Vertebrate assemblages from the early Late Cretaceous of southeastern Morocco: An overview

Journal of African Earth Sciences 57 (2010) 391–412

Contents lists available at ScienceDirect

Journal of African Earth Sciences

journal homepage: www.elsevier .com/locate / ja f rearsc i

Geological Society of Africa Presidential Review No. 16

Vertebrate assemblages from the early Late Cretaceous of southeasternMorocco: An overview

L. Cavin a,*, H. Tong b, L. Boudad c, C. Meister a, A. Piuz a, J. Tabouelle d, M. Aarab c, R. Amiot e,E. Buffetaut b, G. Dyke f, S. Hua g, J. Le Loeuff f

a Dpt. de Géologie et Paléontologie, Muséum de Genève, CP 6434, 1211 Genève 6, Switzerlandb CNRS, UMR 8538, Laboratoire de Géologie de l’Ecole Normale Supérieure, 24 rue Lhomond, 75231 Paris Cedex 05, Francec Faculté des Sciences et Techniques, BP, 509, Boutalamine, Errachidia, Moroccod Musée Municipal, 76500 Elbeuf-sur-Seine, Francee IVPP, Chinese Academy of Sciences, 142 XiZhiMenWai DaJie, Beijing 100044, Chinaf School of Biology and Environmental Science, University College Dublin, Belfield, Dublin 4, Irelandg Musée des Dinosaures, 11260 Espéraza, France

a r t i c l e i n f o

Article history:Received 16 June 2009Received in revised form 9 December 2009Accepted 11 December 2009Available online 23 December 2009

Keywords:AfricaMesozoicVertebrataPalaeoecologyPalaeogeography

1464-343X/$ - see front matter � 2009 Elsevier Ltd. Adoi:10.1016/j.jafrearsci.2009.12.007

* Corresponding author. Fax: +41 22 4186301.E-mail address: [email protected] (L. Cavin).

a b s t r a c t

Fossils of vertebrates have been found in great abundance in the continental and marine early Late Cre-taceous sediments of Southeastern Morocco for more than 50 years. About 80 vertebrate taxa have so farbeen recorded from this region, many of which were recognised and diagnosed for the first time based onspecimens recovered from these sediments. In this paper, we use published data together with new fielddata to present an updated overview of Moroccan early Late Cretaceous vertebrate assemblages. The Cre-taceous series we have studied encompasses three Formations, the Ifezouane and Aoufous Formations,which are continental and deltaic in origin and are often grouped under the name ‘‘Kem Kem beds”,and the Akrabou Formation which is marine in origin. New field observations allow us to place fourrecognised vertebrate clusters, corresponding to one compound assemblage and three assemblages,within a general temporal framework. In particular, two ammonite bioevents characterise the lower partof the Upper Cenomanian (Calycoceras guerangeri Zone) at the base of the Akrabou Formation and theupper part of the Lower Turonian (Mammites nodosoides Zone), that may extend into the Middle Turonianwithin the Akrabou Formation, and allow for more accurate dating of the marine sequence in the studyarea. We are not yet able to distinguish a specific assemblage that characterises the Ifezouane Formationwhen compared to the similar Aoufous Formation, and as a result we regard the oldest of the four ver-tebrate ‘‘assemblages” in this region to be the compound assemblage of the ‘‘Kem Kem beds”. Thiswell-known vertebrate assemblage comprises a mixture of terrestrial (and aerial), freshwater and brack-ish vertebrates. The archosaur component of this fauna appears to show an intriguingly high proportionof large-bodied carnivorous taxa, which may indicate a peculiar trophic chain, although collecting biasesalter this palaeontological signal. A small and restricted assemblage, the OT1 assemblage, possibly corre-sponds to a specific, localised ecosystem within the Kem Kem beds compound assemblage. Microfossilsand facies from the Aoufous Formation, corresponding to the top of the compound assemblage, provideevidence of extremely abiotic conditions (hypersalinity), and thus of great environmental instability. Atthe base of the Akrabou Formation the first ammonite bioevent, Neolobites, corresponds to the onset ofthe marine transgression in the early Late Cenomanian while the Agoult assemblage (Late Cenomanian?)contains a variety of small fish species that have Central Tethyan affinities. Finally, the youngest Mam-mites bioevent in the late Early Turonian corresponds to a deepening of the marine environment: thissequence is isochronous with the Goulmima assemblage, a diverse collection of fish and other marinetaxa, and shows affinities with taxa from the South Atlantic, the Central Tethys and the Western Interiorseaway of North America, and further highlights the biogeographical importance of these North AfricanLate Cretaceous assemblages.

� 2009 Elsevier Ltd. All rights reserved.

ll rights reserved.

http://dx.doi.org/10.1016/j.jafrearsci.2009.12.007

mailto:[email protected]

http://www.sciencedirect.com/science/journal/1464343X

http://www.elsevier.com/locate/jafrearsci

392 L. Cavin et al. / Journal of African Earth Sciences 57 (2010) 391–412

Contents

FCv

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3922. Geological setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3933. Vertebrate assemblages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394

ig. 1.archaribraye

3.1. The compound assemblage of the Kem Kem beds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394
3.1.1. Faunal list . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3953.1.2. Age, palaeoenvironment and palaeogeography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399
3.2. The OT1 assemblage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402
3.3. The marine transgression or Neolobites bioevent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4023.4. The Agoult assemblage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402


3.5. The Goulmima assemblage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405
4. Discussion and conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 408Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409

1. Introduction

The Cretaceous deposits surrounding the northern, eastern andsouthern borders of the Palaeozoic Tafilalt and Mader basins inMorocco have yielded numerous vertebrate remains for more than50 years. The first discoveries of fossil vertebrates in the ‘GrèsInfracénomaniens’ of the Kem Kem region were made by Frenchgeologists, including Choubert who found tooth plates of lungfishand ganoid scales south of Taouz in 1938 (Choubert, 1948; Chou-bert et al., 1952). Indeed, the first illustrations of vertebrate fossilsfrom the Kem Kem beds were published by Choubert et al. (Fig. 1A)

Plates from Choubert et al. (1952): (A) the first illustrations of vertebrateodontosaurus (not Megalosaurus) saharicus. Fig. 3, labelled as Crocodilus sp., is probaanus (Fig. 3).

in 1952. In the late 1940s and early 1950s, Lavocat (1948, 1949,1951, 1954a,b) published a series of notes on his palaeontologicaldiscoveries in this area of Morocco, with special emphasis on hisdinosaur finds. In 1954, he published a geological synthesis ofthe Kem Kem area (Lavocat, 1954b) which is still one of the bestaccounts of the local geology. The first record of marine fish frag-ments in the Turonian of the High-Atlas and Midelt area had beenmade slightly earlier by Dubar (1949) and over the following dec-ades only a few vertebrate remains from the ‘mid’ Cretaceous ofthis area of Morocco were described, including isolated fish frag-ments described by Tabaste (1963). In 1971, a German team from

fossils from the Kem Kem beds. Fig. 2 shows a theropod tooth referable tobly a Spinosaurus tooth; (B) invertebrates from the Cenomanian including Neolobites

L. Cavin et al. / Journal of African Earth Sciences 57 (2010) 391–412 393

the University of Göttingen discovered vertebrate fossils in theTaouz area, including fish (Wenz, 1980, 1981) and dinosaurs(Buffetaut, 1989); over the last 20 years, local people in this partof Morocco have engaged in very active excavation work to collectCretaceous fossils for commercial purposes (while Palaeozoicinvertebrates from the Tafilalt and Mader basins are also being ex-tracted and sold). Because of such commercial activity, beautifullypreserved fossil specimens have been discovered in both the KemKem beds and in the Goulmima area, but often with little associ-ated information about sedimentology, stratigraphy and geograph-ical location. Moreover, because of the rudimentary techniquesused by local collectors to remove fossils from the field, specimensare often incomplete, and numerous specimens or fragmentaryfossils with no commercial value are often neglected and not col-lected (McGowan and Dyke, 2009). Although important for the lo-cal economy, this collecting activity is harmful for the preservationof Moroccan palaeontological heritage and for scientific studies,even though natural history museums worldwide have acquiredMoroccan specimens from fossil dealers, allowing some heritagepreservation and research to proceed. In any case, we know thatspecimens corresponding to about 80 early Late Cretaceous verte-brate taxa from this region housed in various public collectionshave been described so far. Most of these fossils, however, werefound by local people and thus have little associated field data.

In this paper we present the results of fieldwork sporadicallyconducted by us over the course of 15 years, specifically the firststratigraphic framework for understanding the succession ofMoroccan Cretaceous vertebrate assemblages. Although we foundfew complete specimens of vertebrates during our field trips, alarge amount of fragmentary material was collected, together withstratigraphic and geographical observations. These data are veryimportant because they allow us to present, for the first time, a re-gional synthesis of the succession of vertebrate assemblages in theearly Late Cretaceous of southeastern Morocco.

2. Geological setting

At the beginning of the Mesozoic, the mountains of the Variscorogeny located in the region corresponding to the modern Wes-tern Sahara were peneplained. At the same time rifting associatedwith the opening of the Central Atlantic, at the origin of the CentralAtlantic Magmatic Province (CAMP) 200 Ma ago, caused an uplift ofthe Moroccan side of the rift that prevented continuous depositionof Mesozoic sediments in western Morocco, ultimately leading tothe extensive exposures of Palaeozoic and Precambrians rocks inthis region (Fabre, 2005; Michard et al., 2008; Frizon de Lamotteet al., 2009). Although the movement of Africa relative to a fixedEurope was an eastward lateral displacement from about 175 Macaused by the opening of The Central Atlantic, continental (terres-trial and freshwater) biogeographic affinities with South Americaremained strong in this region during the whole of the Early Creta-ceous. Because of the opening of the South Atlantic, the movementof Africa relative to Europe changed progressively into a N–S con-vergence between ca. 92 and 46 Ma and eventually led to the sur-rection of the Atlas system, which was contemporaneously upliftedfrom Morocco to Algeria and Tunisia from the Middle Eocene to Re-cent (Frizon de Lamotte et al., 2009).

During the Middle and Late Jurassic, most parts of Moroccowere emerged as shown by series of continental and brackishdeposits in the central and eastern High-Atlas, as well as in theMiddle-Atlas regions (Charrière et al., 1994, 2005). Followingemergence, the first marine ingression is Early Barremian in age(Charrière et al., 2005). During the Barremian–Aptian two narrowelongated marine gulfs extended northwards along the Middle-At-las and from the western Essaouira basin (Frizon de Lamotte et al.,

2009). In the Early Cretaceous, North Africa was extensively cov-ered with deltaic and brackish deposits, which were first referredto as the ‘Continental Intercalaire’ by Kilian (1931). As a package,this series of sediments has proved difficult to date because ofthe absence of marine incursions, but recent attempts to do sobased on vertebrate assemblages indicate that Nigeran outcropsof the ‘Continental Intercalaire’ [which does not comprise a Palae-ozoic component in this country, the so-called ‘Continental post-tassilian’ of Furon and Lombard (1964)], might be as old as lateMiddle Jurassic (Rauhut and López-Arbarello, 2009). In Morocco,the ‘Continental Intercalaire’ occurs in the southeastern part ofthe country, and apparently contains the uppermost part of theseries only, which is Cenomanian in age (see below). The succes-sion is topped by coastal and then open marine sediments that cor-respond to the great Cenomanian–Turonian worldwidetransgression (Gale, 2000).

In Morocco the ‘Continental Intercalaire’ – locally called the‘Kem Kem beds’ – belongs to the so-called ‘hamadas Cretaceous–Tertiary palaeogeographic domain’ in contrast to other Creta-ceous–Tertiary Plateaus in the region, including the Phosphate Pla-teau north of the Atlas Mountains and the coastal Laayoune-BouCraa Plateaus in the southwest. The hamada domain extends tothe south of the High-Atlas, mainly over the Sahara craton (Zouhriet al., 2008) and belongs to the Plateforme Préafricaine (Ferrandiniet al., 1985; Ettachfini and Andreu, 2004). It comprises the Iminiand Goulmima narrow plateaus wedged between the Anti-Atlasand the High-Atlas and is limited to the north by the majorSouth-Atlasic fault, the Guir Hamada in the East, the Kem Kem Pla-teau in the South, and the Draa Hamada in the southwest (Zouhriet al., 2008). In the Moroccan Hamada domain, the time intervalspanned by the known vertebrate occurrences, the Cenomanian–Turonian, is characterised by important changes in lithofacies thatreflect a marked shift in depositional environments.

It has also proved difficult to precisely define many of the Cre-taceous Formations in Southeast Morocco. In the Anti-Atlas area,Choubert (1948) recognised the ‘trilogie mésocrétacée’ with, frombottom to top, the ‘Grès Infracénomaniens’, the ‘Marnes à gypsescénomaniennes’, and the ‘Calcaires cénomano-turoniens’. Althoughthe lowest unit had not yielded fossils at the time, Choubert (1948)regarded it as probably exclusively Albian in age and suggestedthat it corresponds to the ‘Continental Intercalaire’ defined by Kil-ian (1931). In the middle unit, Choubert (1948) mentioned theoccurrence of Cératodes[sic] africanus and ganoid scales (identifiedby Camille Arambourg), and in the upper part the occurrence of theammonite Neolobites vibrayeanus. Later, Choubert et al. (1952)mentioned the recent discoveries of fish and reptiles in the ‘GrèsInfracénomaniens’ by Lavocat, and noted again that this faunalooked similar to the assemblage from the ‘Continental Interca-laire’. Interestingly, Choubert et al. (1952) observed that thisassemblage is also similar to the Cenomanian assemblage fromEgypt discovered by Stromer at the beginning of the 20th Century.Although they all noticed resemblances with Cenomanian assem-blages, Choubert and other authors, such as Lavocat, still consid-ered this unit to be infracenomanian in age. More recently, thelowest two parts of the trilogy were gathered into the informallynamed Kem Kem beds (Sereno et al., 1996), with a lower unit (cor-responding to the ‘Grès Infracénomaniens’) and an upper marlyunit (corresponding to the ‘marnes à gypses cénomaniennes’).These names have been widely used ever since.

In the ‘Sillon Préafricain’, along the southern slope of the High-Atlas, Dubar (1948) also defined three formations in the Cretaceousseries which are, from bottom to top: the sandstone Ifezouane For-mation, referred to the Albian without palaeontological evidence,the Aoufous Formation, mainly clayey sandstones and green marlswith gypsum, referred to the Albian–Cenomanian, and the AkrabouFormation which corresponds to the Cenomanian–Turonian


limestone hamada. Sereno et al. (1996, note 21) also remarked thatthe underlying beds west of Goulmima consist of a marginal mar-ine-evaporite facies characterised by alternating marl, limestoneand gypsum that do not form two distinct, subequal subunits. Sec-tions published by Ettachfini (2008) and our own observations inthe southeastern vicinity of the town of Goulmima indicate, how-ever, that the Ifezouane Formation (the lower sandstone unit) oc-curs in most of the eastern and southern outcrops of theCretaceous series of the ‘Sillon Préafricain’ (as already mentionedby Choubert, 1948), although it is generally less well exposed thanthe ‘Grès Infracénomaniens’ (or ‘lower unit’) in the Kem Kem area.

From a palaeogeographical point of view, Choubert (1948) con-sidered that Moroccan Cretaceous sediments were deposited intwo different basins: the Atlasic basin (including the ‘Sillon Préa-fricain’), covering the present Atlas range and reaching the Atlanticin the west, and the Taouz Basin, formed by the Taouz hamada andthe Kem Kem. He suspected that sporadic communications be-tween both basins were probable, possibly along north–south ori-

Fig. 2. Satellite photograph (A) and simplified geological map (B) to show thelocalities discussed in the text (stars). 1, Tadirhoust-Asfla; 2, Tadirhoust; 3, spring ofGoulmima; 4, road of Goulmima; 5, Ziz; 6, Takemout; 7, Zaouia; 8, Douira; 9, JebelAl Qabla; 10, Slilim; 11, Khetitila Srhira; 12, Bou Laâlou; 13, Chaaft; 14, El Begâa; 15,El Begâa – Taouz; 16, Tizi Tazguart; 17, Tizi Momrad; 18, Gara Sbâa – Agoult; 19,Tazoughard – Belkasem; 20, Tazougard SW. Black stars indicate localities corre-sponding to the stratigraphic column shown in Fig. 3.

ented depressions running across the Tafilalt basin (Choubert et al.,1952). Lavocat (1948, 1954b) then showed that the ‘Infracénoma-nien’ is present below the Tertiary deposits of the Guir Hamada,thus demonstrating a connection between the basin of the ‘SillonPréafricain’ in the north and the Taouz Basin in the south.

