As an engineer, I learnt that you could, in principle, have a flying machine which was unstable and was stabilised by having - if you like - sense organs or sensors which told it what was happening, and a very quick built-in response to those sensors. And I didn't think any more about it until years later, when I was an undergraduate at University College, studying zoology, I realised that flying animals are probably unstable. I knew that... it was clear that birds are, it's rather obvious. In the moulting season, you'll occasionally see a sparrow with no tail, and it flies fine. No aeroplane would fly without a tail. Then, there were much more elegant experiments by a guy called Pringle, who later become Professor at Zoology at Oxford, who showed that two winged flies, Diptera, are unstable, but... in a static sense. And yet they have these little sense organs, the halteres, which are their modified hind wings, which give them information about how they're moving, and they have reflexes to this which enables them to fly. And it's kind of amusing, if you cut the halteres off, the fly can't fly, it just fizzes to the ground. But then, you can make it stable again by giving it a fin. The easiest sort of fin to give it is the tip of a cock's hackle feather, I had them because I used to tie trout flies. And you stick one of these little feathers on the back of the fly and it can fly again. It doesn't fly very well, it zzzzzz, like this, but it can fly. And so I realised that what was true of aeroplanes was true of animals. But interestingly, the most primitive flying animals were... were stable, they didn't have this sensory device, they had great long tails.

[Q] Pterosaurs?

The first Pterosaurs have long tails, Rhamphorhynchus has a great big long tail.

[Q] A great ping-pong bat on the end.

Yes, that's right. What's it called... Archaeopteryx has a great long tail. I don't have any primitive bat with a long tail, but we don't have any.

[Q] There were some late Pterosaurs that didn't have any tail at all, they must have been computationally controlled.

They were computationally controlled, yes. The later ones are fine, but the first insects, the first birds and the first Pterosaurs all had long tails, the one group we don't know about are the bats. They're clearly unstable now, but were they primitively stable? I'd love somebody to find a bat with a long tail.