John Elihu Hall.

The American journal of science online

. (page 27 of 61)
Online LibraryJohn Elihu HallThe American journal of science → online text (page 27 of 61)
Font size
QR-code for this ebook


spar, developed, perhaps, during the late magmatic period, or
during a subsequent period of metamorphism. It possesses a
very marked pleochroism and absorption : light brown or almost
colorless to deep reddish brown. The axial angle as measured
with the microscope was found to be 10 degrees. It resembles
closely mica, often observed in a somewhat similar connection
elsewhere, and is doubtless an iron rich variety. Alteration
changes it to a chlorite. The spinel is of a dull green color,
feeblv translucent, isotropic, and is to be referred to the vari-
ety pleonast. Its grains are entirely without crystalline out-
lines, and it occurs both with the ilmenite and about the feld-
spar grains. Quantitative estimates show that considerable
portions of the ore will not contain over 2 to 2*5 per cent cf
accessories, but the general run will carry from 5 to 6 per cent.

Stimcture and Composition of the Ilmenite Grains. — Exam-
ined with a strong direct illumination, polished surfaces of the
ilmenite show that individual grains are not of homogeneous
composition, although all are identical in character, 'fhey are
made up of a very fine lamellar intergrowth of two kinds of
material. One kind, comprising what is roughly estimated as
one-fifth to one-quarter of the whole, is of a bright steel-gray
color and follows, as a rule, a nearly straight course across the
grains. Many of them pinch out within the grain and in some
the strips have the form of very flat, lensif orm bodies ; again
they are slightly curved in outline. In width they vary from
0*003 to 0*02"". The second series are uniformly broader
than the first and run from 0*036"" to 0*09"" in width. These
are of a dull black color. The whole intergrowth suggests in

Am. Jour. Sci.— Fourth Series, Vol. XXXJII, No. 195.— Maroh, 1912.
18



Digitized by VjOOQ IC



266 Warren — Ilmenite Rocks near St. Urhaiuy Quebec ;

appearance a fine microperthite structure. If a polished sur-
face be immersed in cold dilute hydrochloric acid, the latter
very soon shows the characteristic color of ferric chloride,
becoming in the course of two or three hours (juite stronecly
colored. " The solution gives reactions for ferric iron only.
Pure magnetite, similarly treated, dissolves somewhat more
readily, but the solution reacts strongly for both ferric and
ferrous iron. The steel-gray lamellfle are deepened by the
action of the acid forming tiny grooves, indicating that it is
this part of the intergrowth that is dissolved. As will be seen
from the chemical analysis of the ore given later, the percent-
age of ferric oxide present is in general agreement with the
percentage of the steel-gray lamellfle. These facts point
strongly to the conclusion that the latter are hematite in inti-
mate crystaUographio intergrowth with ilmenite. It may also
be noted that the ilmenite grains are not aflEected by an ordi-
nary magnet, and indeed require a strong field on an electro-
magnet in order to be attracted to the point of picking up.
Their weak magnetic properties indicate quite clearly that
there is no magnetite present The very fine powder rubbed
on smooth white paper gives a dull brownish black streak
(with perhaps a sliglit reddish tint) identical in appearance
with the streak similarly obtained from a mixture of f pure
magnetite (black streak) and J hematite (dark red streak).

To the writer this intergrowth appears to have some interest
in connection with the composition of "titanic-irons" gener-
ally, and the disputed question of the isomorphism of hematite
and ilmenite. While it is impossible to make an exact estimate
of the amount of ferric oxide present as such in the inter-
growth, the approximate estimate given above agrees approx-
imately with the per cent of ferric oxide found by analysis.
From this it appears that the amount of FOjO, mixed isomorph-
ously with the ilmenite must be small. The excess of Fe,0, so
commonly reported in ilmenite analyses has generally been
accepted to mean that the ilmenite and hematite molecules are
isomorphous. If the ilmenite grains in the present instance

f)osses8 the structure originally assumed by them on crystal-
ization from the magmatic condition, the idea at once presents
itself that the Fe^O, present in ilmenite may be always in large
part, at least, present in the form of a tine intergrowth, and
the desirability of examining carefully prepared polished and
etched specimens of ilmenite whose chemical composition is
accurately known is at once apparent. It is of course by no
means certain that the intergrowth is an original structure.
The two molecules may have crystallized originally as an iso-
morphous mixture and subsequently, under changed conditions
of temperature, etc., being no longer stable in the isomorphous



Digitized by VjOOQ IC



New Occurrence of Entile and Sapphirme. 267

state, separated, forming the intergrowth described. Such a
change may have been considerably facilitated by the meta-
morpnism to which this rock has to some degree been sub-

