is also shown when minerals become magnetic
after being heated in the reducing flame of the blowpipe.

H. Calcium. The ammoniacal filtrate from the
hydroxides of alumina and iron obtained in F, or the
clear liquid in case ammonia failed to precipitate, may
contain lime salts in solution. To prove the presence of
lime it should be heated to boiling and some ammonium
oxalate added, when the formation of a precipitate of
oxalate of lime proves the presence of this element. If
it should be desired to further test the solution for mag-
nesia the lime oxalate must be removed by filtration; it is
allowed to stand for some time and then filtered; if the
filtrate runs through turbid it should be again passed
through the paper until the liquid is clear. To this a
little more ammonium oxalate is added to prove the com-
plete precipitation of the h'me.

I. Magnesium. Ordinarily this element should not
be tested for until the alumina, iron and lime have been
removed, as directed in F and H, or their absence ascer-
tained. To the solution thus -obtained some sodium
phosphate and a considerable quantity of strong ammonia
are added. The formation of a precipitate, ammonium
magnesium phosphate, proves the presence of this element.
If a precipitate does not form at once it is not, how y ever,
safe to consider magnesia absent, for if the amount is
small and the solution warm it may not appear until the
liquid has become cold and has stood for some time. It
is then apt to appear as a crystalline precipitate attached
to the walls of the vessel.

J. Sodium. A mineral containing sodium when heated
before the blowpipe colors the flame bright yellow.
The best effect is obtained with silicates when the



120 ROCKS AND ROCK MINERALS

powdered mineral is previously mixed with an equal
volume of powdered gypsum and a little of this taken upon
a clean moistened platinum wire which has been previously
tested. The reaction is, however, so delicate and pro-
duced so strongly by minute quantities of the element or
accidental traces that great judgment must be used in
employing it. It is only when the coloration is very
intense and prolonged that the mineral should be inferred
to contain soda as an essential oxide.

K. Potassium. This element may be detected by the
violet color it communicates to the Bunsen or blowpipe
flame. In silicates it is best obtained by powdering the
mineral and mixing it with gypsum, as mentioned under
sodium in J. The flame color is delicate and entirely
obscured by any sodium present; in this case it can be
seen by viewing it through a piece of thick, dark blue
glass which cuts off all but the potash flame. Through
this it will appear of a violet or reddish purple.

Another test is to take a small portion of the final
filtrate obtained in D or ED combined, evaporate it to a
very small volume, add an equal volume of alcohol and
if turbid filter it. Then add a few drops of hydrochlor-
platinic acid, B^PtCle, and if a heavy yellow or orange
colored crystalline precipitate, potassium platinic chloride,
K^PtCle, forms it shows the presence of this element. No
ammonium salt must be present or it will yield a similar
precipitate.

L. Hydrogen Water. If a little of the powdered
mineral be placed in a glass tube, one end of which has
been closed by fusion and drawing off, and gently heated
below redness, the evolution of water, which collects on the
upper walls of the tube, shows that it contains loosely
attached water of crystallization. This occurs with
zeolites, such as analcite, NaAlSi 2 O 6 + H 2 0, and with
gypsum, CaSO 4 + 2 H 2 O. On the other hand, some
minerals, and many silicates are among them, contain
hydrogen and oxygen firmly attached in the form of



THE DETERMINATION OF ROCK MINERALS 121

hydroxyl OH, and this is only given off as water at a
very high heat. Indeed with some, as for instance
staurolite and talc, Mg3Si 4 10 (OH)o, it is necessary to
subject the assay to intense ignition by heating it white
hot before the blowpipe before the water is given off.
This difference in behavior will often serve as a useful
test in determining minerals. Many minerals which
contain hydroxyl also contain fluorine and in this case it
will be often found that the water evolved in the tube
gives an acid reaction to test paper and the glass may be
etched. Unless the latter occurs the test is not however
decisive of the absence or presence of fluorine.

