Remember the Alamosaurus?

I do! I recall my first encounter with the Late Cretaceous sauropod material dubbed Alamosaurus at the Smithsonian Institution on June 27th, 1995. I visited specifically to photograph and measure the caudal vertebral series of USNM 15560. This Late Cretaceous Utah North Horn Formation specimen was referred by Gilmore in 1946 to Alamosaurus sanjuanensis, the holotype of which was also at the Smithsonian. 

Gilmore originally (1922) named Alamosaurus from two bones found in the Ojo Alamo Formation of New Mexico. He designated USNM 10486 as the holotype, making it a scapulotype (aka a holotype consisting solely of a scapula :-)), and assigned an ischium (USNM 10487) found nearby as the paratype. For the 1946 publication, Gilmore compared the now three New Mexico elements (a caudal vertebra had been found by then) to the North Horn Formation, Utah material, which included a scapula and an ischium, and concluded that they were the same species. So things have sat ever since, a highly variable holotype bone upon the shoulders of which the entirety of Late Cretaceous North American sauropods had been ascribed to…

My dance with Alamosaurus began in earnest when I referred three bones excavated near the USNM 15560 Utah quarry in 2017 to Alamosaurus. While writing that short paper, I strongly considered making the North Horn material a holotype of its own genus, but I let it go. Why? I respected Gilmore’s decision and am a lumper by philosophy. So I left it alone, figuring Alamosaurus is like Diplodocus, we diplodocid researchers know YPM 1920 (the Diplodocus longus holotype) is a TERRIBLE holotype, so we vaguely acknowledge it and move on, knowing we are really talking about AMNH 223. That knowing wink is what everyone has done with the Late Cretaceous sauropod material. We haven’t found enough of a single skeleton with representative elements of the entire vertebral column to help us tease apart what is going on in the Late Cretaceous of North America, so we use Alamosaurus as a waste-basket name, a la Pleurocoelus of the Early Cretaceous (and Camarasaurus ;-) ?). Recently, I thought I’d found the proverbial proof of a second taxon, but, as you will read, re-search led me to realize I was wrong.

This blog turned into a novella! What started as a retelling of the history of this taxon became a commentary on variation and thoughts on the Paul 2025 Utetitan paper, in which the holotype was designated as the Utah North Horn Formation specimen USNM 15560, which I had considered naming in 2017.

The Gilmore Era

Alamosaurus sanjuanensis causes me heartburn. Why? As usual, the issue stems from the “Copper Age” of dinosaur paleontology, 1922 in this case. Charles Gilmore, a lad I deeply respect, named Alamosaurus from two bones collected from the Upper Cretaceous Ojo Alamo Formation in New Mexico. These were the first North American Late Cretaceous sauropod bones discovered. Here are the two bones in his paper; happily, he provided photos rather than simple outlines.

From Gilmore, 1922, Plate 1, the holotype left scapula USNM 10486

The holotype of Alamosaurus sanjuanensis is a single bone, USNM 10486, the incomplete left scapula shown above. Don’t have a scap to compare it to? Then you are out of luck! That is the root cause of this mess: a single, incomplete, distorted, highly variable element was used to erect a taxon.

Gilmore assigned a paratype, USNM 10487, a mostly complete right ischium. Here it is, shown from two sides.

From Gilmore, 1922, Plate 2, paratype right ischium USNM 10487

Alas, there is zero guarantee these two bones go together. Gilmore, in his 1922 paper, wrote, “While it is quite possible that both may pertain to the same individual, it is thought best to regard them as distinct until their closer relationship can be more positively established.” He ends his discussion section with “That both of these bones pertain to the same individual cannot be proven, but that both are Sauropod in aspect seems certain.” 

Both belong to sauropods; on that, he was right, but how can we test if they belong to the same individual animal? Compare this neat table of scapula-to-ischium length ratios I built from the sauropod assassin Gunnar Bivens's data. These are specimens where we know the two bones belong to the same individual. The problem, though, is that I have measured none of these, so we are comparing measurements between authors. We don’t know where they started or ended their tape measure, whether they measured straight, diagonally, or around a curve, or whether they estimated lengths in places. People typically report when they have estimated, but I can’t be 100% sure unless it is stated. We also can't be 100% sure they omitted the coracoid in the measurements. Thus, take this table with grains of salt; it is the best I can do until the same person measures each bone exactly the same way.

Table of sauropod scapulae, ischia, and the ratio between them within the same individual (numbers from Gunnar Bivens by way of the primary literature)

I used Gilmore’s conservative estimate of 170 cm (in red) because, in my experience, he seems to have been one of the more careful and observant of the monograph writers (note his Apatosaurus realizations ;-) ). The ratios betwixt the New Mexico and Utah specimens don’t match. Alas, I don’t have multiple specimens of the same taxon to determine the range of variation to expect across any of these titanosaurs. Does a .05 difference mean anything? Until I did this chart, I was thoroughly convinced the bones didn’t go to the same animal as they were found 200’ apart. Now I am not as sure, or at least I have an idea of how to test further: let’s see how variable these ratios are across known taxa, ideally from the same quarry. Emmanuel Tschopp provided me with measurements for two specimens of Diplodocus (CM 84 (0.76), USNM 10865 (0.75)). However, two specimens of Galeamopus (CMC VP 7573 (.66), unnumbered (.59)) differed substantially. I'm sure we have more Morrison Formation specimens I can test. Aside, what is going on with Baurutitan, a 1.0 ratio???

Discussions with various SauroSquad Europe members led to the hypothesis that an osteohistological analysis could be conducted. Work done over the last decade on this very topic has produced mixed results, but it is certainly worth serious consideration if anyone from the Smithsonian is reading this.

What can we do with this info? Well, considering Gilmore called the ischium a paratype, not much, since a paratype simply means it was a nearby bone and has no taxonomic status to help us out. But I like such thought experiments, if for no other reason than we can perhaps come up with a ratio to describe a character rather than relying on qualitative descriptors. More on that later…

Why Name a Dinosaur From a Single Bone? Gilmore’s Tangled Web

The finding of a Late Cretaceous sauropod dinosaur bone in North America was unique in the 1920s. None had been found there, yet they were known from other parts of the world. That's why Gilmore in 1921 wrote a brief blurb that appeared in the journal Science about the discovery of an “almost complete” scapula of “fairly good state of preservation” with a “close general resemblance to the described scapulae of the Sauropoda” which “have not been known before above the early Lower Cretaceous in North America” that “is of the greatest interest.” He advised that he would provide a more detailed description once it was prepared, which he did the following year.

In his 1922 description, he noted that the scapula's shape differed from those of Camarasaurus, Diplodocus, and Haplocanthosaurus, correctly using sauropod scapulae for what they are best used for: family-level delineation. Alas, their tremendous variability (individual left-to-right-side variations, within species, and taphonomic, the latter due to the innate challenges of properly preserving a long, thin, semi-curved element without various deformation) makes their utility beyond family ID all but useless (just strike up a conversation with Dr. Ray Wilhite if you’d like to get *lots* of specifics… make sure you allocate a sauropod-sized block of time!). I’d certainly not hang a new genus off of a mostly complete scapula (don’t get me started on Supersaurus; that is a pain in my neck… and shoulder).

The as-preserved length of 155 cm (5’) led Gilmore to note the bone “rivals the largest of the Camarasaurus scapula,” and he gave it a “conservatively estimated” total length of at least 170 cm (5’ 8”), making it a veritable giant among sauropods of the time. It is that number, 170 cm, that I used in my table above.

He designated an ischium as a paratype, which is code for “Hey, we found this bone near(ish) the holotype, and we can’t prove it goes with it, but we thought it’d be cool to point out, even though it has no bearing on the holotype’s status.” Yes, the ICZN doesn’t phrase it that way, but that is what it means in practice. 

