Hello! This page has unofficial supplementary information for my new paper, “Evidence for bird-like air sacs in saurischian dinosaurs”, published in February 2009 in the Journal of Experimental Zoology. High resolution versions of the figures are below, and a link to the PDF of the paper. I may write some more explanatory information about the paper if time permits; that information might go on this page or in regular posts on the blog, one of which is already up.
Full Citation and Link to PDF
Wedel, M.J. 2009. Evidence for bird-like air sacs in saurischian dinosaurs. Journal of Experimental Zoology 311A:611-628.
High Resolution Figures
Here are all of the figures from the paper, plus a couple of freebies (click for full-size versions). Please feel free to use them for non-profit educational purposes, just credit them to me. For use in periodicals, books, documentaries, or any for-profit endeavors, please contact me about reproduction rights.
Fig. 1. Evidence for air sacs in fossil archosaurs. Letters next to each taxon indicate that they have patterns of PSP that are diagnostic for certain air sacs: C, cervical air sacs; A, abdominal air sacs; Cl, clavicular air sacs. See text for diagnostic criteria. In both sauropodomorphs and theropods, cervical air sacs pneumatized the skeleton before abdominal air sacs. Morphological evidence for PSP in Archaeopteryx is equivocal (O’Connor, 2006). A naı¨ve reading of the fossil record would suggest that different air sacs evolved independently several times—for example, independent origins of abdominal air sacs in Mamenchisaurus and Neosauropoda. However, it is more parsimonious to infer that cervical and abdominal air sacs, at least, were present in all members of Eusaurischia, and simply failed to pneumatize the postcranial skeleton in some taxa (e.g., most basal sauropomorphs). It is possible that an air sac system is primitive for Ornithodira, but the total absence of PSP in Ornithischia, a diverse and long-lived clade, is problematic (see Wedel, 2007). Phylogeny based on Gauthier (’86), Wilson (2002), Zhou and Zhang (2002), Upchurch et al. (2004, 2007), and Yates (2007).
Fig. 2. MAL-200, an anterior caudal vertebra of Malawisaurus dixeyi. (A) The vertebra in left lateral view showing the position of CT slices. (B–D) CT cross sections. Matrix was erased from the internal chambers using Photoshop 5.5. Pneumatic foramina on the neural arch and spine are connected to a network of internal chambers, but the centrum is apneumatic.
Fig. 3. Pneumatization of the vertebral column in the chicken, Gallus gallus. Pneumatic vertebrae are shown in black. Data are from Hogg (’84b); vertebrae are shown at earliest date of complete pneumatization. Some rare variations are not shown; for example, the second and third dorsal vertebrae were pneumatized in one individual (from a total of 44) examined by Hogg (’84b). The spread of PSP along the vertebral column in the chicken parallels the evolution of PSP in nonavian theropods and sauropods; compare to Wedel (2007, Text-Fig. 2).
Supplementary Figure 1. The distribution of fossae and pneumatic chambers (black boxes) along the vertebral column in sauropodomorphs. Only the lineage leading to diplodocines is shown here. A similar caudal extension of pneumatic features occurred independently in macronarian sauropods (Table 1) and several times in theropods. It also parallels the development of vertebral pneumaticity in birds; compare to Fig. 3. After Wedel (2007:text-fig. 2).
I cut this figure from the final paper, but I’m including it here for easy comparison to Fig. 3, above.
Fig. 4. Pneumatization of the vertebral column in the chicken, Gallus gallus. Pneumatic vertebrae are stippled. The vertebral column is pneumatized by diverticula of the cervical air sacs, lungs, and abdominal air sacs. A pneumatic hiatus is one or more apneumatic vertebrae that are bordered anteriorly and posteriorly by pneumatic vertebrae. These hiatuses are produced if the diverticula from the different parts of the respiratory system do not meet. Supporting data come from King (’57), King and Kelly (’56), Hogg (’84a,b), and from personal examination of museum specimens. Inspired by King (’57, Fig. 1). Abbreviations: CaS, caudosacral; CaD, caudal dorsal; CvD, cervicodorsal.
Fig. 5. A pneumatic hiatus in a chicken. The notarium of UCMP 119225 is composed of four vertebrae. The three anterior vertebrae are pneumatic, but the fourth is not. (A) The specimen in left lateral view under normal lighting. (B) The specimen lit from behind to show the pneumatic (translucent) and apneumatic (opaque) regions. (C) A micro CT slice through a pneumatic vertebra. (D) A micro CT slice through the apneumatic vertebra. Note the density of the trabeculae in D compared to C. The anterior synsacral vertebrae of this individual are pneumatic. The apneumatic vertebra is bordered anteriorly and posteriorly by pneumatic vertebrae, and constitutes a caudal dorsal pneumatic hiatus.