Consequently, there is no sedimentological or palaeogeograph-ical evidence in favour of separating the early Late Cretaceous ser-ies located in the ‘Sillon Préafricain’ from the series located in the‘Taouz Basin’ or the Kem Kem area. Following Ettachfini and And-reu (2004), we thus propose use of the names Ifezouane, Aoufousand Akrabou formations to classify the Cretaceous sediments inthe area. We also suggest retention of the term ‘Kem Kem beds’to characterise the vertebrate assemblages of both the Ifezouaneand Aoufous formations as so far we have not been able to distin-guish two different sets of taxa in this package of sediments. Thusthe expression ‘compound assemblage’ for the Kem Kem beds isused in this paper to characterise the vertebrate taxa found inthese two formations.

Because this succession of vertebrate assemblages was depos-ited in a single sedimentary basin, and because no large hiatusesare present in the deposits, we consider that they mirror the bioticand abiotic evolution that occurred in this region of northwestAfrica between the Early Cenomanian and the Middle Turonian,over a timescale of around 8 million years. Localities visited bythe authors during the last decade are shown in Fig. 2; in May2008 eight stratigraphic sections were measured (Fig. 3).

3. Vertebrate assemblages

3.1. The compound assemblage of the Kem Kem beds

The Ifezouane Formation is composed mainly of detritic sand-stones with cross-stratified structures (Fig. 4B) and is the richestin terms of numbers of disarticulated vertebrate fossils, especiallyat localities situated in the southern part of the outcrop area(Fig. 4A). The Ifezouane Formation rests unconformably on Palaeo-zoic rocks and its base – composed of conglomerates and breccia –fossilised the peneplain. Its thickness is very variable, from 0 to250 m, and normally decreases from the south to the north (Chou-bert, 1948). In contrast the Aoufous Formation is characterised bymarls that contain intercalations of detritic sandstone and micro-conglomerates (Fig. 4D). In localities in the southern Kem Kemarea, this formation has yielded only a few isolated vertebrate re-mains, mainly Onchopristis teeth, while localities to the north ap-pear to be richer (Fig. 2): at the Douira locality, for example,abundant vertebrate remains are found in the upper unit, wherethey are systematically excavated by local collectors (Fig. 4D).The thickness of the Aoufous Formation is between 100 and200 m; its facies and microfossil content are described below (Sec-tion 3.1.2). The vertebrate occurrences that are known from boththe Ifezouane and Aoufous formations are discussed together inthis paper as a single compound assemblage, that of the KemKem beds.

Vertebrate fossils that comprise the Kem Kem compoundassemblage have been collected along a series of outcrops thatspan some 250 km along a westwardly inclined ‘U’ around thewestern end of the Anti-Atlas belt (Fig. 2). The more northerly ofthe recognised fossiliferous outcrops have been recorded north-west of Erfoud (at the localities of Tarda and Jorf, not shown inFig. 2), while to the east localities have been recorded in the vicin-ity of the village of Douira (Takemout, Zaouia, Douira) and then ex-tend to the east (Hasi Beraber) and south, especially to the east ofthe town of Taouz (El Begâa, El Begâa – Taouz). Fossiliferous out-crops have also been traced southwest (Ouzina) towards the KemKem area (Tizi Tazguart, Tizi Momrad, Gara Sbâa) and occur farther

Fig. 3. Stratigraphic column measured through the eight localities in Fig. 2.


south (Lavocat, 1954b); we have not yet visited this region and nopublished data are available at present.

As mentioned by Lavocat (1954b), fossils occur in low density inthese deposits which makes systematic excavation difficult; how-ever, some areas may yield rich collections of bones and bone frag-ments, as noted initially by Lavocat (1954b, Fig. 5A) and confirmedby two of us (H.T., J.T.) during a three-week series of systematicexcavations conducted at several localities in the Kem Kem beds(May 2008). According to Lavocat (1954b), fossils are sometimesconcentrated at the surface by erosion but most of these concen-trations have now vanished as a result of continuous surface col-lecting. At these more ephemeral localities fossils are associatedwith the kerkoub facies (Fig. 4C), characterised at the surface byconcentrations of sandstone spheres that range from around 2 to10 cm in diameter (Lavocat, 1954b).

3.1.1. Faunal listThe most abundant vertebrate remains found in the Kem Kem

beds are large isolated rostral teeth of the sclerorhynchid sharkOnchopristis numidus (Fig. 4C). For the most part these teeth arefound isolated, but fragments of rostral cartilage, sometimes asso-ciated with rostral teeth, are also quite common and an almost

complete skull has recently been reported (Dutheil and Brito,2009). In addition to O. numidus, Sereno et al. (1996) and Dutheil(1999a) recorded seven other, less abundant, elasmobranch taxafrom the Kem Kem beds: Asteracanthus aegyptiacus, Distobatus nu-tiae, Tribodus sp., ’Lissodus’ sp., Serratolamna amonensis, Cretoxyrhi-nidae indet., and Marckgrafia lybica (Dutheil, 1999a). Fin spines ofhybodonts are also rather common: Dutheil (1999a) recognisedtwo forms that he referred to Asteracanthus and to Tribodus/‘Lisso-dus’. In addition, the dental plates of lungfish are abundant in theKem Kem beds (recorded from the localities of Douira, El Begâa,Tizi Tazguart and Tizi Momrad, among others); indeed, in his pa-pers Lavocat variously referred to these deposits by the names‘grès à Onchopristis’ or ‘grès à Ceratodus’. Lungfish remains werefirst referred to Ceratodus humei by Tabaste (1963) – a taxon in-cluded in the genus Protopterus by Martin (1984a) but later re-ferred back to Ceratodus by Churcher and De Iuliis (2001) andChurcher et al. (2006) – while others have been referred to the neo-ceratodontid ‘Neoceratodus’ africanus (Martin, 1982, 1984b). Coela-canth skull fragments are also commonly encountered at thesesites (i.e., Douira, Tizi Momrad); interestingly, the one currentlyrecognised Kem Kem coelacanth was referred to Mawsonia lavocatiby Tabaste (1963) and Wenz (1980, 1981), and represents a genus

Fig. 4. Cretaceous outcrops and facies of SE Morocco. (A) outcrops in the Ifezouane Formation, with galleries that have been dug for collecting fossils (arrows) at El Begâa; (B)cross-bedding stratification in the Ifezouane Formation at El Begâa; (C) kerkoub facies with Onchopristis and theropod teeth (arrows), top of the Ifezouane Formation atDouira; (D) outcrops in the Aoufous Formation at Douira; (E) Akrabou Formation, showing the bank with Neolobites vibrayeanus, sea-urchins and mollusks, excavated by localresearchers (arrows). The top of the Aoufous Formation lies to the bottom left under the mass of falling rocks. East of Taouz; (F) Excavation for fossils from the Agoultassemblage; (G) Close up showing siliceous beds (arrows); (H) View of the Cretaceous deposits between the cities of Errachidia and Goulmima. The reddish lowest part andthe middle part covered with fallen rocks of the cliff correspond to the Aoufous Formation. The carbonate beds that top the cliff correspond to the Akrabou Formation. Notethe Jurassic southern slope of the High-Atlas in the background; (I) Cliff located 10 km to the north of the town of Goulmima (spring of Goulmima), showing beds excavatedby local researchers (arrows); (J) close up of outcrop in the Akrabou Formation showing the succession of limestone and marly limestone beds with nodules (arrows). Thesesediments sometimes contain ammonites and fishes from the Goulmima assemblage.


that is otherwise only known from South America, characterising awest-Gondwana Province. In agreement with this ‘South Americanconnection’, some more recently described skull remains also showfeatures characteristic of the Brazilian Cretaceous genus Axelrod-ichthys (Cavin and Forey, 2004).

Several taxa of Cladistia (polypterids) have been recorded fromthe Aoufous Formation; two are represented by isolated pinnulaefrom detritic beds (Dutheil, 1999a) while three others are knownon the basis of articulated specimens from the clayey localityOT1 (see below). In addition, at least two semionotiforms are

Fig. 5. Vertebrate macrofossils from the Kem Kem compound assemblage. (A) excavation conducted by René Lavocat (sitting on the ground) at Gara Sbâa in 1950 (Courtesy:late René Lavocat); (B and C) Stromerichthys aethiopicus, jaw fragment (B) and scales(C); (D) Concavotectum moroccensis, posterior part of a braincase in left lateral view; (E andF) Galianemys sp: first left costal plate (E); nuchal (F); (G and H) caudal vertebra of an indeterminate lithostrotian MDE-D108 in lateral (G) and dorsal (H) views. Scales barsequal 2 cm.


known to occur in the Kem Kem assemblage: one is a Lepidotes-likespecies known from undescribed skulls housed in the Natural His-tory Museum, London (Peter Forey, personal communication,2007), and on the basis of numerous isolated scales found in situat most of the localities visited. Some of these scales show the‘double-peg pattern’ regarded as a possible synapomorphy forthe family Semionotidae (Cavin et al., 2009). Indeed, the sheer sizeof the preserved skulls and some loose scales indicates the pres-ence of very large individuals, probably up to several metres inlength. The only other currently known semionotiform fish is thegar Oniichthys falipoui (Cavin and Brito, 2001): fragments and iso-

lated scales referable to this species are common in many partsof the Kem Kem area (e.g., the unknown type locality, fragmentsat El Begâa, Gara Sbâa, etc.).

The halecomorph species Calamopleurus africanus was describedbased on a complete articulated skull of uncertain origin collected‘close to Taouz’ (Forey and Grande, 1998); during our 2008 fieldseason several additional isolated fish fragments referable toStromerichthys aethiopicus were also recovered. This record de-serves comment because this genus, described by Weiler (1935)on the basis of material from Bahariya, Egypt, and now housed inthe Bayerische Staatssammlung für Paläontologie und Geologie in


München, Germany (part of the material was destroyed by thebombing of the Museum during World War II, but some materialis still kept in the collection, in particular a fragmentary maxillaand one scale) (O. Rauhut and A. López-Arbarello, personal com-munication, 2009), is of uncertain systematic relationships. Mostmaterial of Stromerichthys from the Kem Kem beds consists of iso-lated teeth and scales which have a characteristic shape (i.e., teethand jaw fragments from El Begâa, Tizi Tazguart and Tizi Momrad;scales from Boudarite, Takemout, El Begâa, Tizi Tazguart and TiziMomrad). These teeth are often rather large and have a crown thatreaches 13 mm in length in the largest specimen, shaped as aslightly curved cone. This curvature affects either the whole crownof the tooth or the tip only, while the surface bears very fine stria-tions and the apex has a well-marked acrodine tip. However, onefragment of maxilla (Fig. 5B) containing two complete teeth andthe base of two others is very reminiscent of the specimen de-scribed by Weiler (1935: pl. II, Figs. 17 and 23). Scales, already re-ported from this deposit by Tabaste (1963) and Dutheil (1999a),typically have an ornamentation consisting of strong longitudinalridges with a larger central one (Fig. 5C). Fragments with strongossifications and coarse ornamentation, slightly reminiscent ofbones from mawsoniid coelacanths, may also be referrable toStromerichthys based on comparisons with material describedand figured by Weiler (1935: Fig. 2). This taxon, referred to thefamily Gigantodontidae by Weiler (1935), certainly deserves fur-ther study.

Teleosteans from the Kem Kem beds are represented by the ich-thyodectiform Cladocyclus pankowskii (Forey and Cavin, 2007; fromunknown type and other localities;), by the weird osteo-glossomorph Palaeonotopterus greenwoodi (Forey, 1997; Taverneand Maisey, 1999; Taverne, 2000, 2004; Cavin and Forey, 2001;type locality unknown but fragments recorded at least at Hasi Ber-aber), by a new taxon of notopterid (Brito et al., 2009), by the pos-sibly basal ostariophysan Erfoudichthys rosae (Pittet et al., 2009;unknown type locality), and by the tselfatiiform Concavotectummoroccensis (Cavin and Forey, 2008; unknown type locality, frag-ments collected at Douira and Tizi Momrad including severalbraincases from field work in 2008) (Fig. 5D). The lack of well-doc-umented locality information for most teleostean fish also repre-sents an area for future work.

Rage and Dutheil (2008) recently described a number ofamphibians and squamates from the Kem Kem beds, includingamphibians – the sirenid Kababisha sp., the pipid frog Oumtkoutiaanae and non-pipids frogs – an indeterminate lizard, snakes thatappear similar to Simoliophis libycus, indeterminate madtsoiidsand nigerophiids. The known turtle fauna of the Kem Kem com-pound assemblage is also diverse, with four families of pleurodi-rans represented. Of these, euraxemydids comprise Dirqadimschaefferi, the podocnemidids Hamadachelys escuilliei, the bothre-mydids Galianemys whitei and G. emringeri, and the araripemydidsAraripemys sp. (Gmira, 1995; Tong and Buffetaut, 1996; Gaffneyet al., 2002, 2006). So far, because all the described chelonian fos-sils were collected by local people and purchased by museums, theexact locations and stratigraphic horizons of these taxa are un-known. However, two large turtle shells figured by Gaffney et al.(2006) and confidently assigned to Galianemys sp., were collectedat Tizi Tazguart (Fig. 2). According to the nomad who discoveredthis locality more than 30 large turtle shells have been unearthedat this site; during our field work, isolated plates of Galianemyssp. were collected in Tizi Tazguart and Tizi Momrad (Fig. 5E andF) and a few fragmentary plates of Araripemys sp. were encoun-tered at El Begâa and Gara Sbâa.

Crocodilian remains, notably isolated teeth (Larsson and Sidor,1999), are very abundant in the Kem Kem beds and indicate a highdiversity of taxa with various dietary adaptations as well as severalforms that remain undescribed. Although Lavocat (1955a) reported

a species of the long-snouted eusuchian Thoracosaurus, T. cherifien-sis, from the Kem Kem, his fossil material was subsequently shownby Broin (2002) to belong to an advanced, mesosuchian-gradecrocodilian described as Elosuchus cherifiensis. Buffetaut (1976)also reported the presence of Libycosuchus in the Kem Kem bedson the basis of the posterior part of a skull, which, according toLarsson and Sues (2007), belongs to another genus which they callHamadasuchus. This is worthy of further discussion because Hama-dasuchus rebouli was described by Buffetaut (1994) on the basis ofa dentary, interpreted as belonging to a small, ziphodont tremato-champsid while Larsson and Sues (2007) subsequently referredadditional specimens to Hamadasuchus, including a well-preservedskull. It appears likely that Larsson and Sues’s (2007) attribution ofadditional specimens to Hamadasuchus is erroneous and that thematerial in question should be placed in a distinct genus, appar-ently belonging to the Trematochampsidae: Larsson and Sues(2007) refer the skull they describe to the Peirosauridae, but thistaxon is in all likelihood a junior synonym of Trematochampsidae(Buffetaut, 1988). An additional well-preserved skull (Hua et al., inprep.) belongs to yet another taxon of trematochampsid and sup-ports the idea of a diverse trematochampsid radiation in the Ceno-manian of Africa. The recent descriptions by Sereno and Larsson(2009) of Araripesuchus rattoides and Laganosuchus maghrebensisfurther increases the known diversity of crocodilians from theKem Kem beds. Indeed, Araripesuchus rattoides provides an addi-tional example of clear similarities with South American faunas;Araripesuchus was originally described from Brazil, and later fromNiger (Buffetaut and Taquet, 1979), whereas the stomatosuchidLaganosuchus provides another faunal element indicating similari-ties with the Bahariya fauna of Egypt, which has yielded Stoma-tosuchus inermis (Stromer, 1925).

Above all, however, the Kem Kem beds are now well known fortheir dinosaur assemblage. Lavocat (1954a) described the rebba-chisaurid sauropod Rebbachisaurus garasbae on the basis of a par-tial skeleton which has still been only partially described. Forexample, a scapula and a dorsal vertebra were figured, but Lavocat(1954a, 1955b) mentioned the discovery of 11 vertebrae (includingsix articulated caudal vertebrae), a sacrum, 10 ribs, a humerus andpelvic bones at the Gara Sbâa locality between 1949 and 1952. Avery large dorsal vertebra from the Kem Kem beds – from an un-known locality – kept at the Musée des Dinosaures, Espéraza(MDE), is also referrable to R. garasbae, together with a large num-bers of isolated teeth.

