(Last updated July 5th, 2026)

Pittet 2026 wrote a fantastic ~23,000-word paper using CT-scans to investigate the potential sensitivity of Niger spinosaurid snouts. Delightfully, he CT-scanned Crocodylus moreletii and Gavialis gangeticus in the process, and produced extraordinary CT-derived images in his paper. However, it is the section (<700 words total) where he suggests Suchomimus is more properly called Cristatusaurus that garnered an inordinate amount of attention in the paleo community at large. This gave me the opportunity to add to my spinosaur data acquisition after the S. mirabilis diggings (I need to blog that up someday), and to use this as an example of how much in paleo is decidedly not cut-and-dried.

What is a Holotype, and Why Does It Matter?

The holotype, aka name-bearing bone(s), is the most important specimen to a taxonomist because it is the reference material one uses to compare to all other bones; essentially, it lets you know when you've found another of that kind of animal.  The International Code of Zoological Nomenclature maintains rather detailed rules on holotypes. 

When paleontologists argue over whether a specimen belongs to a species, is a different species, or invalidates an old name, the comparison ultimately comes back to the holotype. Because so many species are known only from incomplete skeletons, skull fragments, teeth, or isolated bone(s), this causes no small amount of heartburn and frustration inside the academic community, and tis magnified 10x+ outside as we live in a world that wants rigid answers, not "Well, maybe, but that presupposes conspecificity! Did you consider variation?" among a host of reasons why a holotype might be challenged. Mix in the Copper Age, when what I dub paucitypes were the norm (and, alas, are still in use): a single, partial bone or tooth receives a name. Doh!

No dinosaur is 100% complete, so we inevitably have missing elements. Typically, we use comparative anatomy and "daisy chain" specimens to infer the morphology of the missing elements using a third+ specimen. What do we do when we only have two non-overlapping bones from the same formation? How do we decide whether they belong to the same genus or are separate? This vexing issue is in play throughout the rest of this write-up.

A Brief(ish) History of Cristatusaurus, a Spinosaurid You Might Not Have Heard Of

Taquet 1976

The first mention I can find of the bone (GDF366) that would become the holotype of Cristatusaurus was in Taquet's 1976 paper on the geology and paleontology of Gadoufaoua, Niger. I've provided a translation of the theropod portion of his paper at the end of this document.

Before I get into spinosaurs I want to ask a question. Taquet mentions finding a layer, GAD 5, that lacks carcharodontosaurid teeth. My translation of his quote: "The teeth of Carcharodontosaurus saharicus have therefore never to date been found in pre-Albian levels, and, until proven otherwise, this species is absent from the Gadoufaoua locality." Has this held true? I don't know, but I figure there are Gondwanan experts who can say.

Taquet mentions he found a huge claw in 1965, as preserved 23 cm long (~25 cm along the outer curve), and he speculated it would have been 29 cm whole, the size of your average adult Diplodocus caudal vertebra where the caudal rib ends (aka the longest vert in the tail). I've pasted a Ricqlès (1967) image, as well as the Rozhdestvensky (1970) oft-reproduced drawing. Me being me, I overlaid the two (after flipping the Ricqlès image around) and noted that the illustration and the photo are not a 1:1 match. I suspect that the thumb caused some grief. Even so, the top and bottom curves never aligned 100%. It always amazes me when a drawing of a photograph isn't a precise overlap. Perhaps it is because I'm a simple lad and would have used some tracing paper to reproduce it :-)? I was delighted to read that Osmólska and Roniewicz (1970) identified it as a manual phalanx and suggested it belonged to a giant theropod with a "strongly developed forelimb."

My favorite part of Taquet (1976) was, of course, about a neural arch found in 1970. Alas, he doesn't explicitly state that it's a dorsal vert, but I'm presuming it is. He wrote, "...whose neural spine is incomplete but must have been of great length..." It reminded him of Altispinax dunkeri, and he said twas similar to those of Spinosaurus aegyptiacus. 

He also noted a postorbital found in 1972 that resembled Acrocanthosaurus from the locality, and made a passing comparison to another from a different-aged locality, the morphology of which was not identical to the first yet also reminded him of Acro. No specimen numbers were provided, so I can't tell from Taquet and Russell (1998) when (or if) either of the photographed verts was one of those found in 1970.

Taquet writes, "The material from the Gadoufaoua locality attributed to this large carnosaur therefore approaches the pieces described under the names Acrocanthosaurus, Altispinax, and Spinosaurus. Now, these three Lower Cretaceous genera have been placed in a distinct family, that of the spinosaurids. The representatives of this family are relatively poorly known, but have in common the fact of possessing long neural spines. The carnosaur from the Gadoufaoua locality appears to be assignable to this family."

He concludes this section with "The study of the theropod material from Gadoufaoua, combined with that of the material from the other localities of the Continental Intercalaire of the Sahara, in particular from Gara Samani, will certainly help shed light on the family Spinosauridae." [emphasis mine]. I'm not sure about shedding light, but the material he found certainly has caused taxonomic controversy!

Before Cristatusaurus, There Was GDF365 and GDF366

Taquet (1984) describes "Two anterior extremities of mandibles" found at the "Gadoufaoua locality (level of the Innocents, El Rhaz Formation, Tegama series)." Anatomically speaking, the mandible is the lower jaw, and his description is based on the interpretation that these are the lower jaws of a spinosaurid. He writes, "The ventral margin is convex toward the front and continues in a straight line toward the back." 

These elements (GDF366 and GDF365, you'll see lots of pics of these very soon :-)), which he figures in the paper, are actually premaxillae. The "ventral margin" is the premaxillary crest that becomes important when comparing to other spinosaurs. He notes of GDF365, "...the two dentaries are tightly fused to one another toward the back, while the suture that separates them is still visible toward the front."

He writes of GDF366, the bone pair that Taquet and Russell (1998) will name the holotype of Cristatusaurus, "...it was possible to separate the two dentaries in order to observe the lingual surface: this surface is perfectly flat" and "...the extremity of the mandible is relatively broad at the level of the occlusal surface; it narrows ventrally and posteriorly into a narrow keel, which gives it a narrow V-shaped profile." He also notes "numerous vascular openings and does not present ornamentation of the crocodilian type." 

Modified from Taquet 1984. He had them upside down (dentary vs premax)

Flipped into life position

Why did Taquet conclude that these were dentaries rather than premaxillae? His words are below, and I've underlined the key component to his error, that being he didn't realize spinosaurs "break the theropod rules" with premaxillary teeth, having more than five:

"They cannot be premaxillae, for they bear no trace at their posterior extremity of sutures proving a contact between premaxillae and maxillae. Moreover, the dental rows are continuous at least as far as the seventh tooth, and it is known that the premaxilla of theropods, just like that of crocodiles, bears only three to five teeth. These are therefore indeed anterior extremities of dentaries."

Charig and Milner (1986), in their Baryonyx paper, noted these bones were "...described as mandibular symphyses of a spinosaurid. They are, however, identical to the conjoined premaxillae of Baryonyx, and comparison with Spinosaurus does not suggest spinosaurid affinities." In one sentence, they remedied two errors: these are premaxillae, and they aren't spinosaurids a la Spinosaurus; they are baryonychines.

I found this line from Taquet interesting: "If the piece described by Stromer had not undergone crushing..." I haven't studied spinosaurs, but am curious whether, and what kind of, crushing the Spinosaurus dentary had endured. He also wrote, "...Stromer, not having in the Bahariya material the right and left dentaries in connection, probably underestimated the length of the symphysis." All of this must have been concluded from the Stromer illustration.

Taquet also suggests there are likely two species of spinosaurs present, based on dentition: 

"It is probable that in the case of the pieces from Niger we are dealing with two distinct species of a new genus of spinosaurid, one of the species possessing smooth, flattened teeth with serrated margins, the other possessing grooved (fluted) teeth with serrated margins."

