What is Rhamphocephalus? An earlier bird.

Some confusion in the academic literature today
as a Middle Jurassic fossil known since the 19th century is grossly misidentified.

Figure 2. Rhamphocephalus in situ, traced by Seeley, traced by O'Sullivan and Martill and Rhamphorhynchus graphic from Wellnhofer 1975.

Figure 1. Rhamphocephalus in situ, traced by Seeley, traced by O’Sullivan and Martill and, for comparison sake, Rhamphorhynchus graphic from Wellnhofer 1975, all appearing in O’Sullivan and Martill 2018. Rhamphocephalus has been traditionally identified as a pterosaur. That paradigm was challenged by O’Sullivan and Martill 2018, but that challenge is challenged again here.

Today a paper by O’Sullivan and Martill 2018
redescribes several fossils from the Middle Jurassic (165–166 mya) of England, traditionally ascribed to the wastebasket pterosaur taxon, Rhamphocephalus prestwichi (type, Seeley, 1880;  OUM J.28266; Figs. 1–4). Most of the disassociated specimens (individual jaws, limbs) are clearly pterosaurian. One (the goose-sized skull roof) is clearly not pterosaurian.

Figure 2. Rhamphorhynchus compared to a large choristodere, Simoedosaurus, and to a large thalattosuchian, Pelagosaurus. There is absolutely no match here.

Figure 2. O’Sullivan and Martill compared Rhamphocephalus to a large choristodere, Simoedosaurus, and to a large thalattosuchian, Pelagosaurus. There is absolutely no match here, either in size or morphology. Colors and ‘to scale’ Rhamphocephalus images added for clarity.

The holotype of Rhamphocephalus prestwichi,
“an isolated skull table, is found to be a misidentified crocodylomorph skull,” according to O’Sullivan and Martill, who illustrated the 10x smaller specimen alongside a dorsal view of the 3m long thalattosuchian (marine) croc, Pelagosaurus, from the Lower Jurassic of England and, perhaps to cover all their bases, flipped anterior-to-posterior alongside the Paleocene choristodere, Simoedosaurus (Fig. 2). Note: the authors did not illustrate their comparative taxa to scale (as shown above), perhaps because the taxa are 10x larger and are morphologically dissimilar. So why make such comparisons? I don’t understand the logic of these paleontologists making such readily disprovable comparisons.

Figure 1. The skull roof named Rhamphocephalus here with bones and teeth colored.

Figure 3. The in situ specimen of Rhamphocephalus here with bones and teeth colored. At standard monitor 72 dpi resolution, this image is 2x life size. Perhaps this skull can be µCT scanned for buried data. Some palatal elements are peeking out from the antorbital fenesrae and nares. The dentary teeth make a few appearances, too. This is a sharp-tipped taxon.

Traced here
using DGS methods (Fig. 3) and phylogenetically tested in the large reptile tree (LRT, 1321 taxa) goose-sized Rhamphocephalus nests with the hummingbird-sized, Hongshanornis (Fig. 2), an Early Cretaceous toothed bird from China. Hongshanornis is one of the few toothed birds in which the orbits are further forword, creating a longer cranium to match that of Rhamphocephalus. A suite of other skull traits are likewise most closely matched to Hongshanornis. The Rhamphocephlaus specimen appears to be complete without obvious breaks either at the toothy tip of the skull or the occiput. More teeth and bones were identified here.

Figure 2. Rhamphorcephalus in situ compared to Hongshanornis in situ to scale and enlarged to match.

Figure 2. Rhamphorcephalus in situ compared to Hongshanornis in situ to scale and enlarged to match skull length. To scale image (above) is 1.25x actual size, much too small for sea crocs. similar in size to pre-birds. Hongshanornis is a tiny bird, similar in size to a hummingbird.

the authors report, “The earliest known record of Bathonian pterosaurs is an account of “fossil bird bones” from the Taynton Limestone Formation of Stonesfield by an anonymous author A.B., appearing in the March edition of the Gentleman’s Magazine of 1757.” For this specimen, and only this specimen, A.B. got it right. The other specimens are clearly pterosaurian.

the authors report, “This specimen is exposed on a limestone slab in dorsal view and was assigned to Pterosauria based on its perceived thin bone walls. Seeley (1880) noted that the arrangement of bones was more crocodilian than pterosaurian and considered this construction diagnostic of the new taxon. Significantly he (Seeley 1880: 30) stated: “I shall be quite prepared to find that all the ornithosaurians from Stonesfield belong to this or an allied genus which had Rhamphorhynchus for its nearest ally.” In the LRT crocodilians are closer to birds than pterosaurs are.

Figure 6. Rhamphocephalus chronologically precedes the Solnhofenbirds by several million years making it the oldest known bird.

Figure 6. Rhamphocephalus chronologically precedes the Solnhofenbirds by several million years making it the oldest known euornithine bird.

