Jurassic Dinosaur Footprints from Southern Italy - CiteSeerX

534

RESEARCH REPORT

Jurassic Dinosaur Footprints from Southern Italy: Footprints as Indicators of Constraints in Paleogeographic Interpretation MARIA ALESSANDRA CONTI Dipartimento di Scienze della Terra, Universita` ‘‘La Sapienza’’, P.le A. Moro 5, 00185, Roma, Italy MICHELE MORSILLI Dipartimento di Scienze della Terra, Universita` di Ferrara, C.so Ercole I d’Este 32, 44100, Ferrara, Italy UMBERTO NICOSIA*, EVA SACCHI Dipartimento di Scienze della Terra, Universita` ‘‘La Sapienza’’, P.le A. Moro 5, 00185, Roma, Italy; Email: [email protected] VINCENZO SAVINO, ALEXANDER WAGENSOMMER, LEONARDO DI MAGGIO Speleo Club Sperone, San Giovanni Rotondo, 71013, Foggia, Italy PIERO GIANOLLA Dipartimento di Scienze della Terra, Universita` di Ferrara, C.so Ercole I d’Este 32, 44100, Ferrara, Italy

PALAIOS, 2005, V. 20, p. 534–550

DOI 10.2110/palo.2003.p03-99

Three loose blocks, rich in dinosaur footprints, were found in a small pier at Mattinata (Gargano Promontory, Foggia, Italy), most probably quarried from the Upper Jurassic Sannicandro Formation. All of the footprints in the blocks are ascribed to medium-sized theropod trackmakers. Recent track discoveries from both the Early Cretaceous San Giovanni Rotondo Limestone and the Late Cretaceous Altamura Limestone, as well as this new discovery, reveal the consistency of terrestrial associations along the southern margin of the Tethys Ocean in the peri-Mediterranean area at the end of Jurassic through Cretaceous times. The presence of these dinosaur-track-rich levels within marine sediments of the Apulia Platform underlines the relevance of dinosaur footprints as a means of constraining paleogeographic reconstructions.

Sasso, 2003). The discovery of terrestrial animal tracks imprinted in marine carbonates is considered important, and the Mattinata footprints appear distinct from the material already known from this same region. GEOLOGICAL AND STRATIGRAPHICAL SETTING

*Corresponding Author

The Gargano Promontory is part of a large paleogeographical unit known as the Apulia Carbonate Platform, which, during the Mesozoic, was part of the southern margin of the Tethys Ocean. The Apulia Platform is considered one of the so-called peri-Adriatic platforms that are quite similar in facies architecture, size, and shape to the modern Bahamian Banks (Bernoulli, 1972; D’Argenio, 1976; Eberli et al., 1993). The Gargano Promontory and other parts of the Apulia Region make up part of the foreland of the Apennine chain. Structurally, the Gargano area is folded into a gentle anticline with a WNW axis (Martinis, 1965). This broad structure contains numerous faults with various trends and kinematics (Funiciello et al., 1992; Bertotti et al., 1999). The most prominent structural feature is the Mattinata fault, a regional E–W shear zone that crosses the entire Gargano area, and continues offshore for many tens of kilometers (De Dominicis and Mazzoldi, 1989). The Gargano area, together with Maiella Mountain, is the only place where the transition between platform and slope-to-basin facies crops out. In other areas, the eastern margin of this platform lies offshore under the Adriatic Sea, about 20 to 30 km from the present coastline (De Dominicis and Mazzoldi, 1989). The stratigraphy of the Gargano area has been reinterpreted frequently in the last two decades. In the late 1960s–early 1970s, geological surveys introduced many formational units (Martinis and Pavan, 1967; Merla et al., 1969; Cremonini et al., 1971), complicating stratigraphical correlations in this area. Other reviews and original stud-

Copyright Q 2005, SEPM (Society for Sedimentary Geology)

0883-1351/05/0020-0534/$3.00

INTRODUCTION In early 2001, two of the authors (V.S., A.W.) found dinosaur footprints on the surface of three loose limestone blocks in a small pier that protects the entry to the small harbor of Mattinata on the Gargano Promontory (Puglia, southern Italy, Fig. 1). The blocks were removed and are stored in the Mattinata town museum (Museo Civico Storico-Archeologico). This is the third find of dinosaur footprints in the Apulia region, where several single dinosaur footprints and trackways have been found previously. The first site was found in 1999 in the Murge area near the town of Altamura (Andreassi et al., 1999; Nicosia et al., 2000a, b; Dal Sasso, 2003), and other footprints were discovered in the Gargano area (Gianolla et al., 2000a; Dal

DINOSAUR FOOTPRINTS FROM SOUTHERN ITALY

535

FIGURE 1—Location map of the Gargano Promontory with main roads and towns.

ies defined new stratigraphical schemes for this area concerning platform and slope-to-basin units (e.g., Luperto Sinni and Masse, 1986, 1994; Bosellini et al., 1993, 1999; Claps et al., 1996; Luperto Sinni, 1996; Bosellini and Morsilli, 1997, 2001), and proposed new models for the evolution of this carbonate platform (Masse and Borgomano, 1987; Bosellini et al., 1993, 1999; Morsilli and Bosellini, 1997). The successions cropping out were divided into different second-order stratigraphical sequences, bounded by unconformities, of different types and origins (see Bosellini et al., 1999, for a review). Herein, the focus mainly is on the lithostratigraphical units that contain shallow-water sediments in which footprints of terrestrial reptiles could have been impressed.

The main sequences that contain shallow-water sediments are the Monte Sacro Sequence and the Mattinata Sequence (sensu Bosellini et al., 1999; Fig. 2). During the Late Jurassic–Early Cretaceous, two inner-platform units were deposited in the Gargano area: the Sannicandro Formation and the San Giovanni Rotondo Limestone. Innerplatform facies of Late Cretaceous age also are present in some small outcrops, and have been named the Calcari di Casa Lauriola (Merla et al., 1969); these units correspond to the well-known Altamura Limestone (Laviano and Marino, 1998; Bosellini and Morsilli, 2001). Sannicandro Formation The Sannicandro Formation is poorly known when compared to other units in the area because previous works in-

FIGURE 2—Chronostratigraphic chart showing the Upper Jurassic–Lower Cretaceous formations of the Gargano Promontory in an idealized E–W transect. 1 5 inner-platform facies; 2 5 margin facies; 3 5 slope to base-of-slope facies; 4 5 basin facies; 5 5hiatuses.

536

CONTI ET AL.

cluded part of this formation in the San Giovanni Rotondo Limestone (Mattavelli and Pavan, 1965; Pavan and Pirini, 1966; Cremonini et al., 1971). More recent investigations (Luperto Sinni and Masse, 1986; Claps et al., 1996) only examined the Lower Cretaceous succession of inner-platform facies. Luperto Sinni and Masse (1994) proposed that the Sannicandro Formation should be confined to innerplatform facies of Jurassic age, and other units (Formazione di Monte La Serra, Calcari di Sannicandro, and Calcari di Rignano; sensu Cremonini et al., 1971) should be included into the Sannicandro Formation. Based on a Valanginian-age-drowning unconformity (Bosellini and Morsilli, 1997), Morsilli and Bosellini (1997) suggested that the Sannicandro Formation could be extended to include the Berriasian–Early Valanginian interval. The Sannicandro Formation crops out only in the western and central sectors of Gargano. The base of this unit is unknown in outcrop, and it is overlain by the San Giovanni Rotondo Limestone. Its lateral eastern boundary is difficult to map because it passes very gradually into the Monte Spigno Formation. The exposed thickness is estimated to be at least 400–500 m (Morsilli, 1998). The lowermost-known tract of this unit crops out in the area of Rignano Garganico (description after Morsilli, 1998). At the base of the unit, thick beds of whitish, peloidal-bioclastic mudstone to wackestone with oncoids alternate with peloidal-bioclastic packstone beds (10–30 cm thick). This lower part of the unit is frequently dolomitized. At the top of the unit, there are peritidal cycles (1–2 m thick) composed of peloidal wackestone, with localized packstones containing micritic intraclasts, and cryptomicrobial laminites (sensu Demicco and Hardie, 1994) showing a planar or domal shape. The same lithofacies organization has been recognized in the Monte Calvo area near San Giovanni Rotondo, where it is associated with thin beds of oolitic grainstone that become more frequent towards the transition to the adjacent Monte Spigno Formation. In this area, there also are numerous fenestral structures and small lenses of flat-pebble breccia (clasts 3–4 cm). The upper part of the Sannicandro Formation mainly consists of subtidal lithofacies and rare, thin planar-stromatolite beds. The main textures are peloidal wackestone and bioclastic packstone, locally very rich in dasycladalean algae (Campbelliella striata and C. milesi). The microfossil association partially covers the Callovian– Berriasian time interval, but the peritidal facies indicates a Kimmeridgian–Tithonian age. The facies associations of this formation could be referred to an inner-platform setting with protected lagoons, tidal-flats, and supratidal depositional environments, sometimes affected by storm events (Morsilli and Bosellini, 1997). Many of the lithofacies in the Sannicandro Formation are the same as the ones observed on the Mattinata blocks. Dinosaur footprints from this formation have not been found in place, but more research could fill this gap. Therefore, the potential presence of dinosaur footprints or bones in the Monte Spigno Formation, in particular in the transition area with the Sannicandro Formation, cannot be excluded. San Giovanni Rotondo Limestone This unit has been studied in detail both in the type area (Luperto Sinni and Masse, 1986, Claps et al., 1996)

