| ILLINOIS STATE |

| LABORATORY |

I OF NATURAL HISTORY 1

LIBRARY I

L I B RAR.Y

OF THE

U N I VER.SITY

Of ILLINOIS

h^TURAL HISTORY SURVEV

550.5

FI
v.3 / cop.Q.

REMOTE

■-.1 1

LIBKAKY

UNIVERSITY OP ILLINOIS

Ua3Ai!A

Field Columbian Museum.

Publication 109.

Geological Series. Vol. Ill, No. 2.

THE

SHELBURNE AND SOUTH BEND

METEORITES.

Oliver Cummings Farrington, Ph. D.
Curator, Department of Geology.

to

CD

CD

TT

<3C

Chicago, U. S. A.
February 1, 1906.

THE SHELBURNE METEORITE.

BY OLIVER CUMMINGS FARRINGTON.

The Shelburne meteorite fell about three miles from Shelburne,
Ontario, at 8 p. m. August 13, 1904. Two stones were obtained
from the fall, one of which weighed 12.6 kg. (27^ lbs.) and the other
5.6kg. (12^ lbs.). The latter of these stones came into the pos-
session of his Museum, where it is preserved under the Museum
number, Me. 606.

The general phenomena of the fall and the larger stone have
been described by Borgstrom** (Trans. Royal Astronomical Society
of Canada, 1904, pp. 69-94). It remains to describe the smaller stone
and give some additional general observations regarding the fall.
The distance between the points of fall of the two stones was about
three-quarters of a mile and the direction between them a southeast-
northwest one, the smaller stone being at the southeast. The latter
stone fell within eighteen inches of the rear porch of the resi-
dence of Mr. John Shields. The phenomena of the fall, as stated
by Mr. Shields to the writer, were that sounds like a muffled
drum-beat were heard by various members of the family who were
in the house at the time, followed by a dull thud at the rear of the
house. A man at the barn, two or three rods west of the house,
also saw a momentary light. Mr. Shields' impression from the
noise was that an old shed in the rear of the house, shown in Fig.
1, had fallen. He accordingly investigated to see if this were true.
The shed proved to be intact, but a hole newly made was noticed
in the soil near it. It was also noticed that the side of the house
south of the hole was splashed .with mud. No one investigated
farther at the time, but on the morning of the second day (August
15) Mr. Shields dug into the hole and at a depth of eighteen inches*
found the stone here to be described. A portion of a burdock
leaf, which had evidently been carried into the hole with it, lay
under the stone. This showed no evidence of charring or burning.

♦(4m. or 16 inches a -cording to Borgstrom, but judgment might well differ as to the exact
depth.)

8

Field Columbian Museum — Geology, Vol. III.

The character of the soil in which the meteorite fell was clayey,
and no other stones were observed while digging it up. The mete-
orite lay with the side shown in Plate V, down. The weather at the
time was fair, but there had been a shower a few hours previous.
The mud splashed by the meteorite against the house was seen by
the writer when he visited the locality six months after the fall, in
February, 1905. The mud had been thrown in considerable quan-
tity across the porch, a distance of about three feet, upon a

Fig. 1. Place of fall of the Shelburne meteorite. A board marked by the star stands
in the hole from which the meteorite was dug.

window, and was even to be seen on the lower side of the roof
of the porch at a height of about eight feet. The direction in
which the mud was thrown was southeast from the point where
the meteorite fell. The shed shown in Fig. 1, was six and a half
feet from the point of fall. The height of this shed is twelve feet.
It stands north of the house, and northwest of the point of fall of
the meteorite. A line drawn to the point of fall of the other stone
would pass directly over the roof of this shed so that had the di-
rection of fall of the smaller stone been at a low angle with the
horizon, it would have struck the roof. Calculation shows that the
angular altitude of the meteorite must have been at least 26 ° to

Feb., 1906. The Shelburne Meteorite. 9

allow it to clear the shed, if it came from the northwest. If its
movement was in the opposite direction, i. e. toward the north-
west, it must have fallen nearly vertically to have avoided striking
the roof of the porch. This fact, together with the noticeable
throw of mud to the southeast, indicates that the path of the
meteor was toward the southeast. If this view be correct, the
larger stone fell first, which is contrary to the usual rule, and, con-
trary to what would be expected, since the greater momentum of
larger stones usually carries them farther. It is possible in this
case that the bursting of the meteor caused a deviation of motion
which brought the larger stone to the ground first. The accounts
of those who saw the meteor pass seem to be of no value for deter-
mining the direction of motion. In the reports quoted by Borg-
strom four observers assert that the meteor was traveling north-
west and three that it was traveling southeast. A similar conflict
of opinion was found by the writer to exist among those at Shel-
burne who saw the meteor. A point on which all witnesses
agreed, however, was that several reports were heard, at least
as many as three. This would indicate that the meteorite broke
into more pieces than were found.

