The so-called impact craters of Mauritania
Aouelloul, Tenoumer and Temimichat

Norbert Brügge, Germany
Dipl. - Geol.

Update: 27.10.2016

Geology of Mauritania

Mauritania comprises four major geological domains. The central part of the Reguibat Shield - an uplifted part of the West African Craton which has been stable since 1700 Ma and dominates the northern third of the country's surface geology.
The N-S Mauritanides Orogenic Belt. The western margin of this orogen is concealed beneath coastal basin sediments.
The western part of the Taoudeni Basin infilled by Neoproterozoic to Devonian sedimentary strata unconformably overlying crystalline rocks of the Palaeoproterozoic to Archaean West African Craton. The continental to shallow marine Taoudeni Basin does not exceed 5000 m in thickness, and though it varies somewhat, comprises a remarkably homogenous lithological sequence.
A western Mauritanian Coastal Basin infilled by Mesozoic-Cenozoic sediments.


Reguibat Shield

The West African Craton is primarily composed of two shields, the Reguibat Shield in the north of Mauritania and the Leo Shield in Liberia and Sierra Leone, separated by the sedimentary (Upper Proterozoic to Palaeozoic) Taoudeni Basin.
The Reguibat Shield dominates the northern third of the country's surface geology. It constitutes a very extensive portion of Precambrian crust, bounded on its western side by the Mauritanides Belt and to the south it is overlain by late Precambrian sedimentary rocks of the Cambro-Ordovician and Devonian series.
The Reguibat Shield is divided into two main parts:
The eastern shield or 'Eburnean terrane', with ages of 2000 ±200 Ma corresponding with the Birimian tectonothermal events and predominantly comprising intrusive granites in the west and volcanic formations in the east.
The western shield or 'Archaean terrane' mainly with ages of ≥2500 Ma and predominantly comprising migmatites gneisses, granitoids, ferruginous quartzites and Banded Iron Formation (BIF), amphibolites and feldspathic gneisses. The highest grade gneissic and granulite terrain is the Archaean basement located in the western part of the shield in the Akjoujt area where it outcrops as poorly granular, migmatic felsic gneiss and coarsely foliated orthogneisses.

Taoudeni Basin

The Taoudeni Basin has experienced only one period of modern oil exploration. Agip and Texaco were each awarded blocks in 1970 with seismic and in 1974 with two drillings.
The Taoudeni Basin is a broad intracratonic sag which appears to have formed in response to the Pan-African Orogeny. The basin initiated in the Infracambrian, and continued to develop throughout the Palaeozoic until the Carboniferous. Despite the proximity of the Pan-African, Caledonian and Hercynian Orogenic belt, there is no evidence of significant deformation in the basin. For example, the Hercynian Orogeny, which formed the Mauritanide Fold Belt, is pinned by a fore-thrust in the eastern part of the fold belt, and significant deformation does not appear to have propagated eastwards into the present day Taoudeni Basin.
There are some broad, long-lived structural highs through the basin: The northeast striking Abolag–Ouasa High compartmentalises the basin into two depocentres: The Maqteir Depression to the west, and the main Taoudeni Depocentre to the east. These are broad, relatively unstructured depocentres, although there are some large features evident in the 2D seismic data (El Mrayer High).

In southernmost Mauritania, a Cretaceous rift basin is superimposed on the Taoudeni Basin. This poorly known basin is only inferred from gravity data. The Nara Rift is a northeast-striking feature straddling the southern Mauritanian border.

Crater Aouelloul

Crater Aouelloul: Latitude 20° 15' N; Longitude 12° 41' W

The crater Aouelloul is located in the Adrar region of the western Sahara Desert, Mauritania. The crater is one of the smaller known craters, having a diameter of 390 m. The rim is well-defined and rises 15-25 m above the local topography, and 53 m above the crater floor. The crater is located in Ordovician sandstones and quartzite, and is filled with sandy silt and Aeolian sand. Estimates of the thickness of this sedimentary fill are around 23 m.