and in the Apricena-Poggio Imperiale area (Luperto Sinni and Masse, 1986). This unit partially corresponds to the Bari Limestone (Azzaroli et al., 1968; Ricchetti, 1975) in terms of age and sedimentary environments, and represents Lower Cretaceous, inner-platform facies of the Apulia Platform. The San Giovanni Rotondo Limestone is a thick succession (500–600 m) of shallow-water limestone that can be subdivided into three members (see Claps et al., 1996 for a more detailed description). Member 1 (Borgo Celano Member of Luperto Sinni and Masse, 1986) consists of a monotonous and acyclic subtidal unit with meters-thick mudstone to wackestone that is intensely bioturbated. This member can be referred to a shallow-subtidal, lagoonal setting. Member 2 (Loferitic Member of Luperto Sinni and Masse, 1986) is a thick, cyclic unit characterized by quasi-periodic alternation of loferitic beds and centimeter-thick layers of green shale. Many lithofacies have been documented in this unit: mudstone–wackestone beds (0.5–2 m thick), intensely bioturbated with occasional black pebbles at the base, green clayey interlayers, planarcryptomicrobial or domal-shape laminites (LLH-type sensu Logan et al., 1964), and thin lenses of ooidal grainstone. Sometimes karst infilling, characterized by reddish shales (terra-rossa-like soils) with floating limestone clasts, disconformably cuts the previous lithofacies. These features can be referred to a tidal-flat setting with subaerial exposure. Member 3 consists of various lithofacies, such as thin-bedded packstone to grainstone with peloids, bioclasts, and micritic intraclasts that are normally graded and evenly laminated. Lags of ostreids, requienids, and nerineids also may occur. The peritidal cycles consist of peloidal–bioclastic packstone to wackestone at the base, and planar stromatolites (20–100 cm) at the top. There are lens-shaped beds (30–70-cm thick) of clast-supported carbonate breccia with clasts of various lithofacies and size (reaching 20–25 cm). The age of the San Giovanni Rotondo Limestone is between Late Valanginian and Early Aptian (Claps et al., 1996). Many dinosaur footprints were discovered in this formation in a quarry near the Borgo Celano section (Gianolla et al., 2000). They occur in three distinct layers of Member 2 of the San Giovanni Rotondo Limestone. Member 2 belongs to the Campanellula capuensis Zone of Late Hauterivian–Early Barremian age (Claps et al., 1996). Altamura Limestone This unit comprises the uppermost track-bearing horizon in the Apulia region. The unit, previously known as Calcari di Casa Lauriola, only crops out in two small areas in the Gargano region south of San Giovanni Rotondo and near Apricena (Merla et al., 1969; Luperto Sinni and Masse, 1986). Laviano and Marino (1996) pointed out the correspondence in terms of facies and age of this unit with the better-known Altamura Limestone. The Altamura Limestone represents the return of a transgressive marine succession after the prolonged mid-Cretaceous emersion of the Apulia platform (D’Argenio et al., 1987; Mindszenty et al., 1995), which is indicated by the presence of bauxitic deposits (Crescenti and Vighi, 1964). In the San Giovanni area, it consists of mudstone to wackestone with thin layers of green shale of late Turonian?–Coniacian age (Luper-


537

to Sinni, 1996). In the Apricena area, it consists of thick beds (1–2.5 m) of peloidal to bioclastic mudstone–wackestone alternating with planar or domal (also LLH) stromatolite beds (Morsilli, 1998). In the upper part of the exposed succession, there are many beds rich in radiolitidae in life position, organized into bouquets or clusters (sensu Ross and Skelton, 1993). This interval in the Apricena area is Late Turonian–Early Senonian in age (Laviano and Marino, 1996; Morsilli, 1998; Morsilli et al., 2002). DISCOVERY AND STUDY OF THE FOOTPRINTS This is not the first case of footprints discovered in a pier. Thus, this study followed the approach previously chosen by Dalla Vecchia and Venturini (1995). Although the pier was built some tens of years ago, attempts were made to trace the source of the material back to the quarries that furnished the blocks by examination of documents from the companies that built the pier. Unfortunately, this was not as successful as it was for the above authors, and a definitive result was not reached because the company documents only mentioned the source area for the material and not the quarry. Consequently, the place of origin of the footprint-bearing material is still unknown. Moreover, the blocks with the footprints are the only three that differ in lithological characteristics from the other hundreds of blocks in the pier. How and why they reached their place in the pier is still a mystery. Subsequently, micropaleontological analyses were attempted, but samples examined in thin section were practically barren and lacked diagnostic paleontological evidence for age determination. Consequently, a detailed study of the sedimentological features of the Mattinata blocks was conducted in order to compare the various lithofacies and sedimentary structures recognized on the blocks with ones present in outcrops that showed similar features. The blocks were designated as MPA, MPB, and MPC, respectively. Blocks MPA and MPC include footprints preserved as natural molds, while MPB reveals the presence of natural casts. Blocks also were labeled on each side in order to have reference points for subsequent studies. The trampled surfaces were carefully mapped (Figs. 3, 4) and each footprint was numbered and drawn; some specimens also were cast. Morphologic and sedimentologic features of blocks MPA and MPC revealed that these two blocks originally were joined together (side B of the MPA block with side D of the MPC block). Contiguous trampled surfaces were found between the two blocks with only a few centimeter-sized fragments missing. No correspondence was found between prints (MPA, MPC) and natural casts (MPB), so the surface of block MPB represents a different portion of a track-bearing surface with respect to MPA and MPC. Nearly 15 m2 of trampled surface were examined. The amount of trampling is quite low (;20%), although it reveals an area where considerable activity occurred. Sedimentological Features of the Blocks A detailed study of the sedimentological features was carried out on the Mattinata blocks. Sedimentological features were observed on polished slabs taken from edges

FIGURE 3—Map of two of the Mattinata blocks. (A) MPA. (B) MPC.

538

CONTI ET AL.

FIGURE 4—Map of the Mattinata block MPB.

and sides of the blocks to obtain measurements and descriptions for stratigraphical logs, as well as to make detailed observations of fresh surfaces of the lithofacies. Thin sections also were made and used for descriptions. Bed-by-bed measurements of the various sides of the blocks revealed great heterogeneity at a centimeter scale, particularly in blocks MPA and MPC. Block MPA: MPA is ;2.5 m long and 1.8 m wide. The average thickness is ;70 cm (Fig. 5A). Side B of block MPA fits very well to side D of block MPC. An erosional surface or other irregularities are present in some layers. A continuous polished slab was obtained by cutting the edge between side A and B (Fig. 6). Well-laminated beds with a planar shape occur at the base. The texture of this limestone is a peloidal wackestone–packstone, with abundant rhombohedral dolomite crystals. An interval with a brecciated texture is visible in this lower part. The breccia is divided into two parts by a small bed showing the same characteristics as the unit previously described. Above the breccia interval, the layers are characterized mainly by very thin laminations that are more-or-less planar to slightly wavy. In the upper part, some layers are not laminated, and other parts are only a little disturbed. The top layer has features typical of cyclic facies from shallow-water settings. These features can be correlated physically to those in block MPC. A very peculiar structure occurs in the upper part of a parallel polished slab (Figs. 6, 7), where the section of the bottom layer and the disturbance of the entire upper layer are clearly visible. This structure is very similar to tridactyl imprints (Dalla Vecchia and Venturini, 1996; Avanzini, 1998; Dini et al.,

FIGURE 5—Side views of the blocks. (A) MPA—side C, block thickness 70 cm; (B) MPB—side B, block thickness 65 cm; (C) MPC—side C, block thickness 75 cm.

1998). In general, this could be referred to as typical dinoturbated structure. Block MPB: MPB has a trapezoidal shape, with the base of the triangle ;1.3 m long and a height of ;2.5 m. The average thickness is ;65 cm (Fig. 5B). Although this block is stored upside down in the museum to show the natural casts, the description here follows the stratigraphic order. Two slabs were cut to obtain a continuous interval; the slabs were then polished and described. The first was taken at the edge between sides A and C (upper interval) and the second at the edge between B and C (lower interval; Fig. 8). A schematic log also was measured on side B (Fig. 9). From the base, corresponding to the tracked layer,


FIGURE 6—Polished slab of the block MPA cut on the edge between side C and D; scale bar 5 20 cm. In this figure, the sedimentary features and the main layers of this block are clearly visible.