Aside, the ICZN doesn’t care if one’s characters are valid when naming a taxon, just as long as one has characters and puts the holotype in a ‘collection’ (which can be a closet in my house…). This fact about the ICZN means that Paul’s 2025 renaming of the North Horn Formation material USNM 15560 to Utetitan will likely stand the test of time. I suppose a successful petition to designate USNM 15560 as a neotype for Alamosaurus would “sink” the new taxon, Utetitan. Still, I’d wager heavily it wouldn’t be approved, if for no other reason than the holotype isn’t lost and too much has been made of the presence of two ventral “runners,” which may very well be variation or taphonomy.

Sauropod ischia are quite useful for family-level sorting; Gilmore’s struggles to identify USNM 10487 as an ischium reflect their complexity and utility. He wrote in 1922, “This bone differs so from other Sauropod ischia that its true nature was determined with difficulty.” Titanosaur ischia are decidedly different from the Morrison Formation specimens he was familiar with, and it took him a bit to suss out what it was. Ischia are orders of magnitude better than scapulae for determining genus, but not so significant that, if he had made it the holotype, our Alamosaurus lives would be any better, even though we have many more ischia than scapulae at our disposal for comparison.

Gilmore noted that many other bones were found with these elements. Alas, I know not of any quarry map that was created of this field of broken dreams. He wrote, “Directly associated with the bones of Alamosaurus are many other fragmentary and undeterminable dinosaur bones, teeth of carnivorous and Ceratopsian dinosaurs, dermal plates of an armored form, turtle fragments, and crocodile bones.” Some of these elements have been recognized as belonging to a sauropod and are logically referred to as Alamosaurus. Check out Mateer's 1976 paper, where he designated as topotypes an ilium fragment and an articulated last sacral/first caudal/partial second caudal vertebra section (the latter seems to have been misinterpreted and is actually three sacral vertebrae). A topotype is a bone collected from the type locality.  Like the paratype ischium, topotypes have no "legal" standing in determining what is/isn't an Alamosaurus.

In 1929, von Huene wrote [my translation], “In North America, apart from Alamosaurus in New Mexico, no Upper Cretaceous Titanosauridae are known.” This is the first time the New Mexico material was referred to a family, as Gilmore didn’t do so, which surprised me. That family, the Titanosauridae, was named by Lydekker in 1893. I have pasted the note in the image below.  

Lydekker's 1893 creation of the Titanosauridae

In 1938, Gilmore wrote, again in Science, that a 1937 Utah expedition in the Late Cretaceous North Horn Formation “made a discovery of more than ordinary interest.” (I love how they wrote ‘back in the day’!). He continued, “The Sauropod specimen is of special interest not only because of its unexpected appearance here, but also because it pertains to the family Titanosauridae, the first recognized occurrence of this family in North America.” He noted that, “...comparisons as have been possible appear to show its affinities to be near, if not belong to that [Alamosaurus] genus.” and advised he’d be describing the specimen in greater detail. Perhaps he wasn’t aware of von Huene’s 1929 work, which had already referred the New Mexico material to the Titanosauridae. He does reference it in his posthumous 1946 work, though.

In 1946, Gilmore (posthumously) published on what turned out to be a spectacular sauropod excavated from Utah’s Upper Cretaceous North Horn Formation. He even noted in his introduction section, “(3) it permits for the first time an adequate diagnosis of the genus Alamosaurus,” meaning he knew full well he was causing the world trouble by his single-bone naming back in 1922. He further confirms this sentiment by writing in the section describing the holotype, “The genus Alamosaurus was established on a meager specimen… consists of two bones, a left scapula and a right ischium, somewhat incomplete and perhaps pertaining to two individuals. The rather unusual procedure of founding a genus on such scanty evidence was justified on the ground that this was the first authentic record of the occurrence of sauropodous dinosaur remains in the Upper Cretaceous of North America.”

I applaud his honesty, but why not simply announce “first occurrence of a sauropod in North America’s Late Cretaceous” in 1922? Alas, the practice of naming a dinosaur from a single bone continues to the present day and will persist as long as paleontology exists, because naming a specimen draws far more attention than one assigned to a family without a moniker. It immortalizes the namer, and, especially in today’s media-driven world, accompanying it with splashy artwork can give a new taxon a few seconds of a viewer's attention, possibly leading to future funding. Cynical? Yeah, re-reading it certainly makes it sound that way, but I’d position it as simply a fact. Feel free to ascribe whatever emotion you’d like :-). My goodness, I love being able to ramble in my own blog… cathartic!

Gilmore continues in the 1946 paper, “The present specimen [USNM 15560], is in close agreement with the type in all important particulars, and as the slight differences observed could not be considered of specific importance, I refer this specimen to Alamosaurus sanjuanensis.” Gilmore doubles down on his ID, “Contributory evidence of the correctness of this conclusion is furnished by a second specimen from the type area–No. 15658, United States National Museum. This specimen consists of approximately the twenty-first caudal centrum and a caudal neural spine,” collected in the same general area of New Mexico. “These bones are in perfect agreement with the caudal vertebrae of the Utah specimen, down to the smallest details.” Sigh. Where to begin?

I have chased tail around the globe and can say a “twenty-first caudal centrum” doesn’t provide much info past, “yep, a titanosaur died here.” Gilmore, you are better than that! His logic is as follows: “I found a New Mexico Late Cretaceous sauropod scapula, which I made the holotype of a new genus because none were known before from here. I found an ischium that I don’t think goes with the individual I named the holotype after, so I’ll call it a paratype. I know I shouldn’t, but I am doing it anyway. Thankfully, I found this cool skeleton that has both elements, but to seal the deal, I’m going to take a caudal vertebra found in the same general New Mexico area and say, since it matches this new cool Utah specimen, we now have a great match!” Yes, Charles Gilmore, that is one possibility; another is that you have two taxa. The fact that we can’t know is why we shouldn’t name dinosaurs from scraps to begin with.

I have had the distinct pleasure of photographing and measuring USNM 15560, the Utah specimen that is now the Utetitan holotype. It consists of 30 articulated caudal vertebrae, whose positions we know because the first biconvex caudal is present. 25 chevrons, a pair of ischia and paired sternal plates, a left scapula with an attached coracoid, a complete right forelimb, and three ribs round out the material. Gilmore wrote, “Although not collected, the sacrum was observed in the field…” which led Jim Jensen to track down the locality, thinking he could collect it (he couldn’t), and me writing a brief 2017 note describing the colorful history of relocating the site as well as providing figures for the proximal ends of the humerus and femur Jensen collected from nearby. It was while writing that paper that I seriously considered giving the North Horn material a new genus name and/or making the holotype a nomen dubium. My Jack McIntosh training precluded me from doing so. I kinda wish I had now :-).

If you are playing along at home, you'll see that Gilmore chose to retain his taxon because he felt the scapulae and ischia were identical. Let’s take a deeper dive into Gilmore’s comparisons. He noted his “about 76 inches, approximately 2 meters” Utah scapulocoracoid differs from the New Mexico holotype in that the latter “has a more gradual slope to this edge, as shown in plate 10, figure 1.” He wrote, “In general the scapula closely resembles the type of Alamosaurus sanjuanensis, differing only in a few details such as having a more sinuous posterior border, a more rapid slope from the thickened spine upward to the border, and slightly smaller size. None of the differences observed could be considered as more than individual variation. For that reason, substantiated by the close resemblance of the ischia to the paratype, this specimen is referred to Alamosaurus sanjuanensis.” 

I found it rich that Gilmore took bones from (possibly) two different individuals from the same New Mexico locality, mixed in a caudal vertebra from a nearby (same?) locality, and combined them into one taxon, all while chiding Huene’s 1929 Titanosaurus australis scapula for being assigned to Titanosaurus “...because of the scattered occurrence of the South American materials, the differences in the scapula would suggest an accidental interchange of skeletal parts.” At least he used ‘accidentally’! He took a similar shot at Lydekker, “Insofar as one can determine from the illustrations and description of the meager type materials, the Indian Titanosaurus and the North American Alamosaurus seem to be generically identical…” So why didn’t he rename his specimen Titanosaurus in 1946? Gilmore continues, “...yet it must be kept in mind that other parts of the skeleton might display points of distinction.” He didn’t take this advice in his own work. Instead, he noted “...the wide geographical separation leads to a decision to continue the use of the term Alamosaurus to denote the North American representative of the Titanosauridae.” Define wide, Charles! The distance from New Mexico to Utah could be considered wide by some…

For fun, here is an image of the Titanosaurus indicus holotype from Mohabey et al. 2013 (a login is required) compared to a 1995 shot I took of USNM 15560 caudal 12. I certainly wouldn’t say identical.

