Last time I promised you a ring of islands with nothing in the middle, tracing the outline of a mountain that sank out from under it — a shape that exists only because the thing holding it up has been falling away the entire time.
In 1835 Charles Darwin stood on a height in Tahiti and looked across the strait at Moorea: a mountain wrapped in a ring of white surf. What he later wrote about it reads like an instruction. Remove the central land, he said, and what is left is an atoll. The ring of breakers is the same ring either way. Only the mountain is optional.
The mountain is the part that leaves
Reef coral lives only in warm, lit, shallow water. Darwin’s answer to a reef in mid-ocean, over water miles deep, was that the foundation went down and the reef went up. A volcano rises out of the sea and coral fringes its shore. The volcano begins to sink, and the coral, which can only live at the top, keeps building on its own dead to stay there. The fringing reef becomes a barrier reef; the channel inside it widens as the mountain shrinks. Then the last peak goes under, and the ring is all that is left: a coral wall standing on a kilometre of its own skeleton, drawing the outline of something that has not been there for a very long time.
The lagoon is the hole where the mountain was. Darwin had noticed that the solid reef grows only on its outer edge, the side the breakers hit; so as the ground fell away, the rim kept pace and the interior did not.
Nobody could dig deep enough
Darwin, a year before he died, hoped some millionaire would bore an atoll and bring up cores from five or six hundred feet. The Royal Society tried at Funafuti in the 1890s, reached 1,114 feet when the last of the drill’s diamonds ran out, and was still in reef rock.
The bottom was found in 1952, on Enewetak in the Marshall Islands, by the nuclear test programme’s drillers. One hole went to 4,630 feet and struck hard basement. Another brought up olivine basalt between 4,208 and 4,222 feet, with shallow-water limestone directly on top of it, and that limestone was Eocene: reef had been stacking up on that sinking mountain for tens of millions of years, its top mostly within reach of the light. Darwin was right.
The same cores said something else
The Enewetak cores also held leached zones — bands where the coral’s mineral had been dissolved out or recrystallised — which the drillers read as periods when the top of the atoll had stood above sea level, in the rain. Because sea level is not still. Darwin could not have known: the ice ages were being argued into existence as he wrote, and nobody yet understood that a glaciation pulls the ocean down by more than a hundred metres and gives it back.
So one modern reading, argued outright in a 2021 review, separates the two answers his drill was supposed to give at once. The kilometre of limestone is subsidence, and it is his. But the ring on top — the rim, the lagoon, the hole in the middle — is much younger. Each time the sea fell, the atoll was a flat limestone plateau standing in the rain, and rainwater dissolves the middle of a limestone plateau faster than its edges. Each time the sea came back, coral reoccupied the raised rim and grew straight up to meet it, and the dissolved middle flooded into a lagoon. The review’s authors put it bluntly: the volcano has had no influence on what the modern atoll looks like. How much is dissolution and how much is the rim simply outgrowing the interior is still argued. Either way, the shape we see is an ice-age form with coral on the lip.
Only the rim kept pace, exactly as Darwin said. Not for his reason.
The race has a finish line
The Hawaiian islands ride the Pacific plate north-west off the hotspot that made them, sinking and cooling as they go: a live volcano at one end, then eroded high islands, then near-atolls with a last pinnacle in the lagoon, then true rings. Kure, the northernmost atoll in the world, sits at about the latitude a 1982 paper named the Darwin Point, where coral growth slowed by cooler water is just matched by the rate of subsidence. Beyond it the chain continues as drowned platforms. Those are the rings that lost.
There is no correcting loop in this, nothing like No. 88’s marsh, which climbs back by flooding deeper. A reef grows as fast as light and warmth allow and the sea surface planes it off — the marsh has a governor, the reef has a lid — and past the point where deeper means darker and darker means slower, the losers are permanent. The sea we are now raising is the same race, run again.
A ring in the open sea, drawn around nothing, because for all that time it was the only part fast enough to stay.
≈
One loop I’m watching
Next: a forest so large it makes a good part of its own weather — rain that falls, is lifted back into the air by the trees, and falls again downwind, over and over, before it ever reaches the mountains; and the argument over whether the trees are merely recycling the rain or pulling it in from the sea. Next time.