539

there is an interval with wavy laminae, sometimes disturbed or poorly visible. A thick interval in which the original texture has been obliterated completely by dissolution and precipitation phenomena (karst features) is visible in the middle part. The top consists of a massive dolomite with evidence of bioturbation. Block MPC: MPC is ;3 m long and 1.5 m wide. The average thickness is ;75 cm (Fig. 5C). Based on certain irregular layers, a detailed reconstruction of some bed geometries has been carried out. Two logs were compiled from the middles of sides A and D. Because these two logs are not representative of the entire heterogeneous block, a draft of the entire geometry of side A has been reconstructed. The main feature is a small, channeled bed that has its long axis oriented between side A and C. It pinches or thins out in the other directions. Two stratigraphic logs were measured on various sides of the blocks. Log A–A’ shows an irregular base with very thin beds of limestone, sometimes with stylolitic boundaries (Fig. 10A). The textures are mainly wackestone with some dolomitic crystals. Some laminae may have been related to a weak traction current. Above this thin interval of limestone (the only part of the block that is not dolomitized), there are dolomitic, millimeter-thick, slightly undulating laminae. The laminae are truncated both laterally and in the upper part by an irregular, channelized, dark-gray, flat-pebble breccia with centimeter-sized clasts (maximum size of 5 cm). The breccia is concentrated at the bottom and top of the bed and is separated by a matrixrich interval. On top of the breccia bed, very thin, undulating laminae have a hemispheroidal shape and flatten out in the upper part of this layer. A poorly laminated or bioturbated interval in the middle part of the succession is followed by a very well defined laminated interval with a wavy to planar shape. In the upper part of the block, there is an interval without lamination containing small depressions or deformations that can be related to pressure on a weak substrate. On the top of block MPC, where the footprints occur, laminations frequently are disturbed by dinoturbation. Log B–B’ on the same block shows a different organization, mainly in the lower part (Fig. 10B). The breccia interval here is very thin and some flat clasts are still in place (mud cracks). Other deformations are very clear and are considered good examples of dinoturbation. The remarkable heterogeneity of this block is apparent

FIGURE 7—Dinoturbated structure. This part corresponds to the top part of the previous figure. The deformation of the layers and the chaotic infill of the depression caused by dinosaur imprint (maximum width 5 40 cm) are clearly visible.

540

CONTI ET AL.

FIGURE 9—Log of MPB measured along the side B. The lower part of this block is characterized by the presence of thin lamination. The middle part shows karst features, while the upper part is comprised of dolomitic limestone without sedimentary structures, probably related to bioturbation of a subtidal interval.

when viewing all of side A (Fig. 11). The most impressive feature is the channeled breccia body that abruptly cuts the lower part of the layers and has a more-or-less flat upper boundary. This body is elongated in the direction of sides A and C, and gives some information about the shape of this bed and its formation (Fig. 12). Paleoenvironmental Interpretation of the Track-Bearing Blocks Some lithological features are useful for reconstructing the depositional environment of the dinosaur footprints. The various lithofacies found in the Mattinata blocks could be interpreted as indicators of a subtidal (shallow-lagoon) setting, and intertidal- to supratidal-flat settings. A small tidal channel is indicated by the lens-shaped, flatpebble breccia bed along side A of Block MPC. Desiccation cracks and the abundance of flat clasts testify to the common occurrence of supratidal conditions. Laminated inter← FIGURE 8—Polished slab of the block MPB cut on the edge between side A and C; scale bar 5 20 cm. The main feature to note is the interval with karst structures (halfway up the slab) that obliterate the original rock texture.


541

vals, interpreted as cryptomicrobial laminites, indicate the presence of algal or microbial mats. Most of the dinosaur footprints, therefore, must be related to an intertidal to supratidal setting. DINOSAUR FOOTPRINTS The Mattinata pier material is composed of 29 natural molds and 8 natural casts; 14 molds and 3 casts are preserved well enough to allow good descriptions. All of the footprints lack convex rims. This seems to confirm the presence of a microbial mat on the track-bearing surface. On blocks MPA and MPC, footprints are obscured in places by laminae that partially fill the more deeply impressed areas. This suggests the footprints were made on the exposed surfaces and are not undertracks. Moreover, overprinting is very rare, and no extramorphological features, such as retro-scratching, reflux of sediment inside the footprints, or sliding traces in elongated footprints were recognized. Consequently, most of footprints correspond well to the underside morphology of the trackmakers’ feet. The better-preserved footprints are discussed below, following Thulborn (1990), Leonardi (1987), and Padian (1992), and can be separated easily into three types. Type 1

FIGURE 10—Logs of block MPC measured along (A) side A and (B) side D. This block is characterized by the presence of a breccia bed with irregular geometric features (see Fig. 11), and the lower part is less dolomitized. The other lithofacies are an alternation of massive and laminated dolostone.

Material referred to this type consists of 12 footprints, nearly 30 cm long, which can be ascribed to a mediumsized bipedal dinosaur. The material includes both natural molds (MPB4, MPB5) and natural casts (MPA1, MPB2, MPC1, MPC2, MPC3, MPC5, MPC11, MPC12, MPC13, MPC19), which allow reconstructing a trackway from incomplete and ambiguous evidence. Three specimens (MPB4, MPB5, MPC2) have a very pronounced metapodium impression, as long (or longer) than digit III, so that the length of these footprints is nearly double that of the other Type 1 specimens. The proportions suggest these impressions are from a large part of the metatarsals made during plantigrade gait or squatting. There are no examples of paired parallel prints, therefore suggesting a sitting posture. This elongation and the hallux impression are evident in four footprints (MPC3, MPC2, MPC12, MPC1) associated in an irregular sequence, evidently made during a very slow progression (Fig. 13). The resulting trackway morphology is quite wide, with short steps, and with a large part of the metatarsals and the hallux touching the ground. Wide-gauge theropod tracks recently were described by Day et al. (2002, 2004) and were associated with a slow-walking phase of locomotion. A partial trackway of two consecutive footprints (MPC19, MPC13) was reconstructed graphically (Fig. 14), and both trackways suggest normal trackmaker progression, not resting phases. This type of gait may be unusual: thus, trackway parameters should be considered with caution. Footprints MPA1 (Fig. 15), MPC3, and MPC11, preserved as natural molds, clearly show phalangeal pads on digits I, II, and III, and less clearly on digit IV, with the digit II metatarsophalangeal pad being diagnostic. MPA 1 seems to show a web-like structure between digits III and IV, most probably due to folding of thin microbial laminae around the footprint. In three cases (MPB4, MPB5, MPC2) the impression of a long and narrow metapodium

542

CONTI ET AL.

FIGURE 11—Drawing of side A of the block MPC. The main feature to note is the shape of the flat-pebble breccia bed, shaded gray in this figure, which is a channelized body with an erosional base. The irregularities in the upper part of this block are related to dinoturbation, with normal print (compare with Fig. 7), and underprint deformation.

FIGURE 12—Detail of the breccia bed of Block MPC (maximum width 25 cm, see Fig. 11 for location).

FIGURE 13—Four consecutive tracks referred to the same dynamic action of an undecided or sliding trackmaker (two tracks of the left pes precede the impression of two tracks of the right one).


543

FIGURE 15—Photograph and outline drawing of Type 1 footprint (MPA1).

Footprints of functionally tridactyl peds with hallux impressions often are present among ornithischian tracks (e.g., Anomoepus and/or Moyenisauropus; see Olsen and Rainforth, 2003), but they also are known in theropod footprints (e.g., Bu¨ckeburgichnus sensu Lockley 2000; Eutynichnium; ‘‘Gigandipus’’; ‘‘Hyphepus’’; Jalingpus; Kayentapus soltykovensis; Picunichnus; Saurexallopus; Theroplantigrada; Tyrannosauripus; and an unnamed theropod from Middle Jurassic of Morocco; Nouri et al., 2000). The hallux often is undervalued as a character (see Harris et al., 1996 for a review) because it is considered a kind of extramorphological feature in typical tridactyl forms, and only is visible in the deepest impressions. A superficial resemblance between Apulian specimens and Anomoepus (related to ornithischian trackmakers) concerns partial metatarsal impressions. However, the Apulian footprints are not the traces of resting animals with both hind limb impressions parallel to each other (Lull, 1904; Gierlinski, 1994, 1996; Avanzini et al., 2001; Lockley et al., 2003). The impression of digit I is a common feature in old figures of Eutynichnium lusitanicum, from the Late Jurassic of Portugal, in which ‘‘depth. . . [10–15 cm]. . . accounts for the preservation of hallux traces’’ (Lockley et al., 2000a, p. 323; but see dos Santos, 2002). A small footprint figured by

FIGURE 14—Footprints (MPC19, MPC13) geometrically completed by a third virtual footprint (MPv) to complete a stride.

also is present, as sometimes happens in squatting dinosaurs. In these cases, the interdigit divergences are larger and much mud was displaced, hiding the metatarsal-phalangeal region (Fig. 16). MPC5, although apparently tridactyl and slightly smaller, could be ascribed to a partial impression of this type.

FIGURE 16—Natural cast of a very elongated footprint (MPB4). Note the central area masked by displaced carbonate mud.

