The Rodeo meteorite

5^0

| ILLINOIS STATE |

| LABORATORY |

* OF NATURAL HISTORY 1

LIBRARY |

X I B RARY

OF THL
U N I VE.R.SITY
Of ILLINOIS

m-WRAL HISTORY SURVEy

550.5

FI
v.3,top.Q.

REMOTE ST

LIBRARY

UNWERSW OF ILUNUib

URBANA

Natural History Survey
Library

Field Columbian Museum

Publication ioi.
Geological Series. Vol. Ill, No. i

THE RODEO METEORITE

BY

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

Chicago, U. S. A.
March i, 1905

Field Columbian Museum

Publication ioi.

Geological Series. Vol. Ill, No. i.

THE RODEO METEORITE

BY

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

Chicago, U. S. A.

March I, 1905

THE RODEO METEORITE.

BY OLIVER CUMMINGS FARRINGTON.

This meteorite is an iron mass found about 1852 by a goat-herder
in an arroya north of the Nazas River, 12 kilometers northwest of
the hamlet of Rodeo, State of Durango, Mexico. The location is
approximately 250 20' N. lat. and 1040 40' W. long. Upon discovery
of the iron it was made to do duty as an anvil at a forge for many
years. As received at the Museum, evidence of its industrial use
was to be seen in its having been beaten flat and smooth on one side.
The surface so treated is apparent in Plate II., by its smoothness and
turned-over edges. The meteorite as a whole is irregular in form
and without marked orientation. Its extreme dimensions are 12 x
9x8 inches (30 x 23 x 20 cm.). Its weight when received was 97
pounds (44.1 kgs.). An attempt had evidently been made at some
time to cut off a portion of the mass with a cold-chisel, thus producing
the incision shown in Plate I. Above this a small surface appears
that was filed smooth for etching. In other respects the surface of
the meteorite has the natural contours. The surface in general,
though irregular, is everywhere rounded, showing no angular or
sharp edges. There are many partially defined pittings of various
depths and diameters, the largest of these having an elliptical out-
line and being 4 inches (10 cm.) in length, 3 inches (8 cm.) in width,
and about 1% inches (4 cm.) in depth. The position and character
of this pit are shown in Plate I. In color the surface of the meteorite
is darkened by exposure, but it has nowhere rusted deeply, and in
several places the nickel- white color of the metal is visible. In such
places Widmanstatten figures often can be seen also. On any pol-
ished surface of the meteorite, too, the figures appear nearly as
plainly as after etching.

Several complete sections of the meteorite were made in order to
determine its interior structure. All show on etching well-defined
figures octahedral in character. A photograph of one of these etched
sections is shown in Plate III. The bands (Balken) are more numer-
ous than the meshes (Felder), yet the latter occupy a considerable
amount of the total area. Through a belt about two inches (5 cm.)
in width running across the middle of most of the sections a minutely
dotted appearance is presented resembling that described by Brezina

2 Field Columbian Museum — Geology, Vol. III.

as characterizing Charcas* and referred by him tentatively to
minute inclusions of troilite. An examination of the dots in Rodeo
with the lens shows them to be minute, shallow, saucer-shaped
pits. They are scattered irregularly along the bands of kamacite,
and are to be seen in some of the swathing kamacite, but never
in the plessite. The tendency of the iron to rust at these points
is greater also than at others. They appear therefore, to mark
the occurrence of some more soluble ingredient in the kamacite.
This is probably not troilite, but may be an iron containing less
nickel than the kamacite. The lamellae of the meteorite may be
grouped into two classes; one about i mm. in width, swollen, and
with wavy outlines, and the other about half as wide, and with more
nearly rectilinear outlines. As a rule, these two kinds of lamellae
have a different orientation as compared with each other. The
kamacite is granular, much lighter in color than the plessite. A con-
siderable quantity of swathing kamacite is present. While in general
it follows the outline of the inclusions and forms a narrow border to
them, at times its outer border is quite independent of the shape of
the inclusions and it covers relatively broad areas. The taenite is
well developed, silver-white in color, and displays the structure of a
section brilliantly on holding one at an angle to the light. The ples-
site is not depressed by etching as is the kamacite. At times it occu-
pies the meshes alone, while again the meshes may display elaborate
combs resulting from skeleton growths of taenite. Scattered irregu-
larly through the sections and forming an important feature in the
structure of the meteorite, occur numerous inclusions of schreibersite.
The form of these inclusions, especially those of large size, is in gen-
eral elongated, and rectangular or spindle-shaped. Some of the
smaller inclusions, however, are star-shaped, while others have no
well-defined form. The largest inclusion noted (shown in the upper
right-hand corner of Plate III.) has a length of one and a half inches
(4 cm.) and a width of one-fourth of an inch (.5 cm.). The schreiber-
site is tin-white in color, brittle, and magnetic, and affords the usual
blow-pipe and chemical tests for that mineral. The inclusions are
always bordered by a band of swathing kamacite about 1.5 mm. in
width. The inclusions, while having no apparent regularity of
arrangement among themselves, are usually disposed, especially the
elongated ones, parallel to the Widmanstatten figures, or in other
words, the octahedral structure of the meteorite. This can well be
discerned by a study of Plate III. Another inclusion of an interesting
character found in one of the sections was a nodule about one centi-
*Wiener Sammlung, 1895, p. 275.

