If solids are the well behaved grandparents of the matter world, then plasmas are the insolent children - they are often badly behaved, refuse to sit still, are prone to energetic outbursts and have a tendency to do the opposite of what one wants. Taming the "naughty child" is the subject of thermonuclear-fusion research - a topic that was discussed in detail at the American Physical Society plasma-physics meeting held in Long Beach, California, late last year.

The benefits of thermonuclear fusion are clear - it offers a source of near-limitless and clean energy. The downside is that all the criteria needed to build a fusion reactor are rather difficult to achieve simultaneously. Fusion reactions are not easy to start or to sustain, which is a good thing otherwise all the stars would have burned out long ago, leaving the universe bereft of life. However, a number of potential methods do exist for achieving controlled thermonuclear burning on Earth and they are the subject of a substantial global endeavour.

In the February issue of Physics World, Rob Akers of UKAEA Fusion, Abingdon, UK outlines key advances discussed at the conference in the two main approaches towards achieving thermonuclear conditions in the lab - inertial-confinement fusion and magnetic-confinement fusion.