544

Kuban (1989, p. 68, fig. 7.17J) as ‘‘Jalinapus’’ [sic], originally attributed to an ornithischian, but more recently to a theropod (Gierlinski, 1994), also is similar. The type specimen (in Zhen et al., 1983) is a single footprint with a hallux impression in a well-impressed series of 38; it appears to be a rare case within this sample. A partial resemblance in the position of digit I and the narrowing at the base of the digit II also can be recognized in Picunichnus benedettoi from Cenomanian sediments of Argentina (Calvo, 1991). The central posterior elongation and the hallux impression resemble Kayentapus soltykovensis from Lower Jurassic of Hungary (Gierlinski, 1996), but specimens differ in relative proportions and in digit III-IV divergence. Megalosauripus footprints (sensu Lockley et al., 2000a) sometimes show the hallux impression when deeply impressed, although detailed published information is not available. The metapodium, or posterior elongation of the footprints, appears in Mattinata material consistently. Pittman (1989, figs. 15.8, 15.9) illustrated a similar footprint interpreted as a very deep impression, and rejected the hypothesis of an animal walking on its metatarsals. An example of elongate footprints from a walking dinosaur interpreted as the tracks of ‘‘an animal slipping on a wet substrate’’ was described by Kvale (2001, p. 250). Elongate theropod footprints are quite widespread (e.g., Kuban, 1989; Gierlinski, 1994; Pe´rez-Lorente, 1994) and have been described either as elongated, or as footprints of plantigrade dinosaurs. As noted above, most of the footprints, including Kayentapus (sensu Gierlinski 1996), Jalingpus, and Megalosauripus, sometimes show digit I and elongate metapodium impressions. Deep impressions, or a peculiar semidigitigrade and plantigrade gait may explain the origin of both features. The closest fit between the new material and known footprints is with unnamed footprints described and figured by Aguirrezabala and Viera (1980, fig. 24) from Late Jurassic (Kimmeridgian) sediments of Bretun (Soria, Spain). In these footprints, digit I seems to lie in a slightly different plane and is less deeply impressed. The similarity is strong enough to consider ascribing this material to the same trackmaker group. The same material, subsequently figured by Pe´rez-Lorente as trackways C and G (Pe´rez-Lorente, 1994, figs. 1, 2), allows favorable comparison of trackway characters with the Mattinata material. The Spanish trackways slightly differ in having a higher pace angulation (narrower interpedes distance). Theroplantigrada, an ichnotaxon based on Aptian tracks from La Rioja (northern Spain) ascribed to a theropod (Casanovas Cladellas et al. 1994), is similar to the Mattinata material and to trackways described by Aguirrezabala and Viera (1980). This monotypic ichnogenus is based on a single trackway in which, except in size, the footprints seem quite similar to the Mattinata material in the presence and position of the hallux and the metapodium impression. All of the Spanish material shows very elongated footprints. The Apulian material resembles trackway A and some footprints of trackway D (Casanovas Cladellas et al., 1994, figs. 3 and 15, respectively). Casanovas Cladellas et al. (1994) considered the presence of an interdigital web as the main diagnostic feature of Theroplantigrada. This ichnotaxon differs from the ma-

CONTI ET AL.

FIGURE 17—Specimen MPA1. A thin microbial lamina, in front of hipex III–IV, mimics a web-like structure.

terial of Aguirrezabala and Viera (1980) in age, but this interdigital web cannot be considered a defining character. First-hand observation of the Spanish material does not permit confirmation of whether web traces truly are present. Apparent web traces are lacking in all the Mattinata footprints described herein, except in MPA1, just in front of the hypex III-IV (Fig. 17). However, this structure was generated by the inflexion of microbial laminae some centimeters away from the real boundaries of the footprint. If webbing really is present in Theroplantigrada, it would be a rare case among dinosaurs. More likely, it is an extramorphological feature. Only three purported cases were cited by Thulborn (1990)—Otouphephus magnificus, Swinnertonichnus mapperleyensis, and Talmontopus tersi—and in all three cases, the web traces subsequently were interpreted as extramorphological features. For Otouphephus magnificus, ‘‘the web-like trace . . . was later shown to be an artefact’’(Thulborn, 1990, p. 80). Swinnertonichnus mapperleyensis, based on a single tridactyl coelurosaur footprint with web-like traces (Sarjeant, 1967), subsequently lacks web traces in the figures of Haubold (1971, pl. 42, fig. 14), and was reinterpreted as a crocodilian imprint, because ‘‘webbing. . . (is). . . a feature unknown in dinosaur footprints. . . ’’(Sarjeant, 1996, p. 14). The same specimen, re-examined by King and Benton (1996), was ascribed more convincingly to Chirotherium, an ichnotaxon that lacks a web. Regarding Otouphephus and Talmontopus, King and Benton (1996, p. 221) reported


FIGURE 18—Photograph and outline drawing of the best-preserved Type 2 footprint (MPC10).

that ‘‘. . . Lockley (pers. comm., 1994) cannot confirm the presence of webbing in either taxon.’’ In the opinion of some authors, ichnogenera should correspond to footprints revealing diagnostic differences in definite skeletal structures of the trackmakers (Carrano and Wilson, 2001). Consequently, if the supposed webbing in Theroplantigrada is extramorphological, the main diagnostic features of Theroplantigrada would be the presence of hallux and metatarsal impressions that correspond to skeletal structure; the presence of web traces conceivably could be considered a character at the ichnospecies level. If the above rationale is accepted, the Mattinata material, trackways C and G described by Aguirrezabala and Viera (1980), trackway D of Casanovas Cladellas et al. (1994), and the trackway ascribed to Theroplantigrada encisensis all could be included in the same ichnogenus. Consequently, the Mattinata material described here is classified as cf. Theroplantigrada isp. Comparisons are possible with similar tracks. Similar trackways and tracks are known from the Upper Jurassic of Soria, Spain (Aguirrezabala and Viera, 1980); ‘‘Eutynichnium’’ lusitanicum in the Uppermost Jurassic of Portugal; an unnamed trackway from Middle Jurassic deposits of Morocco (Nouri et al., 2000); Jalingpus yuechiensis from the Upper Jurassic of Sichuan, China; Kayentapus soltykovensis from the Lower Jurassic of Poland, Sweden, and Hungary; and Theroplantigrada encisensis from Early Cretaceous beds (probably Aptian in age) of northern Spain. All represent medium-sized dinosaurs that left hallux and metapodium traces. This kind of footprint cannot be ascribed easily to a zoological group below the level of theropod trackmaker or, perhaps, to a cursorial, light ornithopod (Thulborn, 1990; Viera and Torres, 1992; Farlow and Lockley, 1993). Accepting the more probable hypothesis of a theropod trackmaker, it is inferred that these footprints represent a medium-sized, non-derived theropod (a ceratosaur) or a basal tetanuran where digit I is not reduced. Type 2 Only three stout, tridactyl footprints preserved as natural casts (MPC4, MPC10, MPB7) are included in this type (Fig. 18). MPC4 and MPC10, although deeply impressed, lack many details of pads; MPC10 and MPB7

545

FIGURE 19—Outline drawings of other Mattinata footprints. (A) Type 3 footprint (MPC17). (B) Undetermined specimen (MPA3).

show a characteristic, folded proximal margin and welldeveloped claws. Often, similar material has been described under a plethora of names or, more frequently, has remained unnamed, perhaps due to the generalized shape of the prints (e.g., Olsen and Galton, 1984). However, recent papers by Lockley et al. (1996; 2000a) and Olsen et al. (1998) have shown that detailed analyses can allow separation of at least some theropod footprints and trackways. According to older literature, material similar to the Mattinata prints could be referred to Eubrontes or to Megalosauripus (for a review on the debate around this last name and a suggested solution, see Lockley et al., 2000a; Lockley and Meyer, 2000). However, studied specimens lack the socalled heel, or proximal area, made by the impression of metatarsal-phalangeal pad of digit IV. The Mattinata specimens differ in this character from most other theropod footprints, which are characterized by the impression of the metatarsophalangeal pad of digit IV. Two exceptions are Carmelopodus untermannorum, a small footprint from Middle Jurassic deposits of northeastern Utah (Lockley et al., 1988), and Skartopus australis from the Cretaceous of Australia (Thulborn and Wade, 1984). In the diagnosis of Carmelopodus, the ‘‘lack of any impression of a fourth proximal pad on digit IV is stressed’’ (Lockley et al., 1998, p. 260). The Mattinata footprints and the Utah and Australian prints differ in other characters, but it is interesting that this is another example of the lack of such an important feature in mid-Jurassic material. Consequently, the Mattinata specimens may share this functional character with Carmelopodus. Due to the lack of related ichnotaxa, no stratigraphical inferences are possible. It is only possible to note that Carmelopodus comes from Middle Jurassic deposits, Skartopus from the Mid-Cretaceous, and there are similar subdigitigrade theropod tracks in the Lower Jurassic of Poland (Gierlinski and Pienkowski, 1999). Type 3 Only two middle-sized footprints (MPC9, MPC17; natural casts) pertaining to the same partial trackway (Fig. 19A) are ascribed to Type 3 prints. These footprints resemble Therangospodus, a recently formalized ichnogenus attributed to a theropod track-maker that lacks well-defined pads on the digit impression (Lockley et al., 2000b), and

