SMITHSONIAN CONTRIBUTIONS TO THE
EARTH SCIENCES
NUMBER 5
Roy s.ciarke, jr., The Allende, Mexico,
Eugene larosewich, Brian Mason. ^?- , ? o i
JosephNelen.ManuelGomez, Meteorite Shower
and Jack R. Hyde
SMITHSONIAN INSTITUTION PRESS
CITY OF WASHINGTON
I97O
ABSTRACT
Clarke, Roy S., Jr., Eugene Jarosewich, Brian Mason, Joseph Nelen, Manuel
G6mez, and Jack R. Hyde. The Allende, Mexico, Meteorite Shower. Smithsonian
Contributions to the Earth Sciences 5:1-53. 1970.?The Allende meteorite fell near
Parral, Chihuahua, Mexico, between 0105 and 0110 Central Standard Time on
Saturday, 8 February 1969. The fireball approached from the south-southwest
(S37?W), and broke up in the atmosphere, producing thousands of fusion-crusted
meteoritic stones. The smallest individuals were recovered 4 km east of Rancho
Polanco (26?43'N, 105?28/W), and the largest near Rancho El Cairo (27?06/N,
105?12/W), some 50 km to the north-northeast across the Parral-Jimin^z highway.
Specimen size increases generally as one moves to the north-northeast within the
field, and many large specimens (5-15 kg) were recovered in and around the area
enclosed by Pueblito de Allende, San Juan, Rancho Blanco, and Santa Ana. At
least two tons of meteoritic stones have been recovered, with crusted individuals
ranging in weight from approximately 1 g to one individual of 110 kg. Specimen
shapes are mainly fragmental, due to one major disruption of the parent body,
followed by minor subsequent fragmentation. Individual stones have primary and
secondary fusion crust, and some fresh fracture surfaces due to late-stage breaking.
A small percentage of stones shows strong ablative shaping due to oriented flight.
The elongate strewnfield possibly exceeds 300 km2 in area, making Allende the
largest recorded stony meteorite fall both in its areal extent and in total weight
of recovered meteorites. Allende fell near the sites of find of two major iron
meteorites, Morito and Chupaderos.
Chemical and mineralogical compositions establish that Allende is a Type III
carbonaceous chondrite. Three distinct components can be recognized: fine-
grained black matrix (~60%), chondrules (~30%), and irregular white ag-
gregates (^10%). The matrix consists almost entirely of iron-rich olivine
(average 50% Fe2SiO4), with minor amounts of troilite, pentlandite, and taenite,
rendered opaque by dispersed carbonaceous material. Most of the chondrules are
magnesium-rich, and consist of olivine (average 9% Fe2SiO4) with minor amounts
of clinoenstatite and some glass; a few chondrules are rich in calcium and alumi-
num, and are made up largely of anorthite, gehlenite, augite, and spinel. The
irregular aggregates are also rich in calcium and aluminum, and contain anorthite,
gehlenite, augite, spinel, nepheline, grossular, and sodalite (the last two minerals
have not previously been recorded from meteorites). Complete chemical analyses
have been made of the bulk meteorite, a dark inclusion, the matrix, a chondrule
concentrate, two individual chondrules, and a single aggregate.
Official publication date is handstamped in a limited number of initial copies and is recorded
in the Institution's annual report, Smithsonian Year.
UNITED STATES GOVERNMENT PRINTING OFFICEWASHINGTON : 1970
For sale by the Superintendent of Documents, U.S. Government Printing OfficeWashington, D. C. 20402 - Price $1.25
Contents .
Page
Abstract ? ? "
Acknowledgments 1
Introduction -. 1
Specimen recovery and distribution for study 2
Published reports ? 2
Proposed name : 2
Strewnfield '. ? 2
' Distribution of specimens 6
Recovery of individual stones 9
Other meteorites from the Parral-Jim?nez area. 15
Morphology ...; 1?
Mineralogy and petrography 32
Fusion crust ' 42
Chemical composition 44
Comparative chemical data 47
Discussion 47
Literature cited 52
Tables
Page
1. Minerals in the Allende meteorite 33
2. Microprobe analysis of gehlenite, fassaite, and anorthite in a large chondrule 39
3. Analytical data on the Allende meteorite 45
4. Trace element data on the Allende meteorite 47
5. Comparison of chemical data on the Allende meteorite 50
6. The Type III carbonaceous chondrites 51
Illustrations
Page
1. Location of Allende meteorite strewnfield 3
2. Plan of the Allende meteorite strewnfield 4
3. Field recovery by Brian Mason of a 12-kg Allende specimen (NMNH 3527) near
San Juan 8
4. An Allende individual (NMNH 3491) found in place near Rancho Santa Ana 10
5. An Allende individual found by Manuel G6mez near Rancho Santa Ana
(NMNH 3493) 11
6. An Allende stone found partially embedded at base of cactus plant (NMNH 3494) 12
7. An Allende individual (NMNH 3502) found by local boy near Rancho Blanco 13
8. An Allende individual found near schoolhouse in San Juan (NMNH 3517) 14
9. Location of meteorites of the Parral-Jim6nez area 16
10. A blocky, crusted Allende individual (NMNH 3523) found between Pueblito de
Allende and San Juan 18
11. An Allende individual (NMNH 3500) found near Rancho Blanco 19
12. An Allende individual (NMNH 3501) found near Rancho Blanco 20
13. An Allende stone (NMNH 3494) found at base of cactus near Rancho Santa Ana 20
14. An unusually large, oriented Allende individual (NMNH 3527) found near San Juan.. 21
15. An aerodynamically shaped Allende individual found near Torreon de Mata
(NMNH 3522) 22
iii
iv
Page
16. An aerodynamically shaped Allende stone (NMNH 4018) found near the Ceniceroses.. 23
17. An aerodynamically shaped Allende individual found near Cienega de Ceniceros
de Arriba (NMNH 4083) 24
18. An aerodynamically shaped Allende individual found near Cienega de
Abajo (NMNH 4084) 25
19. An aerodynamically shaped Allende individual found near the Ceniceroses
(NMNH 4157) 26
20. An unusually large Allende stone (NMNH 4006) found near Cienega de
Ceniceros de Abajo 27
21. Lustrous fusion crust on an Allende stone (NMNH 3812) found near the Ceniceroses.. 28
22. Lustrous fusion crust on an Allende stone (NMNH 3813) found near the Ceniceroses.. 29
23. Lustrous fusion crust on an Allende stone (NMNH 3814) found near the Ceniceroses.. 30
24. Broken surfaces of an Allende fragment (NMNH 3529) 31
25. Polished surface of an Allende slice (NMNH 3529) 32
26. Electron micrographs of Allende meteorite matrix 34
27. Composition of individual olivine grains in Allende chondrules 35
28. Irregular aggregates and chondrules in black matrix 36
29. Allende meteorite chondrule of prismatic olivine crystals 37
30. Allende meteorite barred olivine chondrule 37
31. Three juxtaposed barred olivine chondrules 38
32. Allende meteorite granular olivine chondrule 39
33. Gehlenite-anorthite-fassaite-spinel chondrule 40
34. Anorthite-forsterite-spinel chondrule 41
35. A mosaic of photomicrographs of lustrous fusion crust (NMNH 3811) 43
36. SiOa plotted against MgO for chemical analyses of chondrites 48
Roy S.Clarke, Jr.,
Eugene Jarosewich, Brian Mason,
Joseph Nelen, Manuel Gomez,
and Jack R. Hyde
The Allende, Mexico,
Meteorite Shower
Introduction
In the early morning of Saturday, 8 February 1969
(between 0105 and 0110 Central Standard Time), a
brilliant fireball was observed over much of north-
ern Mexico and adjacent areas of Texas and New
Mexico. The most spectacular phenomena were
centered around the city of Hidalgo del Parral in
the south-central part of the state of Chihuahua.
The fireball approached from the south-southwest,
and as it neared its terminal point the brilliant
light was accompanied by tremendous detonations
and a strong air blast. Thousands of individual
meteoritic stones rained down over a large area of
rural Mexico. One weighing 15 kg fell within four
meters of a house in the town of Pueblito de Al-
lende, 35 km east of Parral. This stone was broken
up, and pieces taken to the office of the newspaper
"El Correo de Parral" the same day; the news of an
important meteorite fall was published that eve-
ning. Dr. E. A. King of the NASA Manned Space-
craft Center in Houston, Texas, heard of these
events on the morning of 10 February 1969 and de-
parted promptly for the scene. He returned with
meteoritic material on the morning of 12 February,
and he and his colleagues have published a prelimi-
nary report on their findings (King et al., 1969).
Our initial field investigation was undertaken
after conferring by telephone with Ing. Diego
Cordoba, Director del Instituto de Geologia, Mexico
City. Roy S. Clarke and Brian Mason arrived in
Parral on Wednesday, 12 February and spent the
next five days investigating the fall and collecting
Roy S. Clarke, Jr., Eugene Jarosewich, Brian Mason, and
Joseph Nelen, Department of Mineral Sciences, National Mu-
seum of Natural History, Smithsonian Institution, Washing-
ton, D.C. 20560. Manuel Gdmez, ASARCO Mexicana SA.,
Parral, Chihuahua, Mexico, and Jack R. Hyde, Smithsonian
Institution Meteorite Recovery Project, Lincoln, Nebraska
68508.
material for research and preservation. Ing. Manuel
G6mez of ASARCO Mexicana S. A., Parral, and Ing.
Ce'sar Gallardo of the University of Chihuahua
joined the field party and contributed importantly
to its success. Clarke returned to the field 24-27 May
with Jack R. Hyde of the Smithsonian Institution's
Meteorite Recovery Project, Lincoln, Nebraska.
Hyde remained in the area for several weeks con-
Acknowledgments
We are indebted to Ing. Diego Cordoba, Di-
rector del Instituto de Geologia, Mexico City,
for permission to investigate the fall area and
to collect material for scientific investigation;
to R. C. Byrd, Superintendent, ASARCO Mex-
icana, Parral, for providing facilities for our
work; to R. E. McCrosky and Gunther Sch-
wartz of the Smithsonian Institution Astro-
physical Observatory for field assistance and
informative discussion and interpretation of
phenomena of fall observations. Our co-work-
ers within the Department of Mineral Sciences,
National Museum of Natural History, have
contributed to this work in various ways: John
S. White, Jr. prepared many X-ray powder
photographs; E. P. Henderson prepared photo-
micrographs and provided helpful discussion;
Grover C. Moreland prepared critical sections
for study; Sarah W. Scott cataloged field collec-
tions and assisted in preparation of the manu-
script; and William G. Melson critically re-
viewed the manuscript and provided helpful
discussion. We are indebted to the National
Aeronautics and Space Administration for
grants that covered part of the costs of this
investigation.
SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
ducting studies of specimen distribution, and he
returned to the field for an additional three weeks
in late October and November.
SPECIMEN RECOVERY AND DISTRIBUTION FOR STUDY.?
Information related to the fall and recovery of the
Allende meteorite was disseminated rapidly to in-
terested scientists throughout the world by the
Smithsonian Institution's Center for Short-Lived
Phenomena, Cambridge, Massachusetts. Specimen
material reached the Division of Meteorites, Na-
tional Museum of Natural History, Washington, on
19 February, and its distribution to interested scien-
tists began immediately. All reasonable requests
were granted; scientists in thirty-seven laboratories
in thirteen countries were sent material within a
few weeks of the date of fall. Dr. E. A. King's group
in Houston had also distributed material to other
research laboratories.
Many individual meteorite specimens have been
recovered from the strewnfield, and Allende ma-
terial is now widely distributed in collections and
research laboratories. We obtained about 150 kg of
meteorites on our first trip into the field and smaller
amounts on the second and third trips, and we were
shown many large specimens that we were unable
to acquire. Several other scientific groups from in-
stitutions in Mexico and the United States are
known to have visited the area and acquired ma-
terial, as has one Canadian group. Commercial and
private collectors have been active in the area and
have removed indeterminate amounts of specimen
material. One dealer reported handling approxi-
mately 500 kg of Allende meteorites on a wholesale
basis by early November 1969. An aggressive ama-
teur collector visited the field in late February and
obtained 110 kg of excellent material for his collec-
tion. Information of this type leads us to conclude
that during the first five months after the fall two
tons of meteorite specimens were removed from the
strewnfield. The recovery of this large quantity of
specimen material establishes the Allende meteorite
as the largest stony meteorite shower on record.
PUBLISHED REPORTS.?The immediate and broad
distribution of Allende meteorite specimens has re-
sulted in the prompt publication of several studies
that are not referred to directly in this paper. Fire-
man (1969) published a preliminary account on the
field recovery and laboratory study of Allende. A
report on the atmospheric collection of meteoritic
debris associated with the Allende fireball has been
prepared by Carr (1970). The study of a number
of radionuclides in Allende by gamma-ray spectrom-
etry has been reported by Rancitelli et al. (1969).
Fireman et al. (1970) have reported several types of
ages for Allende based on determination of rare
gases^ K, U and Th, and Durrani and Christodou-
lides (1969) have used thermoluminescence as a
means of estimating an exposure age for Allende.
PROPOSED NAME.?We propose the name "Allende"
for this meteorite, although "Pueblito de Allende"
was used in the preliminary report of King et al.
(1969), because this town was the location of the
first stone known to have been recovered. The ex-
tent of the strewnfield was not fully realized at that
early date, and it is now clear that the name Allende
has an association with a large area in the heart of
the strewnfield. Specimens have been recovered
within a kilometer of the historic and beautiful
town of Valle de Allende, from along a 15-km seg-
ment of the valley of the Rio del Valle de Allende,
as well as in the environs of Pueblito de Allende.
Hey (1966) comments in the introduction to his
standard Catalogue of Meteorites that "meteorite
names are essentially labels, of no intrinsic signif-
icance." Labels should be simple and concise, and
the name Allende satisfies these criteria admirably.
The name honors Ignacio Jose* Allende (1779-1811),
a Mexican army officer and patriot of the uprising of
1810-1811.
Strewnfield
The Allende fireball approached from the south-
southwest, the azimuth of the radiant (direction of
approach) being approximately 215? and its alti-
tude 13? (preliminary values, McCrosky et al.,
1969). The resulting meteorite strewnfield is located
540 km (340 statute miles) south (S10?E) of El
Paso, Texas, and 200 km south (S20?E) of the city
of Chihuahua, to the east of the Sierra Madre Oc-
cidental (Figure 1). Its head lies within and some-
what to the east of a small group of hills, the Sierra
de Almoloya, just north of the Interamerican High-
way (Route 45), halfway between the cities of
Parral and Jimenez. Small individual meteorite
specimens from the tail of the field have been re-
covered as far south as 4 km east of Rancho Polanco,
26?43'N and 105?28/W (Figure 2). Approximately
50 km to the north-northeast a large individual
NUMBER 5
ARIZONA !
!
TUCSON |
. y.s__ j
MEXICO
?ROSWELL
NEW MEXICO
? LAS CRUCES ? ABILENE
. 1^ E
I JUAREZ f\
\
\
\
SAN ANTONIO
TORREON
ALLENDE MEXICO METEORITE STREWNFIELD
?. DIRECTION of FIREBALL
KILOMETERS
FIGURE 1.?Location of the Allende meteorite strewnfield, with direction of fireball approach indicated.
stone of approximately 110 kg, broken into many
fragments on impact, was discovered at 27?O6'N and
105?12'W, just to the east of the Sierra de Almoloya.
The specimen-rich Allende meteorite recovery
area, 50 km in length and possibly exceeding 300
km2 in area, is the largest stony meteorite strewn-
field that has been studied in any detail.1 The map
1 The recent fall of the Belfast (Bovedy), Northern Ireland,
meteorite resulted in the recovery of two stones separated by
a distance of 62 km (Meighan and Doughty, 1969), and the
suggestion that the main mass continued on for 150 km into
the Atlantic Ocean (Hindley and Miles, 1970). Kulik (1922)
reported that stones from the Saratov, Russia, meteorite fall
were found over a distance of 130 km, but details are lacking.