This presupposes a few things, that they have homodont dentition (do they?) and that there is no variation of this feature on teeth (I recall Sereno et al. 2026, in the S. mirabilis character, score this character as both 0 and 1 for S. aegyptiacus).

This non-assumption of conspecificity is important in the debate of Cristatusaurus and Suchomimus. Do we have one species of predator roaming Niger? Or do we have two closely related forms? Think grizzly versus black bear, both of which are quite similar skeletally, and both live in proximity of one another today in Yellowstone.

Taquet 1984 was the First to Propose Spinosaurs as Herons/Storks

Taquet (1984) wrote, "...the skull of these theropods must have possessed a very characteristic look, quite unusual if one compares it to the skulls of the other known carnosaurs." This builds upon Stromer's own (1936) comments that spinosaurids likely had an unusual (for theropods) diet, at least that is my interpretation of these sentences:

"...the teeth of Spinosaurus are very little flattened laterally, and a front and back edge is scarcely developed (Stromer II 3, 1915, pl. I, figs. 7–11). What we are dealing with here, then, are very simple, pointed, conical teeth, following one another at small intervals (loc. cit., fig. 6). Such a dentition could not have served for tearing prey to pieces, but was essentially a catching-rake [Fangrechen], quite similar to that of crocodiles, in which, moreover, likewise strongly enlarged, canine-like conical teeth occur far to the front (see text-fig. 10B, p. 69). A peculiarity, however, is the unusually strong differentiation in tooth size."

and

"Spinosaurus, indeed, could catch fish with its dentition just as well as crocodiles do."

[I provide translations of parts of Stromer 1915 and 1936 at the end of this document]

Based on "elongated form of the snout," at (1) Bahariya (lower Cenomanian of Egypt), (2) Gadoufaoua (Aptian of Niger), (3) Gara Samani (Albian of Algeria), (4) Abangarit (Albian of Niger), and the (5) Kem Kem (Albian of Morocco) "affirm the presence, in the Lower Cretaceous, of an extremely specialized family of theropod dinosaurs." [FYI, Kem Kem is now considered Cenomanian]

Taquet noted (via my translation), "Longirostriny in this case goes hand in hand with a piscivorous dietary regime, as one can observe with the eusuchian crocodile living today in the Ganges, Gavialis gangeticus. Nothing therefore forbids supposing the existence of piscivorous theropod dinosaurs. All the Cretaceous localities of the African platform abound in fish remains (genera Lepidotes, Mawsonia, Onchopristis, Hybodus), and it is tempting to imagine spinosaurid dinosaurs fishing along the rivers or at the edge of the lakes in the manner of herons or storks."

The original French:

Rien n'interdit donc de supposer l'existence de Dinosaures Théropodes piscivores. Tous les gisements Crétacés de la plate-forme africaine abondent en restes de Poissons (genres Lepidotes, Mawsonia, Onchopristis, Hybodus) et il est tentant d'imaginer des Dinosaures Spinosauridés pêchant le long des fleuves ou au bord des lacs à la manière des Hérons ou des Cigognes.

 

Charig and Milner (1986) reference Taquet's behavioral suggestion but claim they independently derived the idea: "Nevertheless the elongated snout with its terminal expansion and the large number of finely serrated teeth may well be indicative of ichthyophagy. Taquet's article on the Niger theropod made the same suggestion, comparing the elongated snouts with those of the Recent Gavialis gangeticus and commenting that "il est tentant d'imaginer des Dinosaures Spinosaurides pêchant le long des fleuves ou au bord des lacs a la maniere des Herons ou des Cigognes." We independently conceived a similar idea for Baryonyx, though as a quadrupedal predator crouching on the bank rather than as a biped stalking through the shallows."

I chuckled that they felt the need to point out they thought of it, too, all on their own. I've been there; I have thought of something novel, only to find out Gilmore had written it down in 1936 or Lull in 1919. I don't think I've ever pointed out I independently derived the idea... maybe I should :-).

Buffetaut (1989a) gave Taquet credit for suggesting fish-eating dinosaurs, "The idea of fish-eating dinosaurs was suggested by TAQUET (1984) on the basis of the jaw remains from Niger..." and added, "Spinosaurus and Baryonyx may thus have been more or less amphibious fish-eating theropods with crocodile-like teeth and jaws, just as, conversely, various extinct crocodilians with theropod-like deep narrow snouts and compressed serrated teeth probably were terrestrial predators."

This sentence accepts Charig and Milner's "Niger material is Baryonyx" position and notes we have evolutionary convergence of dentition and prey, of dinosaurs and crocs in water and on land ending up with similar tooth morphology.

I didn't find Taquet's 1984 quote referenced in any of this century's spinosaur papers I have read. Anyone know why? It gets called a heron and stork quite often. Maybe the quote was overlooked because it was in French?

 

GDF365, 366 Figured in the Angaturama Paper

Kellner and Campos (1996), while naming Angaturama, provide photographs of GDF365 and GDF366 (Figs. 5-7). They curiously wrote, "No evidence of a premaxillary sagittal crest could be observed."

I wonder if they were working from figures or actually handling the bone. We should require all authors to denote if they handled a specimen directly or, if not, which images they used to reach their conclusions.

Many authors note the presence, or absence, of serrations and suggest these are important. I kept the quote above because the second paragraph is another vote for serrations indicating "at least two species" in Africa.

 

 

Kellner and Campos (1996) argue that GDF365 and 366 are morphologically similar, but they also sought to determine whether they could be from the same or different taxa. They noted, "The major differences are the ornamentation of the bone surface (more smoother on the smaller specimen: MNHN, GDF 366), the size, and the teeth ornamentation."

They stated, "The first two differences could be related to the different ontogenetic stages of those specimens, with the larger (MNHN, GDF 365) representing possibly a mature animal, while the smaller (MNHN, GDF 366) likely represents a juvenile. The fact that the premaxillae of the smaller specimen are not fused or tightly connected seems to support this interpretation." Everyone I've read thus far agrees this is likely the case.

They comment on the differences between tooth ornamentation that Taquet (1984) noted: "...it must be remembered that no basal portion of a tooth from the larger specimen (MNHN, GDF 365) can be examined. In the smaller specimen (MNHN, GDF 366) the longitudinal ridges are comparatively well developed, but essentially on the lingual basal portion of the crown (Fig. 7B). The top part of the seventh left replacement tooth also lacks strongly developed longitudinal ridges. " It felt like they were testing if these two elements were conspecific.

Seemingly finding the tooth differences potentially of great value, they write, "Therefore we conclude that although the Gadoufaoua specimens probably represents animals of different ontogenetic stages, it is not sure at this point if they represent two different taxa." This is the earliest I can recall reading that these elements may not be conspecific.

 

1998: The Niger Holotypes

Cristatusaurus Taquet and Russell 1998

Taquet and Russell (1998) named Cristatusaurus, the "Crested Lizard," thanks to a crest on the premaxillae. GDF366, interpreted as a juvenile specimen, was designated the holotype, while GDF365, a better preserved, older individual, is referred to the taxon. Its association with GDF366 is never stated. Was it found nearby? 100 miles away? It presumes conspecificity of spinosaurid-looking taxa. And why use a juvenile as a holotype???

I've grabbed images of GDF366 and overlaid them on one another, using the dorso-anterior curve as my reference to tie them all together. If you have heard me present or read some of my missives, it shouldn't come as a surprise that none of the images overlap identically. One caveat: Taquet 1984 and Hendrickx et al. 2016 are "left reversed," which is why they have different scratches, bumps, and holes. I kept them as a reminder that when folks are using a "reversed" image, there will be differences; these elements don't come out of the ground pristinely bilaterally symmetrical, thus always take extra caution when something has been flipped, especially if one is using outlines.