{'ected. It is to be noted that, although the two minerals
lave very nearly the same crystallographic constants, they
differ somewhat in symmetry, and the accepted formute for
the two are not strictly analogous, ilmenite being RTiO, and
hematite Fe^O,. It may, therefore, be questioned whether the
crystallographic and chemical analogies of the two are suffi-
ciently close to permit of isomorphous mixture to more than a
very limited degree, but close enough to condition an intimate
and definite crystallographic intergrowth. It is hoped that a
further study of the relations existing between ilmenite and
hematite and also magnetite when these molecules occur
together may soon be carried out.

Deposit with Rutile-Sapphiriner-Bearing Ilmenite, — The
second, and in the present instance, the most interesting
deposit, is located near the top of the sanie ridge as the pre-
vious one, about one-half a mile to the southwest. It has been
pretty well exposed for a length of about 300 feet and for 50
feet in width. At one point the ilmenite rock has been opened
up to a depth of about 15 feet and to a less depth in several
others. Its contacts with the anorthosite, where exposed, have
a roughly east and west trend and are nearly vertical. In the
ore are several streaks of anorthosite which have also a nearly
vertical extension and a more or less marked schistosity follow-
ing much the same direction. There is also in the ore in
places a feebly marked banding with the same trend. Further
west and southwest of this deposit several small dike-like
masses of ilmenite are exposed with the same orientation, but
these carry no rutile, etc.

The mineralogical character of the greater part of the ilmen-
ite rock in this mass is essentially the same as that previously
described. The grain, however, appears on the average to be
a little finer. A portion of the deposit differs from the rest
and from other known bodies of ilmenite associated with anor-
thosite rocks, in containing a notable percentage of rutile and a
smaller amount of the rare mineral sapphirine.

The rutile-bearing portion was first observed as a streak two
feet wide, with a nearly vertical dip and an indistinctly marked
banding parallel to the walls which ran east and west. The
passage of this streak into the rutile-f ree ilmenite on both sides
was very sudden. Toward the west the rutile-bearing portion
widened, and was somewhat less sharply defined, and there is
evidence which points to the occurrence of rutile-bearing bands
and patches. It appears, however, to always change auite
sharply into the rutile-f ree rock. It was traced for several nun-
dred feet.



Digitized by VjOOQ IC



268 Warren — Ilmenite Rocks near St. Urhain^ Quebec;

Cutting the ore-body in the ru tile-bearing portion is a streak
of anorthosite rock which itself carries more or less rutile.
The rutile makes up from two to three per cent of this rock as
nearly as it was possible to estimate it. It is associated with
ilmenite and a considerable amount of biotite. These minerals
are arranged along distinct lines of schistosity. No sapphirine
has been noted in this rock.

The rutile-hearmg rock is of a brownish black color and con-
sists of a rather finely granular ilmenite thickly sprinkled with
grains of an orange-red rutile, a smaller amount of feldspar,
biotite, sapphirine, or their decomposition products, and spinel.
The sappnirine cannot be distinguished without the aid of a
good lens, and then only upon very close inspection, about
the feldspar and ilmenite grains in the form of very dark,
greenish black grains. The less altered ore is firm, but
weathered portions are somewhat friable. All of the material
collected shows more or less limonite along cracks and joints.
In a limited portion of the rutile-bearing rock fairly numerous
plagioclaee grains or groups of grains, often larger than the
average in size, are present. These feldspars sometimes reach
a length of two or three centimeters and one centimeter in
width, and are characteristically associated with a strong devel-
opment of biotite plates.

Microscopic thin-sections of the rutile-bearing rock disclose
a highly xenomorphic texture for all of the constituent
minerals with the one exception of the spinel inclusions in
some of the feldspar grains. The ilmenite forms an almost
continuous background in which the other minerals lie. Its
grains, although irregular in outline, are roughly equidimen-
sional and are fairly uniform in size, their average cross-section
being about 3™". They consist of the same lamellar inter-
growths as previously described, but the two sets of lamellffi are
narrower than in the former case, conforming to the smaller
average size of the grains. The rutile is in the form of simple
crystal grains or clusters of such, and is of a beautiful orange or
golden brown color with a barely perceptible pleochroism. The
cleavages are prominently developed. Twinning is rare. In-
dividual grains attain a diameter of 3'5""". From this size they
run down to mere specks, the average being in the neighbor-
hood of O-G™"^. It is distributed quite uniformly through the
ilmenite and occurs also with the other minerals, being some-
times enclosed in their grains. The spinel is rather sparingly
present and forms grains comparable in size to those of tlie
rutile. In the feldspar-rich portions of the rutile-bearing rock,
it also occurs included in the feldspars in the form of exceed-
ingly minute crystals. These inclusions deserve a brief
description. Many of the feldspars are crowded with them.