The presence of additional sauropod groups in the Kem Kemassemblage was first indicated by Russell (1996) who suggestedthe presence of andesaurine sauropods on the basis of amphypl-atian middle and posterior caudals, although others have sug-gested that these elements may belong instead to arebbachisaurid (Salgado et al., 2004). In addition, a single isolatedprocoelous middle caudal vertebra (MDE-D108) can be referredto an indeterminate lithostrotian; very similar lithostrotian cau-dals have been reported from the Cenomanian of Sudan (Werner,1994; Rauhut, 1999). By way of comparison, the Bahariya localityin Egypt has yielded Aegyptosaurus aegyptiacus (Stromer, 1932),while Stromer also suggested the presence at Bahariya of a secondlarge sauropod belonging to the family Dicraeosauridae, and mostrecently Smith et al. (2001) described Paralititan stromeri, from thesame locality. Although the currently known fossil material fromthe Kem Kem is too incomplete to allow identification of the Egyp-tian taxa, it nevertheless does demonstrate the coexistence of atleast two sauropod families in the area and further reinforces sim-ilarities between the Egyptian and Moroccan assemblages. Finally,O’Leary et al. (2004) described elements from an armoured salta-saurid from the ‘Continental Intercalaire’ in Mali, although thesedeposits are poorly dated and might be older than those of theKem Kem beds (Cenomanian, see below).


The remains of theropod dinosaurs are very abundant indeed inthe Kem Kem beds, especially isolated teeth. As early as 1948,Lavocat reported the presence of ‘megalosaurid’ teeth in thesedeposits. Indeed, as the teeth of large carnivorous theropods areespecially sought after by fossil collectors, specimens are preferen-tially retained by local dealers and has lead to the observation thattheropods are over-represented in commercially obtained collec-tions of Kem Kem beds fossils (McGowan and Dyke, 2009). In spiteof this collectorship effect, however, theropod diversity is still ex-tremely high in the Kem Kem beds; a number of taxa have been re-ported, and were likely present, but are so far known only fromincomplete, fragmentary material. One of the first specimens tobe described in detail was a tibia, which according to Lavocat(1954a), closely resembles Elaphrosaurus bambergii, from the LateJurassic of Tendaguru (Janensch, 1925), a taxon now usually con-sidered to be a ceratosaur (Tykoski and Rowe, 2004). This tibianeeds to be re-examined in light of the confirmed presence ofabelisaurid theropods in the Kem Kem assemblage (see below).

Most famously, large-bodied spinosaurs (Spinosauridae) arewell-represented in the Kem Kem beds, particularly again on thebasis of isolated teeth. However, it seems that early palaeontolo-gists who worked in these sediments did not recognise these teethas belonging to theropods because they closely resemble those oflarge crocodiles; indeed in the first published illustrations of verte-brate fossils from the Kem Kem beds (Choubert et al., 1952, pl. VI;Fig. 1A), a tooth designated as ‘Crocodilus sp.’ seems rather to befrom a spinosaur (Fig. 1). Taquet (1984) was the first to mentionthe occurrence of spinosaurids in the Kem Kem, while later, Buffe-taut (1989) described an upper jaw fragment from the Taouz areaas Spinosaurus cf. aegyptiacus, Milner (2003) referred more com-plete Kem Kem beds jaw remains to S. cf. aegyptiacus, and Dal Sassoet al. (2005) described the upper jaw of a large individual as Spino-saurus cf. aegyptiacus. In fact it seems as though two species ofSpinosaurus may well have been present in the Kem Kem beds:Russell (1996) named Spinosaurus maroccanus on the basis of anisolated cervical vertebra and referred a few other isolated re-mains, including a jaw fragment, to this taxon. Its status is ques-tioned, however, as Sereno et al. (1998) and Buffetaut and Ouaja(2002) have considered Spinosaurus maroccanus, which is basedon inadequate material, to be a junior synonym of S. aegyptiacusStromer, 1915. As a result the number of spinosaurid taxa presentin the Kem Kem beds remains uncertain, the situation being fur-ther complicated by the presence of several tooth morphotypes(some teeth, for example, have a smooth surface whereas othersare fluted); it is unclear whether this simply reflects individual var-iation or confirms the presence of several species.

Another large Kem Kem theropod, Sigilmassasaurus brevicollis,was described on the basis of a cervical vertebra by Russell(1996), who also referred various other isolated vertebrae to thistaxon. However the validity of Sigilmassasaurus is highly doubtful;Sereno et al. (1998) are probably correct in referring Russell’s(1996) material to Carcharodontosaurus. However, Novas et al.(2005) noted that the vertebrae are very different from the cervi-cals of South American carcharodontosaurids and it is intruigingthat vertebrae resembling those of Sigilmassasaurus have also beenfound alongside Carcharodontosaurus teeth at Albian localities insouthern Tunisia (Benton et al., 2000).

The large carcharodontosaurid theropod Carcharodontosaurussaharicus, however, is well-represented in the Kem Kem beds onthe basis of abundant isolated teeth and a well-preserved skull(Sereno et al., 1996). This skull was recently proposed as the neo-type for Carcharodontosaurus by Brusatte and Sereno (2007). Inaddition, fossil remains of abelisaurids were first reported fromthe Kem Kem by Russell (1996) who described a dentary fragmentas cf. Majungasaurus. The presence of these theropods was alsoconfirmed by Mahler (2005) on the basis of a maxilla while Delta-

dromeus agilis was described by Sereno et al. (1996) on the basis ofa partial skeleton. Deltadromeus, a lightly-built theropod, has vari-ously been considered Coelurosauria incertae sedis by Holtz et al.(2004), a possible noasaurid by Sereno et al. (2004), and a basalceratosaur by Carrano and Sampson (2008). Finally, the presenceof the family Dromaeosauridae in the Kem Kem assemblage wasestablished by Amiot et al. (2004) on the basis of isolated teethwhile ornithischian dinosaurs have only been reported to date onthe basis of their footprints (Sereno et al., 1996).

Pterosaurs are relatively common in the Kem Kem beds (Well-nhofer and Buffetaut, 1999; Kellner, 2009), although their fossilstend to be fragmentary, which is not unexpected in such detritic,high-energy deposits. However, the presence of isolated teeth –abundant at some localities – and disarticulated material demon-strates a large range of taxa. Wellnhofer and Buffetaut (1999) de-scribed specimens referred to at least four taxa, includingrepresentatives of ?Pteranodontidae, ?Azhdarchidae, Tapejaridaeand Ornithocheiridae, while Mader and Kellner (1999) erectedthe taxon Siroccopteryx moroccanus on the basis of a jaw fragmentreferred to an anhanguerid, which may in fact belong to the genusColoborhynchus (Barrett et al., 2008). Newly discovered jaw mate-rial confirms the occurrence of azhdarchids in the Kem Kem beds.

The single incomplete vertebra considered as avian by Riff et al.(2004) does not preserve any avian synapomorphies: although re-cords of these taxa remain to be confirmed in the Kem Kem bedscompound assemblage, birds were present alongside pterosaurs(GD, personal observation 2008).

3.1.2. Age, palaeoenvironment and palaeogeographyFor many years the vertebrate compound assemblage of the

Kem Kem beds was regarded as Early Cretaceous in age (i.e.,‘Infracénomanien’ for French geologists), because the youngestage of these sediments is constrained by the occurrence of theCenomanian ammonite Neolobites vibrayeanus in the overlyinglimestone (see below). However, on the basis of the known rep-tiles, the Kem Kem compound assemblage has also been consid-ered younger than the Tiouraren Formation of Niger (Taquet,1976; Moody and Sutcliffe, 1991; Russell, 1996), recently datedto the late Middle or early Late Jurassic (Rauhut and López-Arba-rello, 2009) and close in age to the Bahariya assemblage of Egypt(Choubert et al., 1952; Buffetaut, 1989, 2001; Tong and Buffetaut,1996; Russell, 1996; Sereno et al., 1996; Wellnhofer and Buffetaut,1999), particularly because of the occurrence in both assemblagesof Spinosaurus and Carcharodontosaurus. This age seems to be borneout by more recent palaeontological data: based on the sharkassemblage of the Kem Kem beds, Sereno et al. (1996) noted strongaffinities with the Bahariya assemblage. A close relationship be-tween the tselfatiiform fish Paranogmius doederleini from Bahariya,described by Weiler (1935) and re-interpreted by Taverne (2003),and Concavotectum moroccensis from the Kem Kem beds (Cavinand Forey, 2008) provides additional supporting evidence for thesimilar age of both assemblages. The Bahariya Formation is well-dated to the Early Cenomanian (Catuneanu et al., 2006) and as suchwould indicate an Early Cenomanian age for the Kem Kem beds. In-deed, other similar North African vertebrate assemblages have alsobeen considered Cenomanian in age, including the Wadi Milk For-mation, Sudan (Buffetaut et al., 1990; Werner, 1994), and the Ech-kar Formation, Niger Republic. Of these, the Wadi Milk assemblageis thought to be Cenomanian because of the occurrence of teethfrom the shark Asteracanthus aegyptiacus (Werner, 1994) and otherfaunal elements. In particular the dinosaurs of the Wadi Milk aresimilar to those known from the Bahariya and the Kem Kem beds,also reinforcing a similar age interpretation (Werner, 1994; Rau-hut, 1999). In Niger, the abelisaurids Rugops primus (Sereno et al.,2004) and the carcharodontosaurid Carcharodontosaurus iguidensis(Brusatte and Sereno, 2007) found in the Cenomanian Echkar


Formation, indicate taxonomic differentiation when comparedwith the dinosaurs from the Kem Kem beds, although both thero-pod families are common in both assemblages.

Microfossils found in the Aoufous Formation (see below) haveso far not proved helpful for dating these sediments. There is noevidence at the base of the Cretaceous series in the study areafor pre-Cenomanian fossils, although this possibility cannot bedefinitively ruled out. If we are correct, the sediments preservedin this region correspond only with the uppermost part of the Sah-aran ‘Continental Intercalaire’. Indeed, several of the formations in-cluded in the ‘Continental Intercalaire’ in other regions of NorthAfrica have yielded older vertebrate assemblages: Late Jurassic,or Hauterivian–Barremian, taxa from the Tiouraren Formation, Ni-ger (Sereno et al., 1994; Rauhut and López-Arbarello, 2009); ?Ap-tian taxa from the Elrhaz Formation, Niger (Taquet, 1970, 1976,1980; Sereno et al., 1994) (but see discussion of the age of this for-mation in Benton et al., 2000); and Albian taxa from Algeria (Depé-ret and Savornin, 1927), Tunisia (Bouaziz et al., 1988; Benton et al.,2000) and other Saharan localities (Lapparent, 1960).

As a result, the Kem Kem compound assemblage, as definedhere, encompasses fossils from two formations: the lower Ifezou-ane Formation and the upper Aoufous Formation (Ettachfini andAndreu, 2004). Although most of the vertebrate remains describedto date are from the lower Ifezouane Formation, the Aoufous For-mation also contains a number of vertebrate records; howevermany occurrences from across the Kem Kem, because they lackfield data, cannot be assigned to either of these formations withconfidence.

We now describe the sedimentology and microfossil contents ofboth formations separately, and present new data for the AoufousFormation. Finally we discuss the palaeoenvironmental implica-tions of the Kem Kem compound vertebrate assemblage in lightof our new findings.

No new information is presented for the Ifezouane Formation.The main facies of this unit consists of sandstones with cross-bed-

Fig. 6. Microfossils and facies from the Aoufous Formation. (A) microremains, mainly fi(ap300). Scale bar equals 100 lm; (C and D) gastropod lumachelle (D: ap297); (E) SilicifieScale bar equals 50 lm; (G) facies showing two size classes of quartz fragments (ap296

ded stratification, intercalations of pinkish sand and conglomerateswith quartz pebbles, and some calcareous layers containing lamel-libranchs and gastropods (Ettachfini and Andreu, 2004; Ettachfini,2008). Many excavations for vertebrate fossils conducted by localpeople are located in the upper part of the Ifezouane Formation(Fig. 4A).

Our field data indicate that the Aoufous formation is composedof three main facies: (1) variegated (sometimes sandy) claystones(+-hcl-and marlstones?); (2) fibrous, or saccharoid, gypsum layers;and (3) dolomitic limestone.

The micro-bio-components of the Aoufous claystones werepicked from washed and sieved (71, 125, 200 and 400 lm) sedi-ment. Most samples are sterile, but rarely they yield quite abun-dant organic components. Our preliminary study indicatesmostly fish remains (scales, vertebrae, teeth) associated with otherindeterminate vertebrate fragments (Fig. 6A, sample ap299, Douirasection) within these samples. Only apatitic microfossils are pres-ent in the claystone samples and no calcite shells are observed.This absence may be due to either palaeoenvironmental or preser-vational factors. Only a few samples with carbonaceous residues(i.e., sample ap302, Douira section) contain micro rhizocretions(i.e., preferential calcareous cementation around plant rootlets);quartz is frequent throughout and marcassite is sometimes pres-ent, indicative of a reductive environment.

The facies of the carbonaceous Aoufous Formation also containa depauparate fauna. Strong dolomitisation (Fig. 6B) is likely tohave eliminated any organic remains in most samples, althoughat Douira some do contain lumachelles consisting of monospecificgastropods (Fig. 6C and D). Thin sections also show abundant gas-tropods, thick and thin-shelled ostracods, as well as rare silicifiedfish scales, bones (Fig. 6E), and very rare agglutinated foraminifers(Fig. 6F). Ettachfini and Andreu (2004) noted the presence of undu-lated cryptalgal lamination within this facies, while quartz anddolomite are also abundant. Quartz fragments belong to two sizeclasses (Fig. 6G); large (250–600 lm) sub-rounded fragments,

sh, from the Douira locality (sample ap299); (B) dolomitised carbonaceous faciesd fish scale (ap300). Scale bar equals 200 lm; (F) Agglutinated foraminifera (ap297).). Scale bar equals 200 lm. For sample localities, see Fig. 3.


probably of detritic origin, and small (30–50 lm), subanguloushomometric fragments, probably of eolian origin.

When present, microfaunal assemblages are of very low diver-sity but contain rather abundant skeletal grains. This patternmay have been caused by: (1) the destruction of calcareous shellsby dissolution in claystones and subsequent dolomitisation of mostof the carbonaceous layers; (2) extreme abiotic conditions andgreat instability of the palaeoenvironment, with the rare fossil re-mains being the result of short transgressive episodes; or (3) ahypersaline palaeoenvironment, with rare occurrences of gastro-pods, ostracods and arenaceous foraminifers, known for their eury-haline tolerance (Fluegel, 2004). The common occurrence ofgypsum layers also provides evidence for hypersaline conditionsin the Aoufous Formation.

Our new sedimentological and micropalaeontological observa-tions agree with the model that the Aoufous Formation representsa coastal lagoon, or paralic sebkha, deposited in a medio- to supr-alittoral palaeoenvironment, as proposed by Ettachfini and Andreu(2004). Indeed, the known vertebrate assemblage is also in generalagreement with reconstructed continental palaeoenvironments forboth the Ifezouane and Aoufous formations, as evidenced by a par-ticularly abundant terrestrial component (i.e., a lizard, a madtsoiidsnake, dinosaurs and pterosaurs). On the other hand, the aquaticcomponent of these sequences – the sharks, bony fish and someof the crocodiles – is indicative of freshwater or brackish environ-ments. Amongst turtles, extant podocnemidids live in lacustrineand riverine environments while some taxa, in particular thosewith South American affinities (i.e., Tribodus, Calamopleurus, Clado-cyclus, etc.) and Western Tethyan or Central Atlantic affinities (i.e.,Stromerichthys, Simoliophis), may indicate marine influences,although no clear open marine indicators have yet been definitelyidentified. Simoliophis is a marine ‘hind-limbed’ snake restricted tothe Mediterranean Tethys, more widespread on its southern mar-gin, but with rare occurrences to the north (Bardet et al., 2008).