The stone found by Mr. Shields, and now in the possession of
the Museum, has a shape resembling that of a flat-iron. Its length
is 10 inches (25 cm.), its width 5^2 inches (14 cm.) and its thickness
3 inches (8 cm.). The several surfaces show differences of crust
and rugosity, which indicate the orientation of the meteorite. Thus,
of the broad surfaces, one, that shown in Plate VI, is smooth, and
has only broad, shallow pits. This was the surface found upper-
most when the meteorite was dug up, and is plainly the rear side
of the meteorite. The opposite surface, shown in Plate V, is for
the most part peppered with small, irregular pits and the crust is
thinner. It is not as smooth as the side previously described. It
seems evident from the character of the crust and the pittings
that not only was this the front side of the meteorite in falling,
but that a piece corresponding in outline to the rough portion was
split off during the fall. On the lower side of the surface in
the position in which the meteorite stands in Plate V, the in-
terior of the meteorite is seen, over two areas, each covering about
a square inch. Of these areas the one at the right was pro-
duced by a piece having been chipped off for examination when
the meteorite was first found. The one at the left, triangular in
shape, is a natural scaling which, since it is not encrusted, must

ro Field Columbian Museum — Geology, Vol. III.

have been made about the time the meteorite struck the earth.
It passes along the plane of a nickel-iron-troilite vein such as appears
in other parts of the meteorite, and the position of this vein doubt-
less determined the fracture. Of the narrow surfaces of the mete-
orite, one, that shown in Plate VII, has a rugose character and
incomplete crust similar to that of the front side of the meteorite.
Here, evidently, the meteorite split off from some other mass dur-
ing its descent to the earth. The other narrow surfaces, shown in Plate
VIII, have a complete crust and rounded edges. Their pittings
are few and irregular and show rounding and smoothing.

By means of a cast of the larger stone, kindly furnished the
Museum by Dr. Borgstrom, it was possible to determine in what
manner the two stones may have been joined together. The rear
and front sides are so plainly marked on both stones and the surface
of mid-atmospheric fracture so evident in the smaller one that there
is little difficulty in deciding that the stones were joined together in
the manner indicated in Plates IX and X. Of these, Plate IX
shows what, in the view of the writer, was the front side of the
meteorite in its descent and Plate X the rear side. This deter-
mination does not however, accord with that of Borgstrom;
for the larger stone. Borgstrom reverses them,* basing his
determination chiefly on the fact that on what he re-
gards as the rear side, the apparent directions of flow of
molten matter point in a general way toward the center of the
mass. These directions of flow are determined by a heaping of
the crust on the sides of the pits. Such indications, however, are
liable to be deceptive. Several pits on the smaller stone show
drift phenomena in one direction or another which, in the present
writer's view, are to be regarded as remains of flows from the centers
of the pits outward. These flows probably take place in all directions
but leave traces only here and there. All the other characters of
the side regarded by Borgstrom as the rear one seem without ques-
tion to be those of the front. It has a generally arched or conical
character, a relatively large extent of surface, and deep pits. These
are well known characters of the front side of oriented meteorites.
Moreover, if this were regarded the rear side, the side which
must be taken for the front is one having a form concave toward
the direction of movement through the air. It is hardly conceiv-
able that the mass could have come through the air with its con-
cavity foremost. The form obtained by joining the two stones

*0p. cit. p. 84,

Feb., 1906. The Shelburne Meteorite. ii

together in the manner indicated in Plates IX and X is a somewhat
conchoidal one, indicating a scaling off from a larger mass. Such
a form, as is well known, is exhibited by other meteorites, notably
that of Butsura among stones and Canon Diablo among iron mete-
orites. Such a form would be especially liable to fracture during
descent. Borgstrom remarks that the larger stone is characterized
by concave surfaces. This is also true of the smaller one, and of
Butsura as well. The two parts when placed together in the manner
indicated in Plates IX and X correspond perfectly as regards front
and rear sides. The rear side of each is concave, smooth, and has
broad, shallow pits. The front side is concave, rough, and has small,
deep pits.

