THE PRECAMBRIAN OF THE ANTI-ATLAS (MOROCCO): A GEOCHRONOLOGICAL SYNTHESIS (Géoparc Jbel Bani)
le géoparc du jbel bani - tata

THE PRECAMBRIAN OF THE ANTI-ATLAS (MOROCCO): A GEOCHRONOLOGICAL SYNTHESIS (Géoparc Jbel Bani)

Par RENE CHARLOT

Centre Armoricain d'Etude Structurale des Socles, Institut de Géologie, Université de Rennes (France)

(Received May 30, 1975; revision accepted August 19, 1975)

ABSTRACT

Charlot, R., 1976. The Precambrian of the Anti-Atlas (Morocco): a Geochronological synthesis. Precambrian Res., 3: 273--299.

Twenty total-rock Rb/Sr isochrones (ka~Rb -- 1.47" 10-" yr-' ) and many individual data, mainly obtained by the K/At and Rb/Sr methods, on total rocks and minerals show that the orogenic story of the Precambrian of the Anti-Atlas, which was studied in its western and central parts, began with the Eburnean orogeny (1900--1650 m.y.). Up to now there is no good evidence of an older basement, even if the rather high initial ratio (0.707--0.709) of some Eburnean granites might suggest that such a basement exists at depth. There is also no Geochronological evidence of plutonium until in early Panafrican time (680 m.y.) there started a period of intense magmatic and thermal activity which was ended by the emplacement of late post-tectonic granites at 530--520 m.y.

The late Panafrican acid volcanism does not generally give suitable material for Rb/Sr dating, because of the great mobility of the radiogenic strontium in such a material even under very weak thermal conditions.

INTRODUCTION

Situated at the western end of northern Africa, Morocco constitutes the mobile foreland of the West-African craton. All the geological periods are represented, from the recent sediments of the Rif in the north, to the ancient Precambrian formations in the Anti-Atlas chain, in the south, which separates Morocco from the Sahara. It is this latter area, the only showing Precambrian outcrops, with which this paper is concerned.

GEOLOGICAL SETTING

In all of his basic investigations in this area, Choubert defines the Anti- Atlas chain as a complex group of Precambrian formations with different ages, which underlie a Paleozoic sequence that was folded during the Hercynian orogeny and raised for the last time during the Alpine event. This Precambrian basement, which appears (Fig.1) in some massifs or "boutonnieres" under the folded Paleozoic coverage, constitutes the frontal ring of the Precambrian African continent.

During the progress of his investigations, Choubert first defined three (1952), then five (1963), then six (1967--1970) successive Precambrian cycles to which he adds an Infracambrien or "Adoudounian" cycle. Lastly in 1973 he presented a synthesis in which he classified these seven cycles by assigning ages to each of them as follows:

(1) Ancient Archean: systems of the Jbel Ouiharem gneisses (Kerdous area) and of the Oued Assemlil gneisses (El Graara area). These were referred to the Zagorides orogeny with an estimated age older than 2650 m.y.

(2) Recent Archean: Kerdous series (ancient Precambrian 0) and Zenaga series (ancient .Precambrian I), now correlated in age and representing the Berbèrisés orogeny with an estimated age of 2400--2600 m.y.

(3) Lower Proterozoic (or Precambrian II 1): E1 Graara system, ending with the "Bouazzerides orogeny": with an estimated age of 2200--2000 m.y.

(4) Middle Proterozoic (or Precambrian II): a system of limestones and quartzites. Orogeny of the "Anti-Atlas ides": with an estimated age of 1400-- 1300 m.y.

(5) Upper Proterozoic (or Precambrian II 3): The Siroua--Sahro system ending with the orogeny of the "Marocanisée". Estimated age: 1000 m.y.

(6) Terminal Proterozoic (or Precambrian III): the Ouarzazate series. Estimated age: from 1000 to 950 m.y.

(7) Adoudounian: series of Infracambrien dolomites. Estimated age from 950 to 570 m.y.

Following his work in the Bou Azzer--E1 Graara area (central Anti-Atlas), Leblanc expresses doubt on this multiplicity of orogenic events and differentiates in that area only three structural units (Leblanc, 1975) which are:

(1) Polycyclic gneisses overt rusted by the ophiolite of the second unit.

These are the Oued Assemlil gneisses (Archean in the Choubert classification) which Leblanc assigns to the Eburnean orogeny.

(2) Very folded but generally unmetamorphosed formations, beginning with an ophiolite complex, associated with limestones, quartzites and schists overlapped in major unconformity by the Tiddiline Formation which was left metamorphosed and folded only during a late compressive Panafrican phase (Leblanc, 1970--1972). This unit, which corresponds to the lower, middle and upper Precambrian cycles of the Choubert classification, according to Leblanc, was affected by the Panafrican orogeny alone.

(3) A coverage unit overlapping in major unconformity the folded series of the previous unit. It is made up of the Ouarzazate rhino-andesitic formations and the azoic Adoudounian.

These two Stratigraphic and orogenic divisions are schematically represented in Fig.2.

Structurally, the Anti-Atlasic Precambrian area is separated from the Atlasic area by the great fault of the "Accident sud-atlassiques" which, in fact, separates this entire northern part of Africa from the African continent. A second great fault "l'Accident majeur de l'Anti-Atlas" delineates the front of the ancient African continent, against which as suggested by Choubert, the successive Precambrian belts are closely fitted.

Fig.1 Distribution map of the Precambrian windows in the Anti-Atlas (from Choubert,

1963) and situation of the main localities reviewed.

