Chiappe and Calvo, 1994. Neuquenornis volans, a new Late Cretaceous bird (Enantiornithes: Avisauridae) from Patagonia, Argentina. Journal of Vertebrate Paleontology. 14(2), 230-246.

This avian analysis was based on Chiappe's (1992) dissertation and had several revisions until being replaced by Chiappe et al. (1998) and later Chiappe (2001). It was the first analysis to place Patagopteryx as a basal euornithine, and the first to support hesperornithines and Ichthyornis as closer to Aves than enantiornithines are.

Phylogeny

|--Dromaeosauridae
|--Archaeopteryx
`--Ornithothoraces
   |--Iberomesornis
   `--+--Enantiornithes
      `--+--Patagopteryx
         `--Ornithurae
            |--Hesperornithiformes
            |--Ichthyornis
            `--Neornithes

Taxon Issues

Dromaeosauridae- The text states this is only based on Deinonychus.
Iberomesornis- This is recognized today as being a member of Enantiornithes.
Enantiornithes- This is based on Alexornis, Avisaurus, Enantiornis, Gobipteryx, Kizylkumavis, Lectavis, Martinavis? vincei (unnamed at the time), Nanantius, Neuquenornis, Sazavis, Soroavisaurus, Yungavolucris and the unnamed PVL material. While Chiappe and Calvo state Cathayornis (= Sinornis) is a "putative enantiornithine", they also state "further studies are necessary to confirm its alocation to the clade" and do not seem to have used it to code based on obvious characters such as the presence of premaxillary teeth or a pygostyle. The holotype of Sinornis was not included even in the 1996 version, while Concornis was included in 1996 but seems absent in earlier versions since Sanz et al. (1995) added it to the matrix as a separate OTU. All of the above taxa with the possible exception of Lectavis are recognized today as enantiornithines.
Hesperornithiformes- This is more commonly called Hesperornithes now. According to Chiappe's more detailed 1996 paper, this is based on Baptornis, Parahesperornis and Hesperornis.
Neornithes- I follow most current authors in calling this crown clade Aves. Again according to Chiappe (1996), this is based on 34 recent species in 31 families, the basalmost of which are Rhynchotus, Pterocnemia, Struthio, Chauna, Amazonetta, Branta and Bonasa.

Coding Issues

3. Patagopteryx has nasal processes of its premaxillae which extend to the frontals (Chiappe, 2002).

4. Ichthyornis lacks premaxillary teeth (Lamb, 1997).

5. The external nares in Enantiornithes (Gobipteryx- Chiappe et al., 2001) are smaller than their antorbital fenestra. In addition, this is true of Parahesperornis (Martin, 1984) and some basal avians such as Anas (Digimorph), so Hesperornithes and Aves are recoded as polymorphic.

6. The caudal maxillary sinus of Hesperornis and Aves may be homologous to the maxillary fenestra, but is not coded as such here.

7. The naris does not extend posteriorly to the level of the jugal in enantiornithines (Gobipteryx- Chiappe et al., 2001) or hesperornithines (Parahesperornis- Martin, 1984; Hesperornis- Witmer, 1990).

8. Enantiornithines (Gobipteryx- Elzanowski, 1995) have ectopterygoids. The uncinate in Hesperornis and some avians may be homologous, but their apomorphic coding is tentatively retained.

9. The orbital process of the quadrate is broken off in Ichthyornis (Clarke, 2004), making its pointedness unknown.

11. The presence of braincase contact with the quadrate in Deinonychus is unknown, though it is present in Archaeopteryx (Elzanowski, 2002).

12. The presence of quadrate pneumaticity in Deinonychus is unknown. It is present in Archaeopteryx (Alonso et al., 2004).

13. Patagopteryx lacks articular pneumaticity (Chiappe, 1996).

14. Deinonychus (Norell and Makovicky, 1999) and Iberomesornis (Sereno, 2000) have large anterior dorsal hypapophyses. Enantiornithines are tentatively retained as having them, presumably based on Lecho material.

