[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: Coelurosaur Analysis Update for Dilong



> My plan for quantitative characters is to follow Theile's (1993) method.
> First, the minimum and maximum values are found among taxa.  Let's say I'm
> coding metacarpal I/II ratios, and (using made up numbers) the lowest is
> Ichthyornis, at 20%, while the highest is Ornithomimus at 105%.  Now I
> follow this equation-
> Xs = 10x (X-20)/(105-20)
> Xs is the character state, while X is the ratio of whatever taxon I'm
> coding.  Say I'm coding Deinonychus, with a ratio of 40%.  It would be
coded
> state 2.  Every quantitative character ends up with 10 states, which are
> completely objective.  Then to make sure the 10 points of possible
variation
> aren't weighted more than the 2 points of variation a qualitative
character
> gets, quantitative characters all get a weight of 1/10.  This also
prevents
> possible errors in measurement from mattering much.

It's indeed objective, and the weighting is a good idea. But it's probably
not a good idea to have the same number of states for every quantitative
character! For some characters this could result in huge cases of
polymorphism due to individual variation -- like, every species that has
state 2 also has state 3, and perhaps vice versa.
        Now how to find out the number of states... probably the only
scientific method to accomplish this would be to use statistics to find
morphological gaps. In other words, an analysis would never get done. :-| I
don't know how TNT deals with quantitative characters, however. Perhaps
they've found a solution.

> - Adding that new enantiornithine from CJES

Wwwww...which one is that?

>    `--+--+--"Alashansaurus"
>       |  |--Fukuiraptor
>       |  `--+--Megaraptor
>       |     `--"Tanycolagreus"

Hey cool. :-)

>       |--Iliosuchus
>       |--Stokesosaurus

So dissimilar from *Aviatyrannis*?

>       |--+--Coelurus
>       |  `--Therizinosaurus

Ouch. Must be an artifact of the half-done coding...

>          `--+--+--+--Ornitholestes
>             |  |  `--Nqwebasaurus
>             |  `--+--Sinosauropteryx
>             |     `--+--NGMC 2124
>             |        `--+--Huaxiagnathus
>             |           `--'Borsti'
>             `--+--Scipionyx
>                `--+--Proceratosaurus
>                   `--+--Compsopgnathus
>                      `--Maniraptoriformes

Looks neat. :-}

> `--+--+--Erliansaurus
>    |  `--+--Nothronychus
>    |     `--+--Alxasaurus
>    |        `--+--+--Beipiaosaurus
>    |           |  `--Neimongosaurus
>    |           `--+--Segnosaurus
>    |              `--+--Erlikosaurus
>    |                 `--Nanshiungosaurus

Interesting that *B.* is so high up.

> Deinonychosauria
> |--+--Scansoriopterygidae
> |  `--+--Archaeopteryx
> |     `--Wellnhoferia
> [...]
>       `--+--+--Rahonavis
>          |  `--Shenzhouraptor
>          `--+--+--Microraptor gui
>             |  `--+--Microraptor zhaoianus
>             |     `--Cryptovolans
>             `--+--NGMC 91
>                `--+--Sinornithosaurus
>                   `--+--Graciliraptor
>                      `--+--Bambiraptor

Harr, harr, harr!!! =8-) _Just_ get the weird troodontids out of here, and
my predictions come true (...well... a small part of them).
        How did you code *Graciliraptor*? The day I got the latest JVP, the
*G.* paper was finally available, and it has turned out to be rather
disappointing. There's no illustration of the entire fossil, and the paper
is rather short.

>                                  `--+--+--Achillobator
>                                     |  `--+--Utahraptor
>                                     |     `--Velociraptor
>                                     `--+--+--Deinonychus
>                                        |  `--Unenlagia

Interesting combinations.

> Avialae
> |--+--Omnivoropteryx
> |  `--+--Yandangornis
> |     `--Avimimus
> `--+--Sapeornis
>    `--+--+--Jixiangornis
>       |  `--+--Confuciusornis
>       |     `--Changchengornis

Man! Why haven't I got that idea much, much earlier -- that *Yandangornis*
and *Avimimus* could be closely related! What do they share in your matrix?
A big toothless pmx? Extra-long legs?
        Interesting that *Jixiangornis* is so far away from the...
phenetically... almost identical *Shenzhouraptor*.
        Good to see how far apart *Omnivoropteryx* and *Sapeornis* can end
up. I've now coded both, and they differ in 4 characters, but I haven't run
the analysis yet (waiting for time to read Julia Clarke's thesis).

>                         |        `--+--Aberratiodontus

How did you code this vile creature? The photos are horrible. The bones seem
to have no surfaces. I tried to code it but gave up.

>                               `--+--+--Parvicursor
>                                  |  `--+--Patagopteryx
>                                  |     `--Songlingornis

Speechless.

>                                  `--+--+--Yanornis
>                                     |  `--Ornithurae
>                                     `--+--Ambiortus
>                                        `--+--Apsaravis
>                                           `--Yixianornis

How did *Yanornis* end up closer to Ornithurae than the other 3?