While breccias are not found at Aouelloul, impact glass can be found on the south, southeast, and north outer part of the crater rim. Because the glass is enriched in siderophile elements, has a low water content, and contains lechatelierite, it has been interpreted as an impact glass (?).
Proving Allouelloul’s impact origin has proved extremely difficult. Because the impact occurred in sandstone, the shock wave energy that normally forms planar deformation features was largely dissipated within the porous rock. Therefore, different stages of shock metamorphism occur, and PDF's rarely form. Petrographic study of samples from Aouelloul show shattered and fractured quartz, but no distinct PDF's.
Instead of relying on petrography, scientists have used chemical analysis to prove an impact origin here (?). While most of the glass is compositionally similar to the local sandstone, the glass does have high concentrations of Fe, Co, Ni, and Ir. In addition, the Re-OS isotope ratios of the glass were measured. This method examines the admixture of small amounts of recondensed material with low Os ratios to "target" rocks with high Os ratios (?).
An absolute age of 3.1±0.3 million years was obtained by fission track and K-Ar dating of the impact glass.

Note: The chemistry points that this glass is an alkaline-poor rhyolitic volcanic glass


Gucsik et al., 2004 Koeberl & Auer
range (16) average range (7) average


81.70 87.5 85.67 84.90 86.90 86.04


6.85 9.60 7.76 6.36 7.57 6.87


0.14 1.08 0.49 0.43 0.53 0.47


1.43 3.25 2.13 2.20 2.74 2.36


0.03 0.30 0.16 0.02 0.05 0.03


0.15 1.59 0.79 0.98 1.32 1.15


0.19 0.52 0.40 0.31 0.40 0.34


0.02 0.39 0.11 0.17 0.27 0.23


1.61 2.19 1.83 2.06 2.56 2.32


Glasses with included differently clasts

Demolished and impregnated sandstone



The enigmatic Chinguetti "meteorite" fits to this structure because it is a multi-sedimentary breccia.

Sources: &


Cambridge Conference Correspondence:

C. Koeberl, W.U. Reimold, S.B. Shirey: The Aouelloul crater, Mauritania: On the problem of confirming the impact origin of a small crater
METEORITICS & PLANETARY SCIENCE, 1998, Vol.33, No.3, pp.513-517

The impact origin of small craters in sedimentary rocks is often difficult to confirm because of the lack of characteristic shock metamorphic features. A case in point is the 3.1 Ma Aouelloul crater (Mauritania), 390 m in diameter, which is exposed in an area of Ordovician Oujeft and Zli sandstone. We studied several fractured sandstone samples from the crater rim for the possible presence of shock metamorphic effects. In thin section, a large fraction of the quartz grains show abundant subplanar and planar fractures. Many of the fractures are healed and are evident only as fluid inclusion trails. A few grains showed sets of narrow and densely spaced fluid inclusions >trails in one (rarely two) orientations per grain, which could be possible remnants of planar deformation features (PDFs), although such an interpretation is not unambiguous. In contrast, an impact origin of the crater is confirmed by Re-Os isotope studies of the target sandstone and glass found around the crater rim, which show the presence of a distinct extraterrestrial component in the glass (?).

Crater Tenoumer

The Tenoumer crater is about 1,900 m in diameter and is located in the western Sahara desert, Mauritania. The crater is almost perfectly round in shape, and the rim is 110 m high from bottom to top.
Tenoumer is located on a peneplain consisting of Precambrian gneisses and granites. A thin layer of sediments that are Pliocene or younger covers the Precambrian rocks. Tenoumer is located on top of the Pliocene sediments.
Due to the rare occurrence of igneous materials outside the crater a
volcanic origin is possible. Small glass inclusions are also present.

Basement rocks have been found in outcrops outside of the crater, located in individual bodies up to 20 m long. These rocks are dark grey, vesicular, and contain altered gneiss and granite clasts.

According to the analysis data in PRATESI et al. we have a clearly volvanic andesitic melt of a diorite magma.

Oxide (wt%)

 MSP 2314 MSP 2315 MSP 2317 NMNH 113029-15 NMNH 113029-18 NMNH 113029-62


58.02 62.17 58.24 60.92 60.44 61.78


12.51 12.60 12.42 13.41 11.08 12.51


0.83 0.74 0.86 0.39 0.65 0.77


0.09 0.09 0.10 0.04 0.08 0.09


7.39 6.83 7.67 3.89 6.70 7.26


5.68 5.17 5.94 1.85 5.04 5.26


6.89 5.47 6.50 7.80 7.91 5.27


3.39 3.35 3.53 3.59 2.85 3.39


1.72 1.90 1.76 1.98 1.48 1.83


Chemical compositions of impact melt breccias and target rocks from the Tenoumer impact crater, Mauritania
G. Pratesi, M.Morelli, A.P. Rossi, G.G. Ori; Meteoritics & Planetary Science, Volume 40, Issue 11, (2005)