Falconer’s posthumous 1868 images (the three drawings) plus the brown image (top left, from Mohabey et al. 2013) are of the same vertebra, the holotype of Titanosaurus indicus, compared to USNM 15560 caudal 12.

Note the danger of illustrations, the Titanosaurus indicus prezygapophyses aren’t that long in person. Perhaps they were lost over time?

Alas, not the same angles, but USNM 15560 is certainly wide. How much crushing has T. indicus experienced, if any? (That is foreshadowing about a project Dipanjan Munshi and I are working on :-)).

Regarding the ischia, Gilmore felt “...the closest resemblance was found with the paratype of Alamosaurus sanjuanensis…” and “These bones so closely resemble one another in all of their principal features that no doubt is left of their generic identity.” I’ve included photos below in this article… You be the judge.

My takeaway? Gilmore chose to name a scrappy holotype and knew he was doing it. The fact that he took such pains to name the North Horn specimen as Alamosaurus tells us what? I’m not sure. Why didn’t he name it a new genus? Perhaps he desired to keep it Alamosaurus out of spite? He wrote this in his 1938 note to Science (emphasis mine):

“It should be recalled that Alamosaurus was the first sauropod dinosaur to be recognized from the Upper Cretaceous of North America, and the skepticism with which that announcement was received may now be dissipated by this second discovery under circumstances that are even more convincing than the first, if that is necessary.”

He could have happily named it a new genus, and no one would have squawked. Maybe he wanted to say, “Told ya so!” to his detractors? Does anyone have published detractor comments? I don’t recall reading any that specifically called Gilmore out for naming Alamosaurus. Perhaps in the hallways of conferences?

Et tu, Utetitan?

Thus things stood until Paul (2025). Numerous authors (check out Paul 2025 for a nice list of them) have published on nearly all aspects of the skeleton of sauropodous material found from the Late Cretaceous of North America. All of them use a ‘?’ or ‘cf.’ (for compares favorably) or straight-up call the bones Alamosaurus because they were mature enough to know a single or even a few bits of element X or elements B, C, and D weren’t appropriate to name a genus from (see Gilmore, that’s how it is done!). USNM 15560 does have enough to name a genus; of course, it lacks dorsals, cervicals, and lots of other bits, such that the beautiful cervical series on display in Texas (BIBE 45854) at the Perot Museum can’t be compared. In the “Copper Age” or even “Silver Age,” I’d wager that the cervical series would have a name; heck, I’d bet most folks would tack a name on those verts. Thank you, Tykoski and team, for not doing so. In light of recent events, perhaps now is the time…

Yours truly at the Perot with some sweet ?Alamosaurus cervical vertebrae (BIBE 45854)

Philosophical Musings

Should Gilmore get a bigger vote?

If there is any human being who could determine whether the New Mexico and Utah bones belong to the same species, it is Gilmore, the guy who named the New Mexico scapula Alamosaurus. Perhaps he chose characters that are not what we now consider optimal, but he originally named Alamosaurus. He was convinced the Utah and New Mexico material belonged to the same species. This guy wasn’t just your average clown on the street; this was an expert in the twilight of a long, storied career who certainly knew what he was doing. It seems, if nothing else, disrespectful that we are ignoring his explicit wishes by splitting the material into two taxa. Now I don’t know if the ICZN even existed when all of this went down, but if it did, I’m sure he would’ve taken the appropriate steps to prevent the actions that recently transpired. I ask you, if the expert who named the first one and set the second one side by side with the first made the claims, then I can’t see any clearer scenario than leaving them both as the same taxon, especially considering Gilmore was aware of the impact and importance of variation. It is an interesting conundrum, as our discipline gives scientists so much leeway in deciding whether something is a new genus and species. Because Gilmore named the taxon and declared these were the same species, shouldn’t that carry weight?

However, Matthew Mossbrucker (conveyed to me via Mathew Wedel) points out that the “Copper Age” [my term] paleontologists didn’t have a grasp of the actual depth of time. If Gilmore thought the world was 10 million years old, then how could there be enough time to evolve numerous genera, considering he was aware of the range of life, from ancient invertebrates to Pleistocene mammals? Could that be why the “old timers” use species so much? They simply didn’t have the benefit of understanding that we have today, and didn’t think there was enough time for so many genera to evolve. Perhaps if they did, they would have named more genera? Think Gorgosaurus lancensis, the name Gilmore gave to CMNH 7541, the holotype of Nanotyrannus lancensis. Perhaps today, he would have initially given it a new genus name rather than making it a species of Gorgosaurus?

Variation rant alert

There is, to me, an almost complete disdain for the massive range of variation present in extant taxa, which might be a direct result of how we teach anatomy. If one is lucky, one gets an actual cadaver. Once. For a few months. Soft tissue often varies; these are highlighted to med students (it is essential that *they* know that when they cut organisms open, parts might not be where they “should” be), but osteological variation? Nah, not emphasized. Thus, paleontologists from a traditional medical school track may not fully grasp the actual extent of variation. 

Those few who go on to teach human anatomy begin to see, as they grey, that osteological variation “is a thing,” and is greater than what we are taught in our classroom days. Veterinary anatomists see many forms of variation; just ask Dr. Ray Wilhite, who has collected hundreds of examples of variation in alligators, dogs, and other critters.

Perhaps if each anatomy student were given a giant box of examples of bony variation, everyone would be much more cognizant of it. Variation destroys a cladogram (ok, it can destabilize placements, but I like the hyperbole :-)). I know they are built “robust” such that one character shouldn’t break one, but a single character can move taxa, and it is the placement of specific specimens that we end up disputing (Cactus Park skull, anyone ;-) ). As so many characters are susceptible to variation, what to do?

Holotypes are for what, now?

What is the point of a holotype? It lets me know when I have found another of the same species. USNM 10486, the holotype of Alamosaurus sanjuanensis, is the only bone that can be compared to determine whether it belongs to the same taxon. Thus, Gilmore’s fantastic luck when USNM 15560 was found with a scapula. Gilmore opted to make them the same species. Was this because of his philosophy of time? Would he have used a new genus if raised in today’s world? Was it because his tolerance for variation was greater than some folks have today? Perhaps he observed enough scapular variation, noticed the “smooshing” on USNM 10486, and concluded it caused that furrow, thus indicating they were within the same species tolerance? Maybe he was being spiteful? Impossible to know, but likely at least one of those ponderings is correct.

Should A. sanjuanensis receive a neotype? Be dubbed a nomen dubium? For the latter, well, it depends on how much value one places on the scapula. Given my experience, scapulae are far from ideal as holotypes. Thus, what to do? If you say the scapula is too variable and the characters Gilmore used are invalid, then I suppose the name could be done away with. If we struggle over comparing USNM 15560 with USNM 10486, how useful is the holotype? As for a neotype? The holotype isn’t lost. I’d wager Gilmore would have happily made USNM 15560 the holotype if he’d found it first. However, if he had, it wouldn’t have been called Alamosaurus. Thus, we end up in the same situation as USNM 15560, receiving a new name :-). My preference is that USNM 10486 be treated in whatever manner is appropriate when the holotype is non-diagnostic below the family level.

As my friend and brilliant taxonomist and philosopher of science, John D'Angelo, reminds me, there is a difference between taxonomic and nominal species. A taxonomic species refers to a single biological species and is a testable hypothesis, "these animals produce viable offspring," whereas a nominal species is a name attached to a holotype. The nominal species, regulated by the ICZN, exist to standardize the application of names. In practice, John argues that we should start with the taxonomic species concept and determine which skeletons belong to the same species, then apply nominal species rules as "this name goes here." He also reminds me, "The most accurate taxonomy is no taxonomy at all." Like I said, brilliant!

