ORIGINAL PAPER

A Phororhacoid bird from the Eocene of Africa

Cécile Mourer-Chauviré & Rodolphe Tabuce &

M’hammed Mahboubi & Mohammed Adaci &Mustapha Bensalah

Received: 4 May 2011 /Revised: 20 July 2011 /Accepted: 22 July 2011# Springer-Verlag 2011

Abstract The bird fossil record is globally scarce inAfrica. The early Tertiary evolution of terrestrial birds isvirtually unknown in that continent. Here, we report on afemur of a large terrestrial new genus discovered fromthe early or early middle Eocene (between ∼52 and46 Ma) of south-western Algeria. This femur shows allthe morphological features of the Phororhacoidea, the so-called Terror Birds. Most of the phororhacoids wereindeed large, or even gigantic, flightless predators orscavengers with no close modern analogs. It is likely thatthis extinct group originated in South America, wherethey are known from the late Paleocene to the latePleistocene (∼59 to 0.01 Ma). The presence of aphororhacoid bird in Africa cannot be explained by a

vicariant mechanism because these birds first appeared inSouth America well after the onset of the mid-CretaceousGondwana break up (∼100 million years old). Here, wepropose two hypotheses to account for this occurrence,either an early dispersal of small members of this group,which were still able of a limited flight, or a transoceanicmigration of flightless birds from South America toAfrica during the Paleocene or earliest Eocene. Paleo-geographic reconstructions of the South Atlantic Oceansuggest the existence of several islands of considerablesize between South America and Africa during the earlyTertiary, which could have helped a transatlantic dispersalof phororhacoids.

Keywords Aves . Eocene . Algeria . South America .

Paleobiogeography. Transatlantic dispersal

As yet, very little is known about the Paleogeneavifaunas from Africa. The phosphatic deposits fromthe upper Paleocene and lower Eocene of Morocco haveyielded almost exclusively marine birds, belonging to theextinct families Pelagornithidae (pseudo-toothed birds)and Prophaethontidae (early representatives of tropic-birds), and possibly to the recent family Phaethontidae(tropicbirds) (Bourdon et al. 2010 and references therein).A few remains of Pelagornithidae have also been found inthe middle Eocene of Nigeria and Togo (Mayr 2009).Among the terrestrial birds, a primitive galliform has beenrecently reported from the middle Eocene of Namibia(Mourer-Chauviré et al. 2011), and a relatively diversifiedavifauna is known in the late Eocene or earliest Oligocene(Seiffert 2006) of the Jebel Qatrani Fm. of the Fayum inEgypt (Rasmussen et al. 1987; 2001). Because of this poorfossil record, it is hard to say anything significant about

Communicated by: Robert Reisz

C. Mourer-ChauviréLaboratoire de Géologie de Lyon, Université de Lyon,Université Lyon 1, CNRS, UMR 5276,2 rue Dubois,69622 Villeurbanne Cedex, France

R. Tabuce (*)Institut des Sciences de l’Évolution, UMR 5554, cc064,Université Montpellier II,Place Eugène Bataillon,34095 Montpellier Cedex 05, Francee-mail: Rodolphe.Tabuce@univ-montp2.fr

M. MahboubiLaboratoire de Paléontologie Stratigraphique etPaléoenvironnement, Université d’Oran,B.P. 1524, El M’naouer, Oran 31000, Algeria

M. Adaci :M. BensalahLaboratoire de Recherche n° 25, Département des Sciencesde la Terre, Université Abou Bekr Belkaïd,B.P. 119, Tlemcen 13000, Algeria

NaturwissenschaftenDOI 10.1007/s00114-011-0829-5

the evolution of the early Paleogene terrestrial avifauna inAfro-Arabia. In fact, in that continent, very few earlyPaleogene sites have so far yielded an exclusive terrestrial/continental vertebrate fauna. Among them, the Gour Lazibfauna in Algeria is famous for having yielded numerousmammals including several primates, hyraxes, rodents,elephant-shrews, insectivoran-grade mammals, chiropter-ans, creodonts, and “condylarths” (Tabuce et al. 2009 andreferences therein). The mammalian fauna from the GourLazib is the most diverse from the entire Afro-ArabianEocene; it is dated from the lower Eocene (Ypresian) orfrom the very beginning of the middle Eocene (Lutetian).In January 2009, our continuing long-standing fieldworkin the vertebrate-bearing localities of the Gour Lazib hasled to the first discovery of some bird remains. Thesefossils represent the oldest known occurrence of terrestrialbirds in Africa. We describe here a new genus and speciesof a phororacoid bird documented by an almost completefemur discovered from the HGL 51 level of the GlibZegdou outlier, Gour Lazib area (Fig. 1).

Results

Systematic paleontology Aves Linnaeus, 1758; GruiformesReichenbach, 1852; Cariamae Fürbringer, 1888; Phororha-coidea Patterson 1941; Family incertae sedis; Lavocatavisafricana gen. et sp. nov.