Fossil indicators of a strictly freshwater environment for thedetritic unit of the Kem Kem beds include sirenid urodels andfrogs. Russell (1996) proposed that the environment of this partof the sequence was likely a fluvial plain with westward drainage,while Dutheil (1999a) presented rose diagrams which show cross-stratified beds oriented north–south with a northward drainage.Russell (1996) also suggested that the remains of coelacanthsand other fish more than 3 m in length indicates the ‘existence ofsubstantial freshwater bodies upstream, either within the plainor possibly beyond it to the west’. The presence of large bodies ofwater is confirmed by our observations because of the occurrenceof other large fish, including Onchopristis and lungfish, the ‘Lepi-dotes-like’ semionotiform and Concavotectum moroccensis. Of these,the latter had a large buccal cavity with proportionally very longgill rakers (borne by the branchial arches) and was thus a filter fee-der (Cavin and Forey, 2008) reminiscent of the recent paddlefishPolyodon spathula (Grande and Bemis, 1991). Indeed, the large sizeof the freshwater fish and crocodilians from this deposit was alsoused as evidence for the presence of highly productive waters inmid-Cretaceous Saharan ecosystems (Russell and Paesler, 2003).Interestingly, large-bodied aquatic and terrestrial tetrapods arealso common in the Kem Kem compound assemblage and thus adirect ecological link between large fish and large carnivorous tet-rapods can be proposed for ichthyophagous predators, such ascrocodiles and spinosaurid theropods. The large body sizes at-tained by dinosaurs like Spinosaurus may be explained as an adap-tation to catch large fish. Indeed, and as already suggested byRussell (1996), other large theropods like Carcharodontosaurusmay have been attracted to the margins of the streams by the richfish community. Russell (1996) even went so far as to suggest thatthe major food chain linking components of the Kem Kem bedsvertebrate compound assemblage was short, with just two main

levels: fish and small aquatic reptiles at the base and theropodsat the top. This hypothesis, however, requires further quantitativestudy utilising theoretical ecological approaches because theapparent scarcity of herbivorous taxa may instead indicate biasedcollecting in the Kem Kem area (McGowan and Dyke, 2009). Be-cause we also know that at least two sauropod taxa were presentin the Kem Kem assemblage, any trophic network was likely morecomplex.

As shown above, the Kem Kem vertebrate compound assem-blage shows clear similarities with the Cenomanian Bahariyaassemblage in Egypt, the Wadi Milk assemblage in Sudan, and alsowith the Libyan assemblage of Draa Ubari (Nessov et al., 1998;Rage and Cappetta, 2002), suggesting that in the early Late Creta-ceous the continental faunas from western North Africa were notseparated by an epicontinental sea from those of eastern NorthAfrica. Dinosaurs from the Cenomanian part of the Echkar Forma-tion in Niger, however, are not the same as those from Sudanand from northern Africa (Morocco, Egypt). This might be causedeither by: (1) the small number of sampled taxa so far recordedfrom the Cenomanian of Niger; (2) variations related to differencesin the ecosystems; (3) a slightly different age for the Nigeran local-ity; or (4) latitudinal differences. The latter point, however, is un-likely since the Nigeran and the Egyptian localities were almostat the same latitude in the early Late Cretaceous. In addition, andas we have already noted, many of the vertebrate taxa from theKem Kem beds show palaeogeographical affinities with SouthAmerica, including the fish (Maisey, 2000; Cavin et al., 2001; Cavin,2008), pipid frogs (Rage and Dutheil, 2008), turtles (Tong, 2008)and dinosaurs. Among the Elasmobranchii, for instance, Triboduswas first described from the Santana Formation, Brazil, by Britoand Ferreira (1989), then the genus was tentatively reported fromCongo and Egypt (Maisey, 2000), and from Tunisia (Cuny et al.,2001). This clear vicariant pattern, however, was altered by the dis-covery of Tribodus in the Cenomanian of Charente, France (Vulloet al., 2005; Vullo and Néraudeau, 2008a). This French occurrence,together with the occurrence of scales of Stromerichthys (Vullo andNéraudeau, 2008b), may reflect either a vicariant event betweenEurope and Western Gondwana, or dispersal across Tethys. Clearvicariant patterns between taxa from South America and fromthe Kem Kem beds compound assemblage are observed at the spe-cies level: Mawsonia (cf. M. gigas and M. lavocati respectively),Calamopleurus (C. cylindricus and C. africanus respectively), Cladocy-clus (C. gardneri and C. pankowskii respectively), Araripesuchus (A.rattoides and A. gomesii); and at the genus level: euraxemydids(Euraxemys and Dirqadim respectively), podocnemidids (Brasilemysand Hamadachelys respectively), bothremydids (Cearachelys andGalianemys respectively) (Gaffney et al., 2006; Tong, 2008), anddiplodocids (Limaysaurus and Rebbachisaurus respectively) (Salgadoet al., 2004). Other vicariant patterns between both blocks ofWestern Gondwana are observed, although the sister-pairs of taxaon both sides of the split are not well identified. These include theDipnoi, which shows a clear vicariant event for the Recent lepido-sirenids, but the situation is unclear for the neoceratodontids (seeApesteguía et al., 2007; Cavin et al., 2007a; Cione et al., 2007 fordiscussions), as well as for the Cladistia and possibly the Gingly-modi (these two lineages were present in South America untilthe end of the Cretaceous; Gayet et al., 2002). Nevertheless, thesauropod and theropod assemblage of the Kem Kem beds resem-bles that known from Bahariya (Egypt) while recent work suggeststhat the theropod assemblage is similar to more or less coevalassemblages in Brazil (Medeiros, 2006), where spinosaurids andcarcharodontosaurids are also abundant. The pterosaur assem-blage from the Kem Kem beds appears to show a fairly high diver-sity, as do other pterosaur assemblages of similar or slightly oldergeological age (the Aptian Crato and ?Albian Santana Formations ofBrazil and the Albian Cambridge Greensand of the UK), but the


fragmentary nature of the material makes a precise appraisal ofthat diversity difficult.

3.2. The OT1 assemblage

Dutheil (1999a) recorded several fish taxa from a locality calledOT1, 10 km south of the village of Tafraout, preserved as articu-lated specimens in an elongate clay lens alongside plants, molluscs,insects and crustaceans. Although situated in the upper part of unitII of the Kem Kem beds, the Aoufous Formation as used here (i.e. inthe same sedimentary package as the Kem Kem beds compoundassemblage), the faunal composition of this lens is peculiar andwarrants further discussion.

3.2.1. Faunal listDutheil (1999a) described an indeterminate batoid from this

locality, the cladistian Serenoichthys kemkemensis (Dutheil,1999b) and two other, as yet unnamed, cladistians (Dutheil,2009), an indeterminate actinopterygian, Diplospondichthys mor-eaui (Filleul and Dutheil, 2004), and Spinocaudichthys oumtkouten-sis which is the oldest known freshwater acanthomorph (Filleuland Dutheil, 2001). Diplospondichthys in particular exhibits astrange combination of characters, especially diplospondylous ver-tebrae, which led Filleul and Dutheil (2004) to compare this specieswith various taxa among halecomorphs, basal neopterygians andAnguilliformes. Note also that Diplomystus sp., as well as an inde-terminate clupeomorph were figured in an early publication(Dutheil, 1999a) but are still pending description. Rage and Dutheil(2008) also reported the presence of amphibians and squamates atOT1, but because they were found in the clastic sediments situatedjust below and above the clay lens containing the fishes (Dutheil,personnal communication, 2009), they belong instead to the com-pound assemblage of the Kem Kem beds as defined here and notthe OT1 assemblage sensu stricto.

3.2.2. Age, palaeoenvironment and palaeogeographyThe OT1 locality is located within the Aoufous Formation, re-

garded here as being early Cenomanian in age. The environmentof this site has been interpreted as a quiet lake, or pool, with fastfossilisation occurring, evidenced by the preservation of musclesin most of the fish (Dutheil, 1999a). Because the placement of Dip-lospondichthys within actinopterygians is problematic, Spinocaud-ichthys has been hypothesised to occupy an indeterminateposition among basal acanthomorphs, indeed the unresolved posi-tions of both taxa are really due to an unexpected combination ofcharacters, it is possible that the fish assemblage of OT1 is highlyendemic.

3.3. The marine transgression or Neolobites bioevent

The contact between the Aoufous Formation and the overlyingcarbonate Akrabou Formation is generally hidden by masses of fall-en rocks. Where the contact is visible, for example to the east of thevillage of Douira, the Akrabou Formation lies conformably on top ofthe Aoufous Formation. The Akrabou Formation was studied in de-tail by Ettachfini and Andreu (2004) and Ettachfini (2008), who dis-cussed its lithostratigraphy, paleontology, paleogeography, andpaleoenvironments, and found evidence of four transgressive–regressive sequences on this open marine platform.

The Upper Cenomanian transgression is characterised in thisarea by carbonaceous facies, which have been studied using thinsections. Our preliminary study shows that during the Neolobitesbioevent, any benthos was very scarce, calcispheres are common,and planktonic foraminifers are abundant. These taxa are generallysimple, little evolved (no keel) and of very small morphotypes. Sec-tions measure around 70–100 (max 150) lm, whereas the species

described by most authors (e.g. the syntheses of Caron, 1985; Pre-moli Silva and Sliter, 1995; Premoli Silva and Verga, 2004; Roba-szynski and Caron, 1979) tend measure usually between 250 and600 lm. Eleven different types of section are distinguished andcontain small and simple (mostly rounded chambered) specimensthat are difficult to determine in thin section. Nevertheless, consid-ering the geometric evidence (i.e. allowing for some intraspecifictest variability), the regularity of the size of the sections and theconstancy of the wall type and diagenesis, we assume that all sec-tions illustrated in Fig. 7A–U are the result of sections takenthrough Hedbergella (Asterohedbergella) asterospinosa Hamaoui,1964. Sections of various primitive type adults, known to be oppor-tunistic species that rapidly colonise shallow, newly-opened areasduring transgressive periods (Hart and Bailey, 1979; Caron, 1983;Caron and Homewood, 1983; Hart, 1999), can however be homeo-morphous. For example, Hedbergella (Asterohedbergella) asterospin-osa is rarely illustrated (e.g. Abdallah and Meister, 1997; Ettachfini,2008; Ettachfini and Andreu, 2004; Ettachfini et al., 2005; Hama-oui, 1964) and does not appear in the syntheses on planktonicforaminiferas mentioned above. Often only the most characteristicsections (with a spine) are shown; various sections through Hed-bergella (Asterohedbergella) asterospinosa are figured in Fig. 7A–U.

In Morocco (i.e. the western High-Atlas, central High-Atlas,Middle-Atlas, and High Moulouya), Hedbergella (Asterohedbergella)asterospinosa is only known from the Upper Cenomanian; its lastoccurrence marks the base of the ‘imprecision interval’ of theCenomanian/Turonian boundary (Ettachfini, 2008).

The assemblage encountered here is different from the typicalRotalipora cushmani zone assemblage generally assigned by theNeolobites vibrayeanus ammonite found in the same layers (see be-low). The low diversity of plankton and the scarcity, or absence, ofassociated benthic foraminiferas suggests a peculiar environmentwith probable unfavourable bottom waters; oxygen and organicflux are of great ecological importance (Van der Zwaan et al.,1999). Surprisingly, macrobenthos like gastropods, bivalves andsea-urchins are rather abundant.

In the studied area, the first ammonite bioevent is characterisedby Neolobites vibrayeanus (d’Orbigny) associated with the nautilidsEutrephoceras and Angulithes (see Basse and Choubert, 1959; Mei-ster and Rhalmi, 2002) (Fig. 7). In the Southern part of our fieldarea, these taxa are often associated with a rich invertebrate mac-rofauna, including bivalves, gastropods and echinoids (Fig. 4E).This bioevent is short in time and corresponds precisely to the low-er part of the Upper Cenomanian (more precisely the lower part ofthe Calycoceras guerangeri Zone) and is recorded at Gara Sbâa, TiziMomrad, Taouz and Belkassem. This event clearly represents anearly transgressive trend in the area. Indeed, this Cenomanian Neo-lobites bioevent is closely related to the outset of the Upper Ceno-manian marine transgression which spreads on the North Africancraton. To the east, the sea reached the South of the Saharan re-gions [i.e. South Algeria (Amédro et al., 1996), South Tunisia (Mei-ster and Abdallah, 1996, 2005) as far as Damergou in Niger(Meister et al., 1992, 1994)]. To the west, in the ‘Sillon Préafricain’,this bioevent appears not to have crossed a NE–SW line that con-nects Tazougart (Belkassem) to Agoult (Gara Sbâa) and indicatesthe possible western border of the sea at this time. In North Africaand in the Middle East, Neolobites is closely related to shallow mar-ine environments and generally marks the beginning of the influ-ence of the Southern Tethys ocean.

3.4. The Agoult assemblage

Cavin and Dutheil (1999) provided a preliminary study of asmall fish assemblage, preserved as negatives on fine-grainedsandstone slabs, referred as ‘Daoura’. All their described fossilswere found about 100 km south of Erfoud, probably on the escarp-

Fig. 7. Macro- and microfossils from the Neolobites bioevent (base of the Akrabou Formation). (A–U) Various sections of Hedbergella (Asterohedbergella) asterospinosaHamaoui, 1964 from samples ap280, ap281, ap285 and ap305. Scale bars equal 25 lm (A–L) and 50 lm (M–U). For sample localities, see Fig. 3. (V and W) Neolobitesvibrayeanus (d’Orbigny); (X) Angulithes sp. Scale bar equals 100 mm (V–X).


ment of the Kem Kem plateau near the Oued (wadi) Daoura,although precise data on stratigraphical and geographical prove-nance are lacking. Murray et al. (2007) have reported on a resem-bling fish assemblage from southeastern Morocco, but indicatedifferences because the Daoura assemblage is characterised bysandstone preservation, while specimens from the new localityare preserved in carbonates. Murray and Wilson (2009) gave moreinformation about this second locality, near the village of Agoult,southeast of the town of Alnif.

In May 2008 some of us (A.P., C.M., L.B., L.C.) visited the localitymentioned by Murray and Wilson (2009) and found that it is lo-cated at the top of the Gara Sbâa hill, where Lavocat excavated

dinosaur remains in the 1950s. Indeed, dinosaur remains, as wellas other fossils typical of the compound assemblage from the theKem Kem beds (see above), occur in the Aoufous Formation atthe bottom of this cliff, while the Agoult fossil assemblage is pre-served in a series of finely laminated carbonate beds that formthe top of the hill. These fossiliferous beds are intensively exploitedby the local people; several quarries are worked along the borderof the gara (Fig. 4F). Within the carbonate sequence, intercalatebeds with a more siliceous matrix form nodule-like structures(Fig. 3G) and although a more precise sedimentological study isnecessary, we suspect that the fish specimens described by Cavinand Dutheil (1999) come from these siliceous beds, while the

Fig. 8. Fish from the Agoult assemblage. (A) pycnodontiform indet., (B) cf. Lusitanichthys africanus; (C and D) indeterminate teleosts.

Fig. 9. Microfossils from the base of the Akrabou Formation situated between the Neolobites and the Mammites bioevents. (A) Alveolinids indet (ap307); (B) Ovalveolina?(ap307); (C) Prealveolina sp. (ap307); (D) Biconcava bentori (ap282); (E) Praebulimina sp. (ap310); (F) Dicyclina sp. (ap307); (G) Nezzazata sp. (ap282); (H) Nezzazata isabellae(ap281); (I and J) Nummofallotia apula (ap305); (K) Pseudolituonella reicheli (ap282); (L) Pseudorhipidionina casertana (ap305); (M and N) Trochospira avnimelechi (ap282); (Oand P) Actinostomaria stellata (ap306); (Q and R) Likanella? sp. A, nom. prov. (ap286); (S) Genotella pfenderae (ap286); (T) Heteroporella lepina (ap305); (U) Boueina? sp.(ap305); (V) Girvanella sp. (ap305); (W) Neomeris? cf. N. circularis (ap310); (X) microfacies with vadose pisoids (ap264). Scale bars equal 50 lm (G–J and W), 100 lm (D, L–Nand Q), 200 lm (A–C, F, K, O and P, T–V, and X) and 250 lm (R and S). For localization of the samples, see Fig. 3.


specimens described by Murray et al. (2007) and Murray andWilson (2009) come from the carbonate beds. If this hypothesisis correct then Daoura, Agoult and Gara Sbâa all correspond tothe same locality. However, pending additional fieldwork we retain

here the name ‘Agoult’ to refer to this assemblage, geographicallymore justified than ‘Daoura’, and to avoid confusion with fossils re-corded from the Gara Sbâa locality, such as Rebbachisaurus garasbae,which belong to the compound assemblage of the Kem Kem beds.