16. This character (less than eleven dorsal vertebrae) is correlated with character 15 (less than thirteen dorsal vertebrae), so taxa coded as lacking character 15 (Deinonychus and Archaeopteryx) are coded unknown for this one.

17. Deinonychus is polymorphic for this character, as its anterior dorsals do have neural canals over 40% of central height (Ostrom, 1969).

19. Iberomesornis should be coded unknown, as related taxa often have heterocoely confined to anterior cervicals, and only amphicoelous posterior cervicals are known.

20. Enantiornithines have a pygostyle (Gobipteryx- Kurochkin, 1996).

22. Deinonychus has uncinate processes (Paul, 2002), while enantiornithines are best coded unknown as no taxon used by Chiappe has a complete enough ribcage to code.

23. The scapulocoracoid articulation is narrow in Iberomesornis (Sereno, 2000).

24. Iberomesornis lacks a procoracoid process (Sereno, 2000).

26. Iberomesornis' coracoid is exposed in ventral view, so that the level of the scapular articulation is unknown (Sereno, 2000).

27. Iberomesornis' coracoid foramen is displaced medially (perhaps to the point where it no longer penetrates the bone) (Sereno, 2000).

28. Since Iberomesornis' coracoid articulation is not preserved, it is unknown what angle the scapula articulated at (Sereno, 2000). Patagopteryx has a short proximal portion of the coracoid which articulates at a low angle to the scapula (Chiappe, 2002), as in Archaeopteryx. As palaeognaths primitively lack a coracoid which is angled sharply from the scapula, Aves is recoded to be polymorphic.

30. Since palaeognaths primitively have straight scapulae, Aves is recoded as polymorphic.

31. The sternum of Deinonychus is unpreserved, so it cannot be coded for sternal size and shape. Ichthyornis actually has a broad sternum only ~1.32 times longer than wide (Chiappe, 2004).

32. No enantiornithine used by Chiappe is preserved well enough to exclude the possibility it had gastralia, nor are Patagopteryx, Ichthyornis or hesperornithines.

33. The sterna of Deinonychus and Archaeopteryx are unpreserved, so cannot be coded for the presence of a sternal keel. Iberomesornis has a sternal keel (Sereno, 2000). Again, palaeognaths primitively lack the character, so Aves is coded as polymorphic.

34. Hesperornithines (Hesperornis- Marsh, 1880) have wide interclavicular angles. The clavicular fragments of Ichthyornis do not allow the interclavicular angle to be estimated (Clarke, 2004).

35. Enantiornithines are polymorphic for the presence of a twist in the humeral shaft, as Neuquenornis lacks this (Chiappe and Calvo, 1994).

36. Ichthyornis has an ulnohumeral ratio of 96%, comparable to the Munich Archaeopteryx, so is coded as plesiomorphic. Aves is best coded as polymorphic, since palaeognaths, Anas and others have short ulnae.

37. Both Archaeopteryx (Elzanowski, 2002) and Iberomesornis (coding in Chiappe, 2001) have anteriorly concave humeral heads. Aves should be coded as polymorphic, as basal palaeognaths (Struthio, Rhea, Apteryx- codings in Livezey and Zusi, 2007) also have the plesiomorphic condition.

38. Deinonychus (Ostrom, 1969) and Patagopteryx (Chiappe, 1996) have capital grooves on their humeri.

39. The transverse ligamental groove is coded as absent in palaeognaths and galloanserines by Livezey and Zusi (2007), so is here coded absent in Aves.

40. Sereno (2000) notes that in Iberomesornis' humerus, the radial condyle is missing and the ulnar condyle at least rounds into the cranial surface. It is best coded as unknown due to the poor preservation.

42. Archaeopteryx's radius can be slightly less than 70% of ulnar width (Wellnhofer, 1974; 1993), making it polymorphic. Ichthyornis' radius is over 70% of ulnar width (Clarke, 2004). Aves should be polymorphic, as paleognaths primitively have broad radii.

