Perhaps the most eagerly sought measurement in modern science is the Hubble constant – the rate at which the Universe is expanding. An accurate value for the Hubble constant would be a guide to understanding the history and fate of the Universe in which we live.

Over the past few years, estimates for the Hubble constant have varied between 60 and 75 kilometres per second per megaparsec (where a megaparsec is a unit of distance equivalent to roughly three million light years). A high value for the Hubble constant means a young, rapidly expanding Universe; a low value means an old, slowly expanding one. The current estimates for the Hubble constant translate into a Universe between 16 and 13 billion years old. For comparison, the Earth is around 4.5 billion years old.

Measuring the Hubble constant means having to find the absolute distance to an object – typically a distant galaxy – whose rate of recession can also be measured. Measuring the speed at which objects appear to recede (as the Universe expands) is relatively easy; gauging an absolute distance is, by comparison, very difficult.

A sure-fire method that has been used since 1929 is based on the pulsations of giant stars called 'Cepheid variables'. Cepheids are useful because their rate of pulsation is closely linked to their absolute brightness. The difference between their absolute brightness and their brightness as seen from Earth can be used to measure their distance from us – in the same way that a faint streetlamp can be judged to be more distant than one that appears brighter, even if both have the same intrinsic or 'absolute' brightness.

In 1991, the Hubble Space Telescope's Key Project set out to refine measurements of the Hubble constant. So far, it has studied more than 800 Cepheids in 18 galaxies, and the results suggest a value of 68 kilometres per second per megaparsec for the Hubble constant. When Eyal Maoz of the NASA Ames Research Center in Moffett Field, California and the University of California, Berkeley and colleagues used the Key Project's methods to estimate the distance to a galaxy called NGC4258, they came up with a distance of around 8.1 megaparsecs: they report their study in the 23 September issue of Nature.

So far, so good – as more Cepheids in more galaxies are studied, the estimates of the Hubble constant just get better and better, right?

Wrong – the new Cepheid-based distance estimate to NGC4258 is significantly different from a new and remarkably reliable measure of the absolute distance to the same galaxy. The discrepancy may lead to further revisions for the age of the Universe, and make people think twice about the tried-and-trusted Cepheid method.

In a report in Nature last month, Jim Herrnstein of the National Radio Astronomy Observatory, Socorro, New Mexico and colleagues estimated the distance to NGC4258 from observations of the behaviour of masers (the infrared equivalent of lasers) emitted by sources orbiting the galaxy's central black hole. The estimate, around 7.2 megaparsecs, remains the most accurate and reliable measurement of the distance to any other extragalactic object – but it is 12% out from the Cepheid-based measurement to the very same galaxy.

What's gone wrong? It could be that Cepheid-based measures consistently overestimate distances, and the Universe is younger than we thought. The value for the Hubble constant might need to be revised upwards, bringing the age of the Universe down to a sprightly 12 billion years.

But things may not be as bad as they look, counsels Bohdan Paczynski of Princeton University, writing about the work in Nature. Paczynski points out that the distance scale used in the Hubble Key Project relies on the assumption that a nearby galaxy called the Large Magellanic Cloud (LMC) is 50 kiloparsecs (thousands of parsecs) away from us, whereas more recent evidence suggests that it is much closer, at around 44.5 kiloparsecs. Plugging this figure into the calculations shortens Cepheid estimates, bringing Maoz and colleagues' figures for NGC4258 into line with those from Herrnstein and colleagues.

The new maser-based method will allow a sharper, more accurate picture of the Universe to emerge – provided that we can cope with a Universe that appears so young.