Why this blog now?

In the last two years, I have spent far more time than I ever thought I would with Late Cretaceous sauropods from North America. It started in April 2023 when Dr. Cristy Lopez and I visited the Perot Museum, and Dr. Ron Tykoski showed us around the annex. Whilst on our way out, I stopped dead in my tracks, for I saw this bone on the back of a pallet tucked deep in the dark.

That looks for all the world like a diplodocid femur (Diplodocus femur above), or at least a non-titanosaurid femur. Below is what a titanosaur femur from Texas looks like, a big, robust beast as befits a Texas stereotype!

We were hurrying to catch a plane and vowed to return. We did, in November 2023, and it forms one of the pillars of a talk I like to give about why we call what we do re-search. We searched again, this time with the specimen placed in the bright lights of the wide-open lab floor. That is when we discovered this:

It had been restored with a narrow shaft and then painted. In the dark, from a cursory pretzel-twisted view, it certainly looked original. But in the light, it was obvious what had transpired. Examples like this are why I fret over research conducted solely from surface scans, photographs, or even videos. There would have been no way to know for sure that such restoration was present without personal observation (or a CT scan). I went from thinking I had found the smoking gun that would allow us to definitively declare a second sauropod family was present in the Late Cretaceous of North America to being glad I (mostly) kept my mouth shut.

Defeated but not discouraged, we continued this re-search run by way of UT Austin, where I was able to study the "Rough Run" specimen. If one can make an argument that there is a second taxon in the Late Cretaceous of North America, this is the one! However, without a scapula, one must play Frogger, leaping from the Gilmore holotype to the North Horn specimen, accepting it is Alamosaurus, and then showing why it is different, all on a variable element. Ugh!

The above was the extent of my Alamosaurus interest until October 9th, 2025, when I received a copy of Paul 2025, which described the North Horn material under a new genus name, Utetitan. Someone had done what I had mulled over! I was excited to see what characters, i.e., what they used to define the new taxon. Well, that didn’t work out as well as I had hoped...

What left me dumbstruck about the paper was the characters Paul offers up to tell me when I’ve found another Utetitan. After all, to me at least, the point of a diagnosis is to emphatically and unequivocally know when I’ve found another of what is being described. Remember, from my experience, scapulae and ischia are best used for family-level ID. But since all we have is the holotype scapula to “legally” compare to, I was hoping for something I’d missed, some clever non-variable, clearly defined character(s). Alas, I failed to find what I was looking for.

The given diagnosis is as follows, along with my thoughts underneath each bullet. I must add that I haven’t meaningfully touched either of the scapulae or ischia discussed since I last saw them when I was but a pup in 1995 (and was utterly disdainful of all things appendicular at the time, so keep that in mind). However, I spend my days looking at vertebrae, have scored my fair share of skeletons, and know troubled characters when I read them.

Scapular characters from Paul 2025

1-Acromion process strongly prominent. 

What is “strongly prominent,” and how would I know when this character is a known variable one?

2-Posterior profile of scapula strongly sinuous so a prominent triangular glenoid process is present. 

Another not useful-to-me character, especially in light of sauropod variation and the presence of distortion.

3-Paired tubercles dorsal to glenoid fairly prominent and widely separated. Maybe this is useful? 

Remember, a sauropod scapula is a huge, thin bone with various soft-tissue actions transpiring; the variation is amazing. I’m not sure when something is “fairly prominent” and not simply “prominent.” Does the presence of two tubercles matter? They exist in other titanosaurs. I will make sure to photograph the tubercles on each scap up close next time I am in DC.

4-Flare of postero-dorsal edge of blade more prominent.

Yet again, the word “prominent” is used without a definition or range. 

Ischium characters

5-Lateral process of ischium modest in size. 

What is modest-sized? How would I know? Give me a ratio at least!

6-Postero-lateral profile of ischium moderately concave arced. 

What does moderately concave even mean?

Femur

7-Femur robust? [sic]

Is one permitted to allow a guesswork character? Define robust! This one absolutely slayed me... (me trying to use hip new lingo doesn't always land according to my kids :-)).

Let’s tally up the quantitative capacity of those characters: Zero. I will be the first to admit I am a qualitative lad through and through. I find myself knowing when something is different and asking my compatriots to “trust me, it is more robust.” That is fine in conversations (well, fine by me, I’m not sure the SauroSquad is always as enamored of it), but as a holotypic diagnosis? That shouldn’t be permitted to fly in 2025.

How does one have thirty mostly articulated caudal vertebrae, twenty-five chevrons, and an essentially complete forelimb yet provide no characters to help diagnose the genus from those bones??? Yes, titanosaur caudal morphology is mostly conservative, but was there not a suite of characters that could be used? Surely the forelimb and tail have *some* diagnostic characters?

I thought Paul's emphasis on scapular and ischial characters was meant to differentiate it from the Alamosaurus elements, but then a femur appeared in the list. Why a femur but not the caudal vertebrae or forelimb?

Gilmore (1946) mentioned a New Mexico caudal, USNM 15658, which matches well with the Utah specimen USNM 15560. Here are some photos from my 1995 trip. Why not compare the caudal vertebrae as I did below? It feels criminal that the tail was ignored. Granted, I love tails and am thus biased, but I'm sure appendicular folk feel the same about the limb material getting short shrift.

USNM 15658 photo (grey, left) from New Mexico, which Gilmore (1946) says resembles the USNM 15560 Utah specimen (#21, right). Keep in mind the posterior ball can vary in shape within a single individual.

Even more unforgivable to me, yet one I shouldn’t be surprised by, considering the author is an artist as well as a paleontologist, is the intentional omission of photographs in favor of his outlines of elements selected. Disconcertingly, Paul twice notes that photos that don’t match Gilmore's scapula illustration shouldn’t be trusted. At least that was how I read his comments. This is a massive problem because I have learned in my travels that the “Old Timers” often had their scientific illustrators illustrate the bones as they should be, not as they are. I am convinced Gilmore did this to some extent in his 1946 paper. 

To test this, I took photographs of the elements wherever possible and laid them against the line drawings of the two Paul 2025 figures. Disagreement occurs only between the pre-1950s line drawings and modern photos of the same bone. Damage? My preparator hat says, “Not that much!” I then compared an illustration from Cope with those of Osborn and Mook for the same bone, and they aren’t exact matches. Trusting drawings is worse than trusting photos and 3D scans, neither of which can replace physically handling the bones with intention and/or a CT scan.

Err… was this paper fact-checked?

The Paul 2025 paper has an inordinate number of errors. I started writing them down as I was reading, but I gave up on my first pass; there were too many. I will list those below that quite surprised me, but for the pedant in me, see the Errata screenshots at the end of this novella. However, because I had typed some up, I am also going to share those via my writer’s prerogative; you can skip if you’d like.

Paul (2025) writes, “paratype pubis” in his Introduction paragraph, right after he cites the year of my paper erroneously (Curtice is cited as 2016, but the paper’s official citation is 2017, as you can see here). I am three sentences in and already have two errors, one extremely severe (pubis???). This start isn’t boding well.

On the second page, I see Figure 1, a figure I can’t abide whatsoever, despite the author's claim, “The new illustrations are the most accurate images of the fossils yet produced.” WHOA! I had to read that three times to make sure I was understanding that the outlines I see in Figure 1 are the most accurate figures of any of these elements... Outlines are notoriously dangerous, especially in this age of digital capabilities; there simply is no need for them solo. Do they make excellent points at times? Absolutely. But they need to be accompanied by actual photos. To that end, I painfully rebuilt the figure, as you will see below.

