So having been involved in several palaeoart projects of various kinds in various ways of late, it seemed sensible that I put fingers to keyboard and try to come up with something useful. I should stress that of course I know and work with a number of full time palaeoart people in various ways and I’m certainly not trying to teach Luis Rey, Mark Hallett or Bob Nicholls how to do their jobs! But there do seem to be many an aspiring palaeoartist out there and there seem to be some common enough errors or misconceptions out there that are worth flagging up to save them being made or corrected. This will be, I hope, a bit of a starter pack and just generally handy for those trying to draw dinosaurs or other extinct critters. Whatever your style or methods the point is (I assume) to produce a piece that as accurately as possible reproduces what the animal did, or could, have looked like.
Start with the anatomy, and indeed finish with it too. If the head is exactly half the length of the arm, and identical to the length of the thigh then the image will need to represent that. If there should be 15 teeth in the upper jaw then they need to be there. If the tail is limited in lateral motion then don’t give it a huge bend in the middle.
This is likely to mean digging into the literature or consulting experts on the subject as well as checking specimens or photos of specimens. Don’t just rely on old artwork or old reconstructions – that is how mistakes are replicated and reproduced. Check every detail as far as you can – do the wrists bend this way or that, could it rear up, do those toes have claws or not, what kind of scales are there, how big was the nostril etc. There is a lot of literature out there and much of it is now available online so get hold of it and read it.
Investigate the specimen or specimens you are working on. If you want to produce an image of the ‘normal’ adult and only juveniles are preserved you will have to modify the overall scale or proportions of the bones you have to get it right. If there is good reason to think that the bones have been distorted or soft tissues moved etc. then these will have to be restored accordingly.
When doing a life restoration, be aware of what things do and do not impinge on the external appearance of the animal. While doubtless things like Velociraptor, Linheraptor and Tsaagan all looked quite different in real life, if you stripped off the colours and feathers and skin, the three would be very hard to tell apart. Subtle changes in how the bones of the skull fit together are important anatomical details for palaeontologists, but will have no effect on how head of the living animal looked.
Most likely you will not be working from a partial (possibly a very partial) skeleton or skull and obviously you’ll need to bring in extra information from other sources. The best ones to take are naturally the nearest relatives to the animal at hand. Need some missing bits of Spinosaurus, then use Baryonyx or Suchomimus if possible and then go to slightly more distant relatives like megalosaurs. Don’t just stick it on an allosaur or tyrannosaur.
Even after all of this you will likely have missing gaps in your knowledge, or areas of uncertainty or contradictions, and these will have to be completed as best as possible. Things like the amount and distribution of muscles on certain parts of the body, exact degrees of limb movements or postures, exact size of the eye or position of the ear are likely to be impossible to reproduce accurately, but not impossible to reproduce sensibly. Check anatomical texts or using living animals to see how muscle groups are organised and arranged and which muscles are large and which small.
All of this should get you an anatomically accurate looking animal, but of course there is more. First off, you need to get a correct covering and appropriate colour scheme going. While there are few fossils with skin etc. preserved (among the archosaurs at least) a considerable amount is known when taken across many species (even if not all of them are too closely related) and more can be inferred from living animals too (here especially, crocs and birds) to provide a solid basis for good reconstructions. It’ll never be 100% accurate of course, but since we don’t know, getting anything over 50% likelihood is good enough. Even here there are issues of course, you could for example argue whether way that ceratosaurs did or did not have protofeathers since it’s not as certain as it was a few years ago so whatever you pick you might get it wrong (when more evidence comes through) and you’ll likely be criticised by one side or the other. When it comes to colours there are (aside from a few notable exceptions now of course) no animals of known colours in the Mesozoic archosaur fossil record. However, that doesn’t mean certain inferences can’t be made – large animals tend to be more mute than small ones, crests and ornaments tend to be brightly coloured, animals living in forests tend to be dappled or striped, those on plains tend to be more uniform, countershading is common, spots and stripes are rarely mingled and so on. Obviously there is still lots of freedom, but paint a large crest a dull green and then add bright pink spots and yellow stripes to the body and your animal will look weird and implausible.
Unless aiming for a vignette you’ll be wanting to put the animal in an environment and with an appropriate behaviour (even if it is just standing still). Much information on the palaeoenvironment should be available in the literature (wet or dry, rivers or forests) and the flora and fauna present.
All of this should get you into a strong position to produce an accurate extinct animal. Of course, you’ll actually have to do the work of getting hold of the papers and photos of the material, reading, checking, measuring and evaluating it all and then actually draw the damned thing. For an artist I suspect that is the easy part, but if you want your reconstruction to be accurate and appropriate then these considerations are, to my mind, essential. Of course with experience comes knowledge and expertise – if you have drawn 100 tyrannosaurs the next one won’t require much preparation, but to get started on a new species or group you’ll need to know the details of its anatomy and biology.