Holotype UM (Université des Sciences et Techniques duLanguedoc, Montpellier) HGL 51–55, right femur, almostcomplete (Figs. 2 and 3). A cast will be deposited in thecollection of the Universities of Oran Es Senia andTlemcen, Algeria.

Etymology Lavocat, for René Lavocat who first reportedthese fossiliferous outcrops, avis, Latin word for bird, andafricana from Africa.

Type locality and age Hammada Gour Lazib 51. GourLazib area, intermediate member of the Glib ZegdouFormation. The charophyte assemblage, discovered

Fig. 1 a Location of the Gour Lazib area, western part of the AlgerianSahara. In this region, the Tertiary series form vast desertic plateaux, theHammadas. The Gour Lazib area constitutes a group of outliers of theseTertiary series, and the Glib Zegdou (b) is one of those outliers. c Recentexcavation in the intermediate member of the Glib Zegdou Formationled to the discovery of a rich vertebrate fauna associated with a

charophyta flora (Tabuce et al. 2009 and references therein). The levelHGL50 has yielded a diverse mammalian fauna dated from the lateearly or early middle Eocene. The level HGL51, which is situatedjust ∼1 m above HGL50, has yielded a similar mammalian fauna andthe femur of Lavocatavis africana gen. et sp. nov. here studied

Naturwissenschaften

throughout this intermediate member, is composed ofseveral species belonging to the genera Raskyella, Stepha-nochara, Nodosochara, Peckichara, Harrisichara, Nitel-lopsis, and Maedleriella. This floral assemblage indicates alate early or early middle Eocene age (Mebrouk et al.1997). Furthermore, mammalian biostratigraphic data alsosuggests an age earlier than late middle Eocene. Really,several groups (namely, anthracotheriid artiodactyls, an-thropoid primates, and hystricognathous rodents), whichentered Africa during late middle Eocene faunal exchange(s) between Asia and North Africa, are lacking in theintermediate member of the Glib Zegdou Formation (e.g.,Tabuce and Marivaux 2005; Gheerbrant and Rage 2006).

Diagnosis Large-sized femur, straight shaft, caudal face ofthe shaft flattened, condylus lateralis and trochlea fibularisprojecting laterally and distally, great development of thefossa poplitea, of the crista supracondylaris medialis, and ofthe osseous bar which joins the condylus medialis to thecondylus lateralis. This unique combination of characters isfound only in the family Phorusrhacidae. Comparatively tothe Phorusrhacidae of similar or larger size (i.e., all thePhorusrhacidae apart from the Psilopterinae), Lavocatavishas the smallest distal end in relation to the total length ofthe femur. In addition, the proximo-distal diameter of the

caput femoris, proportionally to the total length, is smallerin Lavocatavis than in the members of the familyPhorusrhacidae.

Description and comparisons

The proximal and distal parts of this femur are incompletelypreserved on the cranial side. The bone is better preservedon its caudal side, except for a part of the condylus lateraliswhich is missing. The caput femoris is crushed. The shapeis very rectilinear. The shaft is conspicuously flattened onits caudal face and rounded on its cranial face. At theproximal part, the caput femoris is strongly projectingmedially, the crista trochanteris is present but incompletelypreserved, and on the caudal face there are strong muscularimpressions (impresiones obturatoriae). At the distal part,on the caudal face, the condylus medialis is well preserved,the condylus lateralis is partly missing, and the trochleafibularis is strongly projecting laterally. The laterodistal partof the articulation is laterally elongated and projects fartherdistally than the mediodistal part. The fossa poplitea is ovalshaped, proximo-distally elongated, and very deep. Itpenetrates about 15 mm deep under an osseous bar whichjoins the condylus medialis to the condylus lateralis. Thecrista supracondylaris medialis is situated in the prolonga-

Fig. 2 Femur of Lavocatavis africana gen. et sp. nov. in comparisonto that of Patagornis marshi Moreno and Mercerat 1891, early tomiddle Miocene of Argentina. Patagornis marshi, left femur, reversed,a caudal view, after Andrews (1899, pl. XVII, figure 8) andreconstruction of the Patagornithine Andalgalornis steulleti (Pattersonand Kraglievich 1960) scaled for comparison to a human (Alvarengaand Höfling 2003); femur measurements (Table 1) indicate that P.marshi and L. africana were of similar body size. L. africana, right

femur, holotype, UM HGL 51–55. b Caudal view. c Caudal viewslightly mediocaudally oriented. d Cranial view. Abbreviations: cacaput femoris, cf collum femoris, cl condylus lateralis, cm condylusmedialis, csm crista supracondylaris medialis, ct crista trochanteris,faa facies articularis antitrochanterica, faac facies articularis acetabu-laris, fp fossa poplitea, io impressiones obturatoriae, tf trochleafibularis. Scale bar=2 cm

Naturwissenschaften

tion of the condylus medialis and is extremely developed. Itextends on the mediocaudal angle of the shaft, on about 4/5of the total length of the bone. On the cranial face, bothcondyles are not preserved and it is not possible to see theshape of the sulcus patellaris. There is no visible lineaintermuscularis cranialis.