Nininger (1936) reported that specimens of the Pasamonte,
New Mexico, meteorite shower were separated by a distance
of 28 miles (45 km).
presented here (Figure 2) 2 is based largely on our
own field investigations and may not indicate the
full extent of the field. We feel, however, that the
general shape of the field and the nature of the
distribution of samples within it are fairly well
understood at this time. Large individual specimens
may still be found, extending the field to the north-
east and perhaps modifying the shape of that end
2 The maps presented here (Figures 2 and 9) are based on
1:250,000 sheets compiled in 1964. They were prepared as a
collaborative effort of the Departmento Cartografico Militar
de Mexico, the Inter-American Geodetic Survey, and the Army
Map Service of the United States, based on aerial photographs
taken in 1957-1958. They are available from the Mexican
Government. The two sheets used were:
Hidalgo del Parral, Chihuahua. NG13-5, series F501.
Ciudad Camargo, Chihuahua. NG13-2, series F501.
105-30' 105? 15'
ALLENDE METEORITE STREWNFIELD
CHIHUAHUA, MEXICO
0 12 3 4 5
A METEORITE SPECIMEN SCALE Kilometers
? RANCHO SAN AGUSTIN
HACIENDA OJODE LA SIERRA DEALMOLOYA /
ESTACION TROYA
RANCHO SAN ISIDRO
/RANCHO EL CAIRO
PORVENIR
RANCHO. LAS CAMELIAS
E JIDO LA PORRENA
? SAN JUAN /
/ l2llfl .RANCHO BLANCO
UEBLITO DE ALLENDE
V
SANTA ANA
RANCHO SANTA ISABEL
pRANCHO SAN ANTONIO
ALLE DE ALLENDE
EJ.DOLA PALMA. ^.^NCHO f8 DEUCIA8
LA MAGDALENA.V\. "RANCHO SAN JOSE
CIENEGA DE CENICEROS DE ARRIBA. .,
^^ Jf .v/'-^?CIENEGA DE CENICEROS DE ABA JO
?RANCHO LOS OJUELOS
w-
EJIDO VALSEQUILLO
N37?E N35?E N32?E
IO5?3O'
FIGURE 2.?Plan of the Allende meteorite strewnfield.
of the field somewhat. We do not expect subsequent
work to require major changes in the shape of the
field south of the Rio del Valle de Allende.
The map (Figure 2) shows the outline of the
strewnfield and its relation to major geographic
features and habitations. The strewnfield bound-
aries were arrived at largely on the basis of inter-
views with many local residents who had searched
NUMBER 5
for specimens or who had knowledge of where they
were recovered. Specimen collecting by the local
people started immediately after the fall and con-
tinued with more or less enthusiasm for months.
The intensity of this activity depended somewhat on
the interest stimulated at a given period by out-
siders and by the varying labor requirements of an
agrarian economy. Meteorite collecting proved to
be an interesting and profitable leisure-time activity
for a few individuals in each community, and these
collectors were a particularly reliable source of in-
formation needed to reconstruct the specimen dis-
tribution pattern. All of the habitations indicated
on the map (Figure 2) were visited; the important
ones on many occasions, and most at least several
times over a period of weeks or months.
There is an unavoidable vagueness in much of the
information obtained from local residents. Indi-
viduals were mainly open and frank in their reports
to us, but their meteorite finds had not been docu-
mented with scientific precision. Specimens were
collected over a period and kept together at home in
a box, bag, or sack. Plastic bags which had previ-
ously been used for candy or meal and had not been
well cleaned were a favorite storage method for col-
lections of small specimens. Normally, a given in-
dividual collected from a relatively small area with-
in the strewnfield. Collectors generally had sharp
recollections of the location where unusual speci-
mens or particularly large ones were found, but it
was impossible to revisit each site. Distances and
directions are also subject to question, our practice
being to record reports as we received them. Land-
marks are scarce in the chaparral country, possibly
introducing another uncertainty. Occasionally,
claftns were made that seemed to be at variance with
the developing pattern, and these reports were re-
jected unless corroborative evidence could be found.
Generally speaking, we would receive numerous re-
ports of meteorites being found in a given area, and
equally numerous reports of their absence from
other areas. In particularly critical areas, we con-
ducted limited searches, normally with the assistance
of local residents.
The degree of confidence we have in the bound-
aries developed is indicated on the map (Figure 2)
by use of either a solid or broken line. The area of
most uncertainty is north of the highway, where the
two largest specimens have been found. Does the
strewnfield really extend further to the northeast,
perhaps even beyond the railroad? The area south of
the highway between Rancho Las Camelias and the
town of Salaices seems also to be in question, speci-
mens so far not having been reported from there.
An axis of the strewnfield based on specimen re-
covery was drawn by joining the point where small
specimens were found in the extreme southwest
with the locations of the largest specimen (Figure
2). This line, N32?E, is in good agreement with the
azimuth of 215? (N35?E) calculated by McCrosky
et al. (1969) from fireball observations. The N35?E
line is also indicated on the map, assuming the loca-
tion of the largest specimen to be most representa-
tive of the fireball trajectory. It is interesting to note
that the three specimens found north of the high-
way, separated by a maximum distance of 7 km, plot
closely to both of these lines. The shape of the
strewnfield and the distribution of specimens within
it (discussed below) suggest that the N35?E line ap-
proximates the axis of the strewnfield corrected for
preferential scattering of lighter specimens due to
high-altitude winds. McCrosky et al. (1969) re-
ported that data for the time of fall supplied by the
United States Air Force indicated winds "predomi-
nantly from the west up to an altitude of 30 km
except for the immediate surface layer where north-
east winds may have been occurring." These west
winds would be consistent with an eastward skewing
of the tail of the strewnfield, a situation that seems
to be borne out by the observed distribution of
specimens.
The outline of the strewnfield might be some-
what easier to understand if one assumed that the
actual trajectory of the fireball was from a slightly
more westerly direction than the N35?E bearing of
McCrosky et al. (1969) . Justification for such an
assumption can be derived from considerations of
three key specimens from accurately known loca-
tions within the strewnfield. The 110 kg specimen
from the head of the field has been mentioned above
and its location is indicated in Figure 2. A 40-50
kg individual was found within the Sierra de Al-
moloya 5 km to the southwest of the largest speci-
men (Figure 2, and discussed below). The third
specimen was found 12 km to the southwest of the
110 kg individual (NMNH 3527, discussed below),
near San Juan (Figure 2). This specimen is a 12 kg
aerodynamically oriented individual, and it seems
reasonable to assume that a specimen of this type
would fall with relatively little deflection during
SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
passage through the atmosphere, falling close to
the fireball's actual trajectory. Similar assumptions
for the other two specimens based on their size also
seem reasonable. These three specimens lie on a
straight line with a bearing of N37?E, a bearing that
fits well the configuration of the strewnfield. The
asymmetry of the field along this line can be reason-
ably explained as due to high-altitude winds. The
light material that fell in the tail of the field may
have been moved to the east by 4 km or more, while
the eastern border of the main body of the field
may have been skewed eastward by approximately
2 km. This picture seems reasonable to us on the
basis of our field observations and study of specimen
distribution (discussed below).
The character and use of the land within the
strewnfield is largely controlled by elevation and
drainage patterns. The area is traversed by the
drainage systems of the Rio del Valle de Allende
in the north and the Rio Primero in the south. The
Rio Primero is an intermittent stream, dry for most
of the year. Habitation is concentrated along these
streams at the borders of areas of intense agricul-
tural use. Agriculture and habitation account for
10-15 percent of the land use within the strewnfield,
and there is some mining in the area. With only a
slight rise in elevation away from these streams,
the land changes to brush-covered range country
known as chaparral. Much of the chaparral is rela-
tively open, but it does include areas of dense brush,
penetrable only with difficulty. To the southwest
from the Rio del Valle de Allende, the country
generally becomes more open and drier.
Changes of elevation within the strewnfield are
generally gradual with the marked exception of the
Sierra de Almoloya (2,044 meters maximum eleva-
tion) . Elevations tend to increase to the south-
west and range from 1,500 to 1,750 meters. The Rio
del Valle de Allende meanders across the upper
part of the field for a distance of almost 20 km, ele-
vations along the river ranging from 1,550 meters
east of San Juan and Rancho Blanco to near 1,600
meters just east of Valle de Allende. The elevation
of the Rio Primero where it crosses the tail of the
field near the twin villages of Ceniceros is 1,650
meters. From here elevations increase gradually to
the southwest, reaching 1,750 meters just east of
Rancho Polanco.
DISTRIBUTION OF SPECIMENS.?The weight distribu-
tion of individual, crusted Allende meteorite speci-
mens within the strewnfield approximates the pat-
tern that would be expected, assuming one major
disruption of the parent meteorite followed by lim-
ited further breakup of smaller fragments during
high-velocity flight. Atmospheric drag combined
with the low angle of entry produced unusually ef-
fective size sorting over a long strewnfield. This is
clearly demonstrated by the observed distribution of
specimens on the ground and is further supported
by morphological observations that will be discussed
in a following section. Several of the largest individ-
ual pieces were apparently relatively undeflected
in their flight through the atmosphere and traveled
the greatest distance to the northeast after initial
breakup, well beyond the bulk of specimens. Large-
to medium-size fragments traveled less far and were
apparently deflected by the atmosphere to an extent
depending largely upon their individual shapes.
This resulted in a broadening of the strewnfield in
the area that produced most of the large specimens.
Smaller specimens were retarded more promptly by
atmospheric drag?falling many kilometers behind
the largest individuals?and were not deflected far
from the fireball's trajectory. This summary is, of
course, an oversimplification, but it furnishes a
useful framework for discussion of our field obser-
vations.
The largest specimen known to us is a 100-110 kg
fragmented individual which was discovered eight
months after the fall. It was found approximately
1.5 km N15?E of Rancho El Cairo (Figure 2), in
and around a small hole. The second largest stone
was found in early April in the Sierra de Almoloya
4 km north of Route 45 at a point north of Rancho
Las Camelias. This was a 40-50-kg individual that
also fragmented on impact. We know of only one
other stone recovered north of the highway, a 6-kg
individual. Two large stones were recovered south
of the highway and within a few hundred meters of
it, a 35-kg individual recovered northeast of San
Juan (NMNH 3529, broken on impact) and a 17-kg
individual recovered directly north of San Juan
(NMNH 3509, broken on impact). A number of
specimens in the 5-20-kg size range have been re-
covered around San Juan and Rancho Blanco and
in the area extending southeast from Rancho Blanco
for several kilometers. A somewhat isolated 18 kg
individual was recovered in range country 7 km
south of Rancho Blanco (Donald P. Elston, per-
sonal communication, 1969), and several other large
NUMBER 5
individuals (5 to 11 kg) have been recovered 2 to 3
km to the north and northwest of this 18 kg in-
dividual. Many other large individuals have been
found in the agricultural area enclosed by the four
locations Pueblito de Allende, San Juan, Rancho
Blanco, and Santa Ana. The largest individual
that we know of that landed without damage was
recovered at the edge of the river between Pueblito
de Allende and Santa Ana (NMNH 3528, 18.5 kg).
A 12-kg individual (Figures 3 and 14) was recovered
in a freshly plowed field less than one kilometer
south of San Juan (NMNH 3527, found by B.
Mason on 17 February 1969) . Small specimens have
been found interspersed among the large specimens
at the north end of the field, but in very limited
numbers. An example is a 26 g completely crusted
individual that was found within 2 to 3 km of
Rancho Blanco (NMNH 4651). The range country
south from Rancho Blanco and Santa Ana and ex-
tending west toward Rancho San Antonio produced
many medium-size specimens (1 to 5 kg) and oc-
casional larger ones. Specimens in this size range
were also found as far south as several kilometers
southeast from Valle de Allende, and 1-kg speci-
mens have been found as far south as "Rancho
(ruinas)."
South from Rancho (ruinas) and 10 km south
of Valle de Allende is the comparatively isolated
village of Torre6n de Mata (population, approxi-
mately 50). This village has served as an important
collecting point for the south-central region of the
strewnfield. Specimens were recovered here immedi-
ately after the fall, but the local people do not ap-
pear to have been greatly impressed by the abun-
dance of valuable meteorite material close at hand.
Specimens were found within the village and on
the farmland nearby, and in the range country to
both the north and south. Some of the more en-
thusiastic residents of villages to the north walked
or rode horseback from the vicinities of Pueblito de
Allende and Valle de Allende to Torre6n de Mata
searching for specimens. This activity undoubtedly
attracted attention, but the local people seem to
have felt that the small specimens found in their
area were of little value compared to the com-
mercially attractive large specimens found to the
northeast. Five kilometers farther to the southwest
are the twin villages of Cienega de Ceniceros de
Arriba and Cienega de Ceniceros de Abajo (com-
bined population, approximately 500). They served
as collecting centers for hundreds of small speci-
mens recovered from the tail of the strewnfield.
Little collecting was done in this area prior to
the visits of Jack R. Hyde in June 1969. These
visits stimulated searches that led to the recovery
of an important research collection and to extend-
ing the strewnfield as far south as several kilometers
east of Rancho Polanco.
The general trend of decreasing specimen size
that was observed for the head and center of the
strewnfield holds for the tail. This is demonstrated
by calculating average specimen size for collections
made (1) around Torre6n de Mata, (2) the Ceni-
ceroses, and (3) in the extreme tail of the field.
Collections made around Torre6n de Mata nat-
urally include material collected toward the Ceni-
ceroses, and the converse is true for collections
obtained in the Ceniceroses. Torre6n de Mata ma-
terial, however, has a strong bias for areas to the
north and northeast, while material from the Ceni-
ceroses would not normally have been collected as
far north as Torre6n de Mata. The specimens from
the tail of the strewnfield were collected at least
5 km south of the Ceniceroses and on to the end
of the field. The approximately 200 specimens
known to have been collected in the Torre6n de
Mata area range in weight from 6 to 537 g, with an
average weight of 87 g. A somewhat larger collec-
tion of specimens known to have been collected
around the Ceniceroses range in weight from 2 to
540 g, with an average weight of 32 g. The small
group of specimens (17) from the tail of the field
ranged in weight from 3.1 g (completely crusted
individual) to 43 g, with an average weight of 12 g.
Individual specimen weight on the average decreases
as one moves to the southwest within the strewn-
field.
It is tempting to try and estimate the amount of
Allende meteorite material that reached the ground.
We pointed out above that an estimated two tons
of material was removed from the strewnfield during
the first few months after its fall. What fraction of
the material that actually landed does this repre-
sent? An important factor to consider is the effec-
tiveness of search and recovery operations within
the strewnfield. The areas of maximum concentra-
tion of large specimens fortuitously and fortunately
coincided with or were contiguous to areas of maxi-
FIGURE 3.?The field recovery by Brian Mason of a 12-kg Allende specimen (NMNH 3527) 17
February 1969, approximately 0.7 km south of San Juan: a, specimen in front of notebook,
nearly buried in recently plowed field; b, close-up of specimen in place, with pocketknife handle
for scale. Two large pieces and several small fragments recovered. See Figure 14.
NUMBER 5
mum population density. These areas have been
searched extensively by the local people, although
little in the way of systematic searching has been
undertaken. Large areas of the strewnfield have been
given only cursory attention, while remote and
impenetrable areas of the field remain essentially
unexamined. An indication of the nature of the
problem is the fact that the 110-kg specimen men-
tioned above was not discovered until eight months
after the fall, and specimens as large as 5 kg were
still being found in the intensively searched area
around Rancho Blanco (NMNH 4646) and San
Juan (NMNH 4655) in early June, four months
after the fall. A 14-kg specimen was reported to
have been found near Rancho Blanco in mid-
November. Probably only 40 to 60 percent of the
meteorite material that hit the ground has been
recovered to date, leading us to suggest that the
minimum amount of material that comprises the
Allende shower was of the order of four tons. The
possibility exists, of course, that this estimate is
too conservative. Field investigation should be con-
tinued in the Allende strewnfield as long as sig-
nificant amounts of meteorite material are recover-
able.