The holotype GDF366 also includes a right dentary and maxilla (below). The association of these elements to the holotype is never stated. I'm not even sure what I'm looking at here! Perhaps someone has a clearer photograph? They are described as the medial sides of a right maxilla and dentary. I'll take them at their word and assume the original paper had clear photos, and that this is an artifact of the vintage '90s Xerox copy I obtained via interlibrary loan. I felt like I was taking a Rorschach test, interpreting the dentary lumps at the top left as tops of teeth and wondering if it had a "chin" (presuming the right side in this image was the anteriormost end of the lower jaw). They also remind me of early Mars photographs when viewing the channels of the maxillae...

Characters in The Taquet and Russell (1998) Paper in Tables

I broke down the Taquet and Russell (1998) characters because I became quite confused by how they bounce around among taxa. I made this for me to keep "score" and thought someone else might find it useful to see how I dissect a paper.

Note: this is a blog, not a peer-reviewed paper, so I didn't do a third pass to make sure I got it 100% right. It should be, but I wanted to put that out there in case folks begin citing this in actual peer-reviewed papers.

1. Spinosaurus maroccanus versus Cristatusaurus lapparenti

Character	Spinosaurus maroccanus	Cristatusaurus lapparenti
Premaxilla, lateral profile	Bulbous	Short, strongly "hooked"
Premaxilla height, front to back	Decreases posteriorly	Increases posteriorly
Posterodorsal surface	Rounded in cross-section	Narrows into a crest
1st premaxillary alveolus	Relatively small	Relatively large
Lateral premaxillary alveoli	Grouped into two pairs	Uniformly, closely spaced
Maxillary & dentary tooth cross-section	Circular	Laterally compressed
Maxillary tooth orientation	Procumbent, front and back	Vertically oriented posteriorly
Dentary, mid-section	Markedly constricted; alveoli paired	Slightly constricted; alveoli closely spaced
Carinae	Smooth	Finely serrated

2. Versus other taxa:

Taxon	vs. Baryonyx walkeri	vs. Irritator challengeri	vs. Angaturama
Spinosaurus maroccanus	Lacks serrations (Baryonyx has them)	Irritator's dentition is more massive, less homodont	Premaxilla is longirostrine (Angaturama's is not)
Cristatusaurus lapparenti	Premaxilla is brevirostrine (Baryonyx's is not)	Has serrations (Irritator lacks them)	Has serrations (Angaturama lacks them)

(They spell it "Angaturma" twice in the paper)

3. They compare Milner's serration-based split of taxa versus their own long vs short rostrum morphology

Taxon	Dental serrations	Rostrum form
Spinosaurus maroccanus	Absent	Longirostrine, rounded in section
Cristatusaurus lapparenti	Present	Brevirostrine, crested
Baryonyx walkeri	Present	Longirostrine, rounded in section
Angaturama	Absent	Brevirostrine, crested
Irritator challengeri	Absent	Longirostrine, rounded in section

Sort by serrations, and one gets S. maroccanus, Angaturama, Irritator vs. C. lapparenti, Baryonyx

Rostrum morphology produces S. maroccanus, Baryonyx, Irritator vs. C. lapparenti, Angaturama.

Baryonyx and Irritator agree on rostrum shape but split on serrations

Cristatusaurus and Angaturama agree on rostrum shape, too, but split on serrations

Neither character sorts the group the same way twice, which is why they don't commit to whether Spinosauridae, Baryonychidae, and Irritatoridae are validly distinct families, as both synapomorphies fail to yield the same grouping. 

4. Their efforts to unite them

Proposed synapomorphy	S. maroccanus	C. lapparenti	Baryonyx	Angaturama	Irritator
a) Elongate rostrum, nares set back	Yes	Yes	Yes	Yes	Yes
b) Premaxilla ventral margin concave; dentary tip upturned	Yes	Yes (per "hooked" profile)	Yes	Yes	Yes
c) Teeth subcircular, straight-slightly recurved	Yes	No- laterally compressed	Yes (only slightly flattened)	Yes	Yes
d) Seven premaxillary teeth	Yes	Yes	~Yes (6 left / 7 right)	Yes	Yes¹
e) Orbits inclined posterolaterally	?	?	?	?	Yes
f) Supratemporal fenestra reduced	?	?	?	?	Yes
g) Crest extends back over temporal region	?	?	?	?	Yes
h) Tooth roots occupy ~full maxilla depth	?	Yes	?	?	Yes

¹ Inferred from the paper

Row c) contains a direct contradiction: the same tooth-cross-section axis is used both to unite the family in the Conclusion and to distinguish Cristatusaurus from Spinosaurus in the differential diagnosis. Cristatusaurus is the only taxon that breaks the pattern it's supposedly part of. Also, characters e-g are "possible synapomorphies" resting on a sample size of one (I can probably make that claim, though, for most synapomorphies across Dinosauria...). Irritator's cranium in this list is the only one complete enough to show orbits, supratemporal fenestra, or a temporal crest.

5. Carcharodontosaurus

Character	S. maroccanus (Gara Samani maxilla)	Carcharodontosaurus
Lateral surface, longitudinal curvature	More convex	Less convex
Lateral surface, vertical curvature	Slightly convex	Concave
Alveoli	Rounded	Subrectangular

6. Pelecanimimus (I saw this the summer of 2025, and here I see it again :-))

Characters suggesting spinosaurid	Characters against spinosaurid
Seven premaxillary teeth, a count otherwise unmatched among dinosaurs	Total tooth count over 200
Median crest over the temporal region, as in Irritator	Anterior/premaxillary teeth "D"-shaped, not subcircular
Jugal excluded from the antorbital fenestra margin	Remaining teeth laterally compressed
Maxillary teeth larger than dentary teeth	Manus resembles that of ornithomimids
Dentary teeth unusually numerous
Interdental plates absent
Skull narrow and shallow, facial region elongated

7. Supporting numbers

Comparison	Values
Mid-cervical centrum length: height ratio	S. aegyptiacus ~1.1 vs S. maroccanus ~1.5
Neural spine transverse diameter, comparable to anterior dorsal	S. maroccanus: 25 mm vs Cristatusaurus-referred (GDF359): 15 mm
Dentary tooth count	S. aegyptiacus: 15 C. lapparenti (est.): ~20  Baryonyx: 32 Pelecanimimus: ~75

 

Suchomimus Enters the Fray (You Found Me ;-))

Two months after Taquet and Russell (September) named Cristatusaurus, Sereno et al. (November, 1998) named Suchomimus, designating GDF500 as the holotype, with additional bones referred to Suchomimus. This might be the first paper I have read where the holotype is stated but never described; the paper focuses on GDF501, the referred skull. I've reproduced below the original skeletal and the list of Suchomimus elements from the paper. It isn't clear from the description what bones belong to the Suchomimus holotype in the skeletal, especially with the line "many additional bones and teeth." 

What is immediately apparent is that there are no overlapping holotype elements between Cristatusaurus and Suchomimus. Thus any comparison will have to assume that the skull of GDF501 is conspecific with GDF500, not unreasonable but definitely an assumption.

Suchomimus is diagnosed using the characters listed below; however, I'm unable to test 10 of the 13 (22-25, 29-34). Of the three I can see, none are part of the holotype (GDF500), and it isn't clear how I am to know they are properly associated, the same issue I have with Taquet and Russell (1998).

 

Suchomimus had a partial skull (GDF501, below) referred to it. Assuming the skull is Suchomimus and not another taxon, individuals have then made comparisons to other taxa, specifically Baryonyx and Cristatusaurus. 

 

One of my favorite parts of Sereno et al. 1998 is footnote 17:

"In contrast to the specimens described here, previously known spinosaurid material from Niger has been limited to fragmentary disarticulated bones that are attributable to an as yet indeterminate (6-8) spinosaurid. Recently, however, a new spinosaurid, Cristatusaurus lapparenti, was named on the basis of material from Gadoufaoua (4). The holotypic specimen consists of portions of the premaxillae, maxilla, and dentary, the association of which was not established. The authors state that the material differs from Baryonyx walkeri by the "brevirostrine condition of premaxilla." However, no distinguishing features or proportions are apparent to us or to previous authors (13), who attributed the premaxillae to an indeterminate species of Baryonyx. We therefore regard C. lapparenti as a nomen dubium."