3.4.1. Faunal listCavin and Dutheil (1999) reported the occurrence of a

Pycnodontidae indet., an otocephalian referrable to Clupavus afric-anus (Cavin, 1999a), a clupemorph referrable to aff. Spratticeps, an-other referrable to a paraclupeid indet., and Rhynchodercetis at theDaoura site. In addition, Murray et al. (2007) and Wilson et al.(2009) also mentioned two paraclupeids (one possibly correspond-ing to the aff. Spratticeps, although this taxon is regarded as slightlymore derived than the ellimmichthyiformes by Taverne, 1997), aRhynchodercetis, an euteleostean and an acanthomorph, as wellas a macrosemiid later assigned to Agoultichthys chattertoni byMurray and Wilson (2009). Visiting this locality in May 2008, wediscovered further specimens of macrosemiids, pycnodontiforms,clupavids, clupeomorphs, dercetids and possible tselfatiiforms(Fig. 8), together with plant remains and isopod crustaceans.

3.4.2. Age, palaeoenvironment and palaeogeographyDuring our 2008 fieldtrip, a specimen of the ammonite Neolo-

bites vibrayeanus was found at the base of the limestone sequencethat forms the top of the gara, about 25 m below the fossiliferousoutcrop. This indicates an age younger than early Late Cenomanianfor the Agoult assemblage. On the basis of faunal comparisons be-tween several ‘mid’-Cretaceous assemblages, in particular the JbelTselfat assemblage first described by Arambourg (1943, 1954), Ca-vin and Dutheil (1999) suggested a Late Cenomanian age for boththe Agoult (‘Daoura’) and the Jbel Tselfat assemblages. Rock sam-ples collected during the 2008 fieldtrip at the base of the AkrabouFormation, in a position located under the Mammites bioevent orGoulmima assemblage (see below) but above the Neolobites bio-event (i.e. in the sequence containing the Agoult assemblage), pro-vide new information about the palaeoenvironment of this timeinterval. Above the Neolobites bioevent and its specific planktonicmicrofauna, various benthic (and rare planktic) foraminiferas aswell as stromatoporoids, solenoporaceae, corals, algae, bryozoans,bivalves, ostracods, serpulids, calcispheres, gastropods and echino-derms can be seen to gradually colonise the shelf. The preliminarystudy of our thin sections shows, amongst the sections of benthicforaminiferas: alveolinids (Ovalveolina?, Prealveolina sp., Fig. 9A–C), Biconcava bentori (Fig. 9D), Ammobaculites sp., Biplanata pene-ropliformis, bolivinids, buliminids (Fig. 9E), Chrysalidina gradata,Cuneolina sp., Dicyclina schlumbergeri (Fig. 9F), discorbids, epistom-inids, Haplophragmoides sp., Ichnusella? sp., various miliolids,Montcharmontia appeninica, Nezzazata spp. (Fig. 9G and H), Nezzaz-atinella picardi, Nummofallotia apula (Fig. 9I and J), Peneroplis sp.,Pseudolituonella reicheli (Fig. 9K), Pseudorhapydionina dubia, Pseu-dorhipidionina casertana (Fig. 9L), Pseudotextulariella sp., Rotorbinel-la mesogeensis, Spiroloculina? sp., Spiroplectammina sp., Textulariaspp., Trochammina sp., and Trochospira avnimelechi (Fig. 9M andN). The stromatoporoid is attributed to Actinostomaria stellata(Fig. 9O and P) and the solenoporaceae to Parachaetetes asvapatii,while six species of algae (determination by M. Conrad) have beenattributed to Likanella? sp. A (Fig. 9Q and R, very similar to Likanellahammudai, pl. 4, Fig. 11 in Kuss, 1994), Genotella pfenderae (Konishiand Epis, 1962) Granier et al., 1991, emend. Granier and Berthou,2002 (Fig. 9S), Heteroporella lepina Praturlon 1966 (Fig. 9T), Bouein-a? sp. (Fig. 9U), Girvanella sp. (Fig. 9V), and Neomeris? cf. N. circu-laris Badve and Nayak, 1983 (Fig. 9W). The foraminiferalassociation indicates a shallow environment (complex foramini-fers, alveolinids, etc.) occurring within a restricted area (scarcityor absence of plankton and nekton) for most of the series, andthe associated muddy facies (mudstone, wackestone) suggests aquiet environment. Samples barren of fauna but containing vadosepisoids (Fig. 9X), indicative of probable emersive episodes, arenumerous in the El Begâa – Taouz section.

The described fish assemblage from Agoult shows affinitieswith Central Tethyan faunas, in particular the Jbel Tselfat assem-

blage (Cavin and Dutheil, 1999; Murray and Wilson, 2009), but alsowith Cenomanian assemblages from Slovenia (Cavin et al., 2000)and Lebanon (Forey et al., 2003). Nevertheless it is also very differ-ent from the Early Turonian fish assemblage from Goulmima,which shows mainly South Atlantic affinities (see below).

3.5. The Goulmima assemblage

In his explanations of the geological map of the High-Atlas, Du-bar (1949) reported the occurrence of calcareous nodules with fishfragments associated with an ammonite assemblage in the Turo-nian of the Midelt area. Farther west, in the Goulmima area, theequivalent fossiliferous beds are exploited for commercial pur-poses, mostly near the villages of Tadirhoust and Asfla, north ofthe city of Goulmima. Interesting fossiliferous exposures also lieabout 10 km east and north of Goulmima (Fig. 3I). Vertebrate re-mains are usually preserved in calcareous nodules, either com-pletely embedded in the nodules, or with part of the bodyenclosed. Large specimens, especially marine reptiles, are some-times preserved within several distinct nodules. Although fossilsfrom this assemblage can be chemically prepared by dissolvingthe matrix with formic acid, in many cases the nodule core is com-posed of siliceous material that prevents complete preparation ofspecimens. Ammonites are preserved either within nodules, or liefree in the marly beds (Fig. 4J).

3.5.1. Faunal listChondrichthyans are represented by at least two batoids and a

sclerorhynchid (Underwood et al., 2009). The other fishes includeindeterminate pycnodontiforms preserved either as isolated smallremains found loose in the matrix of nodules and as gut contents(Cavin, 1997a), or as large articulated individuals (under study byL.C.). Individual fish recorded thus far include the ichthyodectidIchthyodectes bardacki, the osmeroidid Osmeroides rheris, the ara-ripichthyid Araripichthys corytophorus (Cavin, 1997b), a small ench-odontid, Enchodus sp., sometimes preserved as gut contents (Cavin,1999b), and the pachyrhizodontid Goulmimichthys arambourgi (Ca-vin, 1995, 2001; Fig. 10C), the most common vertebrate from thisassemblage. Marine reptiles are represented by two polycotylidplesiosaurs, Thililua longicollis (Bardet et al., 2003a) and Manemer-gus anguirostris (Buchy et al., 2005), as well as the mosasauroidTethysaurus nopcsai (Bardet et al., 2003b), and indeterminate prot-ostegid turtles (under study by H.T.) (Fig. 10A and B).

3.5.2. Age, palaeoenvironment and palaeogeographyWithin the Akrabou Formation a second ammonite bioevent has

been defined which corresponds to the layer containing vertebrateremains, or the Goulmima assemblage (Cavin, 1999b). This bio-event is longer in time than the first Neolobites bioevent and is cor-related with the late Early Turonian, mainly with the Mammitesnodosoides Zone. The ammonite fauna is also quite diversified withMammites (Fig. 10D), Fagesia (Fig. 10E), Neoptychites, Choffaticeras,Nannovascoceras, and Hoplitoides (see Basse and Choubert, 1959;Kennedy et al., 2008) all present. The occurrence of Romaniceras(Yubariceras) in the upper part of the section also suggests a MiddleTuronian age. Ammonites have been recorded from the outcropslocated in the northwestern part of the studied area only (Goulm-ima and Asfla) but they have been previously reported in areas lo-cated to the north (Settat area) and to the east (Rekkame area)(Basse and Choubert, 1959). Thin sections sampled across the sec-ond ammonite bioevent show small and poorly diversified micro-fossil assemblages. Buliminids represent 95% of the foraminiferalfauna, always with a diagenetic micritised wall. The calcispheres(calcareous dinoflagellate cysts) collected have at least four differ-ent wall types, possibly corresponding to four species, while pel-lets, gastropods, small bivalves, and vertebrate fragments are also

Fig. 10. Macrofossils from the Goulmima assemblage (or Mammites bioevent), Akrabou Formation. (A and B), indeterminate protostegid turtle in dorsal (A) and left lateral(B) views; (C), nodule with fragment of the skull of a probable Goulmimichthys arambourgi; Mammites nodosoides (D) and Fagesia sp. (E). Scale bars equal 40 mm (A–C) and50 mm (D and E).


present. This poorly diversified association is representative of anenvironment with conditions not favourable for life, indeed thismight indicate a likely reason for the the death, and possibly forthe preservation, of the fishes and other vertebrates present be-cause of the lack of organisms responsible of the decay of car-casses. In contrast, the large numbers of small isolated remainsof juvenile Enchodus led Cavin (1999b) to suggest that this environ-ment of deposition may have been used as a nursery ground forthat species.

In contrast to the Cenomanian bioevent, the Turonian Mammites– Romaniceras (Yubariceras) bioevent is present everywhere in thesection except in the south (Kem Kem area) and in the south-eastalong a line that runs from Zelmou to Belkassem and Tarda. Thissuggests, at least in our area, a reduction of the influence of theopen sea toward the south. Indeed, the composition of the faunais classic for the shallow seas of the southern Tethys; ammonitesare quite ubiquitous and there is no evidence for Atlantic influence.The Goulmima fish assemblage shows biogeographical affinitieswith the northern South Atlantic (Cavin et al., 2001) but also with

the western Tethys as indicated by recent findings. Araripichthys,for instance, was first regarded as an endemic fish from Brazil (Sil-va Santos, 1985; Maisey and Blum, 1991), but its distribution isnow spread as far as Venezuela (Maisey and Moody, 2001) andMexico (Blanco and Cavin, 2003) while its stratigraphic distribu-tion extends from the Aptian–Albian (Brazil, Venezuela) to theTuronian (Morocco, Mexico). Similarly Goulmimichthys, first recog-nised based on specimens from Goulmima and previously regardedas restricted to that locality, now extends as far as Colombia (Para-mo Fonseca, personnal communication) and Mexico (Blanco andCavin, 2003). It is interesting to note that pachyrhizodontids,including Goulmimichthys, are a rather frequently encounteredfamily in the Early and Late Cretaceous, recorded in the SouthernAtlantic, Australia and in the Boreal domain – including the Euro-pean Chalk and northern part of the Western Interior Seaway(Cumbaa and Murray, 2008) – but are not yet known from the veryrich fish assemblages of the southeastern Tethys in Lebanon. Noteas well that the fish taxa from the compound Kem Kem vertebrateassemblage, as well as some of the tetrapod taxa and fish from the

Table 1Overview of the vertebrate taxa found in the four assemblages.

KK OT1 Agoult Goulmima

Euselachii Onchopristis numidus Indet.Batoid 1Batoid 2

Asteracanthus aegyptiacusDistobatus nutiaeTribodus sp.’Lissodus’ sp.Serratolamna amonensisCretoxyrhinidae indet.Marckgrafia lybica

Dipnoi Ceratodus humei‘‘Neoceratodus” africanus

Actinistia Mawsonia lavocaticf. Axelrodichthys

Actinopterygii indet. Stromerichthys aethiopicusDiplospondichthys moreaui

Cladistia Species 1 Serenoichthys kemkemensisSpecies 2 Species 3

Species 4Semionotiformes cf. Lepidotes

Oniichthys falipouiMacrosemiidae Agoultichthys chattertoniHalecomorphi Calamopleurus africanusPycnodontiformes Indet. Indet.Ichthyodectiformes Cladocyclus pankowskii Ichthyodectes bardackiOsteoglossomorpha Palaeonotopterus greenwoodi

Notopteridae indet.Elopomorpha Osmeroides rherisEuteleostei Euteleostei indet.Araripichthyidae Araripichthys corythophorusPachyrhizodontidae Goulmimichthys arambourgiOtocephala Lusitanichthys africanus

Erfoudichthys rosaeClupeomorpha Diplomystus sp.

Paraclupeidae indet.Clupeomorph indet. Aff. Spratticeps

Tselfatiiformes Concavotectum moroccensisAulopiformes Rhynchodercetis sp. Enchodus sp.Acanthomorpha indet. Spinocaudichthys oumtkoutensis

Indet.Sirenidae Kababisha sp.Pipidae Oumtkoutia anaeAnura Indet.Sauria Indet.Mosasauroidea Tethysaurus nopcsaiOphidia Simoliophis libycus

Madtsoiidae indet.Nigerophiidae indet.

Euraxemydidae Dirqadim schaefferiPodocnemididae Hamadachelys escuillieiBothremydidae Galianemys whitei

G. emringeriAraripemydidae Araripemys sp.Protostegidae Indet.Polycotylidae Thililua longicollis

Manemergus anguirostrisMesosuchian indet. Elosuchus cherifiensisNotosuchia Libycosuchus sp.

Araripesuchus rattoidesStomatosuchidae Laganosuchus maghrebensisTrematochampsidae Hamadasuchus rebouli

Indet.Rebbachisauridae Rebbachisaurus garasbaeLithostrotia Indet.Spinosauridae Spinosaurus cf. aegyptiacusCharcarodontosauridae Carcharodontosaurus saharicusAbelisauroidea cf. Majungasaurus

Deltadromeus agilisDromaeosauridae Indet.Pterosauria ?Pteranodontidae

Azhdarchidae indet.Tapejaridae indet.Coloborhynchus sp.

Aves Indet.


Fig. 11. Synthetic chart showing the succession of the ‘mid’-Cretaceous vertebrate assemblages in SE Morocco within a stratigraphic framework and withpalaeobiogeographical affinities (arrows).


Goulmima assemblage, have been considered to share close rela-tionships with taxa from the Santana Formation (Maisey, 2000;Cavin et al., 2001). This Brazilian assemblage is generally consideredAlbian in age, which casts doubt on close taxonomic compositionalsimilarities with northwest African faunas. An age not older than Al-bian (Pons et al., 1990), however, and even Cenomanian has beenproposed (Martill, 1993; Tong and Buffetaut, 1996) for the Santanaassemblage which may go some way to explaining the similaritieswe have discussed. We argue that affinities with the probably EarlyCenomanian Kem Kem assemblage and the well-dated Early Turo-nian Goulmima assemblage reinforces the argument for a Cenoma-nian age for the Brazilian Santana assemblage.

Finally, because diversification of basal mosasauroids occurredon the northern margin of the Mediterranean Tethys and in theWestern Interior Sea during the Late Cenomanian and Early/MiddleTuronian, the occurrence at Goulmima of Thililua is one of thesouthernmost occurrences known for this period (Bardet et al.,2008).

4. Discussion and conclusions

The succession of vertebrate assemblages known from south-eastern Morocco is one of the most diverse in the world from the‘mid’ Cretaceous. Assemblages contain all the main vertebrategroups from aerial, terrestrial, freshwater, and marine environ-ments; Table 1 show an overview of the vertebrate taxa found in

the four assemblages we have discussed and Fig. 11 shows the suc-cession of them in a stratigraphic framework. Of the known assem-blages, the compound fauna of the Kem Kem beds is by far themost diverse, but it is likely that it actually comprises several dis-tinct assemblages that we have been unable to distinguish in ourwork to date. Richest diversity seems to occur in the lower Ifezou-ane Formation, but the diversity of vertebrates from the AoufousFormation is still poorly known.