Comparisons Considering the least shaft circumference ofits femur (79 mm), Lavocatavis had an estimated weight ofca. 32 kg (Campbell and Marcus 1992) or even 45–50 kg(Alvarenga and Höfling 2003). Now, among the recentbirds, the maximum weight attained by the flying formsstands around 14 kg, some exceptionally reaching 20 kg(Dunning 1993; Mayr and Rubilar-Rogers 2010). It istherefore possible to assume that Lavocatavis was veryprobably flightless. The late Mesozoic and Cenozoic birdfossil record offers evidence that several large forms amongprimitive, palaeognathous, and neognathous birds were alsoflightless.

Among primitive forms, the genus Gargantuavis, fromthe late Campanian or early Maastrichtian of France, hasbeen described from a synsacrum and a referred femur, andis placed in the Ornithothoraces (Buffetaut and Le Loeuff1998, 2011). Mayr (2009, p. 21) indicates that thissynsacrum could also be attributed to a very large pterosaur.However, the femur referred to Gargantuavis is verydifferent from the pterosaurs femora which are relativelylong and narrow, with a caput femoris strongly projectingproximally (see Wellnhofer 1980, figure 31), and weconsider it as avian. This femur is short and stout, stronglyincurved mediolaterally, and shows a large crista trochan-teris. The shape of this crista trochanteris, stronglyprojecting laterally and rounded, is due to the strongcompression of the bone during the processus of fossiliza-tion. If one could make a three-dimensional restoration of

its form, it would not be very different from a femur ofGastornithidae. It differs from Lavocatavis, the femur ofwhich is very straight.

The oldest known palaeognath is Diogenornis fragilis,from the late Paleocene of Itaborai, Brazil (Alvarenga1983), which is placed in the recent family Rheidae, but itsfemur is unknown. The other palaeognathous birds of thePaleogene are the Palaeotididae, the Remiornithidae, andthe Eremopezidae. The femur is known in Palaeotisweigelti, from the middle Eocene of Germany (Houde andHaubold 1987). This femur is elongate and shows a strongcurvature, caudally directed, at its distal end. In theRemiornithidae, from the late Paleocene of France (Mayr2009), and in the Eremopezidae, from the late Eocene orearliest Oligocene of Fayum, Egypt (Rasmussen et al. 2001;Seiffert 2006), the femur is unknown but the other knownelements show some similarities with those of other extinctor recent palaeognaths. In the recent large-sized Struthioni-formes, as well as in the extinct Dinornithidae andAepyornithidae, the femur is generally short and stout,and strongly widened at both extremities. The section of thediaphysis is circular in shape and it is not flattened on itscaudal surface. The crista trochanteris is generally projec-ting proximally; it is strongly projecting in the Dinornithi-dae and Aepyornithidae, and moderately projecting in therecent forms with the exception of Struthio where the faciesarticularis acetabularis extends slightly proximally beyondthe crista trochanteris. In the recent large ratites, the cristatrochanteris is also projecting on the cranial side. In therecent ratites and the Aepyornithidae, there is a very deeppopliteal fossa. The osseous bar which links the medialcondyle and the lateral condyle is very oblique compared tothe longitudinal axis of the shaft, while in Lavocatavis thisbar is almost perpendicular to this axis. The crista supra-condylaris medialis is weakly developed and is not situatedalong the mediocaudal border of the shaft.

Several forms of large flightless neognathous birds havebeen described in the Paleogene. In the Gastornithidae,known from the Paleocene to the middle Eocene of Europe,North America, and Asia, with the genera Gastornis andDiatryma, the femur is short and stout, strongly widened atboth ends (Andors 1992; Mayr 2009). The AustralianDromornithidae are known mainly from the late Oligocene,with a tentative record of footprints in the early Eocene(Murray and Vickers-Rich 2004). In these birds, the femuris strongly mediolaterally incurved, both ends are muchwidened, and the crista trochanteris is strongly projecting.In conclusion, the Glib Zegdou femur is very different fromthat of the other flightless birds, extinct or extant, andcannot be attributed to any of these groups.

By contrast, Lavocatavis presents affinities with Phor-usrhacidae. All the authors who have described the femurof the Phorusrhacidae (Andrews 1899; Ameghino 1920;

Fig. 3 Lavocatavis africana gen. et sp. nov., right femur, holotype,UM HGL 51–55. Diagram showing the places where the bone isincompletely preserved, in dotted lines. a Caudal view; b cranial view

Naturwissenschaften

Sinclair and Farr 1932) have insisted on the followingcharacteristics: very straight shape; no curvature in thecraniocaudal direction, nor in the mediolateral direction,with only a slight cranial convexity at the distal end; greatdevelopment of the fossa poplitea and of the crista supra-condylaris medialis; and condylus lateralis projectingdistally and laterally. All these characteristics are presenton the Glib Zegdou femur. The femur of Lavocatavis alsopresents, at its proximal end, a flattened facies articularisantitrochanterica, and there is no projecting trochanterfemoris. The caput femoris is strongly projecting mediallyand the collum femoris is weakly indicated, but this part isnot well preserved.