I dug in deeper on the "6-8" citations. (6) is Taquet 1976, before Taquet and Russell 1998 had named it, (7) Taquet 1984, once more before they had named it, and (8) Kellner and Campos 1996, a third indeterminate note before it was formally named. Taquet and Russell named the indeterminate bones in 1998 (GDF366, holotype; GDF365, referred). Perhaps because they were at the time of 6-8 indeterminate, such language was selected, knowing each of those papers referenced GDF 366 and 365? Cristatusaurus was named two months before the Sereno et al. 1998 paper, so he definitely had to address them in some capacity.

Sereno et al. (1998) note "The holotypic specimen consists of portions of the premaxillae, maxilla, and dentary," which is lifted directly from Taquet and Russell (1998), who wrote, "Type specimen: (figure, f-h) MNHN GDF366, both premaxillae, portion of right maxilla and dentary." Taquet and Russell refer GDF365 and dorsal vertebrae 357-359, 361 to Cristatusaurus, and describe the dentary as slightly constricted vertically in mid-section with closely-spaced alveoli, and the teeth are laterally compressed in cross-section with finely serrated carinae. They speculate that the dentary fragment might have had 20 alveoli if complete (vs. 15 in S. aegyptiacus).  

I find it delectable that Sereno et al. (1998) wrote, "the association of which was not established," in a paper that not only doesn't specify what, exactly, comprises GDF500 but also doesn't show where any of the referred material comes from. They then note, "no distinguishing features or proportions are apparent to us or to previous authors," the previous authors being Charig and Milner in 1986, who described them as "identical," and in 1997 as "virtually identical" to Baryonyx. I suppose there were two authors, and they made comments in two papers, so using the plural of "authors" is fair, but when I first read it, I thought they were citing multiple camps to support their claims.

 

Cristatusaurus is named from only cranial material, Suchomimus from only postcranial elements, though note there is a sliver of pink (the holotype color) in Sereno et al. 2022. If that isn't an artifact, a trick of mixing various layers in Photoshop, it still doesn't appear to be a premaxilla, so one can't compare it directly. Which means we need near-100% evidence that GDF501 belongs to Suchomimus and not, say, a second (or third ;-)) taxon. We'll look at the referred material next.

(When I posted this on Instagram, I was informed that a skeleton with a skull was found in 2018, which will hopefully put paid to this debate.)

 

The Referred Material

Cristatusaurus had referred to it a larger fused premaxillary pair (GDF365) and four dorsal vertebrae (GDF357, 358, 359, 361). Oddly, the Taquet and Russell (1998) figure lists "Anterior dorsal vertebra, left lateral view, GDF 330." for (i) and for (j) as "Neural arch and base of a spine of anterior dorsal vertebra, right lateral view. GDF 359." I don't know where the 330 comes from. It is mentioned in one passage that doesn't provide (at least to me) clarity: "The base of a spine preserved on the neural arch of an anterior dorsal vertebra (figure, j) (GDF 359, cited by Taquet, 1976, p. 53) is less robust than a spine from a neural arch of similar dimensions of S. maroccanus (15 versus 25 mm in transverse diameter, cf. Russell, 1996, figure 8). It is possible that the neural spines were not so elevated in the species from Gadoufaoua. A transitional cervico-dorsal vertebra (figure, i) (GDF 330) bears a heartshaped anterior central facet and a pleurocoel beneath the parapophysis, but lacks a hypapophysis. No other dorsal vertebrae possess pleurocoels." I read that paragraph multiple times and came to the same ending... confusion. Maybe GDF330 is one of those found in 1970?

They continue, "The anterior margin of the transverse process of anterior dorsal vertebrae is linked by a ridge to the anterior zygapophysis; a similar ridge descends from the centre of its ventral surface to the posterior base of the neural arch. Two smaller laminae project anteroventrally from the descending ridge across a deep lateral excavation within the neural arch. The neural arch is apparently often interrupted completely behind the posteriorly descending ridge, and the posterior zygapophyses are supported by a separate pillar of bone rising from the posterior end of the neural arches. Zygosphene-zygantrum articulations are present." The ensuing paragraph provided no more clarity. I wonder if GDF330 is a specimen number that can be matched back to specimens mentioned in Taquet 1976?

Neither Taquet and Russell nor Sereno et al. provide detailed locality maps; thus, it is unclear how close or far apart the holotypes are. On one hand, this doesn't matter; larger animals can roam far and wide; the bigger the beast, the farther it might roam (though there are examples of a fox wandering thousands of miles). On the other hand, there might be more than one spinosaurid taxon present. Yellowstone has black bears and grizzly bears that I have observed just a few miles apart in identical habitats. I suppose I want to see it because, well, I want to know! :-)

I overlaid Baryonyx, Suchomimus, Cristatusaurus premaxillae to see how closely they lined up. As you can see, there are some... issues.

Overlay of Baryonyx, Suchomimus, Cristatusaurus premaxillae

Overlay of Baryonyx, Suchomimus, Cristatusaurus premaxillae

What Others Have Written Through Time

Cristatusaurus Through Time in the Eyes of Other Authors

Charig and Milner (1986) considered Taquet's Niger premaxillae (GDF365, 366) to be Baryonyx sp. and write, "two fragmentary snouts from the Aptian of Niger, described as the mandibular symphyses of a spinosaurid. They are, however, almost identical to the conjoined premaxillae of Baryonyx, and comparison with Spinosaurus does not suggest spinosaurid affinities." Of course, their "comparison with Spinosaurus" is code for "comparison with a drawing of Spinosaurus from a single angle, fixed in time and space forever." Unless they happened to see the "lost" photos of it before they were published this century... in which case they have a second, fixed, data point.

Charig and Milner (1997) wrote of GDF365 and 366, "Each is virtually identical to the conjoined premaxillae of the holotype of Baryonyx (Charig & Milner 1986; 1990: 139) except in that they possess seven alveoli on each side, not six on the left and seven on the right as in R9951. We consider that these snouts, despite their much younger age, are referable to Baryonyx sp. indet."

Buffetaut and Ouaja (2002) wrote, "...spinosaurid premaxillae with serrated teeth from the supposedly Aptian Elrhaz Formation of Niger, originally described by Taquet [1984] as  dentaries, and redescribed by Kellner and Campos [1996] and Taquet and Russell [1998], are extremely similar to the premaxilla of Baryonyx. Taquet and Russell [1998] have described them as a new spinosaurid taxon, Cristatusaurus lapparenti, supposedly differing
from Baryonyx by a “brevirostrine condition of premaxilla”. However, what is meant by this is unclear, and there appears to be no significant difference between the pre-maxillae from Niger and that of Baryonyx. Therefore, as noted by Naish et al. [2001, p. 246], “it is arguable as to whether Cristatusaurus can be distinguished from Baryonyx”, Sereno et al. [1998] consider Cristatusaurus as a nomen dubium, and the material referred to it by Taquet and Russell as belonging to an indeterminate species of baryonychine. We agree with Charig and Milner [1997] that this material can be referred to as Baryonyx sp., so that Cristatusaurus should be considered as a junior synonym of Baryonyx. From the same formation in Niger, Sereno et al. [1998] have described a partial skeleton of a baryonychine, which they have called Suchomimus tenerensis. As mentioned by
Sereno et al., this animal is obviously closely related to Baryonyx. Further comparisons between these two taxa may lead to synonymise them at the generic level (A.C. Milner,
pers.com.)."