Digitized by



Google



New Occurrence of RutUe cmd Sapphirine, 269

They are of a pale, dull green color and as a rule are
definitely orientated with reference to the enclosing crystal.
Just what the orientation is has not been made out. Many
of the spinels have a highly perfect octahedral habit, the
whole crystal coming into view with slight changes of
focus. Other crystals show the characteristic cross-sections
of distorted octahedrons, or form flat, triangular plates. These
attain a diameter of •015"'°', though usually smaller. Again
the spinels have the habit of relatively greatly elongated rods,
or somewhat flattened blades arranged in lines across the feld-
spars. The inclusions are isotropic, although being wholly
enclosed in the feldspar, which often exhibits a slight disturb-
ance in its optical properties about the inclusions, they often
seem to be slightly doubly-refracting themselves. Minute
inclusions of sapphirine have also been noted occasionally-
associated with the spinels, but these have a different color and
are irregular in outline.

The feldspars are an andesine like that of the anorthosite.
They are as a rule quite evenly distributed and of fairly uniform
size comparable ^ith the ilmenite, but occasionally, as noted,
they become more numerous and of larger size, and form groups
of grains. Many of them show evidences of strains and slight
bending. In even the fresher material collected the feldspar
is usuaUy partly replaced by secondary products, particularly
where biotite and sapphirine were present with it, and in more
highly altered specimens it is entirely gone.

The biotite is sometimes quite abundant and has the same
characteristics as previously described. It is most intimately
associated with the plagioclase, and where sapphirine is present
it appears to have developed later than this mineral. Its
position about the margins of the feldspar, or replacing part of
it is here, as elsewhere, strongly suggestive of a later second-
ary origin. Its alteration is to chloritic products.

An occasional grain of apatite has been noted, but it is
hardly present as more than a trace.

The sapphirine^ which is of especial interest here, it being
not only a new 0Gcm*rence of this rare mineral, but also in a
new association^ seems to be confined to portions of the
ilmenite rock wliich carry riitile. Even when alteration has
destroyed both plagioclase and sapphirine, the characteristic
alteration products enable it to be seen that the sapphirine has
been a quite constant associate of the feldspar in the rutile-
bearing portion. That there is some intimate relation be-
tween the feldspar and the sapphirine is shown by the fact that
the latter generally lies between the feldspar and the ilmenite.
It is often seen as a narrow band extending around the feld-
spar. (See tig. 1.) The band may widen out into a larger



Digitized by VjOOQ IC



270 Warren — llmenite Rocks near St Urhavij Qtcebec ;

mass. Again, the feldspar may be surrounded by sapphirine
of nearly or quite equal area. About a single feldspar grain
the sapphirine usually has, throughout the greater portion of
its extent, a uniform orientation. The mineral also includes
ilmenite and rutile grains, and in one instance, at least, has
been noted in contact with spinel. Where the feldspar is more
abundant the sapphirine is also more strongly developed and
its grains have in some instances fairly continuous distribution

Fia. 1.



Fig. 1. Microphoiograph of thin-section, showing plagioclase (white) al-
most entirely surrounded by sapphirine (gray) either as a narrow band or as
larger grains between it and the ilmenite (black) or rutile (dark, mottled).
The plagioclase shows many minute microlites of spinel. Magnif. about
90 ; ordinary light.

through the section, and often lie, with and without included
ilmenite and rutile. between the feldspar crystals, or even
included in them. In such sections as much as 20 per cent
has been observed with about an equal amount of feldspar and
rutile, a little spinel, and the remainder ilmenite. Here, the
larger grains frequently attain a diameter of 3*0'"^, the aver-



Digitized by VjOOQ IC



New Occurrence of Rutile and Sapphirine. 271

age being somewhat smaller (0*4 or O'T'"""). Quantitative
estimates of the amount present indicate something like 8-5
per cent for the general run of the ru tile-bearing rock.

