Enantiornithine ontogeny

[Michael Mortimer <mickey_mortimer111@msn.com>]

David Marjanovic asked-

> Is *Liaoxiornis* diagnostic when its ontogenetic age is taken into account?

I doubt it, but I haven't examined the question since my old DML post.

> How did you code the ontogeny-dependent characters of *Iberomesornis* --  
> as "?"?

They seem to be mostly misinterpretations (Sereno, 2000).  The exceptions 
are the concave lateral sternal margin and proximally convex humeral head, 
both of which are also plesiomorphic for enantiornithines.

Conveniently, I had a section written up on ontogeny...

Several included specimens (Liaoxiornis, Dahlingheornis, IVPP V14238 (the 
Yixian embryo), LP-4450-IEI (the Catalan nestling), GMV 2158 and GMV 2159 
(two Yixian specimens examined by Chiappe 2002)) are obviously juvenile, and 
therefore differ from adult specimens in a number of characters (no sternal 
keel or lateral processes, lack of fusion, etc.).  I initially coded these 
as unknown then ran the analysis to find the juveniles still claded 
together.  After examining the list of apomorphies for their node, I was 
able to identify several other probable juvenile characters.  Once these 
were all coded as unknown for the juveniles, they began intermingling with 
adult specimens.  The juvenile characters identified were-
1.(?) external naris smaller than antorbital fenestra (LP-4450-IEI, 
Liaoxiornis, GMV 2158, IVPP V14238)
2.atlantal hemiarches fusion absent (LP-4450-IEI)
3.less sacral vertebrae (6 in Liaoxiornis, GMV 2158; 6-7 in IVPP V14238; no 
in Dalingheornis)
4.(?) caudal transverse processes shorter (Liaoxiornis, GMV 2159)
5.pygostyle absent (Dalingheornis, IVPP V14238; no in Liaoxiornis, GMV 2159)
6.sternum reduced in size (Liaoxiornis, GMV 2158, GMV 2159)
7.sternal keel absent (Liaoxiornis, GMV 2159)
8.lateral sternal process absent (Liaoxiornis, GMV 2158, GMV 2159; no in 
Dalingheornis)
9.(?) lateral sternal process unexpanded (Dalingheornis)
10.medial sternal process absent (Liaoxiornis, GMV 2158, GMV 2159)
11.higher interclavicular angle (Dalingheornis especially; Liaoxiornis, GMV 
2158, GMV 2159; no in LP-4450-IEI, IVPP V14238)
12.(?) lateral coracoid not convex (Dalingheornis, LP-4450-IEI, Liaoxiornis, 
GMV 2158, GMV 2159, IVPP V14238)
13.capital groove absent (Dalingheornis; no in LP-4450-IEI)
14.humeral head concave in middle (Dalingheornis, Liaoxiornis, GMV 2158, GMV 
2159; uncertain in LP-4450-IEI)
15.(?) humeral pneumotricipital fossa absent (LP-4450-IEI, GMV 2158)
16.(?) deltopectoral crest reduced in depth (Dalingheornis, Liaoxiornis, GMV 
2158, GMV 2159, IVPP V14238)
17.humeral distal condyles undeveloped (Dalingheornis?, Liaoxiornis?, GMV 
2159, IVPP V14238; no in LP-4450-IEI)
18.semilunate ridge on distal ulna absent (Dalingheornis, Liaoxiornis, GMV 
2158, GMV 2159; no in LP-4450-IEI)
19.(?) manus longer than humerus (IVPP V14238; equivocal in Dalingheornis; 
no in LP-4450-IEI, GMV 2159)
20.carpometacarpus fusion absent (Dalingheornis, Liaoxiornis, GMV 2158, GMV 
2159; no in LP-4450-IEI)
21.pelvic fusion absent (GMV 2158)
22.(?) pubic boot absent (GMV 2158, IVPP V14238)
23.(?) ischiopubic ratio <66% (Dalingheornis, GMV 2158)
24.cnemial crest absent (Dalingheornis, GMV 2158)
25.tibiotarsal fusion absent (Dalingheornis, GMV 2158, GMV 2159; no in 
Liaoxiornis?)
26.astragalus fused to calcaneum
27.tarsometatarsal fusion absent (GMV 2158; GMV 2159; IVPP V14238; no in 
Dalingheornis)
28.metatarsal trochlea undeveloped (Dalingheornis)