Sues et al. 2002 "They [Taquet and Russell 1998] distinguished this taxon from Baryonyx walkeri solely on the basis of the ‘‘brevirostrine condition’’ of the premaxilla. Sereno et al. (1998) considered this difference uninformative, and Charig and Milner (1997) identified the material discovered by Taquet as Baryonyx sp. indet. Shortly after the publication of the paper by Taquet and Russell, Sereno et al. (1998) briefly announced the discovery of additional skeletal remains, including a snout and a partial postcranial skeleton, from the same locality. They assigned these specimens to a new genus and species, Suchomimus
tenerensis, which they distinguished from Baryonyx walkeri by the broader and taller neural spines of the dorsal, sacral, and anterior caudal vertebrae, robust humeral tuberosities, much enlarged olecranon that is offset from the humeral articulation,
and hook-shaped radial ectepicondyle. There exists at present no evidence to indicate the presence of more than one taxon of spinosaurid in the faunal assemblage from GAD 5. We concur with Milner (in prep.) that the anatomical differences between the material reported by Sereno et al. (1998) and Baryonyx walkeri only warrant recognition of the former as a distinct species of Baryonyx, B. tenerensis. The generic nomina Cristatusaurus
and Suchomimus should be considered subjective junior synonyms of Baryonyx."

Rauhut 2003 wrote, "The recently described... Cristatusaurus lapparenti Taquet and Russell, 1998, [is] regarded as [a] nomina dubium."

Bertin 2010 wrote, "...cranial and vertebral material were assigned to a species of baryonychine spinosaurid dubbed Cristatusaurus lapparenti by Taquet & Russell (1998), a taxon said to be generically separated from Baryonyx walkeri by the ‘brevirostrine condition’ of the premaxilla. Though not elaborated upon by the authors, this was likely in reference to the characteristic spinosaurid termainal [sic] rosette. Based upon inspection of previously published photographs and line drawings (Charig & Milner, 1997; Sereno et al., 1998; Taquet & Russell, 1998), I find no particular distinction between the holotypic remains of Cristatusaurus lapparenti and the fossil material of other baryonychine spinosaurids, a conclusion also made by several previous authors (Charig & Milner, 1997; Sereno et al., 1998; Sues et al., 2002). Though it may likely belong to Suchomimus tenerensis, the spcimen [sic] is instead listed as ‘baryonychine indet.’ in the database because of the fragmentary nature of the material."

Carrano et al. 2012 wrote, "Cristatusaurus lapparenti Taquet & Russell, 1998. Cristatusaurus lapparenti was the first spinosaurid described from the same formation (Elrhaz Formation;
Aptian–Albian) and general location as Suchomimus (Taquet 1976; Taquet & Russell 1998). It has been widely considered a nomen dubium (Sereno et al. 1998; Sues et
al. 2002; but see Allain 2002) because the holotype material (MNHN GDF 366) is extremely fragmentary and the accompanying diagnosis very general, a conclusion with which we agree. Cristatusaurus can only be identified as an indeterminate baryonychine spinosaurid. However, we also consider it unlikely that two baryonychines are present at Gadoufaoua, and therefore Cristatusaurus and Suchomimus almost certainly represent the same animal."

Hendrickx et al. 2016 wrote, "given the paleogeographic and stratigraphic distribution of Cristatusaurus lapparenti and Suchomimus tenerensis, and because the material referred to the two taxa is almost identical, Suchomimus tenerensis [25] is most likely a junior synonym of Cristatusaurus lapparenti. Yet, no definitive autapomorphy could be found in the Cristatusaurus holotype and this taxon is here considered as a nomen dubium; Figs A and B in S1 File"

In the S1 File, Hendrickx et al. 2016 wrote, "The diagnosis given by Taquet and Russell [1] to define Cristatusaurus is nonetheless considered by many paleontologists as uninformative and/or insufficient to distinguish it from Baryonyx. Taquet and Russell [1], indeed, differentiate Cristatusaurus from Baryonyx on the basis of its ‘brevirostrine condition’ (i.e., a short rostrum). The authors consider that the Cristatusaurus premaxilla is short (which would indeed differ from the anteroposteriorly long premaxilla of Baryonyx), possibly due to the confusion between the deep maxillary notch of the Cristatusaurus premaxilla (Figs A:A, B:C) and the external naris, which is clearly visible in Baryonyx and Suchomimus (Fig A:G, J; Fig B:A‒B, D). Given that both premaxillae of Cristatusaurus are incomplete and only preserved the anterior portion of the bone, it is very likely that the premaxilla was as long in Cristatusaurus as in Baryonyx. We, therefore, agree with most paleontologists and consider that the differential diagnosis provided by Taquet and Russell [1] to distinguish Cristatusaurus from Baryonyx is uninformative."

and, "Despite the numerous similarities between the Cristatusaurus and Suchomimus premaxillae, we could not find a definitive apomorphic character diagnosing Cristatusaurus lapparenti. Indeed, the size (as the number) of the premaxillary alveoli, the depth and extension of the nasal fossa, and the development of a premaxillary crest may all vary ontogenetically and intraspecifically, or may result from sexual dimorphism. We, therefore, tentatively consider Cristatusaurus lapparenti as a Baryonychinae nomen dubium pending a deeper examination of the postcranial material of this taxon. Given the apomorphic characters in the cranial and postcranial skeleton (see Carrano et al. [10]), we, however, consider Suchomimus tenerensis as a valid taxon."

Sales and Schultz (2017) did not resolve Cristatusaurus as definitively valid, but they argued that it should not be automatically synonymized with Suchomimus, identifying possible differences in the secondary palate and maxillary ascending process. Their phylogenetic analysis did not recover Cristatusaurus and Suchomimus as sister taxa. They do ask, "Thus, future analyses on the validity of Cristatusaurus should also focus on the taxonomic significance of these differences in relation to Suchomimus."

Lacerda et al. 2021 "Therefore, we agree with Sales and Schultz (2017) on the validity of C. lapparenti..." but I didn't see where Sales and Schultz came out and explicitly stated that.

Pittet 2026 is all-in on Cristatusaurus: "A third point shared by Cristatusaurus and Suchomimus is the premaxillary crest (which originally gave its name to Cristatusaurus) due to the thinning of the upper margin of the premaxillae by the digging extending the narial fossa anteriorly. Even though the crest of the younger specimen is heavily damaged, we can get an idea of it by
comparing the labial view of this individual with the more mature specimen MNHN.F.GDF 365, especially in relation to the location of his neurovascular network (Appendices 1F-H
and 2F-H). Thus, as suggested in the original diagnosis of this taxon (Taquet 1984; Taquet & Russell 1998) and in the light of our tomographic results, we consider this character to be
a diagnostic feature of Cristatusaurus, especially as it is not present in Baryonyx. Nor is this crest linked to ontogenic or intraspecific variability (Hendrickx et al. 2016), since it is present
in two individuals at different stages of maturity. Finally, the presence of this crest in Suchomimus tenerensis suggests that the latter can be referred to Cristatusaurus and is indeed
a junior synonym of Cristatusaurus. What is more, the two taxa are sympatric: all Baryonyx specimens are European (from England and Portugal) while Suchomimus and Cristatusaurus
shared a relatively similar geographical distribution (Niger)."

And now I'm full circle!

Pittet 2026 Fig. 20 overlapping referred elements of Suchomimus (GDF501, blue) and Cristatusaurus (GDF365, green)

Baryonyx (BMNH R9951 from Charig and Milner, 1997) atop referred Suchomimus (GDF365) premaxilla (from Hendrickx et al. 2016)

 

Size Oddities

 

I wouldn't be me if I wasn't overlaying images, and the above is a great example of the Trouble With Scribbles. The top image, from 1998, was the only information the world had on the holotype of Suchomimus until 2022, when 3D scans were assembled to build a modernized skeletal. I removed all elements but those color-coded as "holotype pink" (aside, why is there a cranial element???). One can easily see rather stark differences in the limbs. I can forgive the vertebral proportions, as one is often guessing at the vertebral count (well, within clade parameters), but the limbs should be pretty close to identical... and they are not. Scale the hindlimb to the tibia, and one gets "Popeye" arms, as you can see in the image below.

