Building a Madtsoiid (Patreon)

I said I had to do more snakes (for The Big Book). And here's one I've been putting off for ages... this, because I needed to make it look BIG and hence knew it would take me ages. It's a madtsoiid, a member of an extinct group that persisted from the Cretaceous to the Pleistocene and maybe even early Holocene. The biggest madtsoiids were among the biggest of all snakes, some reaching 7 metres or so (and perhaps more). I wanted to illustrate one of these big species, so reconstructed Gigantophis garstini from the Eocene of Egypt. It was long said to be 9 or 10 metres long, though recent work by Jason Head and colleagues has brought it down to 6.9 metres... which is big enough.

Not much is known about the life appearance of madtsoiids, really. Their vertebrae and ribs show that the big species were robust, chunky snakes, superficially resembling anacondas or especially big pythons. With this in mind, I started by building a generic mega-snake form - never easy... As you can see, it looks ridiculous.

At this point, note that I gave the animal a boxy, chunky head, because I assumed that it would be built like a gargantuan python. This proved to likely be wrong - read on. I then went through successive improvements as I refined the snake's body and curves, improved its head, and eventually added scalation and then a colour pattern (based on that of ground-living pythons). I'm not good enough to draw scales properly (I'm not Gabriel Ugueto), so they look terrible before I begin colouring - I simply can't get them right when we come to the curved sections of the body. In the end, it doesn't matter that much, but it drives me mad. Note all the lines which show where I've been erasing and redrawing the lines.

So what's the deal with the head? Good madtsoiid skulls are known, in particular for the Australian taxon Yurlunggur from the Oligocene or Miocene of Riversleigh (it was described in Nature by John Scanlon in 2006). While Yurlunggur is not the same thing as Gigantophis, I'm guessing that it gives a better guide to the look of the madtsoiid skull than anything else (a partial skull is also known for the Plio-Pleistocene madtsoiid Wonambi, but it doesn't provide so much information on snout shape). As you can see from the screengrab below, the Yurlunggur skull does not reveal a deep, boxy snout, but a shallow(ish) one where the dorsal margin is very slightly concave. I used the general form of this skull to build the head below...

And here it is. I think it's pretty successful. Nothing is known about the scalation pattern of these snakes, nor is it really possible to reconstruct what it might be like given that the snakes surrounding madtsoiids in the cladogram (like scolecophidians, perhaps) are so highly modified. I sorta liked the idea of making my madtsoiid look like an Amethystine python and started with that, but the very distinctive labial scales of this species (which are massive and really tall) look too distinctive to me, so I ended up dropping this idea. What we have in the end is a snake that's based on pythons but isn't quite like any of them, nor indeed on any living snake.

Add all this together, and here's the final look I adopted for my madtsoiid....

The real fun with these illustrations is to see what they look like when added into the montages I'm building. Gigantophis and other big madtsoiids are, as I've said, among the biggest of snakes, so this animal should look HUGE when compared to most of the other snakes I've illustrated so far. Here's how the montage looks like right (remember that all of this is provisional, and not close to the final version). I think it looks pretty neat and hope you like it :)

As always, this stuff is happening because of YOUR support. Thanks as always. More soon.