From the speed of light to the charge on an electron, the fundamental constants of physics had been assumed to be immutable. But that comfortable assumption is being challenged.

The latest in a series of experiments to question this view suggests that over the past 12 billion years, the ratio of the mass of a proton to that of an electron may have decreased. The result has left physicists curious, sceptical, and more than a little stumped.

Protons are about 1,836 times heavier than electrons. The exact mass ratio can be calculated by observing how a cloud of hydrogen molecules (each composed of two protons and two electrons) absorbs ultraviolet laser light. Wim Ubachs of the Free University in Amsterdam, the Netherlands, and his team have done just that, producing data hundreds of times more accurate than those obtained from previous experiments (see Nature doi:10.1038/news060417-7; 2006).

Using the Very Large Telescope in Paranal, Chile, they compared their results obtained with hydrogen in the lab with observations of light from two distant quasars. This light shines through clouds of hydrogen around 12 billion light years away. The lab result was smaller by 0.002% (E. Reinhold et al. Phys. Rev. Lett. 96, 151101; 2006).

A change of 20 parts per million over 12 billion years isn't large — “not jelly”, as Andy Fabian of the University of Cambridge, UK, puts it. But it could point to previously unknown subtleties in the way the Universe is put together. Such an effect is not explained by anything in physicists' standard model of particle physics.

The result has a confidence level of about 3.5-sigma, a statistical term that translates into a 0.3% possibility that it could be down to chance. That's good enough to be called an “indication”, says Ubachs, but for such an important potential result it is not a cast-iron observation.

They've done the best job of anyone so far on comparing the proton and electron mass.

“You don't want to book a ticket to Stockholm on a 3.5-sigma result,” chuckles John Webb, a physicist at the University of New South Wales in Sydney, Australia, who has also studied changes in the proton–electron mass ratio. “But they've done the best job of anyone so far on comparing the proton and electron mass.”

The mass-ratio effect has until now received less attention than the fine-structure constant, α, a measure of the electromagnetic force that keeps electrons in place inside atoms and molecules. Webb has been at the forefront of efforts to probe whether or not the mass ratio changes over time (J. K. Webb et al. Phys. Rev. Lett. 87, 091301; 2001). He expects to publish his most detailed study later this year, which relies on light from many more quasars than used in previous analyses.

Weird stuff: observations of matter billions of light years away suggest the proton–electron mass ratio has changed over eons. Credit: NASA

But the work is controversial. Webb says he has been criticized by senior astrophysicists for even tackling the problem. “It's as though you're knocking a pillar of physics,” he says.

John Barrow, a cosmologist at the University of Cambridge, adds that astronomers are often more sceptical than physicists simply because they are more aware of how complicated — and potentially error-prone — quasar spectroscopy can be. Fabian is certainly cautious about Ubachs' result. “Extraordinary claims require extraordinary evidence,” he says, pointing out that many results at a similar confidence level turn out to be wrong. The most likely error source lies in assumptions about the behaviour of the distant hydrogen cloud, he says. Some parts of the cloud could be hotter or moving faster than other parts, and the hydrogen might be mixed with a smattering of other elements.

Even if further studies do push the confidence level across the five-sigma threshold that physicists regard as convincing, the reason for the changed mass ratio is not understood, nor whether it is an ongoing effect. If true “the laws of physics as we currently understand them are incorrect at their very core,” says Michael Murphy of the University of Cambridge, who works with Webb. “A new set of physics laws must be found which explain the new observations.”

It is unlikely that protons are simply losing weight. But various versions of string theory suggest that extra dimensions occupied by a particle might affect properties such as its mass. Subtle changes in these dimensions could make physical constants vary slightly, explains Barrow. However, “there's absolutely no observational evidence to support this vast array of ideas,” cautions Fabian. The paucity of hard evidence for string theory may be partly responsible for the upsurge in interest in variable constants, Barrow adds; results like Ubachs' could eventually provide a good way to assess the ideas. “I'm sure we'll see some theory papers about this,” he says. “I might write one myself.”

Fabian agrees that the problem has been receiving more attention over the past few years, but that “it's still a minority interest”. The research needs intensive work on the very biggest telescopes, “a large investment in something that could turn out to be zero”. But he agrees that it is an important problem to tackle: “Let's keep shaking the pillars to make sure they're rigid.