No trace of crystallographic outline has been noted on the
sapphirine and there is only a faint suggestion of two inter-
secting cleavages. The fracture is marked, being developed
in the form of prominent irregular cracks and is highly con-
choidal. The pleochroism is strong, a, pale, smokv brown to
almost colorless, b=c, deep sapphire-blue, sometimes with a
shade of green. The single refraction is strong, a, having been
determined by the immersion method as approximately 1*729
for sodium light. The double-refraction is very low, not ex-
ceeding 0*005. The interference tints in many sections show
deep berlin blues. The optical character is clearly negative,
the axial angle rather large, and the dispersion is marked r<,v.
It is to be noted that the index of refraction is higher than
that of the sapphirine from the original locality in Greenland
in specimens studied by the writer.

The sapphirine from India* corresponds to that of the
present locality very closely in color, and the index for a has
oeen determined by the writer as about 1'726.

The sapphirine from St. Urbain shows nearly always a more
or less extensive alteration not observed in the case of the
Indian and Greenland varieties. This begins as a finely
fibrous or foliated development along the curved fracture lines
and margins. Further alteration develops a fine aggregate
of more distinctly foliated material which suggests talc in
appearance. These products become mingled with chlorite,
biotite and ferruginous material, resulting from the alteration
of the associated minerals, which renders it diflScult to deter-
mine the exact character of the alteration. Nevertheless, the
appearance of the alteration is quite characteristic and as noted
above, permit it to be easily seen that the sapphirine was
originallv more abundant than at present.

The alteration of the sapphirine, as well as its intimate asso-
ciation with the other minerals, made it difficult to obtain
suitable material for a chemical analysis. By means of an
electro-magnet and heavy solutions, some 3 grms. sapphirine
contaminated with the other minerals and alteration products
were separated from a sample of rutile concentrates, generously
placed at the writer's disposal by Mr. E. K. Berry of the
General Electric Company of Lynn. In barium-mercuric iodide
solution, gmins of seemingly pure sapphirine sank just after

* Petrolo^'cal Stady of Some Rocks from the Hill Tracts, Vizagapatam
District, Madras Presidency. Walker and Connor, Records Geological Sur-
vey of India, vol. zxzvi, Part 1, 1907. The writer is indebted to Professors
Walker of Toronto and Palache of Cambridge, Mass., for specimens of the
Indian and Greenland minerals which were studied in this connection.



Digitized by VjOOQ IC



272 Warren — llmenite Rocks near St Urbain, Quebec ;

epidote, specific gravity 3*54. Less pure grains continued to sink
down to a specific gravity of 3*43. Tlie true specific gravity
probably lies somewhat above 3*5.

The material was further purified by fusing its fine powder
with sodium carbonate, which has only a slight action on the
sapphirine but decomposes the other minerals, including the
rutile and the alteration products of the sapphirine. Some-
thing over a gram of material was obtained in this manner,
which, examined under the microscope, was found to be
entirely free from impurities, and upon this a chemical anal-
ysis yielded the results given below in column 1. In the other
columns are given other analyses of sapphirine.

1 . II III IV V VI VII

SiO • 13-44 14-56 14-61 14-86 14-76 12-96 12-88^

A1,0, 62-98 60-49 63-11 63-25 63-23 64-44 65-49 ^

Fe^O, MnO,-53 -93

CaO -38

FeO 9-08 7-65 3-92 199 1-66 1*66 -65

MgO 15-28 17-13 16-86 19*28 19-75 19*83 1978

Ign -66 -49 -34 -31

Total 100-78 100-39 9979 99-38 99*39 99-22 99*79

Sp.grs 3-6 3-542 3*473 3*46 ^3-486 T

I — Mineral from St. Urbain, Quebec : Warren.
II— " " India: Walker and Collins, Records, Geol.

Surv. of India, vol. xxxvi. Part I, 1907.

III — ** ** Fiskernas, Greenland : Stromeyer, Unters.,

Misch. Mineralk., 1871, 391.

IV — " " Fiskernas, Greenland : Damour, Bull. Soc.

GeoL, vi, 316, 1849.
V — " " Fiskernas, Greenland : Schhittig, Inaug. Diss.,

Leipzig, xxii, 1884, and Zs. Kr., xiii, 74.
VI — " " Fiskernfts, Greenland: Lorenzen, Medd.

Gronl., vii, 184.
VII — " *< Fiskernas, Greenland : Ussing, Zs. Kr., 15,

p. 600.

Below are given the molecular ratios derived from the above
analyses : FeO has been combined with MgO and listed as RO,
Fe,6, and Mn,0, as R,0,.