Supplementary Figure 2. Criteria for inferring the presence of abdominal air sacs in non-avian dinosaurs. A sauropod is shown, but the same logic applies to theropods. Pneumatized vertebrae are shown in black. Small arrows show the spread of pneumatic diverticula, and large arrows represent ontogenetic trajectories. A. Pneumatization of the vertebrae by diverticula of cervical air sacs (ca), lungs (L), and abdominal air sacs (aa). Other air sacs (light gray) may be present, but are not known to pneumatize the vertebral column. B. Pneumatization of the vertebrae by diverticula of cervical air sacs alone. C. A hypothetical sauropod with pneumatic hiatuses. Vertebrae posterior to a CaD or CaS hiatus could only be pneumatized by diverticula of abdominal air sacs. Therefore, the presence of these hiatuses demonstrates that abdominal air sacs were present. D. Pneumatization of the posterior dorsal, sacral, and caudal vertebrae does not necessarily demonstrate the presence of abdominal air sacs, because continuous pneumatization of the vertebral column could be produced by anastomosing diverticula of the cervical and abdominal air sacs (as in A) or by cervical air sacs alone (as in B). The A-C and A-D transformations are known to happen in extant birds. The B-D transformation does not happen in extant birds (O’Connor, 2006), but cannot be ruled out in non-avian dinosaurs (because it depends on the behavior of epithelial diverticula that do not themselves fossilize). Modified from Wedel (2003a:fig. 4).
I cut this one, too, since it is just v2.0 of the figure as I used in my 2003 paper
. This figure is basically the whole hiatus story in one shot–there just weren’t any examples back in 2003. Now there is at least one, as the following figures show (actually there are many more, but those papers are still in the pipeline–I’ll keep you posted).
Fig. 6. Pneumatization of the presacral vertebrae in Haplocanthosaurus. (A) X-ray image of a posterior cervical vertebra of CM 879 in right lateral view. (B) A CT slice through the same vertebra. (C) X-ray image of an anterior dorsal vertebra of CM 572 in left lateral view. (D) X-ray image of the same vertebra in anterior view. All of the preserved presacral vertebrae of both specimens have large, sharp-lipped fossae that penetrate to a narrow median septum.
Fig. 7. A pneumatic hiatus in a sauropod dinosaur. The preserved portions of the sacrum (S1–S5) and anterior caudal vertebrae (Ca1–Ca3) of Haplocanthosaurus CM 879 are shown in right lateral (top) and left lateral (bottom) views. All of the preserved cervical and dorsal vertebrae have large, distinct fossae. Distinct fossae are also present on the right sides of the fourth sacral and first caudal vertebrae, and on the left side of the first caudal. The left side of the fourth sacral and both sides of the fifth sacral are waisted but lack distinct fossae (see text for discussion), and constitute a caudosacral pneumatic hiatus.
Fig. 8. The fourth and fifth sacral centra of Haplocanthosaurus CM 879. Above, the centrum of the fourth sacral vertebra in right posterolateral (A), right lateral (B), left lateral (C), and left posterolateral (D) views. Below, the centrum of the fifth sacral vertebra in right lateral (E) and left lateral (F) views. The centrum of S4 has dorsolateral fossae on both sides and a lateral fossa on the right side. The left side of S4 is waisted but lacks a lateral fossa. The centrum of S5 is waisted and lacks both lateral and dorsolateral fossae. Abbreviations: dlf, dorsolateral fossa; lf, lateral fossa.
Fig. 9. Anterior caudal vertebrae of Haplocanthosaurus CM 879 in dorsal view. (A) The first caudal vertebra. (B) The second caudal vertebra.
Fig. 10. The air sacs of Haplocanthosaurus. Preserved elements of CM 879 are shown in right lateral view. The cervical and anterior dorsal vertebrae were pneumatized by diverticula of cervical air sacs (green). Middle dorsal vertebrae were pneumatized by diverticula of the lung (red). Diverticula of the abdominal air sac (blue) pneumatized the posterior dorsal, sacral, and first caudal vertebrae. Other air sacs may have been present (gray), but their presence is not detectable from the preserved elements.