 

"Popeye" arms

When I scaled the animal to the tibia (and, therefore, the femora), the forelimbs ate a can of spinach! Otherwise, the now 3D-scanned hindlimbs are twigs, unlike what was drawn in 1998.

I shared these observations on social media and received lots of feedback; the most surprising, to me, was (and I'll paraphrase) "Skeletals are directionally correct, meant to show what bones are present." Then why include a scale bar? Why not declare "these aren't meant to be scientifically accurate, just eye candy?" I work with skeletal artists who take their craft quite seriously. Their work is amazing. Gunnar Bivens uses photos of bones taken in orthogonal view to craft his skeletons. Daniela Barrera Guevara, I can say with firsthand knowledge, is also quite careful. I doubt Scott Hartman, the most influential skeletal artist, would appreciate his work being categorized as "close." Yet many apparently have that viewpoint, far more than I ever knew. I'm good with either kind of skeletal, as long as we are told somewhere what the underlying driving force of the skeletal was: accurate or in the ballpark. 

In the avid enthusiast camp, there is a cottage industry of folks that ping me, "Dr. BC, you are wrong, it weighed 1,446.23469 kilograms and was 13.62394 meters long, my proof is so-and-so's (or my own) skeletal and math equation." I LOVE the passion, and don't ever dismiss anyone out of hand, but it struck me that the crowd of Discord lovers is hanging on every pixel of a skeletal. I always ask them, "Did you measure from the inside or the outside of a bone's drawing, as line thickness matters?" Keep in mind I work on sauropods, across 30 meters, such issues matter (as does cartilaginous pad thickness assumption and zygapophysial overlap, but I digress :-)). 

Irritator vs Angaturama

Someday I'll dive into this dastardly duo; I certainly gathered many images of 'em, from Rorschach-test images in the holotype description to quite nice ones in recent years. We find ourselves in a similar conundrum, made even more tragic by the fact that with Angaturama we can count the teeth with certainty, but with Irritator we can't. Spinosaurids display variation in tooth count (sometimes by side, I'm looking at you, Baryonyx), and the "bone lip" margin is consistently inconsistent. This means when using tooth counts of Irritator, we have multiple built-in assumptions: the tooth position number itself and conspecificity.

 

I want to give a special shout-out to Machado and Kellner (2005) as it is the first image I have found in a long time that matches the photo. It can be done!!!

 

WHY? Why Do We Find Ourselves in this Situation?

If you are wondering why anyone would name a single bone the holotype, well, that merits its own field of study :-). Some folks like to name animals; it gets press, which leads to grants, which leads to press... Others note that if you find a new type of animal in a place it hasn't been found before, well if you simply describe it with only a specimen number, journals aren't excited and so you end up in a "lower tier" publication. Yep, journals have "point values," with Nature and Science at the top and dozens of regional journals at the bottom, the latter of which don't help with tenure, grants, or the press.

Saying "we found this cool spinosaurid snout in Brazil" will get a yawn, but saying, "We found Angaturama!" (even though it is but a snout) gets far more eyes on it immediately. It also ends up in other papers' faunal lists. How many lists come to mind that say "spinosaurid, carcharodontosaurid, etc." versus listing genus names in a formation? Not making those faunal rolls means people aren't aware of a bone's existence.

Logically, if there is any amount of time (say 500k+ years) betwixt when specimens lived, why couldn't they be new taxa? Species can evolve in a quick amount of time, especially if the environment is changing.

The pendulum has shifted to naming everything a new genus. Why? In part because it is easier on the cladists :-). Here is an example. Let's say later on Spinosaurus mirabilis repeatedly falls into the Suchomimus camp (it won't, but stay with me). Now the cladograms will say "Spinosaurus mirabilis and Suchomimus are sisters," but there is another Spinosaurus not part of that grouping. That looks terrible, reads awful, and can even cause node-naming heartburn. But, if it was Unicornisaurus mirabilis, then regardless of where it fell on the tree, it wouldn't cause these odd naming issues. Keep in mind all of these names are made up by people, for people. In the old days, the software worked for us. Today? We work for the software :-).

BC 

July 5th, 2026. 16 hours (probably more, but I don't want to count the many rabbit holes I went down :-))

 

Many References Cited (I grew weary of pasting them...)

Buffetaut and Ouaja 2002

Charig, A. J. & Milner, A.C. 1986. Baryonyx, a remarkable new theropod dinosaur. Nature,
London, 324: 359-361.

Charig, A.J., and Milner, A.C., 1997. Baryonyx walkeri, a fish-eating dinosaur from the Wealden of Surrey. Bulletin of the Natural History Museum Geology, 53: 11–70.

Hendricx et al. 2016

Kellner, A.W.A., and Campos, D. de A., 1996. First Early Cretaceous theropod dinosaur from Brazil with comments on Spinosauridae. Neues Jahrbuch für Geologie und Paläontologie, Abhandlungen, 199: 151–166.

Osmólska, H. et Roniewicz, E. 1970. - Deinocheiridae, a new family of theropod dinosaurs. Paleont. Polonica, No 21, p. 5-19,4 fig., 5 pI.

Pittet 2026

Ricqlès, A. de, 1967. - La paleontologie de terrain: un bilan international. Atoomes, no 243, p. 337-341, 10 fig.

Rozhdestvensky, A.K. 1970. - Giant claws of enigmatic Mesozoic Reptiles. Paleont. Journ., vol. 4, no 1, p. 117-125,7 fig.

Stromer 1915

Stromer 1936

Taquet, P. 1976 Géologie et paléontologie du gisement de Gadoufaoua (Aptien du Niger), Cahiers de Paléontologie, Éditions du CNRS, 191 p.

Taquet, P. 1984, Cr. hebd. Séanc. Acad. Sci., Paris 299, 217-222 

Taquet 1976 Theropod Section Translation

Suborder THEROPODA

Numerous theropod remains have been found at the Gadoufaoua locality. Some preliminary observations can already be offered regarding the recovered material, which can be divided into two clearly distinct lots of roughly equal numerical importance:

1. a lot comprising the bones of a large-sized carnosaur;
2. a lot comprising the bones of a small-sized coelurosaur.

Infraorder CARNOSAURIA - Family SPINOSAURIDAE

Among the collected pieces are the following elements: about ten teeth, a postorbital fragment, a fragment of the mandibular symphysis, a braincase fragment, an ilium, a proximal portion of a femur, several metatarsals, vertebrae, and phalanges, including four large ungual phalanges. [BC - I wonder if these have all been tracked down]

a) Teeth. The teeth recall those of the North American genus Antrodemus (= Allosaurus); curiously, none of them attains the dimensions and the flattening characteristic of the teeth of Carcharodontosaurus saharicus (Depéret and Savornin 1927).

The teeth of Carcharodontosaurus saharicus figured by Depéret and Savornin (1927, Pl. XII, figs. 1 and 2), Stromer (1931, Pl. I, figs. 1 and 2), and Lapparent (1960, Pl. IV, figs. 1 to 4, 7 and 9) are nonetheless very abundant in a number of localities of the Continental Intercalaire of the Sahara. They are large, very flattened, and possess, in addition to the serrations of the cutting edge of the crown, a series of folds running obliquely downward; these folds are much fewer than the serrations and are set somewhat back from the cutting edge of the tooth, on its lateral flanks.

The teeth of this megalosaurid have been collected in the localities of Timimoun and Gara Samani (Algeria), the Kem Kem and Gara Sba (Morocco), Bahariya (Egypt), and In Abangarit (Niger).