RECOVERY OF INDIVIDUAL STONES.?In this section
notes are given on the circumstances of recovery of
specific stones. The stones selected are from those
for which we have particularly reliable information,
and those that are of unusual interest in developing
the distribution pattern within the strewnfield.
The largest specimen that has been recovered to
date was not found until 8 October 1969, eight
months after it fell. Guadalupe Juarez of Rancho
El Cairo was working in a bean field when he no-
ticed a rabbit run into an adjacent uncultivated
area. He got his gun and pursued the rabbit some
150 yards into the chaparral where he discovered
the 110-kg fragmented meteorite. The soil was hard
and consisted of compacted clay containing approxi-
mately 70 percent small stones. The meteorite had
produced a spoon-shaped hole with a one- to three-
inch raised rim at the north end. It was 13 inches
deep at the north end, the bottom sloping smoothly
to ground level at the south end. The hole was 32
inches wide, 52 inches long, and had an apparent
axis of N21?E, The largest individual piece re-
covered was about 6 kg, and the larger pieces were
mainly distributed 12 to 15 feet from the hole,
around its north end. At least 20 kg of dust and
pieces in the 1- to 3-g size range were recovered,
mainly from within the hole.
Comments on individual specimens are given
below:
NMNH 3491, 1.8 kg, approximately 3 km south to
southeast of Rancho Santa Ana.
A completely crusted individual found in place by
schoolboys in our party just prior to a brief but
heavy rainstorm about noon on 13 February 1969
(Figure 4). It was embedded in hard ground, pene-
trating perhaps 5 cm at maximum. Figure 4c is a
photograph of this specimen oriented similarly to
the position shown in the ground. Thin fusion-
crust development on the flat surface to the upper
right in both photographs establishes that this was
a posterior surface during high-velocity atmospheric
flight. This specimen seems to have hit the ground
at a high angle, falling close to the trace of the
fireball's trajectory (Figure 2) . A small amount of
dirt adhered to the specimen as it was removed
from the hole. It is one of many specimens found in
contact with dry grass that showed no signs of
charring.
NMNH 3492, 22 g, approximately 3 km south-south-
east of Rancho Santa Ana, within a few hundred
meters of NMNH 3491.
This specimen is a small fragment free of any ob-
vious fusion crust. A cursory search of the immedi-
ate surroundings produced no additional material.
It was found by a member of our party on the
morning of 13 February 1969.
NMNH 3493, 7.5 kg, approximately 3 km south of
Rancho Santa Ana.
This specimen was found by Manuel G6mez while
running for shelter during the rainstorm about
noon on 13 February 1969. It is a large, flatiron-
shaped individual that landed on hard ground,
making a shallow abrasion (Figure 5). The speci-
men broke into two large pieces (4.8 and 2.5 kg),
a small piece (115 g), and several small fragments.
The largest piece was found about 1.5 m from
the impact point and the other large piece was
an additional meter away. Some of the smallest
fragments were found yet farther away and are
being pointed out in the photograph (Figure 5b).
The alignment of the pieces with the point of im-
10 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
FIGURE 4.?A 1.8-kg Allende individual (NMNH 3491) found in place approximately 3 km south to southeast of Rancho
Santa Ana, 13 February 1969; a, in place, pocketknife for scale; b, rolled out of hole; c, enlarged view oriented as in a.
Scale bar in c, 2 cm.
NUMBER 5 11
FIGURE 5.?A 7.5-kg Allende individual found by Manuel G6mez on 13 February 1969 approxi-
mately 3 km south of Rancho Santa Ana (NMNH 3493). Specimen hit abraded area to right of
boy's foot, a, Largest piece marked by sunglasses, a second large piece farther away and in line
with impact point. Small fragments are being pointed to in b.
12 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
FIGURE 6.?A 7.0-kg Allende individual found partially embedded at base of cactus plant on 13
February 1969 (NMNH 3494). Sunglasses for scale. Found 3 to 4 km south of Rancho Santa Ana.
See Figure 13.
pact was in the direction N40?E, in good agree-
ment with the suggested strewnfield axis.
NMNH 3494, 7.0 kg, 3 to 4 km south of Rancho
Santa Ana.
This specimen was found by a Mexican helper dur-
ing the rainstorm on the morning of 13 February
1969 embedded at the base of a cactus plant. It is
possible that the upper part of the plant was dam-
aged by the essentially vertical descent of the speci-
men, but the damage may also have been caused by
cattle (Figure 6). The specimen was buried ap-
proximately one-third its depth in the relatively
soft earth at the base of the cactus.
NMNH 3495, 4.0 kg, approximately 2 km southeast
of Rancho Santa Ana.
This specimen was sighted in a plowed field by a
Mexican helper from a truck moving slowly along
an adjacent road. The specimen was about 25 m
from the road and had buried itself to about half
its depth. It was found in the early afternoon of
13 February 1969.
NMNH 3502, 2.0 kg, approximately 200 m west of
Rancho Blanco.
This completely crusted individual was found un-
disturbed on the afternoon of 14 February 1969
by a boy in our field party. It landed on hard,
pebbly ground in a chaparral area and made vir-
tually no impression (Figure 7). Only minor damage
was done to the under side of the specimen.
NMNH 3509, 17 kg, approximately 1 km north of
San Juan.
This large, completely crusted individual had been
found north of San Juan within a few hundred
NUMBER 5 13
FIGURE 7.-A 2.0-kg Allende individual (NMNH 3502) found
by local boy at edge of Rancho Blanco on 14 February 1969:
a, in place, b, enlarged view; scale bar, 2 cm.
meters of the main road (Route 45). It was re-
ported to have made a small depression in the
ground and was recovered in three large fragments
and three small ones. The fragments fit together
neatly, reconstructing the original morphology.
Surprisingly little damage was done other than the
clean fractures through the body of the specimen.
It is an eight-sided polyhedron, five sides of which
were abraded by the soil. The fusion crust, however,
is essentially intact.
NMNH 3513, 138 g, approximately 200 m south-
west of Rancho Blanco.
This is a completely crusted individual that is
unusually small for this area of the strewnfield.
It was purchased 15 February 1969.
NMNH 3514, 202 g, approximately 200 m south-
west of Rancho Blanco.
A nearly complete individual that is small for this
area of the strewnfield. Probably less than 10 per-
cent of the specimen is missing. It was purchased
15 February 1969.
NMNH 3511, 8.2 kg, approximately 200 m south-
west of schoolhouse in San Juan.
The specimen was found in a plowed field near the
school in San Juan (Figure 8). It has a large broken
surface that is free of fusion crust, suggesting it
separated into two major portions after the terminal
point of its high-velocity flight. It was purchased
from the finder 16 February 1969.
NMNH 3527, 12.0 kg, approximately 0.7 km south
of San Juan.
This completely crusted individual was found un-
disturbed by Brian Mason about noon on 17 Feb-
ruary 1969 (Figures 3 and 14). It was in a plowed
field, having penetrated the ground to approxi-
mately its own depth. One large piece (11.0 kg),
one small piece (1.0 kg), and several small frag-
ments were removed from the hole.
NMNH 3528, 18.5 kg, approximately 0.5 km south-
southeast of Pueblito de Allende.
This specimen was found at the edge of the Rio del
Valle de Allende between Pueblito de Allende and
Santa Ana in late February 1969. It is a completely
crusted individual that landed undamaged in the
stream bed. It is the largest sample that we know
of that landed without breaking.
NMNH 3529, 33 kg, approximately 2 km northeast
of San Juan.
This specimen was found fragmented near the
main highway 2 km northeast of San Juan during
February 1969. The fragments were collected by
Manuel Gomez by purchase from residents of San
Juan. The largest fragment weighed 14 kg and five
other fragments were in the 1- to 5-kg range.
NMNH 3531, 9.3 kg, near Santa Ana.
An excellent, completely crusted individual found
near Santa Ana. It was presented to the Smithsonian
Institution 16 February 1969 by Mr. and Mrs. R. C.
Byrd of Parral.
NMNH 3681, 281 g, near Porvenir.
A fragment purchased in late May 1969 near Por-
venir. The owner reported that it was part of the
50-kg individual found earlier in the Sierra
Almoloya.
NMNH 3682, 617 g of fragments, near El Cairo.
This collection of small fragments was purchased
14 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
FIGURE 8.?An 8.2-kg Allende individual (NMNH 8517) found approximately 200 m southwest
of schoolhouse (white building) in San Juan: a, finder with Manuel G6mez, kneeling, and Cesar
Gallardo, right, examining hole from which specimen had been recovered earlier, b, photograph
of specimen; scale bar, 2 cm.
NUMBER 5 15
near El Cairo in late May 1969. The owner re-
ported they were part of the 50 kg individual found
earlier in the Sierra Almoloya.
NMNH 3935, 1.8 g, Torre6n de Mata-Rancho
(ruinas) area.
An unusually small, completely crusted individual.
One surface of the specimen is a late break covered
with light fusion crust. Purchased in June 1969.
NMNH 4006, 540 g, in riverbed directly north of
Cienega de Ceniceros de Abajo.
This platy specimen is unusually large for this area
of the strewnfield (Figure 20). Purchased in June
1969.
NMNH 4015, 41 g and 39 g, in Cienega de Ceni-
ceros de Abajo.
These two specimens were both recovered within
the town, one on a rooftop and one from a patio.
They were purchased in June 1969.
NMNH 4289, 2 g, near Cienega de Ceniceros de
Arriba.
A completely crusted individual found in zone 1 to 2
km north to 1 to 4 km south and southwest of
Cienega de Arriba. Purchased during June 1969.
NMNH 4646, 4.36 kg, approximately 1.5 km east
of Rancho Blanco.
This completely crusted large individual was found
near Rancho Blanco?an area that had been inten-
sively searched from the date of fall?in mid-June
1969. It was purchased within a day or two of the
date it was found.
NMNH 4651, 26 g, from an area 1 to 3 km south of
Rancho Blanco.
This completely crusted individual is an unusually
small specimen for this part of the strewnfield. It
was purchased in early June 1969.
NMNH 4655, 4.7 kg, at south edge of San Juan.
This completely crusted individual was found in
the stream at the south edge of San Juan on Sunday,
8 June 1969. It was purchased shortly thereafter.
NMNH 4789 to 4799, 11 small specimens, tail of
strewnfield.
These specimens were found in 1.5 km2 area south-
west of airstrip at tail of strewnfield. Eight of the
specimens were completely crusted individuals,
ranging in weight from 7.2 to 16 g. They were col-
lected by Jack R. Hyde and local helpers in June
1969.
NMNH 4801 to 4809, 8 small specimens, tail of
strewnfield.
These specimens were found in a 3 km2 area to the
southwest of the group above, at the extreme end of
the strewnfield. Seven of the specimens can be con-
sidered complete individuals, and they range in
weight from 3.1 to 43 g, with an average weight of
10 g. If the 43 g specimen is rejected, the average
weight is 5 g. These specimens were collected by
Jack R. Hyde and helpers in June 1969.
NMNH 5183, 1.1 g, near the Ceniceroses.
This is the smallest completely crusted individual in
our collection. It is from a large group of specimens
collected by a single person within an area 4 km
to the northeast of Cienega de Ceniceros de Arriba
and Abajo extending 4 km to the south. This speci-
men was among a group purchased in November
1969.
NMNH 5240, 697 g, 4 km south of Valle de
Allende.
This specimen was found during the first week of
November 1969 partially buried in a field. It is a
major part of an individual with one large fracture
surface and fairly large areas where the fusion crust
had spalled. Areas of the specimen that were unpro-
tected by fusion crust and protruding from the
ground were covered with a fine-grained white ma-
terial tinged with blue and green. X-ray powder
diffraction analysis showed that this material was
largely hexahydrite (MgSO4*6H2O). This specimen
was purchased from its finder in November 1969.
NMNH 5268, 1.5 kg, south edge of Pueblito de
Allende.
This specimen was found on 15 November 1969 on
the south edge of Pueblito de Allende in the mud
at the side of an irrigation ditch. It has large areas
that were unprotected by fusion crust and ap-
parently not buried completely in the mud. These
areas are heavily covered with a coating of fine-
grained white material tinged with blue and green.
X-ray powder diffraction analysis showed this ma-
terial may be mainly hexahydrite (MgSO4'6H2O).
This specimen was purchased in mid-November
1969.
OTHER METEORITES FROM THE PARRAL?JIMENEZ
AREA.?Mexico is a country well-known for its many
meteorite finds (Fletcher, 1890; Farrington, 1915;
16 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
TO CAMARGO
and CHIHUAHUA
METEORITES OF THE
PARRAL-JIMENEZ AREA, CHIHUAHUA
MEXICO
A LOCATION OF METEORITES
0 3 10 15 20 25 30
FIGURE 9?Location of meteorites of the Parral-Jimenez area of Chihuahua, Mexico. The Allende meteorite fell between
two of the largest iron meteorites known, Morito and Chupaderos.
Buchwald, 1968). Three out of the ten largest
individual meteorite masses known are from North-
ern Mexico, and two of these, the Morito and Chup-
aderos irons, are from the Parral-Jim?nez area of
southern Chihuahua. The Allende meteorite, the
largest stony meteorite yet observed, fell in between
these two ancient falls, seemingly attesting to the
"peculiar property, difficult of explanation which
the Mexican soil has in attracting meteoritic" ma-
terial (Barcena, 1876).
Morito is a single 11-ton conical mass of iron, a
medium octahedrite (IIIA), known as early as
1600. It was found on Rancho El Morito (approxi-
mate location of find, latitude 27?OO/N, longitude
105?28/W) and at an early date moved to Hacienda
San Gregorio (Figure 9). It has since been moved
NUMBER 5 17
to Mexico City and is preserved there in the School
of Mines. The three Chupaderos irons are medium
octahedrites (HIB) that have been known for cen-
turies (Figure 9). The two largest masses, 14 and
6.5 tons, were found on the east slope of the Sierra
de Chupaderos (approximate location of find, lati-
tude 27?10/N, longitude 104?4(KW). The probable
site of find of the Morito meteorite and the site
of the two Chupaderos masses are separated by ap-
proximately 85 km. A third mass of the Chupaderos
meteorite is the "Adargas" specimen (Buchwald,
1968). "Adargas" is a 3.4 ton mass that was moved
at an early date to La Concepci6n (Figure 9),
possibly coming originally from the area of the
Sierra Las Adargas.
The Parral?Jimenez area is undoubtedly among
the richest meteorite-producing areas in the world.
Within the relatively small area of approximately
4,000 km2, three of the world's major meteorites
have been recovered: the 11-ton Morito iron, the 24
tons of Chupaderos irons, and several tons of the
Allende stony meteorite. The area per meteorite
find for North America as a whole is 30,000 km2,
many times the area of recovery of the Morito-
Allende-Chupaderos meteorites.
The place-names used in our map (Figure 9)
are all taken from the modern base maps described
above. Important names relating to the history of
the Morito and Chupaderos irons have changed in
a number of instances since the early reports and
maps were prepared. Jimenez or Ciudad Jime'nez
was known earlier as Huejuquilla (an Indian name,
pronounced Wa-hu-ke-yah), and a number of varia-
tions on this spelling have been used. Valle de Al-
lende was previously known as El Valle de San
Bartolome", and simply as El Valle. The Rio del
Valle de Allende has been known as Rio de Allende
and Rio de San Bartolome". The Rio la Chona has
been referred to as the Rio de la Concepcidn.