He doesn’t list important specifics about the outlined specimens in Figure 1. For instance, I can’t figure out Figure 1C, the complete cervical he illustrates, as it just says “Figure 3A in Tykoski and Fiorillo, 2017” but their Figure 3A is comprised of *nine* cervical vertebrae, and position matters when making any vertebral argument. Despite my efforts, I couldn’t identify which specific vertebra he traced, nor how he could trace them so clearly, as they are shown mounted in a series. 

Similarly, his Figure 1D states he is showing “Anterior caudals in posterior view”; however, he seems to have copied the anterior views of Tykoski and Fiorillo 2017 and Gilmore 1946, as both centra outlines match the anterior views that are also figured. I highlight those below.

Notice that in Figure 1A, I pasted actual photographs of the scap, and they don’t match the original. He claims the “bone is not as complete,” but I find that nigh impossible to believe; instead, I'd wager Gilmore had his artist draw more than what was present, thus illustrating the dangers of not touching the actual bones in person. His proof that the bone has been somewhat degraded is his request for the reader to examine the line drawing in Figure 6 of Gilmore 1946 and accept it as fact. No thank you! Not after the nonsense of the Marsh plates and many Cope specimens I have observed, much less Williston and others, where the drawings don’t match the bones. The "Old Timers" did not produce photographic-equivalent drawings every time, and that they sometimes did means one has to scrutinize the bones personally... 

The Figure We Deserved

I tracked down the images Paul cites in his outline figure. I don’t want to comment on how long this took, other than to say “You are welcome” :-).  

Modified from Paul 2025 Figure 1.

The reason the “D” figures are so off is that he wrote that his outlines are of “Anterior caudals in posterior view.” However, it seems the tracings he provides are anterior views, not posterior views, at least for the second and third specimens. 

The first (farthest left) is from Fowler and Sullivan 2011, Fig. 2B, and is traced in posterior view. Considering that no anterior view was provided, this one matches his figure caption. 

The traced middle caudal vertebra image on D is from Tykoski and Fiorillo 2017, Fig. 10, specifically 10A, anterior view. Had Paul traced Fig. 10C, he would have copied the posterior view of the first caudal vertebra, and that is the vertebra’s image I am using to overlap his image below. 

The caudal vertebra on the right is from Gilmore 1946, Plate 8, number 2 (caudal 2). That plate consists entirely of anterior views. 

As you can see, providing the anterior views of the middle and right vertebrae provides a much better match (lower image).

I was unable to replicate Paul's ischia overlap, but, since it isn’t germane to my writings (some of the ischia are quite close morphologically to others, but how variable? Crushed?), I won’t be diving much deeper into them. Suffice it to say, they are a rough bunch, as you will see in some figures later. Perhaps he actually copied from a different paper than the one he says he used?

More Errors

Paul 2025 could have finished the paper on page three, the second paragraph in the paper, with what he wrote, “Diagnosing a species based on fossils from a different geographic and stratigraphic location is inherently problematic and must be considered provisional until the systematic unification is verified, or perhaps more probably refuted, on morphological and stratigraphic grounds.” He is echoing diplodocid workers and our loathing of YPM 1920 as the genotype for Diplodocus. We all overtly (or implicitly) acknowledge YPM 1920 isn’t diagnostic beyond “diplodocine” but thanks to the ICZN (the self-appointed group that sustains such ludicrosity) the petition to substitute the type specimen was denied. Keeping the North Horn specimen as Alamosaurus [insert a new species here] would have been far more conservative. I haven't figured out what I, as a researcher, have gained by calling the North Horn Formation material a new genus. Anyone?

In Paul's ensuing paragraph, there is this headscratcher of a sentence, “Even taking possible over splitting of taxa into account, differing stratigraphic levels each feature an array of diplodocid, camarasaurid, and brachiosaurids…” yet there is for North America a single genus, Camarasaurus of camarasaurids, and a single (for now ;-) ) genus of brachiosaurid, Brachiosaurus, in the Morrison Formation. Do I think there will be more of those two groups named? Likely. But as of now, they aren’t, and therefore two of the three major groups of sauropods in the Morrison Formation (the other being the diplodocids) are monogeneric for anything remotely considered complete specimen-wise. Did he mean the Cretaceous sauropod landscape? I am confused!

He has, to me, an odd take on his Figure 4, which outlines several Camarasaurus supremus scapulae excavated from the same quarry. He argues, “If divergence in the North American titanosaur scapulae is greater than in C. supremus, especially from specimens from differing stratigraphic levels, then that is evidence they do not represent a united taxon.“ 

The Camarasaurus scapula example establishes the bare minimum of variation one can expect in a taxon, not the maximum. I’ve included a bit below an Apatosaurus as a nice example of variation in scapulae (thank you, Ray Wilhite, for the raw images!).

Oh yeah, B and E are given the same specimen number in Paul (2025) Figure 4. It took me a bit to straighten that out. Did none of the reviewers catch such errors? Or did he choose not to fix them? It is puzzling.

I overlaid (in red) 5760-3 from Osborn and Mook 1921 Fig. 77, who thought it was the mate of 5761-2, Fig. 76 (blue outline). However, they note the blue-outlined one is “badly crushed.” Notice how “splorchy” the non-crushed one is, and how well the shafts and distal ends align. It is the proximal side where crushing seems really to have taken its toll.

The red outline is Cope’s sketch of 5760-3 (Osborn and Mook 1921, Fig. 10). The black and grey illustration is a re-drawing by Osborn and Mook (1921, their Fig. 77). Notice how the two images don’t agree. I could never align all the edges. Yet another example of the danger of illustrations, as both are of the same bone, yet disagree on the angle of the mid-shaft, the amount preserved at the proximal end, and the shape of the distal end.

The blue outline above is of USNM 10486, the holotype of Alamosaurus. The red is from USNM 15560, the specimen Gilmore said was virtually identical, but Paul 2025 designated as the holotype of Utetitan. Perhaps Gilmore needed glasses, because even applying the world’s strongest parallax, I can’t get these two to match. However, to some, this is well within the range of variation for sauropod scapulae (I might be one of those folks :-)).

The left shows photos of USNM 10486 from Jasinski et al. 2011 overlain with Gilmore’s USNM 15560. The elements on the right are two Apatosaurus from different individuals at a single-taxon locality. Variation in scapulae… am I right?

I aligned the outline of Paul’s USNM 15560 with the photograph from Jasinski et al. 2011 Fig. 12, using the curve above the break as my “truth.” As you can see, I couldn’t get the elements to line up. Notice the white area on the medial side. You wouldn’t expect the above image to have such a smashed side at all (see the photos below). That side could easily have resulted in an artificially deepened area. In fact, I don’t see how it couldn’t have resulted in some deformation. I’m wondering, too, how much of the coracoid is missing. Notice Gilmore drew it as complete, even though it isn’t.

I use Jasinski et al. 2011 and D’Emic et al. 2011 photographs of USNM 10486 to show the impact of parallax and pixels when building outlines, while also showing the overall consistency of the photos. They also show off how much restoration (the white at the bottom of the image on the right) is present. You can see the white on the left, but you’d never guess how much was present until one flipped the scapula over. At which point you have to ask yourself, how much of the scapula has been deformed? It doesn’t preserve any curve, which impacts morphology. Add that much deformation to a "bendy" bone and I’m not inclined to trust any shape-based character(s)…

Regarding scapular variation, Paul 2025 writes about D’Emic et al.’s 2011 note that USNM 10486 and USNM 15560 have differences attributed to distortion, “...but the exact basis of this assertion is not documented. I have examined the specimens and did not observe any distortion after directly comparing the two bones.” Alas, now I have to visit these specimens at the Smithsonian as I last saw them in 1995 and, being those were the days of film and I was studying caudal vertebrae (notorious consumers of film with each requiring six views per vertebra, and film was not inexpensive to a starving graduate student), I took no photos of them myself, and don’t trust my memory of if they are distorted or not. The images above, though, certainly lead me to think otherwise!