Our review demonstrates that the vertebrates from the com-pound Kem Kem assemblage are represented by taxa that encom-pass a wide range of sizes, from amphibians and squamates(represented by microfossils) up to very large saurischian dino-saurs (represented by isolated fragments, bones and partially artic-ulated skeletons). Apparent completeness of the fauna is on theone hand very favourable for reconstructing the Kem Kem palaeo-ecosystem, but it is important to take into account collecting bias,for instance the proportional differences observed in collections ofthe teeth of herbivorous and carnivorous dinosaurs (McGowan andDyke, 2009). The Kem Kem compound assemblage also contains aclear mix of terrestrial and aquatic components, the latter mainlyfreshwater in origin with a few brackish and marine fossils, butno marine ingressions have been recorded within this sequenceso far. The large size of individuals of several aquatic taxa, togetherwith the peculiar diet of some of them (e.g. planktivorous) indicatethe occurrence of large bodies of water. It is likely that a mosaic ofecosystems was probably present, but we have not been able todistinguish them so far. Indeed the OT1 assemblage may represent


an example of such a localised ecosystem in which fish seem to betaxonomically (even perhaps ecologically) distinct from those ofthe compound Kem Kem assemblage. The top of the series contain-ing this assemblage, the Aoufous Formation, was certainly depos-ited in an environment with abiotic conditions less favourable forvertebrate life, and with clues of extreme abiotic conditions(hypersalinity) and a great instability of the environment.

The ammonite Neolobites bioevent marks the beginning of a ma-jor marine transgression in the early Late Cenomanian. In our studyarea, however, this bioevent seems to occur in the eastern partonly, relative to the Southern Tethys. It is associated with a shallowmarine environment in a newly opened sea (transgressive phase),as indicated by abundant small and ‘primitive’ planktonic forami-niferas. Above the Neolobites bioevent, microfossils indicate a shal-low internal environment within a quiet restricted area, with anumber of possibly emersive episodes. The taxonomic compositionand the small size of the fishes from the Agoult assemblage alsocorrespond with interpretation of this quiet shallow marine pala-eoenvironment while at the same time pointing to strong palaoe-obigeographical affinities with the central and the southernmargin of Tethys ocean.

The Goulmima assemblage, isochronous with the Mammitesbioevent, is indicative of a deeper marine environment duringthe late Early Turonian. Although ammonites from this Mammitesbioevent show no clear palaeogeographical affinities, the verte-brates from the Goulmima assemblage show strong affinities withthe South Atlantic, the Western Tethys and even the Western Inte-rior Seaway in North America, which were already directlyconnected.

The succession of vertebrate assemblages described in this paperencompasses a time interval of probably less than 10 million years(from the base of the Cenomanian to the Lower Turonian). It istherefore interesting to note that few vertebrate lineages can be fol-lowed between assemblages, and is likely due to changing environ-mental conditions in North Africa at the time. This may indicatethat abiotic factors were the main driving force shaping the taxo-nomic composition of the assemblages. Indeed, the palaeobiogeo-graphical affinities of Moroccan fauna are strongly constrained bypaleoecological conditions, which permitted connections with thewest, the north and the east at different times. Deep changes inthe faunal composition of the vertebrate assemblages can also berelated to the peculiar global conditions of this time interval – highsea stand, high sea temperature – which may have favoured thebiological radiations of some lineages, in particular fish (Cavinet al., 2007b). Other groups, such as crocodiles and pterosaurs, alsohave high taxonomic diversity in the Kem Kem beds, but furtherstudy is necessary to test whether this effect corresponds to a gen-uine biological event or to a sampling bias.

Our overview further emphasises the huge potential of the ‘mid’Cretaceous fossil vertebrate record from Southeast Morocco forstudying past ecosystems and their evolution through time. In par-ticular, we show that assemblages, usually treated separately andeven regarded as having been accumulated in distinct basins –the ‘Sillon Préafricain’ and the Taouz Basin – actually formed in asingle sedimentary basin and therefore should be considered aspresenting a continuum in time. In conclusion, our review leadsto two issues that have to be addressed in future studies: (1) it willbe a challenge to correlate the southeastern and the northwesternparts of the Moroccan study area with either micro- or macrofaun-as, keeping in mind that we have no palaeontological and sedimen-tological evidence of a complete profile covering the whole LateCenomanian and Early Turonian and (2) systematic prospectingacross the whole area, combined with systematic excavations inparticularly interesting spots, must be conducted in order to pro-vide a better understanding of lateral variations between assem-blages and the total composition of single sites.

Acknowledgments

We thank Pierre-Alain Proz (Museum, Geneva) for making thethin sections, Prof. Roland Wernli (University of Geneva) and Dr.Marc A. Conrad (Geneva) for their help in the identification ofmicrofossils, and the team of volunteers from SESNE Society of El-beuf for assistance with excavations during the 2008 fieldtrip(Daniel Aubry, Laurent Gauthier, Perrine Luquet, Fabien Bricongne,Guy Godefroy). Dr. Oliver W.M. Rauhut and Dr. Adriana López-Arbarello (München) are thanked for their thorough reviews of thispaper; the late Dr. René Lavocat kindly provided photographs of hisfieldwork at Gara Sbâa (Fig. 4A).

References

Abdallah, H., Meister, C., 1997. The Cenomanian–Turonian boundary in the Gafsa-Chott area (southern part of central Tunisia): biostratigraphy,palaeoenvironments. Cretaceous Research 18, 197–236.

Amédro, F., Busson, G., Cornée, A., 1996. Révision des Ammonites du Cénomaniensupérieur et du Turonien inférieur du Tinrhert (Sahara Algérien), implicationsbiostratigraphiques. Bulletin du Muséum d’Histoire naturelle, 4e Série, 18-Section C, 2–3, 179–232.

Amiot, R., Buffetaut, E., Tong, H., Boudad, L., Kabiri, L., 2004. Isolated theropod teethfrom the Cenomanian of Morocco and their palaebiogeographical significance.Revue de Paléobiologie 9, 143–149.

Apesteguía, S., Agnolín, F., Claeson, K., 2007. Review of Cretaceous dipnoans fromArgentina (Sarcopterygii, Dipnoi) with descriptions of new species. Revista delMuseo Argentino de Ciencias Naturales 9 (1), 27–40.

Arambourg, C., 1943. Notes préliminaire sur quelques Poissons fossiles nouveaux.Bulletin de la Société Géologique de France, 5e Série XIII, 281–288.

Arambourg, C., 1954. Les poissons crétacés du Jebel Tselfat. Notes et Mémoires duService Géologique du Maroc 118. 188 p.

Bardet, N., Pereda Suberbiola, X., Jalil, N.-E., 2003a. A new polycotylid plesiosaurfrom the Late Cretaceous (Turonian) of Morocco. Comptes Rendus Palevol 2,307–315.

Bardet, N., Pereda Suberbiola, X., Jalil, N.-E., 2003b. A new mosasauroid (Squamata)from the Late Cretaceous (Turonian) of Morocco. Comptes Rendus Palevol 2,607–616.

Bardet, N., Houssay, A., Rage, J.-C., Pereda Suberbiola, X., 2008. The Cenomanian–Turonian (late Cretaceous) radiation of marine squamates (Reptilia), the role ofthe Mediterranean Tethys. Bulletin de la Société Géologique de France 179,605–622.

Barrett, P.M., Butler, R.J., Edwards, N.P., Milner, A.R., 2008. Pterosaur distributionin time and space: an atlas. In: Hone, D.W.E., Buffetaut, E. (Eds.),Flugsaurier: Pterosaur Papers in Honour of Peter Wellnhofer. ZittelianaB28, pp. 61–107.

Basse, E., Choubert, G., 1959. Les faunes d’ammonites du «Cénomanien–Turonien»de la partie orientale du domaine atlasique marocain et de ses annexessahariennes. In: 20ème Congrès Géologique International de Mexico, 1956,Symposium Crétacé 2, pp. 58–82.

Benton, M., Bouaziz, S., Buffetaut, E., Martill, D.M., Ouaja, M., Soussi, M., Trueman, C.,2000. Dinosaurs and other fossil vertebrates from fluvial deposits in the LowerCretaceous of southern Tunisia. Palaeogeography, Palaeoclimatology,Palaeoecology 157, 227–246.

Blanco, A., Cavin, L., 2003. New teleostei from the Agua Nueava Formation(Turonian), Vallecillo, NE Mexico. Comptes Rendus Palevol 2, 299–306.

Bouaziz, S., Buffetaut, E., Ghanmi, M., Jaeger, J.J., Martin, M., Mazin, J.M., Tong, H.,1988. Nouvelles découvertes de vertébrés fossiles dans l’Albien du Sud tunisien.Bulletin de la Société Géologique de France 4 (2), 335–339.

Brito, P.M., Ferreira, P.L.N., 1989. The first hybodont shark, Tribodus limae n.g, n.s.p,from the Lower Cretaceous of Chapada do Araripe (North-East Brazil). Anais daAcademia Brasileira de Ciências 61, 53–57.

Brito, P.M., Dutheil, D., Meunier, F., 2009. Presence of Notopteridae (Teleostei:Osteoglossomorpha) in the Kem Kem beds, Cretaceous of Morocco. In: Jalil, N.-E.(Ed.), 1st International Congress on North African Vertebrate Palaeontology,Program & Abstracts, Marrakech, 25–27 May 2009, p. 66.

Broin, F.L.de, 2002. Elosuchus, a new genus of crocodile from the Lower Cretaceousof the North of Africa. Comptes Rendus Palevol 1, 275–285.

Brusatte, S.L., Sereno, P.C., 2007. A new species of Carcharodontosaurus (Dinosauria,Theropoda) from the Cenomanian of Niger and a revision of the genus. Journalof Vertebrate Paleontology 27, 902–916.

Buchy, M.-C., Metayer, F., Frey, E., 2005. Osteology of Manemergus anguirostris n.gen. et sp., a new plesiosaur (Reptilia, Sauropterygia) from the UpperCretaceous of Morocco. Palaeontographica, Abteilung A 272, 97–120.

Buffetaut, E., 1976. Der Land-Krokodilier Libycosuchus Stromer und die FamilieLibycosuchidae (Crocodylia, Mesosuchia) aus der Kreide Afrikas. Mitteilungender Bayerischen Staatssammlung für Paläontologie und Historische Geologie16, 17–28.

Buffetaut, E., 1988. The ziphodont mesosuchian crocodile from Messel: areassessment. Courier Forschungsinstitut Senckenberg 107, 211–221.


Buffetaut, E., 1989. New remains of the enigmatic dinosaur Spinosaurus from theCretaceous of Morocco and the affinities between Spinosaurus and Baryonyx.Neues Jahrbuch für Geologie und Paläontologie, Monatshefte 2, 79–87.

Buffetaut, E., 1994. A new crocodilian from the Cretaceous of southern Morocco.Comptes Rendus de l’Académie des Sciences, Paris, Série II 319, 1563–1568.

Buffetaut, E., 2001. On the age of the mid-Cretaceous vertebrate-bearing Formationsof North Africa. In: Ferre, B., Fouray, M., Tabouelle, J. (Eds.), Colloquium on theCenomanian Stage. Bulletin de la Société d’études des Sciences Naturelles duMuséum d’Elbeuf, 2001, pp. 20–22.

Buffetaut, E., Ouaja, M., 2002. A new specimen of Spinosaurus (Dinosauria,Theropoda) from the Lower Cretaceous of Tunisia, with remarks on theevolutionary history of the Spinosauridae. Bulletin de la Société Géologiquede France 173 (5), 415–421.

Buffetaut, E., Taquet, P., 1979. An early Cretaceous terrestrial crocodilian and theopening of the South Atlantic. Nature 280, 486–487.

Buffetaut, E., Bussert, R., Brinkmann, W., 1990. A new nonmarine vertebrate fauna inthe Upper Cretaceous of northern Sudan. Berliner geowissenschaftlicheAbhandlungen (A) 120 (1), 183–202.

Caron, M., 1983. La spéciation chez les Foraminifères planctiques, une réponseadaptée aux contraintes de l’environnement. Zitteliana 10, 671–676.

Caron, M., 1985. Cretaceous planktic foraminifera. In: Bolli, H.M., Sauders, J.B.,Perch-Nielsen, K. (Eds.), Plankton Stratigraphy. Cambridge University Press,Cambridge, pp. 17–86.

Caron, M., Homewood, P., 1983. Evolution of early planktic foraminifers. MarineMicropaleontology 7, 453–462.

Carrano, M.T., Sampson, S.D., 2008. The phylogeny of Ceratosauria (Dinosauria:Theropoda). Journal of Systematic Palaeontology 6, 183–236.

Catuneanu, O., Khalifa, M.A., Wanas, H.A., 2006. Sequence stratigraphy of the LowerCenomanian Bahariya Formation, Bahariya Oasis, Western Desert. EgyptSedimentary Geology 190, 121–137.

Cavin, L., 1995. Goulmimichthys arambourgi n. g., n. sp., un Pachyrhizodontidae(Actinopterygii, Teleostei) d’une nouvelle localité à nodules fossilifères duTuronien inférieur marocain. Comptes Rendus de l’Académie des Sciences, Paris,Série IIa 321, 1049–1054.

Cavin, L., 1997a. Supposed and direct evidence of trophic relationships within thefish community of the Lower Turonian of Goulmima, Morocco. GeologicalSociety of Denmark, On Line Series, 1. <http://www.2dgf.dk/Publikationer/DGF_On_Line/Vol_1/lionelca.htm>.

Cavin, L., 1997b. Nouveaux Teleostei du gisement du Turonien inférieur deGoulmima (Maroc). Comptes Rendus de l’Académie des Sciences, Paris, SérieIIa 325 (7), 19–24.

Cavin, L., 1999a. A new Clupavidae (Teleostei, Ostariophysi) from the Cenomanianof Daoura (Morocco). Comptes Rendus de l’Académie des Sciences, Paris, SérieIIa 329, 689–695.

Cavin, L., 1999b. Occurrence of a juvenile teleost, Enchodus sp., in a fish gut contentfrom the Upper Cretaceous of Goulmima, Morocco. Special Paper inPalaeontology 60, 57–72.

Cavin, L., 2001. Osteology and phylogenetic relationships of the teleostGoulmimichthys arambourgi Cavin, 1995 from the Upper Cretaceous ofGoulmima, Morocco. Eclogae Geologicae Helvetiae 133, 25–52.

Cavin, L., 2008. Palaeobiogeography of Cretaceous Bony Fishes (Actinistia, Dipnoiand Actinopterygii). In: Cavin, L., Longbottom, A., Richter, M. (Eds.), Fishes andthe breakup of Pangaea. Geological Society of London, pp. 165–183 (SpecialPublication 295).

Cavin, L., Brito, P.M., 2001. A new Lepisosteidae (Actinopterygii, Ginglymodi) fromthe Cretaceous of the Kem Kem beds, Southern Morocco. Bulletin de la SociétéGéologique de France 172 (5), 141–150.

Cavin, L., Dutheil, D.B., 1999. A new Cenomanian ichthyofauna from southeasternMorocco and its relationships with other early Late Cretaceous Moroccanfaunas. Geologie en Mijnbouw 78, 261–266.

Cavin, L., Forey, P.L., 2001. Osteology and systematic affinities of Palaeonotopterusgreenwoodi Forey, 1997. Zoological Journal of the Linnean Society 133, 25–52.

Cavin, L., Forey, P.L., 2004. New mawsoniid coelacanth (Sarcopterygii, Actinistia)remains from the Cretaceous of the Kem Kem beds, SE Morocco. In: Tintori, A.,Arratia, G. (Eds.), Mesozoic Fishes III – Systematics, Paleoenvironments andBiodiversity. Dr Pfeil Verlag, pp. 493–506.

Cavin, L., Forey, P.L., 2008. A new tselfatiiform teleost from the mid-Cretaceous(Cenomanian) of the Kem Kem beds, Southern Morocco. In: Arratia, G., Schultze,H-P., Wilson, M.V.H. (Eds.), Mesozoic Fishes IV – Homology and Phylogeny. DrPfeil Verlag, pp. 199–216.

Cavin, L., Jurkovšek, B., Kolar-Jurkovšek, T., 2000. Succession of fish assemblages inthe Upper Cretaceous of the Trieste-Komen Plateau, Slovenia. Geologija 43 (2),165–195.

Cavin, L., Boudad, L., Duffaud, S., Kabiri, L., Le Lœuff, J., Rouget, I., Tong, H., 2001.L’évolution paléoenvironnementale des faunes de poissons du Crétacésupérieur du bassin de Tafilalt et des régions avoisinantes (Sud-Est duMaroc), implications paléobiogéographiques. Comptes Rendus de l’Académiedes Sciences, Paris, Série IIa 333, 677–683.