Three other meteorites are listed in the litera-
ture as coming from the Parral-Jime*nez area:
Parral, Tule, and Sierra Blanca (Hey, 1966). All
three of these meteorites have recently been ex-
amined by V. F. Buchwald (personal communica-
tion, 1969) as part of his comprehensive examina-
tion of iron meteorites. Buchwald reports that Par-
ral appears to be a fragment of Morito, Tule is a
pseudometeorite, and Sierra Blanca may be a trans-
ported fragment of the Toluca shower.
Morphology
The individual Allende meteorite specimens pre-
serve in their ablation crusts and gross morphology
a partial record of the meteoroid's passage through
the atmosphere. This information, combined with
the observed specimen distribution pattern (dis-
cussed above), allows us to derive a probable se-
quence of events. The meteoroid entered the at-
mosphere at a low angle as a single large body with
cosmic velocity (greater than 11 km/sec, and prob-
ably less than 18 km/sec, McCrosky et al., 1969).
Friction between it and the increasingly dense at-
mosphere produced ablation and initiated the lum-
inous fireball stage. Continued penetration into the
atmosphere was followed by a major disruption?
an explosion of the parent body?at a relatively
great altitude. Many large blocky fragments were
produced, as were smaller pieces and a quantity
of rubble. Each individual fragment continued in
luminous, ablative flight until reaching its point
of atmospheric retardation, the point at which its
cosmic velocity was completely dissipated. During
this part of the flight, fragments were subjected to
differential atmospheric drag depending on their
shape and orientation, resulting in size sorting along
the trajectory and lateral broadening of the cluster.
It was during this interval between initial disrup-
tion and retardation that fragments became individ-
ual crusted meteorite specimens. The recovery of
numerous specimens with areas of secondary fusion
crust indicates at least minor fragmentation and
spallation subsequent to initial breakup. Specimens
reached their individual retardation points and
then continued to the ground in free fall. It was
undoubtedly during the free-fall portion of their
descent that wind became a significant factor in
distorting the symmetry of the field (Figure 2) . The
largest specimens traveled the farthest and shattered
on impact, making small holes in the ground, while
small specimens generally fell far behind, producing
no damage and being recovered unscathed. Large-
to medium-size specimens were recovered between
the two extremes, their condition on recovery de-
pending on such factors as preexisting weakness in
the specimen and whether it landed on hard or soft
ground. Many remarkably fragile features ablated
on Allende meteorite fusion crust survived descent.
Individual stones are typically ablation-modified
18 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
fragmental shapes; blocky, pyramidal, and occa-
sionally lenticular. They are covered with a dull
black fusion crust that tends to be chipped or
spalled awav at edges, revealing either fresh frac-
ture surfaces or fracture surfaces covered with a
light secondary fusion crust that is frequently red-
dish in hue. The fusion crust contains occasional
lustrous spots above the calcium and aluminum-rich
aggregates which are characteristic of the meteorite,
and infrequent sooty areas above fine-grained black
inclusions. The amount of aerodynamic shaping of
FIGURE 10.?A 11.4-kg blocky, crusted Allende individual
(NMNH 3523) found between Pueblito de Allende and San
Juan. Scale bar, 5 cm.
an individual fragment is related to the stability of
its orientation during ablative flight. A small per-
centage of the individuals shows evidence of oriented
flight and heavy ablation both in the gross shape
of the specimen and in the character of fusion-crust
development on anterior and posterior surfaces. A
particularly curious feature of Allende specimen
morphology is the presence on five specimens of
hollow filaments of fusion crust encapsulating an
as yet unidentified material, apparently a colorless
hydrocarbon. In the following section individual
specimens that illustrate many of these features will
be described.
NMNH 3523, 11.4 kg, found between Pueblito de
Allende and San Juan.
This large, blocky specimen was purchased in Pueb-
lito de Allende on 16 February 1969. Its fragmental
shape appears to be relatively unmodified by abla-
tion, but its surfaces are essentially covered with a
well-developed fusion crust (Figure 10). Spalling
has occurred along edges, and many of these spalled
areas are partially or completely covered with a
thin reddish secondary fusion crust that does not
show clearly on the photographs. One fairly large
fragment appears to have broken off at a late stage,
leaving a fresh fracture surface (Figure 10c). In-
clusions in the matrix revealed on this fracture
surface appear to have a parallel alignment, an
NUMBER 5 19
FIGURE 11.?A 6.2-kg Allende individual (NMNH 3500) found approximately 700 m southwest of
Rancho Blanco. Scale bar, 5 cm.
alignment that is also parallel to the top and bottom
exterior surfaces in this photograph (Figure 10c).
This suggests that the fracture pattern of the rock
may well be related to its internal fabric. The
lower part of Figure 106 shows the effect of collid-
ing with the ground. Light-colored clay soil is em-
bedded in the fusion crust, but the crust survived
essentially intact. At the upper part of the light-
colored area a well-developed polygonal craze pat-
tern is emphasized by soil filling small cracks. This
surface craze is observed on specimens recovered
very soon after they fell, but is much more pro-
nounced on some specimens that have been exposed
to the elements for a few months.
NMNH 3500, 6.2 kg, found approximately 700 m
southwest of Rancho Blanco.
This large specimen was purchased in Rancho
Blanco on 14 February 1969. It is a fragmental shape
only slightly modified by atmospheric ablation; a
rectangular base, two trapezoidal sides, and triangu-
lar ends (Figure 11). The rectangular surface is
indicated in profile in Figure \\b. Several of the
surfaces are slightly convex, and they are all cov-
ered with a well-developed fusion crust. Spalling
of fusion crust at edges, a common feature in
Allende specimens in general, is particularly pro-
nounced on this stone.
NMNH 3501, 4.3 kg, found approximately 2 km
southwest of Rancho Blanco.
This complete individual was purchased in Rancho
Blanco on 14 February 1969. It is a nine-sided figure,
comprising three triangular and six trapezoidal
surfaces (Figure 12), suggestive from certain aspects
of a distorted octahedron (Figure 12a). Its flat sur-
faces are all heavily crusted, but there are rela-
tively numerous areas of spalled crust at edges.
Most of these spalled areas are covered with a thin
reddish secondary fusion crust. The apex area in
Figure 12a is covered with this very thin crust that
20
is obvious to the unaided eye, but does not show
at all on these black and white photographs.
NMNH 3494, 7.0 kg, found 3 to 4 km south of
Rancho Santa Ana.
SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
This specimen (Figure 13) was found at the base
of a cactus plant on 13 February 1969 (Figure 6).
It is an essentially completely crusted individual, a
fragmental shape relatively unmodified by ablation.
FIGURE 12.?A 4.5-kg Allende individual (NMNH 3501) found approximately 2 km southwest of
Rancho Blanco. Scale bar, 5 cm.
FIGURE 13.?A 7.0-kg Allende individual (NMNH 3494) found at base of cactus 3 to 4 km
south of Rancho Santa Ana. See Figure 6. Scale bar, 5 cm.
NUMBER 5
There are three large trapezoidal surfaces and a
small one at the specimen's apex. A section per-
pendicular to these surfaces would be essentially
triangular. The corner and bottom surface in Fig-
21
ure ISa shows the part of the specimen that pene-
trated the ground on landing. Only slight damage
to the specimen seems to have resulted. Fusion
crust appears to have spalled off of most of the
FIGURE 14.?An unusually large oriented Allende individual (NMNH 3527). Found in place approximately 0.7 km south
of San Juan (see Figure 3), broken into 1.0-kg and 11.0-kg pieces: a, anterior surface to right, zone of fusion-crust accumula-
tion diagonally from left to right, and posterior surface to left; b, anterior surface; c, posterior surface and zone of fusion-
crust accumulation; d, posterior surface and zone of fusion-crust accumulation. Scale bar, 5 cm.
22 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
edges of the specimen, and most of the spalled areas
have a thin coating of secondary fusion crust. One
surface (Figure 13b) has a large area of spalled
fusion crust, the edges of which have this secondary
coating. The large surfaces in Figure 13a have well-
developed flow lines in thick crust.
NMNH 3527, 12.0 kg, approximately 0.7 km south
of San Juan.
This large individual was recovered in two pieces
from a plowed field, buried to its own depth, by
Brian Mason on 17 February 1969 (Figures 3 and
14) . It is of particular interest because it is an un-
usually large oriented individual with spectacular
fusion-crust development. The specimen's basic
shape is a distorted pyramid with one domed sur-
face. All of the photographs in Figure 14 show the
specimen sitting on the same surface. Figure 14a
was taken looking down on the specimen, showing
the edge of the anterior (upper right) and posterior
(lower left) surfaces. Figure \4b is of the complete
anterior surface, while Figure \Ad shows much of
the posterior area. Figure 14c shows one side of the
posterior surface, a portion of the zone of fusion-
crust accumulation, and an edge of the anterior
surface. The surface not shown is a posterior surface
also containing a zone of fusion-crust accumulation.
This specimen is essentially completely crusted, but
it does have a number of small areas where crust
spalled late in flight or where small fragments have
been removed. The 1-kg apex fragment fits securely
in place, leaving no doubt about the preimpact
shape of the specimen. The black anterior surface
(Figure 14fr) contains a number of small spalled
areas, most of which are covered with a thin coating
of reddish fusion crust. This anterior surface is
somewhat irregular, and thick vesicular fusion crust
has accumulated in depressions. A zone of the pos-
terior surface along the edge of the anterior surface
is covered with a thick vesicular accumulation of
fusion crust (Figure \Aa,c,d). This zone has a
width of 6 cm in some areas and shows a distinct
flow pattern. The posterior surface (Figure 14c,d)
has a much thinner fusion crust that is lighter in
color than that on the anterior surface, having a
decided brownish hue. A polygonal craze pattern is
pronounced on this thinner crust.
NMNH 3522, 188 g, found approximately 2 km
north of Torreon de Mata.
This unusually fine aerodynamically shaped speci-
men was purchased in Torre6n de Mata on 16 Feb-
FIGURE 15.?An aerodynamically shaped Allende individual
found near Torre6n de Mata (NMNH 3522): a, side view of
anterior surface showing cone shape; b, looking down on
anterior surface; c, posterior surface with vesicular fusion
crust. Natural size, ammonium-chloride smoked to enhance
detail.
NUMBER 5
ruary 1969. It appears to have entered upon inde-
pendent flight in the atmosphere as a plate-shaped
fragment. Orientation was maintained and aero-
dynamic sculpturing smoothed forward edges and
produced a cone-shaped specimen with pronounced
anterior and posterior sides (Figure 15, ammonium-
chloride smoked). Flow lines in the fusion crust
fan out from the apex of the cone, and inclusions
within the meteorite produced a patchy appearance
due to varying crust composition (Figure 15 a,b).
The outline of chondrules produced a pronounced
pattern on the anterior surface, and small depres-
sions in this surface developed areas of vesicularity.
A sharp boundary marks the edge of the anterior
and posterior surfaces, with fusion crust accumulat-
ing preferentially on the posterior surface near its
* --:-.?
FIGURE 16.?An aerodynamically shaped Allende individual from the area of the Ceniceroses (NMNH 4018): a, looking down
on anterior surface; c, posterior surface with fusion-crust accumulation at edges; b and d, side views showing flow of fusion
crust. Scale bar, 2 cm.
24 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
edge. The posterior surface is uneven, undoubtedly
a result of initial fragmentation. It is completely
covered with a highly vesicular fusion crust that is
somewhat thinner toward the center of the surface.
Fusion crust in general on the posterior surface is
considerably thicker than on the anterior surface.
NMNH 4018, 121 g, found in an area extending 4
km south and southeast of the Ceniceroses.
This specimen resulted from aerodynamic shaping
of a blocky fragment (Figure 16). It has a domed
anterior surface, four lateral surfaces, and a flat
posterior surface. Two of the corners of the back
surface are nearly right angles, the other two being
less regular. A fairly large area of fusion crust at
the apex of the dome has spalled, as have several
smaller areas of crust at edges. One small spalled
area on the anterior surface has been partially
covered with a light coating of reddish secondary
fusion crust. Most of the white material on the
anterior surface is foreign matter due to contact
with the ground (Figure 16a). The fusion crust
on the forward surface is black with a slight sug-
gestion of red in limited areas. The crust on the
posterior surface is well developed and markedly
different in appearance (Figure 16c). It is pock-
marked and dark reddish-brown in color, not par-
ticularly vesicular. The pockmarks appear to be
remnants of large vesicles that have been distorted
by collapse and surface flow. On the lateral surfaces
there are two types of crust flowing over the pos-
terior crust (Figure I6b,d). The reddish-brown
crust of the posterior surface served as an undercoat
for streamers of vesicular black crust flowing from
the anterior surface. There is a particularly marked
buildup of fusion crust at the edge of the posterior
surface (Figure 16c), resulting from flow along the
lateral surfaces. Small beads of black glass were
carried on to the posterior surface by airflow and
can be seen with moderate magnification near the
edge.
NMNH 4083, 165 g, found near cemetery 300 m
south of Cie"nega de Ceniceros de Arriba.
This specimen appears to have initially been a rec-
tangular parallelepiped that has been modified by
oriented flight and one major late-stage fracture. It
has four distinct surface types; anterior, lateral,
posterior, and fracture. Figure 17 gives views of
FIGURE 17.?An aerodynamically shaped Allende individual
found near Cienega de Ceniceros de Arriba (NMNH 4083):
a, anterior surface; b, side view, showing streamers of fusion
crust at the edge of the anterior surface. Natural size, am-
monium-chloride smoked to enhance detail.
ammonium-chloride smoked surfaces; (a) the an-
terior surface, and (b) the edge of the anterior
surface and a lateral surface. The anterior surface
is predominantly black in color with a suggestion
of red that is pronounced in the few small areas
where fusion crust has spalled. Aerodynamic flow
along the surface is recorded as stringers of glass
in the fusion crust (Figure 17a). A surface pattern
due to differential melting of chondrules and matrix
is evident. Figure 176 is of an edge of the anterior
NUMBER 5 25
FIGURE 18.?An aerodynamically shaped Allende individual found southwest of Cienega de Ceniceros de Abajo (NMNH 4084):
a, Anterior surface with spalled apex; b, vesicular posterior fusion crust. Ammonium-chloride smoked to enhance detail.
Scale bar, 2 cm.
surface and a lateral surface. The accumulation of
fusion crust marked by stringers of glass in the
photograph marks the edge of the lateral surface.
The fusion crust on the lateral surface below the
zone of accumulation is thinner and noticeably
redder in color. The posterior surface has a thin
fusion crust covering a relatively unmodified rough
fracture surface. The area in the lower left of Fig-
ure 17a is a late fracture surface, free of fusion
crust. There are areas on the posterior surface and
at edges where fusion crust has spalled late in high-
velocity flight, and some of these are covered par-
tially or completely with a light, reddish coating of
secondary fusion crust.
NMNH 4084, 87 g, found H/2 km southwest of
Cienega de Ceniceros de Abajo.
The basic shape of this specimen seems to have been
a somewhat irregular tetrahedron (Figure 18, am-
monium-chloride smoked). One set of three surfaces
has been smoothed into an anterior surface of an
aerodynamically oriented individual (Figure 18a).
Its anterior and posterior surfaces are clearly de-
fined by the character of their fusion-crust develop-
ment. The anterior surface has fusion crust over
about half of its area, spalling late in flight having
removed fusion crust at the apex. Part of this area
is covered with a light, reddish secondary fusion
crust that is not revealed in the photograph. There
are several fresh fracture surfaces along the edge
joining the anterior and posterior surfaces (Figure
186). The posterior surface is highly vesicular and
has three unusual protruding bodies that appear
to be fusion-crust-covered large chondrules stand-
ing out in relief.
NMNH 4157, 15 g, found in area 1 to 4 km south-
west to south of Cienega de Ceniceros de Arriba.