1 = Coverage; 2 = Adoudounian and Lower Cambrian; 3 = Proterozoic. lfni window: T = Taoulecht and Tiyourzha; A = Alouzad. Bas Dra window : T = A~n Tamousift; A = Aouinet n'A]t Oussa; O = Oued Cha'iba. Kerdous window: A = Oued Amarhous; Tz = Tazeroualt; Ts = Tasserhirt; J = Jbel Ouiharem; Tf = Tafraout; Ta = Tahala. Zenaga window: A = Azguemerzi; T = Tazenakht; H = Zaouia Sidi el Hocein. Bou Azzer -- E1 Graara window: A = Oued Assemlil; B = Bleida. Siroua: A = Askaoun.

GEOCHRONOLOGYSTUDY

The results reported here have been mainly obtained on that part of the Anti-Atlas which is situated south and west of the two great faults described above, particularly in the Kerdous, Bas Drâ and Zenaga areas and also for a minor part in the Bou Azzer and Siroua areas. All of these results have been obtained by the author using the Rb/Sr method. The analyses were made on a Thomson 206 S mass spectrometer, computerized and automated by a Hewlett Packard 9810 system. Age values are calculated on the basis of ~STRb = 1.47.10 -11 yr -1, and strontium isotope ratios normalized to S6Sr/~SSr = 0.1194. The mean value for 6 separate runs of the Eimer and Amend reference standard were 0.70816_+ 0.00009 (20). Isochrones are computed following the Wendt program.

References will be made to the work of other authors who have carried out investigations on the Anti-Atlas as a whole (Cohen et al., 1964, 1965; Choubert et al., 1965) or on some particular part of it: the Ifni area (Benziane, 1974); rhyolites of the Haut Atlas (Juery et al., 1974); the E1 Graara area (Clauer and Leblanc, 1974). In addition reference will be made to measurements by Tisserant at the geochronology laboratory of the Geological Survey of Morocco.

Results will be classified into four groups, corresponding to the subdivisions by Choubert, as follows:

(I) Ancient basement ("Zagorides" and "Berberides" orogenies)

(II) Lower and middle Proterozoic ("Bouazzerides and Anti-Atlas ides")

(III) Upper Proterozoic ("Marocanisée")

(IV) Terminal Proterozoic

I. Ancient basement

A. Kerdous area

(1) Jbel Ouiharem gneisses. The four analyses done on this formation (Fig.3) yield a scattering of points between two reference isochrones at 1700 and 1300 m.y. Because of the scatter and the small number of analyses, it is not possible to give a precise age to these gneisses and it can only be proposed that they are probably older than or equal to 1700 m.y. This would suggest that they belong to an old basement, but does not imply an age older than 2650 m.y.

Fig.3. Rb/Sr data plot diagram for Jbel Ouiharem gneisses and Oued Assemlil gneisses isochron (from Tisserant, 1971).

(2) Tazeroualt granite. Eight total-rock samples yield an isochron (Fig.4) corresponding to an age of 1905-+ 25 m.y. with a relatively high initial ratio (0.7094 + 0.0003), which suggests a crustal origin for this granite in accordance with its analectic nature as shown by Choubert et al. (1974). Measurements made on Muscovites from pegmatite’s cutting this granite give younger ages (1545 and 1045 m.y. by the K/Ar method and 1750 m.y. by the Rb/Sr method) and show that the region has been subjected to thermal influences, either a slow cooling and rising, or successive reheating.

Fig.4. Rb/Sr isochron diagram for Tazeroualt granite.

(3) Oued Amarhous granite. The first radiometric age measurement in the Anti-Atlas was made on this granite (Cohen et al. 1964--1965) and gave a single value of 2600 m.y. -- with (87Sr/S6Sr)o = 0.712 -- on which Choubert based his dating of the Berbérie orogeny. However, the present results of total-rock analyses on eleven samples fall on a secondary isochron (Fig.5), the slope of which corresponds to an age of 930 +- 16 m.y., with an initial 87Sr/86Sr ratio of 0.812 + 0.002. Muscovites from these samples yield much more ancient age values (1500--1800 m.y.); K-feldspars (microcline) show a deficit in radiogenic strontium, giving "age" values of 750--800 m.y., whereas plagioclases show an excess, yielding values of 3500--5600 m.y.

The significance of these data may be considered in the following manner. Individual Rb/Sr ages (1500--1800 m.y.) as well as K/At ages (1300 m.y.: Choubert et al., 1965) of the Muscovites show that the age of 930 m.y. cannot date the granite emplacement. Results obtained on feldspars show that a transfer of radiogenic strontium from microclines to plagioclases caused the rotation of the total-rock isochron, giving an apparent rejuvenation of the microcline-rich samples and an excess age to the plagioclase-lich samples. Thus, these mineral age values have no geological significance.

It has been shown by Charlot (1974) that this granite was probably emplaced at about 1900 m.y. based on a reassemblage of the affected age values for the massif as a whole. The cause of the resetting was an intense shearing (Hassenforder and Jeannette: personal communication, 1974) which caused a recrystallization of the feldspars and a migration of radiogenic strontium from microcline to plagioclase. The microcline could have expelled all of its radiogenic strontium, as proposed for some Australian granite by Arians et al. (1971), and its age value reset to zero; the shearing could then be the result of the Panafrican orogeny, giving the apparent age value for the microcline.

Or, on the other hand, strontium homogenization could have occurred between microcline and plagioclase with the shearing then depending on the Hercynian orogeny, which is suggested by the slope of the secondary isochrones between the microcline and plagioclase. Thus, the age of 930 m.y. does not necessarily correspond to the tectonism. It probably has no geological significance.