M. Fluorine. This is best tested for as described under
topaz, on page 81, in a bulb tube with sodium meta-
phosphate.

N. Chlorine. This occurs in rock salt, apatite and
sodalite. The test is the precipitation of chlorine as
silver chloride, AgCl, in the solution by addition of a few
drops of silver nitrate. The white precipitate turns
bluish gray on exposure to light. The test for chlorine
is very delicate and slight impurities may cause a faint
opalescence in the liquid on addition of the silver salt.
Rock salt is easily told by its solubility in water, taste
and associations. Apatite usually contains only a very
little chlorine yielding a faint test, or chlorine may be
wanting in it. Sodalite dissolves in dilute nitric acid and
silver nitrate produces in this a considerable precipitate
of the chloride; the nitric acid solution also yields gelatin-
ous silica as in D ; these tests suffice to identify the mineral.

0. Sulphuric Acid. Barium chloride produces in the
solution containing a sulphate a heavy white precipitate
of barium sulphate, BaSO.*, insoluble in hydrochloric or
nitric acid. Gypsum, anhydrite and noselite contain
sulphuric acid; they dissolve in hydrochloric acid and
it may be tested for as above. Noselite also yields
gelatinous silica, as in D, and the two reactions serve to
identify it.



122 ROCKS AND ROCK MINERALS

P. Phosphoric Acid. Dissolve the powdered mineral
(see A) in nitric acid and add some solution of ammonium
molybdate, a yellow precipitate of ammonium phospho-
molybdate shows the presence of phosphorus. This test
is very delicate. It should be conducted with cold or
nearly cold solutions. The precipitate is soluble in excess
of ammonia. If it is desired to make this test in a mixture
of minerals, as in a fine-grained rock for instance, and
silica may be in the solution, it is best to evaporate the
latter and get rid of the silica as directed in D. The
phosphoric acid can then be tested for in the filtrate
acidified with nitric acid. Apatite is the only common
rock-making mineral containing phosphoric acid, and its
presence in rocks and soils can usually be shown by this
test when it cannot be detected megascopically.

Tables for the Megascopic Determination of Rock Minerals.

The two following tables will be found useful in helping
to identify the commoner rock-making minerals. Beside
those given in the tables there are many less common
minerals which enter into the composition of rocks and
which may at times become of local importance. This
is especially true in metamorphic limestones and schists.
Some of the more important of them have been given in
the preceding chapter on the characters of minerals, but
only about fifty minerals or mineral groups constituting
the kinds which are ordinarily met with in megascopic
rock study are here included. The tables can only be
used to distinguish from one another the minerals which
are named in them; they cannot in general be used to
distinguish them from all other minerals. If doubt arises
and a mineral seems to be other than any of those described
here the larger manuals of descriptive and determinative
mineralogy must be consulted for its identification.

Table 1. This is based solely on the most obvious and
easily determinable physical properties and includes about
thirty common minerals or mineral groups. It may



THE DETERMINATION OF ROCK MINERALS 123

often be used to advantage in the field. The only appa-
ratus required in its use are a lens, pocket knife and frag-
ments of quartz and feldspar, in addition to the hammers
usually carried. It will be of advantage to have one
blade of the knife magnetized. The streak or color of the
powdered mineral can be tested by grinding a small piece
to powder between two hammer faces, pouring it on a
piece of white paper and rubbing the dust with the finger
to observe the color produced. A piece may be cracked
into smaller grains and these examined with the lens to
observe the cleavage if it is not well shown by the mineral
on the fractured rock surface. The transparency or
translucency, if not obvious in the mineral in the rock,
may be tested by holding a fragment or sliver against the
light and observing if light is transmitted through its
thinnest edges. The hardness is best tested on a smooth
lustrous cleavage face with the knife point or a sharp-
pointed fragment of quartz or feldspar, substances which
are usually readily obtainable.

Table 2. This includes about fifty of the prominent
rock minerals or mineral groups whose characters are
treated in the foregoing descriptive portion. It requires
for its use some of the simpler apparatus and reagents
found in every chemical and mineralogical laboratory
and the knowledge of how to use them. They have been
already mentioned on page 12.

The table is based upon those of the Brush-Penfield
Determinative Mineralogy which have been modified to
meet the demands of this particular place, and if further
information is desired that manual may be consulted to
advantage.

