'It was recently discovered that Pluto has two moons beyond the one already known (Charon, discovered in 1978). This was very cool news, but it begs the questions: where did they come from?
Quite a few observations have been made, enough to compile some interesting data: the moons are small, they orbit Pluto farther out than Charon does, and their orbits are fairly circular.
There are several mechanisms that can create a moon. One is to have it form along with the primary (the planet). This seems unlikely for Pluto, given that it's way far out in the solar system, where one assumes matter in the early proto-solar system disk was spread pretty thinly. It's hard to see how you can get more than one object at a time in one place from those meager pickings.
Another way is to capture objects that are in orbit around the Sun. Turns out that's tough too. Most of the time, an object orbiting the Sun won't get captured by a planet: it'll swing on by, leaving the planet with as much velocity as it had coming in. If you already have a biggish moon, then capturing an object is easier, though still pretty tough. And it leaves the question of how the big moon got there in the first place*.
Looking at our own Moon may give you an idea for a third way. The Moon is covered with craters, which are from impacts with other, smaller bodies like an asteroid. When it hits, rock and rubble get ejected, raining down on the surrounding area. If the impact is big enough (and the gravity of the impacted body small enough), then the ejecta can be launched into orbit.
If Charon formed from some impact event early in Pluto's history, then there would have been a lot stuff ejected with it. It's possible the two small moons were ejected bodily from Pluto at the same time, or were formed from smaller pieces which merged. They would have formed close in to Pluto and Charon at first, and then over billions of years moved outward as tides from Pluto and Charon gave them more energy.
An interesting side-effect of this idea is that there may be lots of multiple systems out there in the cold dark beyond Pluto. A while ago, it was crazy to think that these kinds of objects might have moons, but we know collisions happen a lot. If moonlets can form from these collisions, then we expect to see lots of multiples out there. We also would expect that they would be heavily cratered from all the impacts (unless something smooths out the surface, like melting from the heat of impact). Since we're sending a probe to Pluto -- the launch window opens on January 17th -- these predictions can be tested. In a few years we may have a much better idea of how these systems form. That's pretty amazing, given that just a few years ago we didn't even know they existed!
For those with some technical background, a science journal paper has been published which describes all this too.
* I should add that this capture idea works better very early in the solar system. Pluto could have captured something smaller because the proto-solar system disk itself can help slow down the incoming object enough. It's complicated physics, with lots of intermediate steps, but you can think of it as the stuff in the disk dragging on the incoming body, draining away some of its energy of motion, allowing Pluto to hold onto it.'