Fig. 5. Rb/Sr data plot diagram for Oued Amorphous granites.

(4) Tasserhirt granite. Analytical data obtained on fourteen total-rock samples yield a scatter (Fig.6) in which there is two possible alignments:

(a) Nine samples fit along a straight line corresponding to an age of 1680+ 35 m.y. and an initial ratio of 0.7012+ 0.0010.

(b) Four samples give an alignment representing 2550+ 67 m.y., but with so low an initial ratio (0.6975+ 0.0012) that this interpretation cannot have any geological significance.

All of these samples cannot be petro graphically distinguished and come from formations which are cut by pegmatite’s in which Muscovites give Rb/Sr age values of around 1900 m.y. and K/at age values of about 1500 m.y. This value of 1500 m.y. is similar to the Rb/Sr age found for the Muscovites from the granites themselves. Therefore, it seems clear that the granites are at least 1900 m.y. in age. The Tasserhirt granite occurs in a large migmatitic zone, but Choubert et al., (1974) show that the migmatization antedates the granite, so that it is not responsible for the resetting age of 1680 m.y. found for the samples in alignment number I. This age of 1680 m.y. is similar to that found in other massifs, on Muscovites or total-rock isochrones, so that it is not a random age value. If the age of 1900 m.y. or older is taken to be true for the emplacement of the granite, then it appears that an important thermal event occurred at about 1680 m.y., and probably lasted until 1500 m.y. which was the closure age of the

Fig.6. Rb/Sr data plot diagram for Tasserhirt granite.

(5) Tahala granite. This granite has only recently been recognized as being ancient (Chariot, 1973--1975), its intrusive character having made it appear to have been associated with the Tafraout granite. Five total-rock samples define an isochron at 1920+ 40 m.y. with an initial 87Sr/86Sr ratio of 0.7072 + 0.0007 (Fig.7). It thus falls within the error range of the age value for the Tazeroualt granite, but is probably not cogenetic because of the difference in the initial ratios. Therefore, if these two granites are the result of two successive phases (Choubert et al., 1974), the time between them is too short to be distinguished by radiometric methods. The initial ratio at 0.707 found for the Tahala granite, as well as for the Tazeroualt granite, confirms a polygenetic origin for these rocks as seen also in an examination of thin sections (Auvray, personal communication, 1972).

B. Zenaga area

(1) Azguemerzi granite. Considered as belonging to the same orogenic phase as the Tazeroualt granitization, this granite gives an isochron (Fig.8) of 1865 + 25 m.y. from five analyzed samples, with an initial ratio of 0.7052 + 0.0009. Although there is an overlapping of the error range with that of the Tahala and Tazeroualt granites, it is possible that this granite may be slightly younger than the others. This would suggest a slight time lag between tectonic phases affecting the western and central Anti-Atlas in accordance with a distinction made by Choubert (1973) between the western and eastern Berberides.

The muscovite’s of this granite all give Rb/Sr ages of 1660--1680 m.y., while the biotitic are scattered between 1560 and 690 m.y., probably reflecting the Panafrican thermal event.

Fig.7. Rb/Sr isochron diagram for Tahala granite.

Fig.8. Rb/Sr isochron diagram for Azguemerzi granite.

(2) Tazenakht granite. Seven samples of this very alkaline granite give an isochron age of 1735 + 15 m.y. with an initial ratio of 0.7085 + 0.0006 (Fig.9).

Muscovites yield ages slightly younger (1680 m.y.), similar to those found for the Azguemerzi Muscovites. Thus, this age could record either a cooling and uplift time, or a thermal front.

This emplacement at 1735 m.y. is about 200 million years younger than the emplacement of the Tahala granite, clearly indicating the two separate intrusive phases, as opposed to the interpretation of Choubert et al. (1974).

(3) Timrharhine pegmatite. Generally considered to be linked to the Tazenakht granite (Bouladon et al., 1950; Choubert, 1963), this pegmatite outcrops in the south of the Zenaga area. Muscovites have been studied by both the Rb/Sr and K/Ar methods. K/At age values reported by Choubert (1965) are scattered between 1828 and 1180 m.y., while the Rb/Sr age (Cal~.en et al., 1965) is 1650 + 50 m.y., very similar to those found for the Azguemerzi and Tazenakht granites. One single K/at feldspar determination was at 1200 m.y. (Vidal, 1967). It is difficult to interpret the scattering of the argon ages; it may be due either to incomplete readjustment or to argon migrations inside the large muscovite crystals (up to 40 cm in diameter). In the first assumption, the age of 1828 m.y. would be a minimum age so that the pegrnatite would be better linked to the Azguemerzi granite, while the second assumption would not permit any geological or geochronological assessment.

In a very recent work about the Zenaga area, Hurley (personal communication, 1975) studied randomly taken samples of granitic rocks in place and class in Eocambrien or present drainage. The results (Fig.10) are scattered in the range of 1800+ 200 m.y., in good agreement with ours, which were obtained on individual litho logic units. Some samples give individually older ages but are discounted by Hurley as being due to 87Sr migrations and too small a sample size.

In the same study, Hurley finds five samples of pegmatite rocks with feldspar crystals over 5 cm in length and yielding a scatter of points about an average of approximately 1200 m.y. (Fig.ll). It seems that these pegmatite rocks are to be related to the same event as the Timrharhine pegmatite’s and that the mean age value of 1200 m.y. does not represent an intermediate orogenic event between Eburnean and Panafrican orogenic but only the result of the Panafrican heating close by.

Fi .9. Rb/Sr isochron diagram for Tazenakht granite.