The pittings of the two surfaces of the smaller stone shown in
Plates V and VII differ in character from those of the other surfaces.
These surfaces may be said to be rougher than the others in the
sense that the roughness is due to a greater abundance and smaller
size of the pittings. The shape of the pittings is irregular, but in
general, saucer-like with diameters of .5 to 1 centimeter. On the
face shown in Plate VII the pittings tend to become elongated in
character, with the long axes parallel with the long direction of
the surface. The edges which both these rough surfaces make
where they join the other surfaces of the meteorite are much sharper
than the edges of other parts of the meteorite. These sharper edges
and roughness indicate less exposure to fusion and erosion, and
therefore a mid-atmospheric fracture along these surfaces. The
largest pits on the meteorite are on the surfaces shown in Plate
VIII. One depression here shows an area of about 3x3 cm. and a
depth of 1 cm. Secondary pits break the configuration of this, but
all have sloping, rounded edges, showing fusion and erosion during
the entire aerial passage of the meteorite. The depressions on the
rear side, that shown in Plate VI, are still broader and shallower in
character and blend in with the general surface so as to nearly lose
the character of pits.

The crust of the meteorite is uniformly black in color. While
in general smooth in appearance, it is seen even with the naked eye
to be dotted over with minute grains rising above the general surface.
These are for the most part protruding metallic grains whose bright
surface can be discovered by filing. Besides these the crust may
be seen under a lens to be abundantly stippled with clots and threads
which anastomose and blend with one another, producing hollows
and elevations. The threads rarely extend more than a few milli-

12 Field Columbian Museum — Geology, Vol. III.

meters independently and are usually very minute. The appearance
of the substance of the crust is like that of black obsidian, being
black, opaque and of pitchy luster. The crust adheres firmly
throughout to the interior, showing no tendency to scale. There is
a noticeable uniformity in the direction in which the threads of
fused matter run on the different faces. Such directions are
shown in some of the photographs, notably Plates V and VII.
In the face shown in Plate VII it is observable that the
drift is in the direction of greatest length along the middle line,
with diversions to the rear side. If a feather, the barbs of which
had been removed from one side of. the midrib, were laid along the
surface the directions of the remaining barbs would indicate quite
accurately the directions of drift. The drift on this face may there-
fore be described as pinnate. On the face shown in Plate V there is
drift radiating from the center outwards. On the face shown in
Plate VI, or rear side of the meteorite, the drift tends to follow the
direction of greatest length, though modified by radiation outward
from the pits.

The crust studied in thin section under the microscope shows
nearly all the zones described by Borgstrom. The first, third, and
fourth are manifest, but the second zone, or " thin brownish layer,"
which he describes, is not visible in any of the sections which the writer
has examined. The failure of this zone to appear may be due to the
thickness of the sections, but if so it would require unusually thin
sections to show it. The intervention of the colorless, or third, zone
between the dark first and fourth zones is a striking phenomenon
and lends a high degree of probability to Borgstrom's view that the
fourth zone is due to alteration of interstitial glass rather than to
a penetration of molten matter from the surface. The thickness
of the several layers, as observed by the writer, accords with that
noted by Borgstrom, except in the maximum thicknesses which he
quotes. The total crust on the small stone is rarely more than .3 mm.
in thickness. Fragments of the meteorite heated B. B. turn black,
shading to red distant from the flame and fuse on the edges to a
black slag.

The meteorite as received at the Museum was penetrated by
several small cracks extending in a general way at right angles to
its broad surfaces. The courses of some of these can be seen in
Plate V. Mr. Shields, finder of the meteorite, states that he noticed
them the second day after digging the stone up. They probably
indicate therefore a partial shattering of the meteorite due to its

Feb., 1906. The Shelburne Meteorite. 13

impact upon the earth. Borgstrom has calculated from the depth
of the hole which the meteorite made in the earth that the velocity
with which it struck was one of 515 feet (157 m.) per second.*
This is equal to the velocity which a body falling in a vacuum
would acquire in 4600 feet.