The suprafamilial name Phororhacoidea Patterson 1941designates the whole of the fossil birds known as TerrorBirds and, according to different authors, includes eitherseveral families, for example in Patterson and Kraglievich(1960), or only one family, Phorurhacidae, with fivesubfamilies, for example in the recent revision of Alvarengaand Höfling (2003). In the present paper, we haveconsidered Lavocatavis as member of a family incertaesedis in the superfamily Phororhacoidea.

Among the Psilopterinae, which are the smallest Phor-usrhacidae, the femur is known in the genera Psilopterusand Procariama (Sinclair and Farr 1932; Alvarenga andHöfling 2003). In these two genera, the shaft has a morecircular section and is less flattened on its caudal face thanin the other Phorusrhacidae, and the distal part is mediallyincurved. The femur of Mesembriornis (subfamily Mesem-briornithinae, middle-sized forms with very elongatetarsometatarsi) is proportionally stouter than the GlibZegdou one and shows a trochlea fibularis projecting verylaterally and distally (Alvarenga and Höfling 2003, figure33, A and D). In the genera Physornis and Paraphysornis(subfamily Brontornithinae, large-sized forms), the femur isshort, stout, and strongly widened at both extremities(Alvarenga and Höfling 2003). In the subfamily Phorusr-hacinae are found truly gigantic forms with very elongatehindlimbs (Alvarenga and Höfling 2003; Bertelli et al.2007). The femur is known in the three genera of thesubfamily Patagornithinae, namely Patagornis, Andrewsor-nis, and Andalgalornis. The femur of Glib Zegdou looksmore similar to those of the Patagornithinae than to anyother member of the Phorusrhacidae.

In the Patagornithinae, the facies articularis acetabularishas a trilobate appearance. The description of Patagornismarshi by Andrews (1899, p. 78) indicates that the femoralhead presents a large blunt-pointed process situated inproximal position, and that, distally, the articular surface isdivided by a shallow groove into an anterior (cranial) and aposterior (caudal) lobe, so that when looked at from theinner (medial) side, the head appears to be imperfectlytrilobate. This feature is clearly visible on one of the

illustrations (Moreno and Mercerat 1891, pl. XIV, figure 7a,right femur, medial view). On the Glib Zegdou femur, thispart is not well preserved, but it is possible to see, in cranialview, a flattened cranial lobe, ellipsoidal in shape, and inmedial view two lobes, a cranial one and a caudal one,separated by a groove. Distally to these lobes, the collumfemoris is slightly excavated. On the cranial face, the cristatrochanteris is incompletely preserved. On the caudal face,it is possible to see an oblique ridge, extending from thelateral border of the shaft to the middle of the faciesarticularis antitrochanterica. This ridge corresponds to theextremely distinct muscular impressions that are present onthe femur of Patagornis marshi (Andrews 1899, pl. XVII,figure 8).

Ameghino (1920, p. 248) describes the femur ofPatagornis marshi as follows: "In posterior view, the distalpart of the bone is flattened and ends into a very deeppopliteal fossa, which is separated from the intercondylarfossa by an exceptionally developed bony bar" (ourtranslation). These characteristics are also present on theGlib Zegdou femur.

The Patagornithinae are known from the Oligocene tothe Mio–Pliocene boundary (Alvarenga and Höfling2003). Lavocatavis shows many similarities morphologi-cally and metrically (Table 1) with Patagornithinae,although it is much older than the known members ofthis subfamily. Lavocatavis is particularly reminiscent ofPatagornis marshi, but differs from this species by thefollowing characteristics: distal part proportionally nar-rower, medial condyle less projecting medially, lateralcondyle and fibular trochlea less projecting laterally (it isnot possible to know the degree of their distal extension),popliteal fossa oval in shape and with better delimitatedoutline, and proximo-distal diameter of caput femorisproportionally smaller (Fig. 2). In Andalgalornis steulleti,the popliteal fossa is more oval in shape and thus thepopliteal fossa of the Glib Zegdou femur is more similar tothis latter form.

In conclusion, Lavocatavis is more similar to themembers of the Phorusrhacidae than to any other extinctor extant large flightless bird. We place it in family incertaesedis in the superfamily Phororhacoidea because it isknown only by a single bone.