I disagree with Paul’s (2025) comments that the differences are enough to justify splitting them into two genera; the variation is just that, variation within a variable element, and they are minor enough characters that I reject the bone’s value for this level of splitting. Titanosaur? Absolutely! Using it to name a genus, NOPE! I know he appeals to the fact that these differences are greater than the outlines of the scaps show present within C. supremus scapulae, but that is a red herring; those are the absolute minimums that can be expected to be observed based on mine and other SauroSquad members' personal experience. Paul makes this very point: “the morphological divergences are sufficient to indicate two different taxa even if the bones were found at the same level of the same formation,” and I counter by imploring everyone study the Cactus Park collection (or get Ray and I to finish our work!) and see how an MNI of eight Apatosaurus have stunningly wild differences amidst appendicular elements. So much so that, if found in separate locales, they’d be named as at least two genera, when they are simply variations within the same species. Takeaway message? Scaps = bad! Ilia, too, being “pelvic scapula,” suffer from horrendous taphonomic variation. He mentions ilia that are different, and I’d say, “Yep, they are,” but distortion terrifies me with those bones, too.

The red is the Coracoid 5761-1 (Fig. 81A, reversed from Osborn and Mook 1921) and the blue is the Coracoid from Fig. 82, 5760-2 (also from Osborn and Mook 1921). Notice how, by selecting the hole as "truth," I can’t get these coracoids to remotely match, despite being from the same species. It didn’t matter what section I chose; these two coracoids simply have lots of variation, which should surprise no one, considering the amount of cartilage and musculature happening in this section of the body.

Osborn and Mook (1921) recognized variation, as they wrote, on page 269, “...the skeletal differences as due to individual variation" when discussing whether they should keep Cope’s Camarasaurus leptodirus. They opted to believe it was based on variable characters and "sunk" it. Regarding scapulae in particular, they wrote, “The shaft has considerable breadth, though this varies somewhat among the different specimens,” acknowledging tremendous variation within scapulae.

The red is Osborn and Mook’s Fig. 80 (1921), showing the curvature of 5761-6, considered the mate of their Fig. 79, the green outline of 5761-4. Notice how wildly different the thickness is, and how the curvature isn’t in agreement at all (compare the light green to the brighter red at the top left of the image, and the bottom left). These scapulae, by the way, are nearly pristine, if only all scapulae were so well-preserved!

This curvature difference is one of the reasons why scapulae make terrible holotypes; they deform! These Camarasaurus examples don’t have much deformation present, unlike most scapulae, which are preserved flat or nearly so. This fact leads me to ask: how much difference in curvature did the scapulae possess? Ostensibly, the ribs will determine the curvature, so I’d hope in life they’d be comparably angled. But ribs are almost never studied or properly figured, so those won't help us.

However, these are cross-sections drawn by Osborn and Mook from ostensibly the same morphological section (assuming each of these outlines occurs in the position on their figures where they are illustrated), and they do show another aspect of variation. Those in blue are believed to belong to the same individual, a left and a right scapula being preserved, and both are in nice condition to boot. Notice how different each of these shapes is from one another. The shape of the scapular shaft certainly looks variable to me.

Paul 2025 severely underplays the wildly variable nature of sauropod scapulae, and he isn’t alone. Scapulae can vary so much within the same specimen that it makes me wonder why we even bother scoring them. They are quite helpful at the family level, so there is that. Also, Paul's 2025 figures are drawings of drawings… Photos suggest some of the original outlines were “idealized” illustrations rather than exact copies. Plus, angles of illustrations can lead to different interpretations; it is amazing how much 5 degrees can alter perception. The bones showcase much more variation than the figures suggest. Lastly, personal observation of the AMNH material revealed much more deformation and taphonomic intrigue than Osborn and Mook (and Paul) indicated. That isn’t simply my axial bias kicking in; I’ve been in the collections with appendicular specialists, and they pointed out many of what I’ll dub “areas of concern.” Scapulae are the Saran Wrap of sauropod bones; they cling to the ribs, but their shapes are wonky and don’t retain their original (already variable) shape once buried.

Factual Errors

Paul 2025’s Figure 3 shows the Hall Lake Formation, which, if I am reading the chart correctly, is in Arizona. As far as I am aware, there is no Hall Lake Formation in Arizona. Besides referencing my paper in the Journal of the Arizona-Nevada Academy of Sciences, the word Arizona, abbreviated or spelled out, doesn’t appear anywhere in the document. Should AZ have been included in the table? Or was it to help connect Utah to New Mexico?

In his Results and Discussion section, he opens up with yet another stunning error of fact; he writes of the New Mexico holotype (USNM 10486) and paratype (USNM 10487) that they “were found about 300 m from one another (Gilmore, 1922).” Gilmore did not write that; what he actually wrote was that they were found “...some 200 feet distant from one another.” 200 feet is quite a bit apart from 300 meters (~984 feet), a nearly 5x error. Now, do I think a bone 200’ from another belongs to the same individual? It is unlikely, but it could happen. Do I think there is a disturbing pattern of factual errors that the casual reader won’t know about whilst reading the 2025 paper? Yes. And that is a problem. One that is as much an indictment of the journal as it is of the author. Why am I, on a first pass through this document, finding all of these mistakes? Yes, I am a sauropod researcher who, a few weeks ago, deep-dived into Alamosaurus for a UT Austin paleo seminar and was especially sensitive to them at the time I read the paper. However, I shouldn’t be seeing these mistakes in a peer-reviewed published document, should I? I suspect (maybe hope is the word I should be using here?) the reviewers saw (these and likely more) and noted them. Why go to the trouble of reviewing something for a journal if this is the result? I think more and more that places like SV-POW! and, hopefully, this page are where we should publish papers. After all, who better to “road test“ a manuscript than visitors to websites focused on sauropods? Oh yeah, by making video descriptions of bones the norm, we can embrace technology to further democratize paleo and explain how we came to our conclusions. Digression over.

Ischia

Regarding the similarities of the ischia across the various localities, I’ll let you decide. Suffice to say, there is either significant variation present betwixt USNM 10487 (paratype of Alamosaurus) and USNM 15560 (Utetitan holotype) or we are dealing with distortion, or perhaps they are from different taxa. I’d rather not have one bone make the decision.

Despite my best efforts, I couldn’t match Paul’s outline with Gilmore’s 1946 fused USNM 15560 ischia. I tried flipping his image, thinking perhaps he copied the other side, to no avail. Maybe he, or someone else, can show me what I did wrong? Did he copy D'Emic et al. 2011, instead? 

USNM 15560 on left (Gilmore 1946) vs USNM 10487 paratype from Gilmore 1922

TMM 45891 “Rough Run” Titanosaur (from Fronimos 2010) vs USNM 10487 from Gilmore 1922

TMM 45889 Left Ischium from Fronimos 2010 vs USNM 10487 from Gilmore 1922

Lehman and Coulson 2002 TMM 43621-1 vs USNM 10487 from Gilmore 1922

Yep, only the juvenile from Coulson and Lehman looks like the full adult paratype of Alamosaurus sanjuanensis. It figures it’d be a juvenile, one where we have to fret about ontogeny.

Caudal Vertebrae

Paul 2025 states he does not think the caudal differences could be ontogenetic in nature, “...but that is speculative.” Everything else in this paper feels speculative, but somehow only the caudals get a pass as having differences that are “real”? And none of these differences are used to help diagnose it?

The Root Cause? No Specimen With Representative Parts

Even if we designated the North Horn Formation specimen USNM 15560 as a neotype, it wouldn’t resolve its lack of sacral, dorsal, and cervical elements. Thus, we still can’t make meaningful comparisons with much of the known North American Late Cretaceous sauropod material. Since caudal vertebrae are so conservative on these titanosaurs that I’d personally never use them to create a new genus (and, don’t forget, I love tails and try to use them for that purpose at every turn!), presacral element comparison is a must.

The most accurate statement in the Paul 2025 paper, to me, was this: “Even if future information indicates that some of the Texas titanosaurids constitute a distinct taxon, none of the known specimens is of holotype quality.”

 Paul 2025, despite copious typos and errors of fact, is super handy for listing specimens. I’ve amended it with the missing specimens from Ford’s Paleofile.com and Bivens’ database, way below.

