A short hosepipe may have solved one of the many mysteries of Stonehenge, showing that the iconic monument was once a perfect, complete circle.
Dry weather in summer 2013 at the Wiltshire monument revealed marks of parched grass in an area that had not been watered.
According to a report in the journal Antiquity, the patchmarks represent the position of the missing sarsen stones which once completed the Neolithic circle.
"Despite being one of the most intensively explored prehistoric monuments in western Europe, Stonehenge continues to hold surprises," English Heritage steward Tim Daw and colleagues wrote.
Located in the county of Wiltshire, at the center of England's densest complex of Neolithic and Bronze Age monuments, Stonehenge has been the subject of myth, legend and -- more recently -- scientific research for more than eight centuries.
The monument was likely built in several stages with "the unique lintelled stone circle being erected in the late Neolithic period around 2500 B.C.," according to the English Heritage.
Researchers have long investigated each part of the monument -- the massive central trilithons, the smaller bluestone settings, the sarsen circle capped by lintels, the outer bank and ditch -- debating whether the outer prehistoric stones were once completely round.
Indeed, the possibility that Stonehenge was an intentionally incomplete monument, with the sarsen circle only finished on the north-eastern side, has been suggested since the mid 18th century.
Now, a hosepipe too short to cover the outer part of the circle where no stones still stand, may have provided a definitive answer, revealing what excavations and high-resolution geophysical surveys failed to find.
Spotted by Daw, the patches on the ground -- believed to be "stone holes" -- appeared in the sarsen circle exactly where stones were expected to stand.
Such crop marks are produced when plants grow over features that have been buried in the ground for a long time, even long after they have been removed.
Buried or once-buried structures interfere with plant growth and develop at a different rate to those growing immediately adjacent. In this case, deep stone holes may have changed the earth permeability, affecting the grass growth.
"Plants that may have initially benefited from the easily available rooting and moisture in disturbed ground fail as the soil dries out, whilst the shorter-rooting surrounding plants manage to survive by extracting water directly from the porous chalk," Daw and colleagues wrote.
A survey, using Differential Global Positioning System equipment, was immediately undertaken, but unfortunately it suffered from limited time, given the elusive nature of the vanishing marks.
"Ideally the survey would have differentiated between marks caused by parching –- the majority -– and those caused by lusher growth," Daw and colleagues wrote.
"It would have also have graded the marks into 'definitive', 'probable' and 'possible' categories. This was not possible, and the result must therefore be treated with caution," they added.
English Heritage hopes to corroborate the findings with searching historic aerial photographs of the site and with future observation of the same phenomenon in similar weather conditions.