Now that a fairly good illustration of the pygostylian portion of Archaeraptor has been published, its relationships can be examined.
undescribed Pygostylian (Archaeoraptor liaoningensis sensu Sloan in partim)
Barremian, Early Cretaceous
Yixian Formation, Liaoning, China
Material- (IVPP collection) partial skull including premaxilla, nasal, frontal and parietal, partial mandible including dentaries, several cervical vertebrae, a few dorsal vertebrae, dorsal ribs, uncinate processes(?), anterior synsacrum(?), scapulae, coracoid, furcula, sternum, humeri, radii, ulnae, semilunate carpal,ulnare, proximal metacarpal I, phalanx I-1, manual ungual I, incomplete metacarpal II, phalanx II-1, manual ungual II, metacarpal III, partial ilium(?), partial ischium(?), femur, proximal tibia
coracoid basal transverse width more than 70% of proximodistal length
Several other characters could be added, such as the strange acromial structure (see below), but this should wait until elements are identified with certainty.
This specimen was originally part of the holotype of Archaeoraptor liaoningensis, as controversially announced by National Geographic (Sloan, 1999). However, several details let this pass as an official description and the species was nomen nudum. This holotype was later shown to be composed of several specimens artificially combined together, one of which is the counterpart of the eumaniraptoran Microraptor zhaoianus. Although there is further controversy regarding which specimen will receive the name (Olson designated the tail, counterpart of Microraptor zhaoianus, as the holotype), I doubt Archaeoraptor will be the final name for the pygostylian when Czerkas describes it. Thus, it is referred to as "Archaeoraptor" below.
The skull is low and pointed, with toothed jaws. The premaxilla is acutely pointed with a slightly posteriorly shifted naris (comparable to enantiornithines). The nasal is slender and straight. A large posterior element may be a frontal. A smaller element posterior to this may be a parietal. The dentaries are thin and straight, some posterior mandibular elements are also preserved.
Cervical vertebrae are elongate, but poorly illustrated. An isolated dorsal vertebra has large pleurocoels, like many basal pygostylians. Several dorsal ribs are preserved, with a possible uncinate process. An irregular, elongate, tapering element in the pelvic area may be a synsacrum.
An elongate element overlapping the coracoid may be a scapula. It is curved, with a pointed distal end and an ?anterodorsal process that may be the acromion. The only other possible identification for this element is a second metacarpal. The amount of curvature and ?acromion process argue against this however. Another incomplete, but similar, process with no obvious acromion is near certainly a scapula. The coracoid is strut-like with concave lateral and medial margins. The ventral margin is more expanded than most pygostylians. The furcula is a narrow V-shape (~48 degree interclavicular angle) with gently curved rami and no elongate hypocleidium. The sternum is very narrow, slightly expanded posteriorly with a ventral keel. Two pairs of posterior processes are present, an anterolateral one with a distal expansion and a small pointed posteromedial one. Both anterior and posterior margins are pointed.
The humeri are both well preserved. They are expanded proximally and distally in the same plane and lack the large deltopectoral crest of confuciusornithids. The distal end lacks the projected internal condyle of at least some enantiornithines (Alexornis, Enantiornis, Kizylkumavis), the internal tubercle is pointed. The straight radius is much more slender (~65%) than the bowed ulna. Oddly, the semilunate carpal and ulnare are unfused to each other or the metacarpus. A thin radiale may also be present. The metacarpal doesn't appear fused. The first metacarpal is narrower than the second, the third is narrower than both. Intermetacarpal space is limited. The third metacarpal is subequal to the second in length. Digit I has a small curved ungual and projects slightly past metacarpal II, if the digits are correctly articulated (a break occurs through the metacarpus). Phalanx II-1 is flattened and expanded mediolaterally. A small second ungual may be present as well, subequal in size to the first.
Possible ilial and ischial remains are preserved (the latter with a possible proximodorsal process). The femur is bowed, but lacks recognizable details. A proximal tibial fragment is also present.
- tapered distal scapula?
- mobile scapulocoracoid joint
- strut-like coracoid
- interclavicular angle of furcula 70 degrees or less
- ulna subequal or longer than ulna
- reduced manual unguals
basal to Jibeinia+Ornithothoraces
- manual digit I passes metacarpal II
- interclavicular angle 50 degrees or less
- sternal carina extends to anterior edge
- coracoid not laterally convex
- hypocleidium not elongate
- metacarpal III does not extend past metacarpal II
Thus, the "Archaeoraptor" specimen is obviously more derived than confuciusornithids, but further relationships are difficult to discern. There are a few characters that have a more complex distribution than normally thought. The anteriorly extensive sternal carina is usually thought to be a euornithine character, some enantiornithines exhibit this condition (Largirostrornis, Neuquenornis), although most lack it (Cathayornis, Concornis, Eoalulavis). Still, only ornithothoracines are known to have it, so it is evidence "Archaeoraptor" was a member of that clade. Similarily, the carpometacarpus has a very complex distribution, with differences in which metacarpals and carpals are fused, and whether or not they are fused distally. However, at least some enantiornithines (Sinornis, Cathayornis? caudatus, Longchengornis?) have unfused metacarpals, indicating the presence of the plesiomorphic condition in "Archaeoraptor" doesn't bar it from the Ornithothoracines. The interclavicular angle is narrower than Protopteryx and Jibeinia, which is another piece of evidence "Archaeoraptor" might be an ornithothoracine. However, the elongate first manual digit suggests it is more basal than Jibeinia or ornithothoracines. Whatever it is, it does not appear to be enantiornithine. More detailed examination of this specimen once Czerkas describes it will undoubtedly reveal further details that can help determine its affinities.
Those who want a figure of the pygostylian section of Archaeoraptor liaoningensis sensu Sloan can contact me offlist.
Sloan, C. P., 1999. "Feathers for T. rex?" National Geographic 196(5): 98–107 [November 1999].
Rowe, Ketcham, Denison, Colbert, Xu and Currie, 2001. Forensic palaeontology: The Archaeoraptor forgery. Nature 410, 539 - 540 (29 March 2001).