I II III IV V VI VII

SiO, 0-224 0-242 0-242 0-247 0246 0216 0-214 =

R/), -617 -593 -621 '620 -620 'ty'M -643 =

RO -508 -534 -475 -508 -516 -517 -502 =

SiO, 200 2-00 200 200 2-00 2.00 2-00

R,(>, 5-61 4-90 5-13 5-02 5*03 5-84 605

RO 4-53 4-41 3-92 4-11 4-19 4*79 469



Digitized by VjOOQ IC



Ne\o Occurrence of Rutile and Sapphirine. 273

The ratios derived from the analyses of Loreiizen and Ussing
(VI and VII) lead to the formula given by them, Mg^Al^Si^O^,
although it is to be noted that there is a considerable deficiency
in the KO figure in both cases, and in VI of RjO,. The ratios
derived from the analyses of Stromeyer (III), Damour (IV),
and Schluttig (V) agree rather more closely with the formula
Mg^AljoSigOjg. The ratios of the St. Urbain sapphirine lie about
half way between these two formulae, while those of the Indian
mineral depart rather widely from both, although they perhaps
favor the second type. Ussing, in discussing the composition
of the mineral, attributed the relatively higher silica values of
the older analyses of Stromeyer, Damour, and Schluttig to the
presence of mica or other impurities. This criticism cannot
apply in the case of the St. Urbain mineral, nor apparently in
that of the Indian mineral ; and as the analysis of the mineral
presents no diflSculties, it does not seem as if the differences
can be attributed to analytical errors. The composition is
that of a very basic silicate, and does not appear to conform to
any simple formula. It seems probable that there may be some
solid solution relation existing between the molecules which
make up the mineral ; at least this is the only apparent expla-
nation of the variations in composition which it snows.

It may be noted that there are two types of the mineral from
Greenland (see Ussing), one of a light blue color, the other
having a deeper blue color and occurring in a more ferrugi-
nous association. Ussing suggests that tne deeper blue color
may be due to the presence of more iron in this type, an idea
in keeping with the higher iron content of the St. Urbain and
Indian minerals.

Origin of the Sapphirine, — In the case of the sapphirine
from Fiskernas in Greenland, the mineral appears to be of met-
amorphic origin, occurring, according to Steenstrup,* in small
ellipsoidal or irregular masses in mica-schist or gneiss, high
in MgO and Al^O, but low in silica. Walker, in the paper
alhided to, presents strong evidence to the effect that the Indian
sapphirine is also of metamorphic (contact) origin, having been ^
formed by the action of ultra-basic, spinel-bearing rocks of the
charnockite series, on a sillimanite schist. Its occurrence as a
metamorphic mineral in these two instances suggests strongly
that it may be also of metamorphic origin at St. Urbain, an
hypothesis which is perhaps also supported by the peculiar and
intimate association of the mineral with the feldspar. If so, it
must represent a reaction product between the feldspar and the
spinel, although in the absence of any lime-bearing silicate of
a secondary nature, sucli as epidote or hornblende, it is diffi-

*Medd. om Gronl., Kopenhagen, 1884, vii, 15.



Digitized by VjOOQ IC



274 Wa/rren — Ilmenite Rocks near St Urhain, Quebec/

cult to picture the exact character of the reaction which pro-
duced it. It is perhaps possible that it was formed by a
reaction between the alumina and silica of the plagioclase and
the spinel, some sodium and silica being eliminated and the
remaining feldspar crystallizing as a somewhat more basic
lime-soda feldspar. The optical properties of the feldspar asso-
ciated with the sapphirine do not indicate, however, that it
is more basic than that elsewhere in the ilmenite nor in the
enclosing anorthosite. Nor is it at all clear why tl^ sapphirine
should not have formed in those portions of the ilmenite rock,
also containing the same plagioclase but free from rutile, these
having certainly undergone substantially the same metamor-
phism as the ru tile- bearing portions. Taking everything into
consideration, the writer believes that the sapphirine in this
instance is a magmatic mineral.

Quantitative Study of the Rutile -hearing Rocks. — Two

auantitative estimates (ICosival) of the mineral composition of
le rutile-bearing rock have been made on large, thin sections,
one cut from a sample which appeared richer than the average
in rutile, and the other from a rutile-poor sample. The results
were as follows :

Rutile 20-4 11-3

Ilmenite-hematite 73*2 84'5

Sapphirine 3*2 0-7

Rest 3-2 3-2

1000 1000

It is certain, as noted, that the sapphirine as judged by its
characteristic alteration products was originally more abund-
ant than at present, so that a part of the material listed as
"rest" represents sapphirine once present. The average of
the percentages for rutile is 15*8 per cent, which is probably
not far from the true average rutile content of the rutile-bear-
ing portion.

As stated, it has not been deemed worth while to make at



Online LibraryJohn Elihu HallThe American journal of science → online text (page 27 of 61)