Fig.10. Rb/Sr data plot diagram for samples from Zenaga region (from Hurley, personal

Communication, 1975).

Fig.ll. Rb/Sr data plot diagram for Pegmatite of ZaouTa Sidi el Hocein (from Hurley, personal Communication, 1975).     

C. Bas Drâ area

(1) ATn Tamousift granite. Twelve analysed samples yield an isochron (Fig.12) corresponding to an age of 1895 + 20 m.y. with an initial ratio of 0.7032_+ 0.0004 Muscovites from this granite and from pegmatite’s cutting it, give ages concordant with that of the isochron. The low initial ratio points to a mantle origin.

(2) Aou'inet n'A'ft Oussa granite. All of the samples analysed showed similar, high Rb/Sr ratios which did not permit an accurate determination of the initial ratio (0.709 + 0.018). The age therefore has a large error: 1890 + 90 m.y.

The similarity between this and the previous age value suggests that these two granites are contemporaneous and may be cogenetic. Graphically, both alignments are superposable (Fig.12).

(3) Oued Cha'iba granite. This granite which occurs in the form of a wide dome, south of Oued Drâ, gives a total-rock isochron age of 1640 + 32 m.y. (Fig.13) with an initial ratio of 0.7105 + 0.0012, suggesting that it is of crustal origin.

It might represent the end of a second orogenic period which began with the Tazenakht granitization and as a final phase might have also gneissified the Tazenakht granite and affected some samples of the Tasserhirt granite.

Fig.12. Rb/Sr isochron diagram for A~'n Tamousift and Aou’inet neat Oussa granites.

D. Other areas (1) Oued Assemlil gneisses (Bou Azzer and El Graara area). These gneisses, first studied by the K/At method (Choubert et al., 1965) yielded mineral ages (Muscovites and biotitic) in the range of 630--520 m.y. A total-rock Rb/Sr study, by Tisserant (1971) allowed him to draw an isochron (Fig.3) yielding an age of 660 + 60 m.y. (initial 87Sr/86Sr ratio = 0.708 -+ 0.003). Such an age is in complete discordance with the geological analysis by Choubert~ (for whom these gneisses are Archean in age) as well as with Leblanc who gives them, by comparison with the Azguemerzi granite, an Eburnean age. In actual fact, the igneous or sedimentary origin of these gneisses is in doubt, so the interpretation of the Rb/Sr age is difficult. If these gneisses are truly ancient, one must accept that they have been subjected to a complete rehomogenization during the Panafrican orogeny, with a total elimination of the radiogenic strontium already formed, since the isochron shows a relatively low initial ratio. At this time we will merely note the quite good agreement between the early Panafrican total-rock Rb/Sr ages and the muscovite K/At ages.

Fig.13. Rb/Sr isochron diagram for Oued Cha’iba granite.

(2) Alouzad granite (Ifni area). One total-rock sample and its minerals (muscovite, K-feldspars) have been studied by Tisserant and reported by Benziane (1974). The isochron (Fig.14) gives an age of 2217 -+ 44 m.y. with an initial ratio of 0.715 + 0.007. Because it is a mineral and total-rock isochron, Benziane thinks that the age can only be a minimum. In fact, the present 87Sr/86Sr ratio of the total rock which is higher than 1.0, does not permit a much greater age if one stays within the permissible lower limit for the initial ratio. This is the first time that the emplacement of granite in Morocco is found to be older than 1900 m.y.

II. Lower and Middle Proterozoic

These formations are generally ungranitized (Choubert et al., 1970) and have not been submitted to a systematic Geochronological study. Most of the analyses have been on single rock samples by the K/at method, and all have indicated age values in the Paleozoic.

A study of the Bleida schists (Bou Azzer--E1 Graara area) belonging to the Precambrian II has recently been carried out by Clauer and Leblanc (1974).

These monometamorphic sediments give an isochron (total-rock and fine fraction) at 602 + 18 m.y., probably representing an isotopic homogenization imposed by the metamorphism. The monometamorphic of the series and the Panafrican age for this metamorphism suggests that no other orogeny has taken place between the development of the schists and the Panafrican event, leading to the conclusion that the Precambrian II cycle, as defined by Choubert (1970), does not end at 1400--1300 m.y., but at 600 m.y.

Fig.14. Whole-rock and minerals Rb/Sr isochron diagram for Alouzad granite (from Benziane, 1974).

III. Upper Proterozoic

Choubert differentiates two successive periods of igneous rock emplacement:

(1) Acid magmatism at the beginning of the cycle with rhyolites and ignimbrites, and also some intrusive granitic massifs typified by the Tafraout granite (Kerdous area).

(2) Granitization at the end of the cycle which mainly includes the emplacement of the Siroua and Sahro granites.

A. Acid magmatism

(1) Kerdous rhyolites. Hassenforder and Jeannette {1974) have shown that these rhyolites belong mainly to the terminal Precambrian cycle.

In our analyses, the 30 samples of total rocks give a scatter of points with apparent ages in the range 580--380 m.y. (Fig.15). It seems that the rhyolites lost varying proportions of their radiogenic strontium, probably during a time of devitrification and water elimination. We will see below with the Ouarzazate rhyolites that this strontium loss may occur even with a total lack of metamorphism.

Fig.15. Rb/Sr data plot diagram for Kerdous rhyolites.