This table is much more complete and certain in its
identification than Table 1 and should always be used
in preference to it when possible. Table 1 is to be
considered a more or less rough method of approximation
to be used in the field or when no apparatus or reagents
are at hand.



124 ROCKS AND ROCK MINERALS

It should be again repeated that the table cannot be
used for the identification of all minerals which occur in
rocks but only to distinguish the commoner ones, men-
tioned in it, from one another. In most cases the identi-
fication of the mineral is complete, but instances may
occur where some comparatively rare one will give similar
reactions. Thus the rare mineral aragonite would lead to
the same place as calcite, but reference to the description
of the latter would show at once that it differs markedly
in other properties, such as cleavage and crystallization.
This will be usually found to be the case, and if further
information is desired it must be sought elsewhere. But
within the limits imposed the table should serve a useful
purpose to the student of rocks.

TABLE 1.

The mineral has a fine cleavage in one direction; is sometimes
micaceous and may be split into thin leaves by the use of the
knife point. Sec. 1 below.

Has a good cleavage in two directions. Sec. 2.

Has a good cleavage in three directions, forming cubes or
rhombs. Sec. 3.

Has a fine fibrous structure and cleavage. Sec. 4.

No apparent good cleavage. Sec. 5.

SEC. 1. Cleavage in one direction.

A. Micaceous. Cleavage leaves tough, flexible, elastic. Occurs

in crystals, shreds, flakes. Black, brown, gray or white.
Transparent-translucent. Mica, p. 50.

B. Micaceous. Cleavage leaves tough, flexible, non-elastic. In

crystals, shreds, masses. Usually green to dark green.
Chlorite, p. 98.

C. Often micaceous. Leaves flexible but non-elastic. Greasy feel,

very soft, marks cloth. White, greenish, gray. Usually in
foliated masses. Translucent. Talc, p. 102.

D. Leaves somewhat flexible but showing cross cleavage cracks

when bent; in one direction fibrous, the other brittle forming
rhombs. Soft, scratched by finger nail, but not greasy in feel.
Usually colorless, white or reddish; transparent to trans-
lucent. In crystals, masses, seams. Gypsum, p. 111.

E. Leaves have a brilliant metallic luster, like polished steel.

Hematite (micaceous variety) p. 91.



THE DETERMINATION OF ROCK MINERALS 125

F. Leaves brittle; lozenge shaped outline. Usually white, trans-

lucent. Scratched by the knife. Crystals in cavities.
Heulandite, p. 104.

G. Not micaceous, massive, brittle. Very hard, not scratched by

knife or feldspar. Yellowish green to dark green, translucent.
In crystals or masses. Epidote, p. 73.

SEC. 2. Cleavage in two directions.

A. Two cleavages at or very nearly at 90 degrees. Brittle, hard,

not scratched by knife but by quartz. Usually of a light
color, white, pink to red or gray, translucent. In crystals,
grains, masses. Feldspar, p. 34.

B. Usually of a dark color, greenish to black; in grains or short

prisms; sometimes light colored in metamorphic rocks and
then often elongated columnar in cleavage direction. Cleav-
age good but not eminent; prismatic. Cleavage angles 87
and 93 degrees. Usually scratched by feldspar. Pyroxene,
p. 55.

C. Usually of a dark color, greenish to black. Apt to be in crystals

elongated or bladed in cleavage direction. Sometimes light
colored in metamorphic rocks. Cleavage very good with
shining surface. Cleavage angles 55 and 125 degrees.
Usually scratched by feldspar. Amphibole, p. 60.

SEC. 3. Cleavage in three directions.

A. Cleavages not at right angles, forming rhombs. Easily scratched

by knife. Usually white, sometimes tinted various shades
to black; transparent to translucent. In crystals, masses,
veins, etc. Calcite, p. 105, or Dolomite, p. 108. (If rhombic sur-
faces of crystals are curved, probably dolomite.)

B. Cleavages at right angles forming cubes, soluble, strong saline

taste. Transparent colorless or white, rarely tinted. In