This initially small fragment was markedly shaped
by aerodynamic ablation (Figure 19). It is typical
of a number of small individuals recovered from
the strewnfield, generally south of the Torre6n de
Mata area and frequently south of the Ceniceroses.
It has a domed anterior surface (Figure 19a) and
irregular posterior surfaces (Figure 19b,c) reflect-
ing the original fragmental shape of the specimen.
It is completely crusted with the exception of one
small area on a posterior edge. The fusion crust
on the anterior surface is comparatively thin, with
a number of fine cracks. Fusion-crust accumulations
developed at edges of the posterior surface due to
flow from the anterior surface. Accumulations of
26 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
FIGURE 19.?A small aerodynamically shaped Allende individ-
ual (NMNH 4157) found southwest of the Ceniceroses: a,
anterior surface; b, side view; c, posterior surface. Twice ac-
tual size.
vesicular glass are also marked in depressions in
the posterior surfaces (Figure 19b,c). The white
areas in Figure 19b and c are adhering foreign
matter.
NMNH 4006, 540 g, found in the riverbed directly
north of Cie*nega de Ceniceros de Aba jo.
This specimen is remarkably heavy compared to
others found so far south in the strewnfield. The
explanation is undoubtedly atmospheric drag, and
evidence from the specimen itself supports this sug-
gestion. The specimen was undoubtedly initially
a platy fragment that presented to the atmosphere
an exceptionally large surface area in relation to its
mass (Figure 20). Fusion-crust development indi-
cates clearly an anterior (Figure 20fr) and posterior
(Figure 20a) surface. Fairly large areas on two op-
posite lateral surfaces are recent breaks, completely
free of fusion crust (Figure 20c). These surfaces
demonstrate that the area presented to the atmos-
phere during high-velocity flight was certainly
somewhat greater than the area of the present
anterior surface. The edges of the anterior surface
that do not join recent fracture surfaces are rounded
(Figure 206), and flow lines within the fusion crust
on lateral surfaces indicate flow toward the posterior
surface. In several areas where lateral surfaces join
the posterior surface there are accumulations of
vesicular crust (Figure 20a) on the posterior sur-
face. Fusion crust on the anterior surface is rela-
tively thin and noticeably reddish in color. The
posterior-surface fusion crust is similar, but it is
spattered in streaks with a black lustrous glass, sug-
gestive of snail tracks, undoubtedly derived from
areas of fusion-crust accumulation. The result is
that the posterior surface has a sheen that is not
noticeable on the anterior surface.
Of the specimens available to us from the area of
the Ceniceroses, less than 6 percent weigh more than
100 g, less than one percent more than 200 g, and
less than 0.1 percent more than 300 g. The four
heaviest specimens are: NMNH 4006, 540 g, dis-
cussed above, NMNH 4007, 272 g; NMNH 4533,
257 g; NMNH 4575, 232 g. Three of these (NMNH
4006, 4533, 4575) are aerodynamically shaped speci-
mens with a large ratio of anterior surface area to
weight of specimen. This would appear to be a
clear demonstration of atmospheric sorting on the
basis of shape as well as on the basis of mass.
NUMBER 5 27
FIGURE 20.?An unusually large Allende individual (NMNH 4006) found near Cienega de Ceni-
ceros de Abajo. The flat shape resulted in high drag and retardation: a, posterior surface;
b, anterior surface; c, broken surface on bottom side of b. Scale bar, 5 cm.
28 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
NMNH 3812, 46 g, found in zone 1 to 4 km south-
east to north of the Ceniceroses.
This specimen is one of a group of five displaying an
unusual type of lustrous fusion-crust development
(Figure 21). The specimen has areas of normal
fusion crust (^30%), spalled areas that are par-
tially covered with secondary fusion crust (~50%),
and areas of a black, lustrous, blistery appearing
crust. This lustrous crust appears in part to have
flowed over areas of regular fusion crust on to
spalled areas. At the edges of these areas of flow the
crust seems to have accumulated to a greater than
normal thickness, and it has a resinous appearance.
From areas such as this, crust material appears to
have been drawn out, forming straps that adhere to
or bridge across areas of spalled matrix surface,
jumping gaps from a few millimeters to a centimeter
in length (upper left in Figure 21a and right side of
Figure 216). These ropy straps are suggestive of
drawn taffy, and they frequently follow the edges
of concave spalled areas. Extending from these
straps are spines that are drawn out to delicate
filaments, frequently with nobby ends. Filaments
on this specimen are translucent and amber in color
when very thin, and apparently encapsulate a color-
less hydrocarbon material.
NMNH 3813, 34 g, found in zone 1 to 4 km south-
east to north of the Ceniceroses.
This is the most dramatic of the five specimens with
the type of lustrous fusion-crust development de-
scribed above (NMNH 3812). In most respects, this
specimen is a typical small individual, much of its
surface being covered with ordinary and secondary
fusion crust or spalled areas (Figure 226). One end
of the specimen, however, is capped with lustrous
fusion crust (Figure 22a). Protruding from this
area is a spine reminiscent of an antler, extending
a full centimeter. It is amber in reflected light, but
light yellow and translucent when backlighted (Fig-
ure 22c). It is remarkable that features of this
delicacy survived landing, recovery by a local resi-
dent, and shipment to Washington.
NMNH 3814, 77 g, found in zone 1 to 4 km south-
east to north of the Ceniceroses.
A large part of the surface of this specimen is free
of fusion crust (Figure 236), while other areas are
covered with a normal thick, and in part vesicular,
FIGURE 21.?One of a small group of Allende specimens with
an unusual, lustrous fusion-crust development. Found near
the Ceniceroses (NMNH 3812). Twice actual size.
fusion crust. About half of the surface of the speci-
men, overlying both spalled and crusted areas, is a
network of ropy straps of lustrous fusion crust (Fig-
ure 23d,e). In several places the ropy material has
lifted from the surface of the sample and can be
seen in silhouette. Figure 23a is an enlargement of
the area in the upper left of Figure 23b, and Figure
23c is an enlargement of an area in the upper right
of Figure 236.
Broken surfaces and surfaces with thin fusion
crust show abundant chondrules, usually about 1
mm in diameter, but sometimes much larger.
Broken surfaces of a single large fragment from a
33 kg individual are illustrated in Figure 24
NUMBER 5 29
FIGURE 22.-An Allende individual from the area of the Ceniceroses (NMNH 3813) with a horn
of lustrous, translucent fusion crust: a, b, twice actual size; c, horn at four times actual size.
30 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
FIGURE 23.?An Allende individual from the area of the Ceniceroses (NMNH 3814) with a network of ropy, lustrous fusion
crust over spalled and crusted areas. Scale bar 2 cm for b, d. e; 1 cm for a, c.
(NMNH 3529, found 2 km northeast of San Juan).
Irregular aggregates, white to pink in color and
sometimes lensoid in shape, up to several milli-
meters long, are common in the black matrix. Fig-
ure 24a illustrates a normal distribution of ag-
gregates, while the reverse side of the same piece
(Figure 246) show a preferential accumulation of
aggregates and large chondrules. A large polished
surface from this same individual is shown in
Figure 25. Chondrules, aggregates, and a single large
dark inclusion (see below) are shown. This surface
would be considered representative if the dark in-
clusion were excluded. Present in both Figures 24
and 25 are indications of a vague oriented fabric
suggested by subparallel elongation of aggregates.
These elongations may represent an original flatten-
ing, or a secondary flattening produced by later
deformation.
NUMBER 5 31
FIGURE 24.?TWO views of broken fragment from a large Allende specimen (NMNH 3529): a, normal distribution of aggre-
gates and chondrules. b, preferential accumulation of aggregates and large chondrules, and a vague oriented fabric. Actual
size.
32 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
FIGURE 25.?A polished surface of slice of a large Allende individual (NMNH 3529). Note large
dark inclusion at bottom of photograph. Actual size.
Mineralogy and Petrography
Allende is an unusually heterogeneous meteorite,
and a comprehensive description of its composition
and structure is thus a complex task. We propose to
treat it in terms of its major components, as follows:
(a) matrix; (b) chondrules; (c) irregular aggre-
gates; (d) dark inclusions. It has a remarkably diver-
sified mineralogy; a list of the minerals so far identi-
fied is given in Table 1.
Under the microscope, sections of normal thick-
ness (0.03 mm) show numerous chondrules, mostly
0.5 to 2 mm in diameter, and less numerous irreg-
ular translucent aggregates of variable size, usually
a few millimeters in maximum dimension, in a
black opaque matrix. Point-counting on one section
gave the following results: chondrules, 34 volume
percent; aggregates, 9 percent; matrix, 57 percent.
As a section is reduced in thickness, the matrix
NUMBER 5
Name
Kamacite
Awaruite
CoDDer
Troilite
Pentlandite
Chromite
Spinel
Hercynite
Perovskite
Olivine
Enstatite
Clinoenstatite
Clinohypersthene ,
Diopside
Augite (fassaite) . .
Ferroaugite
Anorthite
Gehlenite
Grossular
Nepheline
Sodalite
Cordierite
TABLE 1.?Minerals in the Allende meteorite
Mineral location
Formula Matrix Chondrules Aggregates
(Fe,Ni) x
Ni8Fe x x
Cu x
FeS x x
(Fe,Ni)9S8 x x
FeCr2O4 x
MgAl2O? x
FeAl2O4
CaTiOa
(Mg,Fe)2SiO4 x x
MgSiO8
MgSiOs x
(Mg,Fe)SiO8 x
CaMgSi2O8
Ca(Mg,Al,Ti)(Al,Si)2O6 x
Ca(Fe,Mg,Al)(Al,Si)2Oa
CaAlaSi2O8 x
CaaAlaSiOr X
CaaAlaSisOoa
NaAlSiO* x
NaiAlgSiaOuCl x
Mg^USiBOw
X
X
X
X
X
X
X
X
X
X
X
X
X
33
becomes progressively more translucent, and at a
thickness of about 0.01 mm is seen to consist of
minute transparent equant crystals, interspersed
with a moderate amount of opaque material. X-ray
diffraction shows that the matrix consists largely
of iron-rich olivine (~50 mole percent Fe2SiO4).
Electron micrographs (courtesy Dr. K. M. Towe)
show that many of the minute olivine crystals are
idiomorphic, one to a few microns in greatest di-
mension (Figure 26).
Dispersed throughout the matrix is several per-
cent of sulfides, present as pentlandite with occa-
sional coarser grains of troilite. Localized sulfide
concentrations are frequently found at chondrule
edges, as are the larger grains of both pentlandite
and troilite. Areas of pentlandite and grains of
troilite sufficiently large for microprobe analysis
were examined. The nickel content of the pentland-
ite ranges between 16.5 and 21 percent, with an
average value around 19 percent, and about 0.7 per-
cent cobalt is present. The troilite appears to be pure
FeS, without detectable nickel and cobalt. The rare
grains of metal are generally rich in nickel (67?68%
Ni and 1.6% Co by microprobe), although an oc-
casional grain of low nickel kamacite was observed
within chondrules. These nickel-rich grains would
normally be described as taenite, but Dr. E. P. Hen-
derson (personal communication) has noted that
the microhardness of these grains is notably and con-
sistently less than that of taenite, and points out
that the composition is similar to that of awaruite.
Awaruite approximates the formula Ni3Fe and is
an ordered structure in contrast to taenite. Kullerud
and El Goresy (1969) note that awaruite is stable
below 503?C in the condensed Fe-Ni-S system, but
cannot coexist stably with troilite. Electron micro-
probe investigation established, however, that in the
Allende meteorite metal of awaruite composition
34 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
FIGURE 26.?Electron micrographs of Allende meteorite matrix, courtesy of Dr. K. M. Towe: a, transmission micrograph of
idiomorphic olivine crystal (x 68,000), with electron diffraction pattern (inset), b, c, platinum-carbon replica micrographs of
representative matrix grains, probably olivine (x 30,000).
NUMBER 5 35
exists in intimate association with both pentlandite
and troilite. Other minerals included in the matrix
are minute amounts of chromite, native copper, and
possibly mackinawite (Dr. G. Kullerud, personal
communication). The carbonaceous matter cannot
be recognized as such except by the uniform black
coloring of the matrix; it probably exists as a thin
film coating the individual crystallites of olivine
and other minerals.
A remarkable variety of chondrules has been ob-
served in this meteorite (Figures 28-34). They can
be divided into two groups (I and II) on the basis
of chemical and mineralogical composition, and
into many different textural types. The chemical
groups are:
I. Magnesium-rich chondrules. These consist of
magnesium-rich olivine, sometimes with clinoen-
statite and interstitial glass (which may be partly
devitrified). Individual olivine crystals are fre-
quently zoned, with cores of almost pure Mg2SiO4
and increasing iron content towards the surface.
An X-ray diffractometer trace shows a sharp peak
corresponding to a composition of Fa9, with a broad-
ened base indicating some variation in composition.
The average of the microprobe analyses (100
points) is Fa6 (Figure 27), the commonest value
being Fao_i. The clinoenstatite is close to MgSiO3
in composition. Microprobe analyses of the glass
show that it is rich in calcium and aluminum; a
typical analysis is (weight percent) : SiO2, 47.8;
A12O3, 24.7; CaO, 17.5; MgO, 6.3; FeO, 0.7; Na2O,
2.0; K2O, 0.2. The chondrules usually consist en-
tirely of silicate minerals. A few contain small
amounts of sulfides, which occasionally are present
as spherical aggregates. Some chondrules are rimmed
with small grains of sulfides.
A great variety of textures has been observed in
these chondrules. The commoner types can be
grouped as follows, in approximate order of
abundance:
1. Granular olivine (Figure 28). A tightly
packed aggregate of euhedral olivine crystals.
Within a single chondrule the individual crystals
are fairly uniform in size, but this size varies from
chondrule to chondrule; the range measured is
15 20 25
Mole Percent FeoSi0,,
30 35 40 45
FIGURE 27.?The composition of individual olivine grains in Allende chondrules, determine by
electron microprobe analysis.
36 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
from 0.02 to 0.5 mm in different chondrules. Some
chondrules have a core of coarse crystals and a
mantle of much smaller grains.
2. Granular olivine plus clinoenstatite. Like 1,
but containing minor amounts of clinoenstatite.
Frequently, the grains of clinoenstatite are concen-
trated near the surface of the chondrule and then
show a "snowball" effect, the clinoenstatite prisms
being arranged tangentially to the surface. Clino-
enstatite crystals within the chondrules may be com-
paratively large and poikilitically enclose grains of
olivine.
3. Barred olivine chondrules (Figures 29, 30).
Long prisms of olivine are interspersed with bars
of glass or devitrified glass. The olivine prisms may
be parallel throughout the chondrule, may show a
fan-shaped arrangement, or may form two or more
nonparallel sets. Frequently, a barred chondrule
will have a mantle of olivine, which may be in
optical continuity with the core olivine.
4. Monosomatic olivine chondrules. These chon-
drules consist of a rounded single crystal of olivine,
usually about 0.5 mm in diameter; sometimes this
rounded single crystal is surrounded by an aggre-
gate of small olivine grains.
5. Pyroxene-rich chondrules. These are rare in
Allende; this is reflected in the low content of
normative pyroxene, as shown by the chemical
analysis of a chondrule concentrate (Table 3, col-
umn 6, on page 45).
II. Calcium- and aluminum-rich chondrules. These
are comparatively rare, making up less than five
percent of the total chondrules, but show unusual
chemical and mineralogical compositions. Three
distinct types have been recognized, and probably
others remain to be discovered. The mineralogy
of the three types is as follows:
a. Gehlenite-fassaite-anorthite-spinel.
b. Anorthite-forsterite-spinel.
c. Nepheline-sodalite-fassaite-olivine.
FIGURE 28.?Irregular aggregates and chondrules in black matrix of the Allende meteorite;
diameter of the largest chondrule (granular olivine) is 1.4 mm. Transmitted light.