44. Enantiornithines do not have an extensor process (Enantiornis- Chiappe and Walker, 2002; Neuquenornis- Chiappe and Calvo, 1994), merely a convex medial margin to their metacarpal I.

45. Iberomesornis has a partially fused pelvis, as the puboischial suture is fused (Sanz et al., 2002). Enantiornithines should be coded as polymorphic, since Gobipteryx lacks at least iliopubic and ilioischial fusion (Kurochkin, 1996).

46. Iberomesornis has an acetabulum 18% of ilial length (Sereno, 2000), so is coded as plesiomorphic.

47. This character (ilium, pubis and ischium subparallel) is partly correlated with character 55 (pubis retroverted more than 45 degrees past vertical), as any taxon with 47 must have 55. Thus it is here restricted to the ischial portion.

48. Deinonychus has a large ilial antitrochanter (Novas, 1996). Enantiornithines are polymorphic, as Gobipteryx lacks the character (Kurochkin, 1996).

50. Archaeopteryx is polymorphic for the presence of a cuppedicus fossa on the ilium (Elzanowski, 2002).

51. Archaeopteryx lacks an ischial symphysis (e.g. coding in Turner et al., 2007 and Hutchinson, 2001).

52. Enantiornithes (Gobipteryx- Kurochkin, 1996) have a pubic symphysis. The pubis of Ichthyornis is distally incomplete (Clarke, 2004), making it impossible to code for this character.

53. Iberomesornis (Sereno, 2000; Sanz et al., 2002) and enantiornithines (Gobipteryx- coding in Clarke and Norell, 2002) have non-compressed pubic shafts.

54. Enantiornithines should be coded as lacking a pubic boot based on Gobipteryx (Kurochkin, 1996). This distal pubis is unpreserved in Ichthyornis (Clarke, 2004).

55. Deinonychus' pubic angle is uncertain as the proximal pubis is unpreserved (Ostrom, 1976).

56. Deinonychus has a capital ligament fossa in its femur (coding in Senter, 2007).

58. Iberomesornis lacks a patellar groove (Chiappe, 1996).

59. Archaeopteryx does not have a distally closed popliteal fossa (coding in Clark et al., 2002).

60. This is a composite character, coding for confluence between the lateral condyle and ectocondylar tuber (to form a tibiofibular crest) in the second state, and for the posterior projection of this crest in the third state. Chiappe (2002) states Patagopteryx has a tibiofibular crest, and the illustration shows it is not prominent.

61. Iberomesornis has a posterior trochanter (Sereno, 2000).

62. Iberomesornis has been reinterpreted as having a completely fused tibiotarsus (Sereno, 2000). The coding for Enantiornithes is retained, though Gobipteryx has only partial fusion between its tibia and astragalocalcaneum (Kurochkin, 1996). This may be a juvenile character, as other young enantiornithines and Baptornis are known to lack fusion as well.

63. Enantiornithes is coded as polymorphic for the precence of two cnemial crests, as Alexornis has this character (Brodkorb, 1976).

64. Patagopteryx has been reinterpreted as lacking an extensor canal on its tibiotarsus, with the structure being merely the depressed area between retinaculum extensor tubercles (Clarke and Norell, 2002).

65. Archaeopteryx has a laterally oriented iliofibularis tubercle (coding in Hutchinson, 2002). This also seems true in enantiornithines, based on Gobipteryx where there is a scar on the lateral surface but no raised tubercle (Kurochkin, 1996).

68. Iberomesornis has no free distal tarsals (Sereno, 2000).

69. The character "metatarsal fusion starting distally" can only apply in those taxa which have distal metatarsal fusion, which is covered by character 67. Thus the taxa lacking distal fusion (Deinonychus, Archaeopteryx, Iberomesornis and Enantiornithes) are recoded as unknown for this character.

70. Iberomesornis lacks metatarsal V (Sereno, 2000). As Ichthyornis specimens are largely disarticulated and only preserve one pedal phalanx and no metatarsal Is, the absence of metatarsal V is uncertain in this genus.