A Second Issue: Dates!

The ages of the various formations are somewhat contested. It would be fantastic if we had dates we could be confident in, but alas, we don’t. So, on top of the morphology, we have chronal issues. Ugh!

Update: a new paper dropped (Flynn et al. 2025) while I was writing this, providing the age of the Alamosaurus holotype material as being equivalent to the North Horn Formation (66.4-66 Ma)! If this holds up, it removes one of Paul 2025's levers for naming Utetitan. I appreciate his application of stratigraphy, but like the off-the-Colorado-Plateau Morrison Formation stratigraphy, tread lightly.

Conclusion and More Philosophizing

I hope the (briefish) history of Alamosaurus was entertaining, that the variation images invoke gnawing worry, and the risks of science conducted via pixels and ink are hammered home.

Do I fundamentally disagree that the North Horn specimen is a new taxon? No, how could I, since I contemplated naming it. Should we honor Gilmore’s wishes and treat it as the neotype rather than a new genus? Perhaps, but I don’t think that is likely, and there are solid reasons we shouldn’t, as we have learned much about "deep time" since the 1940s, both in the actual length of time and in better dating. As a science studying evolution, our understanding evolves as we uncover more specimens. Maybe naming USNM 15560 a new species of Alamosaurus was the right move back in my 2017 paper? It certainly wouldn’t have brought as much attention to it as naming it a new genus has. Paul even works in a plug for his book in the 2025 paper: "While preparing a new edition of a field guide..." Well played! 

I soliloquied to some friends: “He’s [Paul 2025] likely right, but for the wrong characters. However, does Paul 2025 encourage shotgun-naming every specimen, in hopes that some will stick? Why not name every kinda-complete dinosaur skeleton from every dinosaur quarry a new genus? The use of vague, squishy, qualitative characters can make almost any quarry laden with new taxa. Mix in the fact that we rarely have good temporal control and that species can change in but a few hundred thousand years, and, well, we find ourselves exposed as a discipline." Yeah, that idea didn't sit well with anyone.

Philosophically, what is a dinosaur name used for, anyway? I use specimen numbers when I chat with my compatriots, after all, the names change, but the specimen numbers stay the same (most of the time ;-) ). The next generation of paleo uses specimen nicknames far more than specimen numbers (I wonder if this will be the philosophical 'soft entry' into studying commercially-owned specimens?). This use of nicknames seemed cool until I was asked to speak about ‘Goliath.’ They meant the Tyrannosaurus I measured the femur of, but I naturally assumed it was ‘Goliath’, the Supersaurus now at the Grandview Museum in China. Hilarity (of a sort) ensued.

Imagine a modern-day “naming” race, where a few individuals scramble to name taxa as fast as they can, a Marsh vs. Cope war all over again. Did the pendulum swing too far the other way, combining all of their "Bone Wars" taxa into too few? I’m a lumper by sentiment; the fact that I’m even typing this is giving me anxiety. But I’m in a reflective state, wondering why I had such visceral reactions to the naming of a specimen that I nearly suggested a name for in 2017? What would be the harm in adding dozens of new sauropod genera? I go back to D'Angelo's taxonomic conversations and those of many of the next generation of paleontologists, who argue that we have overlumped our dinosaurs based on today's diversity. The natural outcome of this pedagogy is more genera and species, more papers, press releases, dinosaur figures, artwork, and careers being made or lost based on the success of one's animals. 

Opinions Vary on the Legitimacy of Alamosaurus

Lucas and Sullivan 2000 argue Alamosaurus is a nomen vanum, "Indeed, based solely on its holotype scapula, we cannot diagnose Alamosaurus from these two South American titanosaurids [Laplatasaurus araukanicus (Huene, 1929, pl. 23) and Saltasaurus loricatus (Bonaparte and Powell, 1980, fig. 4; also see McIntosh, 1990, fig. 16.8) (Fig. 3).]. Therefore, strict application of the type concept indicates that Alamosaurus sanjuanensis is a nomen vanum (nomen dubium)." They continue, "Certainly one, and possibly more, titanosaurid taxa are represented by the thus-far collected specimens of Alamosaurus, but the name A. sanjuanensis is based on an inadequate holotype. If and when a potentially diagnostic specimen is discovered, we believe it will be useful (by preserving longstanding usage) to set aside the holotype and designate a neotype. However, for now it is useful to continue using the name Alamosaurus as a form genus for all North American Late Cretaceous  sauropods (Sullivan and Lucas, 2000a)." I love that phrase, "form genus."

Lehman and Coulson 2002 argue it is legitimate: "TMM 43621-1 includes skeletal elements that have not been previously described for A. sanjuanensis, and thus provides significant new osteological information. The new material further supports the validity of A. sanjuanensis, and allows for improved comparison with other titanosaurid sauropods, as well as an opportunity to provide a full skeletal reconstruction for this species."

I'm not sure what was in the water in 2011, but three papers came out addressing Alamosaurus.

D'Emic et al. 2011 argue it is legitimate based on scapula similarities, "Our observations of the holotypic scapula of Alamosaurus sanjuanensis indicate that it is diagnostic, making Alamosaurus a valid taxon (see below). We also confirm that the holotypic left scapula, the paratypic right ischium (both from the Naashoibito Member of the Kirtland Formation; Lucas and Sullivan, 2000), and the referred partial skeleton from the North Horn Formation pertain to the same species, as proposed by Gilmore (1946). The holotypic and North Horn Formation scapulae bear some minor differences, noted by Gilmore (1946:36), which can be attributed to deformation of the latter specimen." They later write, "We also note here that although Gilmore’s (1922) original designation of an isolated ischium as the paratype of Alamosaurus sanjuanensis was based on provenance alone, it pertains to the same species as the conjoined ischia of the North Horn Formation specimen, whose scapula shares the autapomorphies listed above, based on their nearly identical size and proportions (Fig. 5)."

Fowler and Sullivan 2011 suggest an interesting twist, that it is named on juvenile elements, which is nothing but bad in dino paleo: "The results of our analysis suggest that Alamosaurus sanjuanensis is diagnosed based on immature remains (Gilmore 1946; Upchurch et al. 2004; Wilson 2002). It has been recognized for some time that immature dinosaurs typically exhibit features more similar to their ancestors than to adults of their own species (Rozhdestvensky 1965)."

Jasinski et al. 2011 believe all is Alamosaurus: "All the titanosaurid fossils occur as isolated elements within this restricted unit. Therefore, it is logical to assume that all the titanosaurid remains represent a single taxon (i.e., Alamosaurus sanjuanensis) based on parsimony. We maintain that the argument that there may be more than one titanosaurid taxon within this restricted unit is not defensible. The onus is on those who claim the multiplicity of titanosaurid taxa to unequivocally demonstrate, with sound and acceptable data, that there is more than a single taxon (Alamosaurus sanjuanensis) present in the Naashoibito Member, which is the type stratum for this sauropod dinosaur. Thus, we regard all titanosaurid material from San Juan Basin as Alamosaurus sanjuanensis." I'd point them to the Morrison Formation for the existence of multiple taxa, but also concur that I haven't (yet) observed any elements (apart from the 'Rough Run' specimen) that give me reason to consider the presence of a second genus.

Fronimos and Lehman 2014 write about 'Rough Run' variation: "Differences between the ilia and pubes of the new material and those of previously collected North American titanosaur specimens may be intraspecific or interspecific in nature." They leave 'Rough Run' as incertae sedis while adding at the end of their paper, "Although the identity of the specimens as Alamosaurus sanjuanensis is considered likely, in the absence of autapomorphies shared with the hypodigm no referral is made. The specimens most closely resemble a juvenile  titanosaur from the lower Black Peaks Formation and are recognizably different from appendicular elements recovered from the underlying Javelina Formation, which may reflect intraspecific variation or the presence of a previously unrecognized taxon." Conservative yet telling! I thought this would be a foreshadowing statement. It still might be :-).