(2) Ifni rhyolites and andesitic. For these formations Benziane (1974) has given two isochrones: one for the rhyolites at 631 + 5 m.y. (Fig.16), and the other for the andesitic at 603 + 6 m.y. (Fig.17). It is evident that these two formations have not followed the same isotopic evolution as the Kerdous rhyolites. One possible explanation may be that the rhyo-andesitic series of Ifni might correspond to a syn-Panafrican emission phase, followed by a metamorphism that included a digenesis and devitrification of the rocks, thereby closing the systems against strontium loss. Such a metamorphism is known in the Bou Azzer area, being dated at 602 m.y. on the Bleida schists. The Kerdous rhyolites may represent a later volcanic extrusive event. Some very new measurements, not completed up to now, in the Ifni area show the possibility of finding here the equivalent of the Kerdous series particularly in the southern boundary of the window and also in association with the Taoulecht granite. The dating of this event will be discussed with the Ouarzazate series below.

Fig.16. Rb/Sr isochron diagram for Ifni rhyolites (from Benziane, 1974).

Fig.17. Rb/Sr isochron diagram for Ifni andesites (from Benziane, 1974).

(3) Tafraout granite. This granite presents a total-rock isochron age (Fig.18) of 530 + 15 m.y. with an initial ratio of 0.719-+ 0.003 (Chariot, 1973--1975).

The geological interpretation which places this granite in the upper Proterozoic is thus in question, and it seems that this age value truly represents an emplacement late in the Panafrican event.

Similar ages are found for other granitic massifs in the Kerdous area, as follows:

Agouni Yessene granite (Kerdous): 536-+ 11 m.y. (87/86Sr)0 = 0.718-+ 0.004 (Fig.18)

Ida ou Cougmar granite (Kerdous) : 522 + 10 m.y. (STSrS6Sr)0 = 0.735 + 0.02 (Fig.19).

Tiyourhza granite (Ifni): 541 + 8 m.y. (87Sr/86Sr) 0 = 0.7167 + 0.0005 (Fig.20).

All of these granites have fairly high initial strontium ratios, suggesting crustal origin. From all of these data it is clear that a Late Precambrian or Early Cambrian period of granite intrusion is established, at least in the west Anti-Atlas.

Fig.lB. Rb/Sr isochron diagram for Tafraout and Agouni Yessene granites (from Chariot,1975).

Fig.19. Rb/Sr isochron diagram for Ida ou Cougmar granite.

Fig.20. Rb/Sr isochron diagram for 'riourhza granite (from Benziane 1974/.

B. Siroua and Sahro granitization these granitic emplacements are observed as discrete plutons south of the "Accident majeur", but develop into a strong belt north of it in the Siroua and Sahro areas. Only the Siroua granites have been studied up to now. The measurements carried out in collaboration with Tisserant yield two groups of results corresponding to two different petrography types. There is a broad granodioritic massif intruded by veins or small stocks of fine, light granites, aplites or micropegmatites. The granodiorite gives an isochron of 680 + 35 m.y.

(Fig.21) with an initial ratio of 0.7073 + 0.0010 while the fine-grained granites are clearly younger and are dated at 585 + 35 m.y. with an initial ratio of 0.7047 + 0.0045 (Fig.22). One muscovite of this granite gives a single concordant Rb/Sr age (560 m.y.), while the biotites of both granites are closed later, at about 500 m.y.

Fig.21. Rb/Sr isochron diagram for Siroua granodiorite.

Fig.22. Rb/Sr isochron diagram for Siroua granite.

Similar results are obtained by Hurley (personal communication, 1975), with an isochron at 600 m.y. for samples taken in three different granitic massifs, and by Benziane (1974) in the Ifni area, where the Taoulecht granite associated with rhyolites gives a Rb/Sr total-rock isochron with an "age value of 583 + 7 m.y. and an initial ratio of 0.7079 + 0.0008 (Fig.23).

Fig.23. Rb/Sr isochron diagram for Taoulecht granite (from Benziane, 1974).

Related to the same orogenic event by Choubert, the Ifni granodiorite has been studied by Tisserant and reported by Benziane (1974). Five samples yield an isochron age value of 627 + 14 m.y. with an initial STSr/a6Sr ratio of 0.7055 + 0.0006 (Fig.24).

Fig.24. Rb/Sr isochron diagram for Ifni granodiorite (from Benziane, 1974).

IV. The close of the Precambrian this is represented by the volcano-detritic series of Ouarzazate. All measurements done on the volcanic formations (Fig.25) by the Rb/Sr method, as well as the K/Ar method, show a wide scattering of results with apparent ages in the range 550--380 m.y.

For the same series, but in the Haut-Atlas (north Siroua), Juery et al. (1974) present an isochron at 310 + 14 m.y. with an initial ratio of 0.729 + 0.003; however given our own experience in the Kerdous area, the small number of samples analysed permits the possibility of an accidental alignment, particularly since one of the analyses is plotted high above the isochron. In fact, the actual calculated age values scatter in the same range (520--350 m.y.) as our results and are therefore very similar. The important contribution of these authors, however, is a U--Pb zircon determination which gives the emplacement age of this series at 585 m.y. which now defines the base of the Adoudounian cycle, which is conformably Overlain by the paleontologically dated Lower Cambrian.

Fig.25. Rb/Sr data plot diagram for Ouarzazate rhyolites.

Thus it is found that the Ouarzazate series has lost radiogenic strontium in a manner similar to the Kerdous rhyolites, and strengthens the geological correlation by Hassenforder and Jeannette (1974). The age of 585 m.y. for the extrusion of this series remains in good conformity with the conclusion reached above that explained the difference in the evolution of the Kerdous and Ifni series.