The variety of brecciated rocks so far recognized in the Tenoumer area is quite restricted. The collected Tenoumer samples are all melt-matrix breccias, containing both lithic and melt components.
These breccias are dark gray and display a wide range of porosity, with voids varying in size. Thus, the appearance of the samples may be dense, slightly vesicular or, more rarely, so strongly vesicular that they resemble a pumice. In all cases, the void density is probably related to the original amount of co-existing vapor phase inside the melts.
In particular, no large bodies or thick layers of melt have been observed, although their presence could be obscured by sand, mainly inside the crater. The degree of crystallinity of the matrix of the impact melt rocks varies from almost completely crystalline to clearbrown glass containing only a few volume percent crystals, although large glass areas are very rare.

Many clasts are granite or gneiss, although clasts of mafic precursor rocks (amphibolites) can be observed. The size of the inclusions range from 1 mm up to several centimeters, and many of them show features distinctive of impact (PDFs in quartz; transformation of quartz to lechatelierite; “ballen”-structured quartz). Around the inclusions occur chilled areas, where a greater amount of glass is present. Moreover the glass, which can be clear or light brown, develops apophyses penetrating the fractures of the inclusions. Where the inclusions did not retain their original characteristics (such a conservation only occur in the inner portions of the larger fragments), selective mineralogical transformation may be developed.
The typical melt rocks consist of a fine-grained intergrowth of plagioclase laths, pyroxene crystals, oxides, and glass. The texture is intersertal to aphanitic and there are no features, like alignment of crystals, suggestive of flow structures.

The sample NMNH 113029-15 from the southwestern outer rim of the crater reveals clear textural evidence for liquid immiscibility between the silicate-rich glass of the matrix and the carbonates. Calcite, in particular, forms spherules and globules within the silica-rich glass that sometimes can coalesce totally or partially. Such textures, sometimes referred as ocellar or emulsion textures, have been observed in natural carbonatitic igneous. In the past, the presence of carbonates or sulfates in the impact melt glass has been considered as being the product of filling by secondary phases.

Analyses of major elements in impact melt rocks show lower contents of SiO2, Al2O3, and Na2O, and higher contents of MgO, Fe2O3, and CaO, than the felsic rocks (i.e., granites and gneisses) of the basement. In comparison with the bulk analyses of the impact melt, the glass is strongly enriched in Si-Al, whereas it is depleted both in Mg and Fe; moreover, the impact melt rocks are variably enriched or depleted in some REE with respect to the felsic and mafic bedrock types. Gold is slightly enriched in the impact melt, and Co, Cr, and Ni abundances are possibly due to a contribution from mafic bedrock.
Evidences of silicate-carbonate liquid immiscibility, mainly as spherules and globules of calcite within the silicate glass, have been highlighted.

Petrographic Investigation of Ejecta from the Tenoumer Impact Crater, Mauritania
S.J. Jaret & L.C. Kah, B.M. French; 40th Lunar and Planetares Science Conference (2009)

In 2003, a suite of 8 crystalline rocks was collected from Tenoumer ejecta. Specifically, the suite includes 4 samples collected from the crater rim, 2 samples from immediately outside the crater rim, 1 sample from the upper ejecta blanket, and 1 sample from the lower ejecta blanket.
Rim and near rim ejecta samples show no evidence of shock metamorphism (i.e., shocked quartz) and are indistinguishable in both hand sample and thin section from basement samples collected inside the crater .
The crystalline samples are amphibolite to green-schist facies metamorphosed granitic geneisses. Major minerals include K-feldspar, plagioclase, quartz, and biotite, with minor amounts of amphibole, apatite, and opaque minerals. Pre-impact dynamic recrystallization fabrics dominate the rocks. Quartz forms clean, small smooth crystals in narrow bands between larger feld-spar grains indicating metamorphic conditions between the melting points of quartz and feldspars.
Shock indicators occur primarily in the more distant ejecta samples and the melt rocks. Tenoumer melt rocks can be described as vesicular melt-matrix breccias, containing clasts of granitic basement in a plagioclase microlitic glassy matrix. Within the melt rocks, PDF’s oc-cur primarily within granitic clasts entrained in the matrix and rarely within individual quartz grains. Additional shock related features found include PDF’s in feldspars, lechatelierite, and ballen textures.
Flow structures within melt phases indicate rapid movement during molten stage. Within the shocked quartz grains, up to 5 sets of PDF’s were found, but grains most commonly exhibited 2-3 sets. Decorated PDF’s and heavily toasted quartz in the majority of samples indicates at least minor post shock alteration. Ballen quartz and lechatelierite are extremely common in the melt rocks and often occur together. The mean shock pressures of the Tenoumer melt rocks was found to be much lower than that required to melt quartz, supporting the notion of preferential melting of grains (or parts of grains) resulting from heterogeneities within the host rock.


