All of it.
40mph is faster than the terminal rolling velocity on hill more than a mile long averaging 1 in 8 with no turns and assuming a seriously efficient bike and a rider weighing 100kg. Two eleven year olds would have to be in the top fifth percentile of fat biffers to weigh that much put together and their rusty old racer would be several times less efficient.

And given that the cadence at 40mph for a normally-geared childs racing bike would be comically fast, you'd have a job to keep up let alone add any power. So terminal rolling velocity is as fast as they'd ever get.

The story is made up. I don't know why I'm the one who has to spell this out given that the very first person to respond to the thread said it was a blatant lie.
( Dr. Shambolic je suis charlie, Wed 11 May 2011, 14:01, 1 reply)

I see your point, and I agree that it's not possible for a 1 in 8 gradient
But there's no reason to assume a 1 in 8 gradient. A quick google search turns up plenty of streets around the country that are considerably steeper than that - I could walk a few hundred yards out of my front door and find one. I think the steepest you could reasonably expect (without talking about rare outliers) would be about 1 in 4.

Edit: OK - Taking the formula in this post (quoted as coming from DiPrampero et al., 1979) as a start, I put in a few likely values. Using a 10 degree hill (~1 in 4) with a combined bike and rider mass of 80kg, a front surface area of 0.7 and a drag co-efficient of 0.7, the terminal velocity comes out at 38mph. 0.7 is on the high side for a frontal area, so I think it's safe to assume the terminal velocity won't be less than that. The question then becomes whether or not they would have had time to get close to the terminal velocity - since drag varies as a square law I think it's a fair assumption that they could have come within 10-15% even on a relatively short hill. If the hill was steeper than 10 degrees or the bike was heavier than normal then it becomes easier.

So I was wrong to say it would be easy (I thought the terminal velocity would be higher than that) but I maintain that it's not impossible.
( MatJ LOOK, Wed 11 May 2011, 17:40, closed)

So even with a hypothetical mile long 1 in 4 straight hill and assuming the pillion rider has no legs to add to the drag you've still not hit even a theoretical 40mph.
Kids were telling tall tales about being speed trapped on bicycles at least as far back as the 1970s. It was balls then. It's balls now.
( Dr. Shambolic je suis charlie, Wed 11 May 2011, 19:10, closed)

It wouldn't have to be a mile long and they could have tucked down and and and...
OK, I was wrong.
( MatJ LOOK, Wed 11 May 2011, 23:05, closed)

Let's hug until the bad man goes away.

( Dr. Shambolic je suis charlie, Wed 11 May 2011, 23:23, closed)