400 J. Paleont., 80(2), 2006, pp. 400–406 Copyright  2006, The Paleontological Society 0022-3360/06/0080-400$03.00 NEW INFORMATION REGARDING THE HOLOTYPE OF SPINOSAURUS AEGYPTIACUS STROMER, 1915 JOSHUA B. SMITH, 1 MATTHEW C. LAMANNA, 2 * HELMUT MAYR, 3 AND KENNETH J. LACOVARA 4 1 Department of Earth and Planetary Sciences, Washington University, 1 Brookings Drive, Campus Box 1169, St. Louis, Missouri 63130-4899, USA, smithjb@wustl.edu, 2 Department of Earth and Environmental Science, University of Pennsylvania, 240 South 33rd Street, Philadelphia 19104-6316, USA, 3 Pala ¨ontologisches Museum, Bayerische Staatssammlung fu ¨r Pala ¨ontologie und Geologie, Richard-Wagner-Strasse 10/II, D-80333 Mu ¨nchen, Germany, and 4 Department of Biosciences and Biotechnology, Drexel University, 32nd and Chestnut Streets, Philadelphia, Pennsylvania 19104, USA INTRODUCTION I N THE autumn of 1912, the fossil collector Richard Markgraf, with financial support and direction from Bavarian paleontol- ogist Ernst Freiherr Stromer von Reichenbach and the Bavarian Academy of Sciences, discovered the partial skeleton of a bizarre predatory dinosaur in Upper Cretaceous (early Cenomanian, 97 Ma, see Ismail et al., 1989; Barakat et al., 1993; El Beialy, 1994, 1995; Nabil and Hussein, 1994; Ismail and Soliman, 2001; Ibra- him, 2002; Gradstein et al., 2005) rocks of the Bahariya Forma- tion exposed in the Bahariya Oasis of western Egypt (Fig. 1, see also Sereno et al., 1998; Nothdurft et al., 2002). This gigantic theropod, Spinosaurus aegyptiacus Stromer, 1915, possessed highly derived cranial and vertebral features sufficiently distinct for it to be designated as the nominal genus of the clade Spino- sauridae (Stromer, 1915, 1936). Spinosaurids, currently defini- tively known only from Europe, South America, and Africa, are important because of the scarcity of Cretaceous Gondwanan tet- rapod fossils (see Krause et al., 1999, 2003; Carrano et al., 2002; Lamanna et al., 2002). Moreover, fossils of Spinosaurus Stromer, 1915 and other spinosaurids are significant because of controversy surrounding the postulated paleoecology of these taxa (see dis- cussions in Charig and Milner, 1997; Sereno et al., 1998; Sues et al., 2002). Questions related to spinosaurid paleoecology are par- ticularly important in the Bahariya Formation, where Spinosaurus appears to have shared its habitat (see Stromer, 1936; Smith et al., 2001) with at least two other theropods in the size range of Tyrannosaurus Osborn, 1905 (Bahariasaurus Stromer, 1934 and Carcharodontosaurus Stromer, 1931). Unfortunately, the holotype and only known indisputable specimen of S. aegyptiacus (BSP 1912 VIII 19) was lost during the night of 24/25 April 1944 in a British bombing raid of Munich (Nothdurft et al., 2002). The attack severely damaged the building (dating to ca. 1583, No ¨h- bauer, 1987) that housed the Pala ¨ontologische Staatssammlung Mu ¨nchen and destroyed most of Stromer’s Bahariya collection (see Appendix 1). Since 1944, new spinosaurid taxa have been described (see Charig and Milner, 1986; Martill et al., 1996; Ser- eno et al., 1998; Sues et al., 2002) and additional material has been referred to Spinosaurus (e.g., Buffetaut, 1989, 1992; Russell, 1996; Taquet and Russell, 1998; Benton et al., 2000; Buffetaut and Ouaja, 2002), some of it questionably. However, definitive S. aegyptiacus material, or information regarding the original spec- imen of Spinosaurus, has not been forthcoming. We report here on two photographs of the holotype of Spino- saurus aegyptiacus as it was reposited in the Pala ¨ontologische Staatssammlung Mu ¨nchen prior to 1944, which we ‘‘rediscov- ered’’ in the archives of the Pala ¨ontologisches Museum in June 2000, after they were donated to the museum by Ernst Stromer’s son, Wolfgang Stromer, in 1995 (Figs. 2, 3). These are, to our * Current address: Section of Vertebrate Paleontology, Carnegie Museum of Natural History, 4400 Forbes Avenue, Pittsburgh, Pennsylvania 15213- 4080. knowledge, the only surviving photographs of this, the one irre- futable specimen of S. aegyptiacus, which, prior to the initial print releases, which we authorized, of the photograph in Figure 3 (Prendergast, 2001; Glut and Chiappe, 2003), has been repre- sented only by Stromer’s (1915, 1936) drawings (it has recently been suggested that BSP 1912 VIII 19 may represent a chimera of more than one dinosaurian taxon [Rauhut, 2003], but exam- ining that idea is beyond the scope of this paper). Aside from their historical significance, these images are important in that they permit a direct comparison of several of Stromer’s (1915) illustrations with actual photographs of BSP 1912 VIII 19, there- by providing new insight into the skeleton. This is of note because Stromer’s (1915) text and illustrations have been used as sources of character data for Spinosaurus and the Spinosauridae in several theropod phylogenetic analyses (e.g., Sereno et al., 1998; Holtz et al., 2004). It is thus beneficial to be able to determine how closely these illustrations correspond to the actual anatomy of BSP 1912 VIII 19 as reproduced by the photographs, an effort which holds implications for the use of published drawings as sources of systematic information. DESCRIPTIONS OF THE PHOTOGRAPHS Figure 2.1 is a photograph of the right mandibular ramus of BSP 1912 VIII 19 in lateral view. Judging from this photograph, it appears that, in general, Stromer’s (1915) illustration (Fig. 2.2) of this element faithfully reproduces its anatomy and state of pres- ervation, accurately recording the distalmost six dentary alveoli and two preserved crowns (Rd4 and Rd12, based on Fig. 2.1 and data from Stromer, 1915; dental nomenclature after Smith and Dodson, 2003). However, there are some differences between the two illustrations (it should be noted that, in Fig. 2.1, what could be interpreted as the fifth dentary crown is actually the shadow of Rd4 falling against the photographic background). The shape of the mandibular ramus is the same in both, but some of the cracks and foramina on the surface of the dentary are only sche- matically congruent with the photograph (particularly in the cau- doventral area). More importantly, the rostral margin of the den- tary appears distinctly different in the two images. At the symphysis, the rostral margin is more ‘squared-off’ in Figure 2.2 than in Figure 2.1 (angles of 71° vs. 61° measured against a horizontal line set at the ventral surface of the dentary below Rd4). We suspect that this discrepancy might partially result from the two images showing the mandibular ramus from slightly dif- ferent angles, but proving this is difficult. Moreover, the dorsal surface of the dentary at the base of Rd4 is effectively the same angle in each image (72° in Fig. 2.1 and 70° in Fig. 2.2, in an angle measured between the distalmost point on the crown base of Rd4 and its apex). As such, we believe that the rostral margin of the lower jaw of BSP 1912 VIII 19 was incorrectly illustrated in the figure published by Stromer (1915). As the actual orientation of the rostral margin of the right den- tary in BSP 1912 VIII 19 is seemingly more rostrodorsal-cau- doventral than previously believed, the mesialmost two dentary