Sauropod Vertebra Picture of the Week

Sauropod vertebrae in anterior view exhibiting a spectrum of variation in the dorsoventral positions of the neurocentral joint. Wedel and Atterholt (2023: fig. 1).

As described in the last post, Jessie Atterholt is presenting our poster on this project today, at the 14th Symposium on Mesozoic Terrestrial Ecosystems and Biota (MTE14) in Salt Lake City, and the related paper is in the MTE14 volume in The Anatomical Record. Here’s the citation and a direct link to the paper:

Wedel, M.J., and Atterholt, J. 2023. Expanded neurocentral joints in the vertebrae of sauropod dinosaurs. In Hunt-Foster, R.K., Kirkland, J.I., and Loewen, M.A. (eds), 14th Symposium on Mesozoic Terrestrial Ecosystems and Biota. The Anatomical Record 306(S1):256-257.

I’ve been interested in neurocentral fusion in sauropods and other critters for a long time, especially when the neurocentral joint is shifted dorsally or ventrally relative to the neural canal. I noted some instances of those shifted joints in blog posts (one, two, three), but I didn’t know what to do with that information. The impetus to turn those observations into a paper came from two sources. First, working with Jessie got me thinking about shifted neurocentral joints as one more Batman villain in the rogue’s gallery of neural-canal-related weirdness in birds, sauropods, and other archosaurs. Jessie and I kindled the ambition to catalog that entire zoo — results of that mega-project so far are on a new sidebar page. 

Fronimos and Wilson (2017: figure 2)

Second, I read Fronimos and Wilson (2017). This is an extremely cool paper and it’s a shame I haven’t blogged about it before. The authors went through the cervical and dorsal vertebrae of the holotype skeleton of Spinophorosaurus (GCP-CV-4229) and measured the complexity of the neurocentral joints. They found that joint complexity increased toward the base of the neck, maxed out in the anterior dorsals, and decreased in posterior dorsals. That’s consistent with the idea that complex neurocentral joints were an adaptation to increasing biomechanical stress on the vertebrae, which should likewise increase toward the base of the long, cantilevered neck and decrease toward the big anchor of the sacrum. The basic idea is that the complex joints increased the joint surface area and decreased the likelihood of traumatic dislocations — disrupting the joint between the arch and centrum would tend to cause life-ending spinal cord injuries.

Available surface area for the neurocentral joint in its normal position (below) and shifted dorsally, above the neural canal (above). The lower part of the neural arch is H-shaped in cross-section, with anterior and posterior fossae below the zygapophyses. The real-life example this is based on is in the last image in this post.

Reading that paper was a lightbulb moment for me. If the neurocentral joints of sauropods were adapted to resist biomechanical stresses, anything that increased the “contact patch” between neural arch and centrum would be desirable. From the standpoint of a neural arch and centrum trying to stick together, the neural canal is a flaw, a big dumb area of forced non-union. But you can’t get rid of the neural canal, which houses the spinal cord and the developmentally important spinal arteries (see Taylor and Wedel 2021 for more on the latter). The only way to eliminate the gap caused by the neural canal is to get around it by shifting the neurocentral joint dorsally or ventrally. John Gilmore famously said that the internet interprets censorship as damage and routes around it. We hypothesize that in an evolutionary sense, sauropod neurocentral joints interpreted the neural canal as damage and routed around it.

Of course you don’t have to be a sauropod to benefit from the enlarged contact patch between neural arch and centrum, as shown by the ’boutons’ of many mammals, including humans (unfused sheep vertebra shown above). But as Fronimos and Wilson (2017) pointed out, strengthening the neurocentral joints was probably especially important for sauropods, which grew rapidly for a long time and achieved large body size with many joints still unfused (see also Wedel and Taylor 2013: table 1, Hone et al. 2016: table 2). That would also explain why some sauropods went well beyond bouton territory, into having the neurocentral joint entirely dorsal or ventral to the canal.

Hey, it only took me five and a half years to get this idea out of my notebook and into a peer-reviewed paper!

There’s still the question of why the neurocentral joints shifted dorsally in some vertebrae and ventrally in others. The ventral shift in caudal vertebrae makes intuitive sense — the neural arch narrows dorsally, so shifting the joint upward would decrease the surface area, not increase it. Also, shifting the joint ventrally allowed the neural arch to be morticed between the transverse processes, which further increased the contact patch and made the neurocentral joint even stronger.

MB.R.3823, a dorsal centrum of Giraffatitan in posterodorsal view. The neurocentral joint surfaces of the centrum come together dorsal to the neural canal, leaving only a paper-thin gap.

What about dorsal vertebrae? In dorsal vertebrae of Haplocanthosaurus, Camarasaurus, and Giraffatitan, the neurocentral joint is shifted dorsally, to the point that in some Camarasaurus dorsals the joint lies completely above the neural canal. It’s not obvious why that would be more advantageous than shifting ventrally — except possibly that shifting ventrally might have interfered with pneumatization. In some unfused Cam dorsals, like the one shown below, the lateral pneumatic cavities are so big that they excavate right up under the dorsally-shifted neurocentral joint.

MWC 3630, an unfused dorsal centrum of Camarasaurus in right lateral (top) and posterior (bottom) views.

Still, pneumatic diverticula are thought to opportunistically occupy spaces that aren’t being loaded very much (Witmer 1997), so presumably they could make cavities above, below, or in any other direction from the neurocentral joint. We’re not really sure why the joint shifted dorsally in dorsal vertebrae of some sauropods. We know that the developmental program could accommodate shifts in both directions over fairly short distances in the same individual, because in the CM 879 skeleton of Haplocanthosaurus, the neurocentral joints are almost entirely above the neural canals in the dorsal vertebrae, and completely below the canals in the caudals. (The sacrals in that specimen are doing their own weird thing, about which more another time.)

DINO 4970, an unfused neural arch of Camarasaurus in the Carnegie Quarry (“the Wall”) at Dinosaur National Monument. The arch is in ventral view, with anterior toward the top. Note the butterfly-shaped neurocentral joint, with no gap for the neural canal.

Our paper is short and to the point because we don’t have a lot of data on this yet. Our sampling so far is basically limited to stuff we’ve stumbled over that made us go ‘huh!’ As with our work on paramedullary diverticula in birds, we hope that our work inspires more people to look into this weird stuff and document it — we can’t be sure about the rules until we know what all is out there.

References

• Fronimos, J.A. and Wilson, J.A., 2017. Neurocentral suture complexity and stress distribution in the vertebral column of a sauropod dinosaur. Ameghiniana, 54(1), pp.36-49.
• Hone, D.W.E., Farke, A.A., and Wedel, M.J. 2016. Ontogeny and the fossil record: what, if anything, is an adult dinosaur? Biology Letters 2016 12 20150947; DOI: 10.1098/rsbl.2015.0947.
• Taylor, Michael P., and Mathew J. Wedel. 2021. Why is vertebral pneumaticity in sauropod dinosaurs so variable? Qeios 1G6J3Q. doi:10.32388/1G6J3Q
• Wedel, Mathew J., and Michael P. Taylor. 2013. Neural spine bifurcation in sauropod dinosaurs of the Morrison Formation: ontogenetic and phylogenetic implications. Palarch’s Journal of Vertebrate Palaeontology 10(1):1-34. ISSN 1567-2158.
• Witmer, L.M. 1997. The evolution of the antorbital cavity of archosaurs: a study in soft-tissue reconstruction in the fossil record with an analysis of the function of pneumaticity. Journal of Vertebrate Paleontology 17(Supplement 1): 1-76.