Similar dispersions in age values for acid volcanites have already been pointed out in the literature, for example in Norway (Priem et al., 1973), in Massachusetts (Bottino et al., 1970), in New Brunswick, Nova Scotia and Newfoundland (Fairbairn et al., 1966; Cormier, 1969; Fairbairn and Hurley, 1970). In all of these cases, authors impute these losses of radiogenic strontium to weak metamorphism. In the western area of the Moroccan Anti-Atlas (Kerdous and Ifni areas), it is known that the formations of the Anezi series (overlain by the rhyolites) are affected by a slight epizonal metamorphism (Clauer, personal communication, 1971) and that biotites in some of the granites have been rejuvenated to an age of 320--340 m.y. It is therefore believed that even this slight metamorphism is responsible for the opening of the rhyolites to strontium loss. On the other hand, in the central and eastern area no metamorphism has ever been mentioned in the latest Precambrian sedimentary formations, and no isotopic resetting has been noted on biotites from granites which never give ages, even K/At ages, lower than 600--500 m.y.

This suggests that there is easy loss of radiogenic strontium from acid volcanites, with temperature elevations equivalent to burial under only 4000--5000 m.

V. Adoudounian volcanism

The last effusive event of the Precambrian in the Anti-Atlas, this volcanism is particularly well developed in the south of the Bou Azzer--E1 Graara window with the Jbel Bokkho volcano. It is represented by syenites, micro granites and trachytic or andesitic flows interstratified in the lower Adoudounian.

Five measured samples yield an isochron corresponding to an age of 300 + 5 m.y. with an initial ratio of 0.7288 + 0.0005 (Fig.26). Such an age does not correspond to the geological setting well established by Choubert (1963--1970) and now by Leblanc (1975).

Many hypotheses may be envisaged as an explanation. Choubert chooses to think that samples have lost more or less of their radiogenic strontium and gives as a minimum age for the formation the maximum individual age value of 925 m.y. This first hypothesis cannot any longer be considered as valid, because of the U--Pb zircon age found for the Ouarzazate series which underlies the Adoudounian. Another way of thinking would be that the formation had been entirely rehomogenized, causing the high initial ratio, but this would have need of significant thermal event and it is impossible to find any trace of it in the area where all the minerals such as biotites and Muscovites have been definitively closed at 600--500 m.y. even against the argon diffusion.

Thus at this time, and with the present experience of the volcanic in that region, it is probably better to think that the alignment is accidental and that more measurements would finally give a scattering, so that the age of 300 m.y., has no geological meaning.

Fig.26. Rb/Sr data plot diagram for Jbel Bokho volcanites.

CONCLUSIONS

The first orogenic period which may be defined with certainty in Morocco is at about 1900 m.y., without the possibility of differentiating between the two phases suggested by Choubert et al. (1974). These two phases are probably sub contemporaneous. Age values on individual samples from the Tasserhirt granite suggest a more ancient basement, but without isochron data, and in view of the migmatitic zone not yet being completely studied, it is not wise to interpret these points further until more work is done. Thus it appears that there is no good Geochronological evidence for an ancient basement, even if the initial ratios of the Tahala and Tazeroualt granites are high, suggesting that such a basement may exist at depth.

The first orogenic period is found also in the central Anti-Atlas, with the emplacement of the Azguemerzi granite at 1865 m.y. 295

A second magmatic and thermally active period occurred between 1735 m.y. and 1640 m.y. In the central Anti-Atlas, the thermal activity seems to have lasted only a short time with the closure of the Muscovites at 1660--1680 m.y.

In the west there seems to be a greater thermal intensity with the emplacement of the Oued Chaa granite, and a closure of Muscovites occurring at 1500--1450 m.y.

Following this second magmatic activity, there is a long period with no plutonium, continuing until an early Panafrican time marked by the emplacement of the Askaoun granodiorite at 680 m.y. This emplacement starts a period of intense magmatic and thermal activity: gneissification or rehomogenization of the Oued Assemlil gneisses (660 m.y.), and then emplacement of the Ifni granite (630 m.y.), soon followed by the Ifni rhyoandesitic extrusions which were reset in their isotopic evolution by a metamorphism at about 600 m.y. Then there occurs a new volcanic period at 585 m.y. accompanying the emplacement of some veins or small stocks in the Siroua area, also the emplacement of the Taoulecht granite in the Ifni window. The thermal activity is slight, or non-existent since the later series are not metamorphosed (Tiddiline series, Bou Azzer area), or only slightly (Anezi series,

Kerdous area). This very weak metamorphism is attributed to the Hercynian orogeny.

The Panafrican orogeny ends with the emplacement of late post-tectonic granites in the Kerdous and Ifni areas, with no trace of thermal resetting found in the whole central part of the Anti-Atlas following this time, as biotites are closed to strontium and argon at 500 m.y. Hercynian metamorphism becomes evident in proceeding westward, where in the Kerdous area biotites of the Tafraout granite are closed at 320 m.y., or in the Ifni area biotites and K-feldspars of the Mesti granite at 315 m.y. or in the Mirleft granite at 250 m.y.

This Geochronological timing of events is observed on the mineral age histogram (Fig.27) even if the first two periods are not well discernible. On this histogram some rare data appeared during 1500--700 m.y., most of them by the K/at method. It is probable that these are due to incomplete isotopic readjustments during the Panafrican event. This Geochronological defined succession of orogenic periods has geological implications (summarized in Fig.28) leading us to revise the scheme proposed by Choubert.

Thus, the first orogeny of which one can be sure is that which ended the deposition cycle of the Kerdous and Zenaga series. This is the Berberides orogeny which was responsible for the Tazeroualt granitization, dated at 1900--1950 m.y. and thus being equivalent to the Eburnean orogeny. On this scale it is impossible to date precisely the Jbel Ouiharem or Oued Assemlil gneisses, but there is no evidence for their assignment to an earlier cycle.