Tykoski and Fiorillo 2017 write of Alamosaurus legitimacy: "The case was recently made again that Alamosaurus sanjuanensis is a diagnosable taxon (D’Emic et al. 2011) based on features preserved in the holotype scapula (USNM 10486) and referred specimen (USNM 15560), including: the presence of two low processes on the ventral margin of the scapular blade distal to the glenoid fossa versus a single or no process in almost all other titanosauriforms; an asymmetrically concave scapular blade with thicker ventral margin and thinner dorsal margin; dorsal edge of scapular blade nearly straight and ventral edge expanded distally; a circumferential depression limited to the ventral half of the anterior."

They believe in the similarity of many ischia, further uniting them under one species: "The ischium in TMM 41541-1 closely resembles the paratype ischium for Alamosaurus sanjuanensis, USNM 10487 (Fig. 11A, B). The short and broad ischial shaft, deeply concave posterior margin, and long pubic and ischial contacts are all similarities shared between the two specimens. The iliac peduncle is shorter in TMM 41541-1, but it appears this may be the result of post-depositional deformation. The same morphology is also shared with the articulated ischia of USNM 15560 (Gilmore 1922, 1946), as well as with an ischium pertaining to a small, immature individual (TMM 43621-1) from another site in Big Bend National Park (Lehman and Coulson 2002). No autapomorphies of Alamosaurus have been identified in the ischium, but the morphological similarity of the ischia of these specimens supports their assignment to the same taxon."

Tykoski and Fiorillo 2017 argue the scapulae all belong to one taxon: "Among the numerous elements recovered on the nearby site was a left scapula, BIBE 45958, which is missing only the distal half of the scapular blade and part of the contact for the coracoid (Fig. 12). The scapula matches up well with the general morphology of the holotype of Alamosaurus (USNM 10486) and the scapula of the North Horn Formation skeleton, USNM 15560 (Gilmore 1922, 1946). Importantly, it also preserves the two scapular autapomorphies of Alamosaurus sanjuanensis proposed by D’Emic et al. (2011). These include the presence of two low tuberosities on the ventral edge of the base of scapular blade, and an asymmetrical concavity on the medial surface of the base of the scapular blade, resulting in a blade that is thicker ventrally than dorsally. Based on the presence of both characters in BIBE 45958, the specimen is referred to A. sanjuanensis. We also consider the other titanosaur bones from the site to also be remains of Alamosaurus."

Paul 2025 creates Utetitan as a distinct taxon for the North Horn Formation sauropod material based on the characters I covered earlier. Dinosaur paleontology, for better or worse, comes down to the opinion of the author at the time. Paul (2025) provides a name, Utetitan, to a specimen that is in a collection (USNM 15560), and provides, albeit vague-to-me, characters that are (semi-)testable. The onus is on the rest of the paleo community to either accept his statement, ignore his taxon, or show why it is or isn't Alamosaurus. Since Gilmore gave us a hot mess in the form of a partial scapula, which happens to contend with the coracoid as the most variable bone in a sauropod skeleton, it might be unanswerable. Short of making a neotype, as a community, we have to decide if we want to use Utetitan zellaguymondeweyae, Alamosaurus zellaguymondeweyae, Alamosaurus sanjuanensis, cf. Alamosaurus sp. (can one even use that combo?), or some other solution. Someone (I hope not me, but it looks like I have a good start...) can attempt to group all Late Cretaceous North American sauropod elements into morphotypes, then determine which characters stand out and are less affected by variation. 

I emailed the Smithsonian's team to see if they can provide high-res images of the tubercles. Alas, the government is shut down, so a "I'm furloughed" email was the auto-reply. Doh!  Update: They got back to me when it reopened and said it is in the queue! Hooray!!! (or is it queu-ray? ;-))

Other fun observations

The blue/black is from Lehman (1981), while the red was drawn "after Lehman 1981" by Lucas and Hunt. Notice the distal ends (left side of the image) do not match identically. An example of the risk of using illustrations.

The black outline in the above image is of TMM 43621-1, a left ischium, drawn by Coulson in 1998, then Fig. 9 in Lehman and Coulson 2002 (red). Notice they don't match, despite being of the same specimen. The importance of recognizing that illustrations alone do not tell the entire story. Aside, they twice in their figure call the bone an ilium.

Postscript

How does one pronounce Utetitan?

"YOU-tay-tight-en"

"YOU-tight-en"

"YOOT-tight-en"? (The double 'hard T' sound is messing with me, like rolling t's :-))

???

Acknowledgments

Thank you, Cristy Lopez, for your patience! Thank you, curators and collections managers, for allowing me to work on the material. Ron Tykoski and Karen Morton (Perot), Rod Sheetz and Brooks Britt (BYU), Chris Sagebiel (UT Austin), and Michael Brett-Surman (Smithsonian).  To the many individuals I colorfully conversed with, many thanks!!! Gunnar Bivens (conversation, specimen #s, facts, errata), Ray Wilhite (comparo, conversation), Tom van der Linden (philosophy), Emmanuel Tschopp (philosophy, facts), John D'Angelo (philosophy, facts), Taylor Greenfield (ICZN clarification), Mathew Wedel (philosophy), Mike Taylor (philosophy), and Erika Tandy (facts and the awesome TMM 43599-2 pic). If I missed you, mea culpa, please let me know, and I'll happily add you.

Appendix 1: Specimen number comparisons

Only in Tracy Ford Paleofile.com list (Not in Paul 2025)

-MDCH V-043 (MUDECH-2; Chihuahua, México; tooth fragment; Rivera-Sylva et al., 2021). Not listed in Paul’s specimen roster/table. 

-“BURGE, 1997 (pers. comm.)” Utah fragmentary skeleton (no catalog number provided). 

-Carter, Montgomery and Biasatti, 1999 Big Bend material (no catalog numbers provided). 

-Fiorillo, 1998 “3 individuals” (no catalog numbers provided). Not itemized by Paul. 

-Uncollected items (e.g., Lehman 1985 “Uncollected: Femur” at multiple localities). Paul tallies cataloged material only. 

-USNM 15660 - Tracy typo 

-TMM 40597-7 (ischia; the Woodward list). Paul lists TMM 40597-5, not “-7.” 

Only in Paul's 2025 paper (Not in Ford Paleofile.com)

AMNH 21531 (Upper Javelina, TX). 

BYU 9087 (proximal humerus and femur, North Horn, UT). TTU 542 (Javelina, TX). However, Ford lists TTU 546, but not 542. 

TMM 41396-1 (Javelina, TX). 

TMM 40597-5 (Javelina, TX).  (Ford has 40597-7 instead.) 

TMM 45888-1 (osteoderm; Black Peaks, TX). 

TMM 45890-1 and 45890-2 (right tibia; ungual; Black Peaks, TX). 

Only in Gunnar Bivens' database

LACM 157550 (cervical series; Ojo Alamo)
LACM 156591 (left pubis; Bisti/De-Na-Zin)
KUVP 14763 (pelvic material; locality number KU-NM-027) — not detected.
AMNH FARB 3080 (Aguja vertebra)

Appendix 2: More Paul 2025 errata I noticed whilst reading 

Screenshots of typos and grammar

Erroneously Typed Holotype Numbers

USNM 10846 (Paul, 2025 fig. 1 caption) - Belosphys spinosus vertebra on USNM collections page Paleobiology Collections Search

USNM 10847 (Paul, 2025 fig. 1 caption) - Priscodelphinus harlani thoracic vertebra on USNM collections page Paleobiology Collections Search

For what it is worth, I did the same thing in my 2017 paper…

Are the numbers duplicated, or are they typos of other numbers?

Specimen Number Concerns: Where’s the Scapula?

The line drawing is from Ballard and Lehman 2009 Figure 2 (TMM 45855-4, flipped for better blade comparison), and shows where they removed bone for histological analyses. 

The photograph (thank you, Erika Tandy!) is of TMM 43599-2, a scapula from the LK of Texas. Seeing as how the scapula is the holotype, all scapulae matter. 

Time to write: over 20 hours (!)...