Fossil record of Phororhacoidea and paleobiogeographicconsequences

The earliest Phorusrhacidae are known from the latePaleocene of Itaborai, Brazil, and the Eocene ofAntarctica (Mayr 2009). A fragment of bill and a distalpart of tarsometatarsus of a phororhacoid bird have beendiscovered in the late Eocene, age about 36 Ma, of

Naturwissenschaften

Seymour Island, Antarctic Peninsula (Case et al. 1987;Tambussi and Acosta Hospitaleche 2007). Footprints of avery large, three-toed, bird have been reported from FidesPeninsula of King George Island (South Shetland Islands,northern tip of the Antarctic Peninsula) in early Eocenedeposits (Case et al. 1987). These footprints have beenattributed either to a ratite or to a phororhacoid. A femurhas been discovered in the early Maastrichtian of VegaIsland, Antarctic Peninsula (Case et al. 2006). This femurshows similarities with that of cursorial birds and couldcorrespond to an ancestral form of both cariamids andphororhacoids, or to a basal cariamid, ancestral tophororhacoids. During the Neogene, after the establishmentof the Panamanian land bridge, when South America becamelinked to North America, some taxa arrived in North Americawhere they are known from the Pliocene of Texas, and the latePliocene/early Pleistocene of Florida (MacFadden et al.2006). The latest Phorusrhacidae are documented from thelate Pleistocene of Uruguay (Alvarenga et al. 2010).Phorusrhacidae had been reported from the EuropeanPaleogene, but this attribution has been contested and theconcerned taxa are now placed in a distinct family,Ameghinornithidae (Alvarenga and Höfling 2003; Mayr2005). The Ameghinornithidae are now considered asbelonging to the suborder Cariamae but outside a cladeincluding the extinct Idiornithidae and Phorusrhacidae,and the recent Cariamidae (Mayr 2009). As a result, thePhororhacoidea were absent from the Laurasiatic continentduring the Paleogene (Alvarenga and Höfling 2003; Mayr2009), and they were also unknown in Africa up to now.The discovery of a probable flightless phororhacoid atGlib Zegdou as well as the geographic distribution of thegroup could be explained by either vicariance or oceanicdispersal between South America and Africa. Such adispersal through the Laurasiatic continent can be exclud-ed because they are unknown as fossils in the richavifaunas of the Northern Hemisphere.

In this perspective, the paleobiogeographic history ofseveral continental vertebrate groups from the earlyPaleogene of Africa and South America needs to bediscussed here. In addition to Lavocatavis, the Glib Zegdou

fauna includes several fishes and mammals, plus a turtle.The latter is a representative of the podocnemidid turtles, afamily known since the Cretaceous to Recent in SouthAmerica, Europe, Caribbean, and Africa. Although dis-cussed, this distribution could be linked to a vicariant eventafter the mid-Cretaceous opening of the southern AtlanticOcean (Cadena et al. 2010). The vertebrate fauna from theGlib Zegdou also comprises some freshwater fishes (poly-pterids and protopterids) which have their closest relativesin South America, indicative of some vicariant event duringthe Gondwana break-up (Gheerbrant and Rage 2006). Afterthis major tectonic event, and due to tectonic isolation ofAfrica and South America, the early Cenozoic terrestrialvertebrates in these two landmasses have had a long historyof endemic evolution. This endemism gave rise to theemergence of some emblematic clades such as the AfricanAfrotheria and the South American “ungulates”, Xenarthra,and phororhacoids. Interestingly, some of the SouthAmerican “ungulates” (Notoungulata and Astrapotheria atleast) were recently proposed as closely related to theAfrotheria ( Agnolin and Chimento 2011; but see Billet andMartin 2011). Likewise, we must address the Atlantogenataclade, a molecular group that unites the mammaliansuperorder Afrotheria with Xenarthra (Tabuce et al. 2008).According to this molecular hypothesis, the assumedCretaceous divergence of Xenarthra and Afrotheria islinked to the Gondwanian break-up (Wildman et al.2007). Afrotherians are widely distributed during theEocene in Africa and one of their representatives, thehyraxes, are abundant in the Glib Zegdou fauna. Afrother-ian mammals as well as Xenarthra and South American“ungulates” first occur only during the Paleocene (Cifelli1983; de Muizon and Cifelli 2000; Gheerbrant 2009). So, ifthe hypothesis of the common ancestry between Afrotheria,South American “ungulates” and Xenarthra is correct, adeep Cretaceous history of vicariant evolution betweenAfrican and South American mammals is currently unsup-ported by the fossil record.

A vicariance mechanism can also be ruled out for thephororhacoids because they first appeared in South Amer-ica well after the onset of the Gondwana break-up, around

Table 1 Measurements (in millimeters) of the femur of the Glib Zegdou Phororhacoidea, Lavocatavis africana gen. et sp. nov., compared withfemora of Phorusrhacidae (Alvarenga and Höfling 2003)

Lavocatavis africanaUM/HGL 51–55

Patagornis marshiBMNH A 516

Andrewsornis abbottiFMP 14678

Andalgalornis steulletiMACN 69321

Total length 220 as preserved 227 220 –

Proximal width 67.0 as preserved 59 49 –

Width of the shaft in the middle 26.0 25 23 Est. 26

Depth of the shaft in the middle 22.3 – – –

Distal width 52.6 62 – 52

Naturwissenschaften

100 Ma. Moreover, according to the molecular data, thesplit which gave rise to the clade including the higher landbirds, and among them the Cariamidae, occurred slightlybefore the K/T boundary (Ericson et al. 2006; Mayr 2009).