Cavin, L., Suteethorn, V., Buffetaut, E., Tong, H., 2007a. A new Thai Mesozoic lungfish(Sarcopterygii, Dipnoi) with an insight into post-Palaeozoic dipnoan evolution.Zoological Journal of the Linnean Society 149, 141–177.

Cavin, L., Forey, P.L., Lécuyer, C., 2007b. Correlation between environment and LateMesozoic ray-finned fish evolution. Palaeogeography, Palaeoclimatology,Palaeoecology 245, 353–367.

Cavin, L., Deesri, U., Suteethorn, V., 2009. The Jurassic and Cretaceous bony fishrecord (Actinopterygii, Dipnoi) from Thailand. In: Buffetaut, E., Cuny, G., Le

Loeuff, J. (Eds.), Late Palaeozoic and Mesozoic Continental Ecosystems of SEAsia. Geological Society of London, pp. 123–137 (Special Publication 315).

Charrière, A., Dépêche, F., Feist, M., Grambast-Fessard, N., Jaffrezo, M., Peybernès, B.,Ramalho, M., 1994. Microfaunes, microflores et paléoenvironnements successifsdans la Formation d’El Mers (Bathonien–?Callovien) du synclinal de Skoura(Moyen-Atlas, Maroc). Geobios 27, 157–174.

Charrière, A., Haddoumi, H., Mojon, P.O., 2005. Découverte de Jurassique supérieuret d’un niveau marin du Barrémien dans les «couches rouges» continentales duHaut Atlas central marocain: implications paléogéographiques et structurales.Comptes Rendus Palévol 4, 385–394.

Choubert, G., 1948. Essai sur la paléogéographie du Mésocrétacé marocain. VolumeJubilaire de la Société des Sciences Naturelles du Maroc, 307–329.

Choubert, G., Clariond, L., Hindermeyer, J., 1952. Livret-guide de l’excursion C 36.Anti-Atlas central et oriental. Congrès Géologique International, XIXe Session(Alger 1952), Série: Maroc, No. 11. Rabat, 89 p.

Churcher, C.S., De Iuliis, G., 2001. A new species of Protopterus and a revision ofCeratodus humei (Dipnoi: Ceratodontiformes) from the Mut Formation ofeastern Dakhleh Oasis, Western Desert of Egypt. Journal of Palaeontology 44,305–323.

Churcher, C.S., De Iuliis, G., Kleindienst, M.R., 2006. A new genus for the dipnoanspecies Ceratodus tuberculatus Tabaste, 1963. Geodiversitas 28 (4), 635–647.

Cione, A.L., Gouiric, S., Goin, F., Poiré, D., 2007. Atlantoceratodus, a new genus oflungfish from the upper Cretaceous of South America and Africa. Revista MuseoLa Plata, Paleontología 10 (62), 1–12.

Cumbaa, S.L., Murray, A.M., 2008. New Late Cretaceous pachyrhizodontid andenchodontid fishes and associated ichthyofauna from the Northwest Territories,Canada. In: Arratia, G., Schultze, H.-P., Wilson, M.V.H. (Eds.), Mesozoic Fishes 4 –Homology and Phylogeny. Verlag Dr. Friedrich Pfeil, München, Germany, pp.229–256.

Cuny, G., Suteethorn, V., Buffetaut, E., Ouaja, M., 2001. Freshwater hybodont sharksin the Aptian–Albian of Tunisia and Thailand. In: Ferre, B., Fouray, M., TabouelleJ. (Eds.), Colloquium on the Cenomanian Stage. Bulletin de la Société d’étude desSciences Naturelles d’Elbeuf, pp. 27–29.

Dal Sasso, C., Maganuco, S., Buffetaut, E., Mendez, M.A., 2005. New information onthe skull of the enigmatic theropod Spinosaurus, with remarks on its size andaffinities. Journal of Vertebrate Paleontology 25 (4), 888–896.

Depéret, C., Savornin, J., 1927. La faune des reptiles et de poissons albiens deTimimoun (Sahara algérien). Bulletin de la Société Géologique de France 27,257–265.

Dubar, G., 1949. Carte géologique provisoire du Haut Atlas de Midelt, échelle au 1/200 000e. Notice explicative, Notes et Mémoires du Service Géologique duMaroc, 59bis, 60 p.

Dutheil, D.B., 1999a. An overview of the freshwater fish fauna from the Kem Kembeds (Late Cretaceous, Cenomanian) of southeastern Morocco. In: MesozoicFishes 2 – Systematics and Fossil Record. Friedrich Pfeil, München, pp. 553–563.

Dutheil, D.B., 1999b. The first articulated fossil cladistian, Serenoichthyskemkemensis, gen. et sp. nov., from the Cretaceous of Morocco. Journal ofVertebrate Paleontology 19 (2), 243–246.

Dutheil, D.B., 2009. Two new short-bodies Cladistia (Actinopterii) from the KemKem beds (Cenomanian of Morocco). In: Jalil, N.-E. (Ed.), 1st InternationalCongress on North African Vertebrate Palaeontology, Program & Abstracts,Marrakech, 25–27 May 2009, p. 31.

Dutheil, D.B., Brito, P.M., 2009. Articulated cranium of Onchopristis numidus(Sclerorhynchidae, Elasmobranchii) from the kem Kem beds, Morocco. In:Jalil, N.-E. (Ed.), 1st International Congress on North African VertebratePalaeontology, Program & Abstracts, Marrakech, 25–27 May 2009, p. 66.

Ettachfini, M., 2008. La transgression au passage Cénomanien–Turonien sur ledomaine atlasique marocain. Strata Série 2. 225 p.

Ettachfini, M., Andreu, B., 2004. Le Cénomanien et le Turonien de la plate-formePréafricaine du Maroc. Cretaceous Research 25, 277–302.

Ettachfini, E.M., Souhel, A., Andreu, B., Caron, M., 2005. La limite Cénomanien–Turonien dans le Haut Atlas central, Maroc. Geobios 38, 57–68.

Fabre, J., 2005. Géologie du Sahara occidental et central. Tervuren AfricanGeoscience Collection 108. 572.

Ferrandini, M., Philip, J., Babinot, J.-F., Ferrandini, J., Tronchetti, G., 1985. La plate-forme carbonatée du Cénomano–Turonien de la région d’Erfoud-Errachidia(Sud-Est marocain), stratigraphie et paléoenvironnements. Bulletin de laSociété Géologique de France I (8), 559–564.

Filleul, A., Dutheil, D.B., 2001. Spinocaudichthys oumtkoutensis, a freshwateracanthomorph from the Cenomanian of Morocco. Journal of VertebratePaleontology 21 (4), 774–780.

Filleul, A., Dutheil, D.B., 2004. A peculiar diplospondylous actinopterygian fish fromthe Cretaceous of Morocco. Journal of Vertebrate Paleontology 24 (2), 290–298.

Fluegel, E., 2004. Microfacies of Carbonate Rocks. Analysis, Interpretation andApplication. Springer Verlag. 976 p.

Forey, P.L., 1997. A Cretaceous notopterid (Pisces, Osteoglossomorpha) fromMorocco. South African Journal of Science 93, 564–569.

Forey, P.L., Cavin, L., 2007. A new species of Cladocyclus (Teleostei, Ichthyo-dectiformes) from the Cenomanian of Morocco. Palaeontologia Electronica 10(3), 10 p. <http://palaeo-electronica.org/paleo/2007_3/133/index.html>.

Forey, P.L., Grande, L., 1998. An African twin to the Brazilian Calamopleurus(Actinopterygii, Amiidae). Zoological Journal of the Linnean Society 123, 179–195.

Forey, P.L., Lu, Y., Patterson, C., Davies, C.E., 2003. Fossil fishes from the Cenomanian(Upper Cretaceous) of Namoura, Lebanon. Journal of Systematic Palaeontology 1(4), 227–330.

http://www.2dgf.dk/Publikationer/DGF_On_Line/Vol_1/lionelca.htm

http://www.2dgf.dk/Publikationer/DGF_On_Line/Vol_1/lionelca.htm

http://palaeo-electronica.org/paleo/2007_3/133/index.html


Frizon de Lamotte, D., Leturmy, P., Missenard, Y., Sami Khomsi, S., Geoffrey Ruiz, G.,Saddiqi, O., Guillocheau, F., Michard, A., 2009. Mesozoic and Cenozoic verticalmovements in the Atlas system (Algeria, Morocco, Tunisia): an overview.Tectonophysics 475, 9–28.

Furon, R., Lombard, J., 1964. Geological Map of Africa: 1/500 000; Explanatory Note.Natural Resources Research, vol. 3. UNESCO, Paris, 39 p.

Gaffney, E.S., Tong, H., Meylan, P.A., 2002. Galianemys, a new side-necked turtle(Pelomelusoides, Bothremydidae) from the Late Cretaceous of Morocco.American Museum Novitates 3379. 20 p.

Gaffney, E.S., Tong, H., Meylan, P.A., 2006. Evolution of the side-necked turtles, thefamilies Bothremydidae, Euraxemydidae, and Araripemydidae. Bulletin of theAmerican Museum of Natural History 300. 698 p.

Gale, A.S., 2000. The Cretaceous world. In: Culver, S.J., Rawson, P.F. (Eds.), BioticResponse to Global Change: The Last 145 Million Years. Cambridge UniversityPress, Cambridge, pp. 4–19.

Gayet, M., Meunier, F., Werner, C., 2002. Diversification in Polypteriformes andspecial comparison with the Lepisosteiformes. Palaeontology 45 (2), 361–376.

Gmira, S., 1995. Étude des chéloniens fossiles du Maroc. Cahiers de Paléontologie,140 p.

Grande, L., Bemis, W.E., 1991. Osteology and phylogenetic relationships of fossil andrecent paddlefishes (Polyodontidae) with comments on the interrelationshipsof Acipenseriformes. Society of Vertebrate Paleontology, Supplement to Journalof Vertebrate Paleontology 11 (1). 121 p.

Hamaoui, M., 1964. On a new subgenus of Hedbergella (Foraminiferida). IsraelJournal of Earth-Sciences 13, 133–142.

Hart, M.B., 1999. The evolution and biodiversity of Cretaceous planktonicForaminiferida. Geobios 32 (2), 247–255.

Hart, M.B., Bailey, H.W., 1979. The distribution of planktonic Foraminiferida in themid-Cretaceous of N.W. Europe. Aspekte der Kreide Europas, IUGS Series A (6),527–542.

Holtz, T.R., Chapman, R.E., Lamanna, M.C., 2004. Mesozoic biogeography ofDinosauria. In: Weishampel, D.B., Dodson, P., Osmólska, H. (Eds.), TheDinosauria, second ed. University of California Press, Berkeley, pp. 627–642.

Janensch, W., 1925. Die Coelurosaurier und Theropoden der Tendaguru-SchichtenDeutsch-Ostafrikas. Palaeontographica 1 (Suppl. 7), 1–99.

Kellner, A.W.A., 2009. Brief review of the pterosaur record in Africa. In: Jalil, N.-E.(Ed.), 1st International Congress on North African Vertebrate Palaeontology,Program & Abstracts, Marrakech, 25–27 May 2009, p. 34.

Kennedy, W.J., Gale, A.S., Ward, D.J., Underwood, C.J., 2008. Lower Turonianammonites from Goulmima, southern Morocco. Bulletin de l’Institut Royaldes Sciences de Belgique, Sciences de la Terre 78, 149–177.

Kilian, C., 1931. Des principaux complexes continentaux du Sahara. ComptesRendus Sommaires de la Société Géologique de France 9, 109–111.

Kuss, J., 1994. Cretaceous (Albian–Turonian) Calcareous algae from Egypt andJordan – systematics, stratigraphy and paleogeography. Abhandlungen derGeologischen Bundes Anstalt 50, 295–317.

Lapparent, A.F.de, 1960. Les Dinosauriens du ‘Continental Intercalaire’ du Saharacentral. Mémoire de la Société Géologique de France (N.S.) 88A, 1–57.

Larsson, H.C.E., Sidor, C.A., 1999. Unusual crocodyliform teeth from the LateCretaceous (Cenomanian) of Southern Morocco. Journal of VertebratePaleontology 19 (2), 398–401.

Larsson, H.C.E., Sues, H.-D., 2007. Cranial osteology and phylogenetic relationshipsof Hamadasuchus rebouli (Crocodyliformes, Mesoeucrocodylia) from theCretaceous of Morocco. Zoological Journal of the Linnean Society 149 (4),533–567.

Lavocat, R., 1948. Découverte de Crétacé à vertébrés dans le soubassement del’Hammada du Guir (Sud marocain). Comptes Rendus de l’Académie desSciences 226, 1291–1292.

Lavocat, R., 1949. Les gisements de vertébrés crétacés du Sud marocain. ComptesRendus Sommaires de la Société Géologique de France 19 (5), 125–126.

Lavocat, R., 1951. Découverte de restes d’un grand Dinosaurien sauropode dans leCrétacé du Sud marocain. Comptes Rendus de l’Académie des Sciences 232,169–170.

Lavocat, R., 1954a. Sur les dinosauriens du Continental Intercalaire des Kem-Kem dela Daoura. Comptes Rendus, 19th International Geological Congress 1, 65–68.

Lavocat, R., 1954b. Reconnaissance géologique dans les Hammadas des confinsalgéro-marocains du Sud. Notes et Mémoires du Service Géologique du Maroc116. 147 p.

Lavocat, R., 1955a. Titres et travaux scientifiques. Imprimerie Centrale, Romorantin.95 p.

Lavocat, R., 1955b. Découverte d’un Crocodilien du genre Thoracosaurus dans leCrétacé Supérieur d’Afrique. Bulletin du Muséum National d’Histoire Naturelle,Paris 2 (27), 338–340.

Mader, B.J., Kellner, A.W.A., 1999. A new Anhanguerid pterosaur from theCretaceous of Morocco. Boletim do Museu Nacional (Nova Serie), Geologia 45,1–11.

Mahler, L., 2005. Record of Abelisauridae (Dinosauria, Theropoda) from theCenomanian of Morocco. Journal of Vertebrate Paleontology 25, 236–239.

Maisey, J.G., 2000. Continental break up and the distribution of fishes ofWestern Gondwana during the Early Cretaceous. Cretaceous Research 21,281–314.

Maisey, J.G., Blum, S., 1991. Araripichthys. In: Maisey, J.G. (Ed.), Santana Fossils: AnIllustrated Atlas. TFH Publications, Neptune, NJ, pp. 208–217.

Maisey, J.G., Moody, J.M., 2001. A review of the problematic teleost fishAraripichthys, with a description of a new species from the Lower Cretaceousof Venezuela. American Museum Novitates 3324, 1–27.

Martill, D.M., 1993. Fossils of the Santana and Crato Formations, Brazil.Palaeontological Association Field Guides to Fossils 5. 159 p.

Martin, M., 1982. Nouvelles données sur la phylogénie et la systématique desDipneustes postpaléozoïques. Comptes Rendus de l’Académie des Sciences,Paris, Série II 294, 611–614.

Martin, M., 1984a. Deux Lepidosirenidae (Dipnoi) crétacés du Sahara, Protopterushumei (Priem) et Protopterus protopteroides (Tabaste). PaläontologischeZeitschrift 58, 265–277.

Martin, M., 1984b. Révision des Arganodontidés et des Néocératodontidés (Dipnoi,Ceratodontiformes) du Crétacé africain. Neues Jahrbuch für Geologie undPaläontologie, Abhandlungen 169 (2), 225–260.

McGowan, A.J., Dyke, G.J., 2009. A surfeit of theropods in the Moroccan LateCretaceous? Comparing diversity estimates from field data and fossil shops.Geology 37, 843–846.

Medeiros, M.A., 2006. Large theropod teeth from the Eocenomanian of northeasternBrazil and the occurrence of Spinosauridae. Revista Brasileira de Paleontologia9, 333–338.

Meister, C., Abdallah, H., 1996. Les ammonites du Cénomanien supérieur et duTuronien inférieur de la région de Gafsa-Chotts, Tunisie du centre-sud. Geobios,Supplément au Fascicule 29 (5), 3–49.

