Protopteryx Zhang and Zhou 2000
P. fengningensis Zhang and Zhou 2000
Etymology- "primitive feather from Fengning County", after the locality the holotype was discovered in.
Barremian, Early Cretaceous
Yixian Formation, Liaoning, China
Holotype- (IVPP V 11665) skull (28.3 mm), lower jaw, seven or eight cervical vertebrae (18.5), twelve dorsal vertebrae (33.9 mm), dorsal ribs, sacrum, seven free caudal vertebrae (9.4 mm), pygostyle (11.3 mm), scapulae (21.7 mm), coracoids, furcula (14.7 mm), sternum, forelimbs, ilium (14.8 mm), pubis (22.3 mm), ischium (12.9 mm), hindlimbs (86.3 mm), feather impressions
Paratype- (IVPP V 11844) skull, lower jaw, seven or eight cervical vertebrae (19.1), twelve dorsal vertebrae (34 mm), dorsal ribs, sacrum, seven free caudal vertebrae, pygostyle, scapulae, coracoids (12.7 mm), furcula (14 mm), sternum (15.9 mm), forelimbs, ilium (15.3 mm), pubis, ischium, hindlimbs, feather impressions
Diagnosis- one tooth per premaxilla; procoracoid process.
The skull is complete and resembles other basal avians in basic shape, being triangular with large circular orbits. The fused premaxillae possess long nasal processes that extend to the lacrimal and one tooth each. The external naris appears enlarged and retracted somewhat. A maxillary fenestra is still present. The postorbital is Y-shaped and has a rodlike jugal process that probably contacted the jugal. The quadrate lacks an orbital process, is not pneumatic and is not sutured to the quadratojugal. The lower jaw is slender and straight, with possible hyoids preserved. Two teeth are preserved in the dentary. The teeth are conical, unserrated and one possesses a reabsorbtion pit.
Seven or eight cervical vertebrae are present, the last of which is very elongate. The twelve dorsal vertebrae have craniocaudally short rectangular neural spines. Their centra are strongly grooved laterally and have centrally placed parapophyses. There are twelve pairs of dorsal ribs, the first very short. Uncinate processes are absent. Seven vertebrae make up the sacrum. The last three have long transverse processes and the last two have fused neural spines. The six or seven free caudals also have transverse processes and are slightly shorter than the pygostyle.
The scapula is curved and tapers distally. The coracoid is articulated with the scapula through a flat mobile joint and at an acute angle to it. It is elongate and strut-like, with a concave lateral edge. There is a procoracoid process and a prominent lateral process distally. Dorsally, the coracoid is concave. The sternum is roughly diamond-shaped, with a much shorter anterior end. There is a keel posteriorly and a single pair of lateral processes. These processes project posterolaterally from slightly behind the lateral corners. They are proximally forked, the lateral processending abruptly, the posterior process extending almost past the posterior sternal edge and expanding slightly distally. The furcula has an interclavicular angle of 65 degrees (the paper says 50, but the figure and photo show 65) and a very long hypocleidium, more than half as long as an arm. It is laterally excavated.
The humerus is subequal to the ulna in length. It has a well developed capital groove and cranioventrally projected bicipital crest, but lacks a well developed transverse ligamental groove. The proximal and distal ends are expanded in different planes and the distal condyles are located mainly cranially. It lacks the expanded deltopectoral crest of confuciusornithids. The ulna is bowed with the dorsal condyle developed as a semilunar ridge. The radius is 55% of ulnar width. Two carpals are preserved, the semilunate and a smaller carpal contacting metacarpal III. The metacarpus is unfused and lacks an extensor process. Metacarpal I is 23% as long as metacarpal II. It has an elongate phalanx that reaches slightly past metacarpal II and a reduced ungual. There are three phalanges on digit II. The first is shortest and slightly expanded distally, while the ungual is the largest on the manus. The third metacarpal is slender, bowed and longer than metacarpal II. It has two phalanges, the last a tiny ungual.
The pelvic elements are unfused. The ilium has an elongate rounded preacetabular process and a short slender postacetabular process. The acetabulum is more than 11% of ilial length. There is no fossa for the m. cuppedicus. The pubis has an oval section, distal foot and short symphysis. The plate-like ischia taper distally and lack a symphysis. They are 58% of pubic length.