It became obvious that the supposedly primitive characters of a few taxa 
(Jibeinia, Protopteryx, Longipteryx) were also potential juvenile 
characters.  This isn't necessarily true however, as many of these 
characters' juvenile states are primitive for Enantiornithes and more 
inclusive clades.  I created a new ontogeny matrix with these 28 characters, 
and included the juveniles, three taxa mentioned above, as well as some 
others which have apparent juvenile characters (Gobipteryx "N. valifanovi" 
specimen, Hebeiornis, "Cathayornis" caudatus).  Iberomesornis has been 
suggested to be juvenile in the past, so was also included.  Sinornis was 
used as an example of an adult enantiornithine.  Running this created an 
'ontogram', as Carr (2005) and Tykoski (2005) have done for tyrannosaurids 
and coelophysoids respectively.  The embryo was used as an outgroup.  There 
was some homoplasy, but the resulting ontogram was-
|--IVPP V13238
`--+--Liaoxiornis
  |--GMV 2158
  |--GMV 2159
  `--+--Dalingheornis
     `--+--Jibeinia
        `--+--Protopteryx
           |--Longipteryx
           `--+--Gobipteryx
              |--Hebeiornis
              `--+--LP-4450-IEI
                 |--"Cathayornis" caudatus
                 |--Iberomesornis
                 `--Sinornis
It should be read as a progression of age, with more "derived" taxa being 
older.  Interestingly, LP-4450-IEI groups close to the adult.  It does 
possess juvenile bone texture, so we know it wasn't fully grown.  The taxa 
'less derived' than it on the tree should be examined for this texture too.

The transformations occuring at each age level are as follows-
Specimens older than IVPP V14238-
- manus shorter than humerus (not in Protopteryx, Longipteryx).
Specimens older than Liaoxiornis, GMV 2158, GMV 2159-
- eight sacral vertebrae (not in Protopteryx); lateral sternal processes 
ossify; **metatarsals fuse.
Specimens older than Dalingheornis
- **pygostyle fused (also in Liaoxiornis, GMV 2159; not in "Cathayornis" 
caudatus); **sternum enlarges; medial sternal processes ossify (not in 
Protopteryx); pneumotricipital fossa; semilunate ridge on distal ulna.
Specimens older than Jibeinia-
- naris longer than aof (not in Gobipteryx); sternal keel; capital groove; 
**humeral distal condyles develop; pubic boot (not in Gobipteryx); ischium 
lengthens.
Specimens older than Protopteryx, Longipteryx-
- low interclavicular angle (also in IVPP V14238); lateral coracoid convex 
(also in Jibeinia; not in LP-4450-IEI, Iberomesornis); humerus concave in 
middle? (also in Longipteryx; not in Iberomesornis); **carpometacarpus 
fused; cnemial crest (not in "Cathayornis" caudatus); **astragalus fused to 
calcaneum; **metatarsal trochlea developed.
Specimens older than Gobipteryx, Hebeiornis-
- lateral sternal processes expanded distally (also in Longipteryx); pelvis 
fused; **proximal tarsals fused to tibia.

Jibeinia, Protopteryx and Longipteryx were then recoded with the characters 
they possess juvenile morphology for coded as unknown.  These alternative 
"adult" versions were run separately and together to determine their effect 
on the taxon's placement.  Characters marked by asterisks are those in which 
Confuciusornis and Sapeornis possess the derived/adult state, making it 
unlikely (though not impossible) that a basal enantiornithine would have the 
basal state due to phylogeny instead of ontogeny.  When taxa have 
juvenile/basal states which overall younger specimens lack, it could mean 
they kept the juvenile/basal states into adulthood.  For instance, 
Protopteryx and Longipteryx may have had long manus even as adults, while 
adult Gobipteryx may have had small external nares and no pubic boot.  
Protopteryx and Longipteryx ended up having the same position whether 
juvenile states were coded for them or not, but Jibeinia was quite a bit 
more derived if assumed to be a juvenile.  As Jibeinia was found to be 
younger in my test above, perhaps that marks the point in life where 
enantiornithine taxa can be safely analyzed without the need to adjust their 
codings for age?

Mickey Mortimer