The substance of the meteorite as a whole is fairly coherent,
crumbling slightly under pressure by the fingers, but only slightly.
It is sufficiently coherent to take a good polish. The specific gravity
of the meteorite was determined in three ways, the determinations
being made by Mr. H. W. Nichols. The first two determinations
were made with a view to finding the apparent specific gravity, by
which the porosity of the stone is shown. This determination was
made in two ways. First, a cast of the meteorite was im-
mersed in a vessel full of water, and the weight of the water thus
displaced compared with that of the meteorite. This gave G = 3.288.
For the second determination the volume of the meteorite was deter-
mined by comparing the weight of a cube of unit size made of the
same plaster as the cast with the weight of the cast. The weight
of an equal volume of water compared with the weight of the mete-
orite gave G= 3.278. The third determination was made by the
ordinary method of comparing the weight of a piece of the meteorite
immersed in water with the weight of the same in air. From a
slab of the meteorite weighing 480 grams and partially bordered
with crust, the specific gravity obtained by this method was G =
3.504. This corresponds almost exactly with the result obtained
by Borgstrom, which was G = 3.499. Comparison of a mean of the
two values for apparent specific gravity with the specific gravity
as determined by the ordinary method, shows, using the formula
given by Kingf the porosity of the meteorite to be 6.3 per cent.

The interior of the meteorite is in color light ash-gray, flecked
with rusty -brown about the metallic grains, which are nickel-white
to brass or bronze-yellow. Numerous circular spots of light and
dark gray color indicate chondri. Those of dark gray are generally
enstatite, those of light gray chrysolite. The diameter of the
chondri sometimes reaches 6 mm. The metallic grains are for
the most part rather uniform in size and distribution, appear-
ing as metallic points scarce exceeding 1 mm. in any dimension.
They may consist of nickel-iron alone, troilite alone or an aggregate of
these. The two compbnents may be readily distinguished by color.

♦Op. cit.p. 75.

t Agricultural Physics p. 115.

14 Field Columbian Museum — Geology, Vol. III.

In several instances a tendency to a ring-like form is observable,
the diameters of such rings averaging about 2 mm. Aggregation of
the metallic matter in the form of veins is also observable, and
constitutes an essential feature of the meteorite. These veins ap-
pear in section as thin, irregular lines about .5 mm. in width, while
their greatest extent in length noted was 5 inches (13 cm.).
There are three such veins to be seen in the stone appearing
entirely distinct from one another. In a general way they run par-
allel to the broad surfaces of the meteorite, although their course
is tortuous and at times becomes somewhat diagonal to these surfaces.
They outcrop on the crust surfaces of the meteorite as more or less
continuous ridges rising .2 to .3 mm. above the surface. On the
face shown to the left in Plate VIII two such outcrops can be seen
nearly parallel with the front side of the meteorite. One of these
is about half an inch (1 cm.) from the edge, and the other about one
inch (2.5 cm.) below the first. As seen in section none of the metallic
veins runs entirely through the body of the meteorite. In some
sections they appear at the outer edges and disappear in the interior,
while in others they appear in the interior but do not extend to the
edges. This irregularity of course and extent tends, in the writer's
opinion, to confirm his previously expressed view that such veins
are phases of structure of the meteorite rather than filled fissures.*
The general appearance of two of these veins in section, also the
nature of the distribution of the metallic grains in general, is shown in
the section represented in Plate XI. Over the triangular surface
shown in the lower left hand corner of Plate v, where, as before re-
marked, a natural scaling along one of the veins has taken place,
the substance of the vein could be examined. The appearance
of the surface here exposed was one of uniformly bronze-yellow color,
there being no differentiation of ingredients according to color.
On removing a portion about 2 cm. square, however, and grinding
it to a smooth surface, some of the metallic portions showed a
nickel-white color while the rest remained bronze-yellow. This
indicated that the vein was made up of aggregated nickel-iron and
troilite, and this indication was confirmed by further tests. The
nickel-iron grains, some of them several square millimeters in area,
were subjected to the action of nitric acid in order to determine
whether they showed Widmanstatten figures. None appeared, how-
ever, although several trials were made. The action of the acid
only produced a minute pitting of the surface of the metal. By

* Am. Jour. Sci. 4, n, pp. 60-62.