It is likely that the Phorusrhacidae appeared in SouthAmerica because it is there that the oldest forms are known(Late Paleocene), and there that this group became stronglydiversified during all the Cenozoic. In addition, it is also inSouth America that are found the Cariamidae which are thepresent-day closest relatives of the Phorusrhacidae (Acosta-Hospitaleche and Tambussi 2005; Mayr 2009).

The oldest known forms, Psilopterinae, were relativelysmall and Tambussi and Noriega (1996) think that theywere still capable of limited flight. It is possible tohypothesize that small forms dispersed from South Americato Africa by flying and then evolved to give large flightlessforms with morphological characteristics very similar tothat of Patagornithinae.

If the possibility of a convergent evolution is excluded,the alternative hypothesis is that flightless phororhacoidsdispersed from South America to Africa by rafting or byisland hopping. Large flightless phororhacoids were alsopresent in the late Eocene of Antarctica (Case et al. 1987;Tambussi and Acosta Hospitaleche 2007), but it is unlikelythat they dispersed to Africa because, as early as the lateCretaceous, Africa was already very far from Antarctica,while it was still relatively close to the North-East of SouthAmerica (Scotese et al. 1988; Smith et al. 1994).

Until now, only six or seven transatlantic dispersalevents are known for terrestrial vertebrates; they notablyconcern the caviomorph rodents, platyrrhine anthropoids,and several groups of reptiles (Flynn and Wyss 1998; Vidalet al. 2010). Platyrrhines and caviomorphs are sister groupsof two African taxa (catarrhines and phiomorphs, respec-tively); they probably arrived in South America around thesame time, during the late Eocene (Sallam et al. 2009).Their dispersal follows the prevailing westward paleocur-rents and paleowinds suggesting that rafting on floatingislands, in addition to island hopping, was their primarytransoceanic modes of dispersal (Houle 1999; Sallam et al.2009). The oldest known South American rodents havebeen identified in the Tinguiririca fauna in Chile, datedfrom 31.5 Ma, while the oldest known South Americanprimates are dated from 26.4 Ma (Flynn and Wyss 1998).The age of the Glib Zegdou phororhacoid is much older,which suggests that these birds would have crossed fromSouth America to Africa during a much older period,during the Paleocene or the earliest Eocene, crossingfurthered by the fact that the Atlantic Ocean was thennarrower.

Moreover, by comparison with caviomorphs and platyr-rhines, the phororhacoid dispersal is an eastward dispersal.Considering the dominant westward South Atlantic paleo-

currents and paleowinds during the early Tertiary, aneastward rafting on floating islands is improbable, suggest-ing that phororhacoids have dispersed to Africa only byisland hopping. Some possibilities exist for such dispersion.The reconstructions of paleobathymetry made by deOliveira et al. (2008) indicate that at 50 Ma the shortestdistance between Africa and South America was around1,000 km in a straight line (from present-day Sierra Leoneto Paraiba State in Brazil). The ocean was wider furthersouth, “but several islands of considerable size (more than200 km in length) persisted along the present-day sub-merged Rio Grande Rise and Walvis Ridge. Between 20and 30°S, at 50 Ma, a long series of close islands stretchedfrom the African shore, and at least one large island (around500 km in length) was formed by the emergent top of theRio Grande Rise” (de Oliveira et al. 2008, p. 59). Anotherset of islands was present along the Brazilian coast at 20°S.These paleogeographic features, which are underwatertoday, might have reduced considerably the distance of apossible migration of terrestrial vertebrates. It is notpossible for the moment to choose between the twocompeting hypotheses, (1) either a dispersal of small forms,still able of limited flight, which later would have evolvedconvergently with the Phorusrhacidae, or (2) a dispersal offlightless forms by island hopping across the AtlanticOcean. We expect that current and ongoing field researchprojects in latest Cretaceous and earliest Paleogene hori-zons in Africa and South America may document (1) otherpunctual faunal exchanges between the two landmasses,and (2) paleontological evidence in favor of the still poorlysupported common ancestry between the so-called endemicAfrican and South American mammals.

Acknowledgments We thank the vice-chancellors of Tlemcenand Oran Universities, and the authorities from Bechar andTindouf districts, who assisted fieldwork in the Gour Lazib area.We thank H. Alvarenga, S. Chapman, and M. Daniels forsupplying information, and A.-L. Charruault for technical assis-tance. We also thank two anonymous reviewers for constructivecomments. This research was supported by the French ANR-PALASIAFRICA Program (ANR-08-JCJC-0017).