Meister, C., Abdallah, H., 2005. Précision sur les successions d’ammonites duCénomanien – Turonien dans la région de Gafsa, Tunisie du centre-sud. Revuede Paléobiologie 24 (1), 111–199.

Meister, C., Rhalmi, M., 2002. Quelques ammonites du Cénomanien–Turonien de larégion d’Errachidia-Boudnid-Erfoud (partie méridionale du Haut Atlas Central,Maroc). Revue de Paléobiologie 21 (2), 759–779.

Meister, C., Alzouma, K., Lang, J., Mathey, B., 1992. Les ammonites du Niger (Afriqueoccidentale) et la transgression transsaharienne au cours du Cénomanien–Turonien. Geobios 25 (1), 55–100.

Meister, C., Alzouma, K., Lang, J., Mathey, B., Pascal, A., 1994. Nouvelles données surles ammonites du Niger oriental (Ténéré, Afrique occidentale) dans le cadre dela transgression du Cénomanien–Turonien. Geobios 27, 189–219.

Michard, A., Frizon de Lamotte, D., Saddiqi, O., Chalouan, A., 2008. An outline of thegeology of Morocco. In: Michard, A., Saddiqi, O., Chalouan, A., Frizon de Lamotte,D. (Eds.), Continental Evolution: The Geology of Morocco. Springer Verlag,Berlin, Heidelberg, pp. 1–32.

Milner, A.C., 2003. Fish-eating theropods, a short review of the systematics, biologyand palaeobiogeography of spinosaurs. In: Huerta Hurtado, P., TorcidaFernandez-Baldor, F. (Eds.), Actas de las II Jornadas Internacionales sobrePaleontologıa de Dinosaurios y su Entorno (2001). Salas de Los Infantes,Colectivo Arqueologico–Paleontologico de Salas, pp. 129–138.

Moody, R.T.J., Sutcliffe, P.J.C., 1991. The Cretaceous deposits of the Iullemme-den Basin of Niger, central West Africa. Cretaceous Research 12, 137–157.

Murray, A.M., Wilson, M.V.H., 2009. A new Late Cretaceous Macrosemiid fish(Neopterygii, Halecostomi) from Morocco, with temporal and geographicalrange extensions for the family. Palaeontology 52 (2), 429–440.

Murray, A., Wilson, M., Chatterton, B., 2007. A Late Cretaceous actinopterygianfauna from Morocco. Journal of Vertebrate Paleontology 27 (3), 122A.

Nessov, L.A., Zhegallo, V.I., Averianov, A.O., 1998. A new locality of Late Cretaceoussnakes, mammals and other vertebrates in Africa (western Libya). Annales dePaléontologie, Paris 84, 265–274.

Novas, F.E., de Valais, S., Vickers-Rich, P., Rich, T., 2005. A large Cretaceous theropodfrom Patagonia, Argentina, and the evolution of carcharodontosaurids.Naturwissenschaften 92, 226–230.

O’Leary, M.A., Roberts, E.M., Head, J.J., Sissoko, F., Bouaré, M.M., 2004. Titanosaurian(Dinosauria, Sauropoda) remains from the ‘‘Continental Intercalaire” of Mali.Journal of Vertebrate Paleontology 24, 923–930.

Pittet, F., Cavin, L., Poyato-Ariza, F.J., 2009. A New Ostariophysan Fish from the earlyLate Cretaceous (Cenomanian) of SE Morocco, with a Discussion of itsPhylogenetic Relationships. In: Grande, T. (Ed.), A Comprehensive Review ofGonorynchiformes and of Ostariophysan Relationships. Science Publishers, Inc.,pp. 332–355.

Pons, D., Berthou, P.-Y., Campos, D.A., 1990. Quelques observations sur lapalynologie de l’Aptien Supérieur et de l’Albien du Bassin d’Araripe (NE duBrésil). In: Atas 1 Simp. Bacia do Araripe e bacias interiores do Nordeste, Crato,pp. 241–252.

Premoli Silva, I., Sliter, W.V., 1995. Cretaceous planktonic foraminiferabiostratigraphy and evolutionary trends from the Bottaccione section, Gubbio,Italy. Palaeontographia Italica 82. 89 p.

Premoli Silva, I., Verga, D., 2004. Practical Manual of Cretaceous PlanktonicForaminifera. In: Verga, D., Rettori, R. (Eds.), Internationl School on PlanktonicForaminifera, 3� Course, Cretaceous. Universities of Perugia and Milan,Tipografia Pontefelcino, Perugia, Italy.

Rage, J.-C., Cappetta, H., 2002. Vertebrates from the Cenomanian, and thegeological age of the Draa Ubari fauna (Libya). Annales de Paléontologie 88,79–84.

Rage, J.C., Dutheil, D.B., 2008. Amphibians and squamates from the Cretaceous(Cenomanian) of Morocco. A preliminary study, with description of a new genusof pipid frog. Palaeontographica, Abteilung A 285, 1–22.

Rauhut, O.W.M., 1999. A dinosaur fauna from the Upper Cretaceous (Cenomanian)of northern Sudan. Palaeontologia Africana 35, 61–84.

Rauhut, O.W.M., López-Arbarello, A., 2009. Considerations on the age of theTiouaren Formation (Iullemmeden Basin, Niger, Africa), implications forGondwanan Mesozoic terrestrial vertebrate faunas. Palaeogeography,Palaeoclimatology, Palaeoecology 271, 259–267.


Riff, D., Mader, B., Kellner, A.W.A., Russell, D., 2004. An avian vertebra from thecontinental Cretaceous of Morocco, Africa. Arquivos do Museu Nacional Rio deJaneiro 62 (2), 217–223.

Robaszynski, F., Caron, M. (Eds.), and European Working Group on PlanktonicForaminifera, 1979. Atlas of Mid Cretaceous Planktonic Foraminifera (Boreal Seaand Tethys). Parts 1–2, Cahiers de Micropaléontologie, 366 p.

Russell, D.A., 1996. Isolated Dinosaur bones from the Middle Cretaceous of theTafilat, Morocco. Bulletin du Muséum National d’Histoire Naturelle, Paris, 4eSérie, Section C, 18 (2–3), 349–402.

Russell, D.A., Paesler, M.A., 2003. Environments of Mid-Cretaceous Saharandinosaurs. Cretaceous Research 24, 569–588.

Salgado, L., Garrido, A., Cocca, S.E., Cocca, J.R., 2004. Lower Cretaceousrebbachisaurid sauropods from Cerro Aguada Del León, Neuquén Province,northwestern Patagonia, Argentina. Journal of Vertebrate Paleontology 24 (4),903–912.

Sereno, P.C., Larsson, H.C.E., 2009. Cretaceous Crocodyliforms from the Sahara.ZooKeys 28, 1–143.

Sereno, P.C., Wilson, J.A., Larsson, H.C.E., Sues, H.-D., 1994. Early Cretaceousdinosaurs from the Sahara. Science 266, 267–271.

Sereno, P.C., Dutheil, D.B., Iarochène, M., Larsson, H.C.E., Lyon, G.H., Magwene,P.M., Sidor, C.A., Varricchio, D.J., Wilson, J.A., 1996. Predatory Dinosaurs fromthe Sahara and Late Cretaceous Faunal Differentiation. Science 272, 986–991.

Sereno, P.C., Beck, A.L., Dutheil, D.B., Gado, B., Larsson, H.C.E., Lyon, G.H., Marcot, J.D.,Rauhut, O.W.M., Sadleir, R.W., Sidor, C.A., Varrichio, D.D., Wilson, G.P., Wilson,J.A., 1998. A long-snouted predatory dinosaur from Africa and the evolution ofspinosaurids. Science 282, 1298–1302.

Sereno, J.A., Wilson, J.A., Conrad, J.L., 2004. New dinosaurs link southern landmassesin the Mid-Cretaceous. Proceedings of the Royal Society of London B 271 (1546),1325–1330.

Silva Santos, R. da, 1985. Araripichthys castilhoi novo gênero e especie de teleostei daFormação Santana, Chapada do Araripe, Brasil. In: Campos, D.A., Ferreira, C.S.Brito, I.M., Viana, C.F. (Eds.), Coletânea de Trabalhos Paleontológicos, SérieGeologia 27 (2). Rio de Janeiro: Ministério das Minas e Energia—D.N.P.M., pp.133–140.

Smith, J.B., Lamanna, M.C., Lacovara, K.J., Dodson, P., Smith, J.R., Poole, J.C.,Giegengack, R., Attia, Y., 2001. A giant sauropod dinosaur from an UpperCretaceous mangrove deposit in Egypt. Science 292, 1705–1706.

Stromer, E., 1915. Ergebnisse der Forschungsreisen Prof. E. Stromer in den WüstenAgyptens. II. Wirbeltier-Reste der Baharîje-Stufe (unterstes Cenoman). 3. DasOriginal des Theropoden Spinosaurus aegyptiacus nov. gen., nov. spec.Abhandlungen der Königlich Bayerischen Akademie der Wissenschaften,Mathematisch-Physikalische Klasse 28, 1–32.

Stromer, E., 1925. Ergebnisse der Forschungsreisen Prof. E. Stromers in den WüstenÄgyptens. II. Wirbeltier-Reste der Baharije-Stufe (unterstes Cenoman). 7.Stomatosuchus inermis Stromer, ein schwach bezahnter Krokodilier und 8. EinSkelettrest des Pristiden Onchopristis numidus Huag sp. Abhandlungen derKöniglich Bayerischen Akademie der Wissenschaften, Mathematisch-Physikalische Klasse 30, 1–22.

Stromer, E., 1932. Wirbeltiere-Reste der Baharije-Stufe (unterstes Cenoman). 11.Sauropoda. Abhandlungen der Bayerischen Akademie der WissenschaftenMathematisch naturwissenschafiliche Abteilung, Neue Folge 10, 3–21.

Tabaste, N., 1963. Étude de restes de poissons du Crétacé saharien. Mémoire, IFAN68, mélanges Ichthyologiques, 437–485.

Taquet, P., 1970. Sur le gisement de dinosauriens et de crocodiliens de Gadoufaoua(République du Niger). Comptes Rendus de l’Académie des Sciences, Paris, (D)271, 38–40.

Taquet, P., 1976. Géologie et Paléontologie du gisement de Gadoufaoua (Aptien duNiger). Cahiers de Paléontologie, CNRS Paris, 91 p.

Taquet, P., 1980. Succession et repartition des gisements de vertébrés du Crétacé duSahara. Mémoire de la Société Géologique de France, N.S. 139, 185–186.

Taquet, P., 1984. Une curieuse spécialisation du crâne de certains Dinosaurescarnivores du Crétacé, le museau long et étroit des Spinosauridés. ComptesRendus de l’Académie des Sciences, Paris Série II 299, 217–222.

Taverne, L., 1997. Les clupéomorphes (Pisces, Teleostei) du Cénomanien (Crétacé)de Kipala (Kwango), Zaire), ostéologie et phylogénie. Belgian Journal of Zoology127 (1), 75–97.

Taverne, L., 2000. Nouvelles données ostéologiques et phylogénétiques surPalaeonotopterus greenwoodi, notoptéridé (Teleostei, Osteoglossomorpha) duCénomanien inférieur continental (Crétacé) du Maroc. Stuttgarter Beiträge zurNaturkunde, Serie B (Geologie und Paläontologie) 293, 1–24.

Taverne, L., 2003. Redescription critique des genres Thryptodus, Pseudothryptodus etParanogmius, poissons marins (Telostei, Tselfatiiformes) du Crétacé supérieurdes Etats-Unis, d’Egypte et de Libye. Belgian Journal of Zoology 133 (2), 163–173.

Taverne, L., 2004. On a complete hyomandibular of the Cretaceous Moroccannotopterid Palaeonotopterus greenwoodi (Teleostei, Osteoglossomorpha).Stuttgarter Beiträge zur Naturkunde Serie B (Geologie und Paläontologie) 348,1–7.

Taverne, L., Maisey, J.G., 1999. A notopterid skull (Teleostei, Osteoglossomorpha)from the continental early Cretaceous of Southern Morocco. American MuseumNovitates 3260, 1–12.

Tong, H., 2008. Les tortues du Cénomanien continental des Kem Kem et leursimplications paléobiogéographiques. Deuxième Rencontre sur la Valorisation etla Préservation du Patrimoine Paléontologique, Meknès, 9–13 May. Abstractsvolume, pp. 69–70.

Tong, H., Buffetaut, E., 1996. A new genus and species of pleurodiran turtle from theCretaceous of southern Morocco. Neues Jahrbuch für Geologie undPaläontologie, Abhandlungen 199 (1), 133–150.

Tykoski, R.S., Rowe, T., 2004. Ceratosauria. In: Weishampel, D.B., Dodson, P.,Osmólska, H. (Eds.), The Dinosauria, second ed. University of California Press,Berkeley, pp. 151–168.

Underwood, C., Cleason, K.M., Ward, D.J., 2009. Batoids from the Turonian of SEMorocco. In: Jalil, N.-E. (Ed.), 1st International Congress on North AfricanVertebrate Palaeontology, Program & Abstracts, Marrakech, 25–27 May 2009, p.43.

Van Der Zwaan, G.J., Duijnstee, I.A.P., Den, D.M., Ernst, S.R., Jannink, N.T.,Kouwenhoven, T.J., 1999. Benthic foraminifers; proxies or problems? A reviewof paleocological concepts. Earth-Science Review 46 (1–4), 213–236.

Vullo, R., Néraudeau, D., 2008a. When the ‘‘primitive” shark Tribodus(Hybodontiformes) meets the ‘‘modern” ray Pseudohypolophus (Rajiformes),the unique co-occurrence of these two durophagous Cretaceous selachians inCharentes (SW France). Acta Geologica Polonica 58 (2), 249–255.

Vullo, R., Néraudeau, D., 2008b. Cenomanian vertebrate assemblages fromsouthwestern France, a new insight into the European mid-Cretaceouscontinental fauna. Cretaceous Research 29 (5–6), 930–935.

Vullo, R., Néraudeau, D., Allain, R., Cappetta, H., 2005. Un nouveau gisement àmicrorestes de vertébrés continentaux et littoraux dans le Cénomanieninférieur de Fouras (Charente-Maritime, Sud-Ouest de la France). ComptesRendus Palevol 4, 95–107.

Weiler, W., 1935. Ergebnisse der Forschungsreisen Prof. E. Stromers in den WüstenÄgyptens. II. Wirbeltierreste der Baharîje-Stufe (unterstes Cenoman). NeueUntersuchungen an den Fischresten. Abhandlungen der Bayerischen Akademieder Wissenschaften, Mathematische-naturwissenschaftliche Abteilung, NeueFolge 32, 1–57.

Wellnhofer, P., Buffetaut, E., 1999. Pterosaur remains from the Cretaceous ofMorocco. Paläontologische Zeitschrift 73 (1/2), 133–142.

Wenz, S., 1980. A propos du genre Mawsonia, Coelacanthe géant du Crétacéinférieur d’Afrique et du Brésil. Mémoire de la Société Géologique de France,N.S. 139, 187–190.

Wenz, S., 1981. Un coelacanthe géant, Mawsonia lavocati Tabaste, de l’Albien–basedu Cénomanien du Sud marocain. Annales de Paléontologie (Vertébrés) 67 (1),1–20.

Werner, C., 1994. Die kontinentale Wirbeltierfauna aus der unteren Oberkreide desSudan (Wadi Milk Formation). Berliner Geowissenschaftliche Abhandlungen (E)13, 221–249.

Wilson, M.V.H., Murray, A.M., Chatterton, B.D.E., Gibbs, S., 2009. Late Cretaceousactinopterygian fishes from Morocco. In: Jalil, N.-E. (Ed.), 1st InternationalCongress on North African Vertebrate Palaeontology, Program & Abstracts,Marrakech, 25–27 May 2009, p. 44.

Zouhri, S., Kchikach, A., Saddiqi, O., El Haïmer, F.Z., Baidder, L., Michard, A., 2008. TheCretaceous–Tertiary Plateaus. In: Michard, A., Saddiqi, O., Chalouan, A., Frizonde Lamotte, D. (Eds.), Continental Evolution: The Geology of Morocco. SpringerVerlag, Berlin, Heidelberg, pp. 331–358.

Vertebrate assemblages from the early Late Cretaceous of southeastern Morocco: An overview

Documents

Transcript of Vertebrate assemblages from the early Late Cretaceous of southeastern Morocco: An overview