546

CONTI ET AL.

includes only two ichnospecies: T. pandemicus and T. oncalensis. Gierlinski et al. (2001, p. 445) reported a specimen with more distinct phalangeal pads, but stated that they were ‘‘not sure if Therangospodus should be distinguished from Megalosauripus.’’ Specimens that can be compared to the Mattinata material are ascribed to deposits of Late Jurassic age. Therangospodus pandemicus has been described from the Upper Jurassic of the United States (Utah) and Turkmenistan (Lockley and Meyer, 2000), and from the Upper Jurassic of Portugal (Lockley et al., 2000b). The correct age of Therangospodus oncalensis, an ichnospecies first considered Early Cretaceous in age, still seems questionable (see Lockley et al., 2000b). Other Footprints The remaining footprints, although easily recognizable as dinosaur traces, are preserved too poorly to be ascribed to a particular group. MPA5 (Fig. 19B) resembles an ornithopod footprint in the II-IV digit divergence (928), relative shortness of digits, and lack of pads. However, it also shows sharp claw traces and thus could represent a theropod trackmaker. MPC6 and MPA4 could be ascribed to Type 1 traces, and MPC20 and MPC21 likely are two theropod traces in a lightly impressed trackway. This material provides data to evaluate the number of individuals that crossed the area. Notes on the Ichnocoenosis Although the Mattinata pier material still presents many unsolved problems, it allows inferences to be made about paleobiogeography (discussed below), and throws light on the presence of a theropod-dominated track assemblage with inferred correspondence to the assemblage described by Lockley et al. (2000a) and by Lockley and Meyer (2000), which includes Megalosauripus and Therangospodus. Such ichnocoenoses have been recorded from Uzbekistan, Spain, and North America, and are considered characteristic of the early Late Jurassic. In the present case, the dominance of different tracks related to Theroplantigrada is recorded. This difference could reflect age or facies controls affecting the Apulian Platform ichnocoenosis. CONCLUSIONS Age of the Blocks Direct biostratigraphic or chronological calibration of the age of the material is lacking, but three lines of evidence allow inference of the approximate geologic age of the Mattinata blocks: (1) provenance of the blocks, (2) the lithofacies recognized in the blocks, and (3) the footprint types from the blocks. (1) Documents about the pier building are related to the extraction of blocks from quarries in the San Giovanni Rotondo area and Apricena. The former were opened mainly in the San Giovanni Rotondo Limestone, and this unit is never dolomitized. In the Apricena area, the only interval where block extraction continues is in the San Giovanni Rotondo Limestone, where no dolomitic layers are known.

(2) Some beds in small, abandoned quarries near Monte Calvo (Fig. 1) in the Sannicandro Formation reveal similar sedimentological features to the footprint-bearing blocks, and are partially dolomitized. The age of this part of the Sannicandro Formation is referred to the Kimmeridgian–Tithonian interval (Morsilli, 1998). The same lithological and sedimentary characteristics also can be found near the San Nazario area (see Fig. 1) in the northwest part of the Gargano. Here, some old quarries, in some cases dismantled by the construction of a road, reveal very similar lithological and sedimentary features. The age of this succession is the same as the one in the Monte Calvo area (Luperto Sinni and Masse, 1994). (3) Although cautiously avoiding attribution of the footprints to named ichnospecies, the overall character of the assemblage suggests a Late Jurassic age, although these forms were not exclusive to that time interval. In conclusion, these lines of evidence suggest that the studied material comes from the Late Jurassic Sannicandro Formation. Paleogeographic Inferences and Problems During Middle–Late Jurassic and Cretaceous times, the Apulian Platform has been interpreted as a small, isolated carbonate platform, surrounded by the Tethys Ocean, separated from other peri-Adriatic carbonate shelves (e.g., the Laziale—Abruzzese, Campana, Sazani, and Kruja carbonate platforms) by deep-sea areas. It is inside the ApuloDinaric structural unit, and is isolated from both the southern and northern continents by deeper basins (D’Argenio, 1976; Zappaterra, 1990; Eberli et al., 1993; Masse et al., 1993). This model obviously was constructed before repeated discoveries of dinosaur footprints on the Apulian Platform. Recent finds of dinosaur trackways from different places and stratigraphical intervals are very strong paleogeographic constraints that necessitate reconsideration of previous interpretations. At first, this new evidence may be considered weaker than geophysical or structural data, but dinosaur footprints may serve as powerful new tools to change the best-built paleogeographic models. Other workers have pointed out that paleogeographic reconstructions must account for dinosaur footprints. For example, Meyer and Lockley (1997, p. 425) stated that ‘‘recurrent emergent areas must have been present that connected the southern part of the London-Brabant mass with the northeastern part of the Massif Central’’; Kvale et al. (2001, p. 233) claimed that ‘‘a major change in the paleogeographic reconstructions for Wyoming’’ is needed; and a connection of ‘‘emerged area with larger land masses’’ in the Holy Cross Mountains (Poland) was suggested for the Late Jurassic by Gierlinski and Niedzwiedzki (2002, p. 58A). At present, three different track assemblages in the Apulian region have been found. The first, described herein, is a theropod-dominated ichnoassociation, probably Late Jurassic in age. The second, described from the Early Cretaceous (Late Hauterivian–Early Barremian) of Borgo Celano (Gianolla et al., 2000), contains footprints ascribed to theropods, ornithopods, and perhaps sauropods; and the third, described from Late Cretaceous (Santonian) from Altamura (Andreassi et al., 1999; Nicosia et al.,


2000a, b), is ornithopod dominated. These are the most striking evidence of widespread, terrestrial vertebrates in the area. Other data, such as the so-called Ruvo varanoid (Varola, 1999) and vertebrate remains recorded from the Melissano Limestone (Medizza and Sorbini, 1980), also are evidence of terrestrial vertebrates in this area. Finds of bauxite levels (D’Argenio et al., 1987) and land-plant remains (De Cosmo and Morsilli, 2002; Morsilli et al., 2002) also are evidence of soil development and emergent land. These data span different time intervals, and show different evolutionary levels consistent with biological events recognized elsewhere. Moreover, this data set probably is biased by taphonomic events and by reduced availability of Jurassic and Cretaceous rock outcrops with track-bearing potential. This kind of research needs extensive, undisturbed, well-exposed bedding surfaces, a situation not found in Mesozoic deposits in this part of southern Italy. Available data record three land-vertebrate assemblages in the area, but they must be analyzed cautiously. If the trackways only are exceptional occurrences, they could be related to short-term, repeated connections to larger land areas. On the other hand, if they are considered a rare record of normal events, then continuous, long-lasting colonization by land dwellers must be hypothesized. From a paleontological point of view, these hypotheses correspond either to repeated immigration pulses, or to a single, early colonization followed by endemic evolution. The first hypothesis implies a continuous connection and ongoing immigration between the Apulian Platform and large continental areas. Thus, the carbonate-shelf shallow-sea areas were not separated by deep basins, and sea-level drops would have allowed emergence of the platform followed by land-vertebrate colonization. In the source area, evolutionary processes could proceed normally, with the immigrants reflecting worldwide evolutionary trends. This explanation may be easier to accept because it does not need complex evolutionary hypotheses. The second hypothesis must, in turn, be subdivided to two sub-hypotheses: either the platform was as small as current models suggest, with different portions of the inner platform emerging from time to time, or the emergent area was much larger than thought, although no traces of it are preserved. In the first scenario, lacking any evidence of a single, persistent island area large enough to sustain a balanced community of huge animals, land vertebrates must have been wandering continuously from one zone to another within the same platform area—an idea that does not consider the need of dinosaurs for vegetation, fresh water, and nesting sites. Simple emergent episodes, even if repeated and prolonged, thus are not sufficient to explain the data. In the second case, the scenario must include a large, isolated, complex environment (including fresh-water sources and nesting sites), in which the persistence of a balanced fauna was possible for ;130 Ma, with all traces of such a large area removed by subsequent geological events. In both cases, balanced ecosystems imply the presence and the coevolution of plants, herds of planteating animals, and meat-eating dinosaurs. It also must be assumed that a complete suite of scavengers and invertebrate organisms, as well as the proper physical, chemical, and biological conditions, were present to allow evolutionary processes. The carbonate-platform environment raises particular difficulties in this regard because of lim-

