Marathon runners tend to be tough and wiry while sprinters carry more muscle to power them to the finishing line. One build doesn't fit all sports,but this `evident truth' has proved elusive. When Robbie Wilson and his colleagues in Antwerp began looking at frog performances, they also failed to find that faster frogs lacked endurance, until they looked deeper into the animals at their leg muscles. When Wilson and Rob James tested the muscle's peak force versus stamina, the elusive relationship was there(p. 1145).

At first, Wilson had set off down the same track as everyone else; he'd looked at the whole animal's performance. He set the frogs two tasks. First he startled them with a tap on the bum and watched the frog's explosive escape kick (almost 7 per second!). Then he set them an endurance test, where each frog had to swim to the point of exhaustion against a slowly flowing current. Some frogs only managed a few minutes against the flow, but a few kept on swimming for more then half an hour! If the correlation was there, Wilson expected the champion sprinters to fail first at the endurance test.

However, even after analysing the fastest sprint and the longest swim, the combined performances of over 50 frogs failed to produce the expected correlation. Wilson had drawn the same blank as everyone else. What was going on?

Wilson wondered if the difference was more then skin deep. But this meant looking directly at the muscle's performance during the speed versusendurance tasks. Wilson shipped his frogs to Coventry where his colleague Rob James had a force-measuring muscle rig, and began testing the muscle's speed and endurance.

Getting a sprint response out of a single muscle isn't as simple as administering a tap on the bum! James' muscle rig stretched and released the muscle in the same way that the muscle changes shape while the frog is swimming. At the same time the rig electrically stimulating the muscle for the brief period when the shortening muscle would have been contracting during the kick. They simulated both types of activity: the escape response kick by stimulating the muscle at 6Hz, and the endurance swim with a leisurely 2Hz stimulation. During both tests, they measured the forces generated by each contraction to compare each muscle's peak force with its `sticking power' to see if the muscle traded off speed against endurance.

To Wilson's surprise, the correlation was clearly there. But if it was there all along, why hadn't he found it when he was looking at the frog's athletic performances? Wilson explains that there are probably too many other aspects to an animal's performance that have masked the result until now. Wilson puts this in a human context. He says that if you take a champion sprinter and compare his performance with a weekend athlete, the speed versus endurance trade-off is masked. This is because the elite athlete has a better overall performance in both events than the weekend runner, which conceals the endurance trade-off that paid for the sprinter's world-class performance.