Sousa and Uiraúna-Brejo das Freiras are two Cretaceous basins from the Rio do Peixe region that present a great amount of dinosaur footprins. These basins are located on the west of Paraíba State, Northeast Brazil, and their origin are related to fault movements along preexisting structural trends of the basement during South Atlantic Ocean opening.
    The main tetrapod ichnofauna comprises isolate footprints and trackways of large and small theropods, besides ornithopods. There are also invertebrate ichnofossils such as trails and burrows produced by arthropods and annelids. The fossils are palynomorphs, plant fragments, ostracods, conchostraceans, fish scales and crocodilomorph bone fragments. These were preserved in alluvial fans, anastomosing, meandering rivers and shallow lakes deposits of Neocomian age - Berriasian to lower Barremian.
    The paleontological-geological relevance of Sousa and Uiuraúna Basins is the abundance of dinosaurian ichnofaunas. It have alread been identified, as well as mapped 22 ichnofossiliferous sites, and recognized 296 large theropod tracks; 29 smaller theropods; 42 sauropods; 2 quadrupedal ornithischians; 2 small ornithopods; 28 graviportal ornithopods; a set of batrachopodid prints; a lacertoid print; a large number of unclassifiable dinosaurian tracks and a large number of small chelonian half-swimming tracks. Altogether the classified dinosaurian individuals number is more than 395. The main distribution area of dinosaur footprints at Passagem das Pedras (Ilha Farm), in Sousa County is nowadays a natural park - The Dinosaur Valley Natural Monument. The park, with 40 ha of area, is presently one of the best preserved paleontological sites in Brazil. This area is now a tourist complex and offers an entire tourism infrastructure, besides a trained staff to guide tourists and to protect the paleontological site.

    Sousa and Uiraúna-Brejo das Freiras are intracratonic basins of northeast Brazil, that developed along preexisting structural trends of the basement during South Atlantic Ocean opening. The age of its deposits based on palynological material is characteristic of Rio da Serra (Berriasian to Hauterivian) and Aratu (lower Barremian) local stages (Lima & Coelho, 1987; Regali, 1990).
    The sedimentation in these basins was controlled by regional tectonic processes (Lima Filho, 1991). During Dom João time (Purbeckian Stage) due to crustal extension, sigmoidal basins developed at the inflection of the northwest-southwest and east-west faults. During the Rio da Serra time (Berriasian to Hauterivian) under the same tectonic stress pattern, the basinal areas increased and its shapes became rhomboidal. In the last stage, probably at the end of the Aratu time (Lower Barremian Stage), there was a change in the tectonic pattern and the sediment accumulation began to decline.
    These deposits reflect a direct control of the sedimentation by tectonic activity. Along the faulted borders of the basins, deposition consisted of alluvial fans, changing to an anastomosing fluvial system more distally. In the central region of the basins, a meandering fluvial system with a wide floodplain was established, where perennial and temporary lakes were developed (Carvalho, 2000a).
    The palaeontological-geological relevance of Sousa and Uiraúna-Brejo das Freiras basins is the abundance of dinosaurian ichnofaunas that represent parts of an Early Cretaceous widespread megatracksite (Viana et al., 1993; Carvalho, 2000a) established during the early stages of the South Atlantic opening. In this area were found 22 ichnofossiliferous sites, with more than 395 dinosaurian individuals.