References

Acosta-Hospitaleche C, Tambussi (2005) Phorusrhacidae Psilopterinae(Aves) en la Formación Sarmiento de la localidad de GranHondonada (Eoceno Superior), Patagonia, Argentina. Rev EspPaleont 20:127–132

Agnolin FL, Chimento NR (2011) Afrotherian affinities for endemicSouth American “ungulates”. Mamm Biol 76:101–108

Alvarenga HMF (1983) Uma Ave Ratitae do Paleoceno brasileiro:Bacia calcaria de Itaborai, Estado do Rio de Janeiro, Brasil. BolMus Nac 41:1–7

Alvarenga HMF, Höfling E (2003) Systematic revision of thePhorusrhacidae (Aves: Ralliformes). Pap Avul Zool 43(4):55–91

Naturwissenschaften

Alvarenga HMF, Jones W, Rinderknecht A (2010) The youngestrecord of phorusrhacid birds (Aves, Phorusrhacidae) from the latePleistocene of Uruguay. N Jahrb Geol Paläont 265:229–234

Ameghino F (1920) Sur les Oiseaux fossiles de Patagonie, Obrascompletas, vol. XI. Taller de Impresiones Oficiales, La Plata, pp213–359

Andors AV (1992) Reappraisal of the Eocene groundbird Diatryma(Aves: Anserimorphae). Nat Hist Mus Los Angel Cnty Sci Ser36:109–125

Andrews CW (1899) III. On the extinct birds of Patagonia. I. Theskull and skeleton of Phororhacos inflatus Ameghino. Trans ZoolSoc Lond 15:55–86

Bertelli S, Chiappe LM, Tambussi C (2007) A new Phorusrhacid(Aves: Cariamae) from the Middle Miocene of Patagonia,Argentina. J Vert Paleont 27:409–419

Billet G, Martin T (2011) No evidence for an afrotherian-like delayeddental eruption in South American notoungulates. Naturwissen-schaften 98:509–517

Bourdon E, Amaghzaz M, Bouya B (2010) Pseudotoothed birds(Aves, Odontopterygiformes) from the Early Tertiary of Morocco.Amer Mus Novit 3704:1–71

Buffetaut E, Le Loeuff J (1998) A new giant ground bird from theUpper Cretaceous of southern France. J Geol Soc 155:1–4

Buffetaut E, Le Loeuff J (2011) Gargantuavis philoinos: giant bird orgiant pterosaur? Ann Paleont. doi:10.1016/j.annpal.2011.05.002

Cadena E, Bloch JI, Jaramillo CA (2010) New podocnemidid turtle(Testudines: Pleurodira) from the middle-upper Paleocene ofSouth America. J Vert Paleont 30:367–382

Campbell KE Jr, Marcus L (1992) The relationship of hindlimb bonedimensions to body weight in Birds. Nat Hist Mus Los AngelesCo Sci Ser 36:395–412

Case JA, Woodburne MO, Chaney DS (1987) A gigantic Phororha-coid (?) bird from Antarctica. J Paleont 61:1280–1284

Case J, Reguero M, Martin J, Cordes-Person A (2006) A cursorial birdfrom the Maastrichtian of Antarctica. J Vert Paleont 26(3suppl):48A

Cifelli R (1983) Eutherian tarsals from the Late Paleocene of Brazil.Am Mus Nov 2761:1–31

de Muizon C, Cifelli RL (2000) The “condylarths” (archaic Ungulata,Mammalia) from the Early Palaeocene of Tiupampa (Bolivia):implications on the origin of the South American ungulates.Geodiversitas 22:47–150

de Oliveira FB, Molina EC, Marroig G (2008) Paleogeography of theSouth Atlantic: a route for primates and rodents into the NewWorld?In: Garber PA, Estrada A, Strier KB (eds) South American primates:comparative perspectives in the study of behavior, ecology, andconservation. Springer, New York, pp 55–68

Dunning JB Jr (1993) Body masses of birds of the world. In: DunningJB Jr (ed) CRC handbook of avian body masses. CRC, BocaRaton, pp 3–310

Ericson PGP, Anderson CL, Britton T, Elzanowski A, Johansson US,Källersjö M, Ohlson JI, Parsons TJ, Zuccon D, Mayr G (2006)Diversification of Neoaves: integration of molecular sequencedata and fossils. Biol Lett 2:543–547

Flynn JJ, Wyss AR (1998) Recent advances in South Americanmammalian paleontology. Trends Ecol Evol 13:449–454

Gheerbrant E (2009) Paleocene emergence of elephant relatives andthe rapid radiation of African ungulates. Proc Natl Acad Sci USA106:10717–10721

Gheerbrant E, Rage JC (2006) Paleobiogeography of Africa: howdistinct from Gondwana and Laurasia? Palaeogeogr PalaeoclimPalaeoecol 241:224–246

Houde P, Haubold H (1987) Palaeotis weigelti restudied: a smallMiddle Eocene ostrich (Aves: Struthioniformes). Palaeovertebrata17:27–42

Houle A (1999) The origin of the platyrrhines: an evaluation of theAntarctic scenario and the floating island model. Amer J PhysAnthrop 109:541–599