547

ited vegetation productivity, underdeveloped river systems, and soils that need a long time to form from the residuals of karstification. On the other hand, the lack of river environments also could explain the scattered presence of fossils. In fact, a lack of river systems could have reduced the chance for fossilization by reducing deposits available to preserve dinosaur bones and/or footprints. Both cases included in the second hypothesis present another important concern. The recorded Apulian dinosaur ichnocoenoses apparently differ from one another, but may correspond in both the faunal composition and the evolutionary level to age-equivalent dinosaur communities known from Europe, central Asia, and North America. The recognized evolutionary events appear to correspond to events recognized in Europe, in central Asia, and in North America. It is difficult to accept a hypothesis of parallel evolution on the mainlands and isolated carbonate-platform areas. In any case, the presence of a large number of dinosaurs in an area such as the Apulian Platform is problematic. It underlines the inadequacy of paleogeographic models that simply suggest a tectonic disjunction of the platforms. On the contrary, the footprint and other data strongly support a scenario that discards the hypothesis of many small, peri-Adriatic carbonate shelves and interposed basins, and suggests structural continuity and frequent (or continuous) connections of the platform to the mainlands. Consequently, previous hypotheses of dwarf dinosaur faunas that originated by endemic evolutionary phenomena (Dalla Vecchia and Tarlao, 2000; Dalla Vecchia et al., 2002) seem insufficient to explain the evidence fully. If recognized, they might be considered partial explanations or just the record of the final phase of endemic evolution after immigration. Bosellini (2002) used these new data for strong paleogeographic constraints. Reviewing various geological and geophysical data associated with the presence of dinosaurs around the Ionian Sea and surrounding areas, Bosellini (2002) reached the conclusion that the Apulian Platform was connected during the Jurassic and Cretaceous to Africa through the Peloponnesus, Crete, the Cyrene Seamount, and the Medina Ridge. Gierlinski (pers. comm., 2002) also has found similar tracks in the Mesozoic carbonates in Crete. In Bosellini’s (2002) reconstruction, the Apulian Platform, and probably other peri-Adriatic platforms, were not isolated Bahamian banks, but rather were more like the modern Florida peninsula. This model conflicts strongly with other paleogeographic models (e.g., Catalano et al., 2001). In fact, using these data, an eastern connection to the mainland cannot be dismissed. Similarly, more than one immigration route could be hypothesized for these dinosaurs at different stratigraphical levels. Future reconstructions of the area must integrate all data sources, such as geophysical (seismic refraction and reflection profiles of the Apulian Platform margin and surrounding basins), structural and kinematic (relationships between thrust-belt chains and foreland areas), stratigraphical, sedimentological, and paleontological data. ACKNOWLEDGEMENTS Marco Avanzini is warmly thanked for his suggestion about dinoturbation. G. Gierlinski, M. Lockley, and P. Up-

548

CONTI ET AL.

church revised the manuscript and gave important suggestions. Prof. G. Andreassi of the Soprintendenza Archeologica per le Puglie and the Mayor of Mattinata kindly helped our work. REFERENCES AGUIRREZABALA, L.M., and VIERA, L.I., 1980, Icnitas de Dinosaurios en Bretun (Soria): Munibe, v. 32, p. 257–279. ANDREASSI, G., CLAPS, M., SARTI, M., NICOSIA, U., and VENTURO, D., 1999, The Late Cretaceous dinosaur tracksite near Altamura (Bari), Southern Italy: Geoitalia 1999, 28 Forum Italiano di Scienze della Terra, Bellaria 20–23 Settembre; Riassunti, v. 1, p. 28. AVANZINI, M., 1998, Anatomy of a footprint: bioturbation as a key to understanding dinosaur walk dynamics: Ichnos, v. 6, p. 129–139. AVANZINI, M., GIERLINSKI, G., and LEONARDI, G., 2001, First report of sitting Anomoepus tracks in European Lower Jurassic (Lavini di Marco Site-Northern Italy): Rivista Italiana di Paleontologia e Stratigrafia, v. 107, p. 131–136. AZZAROLI, A., RADINA, B., RICCHETTI, G., VALDUGA, U., 1968, Note illustrative della Carta Geologica d’Italia alla scala 1:100.000, Foglio 189 ‘‘Altamura’’: Servizio Geologico d’Italia, Roma, 22 p. BERNOULLI, D., 1972, North Atlantic and Mediterranean Mesozoic facies, a comparison: in Hollister, C.D., and Ewing, J.I., eds., Initial Reports of the Deep Sea Drilling Project, v. 11, p. 801–807. BERTOTTI, G., CASOLARI, E., and PICOTTI, V., 1999, The Gargano Promontory: a Neogene contractional belt within the Adriatic plate: Terra Nova, v. 11, p. 168–173. BOSELLINI, A., 2002, Dinosaurs ‘‘re-write’’ the geodynamics of the eastern Mediterranean and the paleogeography of the Apulia Platform: Earth-Science Review, v. 59, p. 211–234. BOSELLINI, A., and MORSILLI, M., 1997, A Lower Cretaceous drowning unconformity on the eastern flank of the Apulia Platform (Gargano Promontory, southern Italy): Cretaceous Research, v. 18, p. 51–61. BOSELLINI, A., and MORSILLI, M., 2001, Il Promontorio del Gargano, cenni di geologia e itinerari geologici: Quaderni del Parco Nazionale del Gargano, 50 p. BOSELLINI, A., MORSILLI, M., and NERI, C., 1999, Long-term event stratigraphy of the Apulia Platform margin: Upper Jurassic to Eocene, Gargano, Southern Italy: Journal of Sedimentary Research, v. 69, p. 1241–1252. BOSELLINI, A., NERI, C., and LUCIANI, V., 1993, Platform margin collapses and sequence stratigraphic organization of carbonate slopes: Cretaceous–Eocene, Gargano Promontory: Terra Nova, v. 5, p. 282–297. CALVO, J.O., 1991, Huella de dinosaurios en la Formacion Rio Limay (Albiano–Cenomaniano?), Picun Leufu, Provincia de Neuquen, Republica Argentina (Ornithischia-Saurischia: Sauropoda-Theropoda): Ameghiniana, v. 28, p. 241–258. CARRANO, M.T., and WILSON, J.A., 2001, Taxon distributions and the tetrapod track record: Paleobiology, v. 27, p. 564–582. CASANOVAS CLADELLAS, M.L., EZQUERRA MIGUEL, R., FERNAŃDEZ ORTEGA, A., PE´REZ-LORENTE, F., SANTAFE´ LLOPIS, J.V., and TOR´ NDEZ, F., 1994, Icnitas digitı´gradas y plantı´gradas de CIDA FERNA dinosaurios en el afloramento de El Villar-Poyales (La Rioja, Espanã): Zubıá (Monogra´fico), no. 5 (1993), p. 135–163. CATALANO, R., DOGLIONI, C., and MERLINI, S., 2001, On the Mesozoic Ionian Basin: Geophysical Journal International, v. 144, p. 49–64. CLAPS, M., PARENTE, M., NERI, C., and BOSELLINI, A., 1996, Facies and cycles of the S. Giovanni Rotondo Limestone (Lower Cretaceous, Gargano Promontory, Southern Italy): the Borgo Celano section: Annali Universita` di Ferrara, Sezione di Scienze della Terra, v. 7, p. 1–35. CREMONINI, G., ELMI, C., and SELLI, R., 1971 Note illustrative della Carta Geologica d’Italia alla scala 1:100.000, Foglio 156 ‘‘S. Marco in Lamis’’: Servizio Geologico d’Italia, Roma, 66 p. CRESCENTI, U., and VIGHI, L., 1964, Caratteristiche, genesi e stratigrafia dei depositi bauxitici cretacici del Gargano e delle Murge; cenni sulle argille con pisoliti bauxitiche del Salento: Bollettino della Societa` Geologica Italiana, v. 83, p. 285–337. DALLA VECCHIA, F.M., and TARLAO, A., 2000, New dinosaur track