NUMBER 5 37
FIGURE 29.?Allende chondrule (0.7-mm diameter) of prismatic olivine crystals in dark glass,
with a halo of small olivine crystals. Transmitted light.
FIGURE 30.?Barred olivine chondrule with a narrow continuous rim of olivine in Allende
meteorite; diameter 1.2 mm. Transmitted light.
SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
FIGURE 31.?Three juxtaposed barred chondrules in the Allende meteorite, in transmitted (upper)
and reflected (lower) light; the reflected light photograph shows tiny sulfide grains (white) on the
surface of the chrondrules (largest chrondrule is 1.7-mm diameter).
NUMBER 5 39
FIGURE 32.?Allende meteorite granular olivine chondrule (1.3
mm diameter) in reflected light, showing two spherical sul-
fide inclusions.
Type a form prominent chondrules, being un-
usually large, one (Figure 33) measuring 24 mm in
diameter. This provided material for both bulk
chemical analysis (Table 3, column 7) and for
microprobe analyses of the individual minerals
(Table 2). Gehlenite is the most abundant mineral
(~40 percent), and is present as large prismatic
crystals; its refractive indices are ? ? 1.660,
? = 1.655. The microprobe analysis shows that it
is close to pure gehlenite in composition, with a
moderate amount of the akermanite component
(Ca2MgSi2O7). Pyroxene forms about 30 percent of
the chondrule, and is an unusual type, being highly
aluminous and also titanium-rich; in terms of com-
ponents, it contains about 59 percent CaMgSi2O6 by
weight, 24 percent CaAl2SiO6, and 17 percent
CaTiAl2O6. Its refractive indices are a = 1.705,
/? = 1.713, y = 1.735; it is colorless and nonpleo-
chroic, in marked contrast to the purple titaniferous
clinopyroxene in the Angra dos Reis meteorite, and
in terrestrial igneous rocks. Dr. Joan Clark of the
United States Geological Survey has determined its
cell dimensions, with the following results (in A) :
a = 9.725?0.004; b = 8.828?0.003; c = 5.306
?0.003; p = 105?55'?10'; V = 438.1A3. Pyroxene
with a considerable content of the CaAl2SiO6 com-
TABLE 2.?Microprobe analyses of gehlenite, fassaite,
and anorthite in a large chondrule (type a, Figure
33) from the Allende meteorite (J. Nelen, analyst).
Values given in weight percent
SiO,
TiOa
Al,Oa
MgO
CaO
Gehlenite
27.0
0.1
27.7
3.1
41.0
Fassaite
39.8
5.5
18.2
9.7
25.4
Anorthite
43.4
0.1
35.8
0.0
20.2
98.9 98.6 99.5
Iron and sodium were looked for but not detected above the
trace level.
ponent is termed fassaite. Other minerals in the
chondrule are anorthite (about 10 percent), and
spinel (about 20 percent).
The composition of this type a chondrule is
closely similar to the compositions studied by Prince
(1954) in the CaO-MgO-Al2O3-SiO2 system. Using
his diagram for the 10 percent MgO plane in this
system, a melt of approximately the composition of
this chondrule would begin to crystallize spinel at
about 1550?C, followed by gehlenite at about
1400?C, anorthite at 1250?C, pyroxene at 1235?C.
The textural relations of these minerals in the chon-
drule are consistent with this sequence of crystalliza-
tion. Small euhedral crystals of spinel are included
within all the other minerals. Gehlenite is present
as large prismatic crystals, suggesting crystallization
over a considerable temperature interval, whereas
the anorthite and fassaite occur as smaller xenomor-
phic grains in the interstices of the gehlenite crys-
tals, and are evidently a late crystallization product.
A type b chondrule is illustrated in Figure 34.
This chondrule, 2 mm in diameter, consists largely
of prismatic crystals of anorthite, notably larger in
the outer part than in the core. Tiny crystals of
colorless spinel are included within the anorthite
crystals. A little forsterite, interstitial to the anor-
thite prisms, is present. The anorthite contains
about 0.5 percent Na2O, corresponding to about
5 percent of the NaAlSi3O8 component.
A single chondrule of type c has been found. It
was observed on a broken fragment of the meteorite,
40 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
FIGURE 33.?Gehlenite-anorthite-fassaite-spinel chondrule (type a) in the Allende meteorite, 24 mm in diameter; gehlenite is
the major phase and is present as large prismatic crystals, fassaite and anorthite occur as equant grains interstitial to the
gehlenite, and the spinel is disseminated as small crystals throughout the other minerals. Transmitted light.
NUMBER 5 41
FIGURE 34.?Anorthite-forsterite-spinel chondrule type b in the Allende meteorite (2.1-mm diame-
ter); white prismatic crystals are anorthite, the spinel is finely dispersed throughout the chondrule,
and the forsterite is interstitial to the anorthite crystals. Transmitted light.
being unusually large (8 mm in diameter) and a
uniform chalky-white color. It is so fine-grained
that the individual minerals could not be identified
in grain mounts in immersion oils. X-ray powder
photographs provided positive identifications for
sodalite, nepheline, clinopyroxene (probably fas-
saite), and olivine, and this mineralogy is consistent
with the bulk chemical composition of the chon-
drule (Table 3, column 8). By the X-ray procedure
of Yoder and Sahama (1957) the composition of the
olivine was found to be Fa2e, in good agreement
with the FeO/FeO+MgO ratio in the chemical
analysis. This occurrence of sodalite (Na8Al6Si6-
O24C12, with 7.3 percent Cl), the first recorded from
a meteorite, is particularly intriguing, since the
chlorine content of stony meteorites is usually low
(of the order of 100 ppm) and combined as chlora-
patite. Microprobe analyses have confirmed the
presence of nepheline and sodalite, and have pro-
vided some significant compositional data; in par-
ticular, potassium is notably fractionated between
the two minerals, the nepheline containing 1.5 per-
cent K and the sodalite less than 0.1 percent. The
extremely fine-grained texture of this chondrule
suggests that it solidified as a glass and subsequently
devi trifled.
The calcium- and aluminum-rich chondrules fre-
quently show narrow reaction rims against the
matrix, whereas the magnesium-rich chondrules do
not. This can probably be ascribed to the contrast
in composition between the calcium-rich chondrules
and matrix, whereas the magnesium-rich chondrules
are compositionally similar to the matrix. The oc-
currence of reaction rims implies that either the
chondrules or the matrix, or both, were quite hot
when the material aggregated.
A prominent feature of the Allende meteorite,
and one seen in few other chondrites, is the pres-
42 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
ence of irregular aggregates, white, pale gray, or
pale pink, usually up to a few millimeters in maxi-
mum dimension. Many of the aggregates are some-
what elongated or lensoid, and on large surfaces
there is a suggestion of subparallel arrangement of
the individual aggregates. Frequently the as^resrates
show narrow but distinct rims against the matrix;
these rims probably represent reaction zones with
the matrix.
Under the microscope these aggregates are seen
to be very fine-grained, the only prominent mineral
being rounded grains of pinkish spinel, which are
set in a microgranular aggregate of isotropic or
weakly birefringent material. X-ray diffraction, sup-
plemented by chemical analysis, has elucidated the
mineralogy. A bulk analysis of one 250-mg aggregate
(Table 3, column 9) shows that it is composed es-
sentially of calcium and magnesium alumino-sili-
cates, the composition being comparable with the
gehlenite-anorthite-fassaite-spinel chondrules. These
minerals are also prominent constituents of the
aggregates. In addition, some of them contain nephe-
line and sodalite, their occurrence evidently being
conditioned by the presence of appreciable amounts
of sodium in these aggregates. Some aggregates con-
tain considerable amounts of grossular, this mineral
also previously unrecorded in meteorites. The iden-
tification of grossular rests on its X-ray powder
photograph, giving a cell dimension of 11.85A,
identical with that of pure Ca3Al2Si3O12; it is
evidently too fine-grained to be recognized either
optically or with the microprobe, although it is cer-
tainly abundant in some aggregates. Fuchs (1969)
has described one aggregate consisting largely of
cordierite, with associated aluminous enstatite, anor-
thite, spinel, and sodalite. Marvin et al. (1970) have
identified spinel, hercynite, gehlenite, anorthite,
nepheline, diopside, fassaite, ferroaugite, and per-
ovskite from these aggregates.
The presence of grossular has considerable sig-
nificance in the interpretation of these aggregates.
Grossular is not known to crystallize from a melt;
it decomposes above 800?C into a mixture of wol-
lastonite, gehlenite, and anorthite. Yoder (1954)
crystallized it from a glass of the requisite composi-
tion at 800?C. The extremely fine-grained nature
of the aggregates suggest devitrification of a glass,
and the presence of grossular indicates that this
devitrification took place in some of them at or
below 800?C.
On broken surfaces of the Allende meteorite one
occasionally sees small areas somewhat darker than
the enclosing matrix; these areas are fine-grained
and have no visible chondrules, in contrast to the
highly chondritic matrix (Figure 25). They are
usually angular in outline, and range in size up to
about 3 cm across. In thin section under the micro-
scope these inclusions are seen to differ from the
surrounding material only in chondrule size; they
are essentially the same as the main mass of the
meteorite except that the chondrules are much
smaller, ranging up to 0.5-mm diameter, and are
somewhat sparser; the same irregular aggregates
are also present. The overall similarity is confirmed
by a chemical analysis (Table 3, column 4), which
is not significantly different from that of the bulk
meteorite. An X-ray diffractogram shows good oli-
vine peaks indicating a composition of Fa48, the
peaks however being somewhat skewed towards
lower Fa values.
FUSION CRUST.?A unique feature of the Allende
meteorite is the lustrous fusion crust discussed above
in the section on morphology (Figures 21-23). This
type of crust covers both spalled surfaces and areas
of normal fusion crust, indicating that it developed
late in the individual stone's passage through the
atmosphere. Its translucency and amber color led
us to assume first that it was a silicate glass of un-
usual eomposition. One of our specimens (NMNH
3811) was sectioned on this basis for microscopic
and electron microprobe investigation. It was a
great surprise to learn from preliminary examina-
tion of these sections that this strange crust actually
consisted of a thin film of silicate fusion crust en-
capsulating comparatively large volumes of a ma-
terial tentatively identified as a hydrocarbon. Our
main supporting evidence for this identification is
semiquantitative electron microprobe analysis in-
dicating that this "hydrocarbon" contains approxi-
mately 85 percent carbon and no other elements de-
tectable by the microprobe. It contains less carbon
than graphite and considerably more carbon than
the plastic mounting medium used in preparation
of the section. The material is essentially colorless
and is very soft, suggestive of paraffin. Until defini-
tive work can be done, we will refer to this material
as a hydrocarbon.
NUMBER 5 43
FIGURE 35.?A mosaic of photomicrographs of a polished thin section through
lustrous fusion crust, NMNH 3811. A 5-mm length of sample edge is revealed;
transmitted and reflected light.
Figure 35 is a mosaic of low-magnification photo-
micrographs of one of the polished thin sections
through lustrous fusion crust, revealing a 5-mm
length of sample surface. Lustrous fusion crust
overlies all but the extreme right edge of the area
shown. The area of fusion crust at the right of the
photomicrograph is lustrous material extending out
from the meteorite matrix approximately 0.2 mm
and bridging it for a distance of 1 mm. Smaller
areas are seen to the left, and in places this lustrous
crust thins to the point where it cannot be dis-
tinguished from the edge of the matrix. It is, how-
ever, continuous under higher magnification. The
outer edge of this crust assemblage is a thin film
of silicate glass containing a few matrix fragments
and heat-precipitated magnetite. The glass is too
thin to be seen in the photomicrograph, ranging in
thickness from 10 to 30 microns. Underneath the
outer glass layer is a region of hydrocarbon material
that may be continuous with the matrix, or may
adhere to the glass crust, but be separated from the
matrix by voids. The rim seen in the photograph
consists largely of hydrocarbon material colored by
a thin film of dark glass lying mainly below the sur-
face of the section.
The lustrous fusion crust spine protruding at the
left side of Figure 21a detached itself from specimen
NMNH 3812 during the course of this preliminary
investigation, providing material from a second
stone for examination. A longitudinal section was
prepared that yielded a trough-shaped volume of
sample with a surface 4 mm long and from 0.5 to 1
mm wide, ranging from 0.2 to 0.5 mm in depth,
with a total volume of hydrocarbon material of
approximately 0.8 mm3. The total volume of hy-
drocarbon in this one spine prior to loss of material
during section preparation was certainly greater
44 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
than 1 mm3. The glass crust surrounding this vol-
ume was similar in thickness to that noted above,
from 10 to 30 microns, accounting for probably
less than 10 percent of the total volume. If this
small fragment and the sections from specimen
NMNH 3811 are representative of lustrous fusion
crust in general, a large concentration of apparently
indigenous hydrocarbon material is present.
The source of the hydrocarbon material concen-
trated under the crust of specific Allende stones is
an intriguing problem. Bulk carbon analyses
(Table 3) indicate that essentially all of the carbon
in Allende is associated with matrix material and
dark inclusions, average values of several analyses
being 0.36 and 0.37 percent respectively. Work in
other laboratories has yielded some indirect infor-
mation on the nature of the carbon-containing com-
pounds in the Allende meteorite (Han et al. 1969;
Or6 and Gelpi, 1969; Ponnamperuma et al., 1970;
Levy et al., 1970). These workers agree that Allende
contains negligible quantities of solvent-extractable
organic substances. Oro and Gelpi (1969) have re-
ported, however, that direct volatilization or pyroly-
sis, or both, of Allende meteorite produces 10 to 20
ppm of a complex mixture of predominantly aroma-
tic hydrocarbons with smaller amounts of aliphatics
and other compounds. This result is confirmed by
Ponnamperuma et al. (1970). The conclusion of
Han et al. (1969) that an upper limit of 0.1 parts
per billion of possible indigenous organic matter in
Allende seems unwarranted, and our observation
of organic matter concentration within fusion crust
suggests a possible alternative to terrestrial contami-
nation as a source of some of the organic material
they found in surface samples.
The mechanism by which large quantities of or-
ganic material are concentrated under the fusion
crust of specific Allende stones presents a second
intriguing problem. The work on pyrolysis of Al-
lende material suggests that volatile organic com-
pounds may have been formed by pyrolysis during
ablation. A portion of this material would be ex-
pected to be lost to the atmosphere, while a part
of it would be driven into the porous meteorite. A
steep temperature gradient with depth into the
stone would account for condensation and accumu-
lation of volatiles slightly below the molten sur-
face. This zone of accumulation would move into
the meteorite as ablation progressed, concentrating
volatiles from large volumes of pyrolized (ablated)
meteorite. When ablation stopped and the interior
of the sample warmed, the organic accumulation
would expand and flow under the hardening crust,
producing features similar to the observed mor-
phology (Figures 21-23). If 10 to 20 ppm organic
product is realistic for this natural pyrolysis process,
one would expect something of the order of 100 g
of meteorite to be consumed or heated by the abla-
tion process for each milligram of organic material
accumulated. This seems excessive in terms of the
observed accumulations of several milligrams on a
small surface and leads us to wonder if these particu-
lar stones were not associated with inclusions richer
in volatile-producing compounds than normal
Allende material.
Chemical Composition
Chemical analyses have been made of two samples
of the bulk meteorite, a dark inclusion, the matrix,
a concentrate of chondrules, two individual chon-
drules, and a single aggregate. These results are
reported in Table 3.
Two different pieces were selected for the bulk
analysis. The first, 81 g (NMNH 3509), contained
several white aggregates; the second, 45 g (NMNH
3511), appeared more homogeneous. The samples
were ground in a tungsten-carbide mortar to pass
100 mesh. Several metallic grains larger than 100
mesh?totaling, however, less than 0.01 percent of
the sample?remained on the sieve. A split of each
sample was then analyzed for major and minor
constituents, following the general procedure out-
lined in an earlier paper (Jarosewich, 1966). The
two analyses are in excellent agreement with the one
exception that the first sample has somewhat more
A12O3, undoubtedly due to the presence of white
aggregates.