Is the Middle Jurassic too early for a toothed bird?
Perhaps not. Remembet that all of the Late Jurassic Solnhofen birds, traditionally named as one genus, Archaeopteryx, already represent a diverse radiation of taxa, suggesting an earlier genesis for that radiation. Rhamphocephalus indicates that the original bird radiation had its genesis at least 15 million years earlier. 

It is unfortunate
that O’Sullivan and Martill attempted to force fit the skull specimen into a crocodilian clade when no aspect of the thin-walled, goose-sized skull of Rhamphocephalus is crocodilian (Fig. 2)… or choristoderan (when flipped backwards!!). Adding Rhamphocephalus to the LRT gives it a single most parsimonious sister among all the toothed birds and a special Middle Jurassic place in the origin of birds story. All the details fit.

Working with a high-resolution image
of Rhamphocephalus (Fig. 3) copied from a PDF of the paper by O’Sullivan and Martill made this all possible.

Once again, to determine the affinities of a specimen it is more important to have a wide gamut of taxa to work with than to have firsthand access to the specimen itself. No one likes this method, but it clearly works time after time and to not use it invites discredit.

USE THE LRT. That’s what it is here for.

O’Sullivan M and Martill DM 2018. Pterosauria of the Great Oolite Group (Bathonian, Middle Jurassic) of Oxfordshire and Gloucestire. Acta Palaeontologica Polonica 63 (X): xxx–xxx, 2018 https://doi.org/10.4202/app.00490.2018
Seeley HG 1880. On Rhamphocephalus prestwichi Seeley, an Ornithosaurian from the Stonesfield Slate of Kineton. Quart. J. Geol. Soc. 36: 27-30.


7 thoughts on “What is Rhamphocephalus? An earlier bird.

  1. Hello Mr. Peters, I am very interested in your DGS system and I would like to learn how to use it. One question: how do you calibrate it? To know that you’re identifying the right bones? Do you check against CT or X-ray data from fossil specimens? What method do you use to ensure accuracy?

    • Simply put: I color everything. Every splinter. Sometimes I make mistakes. Those I discover by comparisons with sister taxa, scans, drawings, x-rays. Do this and you’ll learn. Since the bones don’t come with labels or boundaries that separate splinters and cracks from sutures, coloring the bones helps you figure this out. Experience helps. A growing number of paleontologists are coloring bones now, especially in µCT scans. Just start and you’ll learn what is right and wrong as time goes by. There are no methods one can use to ensure accuracy other than to strive for accuracy.

      • Very interesting. I am also interested in color-coding. But I wonder. How does interpreting bones from photographs compare to first-hand study? Smithwick et al. (2017) brought up an interesting case study using Sinosauropteryx filaments. As you may know, Alan Fedducia and his colleagues considered the filaments of Sinosauropteryx to be degraded collagen fibers, comparable to those of the ichthyosaur Stenopterygius. However, this paper showed that Feduccia’s observations, which were informed by photographs, were actually misidentifications of scratch marks, shadows, and coloration of the background rock layer. I was wondering how one could avoid these errors using evidence collected from photographs. Should I consult with people who have seen the fossils up close and personal?
        Here is a link to the paper btw: https://onlinelibrary.wiley.com/doi/full/10.1111/pala.12292

      • Indeed, that is a possible issue. Take a look at the many examples shown on the blogpost and website (ReptileEvolution.com) and let me know if you see any false positives. I have made thousands of corrections to scores over the past seven years as I continue to learn. re: Consulting people… I often send data to original authors, as in this case, I sent the new interpretations to Dr. David Martill, someone I know from several symposia.

      • Neil, the bottom line is: put some effort into your work. Don’t be like Wang, Zhou, Zhang and Xu 2002 who traced the simplest outline for their Jeholopterus skull from first hand observation. See: http://reptileevolution.com/jeholopterus.htm (the image that is titled, “Using the Computer Screen as a Microscope.)

        The DGS process has two other steps that must follow. 1) Use the data you glean to make a reconstruction. In that way you can check to see that bones actually fit together like the 3D puzzle pieces they are. 2) Add your data to a phylogenetic analysis to make sure that traits you find are shared by sister taxa. That way you’ll avoid the problems started by Dr. S. Christopher Bennett who reported a giant eyeball preserved edge on in a crushed anurognathid fossil, when no pterosaur shares this trait. That was actually a maxilla. Details here: http://reptileevolution.com/anurognathus-SMNS.htm

  2. This might be your least plausible interpretation yet. No bird skull has such weird frontals with tiny interorbital distance, and as if all those teeth were just missed by the authors. Like your fictional toothy Confuciusornis. And “all the details fit” when no one thinks ornithuromorphs were present in the Middle Jurassic. I don’t claim to know what Rhamphocephalus is, but it doesn’t resemble theropods/birds at all.

    • Try phylogenetic analysis. The sister taxon is preserved in lateral view, so the inter orbital distance is not comparable, nor is it a scored trait. If and when you come up with a better sister, please let me know so a correction can be made.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.