sites in the Albian (Early Cretaceous) of the Istrian peninsula (Croatia). Part II, Paleontology: in Dalla Vecchia, F.M., Tarlao, A., Tunis, G., and Venturini, S., 2000, New Dinosaur Track Sites in the Albian (Early Cretaceous) of the Istrian Peninsula (Croatia): Memorie di Scienze Geologiche, v. 52, p. 227–292. DALLA VECCHIA, F.M., and VENTURINI, S., 1995, A theropod (Reptilia, Dinosauria) footprint on a block of Cretaceous limestone at the pier of Porto Corsini (Ravenna, Italy): Rivista Italiana di Paleontologia e Stratigrafia, v. 101, p. 93–98. DALLA VECCHIA, F.M., and VENTURINI, S., 1996, Le possibili impronte di dinosauro del M. Bernadia e le potenzialita` paleoicnologiche delle sezioni stratigrafiche: Natura Nascosta No. 12, 34–44. DALLA VECCHIA, F.M., VLAHOVIC, I., POSOCCO, L., TARLAO, A., and TENTOR, M., 2002, Late Barrremian and Late Albian (Early Cretaceous) dinosaur track sites in the Main Brioni/Brijun Island (SW Istria, Croatia): Natura Nascosta, v. 25, p. 1–36. DAL SASSO, C., 2003, Dinosaurs of Italy: Comptes Rendus Palevol, v. 2, p. 45–66. DAY, J.J., NORMAN, D.B., GALE, A.S., UPCHURCH P., and POWELL, H.P., 2004, A new Middle Jurassic dinosaur trackway from Oxfordshire: Palaeontology, v. 47, p. 319–348. DAY, J.J., NORMAN, D.B., UPCHURCH P., and POWELL, H.P., 2002, Dinosaur locomotion from a new trackway: Nature, v. 415, p. 494– 495. DE COSMO, P.D., and MORSILLI, M., 2002, Segnalazione di resti di flora ed ittiofauna fossile del Santoniano superiore nell’area di Apricena (Puglia, Gargano): Gruppo informale di Sedimentologia, IX Riunione annuale, Pescara 21–22 Ottobre, p. 39–40. D’ARGENIO, B., 1976, Le piattaforme periadriatiche: una rassegna di problemi nel quadro geodinamico mesozoico dell’area mediterranea: Memorie della Societa` Geologica Italiana, v. 13, p. 137–159. D’ARGENIO, B., MINDSZENTY, A., BARDOSSY, G.Y., JUHASZ, E., and BONI, M., 1987, Bauxites of southern Italy revisited: Rendiconti della Societa` Geologica Italiana, v. 9, p. 263–268. DE DOMINICIS, A., and MAZZOLDI, G., 1989, Interpretazione geologicostrutturale del margine orientale della piattaforma Apula: Memorie della Societa` Geologica Italiana, v. 38, p. 163–176. DEMICCO, R.V., and HARDIE, L.A., 1994, Sedimentary Structures and Early Diagenetic Features of Shallow Marine Carbonate Deposits: Society of Economic Paleontologist and Mineralogists Atlas Series 1, 265 p. DINI, M., TUNIS, G., and VENTURINI, S., 1998, Continental, brackish and marine carbonates from the Lower Cretaceous of Kolone-Barbariga (Istria, Croatia): stratigraphy, sedimentology and geochemistry: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 140, p. 245–269. DOS SANTOS, V.F., 2002, Pistas de Dinosaurio del Jura ´sico–Cretaćico de Portugal. Consideraciones Paleobiolo´gicas y Paleoecolo´gicas: Tesis Doctoral, Univesidad Autońoma de Madrid, 365 p. EBERLI, G.P., BERNOULLI, D., SANDERS, D., and VECSEI, A., 1993, From aggradation to progradation: the Maiella Platform, Abruzzi, Italy: in Simo, J.A.T., Scott, R.W., and Masse, J.-P., eds., Cretaceous Carbonate Platforms: American Association of Petroleum Geologists Memoir 56, p. 213–232. FARLOW, J.O., and LOCKLEY, M.G., 1993, An osteometric approach to the identification of the makers of Early Mesozoic tridactyl dinosaur footprints: in Lucas, S.G., and Morales, M., eds., The Nonmarine Triassic: New Mexico Museum of Natural History and Science Bulletin, v. 3, p. 123–131. FUNICIELLO, R., MONTONE, P., SALVINI, F., and TOZZI, M., 1992, Caratteri strutturali del Promontorio del Gargano: Memorie della Societa` Geologica Italiana, v. 41, p. 861–868. GIANOLLA, P., MORSILLI, M., and BOSELLINI, A., 2000a, First discovery of Early Cretaceous dinosaur footprints in the Gargano Promontory (Apulia carbonate platform, southern Italy): quantitative models on Cretaceous carbonates and the eastern margin of the Apulia Platform: SEPM (Society For Sedimentary Geology) Cretaceous Resource Events and Rhythms Working Group 4 Conference, Vieste-Gargano, Italy, 25–28 September, Abstracts, p. 9. GIANOLLA, P., MORSILLI, M., and BOSELLINI, A., 2000b, Dinosaur footprints: in Bosellini, A., Morsilli, M., and Neri, C., eds., Guide Book: The Eastern Margin of the Apulia Platform: The Gargano Transect: SEPM (Society For Sedimentary Geology) Cretaceous Re-

DINOSAUR FOOTPRINTS FROM SOUTHERN ITALY source Events and Rhythms Working Group 4 Conference, ViesteGargano, Italy, September 2000, p. 34–36. GIERLINSKI, G., 1991, New dinosaur ichnotaxa from the Early Jurassic of the Holy Cross Mountains, Poland: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 85, p. 137–148. GIERLINSKI, G., 1994, Early Jurassic theropod tracks with the metatarsal impressions: Prezglad Geologiczny, v. 42, p. 280–284. GIERLINSKI, G., 1996, Dinosaur ichnotaxa from the Lower Jurassic of Hungary: Geological Quarterly, v. 40, p. 119–128. GIERLINSKI, G., and NIEDZWIEDZKI, G., 2002, Dinosaur tracks in the Late Jurassic of Poland: Journal of Vertebrate Paleontology, v. 22,3, p. 58A. GIERLINSKI, G., NIEDZWIEDZKI, G. and PIENKOWSKI, G., 2001, Gigantic footprint of a theropod dinosaur in the Early Jurassic of Poland: Acta Paleontologica Polonica, v. 46, p. 441–446. GIERLINSKI, G., and PIENKOWSKI, G., 1999, Dinosaur track assemblages from the Hettangian of Poland: Geological Quarterly, v. 43, p. 329–346. HARRIS, J.D., 1997, Four-toed theropod footprints and paleomagnetic age from the Whetstone Falls Member of the Harebell Formation (Upper Cretaceous: Maastrichtian), northwestern Wyoming: a correction: Cretaceous Research, v. 18, p. 139. HARRIS, J.D., JOHNSON, K.R., HICKS, J., and TAUXE, L., 1996, Fourtoed theropod footprints and a paleomagnetic age from the Whetstone Falls Member of the Harabell Formation (Upper Cretaceous: Maastrichtian), northwestern Wyoming: Cretaceous Research, v. 17, p. 381–401. HAUBOLD, H., 1971, Ichnia Amphibiorum et Reptiliorum fossilium: in Kuhn, O., ed., Handbuch der Palaeoherpetologie: Fischer-Verlag, Stuttgart, v. 18, 124 p. KING, M.J., and BENTON, M.J., 1996, Dinosaurs in the Early and Mid Triassic? The footprint evidence from Britain: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 122, p. 213–225. KVALE, E.P., JOHNSON, G.D., MICKELSON, D.L., KELLER, K., FURER, L.C., and ARCHER, A.W., 2001, Middle Jurassic (Bajocian and Bathonian) dinosaur megatracksites, Bighorn Basin, Wyoming, U.S.A.: PALAIOS, v. 16, p. 233–254. KUBAN, G.J., 1989, Elongate dinosaur tracks: in Gillette, D.D., and Lockley, M.G., eds., Dinosaur Tracks and Traces: Cambridge University Press, Cambridge, p. 57–72. LAVIANO, A., and MARINO, M., 1996, Biostratigraphy and paleoecology of Upper Cretaceous carbonate successions in the Gargano Promontory: Memorie della Societa` Geologica Italiana, v. 51, p. 685–701. LEONARDI, G., 1987, Glossary and Manual of Tetrapod Footprint Paleoichnology: Departamento Nacional da Produçaõ Mineral, Brasilia, 117 p. LOCKLEY, M.G., 2000, An amended description of the theropod footprint Bueckeburgichnus maximus Kuhn 1958, and its bearing on the Megalosaur tracks debate: Ichnos, v. 7, p. 217–225. LOCKLEY, M.G., and HUNT, A., 1994, Tracks of the giant theropod dinosaur Tyrannosaurus from close to the Cretaceous/Tertiary boundary, northern New Mexico: Ichnos, v. 3, p. 213–218. LOCKLEY, M.G., HUNT, A., PAQUETTE, M., BILBEY, S., and HAMBLIN, A., 1998, Dinosaur tracks from the Carmel Formation, Northeastern Utah: implications for Middle Jurassic paleoecology: Ichnos, v. 5, p. 255–267. LOCKLEY, M.G., MATSUKAWA, M., and JANJUN, L., 2003, Crouching theropods in taxonomic jungles: ichnological and ichnotaxonomic investigations of footprints with metatarsal and ischial impressions: Ichnos, v. 10, p. 169–177. LOCKLEY, M.G., and MEYER, C.A., 2000, Dinosaur Tracks and Other Fossil Footprints of Europe: Columbia University Press, New York, 322 p. LOCKLEY, M.G., MEYER, C.A., and DOS SANTOS, V.F., 1996, Megalosauripus, Megalosauropus and the concept of megalosaur footprints: in Morales, M., ed., The Continental Jurassic: Symposium Volume: Museum of Northern Arizona Bulletin 60, p. 113–118. LOCKLEY, M.G., MEYER, C.A., and DOS SANTOS, V F., 2000a, Megalosauripus and problematic concept of megalosaur footprints: Gaia, v. 15, p. 313–337. LOCKLEY, M.G., MEYER, C.A., and MORATALLA, J.J., 2000b, Therangospodus: trackway evidence for the widespread distribution of a