Mar., 1905. The Rodeo Meteorite — Farrington. 3

meter in diameter, of a black, amorphous, friable substance resem-
bling graphite. The form of the nodule in the direction of the section
is nearly circular, but in the third dimension its extent is unknown,
as it penetrates into the main body of the meteorite, which has not
yet been cut. No band of swathing kamacite surrounds the nodule,
it being set bodily into the mass of the iron. In appearance and
physical properties the substance of the nodule resembles graphite
fully, but it is magnetic and fuses in the reducing flame at about 4.
Mixed with potassium nitrate it deflagrates readily, but throws out
incandescent sparks in addition to the flaming usual to graphite.
Potassium carbonate results from the reaction. Oxidation with sul-
phuric and chromic acids according to the French method* affords an
appreciable quantity of CO 2. On heating in oxygen the substance
glows and becomes of a red-brown color. It was found to be little, if
any, attacked by the ordinary acids. After a long treatment with aqua
regia, however, and addition of ammonia to the solution, a slight pre-
cipitate of iron hydroxide was obtained. When powdered and added to
a copper sulphate solution, copper was reduced by the substance. Its
specific gravity (obtained by Thoulet's solution) was 2.38. -On account
of the above properties it would appear that the substance is chiefly
graphite, but contains in addition some form of iron, probably a
carbide, intimately mixed with it. Such a mixture should exhibit
the properties of magnetism, reduction of copper from copper sul-
phate and insolubility in acids, which are possessed by this substance.
Such properties seem not to have been possessed by graphite which
has been described from other meteorites. It is common for meteoric
graphite to be accompanied by troilite, as has been noted by Smithf
and other authors. The Rodeo graphite, however, seems to contain no
troilite. A specimen labelled graphite in the Museum collection of what
is probably a portion of a "salamander" from the Isabella Furnace,
Etna, Pennsylvania, was found to exhibit properties much like those
of the Rodeo graphite except that it is heavier, the specific gravity
being 5.56, and the structure is foliated rather than compact.
Since a "salamander" originates by the accumulation of graphite
and graphitic substances in smelting operations, it seems reasonable
to suppose that the Rodeo nodule is a similar segregation of graphite
originally more or less disseminated in the iron.

An analysis of the meteorite was made by Mr. H. W. Nichols,
of the Department of Geology of the Museum. Material for analysis
was secured by drilling a half-inch hole to a depth of seven-eighths

* Blair, The Chemical Analysis of Iron, third edition, p. 136.

t Am. Jour. Sci. 3, 2, p. 304, 433.

4 Field Columbian Museum — Geology, Vol. III.

of an inch, and rejecting the drillings from the crust portion. For
the determination of iron, nickel, and cobalt, a portion of 1.3733
grams was dissolved in strong hydrochloric acid. Solution took
place rapidly and completely, only a few unweighable black flecks
being left after oxidation with nitric acid, evaporation to dryness
and addition of water. Iron was precipitated three times by treat-
ment with ammonia and ammonium chloride as directed by Frese-
nius, except that a large excess of the reagents was used. After
solution with sulphuric acid and reduction with hydrogen sulphide,
the determination was made by the usual titration with potassium
permanganate. Copper was precipitated by hydrogen sulphide from
the filtrate from the iron precipitate and then determined electro-
lytically. Nickel and cobalt were separated in acetic acid solution
as sulphides and separated by potassium nitrite. Nickel was then
determined electrolytically, but cobalt as sulphate, the electro-
lytic determination of this element having proven at times unreliable.
Manganese was tested for in a portion of 2.8248 grams dissolved in
nitric acid and oxidized with potassium chlorate according to Ford's
method. No precipitate was obtained. The treatment with nitric
acid showed the meteorite to be passive until water was added.
Sulphur and phosphorus were determined in a portion of 4.8321
grams dissolved in fuming nitric acid by the slow addition of hydro-
chloric acid. From this sulphur was precipitated as directed by
Blair when iron is present, purified by fusion with sodium carbonate,
and weighed as barium sulphate. Phosphorus was determined by
the acetate method and weighed as magnesium pyrophosphate.
Carbon was determined in a portion of 2.5678 grams by oxidation
with chromic and sulphuric acids and weighed as carbon dioxide.
During the treatment the odor of hydrocarbons was observed, similar
to that obtained in the solution of pig iron. This indicated that
some of the carbon was present in a combined form, while an insoluble
residue showed that some existed as graphite. The analysis gave the
following results:

Fe 89.84

Ni 8.79

Co 0.28

Cu 0.07

P 0.80

S 0.02

C 0.09

99.89
The composition of the meteorite is thus seen to be that usual to
medium octahedrites, with a high percentage of phosphorus. From

Mar., 1905. The Rodeo Meteorite — Farrington. 5

the large amount of schreibersite visible in the sections, such a con-
tent of phosphorus would be expected.