Feb., 1906. The Shelburne Meteorite. 15

continued treatment with strong nitric acid the nickel-iron was en-
tirely dissolved out and the troilite was left free. It was found
to be chiefly in the form of small elongated and flattened nodules and
plates, showing a tendency to faceting at some points, but with no
determinable planes. One of these nodules had a length of 3 mm.
The separation between these nodules and the nickel-iron seemed
complete, there being no intimate intergrowth of the two sub-
stances. The troilite was of dark bronze-yellow color, non-magnetic
and easily fusible to B. B. a magnetic globule.

The microscopic characters of the meteorite have been quite
fully described by Borgstrom, and the features which he points
out are essentially duplicated in the sections before the writer.
The chondritic structure of the meteorite is very marked, and the
chondri exhibit a variety of structures. Especially well represented
are those made up of parallel lamellae of chrysolite and glass.
These lamellae run in different directions in different chondri. In

Fig. 2. Diagrammatic representation of arrangement of chrysolite lamellae in
chondri of Shelburne meteorite.

some they are all parallel and, together with the border of the chon-
drus, extinguish simultaneously. In others they may be found run-
ning in two or more directions, in which case those lamellae which
are parallel extinguish simultaneously, but extinctions are different
for the different groups. In the accompanying diagrams, Fig. 2,
are represented some of the arrangements of lamellae observed.
The first diagram shows a simple single arrangement, the second
two sets of lamellae meeting at angles of 135°, and the third prac-
tically two sets of lamellae meeting at angles of 90 , although
on one side the lamellae are somewhat bent. Extinction in all these
forms is parallel to the long axis of the lamellae. The width
of the lamellae in the chondri of this character is remarkably
uniform, and is about .01 mm. The diameter of the
chondri themselves is from 1 to 1.5 mm. When the individual
lamellae are studied with a high magnifying power their apparent
continuity in the direction of length resolves itself into two

16 Field Columbian Museum — Geology, Vol. III.

or more component lamellae joined end to end. The ends
of these component lamellae are usually rounded. The lamellae
are frequently crossed by fractures which usually run normal to
the length, but are occasionally more inclined.

Between chondri with a structure of the above character and
those which are porphyritic there seem to be all gradations. The
stages are: i. Chondri in which the lamellae are wider and fewer in
number; and 2'. Wide lamellae extending only partially across the
chondrus. If the writer is correct in this observation, it is easy to
see that differences of extinction do not necessarily prove a poly-
somatic origin for a chondrus. The lamellae of each chondrus of the
types figured above are doubtless of a single generation, though
differently oriented. So the crystals of a single chondrus though
differently oriented may be of a single generation. Another arrange-
ment of chrysolite and glass lamellae which was seen in addition
to those noted above was an eccentrically radial one. These la-
mellae are wider than those which are parallel. In this case the
lamellae are wedge-shaped, and are enclosed in a glass so dark
as to be opaque. In the porphyritic chrysolite chondri the crystals
were for the most part uniform in size. In one chondrus, however,
a large crystal with rectangular outline was seen to occupy the
center with smaller ones grouped concentrically about it. In another
large chondrus a smaller one was enclosed. In addition to its
occurrence in chondri chrysolite is to be found in individual crystals
scattered through the mass of the meteorite. These crystals usually
do not appear to be formed in place, but to be fragments consolidated
with the chondri. They show no signs of decomposition or wear, and
are free from inclusions. In outline they are rectangular to polygonal,
and in length measure from .2 to .5 mm.

The enstatite chondri show little variation from the usual fan-
shaped forms. The individual fibers in these forms, however, are
usually much less distinct than the individual lamellae of the chrys-
olite chondri. In one enstatite chondrus an appearance of a system
of fibers crossing the main system at right angles was found on
study with a higher power to be due to a textural change across
the fibers along these lines. Such a change suggests strain. Large
individual crystals of enstatite occur, the largest noted being
lath-shaped and having a length of 4 mm. and a width of 2 mm.
This is truly a remarkable size when compared with that of the
general constituents of the meteorite. The outlines of this crystal
were irregular, yet it was sharply separated from the surround-

Feb., 1906. The Shelburne Meteorite. 17

ing field. Its interior was somewhat corroded and honeycombed,
from what cause does not appear. It showed cleavage in two
directions to which extinction was parallel.

In connection with the occurrence of troilite in the meteor-
ite, it may be noted that one chrysolite chondrus showed grains of
troilite, scattered about its periphery and a vein of the same mineral
extending diametrically across it.