    Some 80 years ago, in the early twenties, Luciano Jacques de Moraes, a Brazilian mining engineer, was working for the DNOCS - Departamento de Obras contra as Seccas (The Department of works against the drought) and surveying the then little known Brazilian North-East. In the West of the State of Paraíba, Moraes discovered two trackways impressed in the rocky pavement forming the bed of the Peixe River (Rio do Peixe) in the Ilha Farm. They were two criss-crossing trackways of different dimensions and very diverse kind.
    Apparently Moraes sent a slab with an original footprint excavated from the SOPP2 track and a plaster cast of a footprint of the SOPP1 track to the U.S.A. to be studied by an unnamed paleontologist; however he did not receive an answer. It was the common opinion at the DNPM in the seventies that the tracks were sent to the American Museum in New York, but they were not found there by G. Leonardi in 1985. In his book, Moraes (1924) classified the largest trackway (SOPP1) to the Stegosauria or secondarily to the Ceratopsia; evidently he interpreted it as quadrupedal. It may be he interpreted the true footprints as fore-prints and the diplacement rims, with their collapsed mud cracks, as hind-prints. He attributed the SOPP2 trackway to a bipedal dinosaur, without deciding between the Theropoda or the Ornithopoda. Moraes described the tracks in detail, with drawings and photographs. He also extrapolated the dimensions of the presumed trackmakers. A skillful work, for the time; however, Moraes was neither an ichnologist nor a paleontologist and his drawings were rather incorrect.
    Moraes sent photographs to F. von Huene of Tübingen, who instead published the drawings from Moraes' book (Huene, 1931). Huene described the first track (SOPP1) as quadrupedal but with total overlap and the second, as bipedal (SOPP2). He attributed the SOPP1 track to ceratopsids or nodosaurids, with preference for the second, and the SOPP2 to the ornithopods, trachodontids or kalodontids. Huene's classification depended on Moraes' drawings. Huene worked in Brasil in the twenties, but he never visited the Paraíba.
    Moraes's trackways at Sousa were seen in 1969 by Llewellyn I. Price (1961) with Diogenes de Almeida Campos of the DNPM.
    In 1975 Giuseppe Leonardi went to Sousa and found that the trackways had been forgotten. Later a local legend arose: the tracks had been discovered by Anísio Fausto Silva, a local farmer. The centenary of this discovery was celebrated in Sousa in 1997. Probably the main tracks had been seen by many farmers and, before them, by indios, because they are quite conspicuous. However, the discovery, for scientific purposes, must be surely attributed to Moraes. Locally the tracks are known as the ox tracks (SOPP1) and the south american ostrich ("ema") tracks" (SOPP2). In 1975-76 Leonardi excavated the two known tracks from beneath two meters of sand and from beneath the rock the SOPP1 trackway as far as the bank of the river, in the process discovering five more trackways (Leonardi, 1979a). In 1977 Leonardi found more tracks at Ilha Farm and some new localities (Leonardi,1980c,1994): Poço do Motor, Pedregulho, Piedade and Juazeirinho in the Sousa Formation; Serrote do Letreiro in the Antenor Navarro Formation, at that time thought to be palaeozoic. An unusual footprint (really two overstepped footprints of Grallator but misinterpreted by Leonardi as Isochirotherium sp.) suggested an incorrect Triassic age for this level (Leonardi, 1979b; 1980b). However, it was later realized that all the Rio do Peixe Group was Lower Cretaceous and that the contact between both formations was heteropic (Carvalho & Leonardi, 1992). Other Leonardi's expeditions followed (28 altogether, cfr Leonardi, 1994, p. 169), almost one complete year in the field altogether, by means of grants from the CNPp (The Nacional Council for Research).
    In the expedition of 1979 Leonardi (1984b, 1994) discovered the sites of Piau-Caiçara and Matadouro (Sousa Formation); Serrote do Pimenta (Antenor Navarro Formation), Mãe d'Água and Curral Velho (Piranhas Formation). In 1980 and 1981 he worked on the site of Piau-Caiçara, focussing on the behaviour of the trackmakers (Godoy & Leonardi, 1985). During 1981 he discovered, together with Geraldo da C.B. Muniz, dinosaurian tracks in the related basins of Lima Campos, Iguatú and Palestina (Leonardi & Muniz, 1985; Leonardi & Spezzamonte, 1994). The 1983 Ligabue expedition was a joint venture with the paleontologists Giancarlo Ligabue (Venice), Philippe Taquet of the Paris National Museum, Diógenes de A. Campos of the DNPM (Rio de Janeiro), and others (Leonardi, 1984a). After 1983 the tracks of the Rio do Peixe basins began to receive visits from several paleontologists, along with many tourists. In 1984 Leonardi discovered the sites of Zoador, Barragem do Domício, Poço da Volta and Engenho Novo in the Sousa Formation and Aroeira and Cabra Assada in the Antenor Navarro Formation (Leonardi 1985, 1994). From 1984 on, Leonardi's associated in the field work were Maria de Fátima C.F. dos Santos and Claude L. de A. Santos of the Museum Câmara Cascudo of Natal (1987a, b), frequently Ismar de Souza Carvalho (Universidade Federal do Rio de Janeiro - Rio de Janeiro, Brazil) (1989; 1992; 1994) and Luis Carlos Godoy (Ponta Grossa, Paraná). In 1985 W. V. Barbosa discovered the site of Varzea dos Ramos. In 1987 Ismar de Souza Carvalho with Leonardi discovered the site of Pocinhos in the Brejo das Freiras basin (Carvalho, 1989; 1996a).
    During all these years, a long and sometimes fruitless struggle was fought to gain protection of the sites from the authorities at all levels and to set up a Natural Park of the Dinosaur Valley. In 1984 Leonardi, with Jo o C. M. Rodrigues of the E. Goeldi Museum of Belém (Pará), began building replicas of the main dinosaurian forms for the park, in a lab in Sousa. In 1988 the sites of Saguim and Piau II were discovered by Leonardi with Anna Alessandrello of the Museo Civico di Storia Naturale of Milan (Italy). Leonardi and Alessandrello explored the Pombal basin with no result. The Paraíso site was discovered in 1992 by Sérgio A.K. Azevedo (Universidade Federal do Rio de Janeiro, Museu Nacional, Rio de Janeiro; 1993). After Leonardi left Brasil in 1989, the ichnologic work in the NE Brazilian basins was systematically continued by Ismar de Souza Carvalho (Carvalho et al., 1994, 1995; Carvalho, 1996a; 2000a, b) working with researchers from Pernambuco Federal University (Maria Somália Sales Viana; Mário Lima Filho) e Rio Grande do Norte Federal University (Narendra Kumar Srivastava).
    In 1992, a State Act of the Paraíba Governor State declared the region of Passagem das Pedras, nearby Sousa County as a Natural Monument (Decreto n^o 14.833, Diário Oficial do Estado da Paraíba, December, 20th of 1992). In this area are found large iguanodontid trackways, theropod and ornithopod trackways and isolated footprints. It was the first ichnocoenosis described in Sousa basin, since 1924 Luciano Jacques de Moraes scientific studies. Later (1996), an agreement of the Ministry of the Environment, Municipal City Hall and Government of the State implanted and built onwards the tourist complex "Dinosaur's Valley".