Including Rodeo, the meteorites now recognized from the State
of Durango are, with dates of their fall or find, as follows:

Avilez, Spherulitic chondrite, Cc 1856

Bella Roca, Fine octahedrite, Of 1888

Cacaria, Hammond octahedrite, Oh 1867

Rodeo, Medium octahedrite, Om 1852

Rancho de la Pila, Medium octahedrite, Om 1804

San Francisco del Mezquital, Siratik ataxite. Ds 1868

The localities of these have been determined as accurately as
possible by the writer from various published accounts, and are
represented as determined on the accompanying map (Plate IV.).
Of these meteorites only one, Avilez, is a stone; the others are all
irons. Of the irons, Cacaria and San Francisco del Mezquital are
sufficiently distinguished by their structure, Cacaria being a Ham-
mond octahedrite and San Francisco del Mezquital an ataxite.
Rodeo and Rancho de la Pila are both medium octahedrites, but the
localities from which they come are about seventy miles apart.
Hence only Bella Roca needs to be compared with Rodeo in order to
determine whether it belongs to the same fall. The localities of
Rodeo and Santiago Papasquiaro, near which Bella Roca is said to
have been found, are in a direct line about forty miles apart. This
is much farther than parts of a single meteor could have been natu-
rally distributed according to our present knowledge. That they
might have been separated by human agency is possible, but not
probable, since the country between these localities is thinly settled
and difficult to travel over. From the appearance of the surface of
the Bella Roca meteorite Brezina concluded* that it could have
lain exposed but a little while. The Rodeo meteorite, however,
is known as far back as 1852. Brezina also describes Bella Roca
as a highly oriented individual showing an almost complete fusion
crust. Neither of these observations would apply to the Rodeo
iron. From the point of view of structure the two irons do not
differ essentially. The appearance of the lamellae is indeed quite
similar, with the exception that Rodeo has a preponderance of the
wider lamellae, sufficient, in the author's opinion, to warrant class-
ing it as a medium octahedrite. Bella Roca is classed as a fine
octahedrite by Brezina. There is also a similarity between the
two irons in the fact that the schreibersite inclusions follow the
octahedral lamellae, in their orientation. The schreibersite in Rodeo,

* Wiener Sammlung, 1895, p. 271.

6 Field Columbian Museum — Geology, Vol. III.

however, is considerably stouter in habit than that in Bella Roca.
The chief point of difference between the two irons, however, and
one which in the writer's view seems alone to warrant their separa-
tion, is that in Rodeo there is an entire lack of the inclusions of
troilite which form so striking and important a feature of the compo-
sition of Bella Roca. Although eight full-sized sections have been
made of Rodeo, no troilite has as yet been observed in it. In Bella
Roca, however, as is well known, troilite is an abundant and charac-
teristic constituent. The chemical analyses of the meteorites do not
show important differences, but this would not be expected as between
medium and fine octahedrites. Still the analyses show a relative
absence of sulphur and hence of troilite, and abundance of phos-
phorus and hence of schreibersite in Rodeo, while the opposite condi-
tion holds in Bella Roca. The analyses compare as follows, that of
Bella Roca being by Whitfield:*

Fe Ni Co Cu P S C

Bella Roca. . 91.48 7.92 0.22 0.21 0.21 0.06 = 100.10

Rodeo 89.84 8.79 0.28 0.07 0.80 0.02 0.09 = 99.89

In view, therefore, of the distance between the localities and the
difference in structure and composition, there seems to be sufficient
reason for regarding Rodeo as a distinct fall.

*Am. Jour. Sci. 3, 37, p. 439.

FIELD COLUMBIAN MUSEUM.

GEOLOGY, VOL. Ill, PLATE I.

The Rodeo Meteorite, x y2. The incision and smoothed surface in the upper right-
hand corner are of artificial origin. The remainder of the surface presents the natural
appearance.

UNIVERSITY OP ILL!

FIELD COLUMBIAN MUSEUM.

GEOLOGY, VOL. Ill, PLATE II.

The Rodeo Meteorite, x l/2. The smoothed surface at the left is
due to the use of the meteorite as an anvil.

LIBKmKV
UNIVERSilY 01- ItLlWV

v.::.\

LIBHAKV

UNIVERSITY Of illwuiS

.....

LIBRARY

universi1y of illinois
urda;:a

..s)IH

. 3 1

ts C C'-