549

Late Jurassic theropod with well-padded feet: Gaia, v. 15, p. 339– 353. LOGAN, B.W., REZAK, R., and GINSBURG, R.N., 1964, Classification and environmental significance of algal stromatolites: Journal of Geology, v. 72, p. 68–83. LULL, S.R., 1904, Fossil footprints of the Jura–Trias of North America: Memoirs of the Boston Society of Natural History, v. 5, p. 461– 557. LUPERTO SINNI, E., 1996, Schema stratigrafico del Cretacico del Gargano basato su risultati di recenti ricerche: Memorie della Societa` Geologica Italiana, v. 51, p. 1019–1036. LUPERTO SINNI, E., and MASSE, J.P., 1986, Donnee´s nouvelles sur la stratigraphie des calcaires de plate-forme du Cre´tace´ infe´rieur du Gargano: Rivista Italiana di Paleontologia e Stratigrafia, v. 92, p. 33–66. LUPERTO SINNI, E., and MASSE, J.P., 1994, Precisazioni micropaleontologiche sulle formazioni di piattaforma carbonatica del Giurassico superiore e del Cretaceo basale del massiccio del Gargano (Italia Meridionale) e implicazioni stratigrafiche: Paleopelagos, v. 4, p. 243–266. MARTINIS, B., 1965, Osservazioni sulla tettonica del Gargano orientale: Bollettino del Servizio Geologico d’Italia, v. 85, p. 45–93. MARTINIS, B., and PAVAN, G., 1967, Note illustrative della carta geologica d’Italia alla scala 1:100.000, Foglio 157 ‘‘Monte S. Angelo’’: Servizio Geologico d’Italia, 56 p. MASSE, J.P., BELLION, Y., BENKHELIL, J., BOULIN, J., CORNEE, J.J., DERCOURT, J., GUIRAUD, R., MASCLE, G., POISSON, A., RICOU, L.E., and SANDULESCU, M., 1993, Lower Aptian (114 to 112 Ma): in Dercourt, J., Ricou, L.E., and Vrielynck, B., eds., Atlas Tethys Palaeoenvironmental Maps, Explanatory Notes: Gauthier-Villars, Paris, p. 135–152. MASSE, J.P., and BORGOMANO, J., 1987, Un mode`le de transition plate-forme-bassin carbonate´s controˆle´ par des pheńome`nes tectoniques: le Cre´tace´ du Gargano (Italie Me´ridionale): Comptes Rendues de l’Academie des Sciences de Paris, v. 10, p. 521–526. MATTAVELLI, L., and PAVAN, G., 1965, Studio petrografico delle facies carbonate del Gargano: Rendiconti Societa` Mineralogica Italiana, v. 21, p. 207–251. MEDIZZA, F., and SORBINI, L., 1980, Il giacimento del Salento (Lecce): in I Vertebrati Fossili Italiani—Catalogo della Mostra: Verona, p. 131–134. MERLA, G., ERCOLI, A., and TORRE, D., 1969, Note illustrative della Carta Geologica d’Italia, alla scala 1:100.000, F8 164 ‘‘Foggia’’: Servizio Geologico d’Italia, 22 p. MEYER, C.A., and LOCKLEY, M.G., 1997, Jurassic and Cretaceous dinosaur tracksites from Central Asia (Uzbekistan and Turkmenistan): The Paleontological Society of Korea, Special Publication v. 2, p. 77–92. MINDSZENTY, A., D’ARGENIO, B., and AIELLO, G., 1995, Lithospheric bulges recorded by regional unconformities. The case of Mesozoic– Tertiary Apulia: Tectonophysics, v. 252, p. 137–161. MORSILLI, M., 1998, Stratigrafia e sedimentologia del margine della Piattaforma Apula nel Gargano (Giurassico superiore–Cretaceo inferiore): Librerie Nazionali di Roma e Firenze, Tesi di Dottorato, Universita` di Bologna, 203 p. MORSILLI, M., and BOSELLINI, A., 1997, Carbonate facies zonation of the Upper Jurassic–Lower Cretaceous Apulia Platform margin (Gargano Promontory, southern Italy): Rivista Italiana di Paleontologia e Stratigrafia, v. 103, p. 193–206. MORSILLI, M., DE COSMO, P.D., BOSELLINI, A., and LUCIANI, V., 2002, L’annegamento Santoniano della piattaforma Apula nell’area di Apricena (Gargano, Puglia): nuovi dati per la paleogeografia del Cretaceo superiore: Gruppo informale di Sedimentologia, IX Riunione annuale, Pescara 21–22 Ottobre, p. 65–66. NICOSIA, U., MARINO, M., MARIOTTI, N., MURARO, C., PANIGUTTI, S., PETTI, F.M., and SACCHI, E., 2000a, The Late Cretaceous dinosaur tracksite near Altamura (Bari, Southern Italy) I—geological framework: Geologica Romana, v. 35, p. 231–236. NICOSIA, U., MARINO, M., MARIOTTI, N., MURARO, C., PANIGUTTI, S., PETTI, F.M., and SACCHI, E., 2000b, The Late Cretaceous dinosaur tracksite near Altamura (Bari, Southern Italy) II—Apulosauripus federicianus new ichnogen. and new ichnosp.: Geologica Romana, v. 35, p. 237–247.

550 NOURI, J., PE´REZ-LORENTE, F., and BOUTAKIOUT, M., 2000, Descubrimiento de una pista semiplantı´grada de dinosaurio en el yacimiento de Tirika (Demnat. Alto Atlas central marroquı´): Geogaceta, v. 29, p. 111–114. OLSEN, P.E., and GALTON, P.M., 1984, A review of the reptile and amphibian assemblages from the Stormberg of southern Africa, with special emphasis on the footprints and the age of the Stormberg: Palaeontologia Africana, v. 25, p. 87–110. OLSEN, P.E., and RAINFORTH, E.C., 2003, The Early Jurassic ornithischian dinosaurian ichnogenus Anomoepus: in LeTourneau, P.M., and Olsen, P.E., eds., The Great Rift Valleys of Pangea in Eastern North America, Volume 2, Sedimentology and Paleontology: Columbia University Press, New York, p. 314–368. OLSEN, P.E., SMITH, J.B., and MCDONALD, N.G., 1998, Type material of the type species of the classic theropod footprint genera Eubrontes, Anchisauripus, and Grallator (Early Jurassic, Hartford and Deerfield Basins, Connecticut and Massachusetts, U.S.A.): Journal of Vertebrate Paleontology, v. 18, p. 586–601. PADIAN, K., 1992, A proposal to standardize tetrapod phalangeal formula designations: Journal of Vertebrate Paleontology, v. 12, p. 260–262. PAVAN, G., and PIRINI, C., 1966, Stratigrafia del Foglio 157 ‘‘Monte S. Angelo’’: Bollettino del Servizio Geologico d’Italia, v. 86, p. 123– 189. PE´REZ-LORENTE, F., 1994, Dinosaurios plantı´grados en la Rioja: Zubıá (Monogra´fico), no. 5, p. 189–228. PITTMAN, J.G., 1989, Stratigraphy, lithology, depositional environment and track type of dinosaur track-bearing beds of the Gulf Coastal Plain: in Gillette, D.D., and Lockley, M.G., eds., Dinosaur Tracks and Traces: Cambridge University Press, Cambridge, p. 135–153. RICCHETTI, G., 1975, Nuovi dati stratigrafici sul Cretaceo delle Murge

CONTI ET AL. emersi da indagini nel sottosuolo: Bollettino della Societa` Geologica Italiana, v. 94, p. 1083–1108. ROSS, D.J., and SKELTON, P.W., 1993, Rudist formations of the Cretaceous: a palaeoecological, sedimentological and stratigraphical review: in Wright, V.P., ed., Sedimentology Review 1, Blackwell Science Publication, London, p. 73–91. SARJEANT, W.A.S., 1967, Fossil footprints from the Middle Triassic of Nottinghamshire and Derbyshire: The Mercian Geologist, v. 2, p. 327–341. SARJEANT, W.A.S., 1996, A re-appraisal of some supposed dinosaur footprints from the Triassic of the English Midlands: The Mercian Geologist, v. 14, p. 22–30. THULBORN, R.A, 1990, Dinosaur Tracks: Chapman and Hall, London, 410 p. THULBORN, R.A., and WADE, M., 1984, Dinosaur trackways in the Winton Formation (mid-Cretaceous) of Queensland: Memoirs of the Queensland Museum, v. 21, p. 413–517. VAROLA, A., 1999, Vertebrati: Pesci e Rettili: in Ricchetti, G., and Pieri, P., eds., Guide Geologiche Regionali, 9 Itinerari, Puglia e Monte Vulture, Prima Parte, a Cura della Societa` Geologica Italiana: BE-MA Editrice, p. 43–45 VIERA, L.I., and TORRES, J.A., 1992, Sobre ‘‘Dinosaurios Coelu´ridos gregarios en el Yacimianto de Valdevajes (La Rioja, Espanã)’’. Nota de re´plica y crı´tica: Revista Espanõla de Paleontologia, v. 7, p. 93–96. ZAPPATERRA, E., 1990, Carbonate paleogeographic sequence of the periadriatic region: Bollettino della Societa` Geologica Italiana, v. 109, p. 5–20. ZHEN, S., LI, J., and ZHEN, B., 1983, Dinosaur footprints of Yuechi, Sichuan: Beijing Natural History Museum Memoirs, v. 25, p. 1– 19.

ACCEPTED DECEMBER 28, 2004