Acknowledgments are due Dr. C. A. Chant, of the University of
Toronto, and the late Arthur Harvey, Esq., of Toronto, for informa-
tion kindly given regarding the meteorite.

THE SOUTH BEND IV1ETEOR1TE.

BY OLIVER CUMMINGS FARRINGTON.

This meteorite was found in the spring of 1893 on a farm about
two miles southeast of the city of South Bend, St. Joseph County,
Indiana. The location of the point of find is 86° 15' W. and 41
38' N. The township in which the find was made is not Portage
township, in which South Bend is located, but the next one east,
Penn township. On account, however, of the close proximity to
the well-known city of South Bend it seems advisable to call the
meteorite by this name. The place of find was a slope of one of
the morainic hills which characterize the area, and the meteorite
was discovered when plowing the soil. It attracted attention as
a curious stone and was therefore thrown upon a pile with other
curious stones, there to lie until its meteoric nature wa's detected
in the fall of 1904 by Mr. George A. Baker of South Bend, Secre-
tary of the Northern Indiana Historical Society. From Mr. Baker
the entire mass was obtained for the Museum. Its weight when
obtained was 5^ pounds (2,374 grams).

The meteorite is seen at a glance to be made up chiefly of iron
and chrysolite, and to be therefore a pallasite.

The shape of the mass may be approximately described as like
that of a baby's shoe. This resemblance is perhaps best shown by
the side view given in Plate XIII. The leg of the shoe, however
does not widen toward the top, but narrows and shows a slight twist.
Following the simile the dimensions of the meteorite may be given
as follows: Length (along sole of shoe), 5>£ inches (14 cm.); height
(from heel to top of leg of shoe), 5 inches (13 cm); width (of sole of
shoe), 1% inches (9 cm.); circumference (around sole of shoe), 15
inches (38 cm.) ; circumference in direction at right angles to above),
12 inches (31 cm.). The appearance of the meteorite from the side
described as the sole of the shoe is shown in Plate XV, and that from
the rear of the shoe, showing the curving of the upper portion, in
Plate XVI.

As all the plates show, the surface of the meteorite is every-
where deeply pitted, giving the entire mass a porous appearance. The

*9

20 Field Columbian Museum — Geology, Vol. III.

pits in general have rounded outlines, and are about half as deep
as broad. A diameter of about half an inch (i cm.) is common,
but occasionally a breadth of one and a half inches (4 cm.) is reached.
At one point the bottoms of two pits on opposite sides meet and
produce the perforation shown in Plate XIV. This perforation is
about one-fourth of an inch (5 mm.) in diameter. Another pit
above this point produces a similar though smaller perforation.
A broader, shallow concavity with subordinate pits occurs upon
this same surface. The diameter of the outer rim of this concavity
is about three inches (8 cm.). The other broad surfaces of the
meteorite tend to be plane or convex. In addition to the pits, which
are confluent at their bases, there are many confluent at their sides,
producing irregular, sinuous depressions all over the surface of
the meteorite. While these cavities are referred to as pits they
should probably not be regarded as due to the aerial course of the
meteorite. On the contrary they are altogether produced, so far as
can be judged, by the weathering out of chrysolite from the metallic
matrix. That they indicate cavities previously occupied by chrys-
olite is shown partly by the spheroidal shape of the pits and partly
by the remains of chrysolite in some of the pits. The edges of the
pits are for the most part rounded so as not to leave sharp, pro-
jecting points. Such roundings may well have been caused by
fusion during the passage of the mass through the atmosphere. Al-
though the substance of the meteorite is tough and firm as a whole,
the surface is considerably rusted and the pits filled to some extent
with sand cemented with iron hydroxide. This indicates that the
meteorite has been exposed for some years to the elements, but
not many, for a moist climate, such as prevails in the region
where it was found, would cause rather rapid decomposition. The
coating of rust on the projecting ridges and points of the meteorite
or in the pits not filled with sand is very thin, a single scratch
with a file serving to reveal bright metal beneath. This rust
is dark brown in color. Where the pits are filled with cemented
sand the color becomes a yellowish-brown. There is no indi-
cation in the contour of the mass of its having been subjected to
movement and pressure, such as it would have undergone had it
been glacially transported. The indications are, therefore, that the
mass fell not many years ago near where it was found.