Noteworthy features of the bulk composition of
the Allende meteorite are the minute amount of
free metal, the presence of most of the nickel as
sulfide (a microprobe test of sulfide-free matrix
showed no detectable nickel in the silicates), the
high content of A12O3 and CaO (about 50 percent
greater than in the common chondrites), and the
low values for Na2O and K2O (about half that in
the common chondrites). The analysis is very similar
NUMBER 5 45
TABLE 3.?Analytical data on the Allende meteorite (E. Jarosewich, analyst). Values in weight percent
unless otherwise indicated
1
Bulk Analyses
Average Dark
of 1 & 2 Inclusion Matrix
6 7 8 9
Chon- Chon- Single
Chon- drule drule Aggre-
drules (Type a)* (Type c)5 gate9
NMNH Cat. Nos. 3509 3511 3509 3510 3510 3529 3509 3510
SiOa
TiOs
A12O8
CrsO8
FeO
MnO
MgO
CaO
NasO
K8O
P,O6
HSO(+)
HaO(_)
C
Cl
FeS1
NiS1
CoS1
Fe
Ni1
Co1
Total
D(g/cm8)
Total Fe
100 FeO
FeO + MgO
34.20
0.16
3.36
0.51
27.22
0.18
24.50
2.65
0.44
0.03
0.23
0.00
0.29
n.d.a
3.98
1.64
0.08
0.15
0.32
0.01
99.95
23.85
34.26
0.14
3.18
0.53
27.09
0.18
24.75
2.57
0.45
0.03
0.23
<0.1
0.00
0.29
n.d.
4.08
1.56
0.08
0.19
0.40
0.01
100.02
23.84
34.23
0.15
3.27
0.52
27.15
0.18
24.62
2.61
0.45
0.03
0.23
0.00
0.29
4.03
1.60
0.08
0.17
0.36
0.01
99.98
3.65
23.85
38
33.42
0.13
2.56
0.56
31.48
0.26
23.91
3.00
0.34
<0.01
0.31
n.d.
0.00
0.37
n.d.
1.95
2.26
0.06
n.d.
n.d.
n.d.
100.61
n.d.
25.71
42
33.11
0.13
3.07
0.55
29.68
0.22
21.42
2.67
0.44
0.03
0.25
n.d.
0.00
0.36
n.d.
5.49
1.05
0.09
0.40
0.85
0.02
99.83
n.d.
26.56
44
41.87
0.26
5.57
0.47
9.44
0.14
34.34
4.06
0.82
0.06
0.11
n.d.
0.00
0.00
n.d.
1.78
0.87
0.03
n.d.
n.d.
n.d.
99.82
n.d.
8.47
13
29.79
0.99
31.61
0.06
0.37
0.02
10.82
26.76
0.11
0.00
0.00
n.d.
0.00
0.06
n.d.
0.00
0.00
n.d.
n.d.
0.03
n.d.
100.62
3.25
0.29
40.2
0.12
17.8
0.2
8.8
0.1
15.2
5.3
10.6
0.6
n.d.
n.d.
n.d.
0.02
2.0*
n.d.
n.d.
n.d.
n.d.
0.05
n.d.
100.6 s
2.80
6.82
25
33.7
1.3
26.6
0.1
2.3
0.0
13.1
21.6
1.1
0.05
0.0
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
0.06
n.d.
99.9
3.18
1.76
n.d. = not determined
1 Calculated on basis of microprobe data for metallic phase: 68% Ni, 31% Fe and 1.6% Co. Metallic Fe was determined
chemically and metallic Ni and Co assigned on basis of analysis given above. Remainder of Ni and Co were calculated as
sulfides. The remaining S was calculated as FeS.
2 M. Quijano-Rico, Max-Planck-Institut, Mainz, determined Cl by neutron activation, obtaining 224 ppm in bulk sample
and 1.5% in chondrule type c. The difference between the 1.5 and 2.0% value is undoubtedly due to our use of a gravimetric
method with a very small sample (33 mg).
'Total of 100.99 corrected by 0.45, the Cl = 0.
4 1.5 g sample weight.
6 0.31 g sample weight.
6Gast et al. (1970) reported the following results on a similar inclusion (all in ppm): K 96.4, Rb 3.5, Sr 180, Ba 47.3,
La 4.63, Ce 11.5, Nd 8.40, Sm 2.82, Eu 1.30, Gd 3.87, Dy 4.90, Er 3.44, Yb 3.96.
46 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
to that of the Efremovka meteorite (Jarosewich
and Mason, 1969), except that Efremovka has about
7 percent nickel-iron and correspondingly less iron
combined in the silicates.
The analysis of the dark inclusion was performed
on a 1.3 g sample. As pointed out above, it is
closely comparable with that of the bulk material
of Allende, and it is very similar to the analysis
of Mokoia (Wiik, 1956), which is another Type III
carbonaceous chondrite.
The analyzed material of the matrix, and the
chondrule concentrate, were prepared by carefully
crushing and sieving a number of fragments of the
meteorite. Gentle crushing releases a large amount
of the matrix in the form of fine dust which passes
through a 325-mesh screen, and a sample of this
was used for the analysis (2.9 g). The chondrules
were concentrated in the ?10 ?|? 30 mesh fraction
(approximately 0.6-2.0 mm diameter), a 1.5 g sam-
ple being obtained. The analyzed chondrules were
not entirely free from matrix; the FeO/FeO-f-MgO
molecular percentage in the analysis is 13, so the
average for the chondrules themselves will be con-
siderably less.
The composition of the matrix is not markedly
different from that of the bulk meteorite, which is
reasonable, since matrix forms about 60 percent by
volume of the meteorite and somewhat more by
weight. The matrix clearly contains most of the
metal and sulfide phases. The FeO/FeO-|-MgO
molecular percentage of 44 is consistent with the
average fayalite percentage (50) in matrix olivine,
since the matrix fraction contains other minerals
(calcium aluminum silicates and spinel) with low
FeO/FeO-f-MgO ratios.
The composition of the chondrule concentrate
shows marked contrasts with that of the matrix,
particularly in total Fe content, MgO content, and
in FeO/FeO-j-MgO molecular percentage. This of
course reflects the fact that most of the chondrules
are made up largely of forsteritic olivine, with minor
amounts of clinoenstatite. The calcium-rich chon-
drules account for the somewhat higher CaO and
A12O3 contents. Sulfides are notably lower in the
chondrule concentrate; they usually occur on the
surfaces of the chondrules, but are sometimes en-
closed within them. Metallic nickel-iron was ob-
served both in matrix and in chondrules.
The aggregates and the calcium-rich chondrules
present the most marked chemical contrasts to the
rest of the meteorite. They contain up to ten times
the amount of calcium and aluminum (and titan-
ium) as in the meteorite as a whole. They are quite
unlike any composition that has yet been recorded
from stony meteorites. The composition is also
quite unlike any terrestrial rocks, except for some
rare calcsilicate hornfels, such as that described by
Knopf and Lee (1957). The nearest analogy is to
some blast-furnace slags. The composition given by
the analysis of the type a chondrule corresponds
to the low-melting region in the CaO-MgO-Al2O3-
SiO2 system, and the association anorthite-gehlen-
ite-fassaite-spinel is consistent with crystallization
of a melt of this composition. The origin of such a
melt from meteoritic matter, however, presents
certain problems. Both calcium and aluminum are
minor elements in average chondritic matter, being
present at about the 1 percent level. One procedure
for enriching them is the removal of large amounts
of magnesium as olivine and clinoenstatite; neither
of these minerals incorporates appreciable amounts
of calcium and aluminum. This can be seen from
the composition of the glass in the magnesium-rich
chondrules, quoted earlier, although this glass is
somewhat richer in SiO2 than the calcium-rich chon-
drules and aggregates. The composition of the sin-
gle aggregate analyzed is rather similar to the type a
chondrule, except for a much higher sodium con-
tent, expressed in its mineralogy by the presence
of nepheline and sodalite. The type c chondrule has
a very distinctive composition, marked by high
sodium and potassium (about twenty times that
in the bulk meteorite), and a remarkable chlorine
content, more than fifty times that in the bulk mete-
orite. These extreme fractionations pose some in-
triguing questions regarding the processes involved
in the formation of such chondrules, presuming the
parent material was essentially undifferentiated. In
this connection it is probably significant that an-
orthite, gehlenite, and spinel are among the first
solids to condense from a gas of cosmic composition.
Larimer and Anders (1970) have convincingly
argued for chemical fractionations in meteorites by
addition or removal of early formed phases. They
find that under the probable conditions within the
cooling solar nebula MgAl2O4 condenses first at
1470?K, followed by Ca, Ti, and Al minerals. At
NUMBER 5 47
1350?K Mg2SiO4 begins to condense, followed by
MgSiO3 and nickel-iron at 1300?K. The presence of
spinel, anorthite, gehlenite, and fassaite as chon-
drules and aggregates in Allende and other Type III
carbonaceous chondrites indicates that this class of
meteorites owes some of its unique characteristics
to the incorporation of this early formed condensate.
COMPARATIVE CHEMICAL DATA.?Additional chemical
data on the Allende meteorite are given in Tables
4 and 5. Table 4 lists concentrations for a number
of trace elements determined by spark source mass
spectrometry. In Table 5 our chemical data are
compared with the published data of King (1969);
Emery et al. (1969), Morgan et al. (1969), and
Wakita and Schmitt (1970). In general, agreement
between workers and the various techniques repre-
sented is excellent. The work reported by Emery
et al. contains several discrepancies, however, per-
haps due either to unrepresentative sampling or
inadequate methods. Table 5 also contains data on
elements that we have not determined.
TABLE 4.?Trace element data on the Allende
meteorite1
Values in parts per million
B 1 Y ..
Sc 11 Zr .
V 70 Nb
Cu 100 Mo
Zn ....... 25 Sn .
Ga 7 Sb .
Ge 15 Ba .
As 2 La .
Br 0.5 Ce .
Rb 0.9 Pr .
Sr 13
2 Nd 0.9
9 Sm 0.5
0.9 Eu 0.1
2 Gd 0.6
0.7 Tb 0.09
0.2 Dy 0.6
5 Ho 0.1
0.7 Er 0.3
1 Yb 0.4
0.2 Pb 1
Elements detected but not of suitable intensity for meas-
urement, or suitable standards were lacking.
Se, Ru, Rh, Pd, Ag, Cd, In, I, Te, Cs, Tm, Hf, W,
Re, Os, Ir, Pt, Au, Tl, Bi, Th, U.
Hg was not detected.
Ta could not be confirmed because of possible contamina-
tion from spark stand.
Lu could not be determined because it was used as an
internal standard.
1B. Mason and A. L. Graham, analysts. Determinations
made by spark-source mass spectrometer while B. Mason
was a Visiting Research Fellow in the Department of Geo-
chemistry, Australian National University, Canberra, 1969.
Gast et al. (1970) have reported on several trace
elements in a calcium-rich inclusion in Allende.
Their values are given in a footnote to our bulk
analysis of one of these inclusions (Table 3).
Discussion
The classification of the Allende meteorite presents
some interesting problems. Visual inspection im-
mediately suggests that it is a carbonaceous chon-
drite, specifically a Type III as defined by Wiik
(1956) ?it is closely similar to others of this type
such as Vigarano and Mokoia. Chemical and min-
eralogical examination confirms this similarity. It
has the Si/Mg ratio characteristic of the carbonace-
ous chondrites, and it falls within the range of other
Type III meteorites (Figure 36). The conjunction
of iron-poor olivine in chondrules with iron-rich
olivine in matrix is also distinctive for these mete-
orites. Yet the carbon content (0.3 percent) is low
for a carbonaceous chondrite, lower in fact than
in a number of chondrites whose Si/Mg ratio places
them in the ordinary (bronzite and hypersthene)
classes.
Recently Van Schmus and Wood (1967) have
provided a revised classification for the chondrites,
based on a two-dimension grid combining chemical
composition and textural and mineralogical fea-
tures. They divide the carbonaceous chondrites into
four types, designated Cl, C2, C3, and C4. Types
Cl, C2, C3 include the same meteorites as Wiik's
Types I, II, III, except that Kaba and Mokoia are
considered C2 rather than Type III. This produces
some ambiguity in the classification of Allende,
since it is very similar to Mokoia in chemical and
mineralogical composition; however, separating Mo-
koia and Kaba from the other C3 chondrites seems
poorly based, since both these meteorites belong
with the other Type III carbonaceous chondrites in
terms of bulk chemical composition. Their matrix
consists largely of iron-rich olivine, whereas in
Type II-C2 chondrites the matrix is hydrated mag-
nesium-iron silicate (serpentine or chlorite). Wiik's
classification appears more suited to the carbonace-
ous chondrites than that of Van Schmus and Wood.
Whereas the latter classification seems well adapted
for the other classes of chondrites, its application to
the carbonaceous chondrites presents certain incon-
sistencies?probably because the sequence Cl to C4
48 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
for the carbonaceous chondrites reflects different
conditions of genesis than the E3-E6, H3-H6,
L3-L6, and LL3-LL6 sequences for the other classes
Van Schmus-Wood classification are known only
in the carbonaceous chondrites; their absence in
the other chondrite classes suggests a fundamental
of chondrites. Significantly, types 1 and 2 in the genetic difference.
16 18 20 22 24
Mg 0, weight per cent
26 28
NUMBER 5 49
Van Schmus (1969) has recognized two distinct
subtypes of the C3 chondrites, provisionally called
the Vigarano and the Ornans subtypes. He writes:
"The most obvious distinction between the sub-
types is textural. The chondrites belonging to the
Vigarano subtype may be characterized as having
large (1-2 mm diameter) 'spongy' chondrules em-
bedded in an abundant, fine-grained, opaque ma-
trix; 'spongy' chondrules contain numerous sphe-
roidal droplets of nickel-iron or sulfide dispersed
through the silicate chondrule. The chondrites
belonging to the Ornans subtype may be character-
ized as a close-packed aggregate of small (0.2-0.5
mm diameter) metal-poor chondrules with fine-
grained, opaque matrix packed between the chon-
drules."
The distinction is a useful one, and can usually
be easily made by visual inspection. The Vigarano
subtype (designated A in Table 6) has large prom-
inent chondrules in a black or dark gray matrix,
whereas the Ornans subtype (B) has an evenly
granular texture of close-packed chondrules and a
uniform light- or medium-gray color. The chemical
analyses of the different meteorites show a close
similarity for the major components, and for many
of the minor and trace elements; carbon, however,
is quite variable, as is H2O-|- (the H2O-(- reported
in the chemical analyses is probably not combined
water but H (and O) combined with carbon in an
organic complex?unweathered meteorites of this
type do not contain hydrated minerals, and H2O-f-
is generally correlated with carbon content). The
recorded densities show a considerable range for
a group of meteorites of rather uniform composi-
tion. Some of this range can perhaps be ascribed
to the difficulties of determining accurate densities
of porous and fine-grained material; however, the
higher densities are characteristic of subtype B mete-
orites, which generally have an appreciable content
(up to 5 percent) of free nickel-iron, whereas sub-
type A meteorites usually have little or no free
nickel-iron and a larger content of carbonaceous
material.
FIGURE 36.?SiO2 plotted against MgO (weight percentages) for chemical analyses of chondrites;
the diagonal lines are for Si/Mg atomic ratios of 1.0 and 1.1. A is the field for 36 analyses of
bronzite chondrites, B the field for 68 analyses of hypersthene chondrites, the black squares
being the means for each group. The analyses of carbonaceous chondrites (-}-) and enstatite chon-
drites (?) are plotted individually, as follows:
Type I
1.