The second active period may be linked to the Tarkwaian (late Eburnean) orogeny (1750--1600 m.y.). It might, in the geological scheme proposed by Choubert, be the orogeny of the Bouazzerides centrals which ends the E1 Graara cycle. On the other hand, in the Leblanc scheme this system cannot be separated from the following series and thus would be younger than 1600 m.y. Therefore, there would not be any representative sedimentary series for this Tarkwaian cycle. The granites which have pointed up the magmatic ally active period were emplaced in the carbonized part of the Anti-Atlas, and have no relationship with the E1 Graara series defined by Choubert in the frontal mobile zone. Thus these age determinations do not permit us to interfere in the discussion of the Stratigraphic position of the E1 Graara series.

It is impossible to find in the results of this work any real traces of the "Anti-Atlas ides" orogeny with a proposed age of 1400--1300 m.y., the rare individual age values found on pegmatitic rocks being more likely related to Panafrican reheating. No magmatic event is noted between 1600 and 680 m.y., thus giving for the last event (the Marocanisée) a younger age (680--585 m.y.) than that proposed by Choubert (1000 m.y.). Caby (1970-1972) and Caby and Leblanc (1973) have already assigned this orogeny to an epizonal branch of the Panafrican orogeny, and this is here fully confirmed.

The last manifestation of this orogeny ends the real Precambrian cycle. It is followed by the deposition of a volcano-(etritic series (Kerdous rhyolites, Ouarzazate series) which Leblanc, and Hassenforder and Jeannette, assign to the beginning of an infra-Cambrian cycle that we are able to date at 585 m.y.

This is subsequently covered by the Adoudounian series, of which the upper part has already been recognized as Lower Cambrian (Boudda and Choubert, 1972; Choubert et al., 1973).

Ultimately, a post-tectonic magmatic activity found in the western part of the Anti-Atlas takes place in the middle Cambriar until about 520 m.y.

Fig. 27. Histogram of the results from single mineral analyses.

ACKNOWLEDGEMENTS

This study has been supported by the Geological Survey of Morocco and the Centre National de la Recherche Scientifique (Centre Armoricain d'Etude Structurale des Socles et R.C.P. 241).

I would like to acknowledge all the workers who helped me throughout the realization of this work: M. Saadi and A. Boudda from the Geological Survey of Morocco; G. Choubert from the world geological map Commission; Prof. J. Cogn~, Ph. Vidal, B. Auvray, R. Capdevilla, Prof. J. Hameurt, J. Cornichet and J. Mac~ from the Centre Armoricain d'Etude Structurale des Socles; M. Leblanc from the Centre G~ologique et G~ophysique de Montpellier;

M. Bonhomme, N. Clauer, B. Hassenforder, D. Jeannette and D. Tisserant from the I n s t i t u t de G~ologie de Strasbourg. Prof. P.M. Hurley is thanked for the permission to use his unpublished data and for making possible the presentation of this paper in good English.

Fig.28. Schematic table of the geochronological events succession in the Anti-Atlas.

REFERENCES

Arriens, P.A., Brooks, C., Bofinger, V.M. and Compston, W., 1971. The discordance of mineral ages in granitic rocks resulting from the redistribution of rubidium and strontium. J. Geol. Res., 20: 4981--4994.

Benziane, F., 1974. Etude p~trologique et g~ochronologique des Granites de la Boutonniere d'Ifni. Thesis, Univ. Grenoble.

Bottino, M.L., Fullagar, P.D., Fairbairn, H.W., Pinson, W.H., Jr. and Hurley, P.M., 1970.

The Blue Hills igneous complex, Massachusetts: whole rock Rb--Sr open system. Geol.

Soc. Am. Bull., 81: 3739--3746.

Boudda, A. and Choubert, G., 1972. Sur la limite inf~rieure du Cambrian au Maroc. C.R. Acad. Sci., Paris, 275 D: 5--8.

Bouladon, J., Jouravsky, G. and Morin, Ph., 1950. Etude preliminaire des pegmatites muscovite et b~ryl du Sud de la plaine de Tazenakht. Notes Mere. Serv. G~ol. Maroc,

120: 37--49.

Caby, R., 1970. La Ch~ine Pharusienne dans le Nord-Ouest de t'Ahaggar (Sahara Central, Alg~rie); Sa Place dans l'Orogen~se du Pr~cambrien Sup~rieur en Afrique. Thesis, Univ. Montpellier, 335 pp.

Caby, R., 1970--72. Evolution pr~-orog4nique, site et agencement de la ch~ine pharusienne dans le Nord-Ouest de l'Ahaggar (Sahara alg~rien): sa place dans l'orogen~se panafricaine en Afrique occidentale. In: Colloque int. sur les correlations du Pr~cambrien. Notes

M~m. Serv. G~ol. Maroc, 236: 65--80; Coll. Int. CNRS, Paris, 192.

Caby, R. and Leblanc, M., 1973. L'orogen~se panafricaine sur les bords Est (Ahaggar nordoccidental) et Nord (Anti-Atlas) du craton ouest africain. Coll. g~ol. afficaine, Florence

(Italie). R~sum~ in Rech. G~ol. Afrique, CNRS, 2.

Cahen, L., Choubert, G. and Ledent, D., 1964. Premiers r~sultats de g~ochronologie sur le Pr~cambrien de l'Anti-Atlas (Sud marocain) par la m~thode strontium--rubidium. C.R. Acad. Sci., Paris, 258: 635--638.