The specific gravity of the meteorite was determined by weigh-
ing carefully the entire mass, first in air and then in water. This
gave the value G= 4.28. Assuming the specific gravity of chrysolite

Feb., 1906. The South Bend Meteorite. 21

to be 3.35 and that of nickel-iron to be 7.70, the ratio of chrysolite
to nickel-iron by weight in the meteorite indicated by this specific
gravity is: —

Chrysolite 78.63

Nickel-iron 21.37

100.00
This result is necessarily too high for the chrysolite and too low
for the nickel-iron as regards the original constitution of the mass,
on account of the fact that some of the original nickel-iron has
altered to limonite, and some of the pores contain more or less sand.
What change should be made in the above figures on this account
in order to express the actual original composition of the meteorite,
however, it is impossible to determine, but it is hardly likely that
a change greater than 5 % should be made.

A piece of the meteorite weighing about 220 grams was removed
by sawing, giving a section having a surface about 2 inches square
available for study. The appearance of this section is shown in
Plate XVII. As indicated by the external characters, the interior of
the mass proves to be a sponge-like body of nickel-iron, the pores
of which are filled with chrysolite. The shape of the pores tends
to be rounded or polygonal, but is occasionally elongated or quite
irregular. A diameter of about half an inch (12 mm.) is a common
one for the pores, and they rarely exceed this. The distribution of
the nickel-iron is rather uniformly tenuous but occasionally bunched
so as to give a square centimeter of surface without chrysolite.
The walls of the pores as seen after removal of the chrysolite are
sinuous rather than angular and have smooth surfaces. A black
graphitic layer about .1 mm. in thickness usually lines the pores,
separating the nickel-iron from the chrysolite. A similar layer
occurring in the Mount Vernon meteorite has been described by
Tassin.* Etching brings out well-defined figures on the nickel-iron
showing that it is made up of the usual alloys of kamacite, taenite and
plessite. The kamacite bands are swollen and very variable in width,
but rarelv exceed 2 mm. in this direction. For the most part the bands
tend to border the chrysolite blebs, following their outlines in vary-
ing course. Bordering the kamacite on the side opposite the chry-
solite occurs a thin ribbon of taenite appearing and disappearing
without regularity, but for the most part quite constant. The
plessite, dark gray in color, fills the spaces between the kamacite
bands, resembling in its irregular shapes, the hieroglyphic figures
assumed by schreibersite in some of the ataxites. At times its

*Proc. U. S. Nat. Museum, 1905, vol.xxxiii, p. 216.

22 Field Columbian Museum — Geology, Vol. III.

structure is comb-like on account of alternating filaments of taenite,
but for the most part it is uniform and homogeneous. The nickel-
iron is malleable but hard.

An analysis of the nickel-iron made by Mr. H. W. Nichols gave
the following results :

Fe

=

90.22

Ni

=

9-35

Co

=

0.26

Cu

=

O.II

P

=

0.05

S

=

0.05

100.04

Schreibersite forms an accessory constituent of the meteorite,
distinguishable from the nickel-iron by its tin-white color and
granular surface. At one point in the section examined it is seen
uninterruptedly over an area about 4 mm. square. At another
point it forms a part of the wall of one of the pores, separating two
chrysolite blebs by a space of about 1 mm. Again it fills about
one-fourth part of a pore, the rest of the filling being chrysolite.
The occurrence of the schreibersite seems to be independent of the
nickel-iron, no swathing kamacite surrounding it. It is brittle,
magnetic, and gives the test for phosphorus with ammonium molyb-
date.

The chrysolite of the meteorite occurs, as previously stated,
filling the pores of the nickel-iron. In spite of the smoothness of
the walls of the pores the chrysolite adheres firmly to them so as to be
removed only with difficulty. The surface of the blebs is usually
rounded and none showed angular facets suitable for measurement.
The color of the chrysolite is generally dark brown to black, though
occasionally a typical olivine green. Often there are variations
of color in the same bleb. As a rule, the color is lighter toward the
center and grows gradually darker toward the periphery, but occa-
sionally there are sectors sharply separated by being darker or
lighter than the remainder of the bleb. Though the blebs generally
appear opaque when seen as a whole, fragments the size of a pin-
head or larger are usually transparent. Under the microscope such
fragments appear clear except for opaque brown or black layers
scattered through them. The fragments are often magnetic before
heating and always so after heating. The individual blebs are mono-
somatic as shown by the uniform directions of their cleavages and
lack of zonal structure. They are considerably fissured and broken,
but not as much so as in the Imilac chrysolite.

