Yi qi and Occam’s razor

Yesterday and earlier we looked at the new dino-bird, Yi qi. This is the third and final note.

Figure 1. Identification errors (in red) on the original Yi qi diagram from Xu et al. 2015.

Figure 1. Identification errors (in red) on the original Yi qi diagram from Xu et al. 2015. LSE = left styliform element. RSE = right styliform element.

To reiterate: 
In Xu et al. (2015) the authors noted a new long bony element in the fore limb of Yi qi (Fig. 1) never before seen on any bird, dinosaur, or archosaur. They labeled the new bone a ‘styliform element’ in accord with similar elements in flying squirrels. Chemical analysis determined that the new bone was indeed bone. This long bone, as long as (or in the opinion of the authors, longer than) any other forelimb element appeared to be attached at one end to the wrist, and at the other end, to nothing else. In certain flying squirrels, the styliform extends laterally to expand the width of the gliding membrane. Xu et al. also compared the unfamiliar bone to similar bones in gliding marsupials, bats and pterosaurs (the pteroid), all of which arise from different parts of the anatomy, frame extradermal membranes and are employed during aerial exercises.

Xu et al. wrote:
“The most striking feature of Yi is the presence of an anomalous, slightly curved, distally tapered, rod-like structure whose length considerably exceeds that of the ulna, associated with each wrist and apparently extending from the ulnar side of the carpus. However, the rod-like bone of the forelimb of Yi is morphologically unlike any normal theropod skeletal element. Indeed, no equivalent of the rod-like bone is known in any other dinosaur even outside Theropoda,” So Yi qi had potentially analogous structures among other tetrapods, but not among birds or other dinosaurs.

The big question is:
was the new bone indeed longer than the ulna? If you trace the elements as they are (Figs. 1, 2), the new bone is not longer than the ulna.


Figure 2. Yi qi tracing of the in situ specimen using DGS method and bones rearranged, also using the DGS method, to form a standing and flying Yi qi specimen. Note the lack of a styliform element, here identified as a left radius and right ulna.

It’s easy to make errors here
The right radius is incomplete and splintered (Figs. 1-3) so that it looks like a radius and ulna. The distal right radius is lost beyond the edge of the matrix. The right so-called ‘styliform element’ extends beyond the elbow and was presumed to also extend beyond the end of the matrix. However, taken ‘as is’ the two so-called ‘styliform elements’ are actually similar in length to each other and similar to the virtually complete left ulna.

Taphonomic churning
Evidently the elements were churned prior to burial with medial digits of the left hand now lateral and the right foot high kicking like a Rockette. Thus Padian’s comment, “Yi qi’s body is not preserved below the ribcage, so reconstructions of the pelvis, hindlimbs and tail must be conjectured from what is known of other scansoriopterygids,” is incorrect or an oversight (see Figs. 1, 2).

Figure 3. Right ulna (former styliform process) of Yi traced using DGS techniques. See Figure 4 for comparison to Epidendrosaurus.

Figure 3. Right ulna (former styliform process, amber color here) of Yi traced using DGS techniques. See Figure 5 for comparison to Epidendrosaurus. The newly identified ulna was flipped prior to burial so that the thicker proximal end is now closer to the lost wrist. The radius (pink and purple) is cracked lengthwise with toothy ridges marking the break. This gave the impression of a radius and ulna, but no intact radius and ulna have tooth-like cracks and ridges. This is a shattered hollow bone. 

On the left forelimb,
the left so-called ‘radius’ (Figs. 1, 4) is barely present. The left so-called ‘styliform element’ gives the erroneous impression of continuing beneath the left humerus. It does not.

Figure 3. Closeup of the former 'styliform element' here identified as a radius in Yi qi.

Figure 4. Closeup of the former ‘styliform element’ here identified as a left radius in Yi qi.

The authors never considered the possibility
that the ulna and radius could have been splintered during crushing prior to burial. Both forearm elements are as hollow as long bones from other dino-birds. And they have been subject to torsion and churning prior to burial. In this scenario the purported left ‘radius’ (Fig. 4) is simply a splinter of the ulna, separated during axial torsion. Similarly, the purported right ‘ulna’ is a splinter of the right radius (Fig. 3).

In this hypothesis
the curved right ‘styliform element’ is in reality an ulna and the straighter left ‘styliform element’ is a radius. As preserved the new radius and ulna are about the same length and similar in diameter to their counterparts. In birds and sister taxa, like Epidendrosaurus (Fig. 5), the radius is typically straighter than the ulna. And the ulna tapers distally.

In the Xu et al. 2015 Supplemental Data,
the ‘styliform element’ lengths each have an asterisk associated with them: 133.5* and 91.3* mm. The ulna likewise has an asterisk measurement, 88.5* mm. The length of the radius was not estimated or shown. The authors note: “* indicates estimated value; the preserved length of the right styliform element is 91.3 mm, but taphonomic information and morphological comparisons between the right and left styliform elements lead us to estimate that the total length of the styliform element is 133.5 mm.” In this case, they overestimated. Perhaps a more precise tracing would have been helpful.

Occam’s razor
As everyone knows, “The principle states that among competing hypotheses that predict equally well, the one with the fewest assumptions should be selected.” Xu et al. struggled with the identity of the odd long bones in Yi qi. Unfortunately they did not select the identity with the fewest assumptions (and autapomorphies). Instead they made headlines around the globe and more than a few eyebrows rise with the invention of a new styliform element, which is really just a misidentified common fore arm element.

In this case
the left ‘styliform element’ articulates with the wrist because the left radius likewise articulates with the wrist. That the distal end no longer articulates with the elbow can be ascribed to taphonomic torsion prior to burial.

The right radius still articulates with the humerus, but the right ulna has flipped lengthwise, such that the narrow distal end (Fig. 4) is now proximal, far behind the elbow. While more difficult to visualize how this may have happened, this bone flip also must be the product of taphonomic churning prior to burial. Perhaps it goes along with the high kick of the foot. Comparisons to the articulated specimen of Epidendrosaurus (Fig. 4) are instructive here. The ulna shapes are virtually identical.

Figure 4. Right ulna of Yi (former 'styliform element') compared to right ulna of Epidendrosaurus.

Figure 5. Right ulna of Yi (former ‘styliform element’) compared to right ulna of Epidendrosaurus.

Padian K. 2015. Paleontology: Dinosaur up in the air. Nature (2015) doi:10.1038/nature14392
Xu X, Zheng X-T, Sullivan C, Wang X-L, Xing l, Wang Y, Zhang X-M, O’Connor JK, Zhang F-C and Pan Y-H 2015.
 A bizarre Jurassic maniraptoran theropod with preserved evidence of membranous wings.Nature (advance online publication)


One thought on “Yi qi and Occam’s razor

  1. Nicely done, David. I’m no expert, but my bet is also on the “normal scansoriopterygid” structure. Occam’s razor also applies to function, and it would be odd for this little group to go so close to the bird feather-wing model and then dabble in bat-like flight mechanics. If Yi really proves to have had a membranous wing supported by a relatively huge stylus, I wouldn’t be disappointed, just surprised.

    The best thing is that your interpretation is probably going to be testable — we’ll likely see more scansoriopterygids, and hopefully more Yi coming to light!

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s