    The dinosaurian ichnofaunas of these basins all have the same stratigraphic-time-paleogeographical context, and represent parts of a widespread megatracksite. Similarities in the lithofacies among the deposits where the footprints occur reflect the same tectonic, climatic and sedimentary processes. The environmental setting at that time was influenced by the initial development of the equatorial Atlantic seaway, with an endemic biota living nearby in ephemeral rivers and shallow lakes under hot climatic conditions (Carvalho, 2000a).
    During the Early Cretaceous, hot climatic conditions were widespread, although there was probably a wide range of humidity (Petri, 1998). According to Petri (1983) and Lima (1983) in the earliest Cretaceuos, climatic conditions were more humid, in regions located to the south of the tropical domain (Recôncavo-Tucano-Jatobá basins). Despite a hotter and drier climate to the north, interpretations of depositional environments and fossils suggest the existence of some lakes which, during the Neocomian, locally provided more humid conditions. At that time, South America was still connected to Africa, and the Atlantic Ocean was at its initial developing phase. In the present northeastern Brazil, in an area of hundreds of kilometers square, ephemeral rivers and shallow lakes were important ecosystems.
    The Sousa basin comprises and area of 1,250 km². The Uiraúna-Brejo das Freiras is a smaller basin with 480km². They are located at the west part of Paraíba State, Brazil, at the municipalities of Sousa, Uiraúna, Poço, Brejo da Freiras, Triunfo and Santa Helena. The basement of these basins is composed of highly metamorphosed Precambrian rocks (aligned structurally in a northwest-southwest or east-west direction). The predominant rocks are migmatites, granites, gabbros and amphibolites.
    A formal lithostratigraphic subdivision of the Cretaceous in the Sousa basin, and the neighbouring Uiraúna-Brejo das Freiras basin, was erected by Mabesoone (1972) and Mabesoone & Campanha (1974). These authors designated the Rio do Peixe Group with a total thickness of 2,870 meters, subdivided into the Antenor Navarro, Sousa and Piranhas formations. The Antenor Navarro and Piranhas formations are composed of immature sediments, including breccias and conglomerates, with pebbles of metamorphic and magmatic rocks in a coarse arkose matrix. These lithologic types are located near the faulted borders of the basins. Toward the basin depocenter, there are conglomeratic and fine sandstones, sometimes interbedded with siltstones and shales. Cross-channel and tabular stratification, climbing-ripples and ripple marks are the main sedimentary structures. The Sousa Formation is composed of reddish sandstones, siltstones, mudstones and carbonate nodules; marls also may occur. Common sedimentary structures include mud cracks, convolute structures, ripple marks, climbing ripples, rain drop marks and bioturbations.
    In the Antenor Navarro and Piranhas formations, the footprints are rare. The lithologies of these lithostratigraphic units are conglomerates, coarse sandstones and sandstones interbedded with siltstones. The litofacies, sedimentary structures and geometry of the beds point to a sedimentation in fan-delta, alluvial fan and anastomosing fluvial environments. Footprints were preserved in fine sediments of subaerial sandy bars in alluvial fans and anastomosing rivers (Figure2). Such deposits are located close to the basin borders, and their accumulation was controlled directly by the regional tectonic activity (Carvalho, 2000a).
    In the Sousa Formation the lithologies are sandstones, shales and mudstones, whose finer grain size was more susceptible for footprint preservation. This essentially microclastic sequence points to lacustrine, swampy and meandering-braided fluvial paleoenvironments. Through the study of conchostraceans, Carvalho & Carvalho (1990) and Carvalho (1993, 1996b) inferred the physical and chemical lake aspects in whose borders the dinoturbation was significant. They were small temporary lakes, hot and shallow, in which the water chemical conditions possessed an alkaline character (pH between 7 and 9). Because of the dimensions reached by some of these conchostraceans (up to 3.5 cm length), there would have existed an ecological optimum, with large amount of nutrients and chemical ions such as calcium and phosphorus.