MacFadden B, Labs-Hochstein J, Hulbert R, Baskin J (2006) Refinedage of the Late Neogene Terror Bird (Titanis) from Florida andTexas using rare earth elements. J Vert Paleont 26(3 suppl):92A

Mayr G (2005) “Old World phorusrhacids” (Aves: Phorusrhacidae): anew look at Strigogyps (“Aenigmavis”) sapea (Peters, 1987).Paleo Bios 25:11–16

Mayr G (2009) Paleogene fossil birds. Springer, BerlinMayr G, Rubilar-Rogers D (2010) Osteology of a new giant bony-

toothed bird from the Miocene of Chile, with a revision of thetaxonomy of Neogene Pelagornithidae. J Vert Paleont 30:1313–1330

Mebrouk F, Mahboubi M, Bessedik M, Feist M (1997) L'apport descharophytes à la stratigraphie des formations continentalespaléogènes de l'Algérie. Geobios 30(2):171–177

Moreno FP, Mercerat A (1891) Catálogo de los pájaros fósiles de laRepública Argentina conservados en el Museo de La Plata. AnMus La Plata Pal Arg 1:1–71

Mourer-Chauviré C, Pickford M, Senut B (2011) The first Palaeogenegalliform from Africa. J Ornith 152:617–622

Murray PF, Vickers-Rich P (2004) Magnificent mihirungs. Thecolossal flightless birds of the Australian dreamtime. IndianaUniversity Press, Bloomington

Patterson B (1941) A new Phororhacoid bird from the DeseadoFormation of Patagonia. Geol Ser Field Mus Nat Hist 8(8):49–54

Patterson B, Kraglievich JL (1960) Sistematica y Nomenclatura de lasAves Fororracoideas del Plioceno Argentino. Publ Mus MunicCien Natur Tradit Mar del Plata 1:1–51

Rasmussen DT, Olson SL, Simons EL (1987) Fossil birds from theOligocene Jebel Qatrani Formation, Fayum Province, Egypt.Smiths Contr Paleobiol 62:1–20

Rasmussen DT, Simons EL, Hertel F, Judd A (2001) Hindlimb of agiant terrestrial bird from the Upper Eocene, Fayum, Egypt.Palaeontology 44:325–337

Sallam HM, Seiffert ER, Steiper ME, Simons EL (2009) Fossil andmolecular evidence constrain scenario for the early evolutionaryand biogeographic history of hystricognathous rodents. Proc NatlAcad Sci USA 106:16722–16727

Scotese CR, Gahagan LM, Larson RL (1988) Plate tectonicreconstructions of the Cretaceous and Cenozoic ocean basins.Tectonophysics 155:27–48

Seiffert ER (2006) Revised age estimates from the later Paleogenemammal faunas of Egypt and Oman. Proc Natl Acad Sci USA103:5000–5005

Sinclair WJ, Farr MS (1932) Part II. Aves of the Santa Cruz beds. RepPrinc Univ Exp Patagonia 7:157–191

Smith AG, Smith DG, Funnell BM (1994) Atlas of Mesozoic andCenozoic coastlines. Cambridge University Press, Cambridge

Tabuce R, Marivaux L (2005) Mammalian interchanges betweenAfrica and Eurasia: an analysis of temporal constraints for theplausible Paleogene anthropoid dispersions. Anthrop Sci 112(3):27–32

Tabuce R, Lehmann T, Asher RJ (2008) Afrotherian mammals: areview of current data. Mammalia 72:2–14

Tabuce R, Marivaux L, Lebrun R, Adaci M, Bensalah M, Fabre P-H,Fara E, Gomes-Rodrigues H, Hautier L, Jaeger J-J, Lazzari V,Mebrouk F, Peigné S, Sudre J, Tafforeau P, Valentin X, MahboubiM (2009) Anthropoid vs. strepsirhine status of the AfricanEocene primates Algeripithecus and Azibius: craniodental evi-dence. Proc R Soc B Biol Sci 276:4087–4094

Tambussi C, Acosta Hospitaleche C (2007) Antarctic birds (Neo-rnithes) during the Cretaceous–Eocene times. Rev Asoc Geol Arg62:604–617

Naturwissenschaften

Tambussi C, Noriega J (1996) Summary of the avian fossil record fromsouthern South America. Münch Geowiss Abh (A) 30:245–254

Vidal N, Marin J, Morini S, Donnellan S, Branch WR, Thomas R,Vences M, Wynn A, Cruaud C, Hedges SB (2010) Blindsnakeevolutionary tree reveals long history on Gondwana. Biol Lett6:558–561

Wellnhofer P (1980) Flugsaurier. Die Neue Brehm-Bücherei 534:1–135

Wildman DE, Uddin M, Opazo JC, Liu G, Lefort V, Guindon S,Gascuel O, Grossman LI, Romero R, Goodman M (2007)Genomics, biogeography, and the diversification of placentalmammals. Proc Natl Acad Sci USA 104:14395–14440

Naturwissenschaften