The cladogram from the supp. inf. of the *Anchiornis* paper

[David Marjanovic <david.marjanovic@gmx.at>]

The recent Nature paper that finds *A.* as a troodontid has 52 pages of 
supplementary information. For those who haven't downloaded it yet 
(unlike some other journals, Nature does not put its supp. inf. behind a 
paywall), I have retyped the cladogram (fig. S5) below.

First in general terms, because it's so large (87 taxa, 363 characters; 
CI = 0.34):

--+--*Allosaurus fragilis*
 `--+--*Sinraptor*
    `--+--Tyrannosauroidea
       `--+--Compsognathidae
          `--+--Ornithomimosauria
             `--+--*Ornitholestes*
                `--+--Segnosauria
                   `--+--Alvarezsauridae
                      `--+--Oviraptorosauria
                         `--+--what they call Avialae
                            `--+--Troodontidae
                               `--Dromaeosauridae

The position of *Sinraptor* probably means that the outgroup was 
erroneously restricted to *Allosaurus*. (If you don't designate an 
outgroup, PAUP* uses the first OTU in the list as the outgroup.)

Then in detail:

Tyrannosauroidea
 |--+--*Tanycolagreus*
 |  `--*Coelurus*
 `--+--*Dilong*
    `--+--*Eotyrannus*
       `--+--*Tyrannosaurus*
          `--*Gorgosaurus*

Jurassic tyrannosaurs all over the place, even though *Guanlong* and 
*Proceratosaurus* are not in the matrix.

Compsognathidae
 |--*Huaxiagnathus*
 `--+--*Compsognathus*
    `--*Sinosauropteryx*

Ornithomimosauria
 |--*Harpymimus*
 `--+--*Deinocheirus*
    `--+--*Shenzhousaurus*
       `--+--*Pelecanimimus*
          `--+--*Garudimimus*
             `--+--*Archaeornithomimus*
                `--+--*Gallimimus*
                   `--+--*Struthiomimus*
                      `--+--*Anserimimus*
                         `--*Ornithomimus*

Weird positions for *Pelecanimimus* and *Anserimimus*.

Segnosauria
 |--*Falcarius*
 `--+--*Beipiaosaurus*
    `--+--*Alxasaurus*
       `--+--*Nothronychus*
          |--*Erliansaurus*
          |--*Nanshiungosaurus*
          `--+--*Neimongosaurus*
             `--+--*Segnosaurus*
                `--+--*Erlikosaurus*
                   `--*Therizinosaurus*

Alvarezsauridae
 |--*Alvarezsaurus*
 `--+--*Patagonykus*
    `--+--*Mononykus*
       `--*Shuvuuia*

Perhaps a bit small, this taxon sample... but the biggest omission is 
not the author's fault. I'll send the SVP abstract next: the place where 
*Guanlong* and *Limusaurus* come from has now yielded an Oxfordian 
alvarezsaurid.

Oviraptorosauria
 |--*Caenagnathus*
 |--*Avimimus*
 |--+--*Protarchaeopteryx*
 |  `--*Incisivosaurus*
 `--+--*Caudipteryx*
    `--+--*Microvenator*
       `--+--*Oviraptor*
          `--+--+--+--*Khaan*
             |  |  `--*Conchoraptor*
             |  `--+--*Ingenia*
             |     `--*Heyuannia*
             `--+--+--*Rinchenia*
                |  `--IGM 100/42
                `--+--*Citipati*
                   `--+--*Elmisaurus*
                      |--*Hagryphus*
                      `--*Chirostenotes*

*Caenagnathus* and *Chirostenotes* as far apart as at all possible! Wow. 
Nice confirmation for the position of *Oviraptor* as the sister-group to 
all other oviraptorids including "elmisaurids". Interesting confirmation 
for the tendency of dinosaurologists to resist any lumping of sister 
species into the same genus (*Incisivosaurus* is still there).

Avialae
 |--+--*Epidendrosaurus*
 |  `--*Epidexipteryx*
 `--+--+--*Archaeopteryx*
    |  `--*Wellnhoferia*
    `--+--*Jeholornis*
       `--+--*Sapeornis*
          `--+--*Confuciusornis*
             `--+--*Protopteryx*
                `--*Yanornis*

Avialae is defined on page 2 of the supp. inf. as "the most inclusive 
clade including Neornithes but not *Deinonychus antirrhopus* and 
*Troodon formosus*". Aves is defined right below as "the least inclusive 
clade including *Archaeopteryx* and Neornithes".

I still don't believe in *Wellnhoferia*. :-)

*Epidendrosaurus* probably includes *Scansoriopteryx*, and *Jeholornis* 
*Shenzhouraptor*.

I wonder how much this topology can be trusted, because among the 
autapomorphies of Avialae that *Anchiornis* lacks is "reversed hallux". 
Ehem.

Troodontidae
 |--*Sinovenator*
 `--+--*Anchiornis*
    `--+--*Mei*
       `--+--*Byronosaurus*
          `--+--*Sinornithoides*
             |--IGM 100/44
             `--+--*Troodon*
                `--+--*Saurornithoides junior*
                   `--*S. mongoliensis*

Note that, although it is the oldest troodontid in this matrix and has 
plesiomorphies like {widely spaced teeth except at the jaw tips}, 
*Anchiornis* is not the basalmost one. There are two ghost lineages high 
up into the Early Cretaceous.

Dromaeosauridae
 |--+--*Rahonavis*
 |  `--+--*Unenlagia*
 |     `--*Buitreraptor*
 `--+--+--*Bambiraptor*
    |  `--+--*Sinornithosaurus*
    |     `--+--*Microraptor*
    |        `--NGM C91
    `--+--*Adasaurus*
       `--+--IGM 100/1015
          `--+--+--*Velociraptor*
             |  `--*Saurornitholestes*
             `--+--*Deinonychus*
                `--+--*Atrociraptor*
                   `--+--*Achillobator*
                      `--+--*Dromaeosaurus*
                         `--*Utahraptor*

So *Deinonychus* is a dromaeosaurine, but *Adasaurus* is not... <head spins>

But at least *Rahonavis* is here to stay, 4 to 6 nodes away from 
*Anchiornis*, *Archaeopteryx* and *Jeholornis*.