Two teeth, collected by Faure in 1957 in the region southeast of the Aïr, were attributed to Carcharodontosaurus saharicus by Lapparent (1960, p. 28). One of these teeth (MNHN, Td.2250) was found 34 km southeast of the Elrhaz well (thus within the Gadoufaoua locality), the other 100 km southeast of the Elrhaz well (MNHN Td.2269). The first of these teeth is of the type that we collected at Gadoufaoua subsequently; it is quite different from the teeth of Carcharodontosaurus saharicus. The second is incomplete, relatively small, and does indeed show lateral folds in addition to the serrations of the crown margin. Now this second tooth, which does appear to belong to Carcharodontosaurus saharicus, in fact does not come from the Gadoufaoua locality but from the Akarazeras cliff (Faure 1966, p. 130), that is, from the GAD 8 level, higher than that of the Gadoufaoua locality (GAD 5); this GAD 8 level is probably Albian. The teeth of Carcharodontosaurus saharicus have therefore never to date been found in pre-Albian levels, and, until proven otherwise, this species is absent from the Gadoufaoua locality. 

b) Ungual phalanges. Several large claws have been collected at Gadoufaoua. The most impressive was found in 1965 north of the Petits Soldats fault (north of the Innocents level). It was isolated, lying on the reg [desert surface]. This claw measures 23 cm long. Its distal extremity is incomplete, and the total length of this ungual phalanx must have been about 29 cm. This piece, figured by Ricqlès (1967), was identified by our Polish colleagues Osmólska and Roniewicz (1970) as a manual phalanx, perhaps belonging to a giant theropod with a strongly developed forelimb, an example of which is provided by Deinocheirus mirificus, described by those authors from the Upper Cretaceous of the Gobi Desert. Deinocheirus was placed in a new family, the Deinocheiridae Osmólska and Roniewicz 1970, itself included in the superfamily Megalosauroidea Walker 1964. In a later publication, Rozhdestvensky (1970, p. 133, fig. 2) reproduces the figure given by Ricqlès, in the course of a study on Cretaceous theropods possessing large forelimb claws.

c) Postorbital. A postorbital, found in 1972, recalls that of the theropod Acrocanthosaurus atokensis from the Lower Cretaceous of North America figured by Stovall and Langston (1950, fig. 2, p. 700). This postorbital is smaller and slightly different from another postorbital found in the Albian of the northern Sahara, in 1971, during a joint expedition with Lapparent and Lefranc to the Gara Samani locality. Both of these postorbitals recall, despite differences, that of Acrocanthosaurus atokensis.

d) Neural arch. A neural arch, whose neural spine is incomplete but must have been of great length, was found in 1970 in the Innocents level. This neural arch is identical to the neural arches of the vertebrae figured by Owen in 1856 (Pl. XIX) and attributed by that author to Megalosaurus bucklandi. Huene (1923) subsequently attributed these vertebrae to Altispinax dunkeri. The neural arch found at the Gadoufaoua locality is likewise very similar to those of Spinosaurus aegyptiacus described by Stromer and originating from the Albian of Bahariya (Egypt).

The material from the Gadoufaoua locality attributed to this large carnosaur therefore approaches the pieces described under the names Acrocanthosaurus, Altispinax, and Spinosaurus. Now, these three Lower Cretaceous genera have been placed in a distinct family, that of the spinosaurids. The representatives of this family are relatively poorly known, but have in common the fact of possessing long neural spines. The carnosaur from the Gadoufaoua locality appears to be assignable to this family.

The study of the theropod material from Gadoufaoua, combined with that of the material from the other localities of the Continental Intercalaire of the Sahara, in particular from Gara Samani, will certainly help shed light on the family Spinosauridae.

Infraorder COELUROSAURIA - Family COELURIDAE - Elaphrosaurus iguidiensis Lapparent 1960

Numerous isolated bones of a small-sized theropod have also been collected at the Gadoufaoua locality. About ten small teeth, very flattened, serrated, with smooth lateral faces, small vertebrae, a few very gracile and hollow limb bones, metatarsals, and phalanges demonstrate the existence in this locality of a coelurosaur. Lapparent (1960, p. 30) attributed to Elaphrosaurus iguidiensis Lapparent 1960 a coelurosaur claw found by Faure southeast of Elrhaz. This coelurosaur is of much smaller dimensions than Elaphrosaurus bambergi Janensch 1925. The coelurosaur from the Gadoufaoua locality, by the dimensions of its claws and vertebrae, is very close to Aristosuchus pusillus (Owen 1876) (= Calamospondylus oweni), which is the only coelurosaur of the Wealden of Europe (Owen 1876, pl. 1).

---

Theropod entries in the faunal list

Order Saurischia, Suborder Theropoda:

• Infraorder Coelurosauria, Family Coeluridae: Elaphrosaurus iguidiensis Lapparent 1960
• Infraorder Carnosauria, Family Spinosauridae: Spinosaurid indet.

---

Theropod points in the general conclusions

The theropod Carcharodontosaurus saharicus, whose large, flattened teeth bearing both serrations and oblique external folds are so characteristic (see the chapter Inventory of the fauna, Theropods), has been found in the following localities: Bahariya, Egypt (Stromer 1931); Timimoun, Algeria (Depéret and Savornin 1927); Gara Samani (Broin, Grenot, Vernet 1971); Kem Kem, Morocco (Lavocat 1948, 1954); In Abangarit, Niger (Lapparent 1953); and Akarazeras (Lapparent 1960). These localities are situated very high in the stratigraphic series, just below the Cenomanian. The teeth of this carnosaur have never been found in levels earlier than those of these localities. All the theropod teeth found in earlier levels are different from the teeth of the type defined by Depéret and Savornin in 1927. All the localities listed above have yielded, sometimes in large numbers (Kem Kem, In Abangarit), theropod teeth similar to those of Carcharodontosaurus saharicus.

 

Stromer 1915 Select Translations

Passage 1:  comparison to normal theropod dentition (p. 11)

Among the dinosaurs, only the theropods come into consideration for comparison. They normally possess laterally compressed teeth, whose distinctly recurved crowns have a finely serrated edge both in front and behind. The number of teeth above and below apparently tends, as a rule, to differ somewhat, and seems often to vary between 12 and 20; their form and size, as well as their spacing, within a given species generally appear fairly uniform — only the rearmost teeth tend to be smaller.

Unter den Dinosauriern kommen nur die Theropoden zu einem Vergleiche in Betracht. Sie besitzen aber normalerweise seitlich komprimierte Zähne, deren deutlich rückgebogene Kronen vorn und hinten je eine fein gekerbte Kante haben. Die Zahl der Zähne ist oben und unten anscheinend in der Regel etwas verschieden und scheint häufig zwischen 12 und 20 zu schwanken, ihre Form und Größe sowie ihre Abstände bei einer Art scheinen im allgemeinen ziemlich gleichartig zu sein, nur pflegen die hintersten Zähne kleiner zu sein.

Passage 2:  the “predator” / tooth-function passage (p. 12)

The form I have described thus fits well, in tooth count, among the Theropoda named here, and was evidently, like them, a predator [Raubtier], in which the enlarged lower 2nd to 4th teeth and their upper counterparts played the role of canine teeth, while the small ones following behind them may have corresponded in significance to the weak anterior molars (diastema teeth) of some carnivorous mammals. In the particular simplicity of its tooth form, however, it stands apart from the normal theropods, and the aforementioned differentiations in size, as well as in the spacing of the teeth, speak for a certain specialization among the theropods.

Die von mir beschriebene Form reiht sich also in der Zahnzahl gut in die hier genannten Theropoda ein und war offenbar wie sie ein Raubtier, dessen vergrößerte untere 2. bis 4. Zähne und ihre oberen Opponenten die Rolle von Eckzähnen spielten, während die kleinen dahinter folgenden den schwachen vorderen Backenzähnen (Lückenzähnen) mancher Raubsäugetiere in ihrer Bedeutung entsprechen mochten. In der besonderen Einfachheit der Zahnform steht sie aber abseits von den normalen Theropoden und die genannten Differenzierungen in der Größe wie auch in den Abständen der Zähne sprechen für eine gewisse Spezialisierung unter den Theropoden.