Crater Tenoumer: Latitude 22° 55' N; Longitude 10° 24' W

Crater: Uplifted granitoide basement rocks on the rimt


Volcanic gas bubbles in a vesicular matrix (pumice)

Source: L. Carion



Glassy crust

Glass-rich melt matrix with flow structures (PRATESI et al.)

Ballenqartz (rudimentary quartz crystals) in greenish Tenoumer glass !!
Source: M. Schmieder


Sedimentary breccia....... impact-breccia

"Clasts of gneiss and granite embedded in a vesicular matrix,
 are found around the crater's rim."  Source:

In the foreground is visible a dyke on the bottom of the crater

Crater Temimichat


Crater Temimichat: Latitude 24°15' N; Longitude 9° 39' W


The Temimichat crater is located in northern Mauritania
(24° 15' N, 9° 39' W). The bedrock is made up of crystalline basement rocks, with no sedimentary cover. The structure has been listed together with other Mauritanian craters or crater-like features. According to POMEROL (1967), mafic rocks have been found in the area.
Basement rocks include granitoid gneisses and gabbros. The crater appears moderately eroded, with a rim height ranging from few meters to few tens of meters. Its diameter is about 700 m.
The crater rim is not completely preserved. Large portions are eroded and dissected, or masked by eolian deposits. The rim is mostly formed by granitoid gneisses. The low-lying portions of the rim correspond to the occurrence of gabbroic dikes, which seem to be more easily erodible, with respect to the dominant granitoid bedrock.The crater interior is covered by recent eolian Sediments. Below this surficial cover, sedimentary deposits are likely filling the crater.
Inside the granitoid gneisses of the rim, structures that strongly resemble pseudotachilite veins crop out discontinuously, and do not occur in the surrounding undisturbed basement. The term pseudotachilite is used with a descriptive meaning only, not implying any genetic process.
The pseudotachilites are dark to greenish veins of glassy material with fluidal texture at places. Inside these veins Small (less than 1 mm size) fragments of the host rock are present. Pseudotachilite veins mainly appear along small-scale faults, with offsets up to few centimeters. Shear zones are also present, with a brittle to brittle-ductile style. Inside the shear zones, appar-ently re-melted granitic clasts (with dimension of few millimeters up to few centimeters) are visible and their shape varies from angular to highly rounded.
The granitoid gneisses have an ipidiomorphic texture with a mineral assemblage formed by quartz, K-feldspar, plagioclase, biotite and some opaque minerals. The gabbroic bodies are composed mainly of plagioclase and amphiboles.
No ejecta blanket appears to be preserved outside the crater. All around the structure only eolian and fluvial deposits are present, with sporadic large rocky blocks that are also visible farer outside the crater, in the surrounding plain.
The time of formation of the structure is still unconstrained, but the present erosional level suggests a relatively old age of formation.

Conclusion: Temimichat crater has been poorly studied during recent times. No ejecta blanket appears to be preserved. The bedrock is formed of crystalline basement rocks, mainly granitoid gneisses, with locally gabbroic bodies. Granitoid gneisses locally show cataclastic deformation effects and are crosscut by dark veins, that show a striking resemblance with pseudotachilites. Veins are often pervasive. They are linked to small scale faults, with centimetric to decimetric offset. Brittle or brittle-ductile shear zones are associated with these veins, in which rounded granitic clasts also occur. Both veins and shear planes appear fresh, suggesting a relatively young age of formation. No regional brittle structures have been observed. The basement deformational style is ductile and no recent tectonic structures can be observed in the area. These observation could be consistent with the hypothesis of an impact origin (??) for these pseudotachilitic veins. No cataclasites or vein appear in the gabbros.