The femur has a trochanteric crest, a posterior trochantor and a distally bound popliteal fossa, but lacks a deep patellar groove and a distinct neck. The tibia is about 120% of femoral length and lacks a cranial cnemial crest and an extensor canal. The fibula is elongate, but it is uncertain whether it contacted the calcaneum. The astragalus and calcaneum are unfused to each other or the tibia. There is a shallow notch between them. A single unfused distal tarsal is present. The metatarsus is fused proximally. Metatarsal I is slightly curved, about 25% as long as metatarsal II and lies on its mediodistal end. Metatarsal II has a wider trochlea than metatarsal III, and metatarsal IV is slightly more slender than the others and bowed a bit laterally. Metatarsal V is present. There is no intercondylar eminence or distal vascular foreman. Digit II is most robust, digit IV is least.
The body is covered with down feathers, which range from 10 to 20 mm long. An isolated down feather shows barbs converged proximally and an afterfeather. There is no evidence for barbulae or hooklets. Flight feathers reach 94 mm long and are asymmetric. An alula is present. Two long central tail feathers lack barbs or rami proximally, but possess both distally.
The authors performed a cladistic analysis with 86 characters and 11 taxa. This resulted in a single most parsimonious tree, with Confuciusornus basal to an enantiornithine group and an euornithine group. The topology of the enantiornithine group was (Protopteryx (Cathayornis (Concornis + Enantiornis))), while the topology of the euornithine group was (Chaoyangia (Hesperornithiformes (Ichthyornithiformes + Neornithes))). It was grouped in the Enantiornithines based on thirteen characters: scapula with taped distal end (ornithothoracine character); V-shaped furcula with long hypocleidium (Jibeinia + ornithothoracine character); ulna subequal or longer than humerus (Jibeinia + ornithothoracine character, also in some basal oviraptorosaurs); ventral tubercle of humerus separated from head by deep capital incision (ornithothoracine character); prominent cranioventrally projecting bicipital crest on humerus; strong lateral grooves on dorsal centra (pygostylian character); parapophyses located on center of dorsal centra; metatarsal IV significantly smaller than metatarsals II and III (the authors state it is only "slightly more slender", a condition also present in Confuciusornis); trochlea of metatarsal II broader than those of metatarsals III and IV; round articular surface in the proximal tibiotarsus; furcula laterally excavated; metacarpal III extends past metacarpal II. As is apparent, the list drops to six characters once they are examined more closely. Are these enough to keep Protopteryx in the Enantiornithines? I compiled a list of twenty-five phylogenetically informative characters (some multiple state) known in Protopteryx and performed an analysis on Protopteryx and its relatives to determine its phylogenetic position. I only included these twenty-five characters and the enantiornithine synapomorphies Protopteryx exhibits because nearly all analyses (Chiappe 1997, in press; Sereno 2000; Zhang and Zhou 2000; my own) agree on the topology (Archaeopteryx (Confuciusornithidae (Enantiornithines + Ornithurae))), so this need not be tested. Any characters which are unknown in Protopteryx are not going to help determine its placement within a known topology. Also, I am not testing enantiornithine monophyly or interrelationships. Where known, all enantiornithines would score identically in my matrix in any case (even the possibly basal genus Eoenantiornis). The same statements apply to various euornithine taxa I could have included, so they were all condensed into Ornithurae. I also added Jibeinia, as my previous study indicated it had a position between confuciusornithids and ornithothoracines, which is a potential position for Protopteryx as well. My analysis of 31 characters and 6 taxa resulted in a single most parsimonious teee (CI .81, HI .19, RI .78). This tree shows Protopteryx is more derived than confuciusornithids based on: mobile scapulocoracoid joint; interclavicular angle of furcula 70 degrees or less; ulna subequal or longer than humerus; less than four phalanges on manual digit III; reduced manual unguals; scapula tapered distally; long hypocleidium; ventral tubercle of humerus separated from head by deep capital incision; ulnar dorsal condyle developed as a semilunar ridge; alula present. It is more primitive than Jibeinia based on: manual digit I longer than metacarpal II; manual phalanx II-2 longer than phalanx II-1. It is more primitive than ornithothoracines based on: less than eight sacral vertebrae; interclavicular angle of furcula more than 50 degrees; more than one phalanx on manual digit III; maxillary fenestra present; postorbital contacts jugal; metacarpal I not fused to metacarpal II; extensor process absent on metacarpus; pelvic elements unfused; metatarsal V present. The six enantiornithine characters are most parsimoniously seen as convergences. Therefore, I would suggest Protopteryx is a pygostylian more derived than confuciusornithids, but less so than Jibeinia and ornithothoracines. Interestingly, confuciusornithids, Protopteryx, Jibeinia and some enantiornithines all have elongate caudal feathers with the characteristic morphology described above. This seems to have been a symplesiomorphy for pygostylians, although I believe it is derived within Theropoda, not indicative of feathers' prototypic state.
Those who want figures of Protopteryx, just ask. That was relatively painless compared to Microraptor, with a stable phylogeny and everything. Next comes Eoenantiornis.