This is an update to my earlier post Funny Bones about the interactive museum exhibit piece I’m working on for the Carnegie Museum of Natural History. I recommend reading that post for context first if you haven’t.


I have been able to spend a bit of time working on the Tricobatrachus robustus (hairy frog) model that I mentioned before. I haven’t gotten to the final version of what I’d like to bring to the museum, but I was able to move through some quick prototyping because I went into it knowing pretty clearly what I wanted to work on.

First, let’s take a closer look at the frog’s secret claw. A 2008 Biology Letters paper by David Blackburn, James Hanken, and Farish Jenkins, Jr. has some excellent imagery that I was able to directly use for my design work.

Specifically, they include this very nice figure in the paper which shows the mechanism of the phalanx claw with a great deal of clarity:

Blackburn paper figure scale bar is 0.5mm

Thirds of the image, from top to bottom:

  1. the actual claw as it appears while extended through the finger pad
  2. “cleared and stained” view of the fourth toe from the side (“medial view”), showing the claw as well as the small piece of bone that it normally rests against prior to being pulled through the skin. The authors call this small piece of bone the “bony nodule.”
  3. section of the frog’s toe, meaning it has been dissected and photographed. Note the “fmt,” the deep digital flexor tendon, is the thing the frog pulls hard on to separate the phalanx from the bony nodule and make it stick out of the skin.

(If you want to know what all of the labels on the figures mean, read the authors’ descriptive caption in the paper!)

Building the claw

I decided to work towards building a lasercut model of the surprise claw, with the goal of making many replicas of it at low cost and quick production to give out to museum visitors. They’ll build their own claw from a simple kit and get to keep it, so hopefully it’ll both be a fun thing to take home and a reminder of the crazy inventiveness of nature.

First cardboard: basic mechanism

I started by cutting out a spike-on-a-stick shape from a piece of cardboard, and then making a simple (albeit very janky) linkage by which it could rotate about a pivot (pencil). I made a small hole in the cardboard to attach a rubber band (the deep digital flexor tendon) and made a piece of cardboard protrude to catch the end of the phalanx, i.e. playing the part of the bony nodule. It’s covered in blue tape in my photo.

first cardboard model

This model worked alright, but the phalanx was too skinny and prone to bending, so I cut a second one that was thicker. A reasonable start, and it showed me that the mechanism at least made sense.

Second cardboard: lasercut phalanges and bony nodule

I used LibreCAD, free open-source 2D modeling software, to trace the image of the finger bones from figure b in the paper. Looked like this in software once I’d done the drawing:

bone tracing

Then I lasercut those parts out of cardboard, assembled them together using a 1/4”-20 nut and bolt as a pivot, and hot gluing the little bony nodule to the cardboard backing. (It is actually very small and hard to attach to things.)

second cardboard model

Note that I traced in pencil the outline of the frog’s actual finger on the background of that image. I realized it made sense for the backing cardboard piece to represent the finger so as to give the user a better understanding of the physical arrangement of all the parts in the animal.

Third cardboard: modifying and tweaking

Finally, I added a pivot that’s offset from the spike phalanx so that it can swing up and down as it wants to instead of being screwed into the neighboring bone in a weird way. I also drew out the finger outline from the picture, cut holes out of parts that needed to have matching holes, and finally cleaned up the whole thing so that it would fit more neatly into the area of the cardboard and not be so wasteful of material. At that point I had this as my design file:

phalanx rev 2

So far, so good. (Also note that I’m using actual layers this time in the software to elaborate the file a bit.) I cut this on the laser and assembled it. It still needs a bit of work: the claw doesn’t quite come out where it should, because I sort of messed up the claw phalanx’s pivot point. So further revisions will be needed. Still, for about an hour and a half soup to nuts, this was a pretty reasonable stopping point for the day:

third cardboard

This time I used smaller hardware, M3 or M4 nuts and bolts.

Time lapse video of my build process

My friend Kevin DeLand is working on a documentation system prototype and asked if he could take a time lapse of the build session described in this post. The system he’s implementing also includes provisions to pull out particular moments of happiness and frustration, and so you’ll see those highlighted in the video. It’s a prototype, folks! It’s a nice idea and I’m thinking I may return to it in February, when I’m considering doing a project every day and will want a well-oiled documentation system in place.

Next up

I’d like to:

  • Reposition the claw phalanx pivot hole so its extending action mimics the real bone better
  • Make the bony nodule a bit bigger because it’s just too small to put a bolt through reasonably (this will make it anatomically incorrect but then again so is a single-point pivot)
  • Consider how better to arrange and affix the rubber band
  • Consider how better to affix the proximal phalanx that doesn’t move
  • Think about scale. How big should the finished thing be?
  • Work on some poster or other quick informative piece of information so that visitors don’t need to hear the whole story from my mouth.
  • Think about users: how will they hold the whole thing, how will they operate it, etc.