Feb., 1906. The South Bend Meteorite. 23

South Bend is the seventh pallasite to be discovered in the United
States, the others recognized being Admire, Anderson, Brenham,
Eagle Station, Mount Vernon and Port Orford. Of these Anderson
is the nearest in locality to South Bend, but it is one hundred and
fifty miles distant. In structure, moreover, it differs. In the char-
acter of its etching figures and the fissured state of its chrysolite,
South Bend resembles the Imilac pallasite more than that of Kras-
noyarsk. It therefore belongs to the Imilac group, and is the first
representative of this group to be found in the United States. To
the meteorites of Indiana it adds a sixth, those now known from
the State being as follows :

Harrison County,

Stone

Cho.

Fell March 28, 1859.

Kokomo,

Iron

Dc.

Found 1862.

Plymouth,

Iron

Om.

Found 1893.

Rochester,

Stone

Cc.

Fell Dec. 21, 1876.

Rushville,

Stone

Cg.

Found 1866

South Bend,

Iron-stone

Pi.

Found 1893

The locations of the points of fall of the above meteorites are
shown on the accompanying map of Indiana, Plate XVIII. A
noticeable feature of the distribution of these falls is that all but
one are along a north and south line close to the meridian of 86°.
The three falls known in Michigan, viz : Allegan, Grand Rapids and
Reed City, also follow closely the same meridian.

LIBRARY
UNIYERSI1Y Of .

X

«

i

a

E/3

LIBRARY
UNIVERSITY OF II

X

LIBRARY
UNIVERSITY OF ILLINOIS

ANA

LIBKAKY

UNIVERSI1Y Of- ILLINUIS

URBAN/*

FIELD COLUMBIAN MUSEUM

GEOLOGY, VOL. III. PLATE IX.

Shelbume meteorite, showing form produced by joining the two stones together.

Front side, X -.->.

LIBRARY
UNIVERSI1Y CI- II

r.

LIBRARY
UNIVERSITY OF ILLII

FIELD COLUMBIAN MUSEUM.

GEOLOGY, VOL. III. PLATE XII.

tM e

81

s eo

?9 .■'S

/ /

rn

^ /

XT^

T""^^^S^^^C______^

isu

^/

GCORG

L \ f ^

^ 7

/ S ~ v> -^ 1

" 1

/

it

/ ~Yt

\

hL

J / LAnC

-CI

/ ' ^

HURON /

N^f y--i3\

/ — ■ — ■ '
/ / " " — ^ w

/ *0

/ 1 Modoc

77. P 1 - *

/ hurlotv.

Of*

/ . TORONTO^T

1 ~~v^-<

/ /

,3 "~~—/— — '

^C_L

LAML ONTARIO

/ Am/ \

/ J \ DeCe "

/ -JJ

— «

LAtvE / fR.it

ONTARIO.

CANADA-.

8i

8-

So 79

78

Map of Ontario, showing location of known meteorite falls.

IV OF ILLmOiS

—

CQ

LIBRARY

UNIVERSITY OF ILLINOIS
URQ

=

ft

o

33

=

LIBRARY
UNIVERSITY Of ILL;

FIELD COLUMBIAN MUSEUM.

GEOLOGY, VOL. III. PLATE XV.

South Bend meteorite. View of base, X %.

UNIVERSE [}i ILLINOIS

UKiini^'

FIELD COLUMBIAN MUSEUM.

GEOLOGY, VOL. III. PLATE XVI.

South Bend meteorite. End view, X %.

LIBRARY
UNIVERSITY OP ILLINOIS
URS

FIELD COLUMBIAN MUSEUM. GEOLOGY, VOL. III. PLATE XVII.

Etched section of South Bend meteorite, showing arrangement of nickel-
iron and chrysolite. The light portions are nickel-iron, the
dark chrysolite, X %.

LIBRARY
UNIVERSI1Y OF ILLINOIS

FIELD COLUMBIAN MUSEUM.

GEOLOGY, III. PLATE XVIII.

N .DIANA

ST-

Map of Indiana, showing location of known meteorite falls.

e