Astronomers have used a new ground-based technique to study the atmosphere of a planet outside our Solar System.
The work could assist the search for Earth-like planets with traces of organic, or carbon-rich, molecules.
Gases have previously been discerned on exoplanets before, but only by using space-based telescopes.
Astronomers reporting in Nature say their method of spotting methane gas on exoplanets could be extended to many other, ground-based telescopes.
Methane was first spotted on an exoplanet named HD 189733b in 2008 by a group led by Mark Swain of Nasa's Jet Propulsion Laboratory in the US.
It is a "hot Jupiter" or gas giant that orbits very close to its parent star, which lies about 63 light-years away from Earth.
It marked the first time that an organic molecule had been detected on an exoplanet. The known planets residing outside our Solar System currently number more than 400.
Dr Swain and his colleagues have now shown that by looking at a different set of light wavelengths, methane and possibly other components can be catalogued using relatively small, Earth-bound telescopes.
They used Nasa's Infrared Telescope Facility in Hawaii to perform measurements of the light emitted by HD 189733b, using a version of the so-called transit method that measures an exoplanet's "secondary eclipse".
At its heart, the approach takes the light received on Earth when HD 189733b is behind its parent star and subtracts it from the light received when it is between its star and the Earth.
What results is the light due solely to the planet. However, the effects of the Earth's atmosphere, with its own atmospheric gases and passing clouds, would typically tend to overwhelm the signal from the distant star.
To overcome this obstacle, Dr Swain and his colleagues decided to look in the infrared part of the light spectrum - in a region that is not currently covered by space-based telescopes - and devised a method to get rid of the effects of Earth's atmosphere.
Their approach made simple assumptions about how errors in the light detected in the telescope are related to each other in terms of the wavelength or colour that is detected, or in terms of the time at which the detection is made. By making these correlations and correcting the signal over and over again, the overall error is whittled down.
With Earth's swirling atmosphere effectively subtracted, the team discovered a peak in their data that corresponded to methane emitting light in a process known as fluorescence.
"Up until this point we've not been able to use ground-based telescopes to detect molecules in an exoplanet's atmosphere," Dr Swain told BBC News.
"These molecules are probes of the conditions and chemistry, and since these planets are too far away for us to send a probe, they are eventually how we're going to answer the question of whether expolanets have a habitable atmosphere to support life."
However, the methane appears to contradict an assumption made about both stars and exoplanet atmospheres before - the existence of a so-called local thermodynamic equilibrium, or LTE. Something is putting more energy into the methane than it can quickly get rid of.
"We don't know what that process is in this case," Dr Swain said. "In our own Solar System, charged particles can cause this fluorescence; the other possibility is some sort of [light in the form of] photons."
Emission of light stimulated by charged particles bumping into atmospheric gases is exactly the mechanism behind the aurora displays seen at visible wavelengths at the Earth's poles.
"If we could show it was charged-particle pumping, you could put constraints on the planet's magnetic field - no-one's been able to do that for an exoplanet before."
In any case, the methane emission is comparatively strong from HD 189733b, leading to two important conclusions. Firstly, other exoplanets may well be experiencing this same process, and detecting any methane or possibly other atmospheric gases would be made easier.
"It's a pretty interesting discovery," said Keith Horne, an astrophysicist and exoplanet expert from the University of St Andrews.
"The main impact is this strong emission line that stands out quite dramatically," he told BBC News.
"You'd be able to detect it on other objects that are farther away [from their parent stars] or are fainter. So far, it's been just the nearest, transiting 'hot Jupiters' that are bright enough to detect this secondary eclipse."
But more than that, the new research shows that some observations that were once only possible from space can now be done using ground-based telescopes.
That vastly increases the number of instruments - far larger than the 3m telescope used in the Nature work - that could be trained on exoplanet atmospheres.
"Larger telescopes could look at this in more detail, because there's so many of them. It potentially allows many different teams to participate; previous detections with the Hubble and Spitzer telescopes are great, but there's only one Hubble and only one Spitzer."
Both Dr Swain and Professor Horne agree that more detailed work is required to be certain that the peak is due to methane, and to establish how the methane violates the LTE assumption.
However, with the new results in hand, further observations in the new spectral region will be easy to justify, at least until the James Webb Space Telescope - the first space-based telescope that can "see" in this part of the infrared - takes to the skies in 2014.
"It's a hard measurement to do, and they seem to have succeeded," Professor Horne said. "The committees that decide how big telescope time is spent will be able to see it's a worthwhile measurement."