On the edges of the crater are distributes large blocks with shatter cones (probably not valid) and such with glassy cover. Some blocks are pasted with a characteristic glassy mass. Blocks of granite have been thrown on the western crater - edge. In this area are numerous breccias visible. (Christian Laroubine, 2005)


Crater rim

Uplifted and crushed granitoid basement material

"Granitic rocks with dark glassy veins are to be found at Temimichat"

"Pseudotachilite veins along shear planes in a faulted granitic block"

Granitoid rocks with marks of  metasomatism



Some remarks: I have not studied the craters at the place. However it is very uncertain, whether the craters of Mauritania have emerged by an impact. There are no clear proofs for an impact. An endogenic origin is probable. There are to find clear proofs for hydrovolcanic metasomatism. The craters are similar of endogenic structures of Libya (Oasis, BP, Arkenu) and Egypt (Gilf Kebir). The Mauritanian structures however are builded by an climbed mafic magma. The crater rimes are pushed up. The material is composed of crushed granitoid and metamorphic basement material on Reguibat Shield (Temimichat, Tenoumer) or Paleozoic sedimentary material (Aouelloul). The craters are in the detail clearly atypically for an impact event. They have no cut edges.
The pushed up material of Basement was changed by high pressure and heat. Magmatic processes are also responsible for the glassy injections in veins. Clear shock metamorphic features are not found in the granitoid rocks of the Basement. Polymicte breccia, glasses and PF's are no clear proofs for an impact event. Glasses and the other forms can emerge by high pressure and heat due to endogenic-magmatic processes.
I believe, the mafic magma plug hidden at the bottom of the craters. On rims of the craters is evident to find mafic material. A volcanic explosion has not occurred. The craters are subvolcanic plugs. They were lifted up contemporaneously along an active tectonic line. At the region of Sfariat along such tectonic disturbances mafic dykes and plugs are spread. On the top of these structures the rock layers were erected in the vertical. That is typically for all structures, which are caused by an uplifted magma (see below). All mafic intrusions (dykes and plugs/craters) have emerged probably in the tertiary period.



Dyke (Sfariat)

Diorite (?) (Sfariat)

Erected layers on a rim (Sfariat)



Guelb er Richat (Mauritania)
21° 04' N # 11° 22' W

Richat enigma: Doming and hydrothermal karstification above an alkaline complex ?

"The stratigraphy of the Guelb er Richat dome includes limestone, dolomite, sandstone, mudstone and chert. Field and petrologic studies indicate that the breccia is composed of angular and rounded heterolithic chert fragments in a silica cement. Silica dissolution and replacement were observed in the breccia whereas dolomitization and sulphidation were observed in the limestones. Early ductile deformation in the chert indicates an early development of silicification. The chert fragments display a bimodality in the boundary fractal dimension (Euclidian distance mapping) that implies two phases of breccia formation. Field observations suggest that the breccia was form by a process of syngenetic karst-dissolution collapse. The most likely interpretation is that the Richat chert was formed during a shelf high-stand and persisted in a sub-emergent setting from the Proterozoic up to the Cretaceous. Thus, the Richat chert appears to be an indication of a long term high-standing structural position."

"A model is proposed in which the intrusion of an underlying magmatic body resulted in the bulging of the overlying crust and production of fluids, thus creating a favorable setting for the dissolution of sedimentary rocks. Formation of voids has led to the collapse and brecciation of overlying units.
Thus, Guillaume Matton and Michel Jebrak of the University of Quebec in Montreal, and James K.W. Lee of Queen’s University in Kingston, Ontario suggest that this unique structure had a terrestrial origin, ultimately forming from the effects of an intrusion originating from Earth's mantle."