Cahen, L., Choubert, G., Delhal, J. and Ledent, D., 1965. Premieres donn~es sur l'~ge du Pr~cambrien ancien de l'Anti-Atlas (Maroc) par la m~thode strontium--rubidium. Notes Serv. G~ol. Maroc, 24: 63--69.

Charlot, R., 1973. Geochronological studies in the Moroccan Precambrian. IIe European Coll. of Geochronol., Heidelberg, 1972; Fortsch. Mineral., 50: 57--59.

Charlot, R., 1974. G~ochronologie du Pr~cambrien de l'Anti-Atlas marocain: probl~me des ages ~ 2000 MA. 2e R~union Annu. des Sciences de la Terre, Nancy, France, p.106.

Charlot, R., 1975. Etude g~ochronologique des granites de Tafraout (Anti-Atlas, Maroc). Can. J. Earth Sci., 12: 19--23.

Choubert, G., 1952. Histoire g~ologique du domaine de l'Anti-Atlas in G~ologie du Maroc, facs. 1, 19e Congr. G~ol. Int. Alger. Monogr. Region~ 3e Ser. Maroc. No.6.

Choubert, G., 1963. Histoire g~ologique du Pr~cambrien de l'Anti-Atlas. Notes M~m. Serv. G~ol. Maroc, 164:352 pp.

Choubert, G., 1967. Divisions du Pr~cambrien en particulier dans l'Anti-Atlas. C.R. Somm. Soc. G~ol. Fr., 4:129 pp. 299

Choubert, G., 1973. Socle et zones mobiles anciennes du Maroc. Correlations. Rapport Sci. Interne CNRS, France.

Choubert, G., Diouri, M. and Faure Muret, A., 1965. Mesures g~ochronologiques r~centes par la m~thode 40A/40K au Maroc. Notes Serv. G~ol. Maroc, 24: 53--61.

Choubert, G. and Faure Muret, A., 1970. Livret guide de I'excursion Anti-Atlas occidental et central. Colloque int. Sur les correlations du Pr~cambrien. Notes M~m. Serv. Gfiol. Maroc, 229:259 pp.

Choubert, G., Boudda, A. and Faure Muret, A., 1973. Essai de chronologie du Pr~cambrien supfirieur du Sud marocain. C.R. Somm. Soc. Gfiol. France, XV: 123--124.

Choubert, G., Faure Muret, A., Hassenforder, B. and Jeannette, D., 1974. Nouvelle interpr~ tation du Pr~cambrien ancien de l'Anti-Atlas (Maroc). C.R. Acad. Sci., Paris, 278: D, 2095--2098.

Clauer, N. and Leblanc, M., 1974. Implications stratigraphiques d'une étude géochronologique rubidium--strontium sur m~tas~diments pr~cambriens de Bou Azzer (Anti-Atlas Maroc). Accepted Note to be published in Notes M~m. Serv. G~ol. Maroc.

Cormier, R.F., 1969. Radiometric dating of the Coldbrook group of southern New-Brunswick, Canada. Can. Earth Sci., 6: 393--398.

Fairbairn, H.W. and Hurley, P.M., 1970. Northern appalachian geochronology as a model for interpreting ages in older orogens. Eclog. Geol. Helv., 63: 83--90.

Fairbairn, H.W., Bottino, M.L., Pinson, W.H. and Hurley, P.M., 1966. Whole rock age and initial STSr/8~Sr of volcanics underlying fossiliferous lower cambrian in the atlantic provinces of Canada. Can. J. Earth Sci., 3: 509--521.

Hassenforder, B. and Jeannette, D., 1974. Rapports entre le Pr~cambrien sup~rieur et l'Infracambrien darts la boutonniere de Kerdous (Anti-Atlas occidental). C.R. Acad. Sci. Paris, 278 (D): 3287--3290.

Juery, A., Lancelot, J.R., Hamet, J., Proust, F. and Allegre, C.J., 1974. L'~ge des rhyolites du P.III du Haut Atlas et le probl~me de la limite Precambrien--Cambrien, IIe R~un.Ann. Sci. Terre. Nancy, France, p. 230.

Leblanc, M., 1970--72. Un complexe ophiolitique dans l~ Pr£~cambrien II de l'Anti-Atlas central (Maroc): description, interpretation et position stratigraphique. In: Coil. Int. sur les Correlations du Pr~cambrien. Notes M~m. Serv. G~ol. Maroc, 236:119--144; Coll. Int. CNRS, Paris, 192, Leblanc, M., 1975. Ophiolites pr~cambriennes et g~es arsenics de cobalt (Bou-Azzer, Maroc). Thesis. Centre g~ologique et g~ophysique Montpellier. 329 pp; Notes Mem. Serv. G~ol. M~roc, in press.

Priem, H.N.A., Boelrijk, N.A.I.M., Hebeda, E.H., Verdurmen, E.A.Th. and Verschure, R.H., 1973. Rb--Sr investigations on Precambrian granites, granitic gneisses and acidic metavolcanics in Central Telemark: Metamorphic resetting of Rb,Sr whole rocks systems. Norge Geol. Unders., 289: 37--53.

Tisserant, D., 1971. Rapport annuel d'activit~ du Service de la Carte g~ologique du Maroc. Vidal, Ph., 1967. M~thode de Datation Potassium--Argon. Mise au Point d'un Appareillage Destin l'Analyse de l'Argon. DiplSme Ing. G~ophys. Inst. De Physique du Globe, Strasbourg, France, 30 pp.

Source web Par : RENE CHARLOT