Figure 4 - Theropod footprint. Serrote do Letreiro, Sousa basin.

    It is not always easy to attribute the track to a trackmaker. However, the traks described above can be be attributed, with some probability, to the following groups: the large theropodian tracks pertain very probably, in the Cretaceous of South America, to large predators of the family Abelisauridae Bonaparte & Novas, 1985 and then probably to the Ceratosauria; the small theropod tracks with the III digit substantially longer than the II and IV, which were classically attributed to the Coelurosauria may probably be referred to some south American theropodian families which assumed in that continent the ecological niche occupied by the Coelurosauria in Laurasia, (e.g. Noasauridae Bonaparte & Powell, 1980); the sauropod tracks were perhaps impressed by Dicraeosauridae, Rebbachisauridae or, more probably, by early titanosaurs (Bonaparte,1986).
    The short quadrupedal trackway of Serrote do Pimenta can very probably belong to an ankylosaur, most likely a nodosaurid, like for example the little Minmi Molnar, 1980 of Queensland. The quadrupedal trackway (undertrack) from Passagem das Pedras could also, perhaps, belong to this group. Ankylosaur trackways are known from Bolivia (Leonardi, 1984a) and more were recently (1998) recorded in the field by Christian Meyer, Martin Lockley and Leonardi and await publication. The large bipedal or quadrupedal tracks with three roundish hooves are attributable to graviportal iguanodontids similar, for example, to Ouranosaurus Taquet, 1976; some rare small ornithopodoid tracks may belong to young iguanodontids or better to some form of small-sized ornithopod, perhaps dryosaurids, already known in South America (Coria & Salgado, 1996).
    The batrachopodidae set of Piau is poorly impressed, however it could, perhaps, be attributed to Crocodylia. The innumerable very small half-swimming tracks of the Piau site are very probably attributable to turtles; the Araripemydae family being the main candidate.

    The main distribution area of dinosaur footprints at Passagem das Pedras (Ilha Farm), in Sousa county is nowadays a natural park. In December of 1992, through an State Act (Diário Oficial do Estado da Paraíba, Decreto n^o 14.833, December, 20^th of 1992), this area was defined as a Natural Monument - named as Dinosaur Valley Natural Monument (Monumento Natural Vale dos Dinossauros) (Figure 5 and 6).
    In 1996 it was established an agreement among the Ministry of the Environment, Paraíba State Government (SUDEMA) and Sousa Municipal City Hall (Convênio MMA/PNMA/PED n^o 96 CV00030/96). The finantial investiments amount are approximately of US$ 800,000.00 (eight hundred thousand american dollars).
The protection of this ichnofossiliferous site comprises (Sônia Matos Falcão, personal information):

Figure 5 - Dinosaur Valley Natural Monument, a natural park with 40 hectares of protect area established by the Paraíba State Government and Sousa Municipal City Hall.

Figure 6 - Dinosaur track at the Dinosaur Valley Natural Monument. Passagem das Pedras (Ilha farm), Sousa county.

Figure 7 - Replicas of the dinosaurs that lived in the Sousa basin. Art of João Carlos M. Rodrigues.

Figure 8 - Informative center of the tourist Complex Dinosaur's

    Mr. Tim Halley of the Zoology Department of the Queensland University at Brisbane (Austrália) and Dr. José F. Bonaparte of Museo Argentino de Ciencias Naturales Bernardino Rivadavia (Argentina) for their critical review.
    Sônia Matos Falcão (Superintendência de Administração do Meio Ambiente - SUDEMA/Governo do Estado da Paraíba) for the information on the finantial and technical investments that have been done in the Dinosaur Valley Natural Monument. This study was supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ) e Fundação Universitária José Bonifácio (FUJB/UFRJ).