Passage 3:  justification for the family Spinosauridae (pp. 27–28)

On the other hand, the peculiarity of the tooth form as compared to the norm for theropods, namely the absence of the distinct recurving of the crown and of the serration of its edges, as well as of stronger flattening, may even justify the establishment of a special new family, the Spinosauridae.

Andererseits kann die Eigenart der Zahnform gegenüber der Norm der Theropoden: das Fehlen der deutlichen Rückbiegung der Krone und der Kerbung von deren Kanten sowie von stärkerer Abplattung, sogar die Aufstellung einer besonderen neuen Familie, der Spinosauridae, rechtfertigen.

Passage 4:  the Djoua/Tunisia teeth misattributed to a fish (pp. 28–29)

The important question of whether associated remains are also known from other localities, I would like to answer for the time being only by repeating my cautious reference (1914, p. 42) to tooth remains from Djoua, south of Tunisia. From strata there, which he attributed to the Albian, and which in my opinion correspond in facies and age to the Baharije Stage, Haug (1905, p. 821, pl. 17, figs. 7, 8) described some teeth, which he questionably attributed to the fish genus Saurocephalus. They could, by their form and size, belong to Spinosaurus aegyptiacus, only their pulp cavity is rather wide, which, however, could be connected with the fact that they are not yet fully grown.

Die wichtige Frage, ob zugehörige Reste auch von anderen Fundorten bekannt sind, möchte ich einstweilen nur durch Wiederholung meines vorsichtigen Hinweises (1914, S. 42) auf Zahnreste von Djoua südlich von Tunesien beantworten. Aus dortigen Schichten, die er dem Albien zurechnete und die meiner Ansicht nach in Fazies und Alter der Baharije-Stufe entsprechen, beschrieb Haug (1905, p. 821, Taf. 17, Fig. 7, 8) einige Zähne, die er als fraglich der Fischgattung Saurocephalus zurechnete. Sie könnten nach Form und Größe zu Spinosaurus aegyptiacus gehören, nur ist ihre Pulpahöhle ziemlich weit, was aber damit zusammenhängen könnte, daß sie noch unausgewachsen sind.

Stromer 1936 Select Translations

The paleoethological remarks on Spinosaurus and Carcharodontosaurus

Regarding Spinosaurus and Carcharodontosaurus, a few comparative and paleoethological remarks remain to be made in conclusion. They show, namely, very strong differences in their quite peculiar dentition. In contrast to the norm for the Theropoda, the teeth of Spinosaurus are very little flattened laterally, and a front and back edge is scarcely developed (Stromer II 3, 1915, pl. I, figs. 7–11). What we are dealing with here, then, are very simple, pointed, conical teeth, following one another at small intervals (loc. cit., fig. 6). Such a dentition could not have served for tearing prey to pieces, but was essentially a catching-rake [Fangrechen], quite similar to that of crocodiles, in which, moreover, likewise strongly enlarged, canine-like conical teeth occur far to the front (see text-fig. 10B, p. 69). A peculiarity, however, is the unusually strong differentiation in tooth size.

Zu Spinosaurus und Carcharodontosaurus sind schließlich noch einige vergleichende und paläoethologische Bemerkungen zu machen. Sie zeigen nämlich sehr starke Unterschiede in ihrer ganz eigenartigen Bezahnung. Im Gegensatze zur Norm der Theropoda sind die Zähne des Spinosaurus sehr wenig seitlich abgeplattet und eine vordere und hintere Kante ist kaum entwickelt (STROMER II 3, 1915, Taf. I Fig. 7–11). Es handelt sich hier also um sehr einfache spitze Kegelzähne, die in kleinen Abständen sich folgen (a. a. O., Fig. 6). Ein solches Gebiß konnte nicht zum Zerfleischen von Beutetieren dienen, sondern war wesentlich ein Fangrechen ganz ähnlich wie bei Krokodilen, wo überdies ebenfalls weit vorn eckzahnartig vergrößerte Kegelzähne vorkommen (s. Textabb. 10 B, S. 69). Eine Besonderheit ist allerdings die ungewöhnlich starke Differenzierung in der Zahngröße.

In Carcharodontosaurus, on the other hand, there is a differentiation in tooth form, and the teeth are unusually broad and laterally flat, and those of the upper jaw are not recurved at all (Stromer II 10, 1931, pl. I, figs. 1, 2, 6a). It is precisely these latter teeth that now essentially resemble those of the well-known shark Carcharodon [the great white]. Such closely set teeth are naturally suited not only for catching and holding, but especially for cutting up large prey animals. Unfortunately, however, almost nothing is known of the mode of life of the living Carcharodon rondeletii, other than that it is an inhabitant of the high seas of warmer oceans. It could thus be a fish-eater, but perhaps it also pursues whales.

Bei Carcharodontosaurus aber liegt eine Differenzierung in der Zahnform vor und die Zähne sind ungewöhnlich breit und seitlich platt und die des Oberkiefers gar nicht rückgebogen (STROMER II 10, 1931, Taf. I Fig. 1, 2, 6a). Gerade die letzteren Zähne gleichen nun im wesentlichen denjenigen des bekannten Haifisches Carcharodon. Solche dicht gestellten Zähne sind natürlich nicht nur zum Fangen und Festhalten, sondern besonders zum Zerschneiden großer Beutetiere geeignet. Leider ist aber über die Lebensweise des rezenten Carcharodon rondeletti fast nichts bekannt, als daß er ein Hochseebewohner wärmerer Meere ist. Er könnte also ein Fischfresser sein, vielleicht aber auch Walen nachstellen.

Diet

As for the Theropoda now, Swinton (1934, p. 67) cites a finding that proves almost conclusively that Allosaurus fed on the sauropod Brontosaurus. We may thus well assume that the large Theropoda of the Baharije Stage likewise preyed upon the very substantial to gigantic Sauropoda found there. Since the latter, however, as we have seen (p. 62), probably lived amphibiously, the former may also have eaten other water-dwelling reptiles, and fish as well. Spinosaurus, indeed, could catch fish with its dentition just as well as crocodiles do, while Carcharodontosaurus, for its part, was also capable of cutting up the carcasses of very large animals. One need not assume, therefore, that these predatory dinosaurs lived far from the coast where their remains were embedded, and that they wandered there only on special occasions, such as during particularly severe drought, or that their carcasses were washed down by a river. This should be noted in relation to my remarks on p. 30. It shows how little unambiguous the evidence regarding the former living conditions of Baharije still is.

Was nun Theropoda anlangt, so führt Swinton (1934, S. 67) einen Befund an, der so gut wie einwandfrei beweist, daß Allosaurus den Sauropoden Brontosaurus fraß. Wir dürfen also wohl annehmen, daß auch die großen Theropoda der Baharije-Stufe den dortigen sehr stattlichen bis riesengroßen Sauropoda nachgestellt haben. Da diese aber, wie wir sahen (S. 62), wahrscheinlich amphibisch lebten, mögen jene auch andere wasserbewohnende Reptilien und auch Fische gefressen haben. Spinosaurus konnte ja mit seinem Gebiß so gut wie Krokodile Fische fangen, Carcharodontosaurus aber war auch imstande, das Aas sehr großer Tiere zu zerschneiden. Man braucht also nicht anzunehmen, daß diese Raubdinosaurier fern von der Küste lebten, an der ihre Reste eingebettet wurden, und daß sie nur bei besonderen Anlässen, etwa bei besonders starker Trockenheit, dorthin wanderten, oder daß ihre Leichen von einem Fluß herabgeschwemmt wurden. Dies ist gegenüber meinen Ausführungen auf S. 30 wohl zu beachten. Es beweist, wie wenig eindeutig noch die Befunde in bezug auf die einstigen Lebensverhältnisse von Baharije sind.

“Fangrechen” (literally “catching rake”) apparently has no clean single-word English equivalent.

“Raubtier” can mean predator/carnivore

 

Thumbnail made using artwork done by the amazing Jonathan Metzger, with his permission.