Guelb er Richat: Latitude 21° 04' N; Longitude 11° 22' W


Internal structure of Guelb er Richat

The 38 km wide Richat structure on the Gres de Chinguetti Plateau in the Adrar region of central Mauritania is an eye-catching feature when seen from space. Although reminiscent of multiringed impact structures, field and laboratory evidence suggests that the Richat structure is a dome of endogenic origin.
The strata dip outward, so that weathering and erosion of the alternating resistant and non resistant Late Precambrian and Early Paleozoic rocks have produced a series of inward-facing cuestas. Quartzite forms the resistant circular ridges, and less resistant rocks underlie the intervening annular depressions, some of which contain seasonal lakes. The centre of the structure, the Guelb er Richat exposes flat-lying limestone and some meta-arkose surrounded by a massive ridge of chert and chert breccia. The overlying part of the stratigraphic section, well exposed outward in the dome, consists of Upper Precambrian/Lower Paleozoic shelf facies rocks.
The quaquaversal dip of the beds increases inward 20 to 25°; locally, dips are as great as 35°. Some minor radial and tangential faults occur in the outer parts of the structure. The north part of the structure is cut by a fault system oriented 30°NE that appears to be unrelated to the formation of the dome. Reconstruction of the eroded beds of the structure suggests that they were uplifted 3 to 4 km to form the dome.
Reconnaissance gravity data do not support the presence at a reasonable depth of an igneous or diapiric intrusive mass nor do they suggest the presence of low-density breccia layers characteristic of impact structures. However, deep erosion could have removed the brecciated lens. Thus, the gravity data neither confirm nor preclude an impact origin for the structure.
Dolerite sills and dikes crop out at several localities within the structure. Associated analcime-rich rock has been interpreted by some to be the result of hydrothermal alteration of rhyolite sills and dikes. No evidence of volcanic activity exists at Richat.
Some geomorphologists in the past have interpreted Richat as an impact structure because of the nearly circular pattern and the reported presence of coesite (?). All geomorphologists now consider Richat to be an endogenic structural dome of purely terrestrial origin:

  • Dips of strata are relatively gentle; strata are even flat-lying at the centre of the structure. Nowhere are beds severely disrupted and contorted.

  • Although breccia is abundant in the structure, the breccia does not have the characteristics of that produced by impact.

  • Injection breccia and pseudotachylite-like material, shatter cones, and other shock-metamorphic effects have not been identified.

  • The coesite reported from the rocks of the structure is almost certainly barite that was misidentified.

  • An unrealistic amount of erosion would be needed if Rich at were an impact structure in order to remove shock-metamorphic effects. Studies elsewhere have indicated that severe effects extend downward below the base of a typical impact crater to about one-fifth its diameter; less severe shock-metamorphic effects extend even deeper. At Richat, then, detectable shock-metamorphic effects might be expected to extend to a depth of perhaps 10 km. The fact that none is observed in the rocks of the structure argues for at least 10 km of erosion if, in fact, Richat is an impact structure. Geologic history of the region does not allow the possibility of that degree of erosion nor does the structure itself show evidence of secondary uplift or deformation, or both. Reconstruction of the beds suggests a maximum uparching of 3 to 4 km, which is a reasonable upper limit to the amount of erosion that has occurred at Richat.

  • The proximity of Aouelloul and Richat is coincidental; they are of significantly different ages. Aouelloul, a small bowl-shaped crater, is probably of Quaternary age, whereas Richat is an ancient structure that has undergone significant planation.







Semsiyat dome, 50 km west-southwest of Richat and centred at latitude 210°'N and longitude 1105°'W, has a diameter of 5 km. Although the style of deformation is similar to that of Richat, Semsiyat is barely detectable on the ground. The structure lies on the Chinguetti Plateau and has only a few meters of topographic relief. Strata dip so slightly that field measurement is difficult. Exposures are poor, and the rocks are extremely weathered. No evidence of shock metamorphism has been discovered at Semsiyat, like Richat, it is believed to be a dome of endogenic origin.

Christian Laroubine, Societe Astronomique de France, has told me: "I sampled breccias on the central zone of Richat. I do not think that we can say that this zone is the central peak of an impact crater, but there is a doubt about a tectonic origin. The components of my samples do not find their origin with a very great depth as could be to it a magma or nearby products. This breccia (left) is astonishing by its structure and its composition. Today, nobody can affirm without controversy if it is of tectonic or impact origin."
I believe, that these breccia can have emerged also in consequence of endogenic processes. In the doubt is the context important. The Richat - structure has unquestionably a endogenic origin. The finds of this breccia and Coesit (?) are the proof for it, that endogenic processes and an impact - event can produce similar forms.

R. F. Fudali: Department of Mineral Sciences, Smithsonian Institution, Washington  - Coesite from the Richat Dome, Mauritania:
A misidentification
The "shattered sandstone" from Richat reported to contain coesite is a tectonic breccia and probably represents a shear zone developed during the structural doming. An optical and x-ray examination of concentrates from this breccia demonstrated that the supposed x-ray reflections of coesite are actually due to barite, introduced into the permeable crushed zone by groundwater.