No giant Egyptian Deltadromeus

The Spinosaurus news has led to me reviewing the African mid Cretaceous taxa on the Database.  This means I've been working my way through the untranslated Stromer (1934), which describes Spinosaurus B and Bahariasaurus among other theropod remains. The result is that what I thought I knew about the supposed giant Deltadromeus remains described there is wrong.  The recent literature would have you believe it's a partial skeleton (Carrano and Sampson, 2008) identical (Sereno et al., 1996) to the Deltadromeus holotype.  Not so.

Sereno et al. (1996) referred the Baharija 'IPHG 1912 VIII' (described by Stromer, 1934) to Deltadromeus, specifying a coracoid, pubes, femur, proximal tibia and fibula as the material. Yet this specimen number corresponds to 32 specimens described by Stromer. The coracoid IPHG 1912 VIII 60 was associated with a scapula that shares that number, the femur is IPHG 1912 VII 69 based on the size Sereno et al. reported, and the fibula must be IPHG 1912 VIII 70 as no others are reported. Yet the pectoral girdle was found in layer m while the femur and fibula were found in layer p. The only proximal tibia reported is IPHG 1912 VIII 78, which is far too small to belong with the other hindlimb elements and from a different locality. Finally, the only pubes with that number are IPHG 1912 VIII 81, which are from yet another locality and much smaller than even the tibia. This materials list agrees with Carrano and Sampson, though note contrary to their statement, it is not a "partial postcranial skeleton". Stromer used the pubes as a paratype of Bahariasaurus, questionably referred the pectoral girdle, femur and fibula to the taxon as they cannot be compared to the holotype, and referred the tibia to aff. Erectopus. Thus all material was not referred to Bahariasaurus, contra Sereno et al..

Comparison of Baharija elements on the left to Deltadromeus holotype on the right, scaled to match in size.  From left to right- pectoral girdles in lateral view, proximal femora in lateral view, tibiae in proximal view, and proximal fibulae in medial view. Modified from Stromer (1934) and Sereno et al. (1996).

The Baharija pectoral girdle actually lacks the anteroposterior expansion considered diagnostic for Deltadromeus by Sereno (length excluding posteroventral process 117% of height vs. 150% in Deltadromeus), which is also found in Elaphrosaurus and Limusaurus. Due to breakage of the posteroventral process, it's uncertain if the coracoid's subacromial notch ('notch in anterior margin' of Sereno et al., as it is the only notch in Deltadromeus' coracoid) is shallow as in Deltadromeus or deeper as in Elaphrosaurus and Limusaurus. Though again, a shallow notch might not be diagnostic of Deltadromeus as it is also found in Ceratosaurus. The pectoral girdles also differ in other ways if scaled to similar overall size, with Deltadromeus having a narrower scapular shaft, a more abruptly expanded acromion, smaller glenoid, and deeper posteroventral process.

Sereno et al. also diagnose Deltadromeus based on its "accessory trochanter" on the distal femoral shaft, which is presumably the mediodistal crest anteriorly. This is common in basal theropods like ceratosaurs (e.g. Berberosaurus, Elaphrosaurus, Limusaurus), but rarer in Coelurosauria which Sereno et al. referred Deltadromeus too. The development of the mediodistal crest is unclear in the Baharija femur. Carrano and Sampson (2008) equated the "accessory trochanter" of Sereno et al. to the M. adductor femoris 1 insertion scar on the posteromedial distal shaft, but this region is unillustrated in the Baharija femur. Finally, the Bajarija femur does have an anterior process on the lateral margin of its medial condyle, stated as diagnostic of Deltadromeus and hinted at in Sereno et al.'s skeletal reconstruction. Carrano and Sampson equated this with the mediodistal crest discussed above, but that projects largely laterally so is probably not the feature Sereno et al. had in mind. While the anterior process could indicate a relationship between the Baharija femur and Deltadromeus, the latter differs in having a fully medially oriented head and an anterior trochanter that extends distally to the fourth trochanter. The Baharija femur is 165% the size of Deltadromeus, which could lead to questions of ontogenetic change, but neither of these characters are known to change ontogenetically in theropods, and they would leave the older specimen with the more basal morphology, which is unlike at least some theropods (tyrannosaurids, dromaeosaurids).

The tibiae are more similar to each other than to Elaphrosaurus, Camarillasaurus, Ceratosaurus or Eoabelisaurus in proximal view (the only available for Deltadromeus), with Deltadromeus differing in having a smaller, triangular posterior groove and larger lateral condyle. The fibulae are roughly similar, though Deltadromeus has a more projected anteroproximal corner and a proximomedial fossa that is less proximally extensive. The supposed pubes of Deltadromeus are actually ischia (Longrich pers. comm. and DML; Carrano and Sampson, 2008), so cannot be compared to the Baharija pubes.

Thus in total, the pectoral girdle and femur are near certainly not Deltadromeus (contrary to Sereno et al.'s claim the remains are identical), the tibia and fibula could be although no described apomorphies are shared, and the pubes cannot be compared. Coincidentally only the pubes can be compared to Bahariasaurus, though they differ markedly from that taxon*. This makes it quite possible Stromer was right in referring the pectoral girdle and femur to Bahariasaurus, and also possible the tibia and fibula belong to that genus. It also may make it more likely the pubis does belong to Deltadromeus, which might get support from study of its undescribed pubic fragments. Because none of the Bajarija material can be said to be more similar to Deltadromeus than the sympatric Bahariasaurus and some are certainly not Deltadromeus, none should be referred to either genus. This also eliminates any evidence Deltadromeus reached gigantic sizes, as there is no evidence the holotype is immature and the completely fused ischial boot would argue against this.

* Bahariasaurus has a less conspicuous and more proximally placed lateral flaring (15% down the shaft, compared to 21%), the distal end is not flared laterally, there is an extensive separation of the pubic shafts distally, and the interpubic foramen is more distally placed (80% down the shaft, vs. 71%).

References- Stromer, 1934. Ergebnisse der Forschungsreisen Prof. E. Stromers in den Wüsten Ägyptens. II. Wirbeltierreste der Baharije-Stufe (unterstes Cenoman). 13. Dinosauria. Abhandlungen der Bayerischen Akademie der Wissenschaften Mathematisch-naturwissenschaftliche Abteilung, Neue Folge. 22, 1-79.

Sereno, Dutheil, Iarochene, Larsson, Lyon, Magwene, Sidor, Varricchio and Wilson, 1996. Predatory dinosaurs from the Sahara and Late Cretaceous faunal differentiation. Science. 272(5264), 986-991.

Carrano and Sampson, 2008. The phylogeny of Ceratosauria (Dinosauria: Theropoda). Journal of Systematic Palaeontology. 6, 183-236.