2.
3.
Alais
Orgueil
Ivuna
Type II
4.
5.
6.
7.
8.
9.
10.
11.
Nawapali
Boriskino
Cold Bokkeveld
Erakot
Mighei
Murray
Santa Cruz
AlRais
Type III
12.
13.
14.
15.
16.
17.
18.
Grosnaja
Kaba
Vigarano
Lance
Karoonda
Mokoia
Ornans
19. Renazzo
20. Felix
21. Efremovka
22. Coolidge
23. Warrenton
Y. Allende (bulk analysis)
Z. Allende (black inclusion)
Enstatite
Chondrites
24. St. Sauveur
25. Indarch
26. AdhiKot
27. St. Marks
28. Abee
29. Atlanta
30. Daniel's Kuil
81. Hvittis
32. Khairpur
33. Blithfield
34. Jajh deh Kot Lalu
50 SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
TABLE 5.?Comparison of chemical data on the Allende meteorite. (Values in
percent unless noted, ppm = parts per million, ppb ? parts per billion)
Element King1
Emery
et al?
Morgan
et al?
Wakita and
Schmitt *
This
Work
O .
Si .
Ti .
Al .
Cr .
Mn
Mg
Ca
Na
K .
C ..
Ni .
Co
Fe .
B .
Sc .
V .
Rb
Zr .
Cd
In .
Cs .
Ba .
Hf
W .
Ir .
Pt .
Au
U .
La .
Ce .
Pr .
Nd
Sm
Eu
Gd
Tb
Dy
Ho
Er .
Tm
Yb
Lu
Y
15.9
0.11
1.75
0.42
0.13
14.5
1.8
0.30
0.025
0.27
1.41
0.070
23.6
0.010
0.017
0.006
0.001
34.7
15.9
1.35
0.42
0.17
13.1
1.88
0.18?0.12
1.15
0.06
27.8
10 ppm
^0.7 ppm
19 ppb
35.9?0.7
15.81 ?0.35
0.39?O.l
0.33+0.01
1.43?0.09
684?17 ppm
612?54 ppm
24.4?0.4
12.2?0.2 ppm
12.0?0.7 ppm
0.16?0.02 ppm
0.15?0.02 ppm
0.71 ?0.03 ppm
1.8?0.2 ppm
0.26?0.01 ppm
1.71 ?0.05
3680?100 ppm
1450?40 ppm
2.0?05
337O?1OO ppm
640?20 ppm
21.9?0.4
11.0?0.5 ppm
130?10ppm
1.3?0.1 ppm
0.19?0.01 ppm
0.027?0.001 ppm
0.06?0.01 ppmB
0.44?0.02 ppm
1.25 ?0.06 ppm
0.20?0.01 ppm
0.91 ?0.05 ppm
0.29?0.01 ppm
0.107 ?0.005 ppm
0.43?0.02 ppm
0.074?0.O05 ppm
0.42?0.02 ppm
0.12?0.01 ppm
0.31 ?0.02 ppm
0.049?0.001 ppm
0.32?0.02 ppm
0.058?0.002 ppm
3.0?0.1 ppm
[36.6]*
16.00
0.09
1.73
0.36
0.14
14.85
1.87
0.33
0.02
0.29
1.39
0.06
23.85
1 ppm
11 ppm
70 ppm
0.9 ppm
9 ppm
5 ppm
0.7 ppm
1 ppm
0.2 ppm
0.9 ppm
0.5 ppm
0.1 ppm
0.6 ppm
0.09 ppm
0.6 ppm
0.1 ppm
0.3 ppm
0.4 ppm
2 ppm
?Oxygen by difference.
1 King (1969).
a Emery et al. (1969).
8 Morgan et al. (1969).
* Wakita and Schmitt (1970).
6 Changed from published values, Schmitt, personal communication (1970).
NUMBER 5 51
TABLE 6.?The Type HI carbonaceous chondrites
Name
Weight
Date of fall recovered, kg D(g/cms) %C
Allende (Mexico)
Bali (Cameroon)
Coolidge (Kansas)
Efremovka (U.S.S.R.) .
Grosnaja (U.S.S.R.) ...
Kaba (Hungary) ......
Leoville (Kansas)
Mokoia (New Zealand)
Vigarano (Italy)
Felix (Alabama)
Kainsaz (U.S.S.R.) ...
Karoonda (Australia)
Lance (France)
Ornans (France)
Warrenton (Missouri)
Subtype A or Vigarano subtype
8/2/1969
22/12/1907
find
find
28/6/1861
15/4/1857
find
26/11/1908
22/1/1910
>2000
3?
4.5
21
3.5
3
8.1
5
16
3.65
3.58
3.53
3.53
3.49
3.40
3.48
3.55
3.42
0.29
0.56
0.19
0.76
0.56
1.99
0.77
0.75
1.15
Subtype B or Ornans subtype
15/5/1900
13/9/1937
25/11/1930
23/7/1872
11/7/1868
3/1/1877
3
>200
42
52
6
50?
3.78
3.76
3.57
3.64
3.61
3.64
0.64
0.61
0.10
0.58
0.35
0.30
The Si/Mg ratio is slightly lower in the Type III
carbonaceous chondrites than in the ordinary chon-
drites, and the FeO content is considerably higher.
This has far-reaching consequences, since the bulk
composition is now very undersaturated with respect
to SiO2, so much so that feldspathoids appear in
place of feldspar. Specifically, sodium in the com-
mon chondrites is present as sodic feldspar (NaAl-
Si3O8, with about 10 percent of the CaAl2Si2O8
component), whereas in Allende (and probably in
other chondrites of this group) sodium is present
as the NaAlSiO4 component (nepheline), and the
feldspar is calcic, being almost pure CaAl2Si2O8.
It has always been something of a puzzle why feld-
spar composition in stony meteorites is remarkably
quantized, being sodic plagioclase in the common
chondrites and calcic plagioclase in most achon-
drites, intermediate compositions (common in ter-
restrial rocks) being essentially unknown in mete-
orites. Possibly this segregation of sodium and
calcium into distinct phases in the Allende mete-
orite is a clue to the solution of this puzzle.
The Allende meteorite is noteworthy as being the
largest of the carbonaceous chondrites yet recovered.
As such, it offers an unequaled opportunity for a
thorough investigation. Our work has emphasized
the significance of the contrasts between matrix,
chondrules, and aggregates. These peculiar aggre-
gates are certainly the most intriguing feature. The
larger ones (those readily visible to the naked eye)
are not uniformly distributed through the meteorite,
being certainly more prominent in some specimens
than others. Similar aggregates have been noted in
other Type III carbonaceous chondrites. Sztr6kay
et al. (1961) noted them in the Kaba meteorite, but
because of scarcity of material they were not closely
investigated. Michel-L^vy (1968, 1969) has described
mineralogically similar aggregates from the Vigar-
ano and Lancd meteorites. We have observed them
in the Bali meteorite. The meteorite which re-
sembles Allende most closely, however, is Leoville.
Only a brief abstract describing this meteorite
has so far appeared (Keil et al., 1969), but conver-
sations with Professor Keil and inspection of speci-
mens of Leoville confirm the remarkable similarity.
Especially significant is the presence of numerous
white aggregates in Leoville, essentially similar in
chemical and mineralogical composition to those
in Allende.
52
Literature Cited
Barcena, M.
1876. On Certain Mexican Meteorites. Proceedings of the
Academy of Natural Sciences of Philadelphia, 28:
122-126.
Buchwald, V. F.
1968. A Reclassification of Mexico's Iron Meteorites. Pro-
gram of the Annual Meeting of the Geological So-
ciety of America, Mexico City, (Abstract), pages
40-41.
Carr, M. H.
1970. Atmospheric Collection of Debris from the Revel-
stoke and Allende Fireballs. Geochimica et Cosmo-
chimia Ada. 34:689-700.
Durrani, S. A., and C. Christodoulides
1969. Allende Meteorite: Age Determination by Thermo-
luminescence. Nature, 223:1219-1221.
Emery, J. F., J. E. Strain, G. D. O'Kelley, and W. S. Lyon
1969. Non-Destructive Neutron Activation Analysis of the
Allende Meteorite. Radiochemical Radioanalytical
Letters, 1:137-141.
Farrington, O. C.
1915. Catalogue of the Meteorites of North America to
January 1, 1909. National Academy of Sciences Mem-
oir, 13:513 pages, 36 plates.
Fireman, E. L.
1969. Freshly Fallen Meteorites from Portugal and Mexico.
Sky and Telescope, 37(5): 1-4.
Fireman, E. L., J. De Felice, and E. Norton.
1970. Ages of the Allende Meteorite. Geochimica et Cosmo-
chimica Ada, 34:873-881.
Fletcher, L.
1890. On the Mexican Meteorites, with Especial Regard to
the Supposed Occurrence of Wide-Spread Meteoritic
Showers. Mineralogical Magazine, 42:91-175.
Fuchs, L. H.
1969. Occurrence of Cordierite and Aluminous Orthoen-
statite in the Allende Meteorite. American Min-
eralogist, 54:1645-1653.
Gast, P. W., N. J. Hubbard, and H. Weismann
1970. Chemical Composition and Petrogenesis of Basalts
from Tranquillity Base. Geochimica et Cosmo-
chimica Ada, Supplement I, Proceedings of the
Apollo 11 Lunar Science Conference, Houston, Texas
January 5-8,1970, 2:1143-1163.
Han, J., B. R. Simoneit, A. L. Burlingame, and M. Calvin
1969. Organic Analysis on the Pueblito de Allende Me-
teorite. Nature, 222:364-365.
Hey, M. H.
1966. Catalogue of Meteorites. The British Museum (Nat-
ural History), London. Printed in Great Britain by
The Alden Press (Oxford) Limited.
Hindley, K. B., and H. G. Miles
1970. Fireball and Meteorite of April 25, 1969. Nature,
225:255-257.
Jarosewieh, E.
1966. Chemical Analyses of Ten Stony Meteorites. Geo-
chimica et Cosmochimica Ada. 30:1261-1265.
SMITHSONIAN CONTRIBUTIONS TO THE EARTH SCIENCES
Jarosewieh, E., and B. Mason
1969. Chemical Analyses with Notes on One Mesosiderite
and Seven Chondrites. Geochimica et Cosmochimica
/f eta, 33:411-414.
Keil, K., G. I. Huss, and H. B. Wiik
1969. The Leoville, Kansas, Meteorite: A Polymict Breccia
of Carbonaceous Chondrites and Achondrite. In
Meteorite Research (editor P. M. Millman). D.
Reidel Publishing Company, Dordrecht, Holland.
Page 217. (Abstract.)
King, E. A., Jr.
1969. Petrography and Chemistry of the Pueblito de
Allende Meteorite. EOS Transactions, American Geo-
physical Union, 50: 459.
King, E. A., Jr., E. Schonfeld, K. A. Richardson, and J. S.
Eldridge
1969. Meteorite Fall at Pueblito de Allende, Chihuahua,
Mexico: Preliminary Information. Science, 163:928-
929.
Knopf, A., and D. E. Lee
1957. Fassaite from near Helena, Montana. American
Mineralogist, 42:73-77.
Kulik, L. A.
1922. Report on Meteorite Expedition Conducted Between
May 19, 1921 and November 1922. Academiia nauk
S.S.S.R., Leningrad. Bulletin, series 6, 16:391-410.
Kullerud, G., and A. El Goresy
1969. Sulfide Assemblages in the Odessa Meteorite. Car-
negie Institution Year Book, 67:187-189.
Larimer, J. W., and E. Anders
1970. Chemical Fractionations in Meteorites?III. Major
Element Fractionations in Chondrites. Geochimica
et Cosmochimica Ada, 34:367-387.
Levy, R. L., C. J. Wolf, M. A. Grayson, J. Gilbert, E. Gelpi,
W. S. Updegrove, A. Zlatkis, and J. Or6.
1970. Organic Analysis of the Pueblito de Allende Meteor-
ite. Nature, 227:148-150.
Marvin, U. B., J. A. Wood, and J. S. Dickey, Jr.
1970. Ca-Al Rich Phases in the Allende Meteorite. Earth
and Planetary Science Letters, 7:346-350.
McCrosky, R. E., A. Posen, G. Schwartz, and C. A. Tougas
1969. Preliminary Comments on the Trajectory, Orbit and
Initial Mass of the Allende Meteorite. EOS Trans-
actions, American Geophysical Union, 50:458 (Ab-
stract).
Meighan, I. G., and P. S. Doughty
1969. Recent Fall of the Bovedy Meteorite, Northern
Ireland. Nature, 223:24-29.
Michel-Levy, C.
1968. Un chondre exceptionnel dans la m?t6orite de
Vigarano. Bulletin de la Socidti francaise de Min-
eralogie et de Cristallographie, 91:212-214.
1969. Etude mineralogique de la chondrite C III de Lance".
In Meteorite Research (editor P. M. Millman). D.
Reidel Publishing Company, Dordrecht, Holland.
Pages 492-499.
Morgan, J. W., T. V. Rebagay, D. L. Showalter, R. A. Nad-
karni, D. E. Gillum, D. M. McKown, and W. D.
Ehmann
NUMBER 5
1969. Allende Meteorite: Some Major and Trace Element
Abundances by Neutron Activation Analysis. Na-
ture, 224:789-791.
Nininger, H. H.
1936. The Pasamonte, New Mexico, Meteorite. Popular
Astronomy, 54(6): 1-7.
Or6, J., and E. Gelpi
1969. Organic Analysis of the Pueblito de Allende Me-
teorite. Meteoritics, 4:287 (Abstract).
Ponnamperuma, C, K. Kvenvolden, S. Chang, R. Johnson,
G. Pollock, D. Philpott, I. Kaplan, J. Smith, J. W.
Schopf, C. Gehrke, G. Hodgson, I. A. Breger, B.
Halpern, A. Duffield, K. Krauskopf, E. Barghoorn,
H. Holland, and K. Keil
1970. Search for Organic Compounds in the Lunar Dust
from the Sea of Tranquillity. Science, 167:760-762.
Prince, A. T.
1954. Liquidus Relationships on 10 Percent MgO Plane
of the System Lime-Magnesia-Alumina-Silica. Amer-
ican Ceramic Society Journal, 37:402-408.
Rancitelli, L. A., R. W. Perkins, J. A. Cooper, J. H. Kaye,
and N. A. Wogman
1969. Radionuclide Composition of the Allende Meteorite
from Nondestructive Gamma-Ray Spectrometric
Analysis. Science, 166:1269-1272.
53
Sztr6kay, K. I., V. Tolnay, and M. Foldvari-Vogl
1961. Mineralogical and Chemical Properties of the Car-
bonaceous Meteorite from Kaba, Hungary. Ada
Geologica, 7:57-103.
Van Schmus, W. R.
1969. Mineralogy, Petrology, and Classification of Types
3 and 4 Carbonaceous Chondrites. In Meteorite Re-
search (editor P. M. Millman). D. Reidel Publishing
Company, Dordrecht, Holland. Pages 480-491.
Van Schmus, W. R., and J. A. Wood
1967. A Chemical-Petrologic Classification for the Chon-
dritic Meteorites. Geochimica et Cosmochimica Ada,
31:747-765.
Wakita, H., and R. A. Schmitt
1970. Rare-Earth and other Elemental Abundances in the
Allende Meteorites. Nature, 227:478-479.
Wiik, H. B.
1956. The Chemical Composition of Some Stony Meteor-
ites. Geochimica et Cosmochimica Ada, 9:279-289.
Yoder, H. S.
1954. Garnets and Staurolite. Carnegie Institution Year
Boofc, 53:120-121.
Yoder, H. S., and T. Sahama
1957. Olivine X-ray Determinative Curve. American Min-
eralogist, 42:475-491.
?ft U.S. GOVERNMENT PRINTING OFFICE: 1970 O?40O-O8B