After a month-long expedition in 1997 to the East Pacific Rise — a ridge in the ocean floor that stretches from California to near Antarctica — as well as several years worth of data crunching, geologist Douglas Toomey at the University of Oregon and an international team came up with some startling results.

At first they assumed they had miscalculated, but after multiple reanalyses to pinpoint the apparent error, the conclusions were still the same. Toomey and co-author William Wilcock of the University of Washington had to consider that the results were legitimate. “Even among the group, there was scepticism initially,” Toomey says. “It took some convincing, but eventually we all got on board.”

Their seismic data suggested that areas of the ridge are volcanically or hydrothermally active not because of a surplus supply of magma pushing its way to the surface (the generally accepted hypothesis), but because magma is coming from the mantle directly underneath active vents (see page 409).

In areas on the ridge where this is not the case, Toomey explains, the Earth's 'plumbing system' redirects the magma underground along the Moho, the boundary between Earth's crust and the upper mantle, which makes volcanic eruptions less likely.