73. Enantiornithes is coded as polymorphic, as Avisaurus gloriae (Varricchio and Chiappe, 1995) has an enclosed distal vascular foramen. Hesperornithines should be coded as plesiomorphic, as basal taxa like Baptornis (Martin and Tate, 1976) and Parahesperornis (Bell and Everhart, 2009) lack enclosed distal vascular foramina.

General analysis conclusions- This was quite a good analysis in many ways. It had a decent number of characters, only three of which are correlated with others and only one of which is a composite. Most characters are also quantified well so that they are unambiguous. One issue is that since every character is coded as present in Aves, the analysis is predisposed to find a topology where every taxon forms a successively closer sistergroup to Aves. This almost eliminates the possibility of finding combinations like Sauriurae, an Enantiornithes including Iberomesornis, or Odontornithes. Only 15% (90/584) of the matrix was miscoded. Many of these are due to discoveries since 1994 such as Gobipteryx postcrania (Kurochkin, 1996) and more complete skulls (Chiappe et al., 2001), and the restudy of Iberomesornis by Sereno (2000). Another recurring issue is the coding for Aves, which is different here due to the recent recognition tinamous are nested deep in Palaeognathae, making ratite anatomy influence the calculated basal state of the clade. Of course, taxa like lithornithids show that Chiappe's view of ratite morphology being a reversal is almost certainly correct, but this should not be assumed given only the taxa he examined combined with a modern topology. Once recoded, the new topology is-

|--Deinonychus
|--Archaeopteryx
`--Ornithothoraces
   |--Iberomesornis
   |--Enantiornithes
   `--Ornithuromorpha
      |--Patagopteryx
      `--+--Hesperornithes
         |--Ichthyornis
         `--Aves

Thus the only difference is that Iberomesornis is in a polytomy with enantiornithines and ornithuromorphs, and the tree is now compatible with the current consensus.

Phylogenetic conclusions- The table shows the number of extra steps needed to accomodate each rearrangement using Chiappe and Calvo's original matrix, and his recoded matrix. A negative number means the arrangement is already most parsimonious, but that many steps are needed to undo it. 86 82

rearrangement original recoded
(Deinonychus,Hesperornithes,Ichthyornis,Aves(Archaeopteryx,Iberomesornis,Enantiornithes)) (Martin, 1983) 29 23
(Deinonychus,Iberomesornis(Enantiornithes,Aves)) -6 0
(Deinonychus,Aves(Iberomesornis,Enantiornithes)) 6 0
(Deinonychus,Hesperornithes,Ichthyornis(Patagopteryx,Aves)) (Alvarenga and Bonaparte, 1992) 12 15
(Deinonychus,Enantiornithes,Ichthyornis,Aves(Patagopteryx,Hesperornithes)) (Chatterjee, 1999) 12 16
(Deinonychus,Archaeopteryx,Aves(Patagopteryx,Enantiornithes)) (Feduccia, 1999) 7 5
(Deinonychus,Archaeopteryx,Hesperornithes(Enantiornithes,Ichthyornis,Aves)) (Cracraft, 1986) 13 18
(Deinonychus,Aves(Hesperornithes,Ichthyornis)) 0 0
(Deinonychus,Hesperornithes(Ichthyornis,Aves)) 0 0
(Deinonychus,Ichthyornis(Hesperornithes,Aves)) 0 0

Both the original codings and the recodings very strongly reject Sauriurae. Iberomesornis is moderately supported as being outside Enantiornithes+Aves in the old matrix, but its position relative to those groups is now ambiguous. Placing Patagopteryx in Aves (as it was originally thought to be) or making it a hesperornithine are both strongly rejected by both matrices. An enantiornithine Patagopteryx is only moderately rejected by both though. Cracraft's old topology placing hesperornithines outside Ornithothoraces is strongly rejected by both matrices. The interrelationships of Ichthyornis, hesperornithines and avians are ambiguous in both versions.