Home > Progress Reports > Status Report

Status Report

The ODP’s team members are amazing. Today we passed 200 verified entries – in fact, we rocketed past it to 228 verified entries! We’ve got 118 unverified entries, and ~45 that are in the que for final checking.

And now, a preliminary plot of femur length vs. tibia length (primarily ornithischians, with a few of the outgroup taxa thrown in). Want to make your own plots? Check out the publicly available data sheet!

Femur Length vs. Tibia Length

Log Femur Length vs. Log Tibia Length. Slope = 0.905, significantly different from 1.

About these ads
Categories: Progress Reports
  1. September 16, 2009 at 10:08 am

    Excellent… though in a log graph both axes should not carry units (mm).
    In the past many arguments has been put forth about the “poor” cursorial ability of dinosaurs (especially therapods) based on a near 1:1 T:F length in comparison with extant ungulates. So, what can we infer from this graph?

  2. Heinrich Mallison
    September 16, 2009 at 11:30 am

    ah, T:F ratio means nothing absolutely, just relatively. After all, dinos are not mammals.

  3. Andy Farke
    September 16, 2009 at 1:26 pm

    Yes, good point Leo – this was a rather “quick and dirty” plot, so the labeling on the axes is not what it could be.

    Re: the T:F ratio, we’re also missing out on the metatarsals in this table, which are another key point in the whole system. One of my pipe-dream goals for this project is to develop a parallel database for mammals, too. . .there are lots of papers talking about evolution of limb proportions, but almost none of them have the data available!

  4. Nathan Myers
    September 16, 2009 at 5:58 pm

    As usual, it’s the ones that fall off the line that are interesting. What’s that hiccup around 2.6 – 2.7mm?

    And where are you finding dinosaurs with such tiny femurs? Did they live in dread of water bears?

  5. September 17, 2009 at 3:15 am

    The single mismatch data point might be an incomplete tibia wrongly entered into the datebase?

  6. William Miller
    September 17, 2009 at 4:41 am

    One useful side product of this database would be to find trendlines to directly compare sizes of different bones. If there are consistent relationships between bone sizes (if the femur is always a predictable percentage of the length of the scapula, for example), this could help improve size estimates for taxa known only from a few bones.

  7. Andy Farke
    September 17, 2009 at 5:34 am

    It’s a log scale – I just forgot to indicate this on the axes. . .sorry! I do rather like the image of Triceratops vs. Tardigrade.

    The hiccup you notice is Zhejiangosaurus lishuiensis, which has the highest femur:tibia ratio (1.9 2.19) of any dinosaur in the database so far! It’s an outlier amongst outliers, but as near as I can tell the data are valid. Talarurus plicatospineus, which has not been completely validated yet so wasn’t included on the plot, has similar proportions. Stegosaurus, Triceratops and Euoplocephalus have nearly as high ratios (~1.7 for each). By contrast, hadrosaurs have a ratio of 1.04 – 1.24.

  8. Andy Farke
    September 17, 2009 at 5:35 am

    Indeed! Very good thought, and perhaps a good side project for someone to work on. As far as I’m concerned, you should go for it! We’ll likely be comparing various bones like this for part of the analysis, but more for the purposes of examining anatomical and functional trends across clades, rather than trying to develop equations to estimate lengths of missing bones.

  9. September 17, 2009 at 7:18 am

    The Zhejiangosaurus paper clearly showed that BOTH hindlimbs are preserved. However, table 2 in the paper failed to indicate WHICH SIDE and the authors neglected (OMG!) to give measurements on both side… femoral length as indicated was 525mm and tibial length, 240mm. Plain T:F ratio is then 2.19.

    In this case it should be prudent to examine whether we should separate Ankylosauridae in a subset analysis… too strange :(

  10. William Miller
    September 17, 2009 at 7:34 am

    I’m not really qualified to do that work myself – I mean, I could determine the trendlines and equations easily enough (that’s just Excel work), but I’m sure there’s more to it…

  11. Andy Farke
    September 17, 2009 at 1:53 pm

    Oops, you are correct on the ratio (which I’ve now updated in my comment) It is correct in the database, I just had mistyped it as the ratio for Talarurus (this is what I get for responding to comments just before bedtime).

    We may do some separate analyses with ankylosaurs (do they scale limb bone lengths similarly to other ornithischians, for instance), but part of the interesting thing here is looking at them in the context of other dinosaurs.

  12. Andy Farke
    September 17, 2009 at 1:57 pm

    For developing equations to predict bone lengths, not really. . .

  13. Nathan Myers
    September 17, 2009 at 6:15 pm

    Z. lishuiensis and T. plicatospineus are interesting as individual outliers, but what I was noticing was a cluster of points well below the line. It’s as if something interesting happens, biomechanically, when your femur reaches a half-meter. The way it is beginning to look, although it will need more data points, is that above .4m, the curve goes stair-stepped, breaking again near .6m and again around .9m. The femur keeps getting longer, but the tibia prefers fixed lengths around .3m, .6m and 1m. Maybe the metatarsals show a complementary pattern?

  14. William Miller
    September 18, 2009 at 2:02 am

    Ah, OK then…

  15. September 18, 2009 at 4:40 am

    I do agree with Andy (commenting above). Perhaps this “stair-step” can be put down to a clade specific phenomenon (presumably, ankylosaurs which happen to be in that size range). An anylasis BY CLADE is important to debug the graph before the meta-analysis of all ornithischian data.

  16. William Miller
    October 24, 2009 at 10:21 pm

    I did some scapula vs. humerus stuff (for all the ceratopsians verified as of Wednesday), and it seemed to work pretty well – Excel gives me a linear trendline with an R-squared of 0.96, which seems pretty good!

  1. October 16, 2009 at 5:20 am

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

%d bloggers like this: