The first ridge of sand behind a beach is held up by a grass that has to be buried alive to keep going. Not because sand feeds it. Because the ground it grew in last year has turned against it, and climbing is the only way out.
Marram — Ammophila arenaria in Europe, American beachgrass on the other side of the Atlantic — is the plant that makes coastal dunes exist. Stiff grey-green blades rolled into tubes against the wind, a rhizome network running through loose sand, and a habit almost nothing else has: bury it, and it comes back up. Sand swamps a shoot, the plant pushes a new vertical stem through the fresh layer, throws out roots into it, and stands there waiting to be buried again. Each burial is knitted in place. The ridge climbs for as long as the wind keeps paying it in sand.
That is the same bargain No. 96’s grazing lawn struck, taken from the plant’s side instead of the animal’s. It is not the bargain No. 88’s salt marsh makes, and the difference is worth having early. A marsh and a foredune are both racing upward inside a frame that will not hold still, and both are paid in sediment. But a marsh is killed by too much of its own currency — sink below the crest of its growth curve and the water that fed it drowns it. A foredune is killed by too little. Nothing on that ridge is at risk of receiving more sand than it can use. Its whole danger is being left alone.
Which is strange, and was a genuine puzzle for most of a century. Plant marram in a sheltered dune where the sand has stopped moving — good ground, by any obvious measure, no wind-blasting, no burial, no salt — and within a few years it thins, yellows and goes. Botanists gave the mystery a name. In 1965 a Journal of Ecology paper reached for another grass entirely as a model for what it called, simply, the Ammophila problem.
The obvious answers had already failed, which is what makes the story good. Dutch coastal engineers, rebuilding foredunes, kept suffering replant failures: culms planted into sand taken from marram’s own root zone grew poorly and died in numbers — and, in the words of the study that finally chased it down, they did so even when fertilizers were applied and competing plant species were absent. So it was not hunger, and it was not neighbours.
In 1988, in Oecologia, Wim van der Putten and two colleagues at a dune research station on the Dutch coast ran the experiment that settles it. They grew marram seedlings in two sands: sand from the rhizosphere of an established marram stand, and sand dredged from the sea floor. Growth in the rhizosphere sand was significantly reduced. Roots came up “brown and poorly developed.” Then they took the same bad sand and sterilised it with gamma irradiation — changing nothing about its chemistry that mattered, only killing what lived in it — and biomass production went significantly up.
Whatever was wrong with the ground was alive. Root-feeding nematodes and soil fungi accumulate around marram roots, specialising on this one host, until the sand a plant has already occupied becomes a place it can no longer grow. The paper’s conclusion uses a word worth stopping on: the decline is caused mainly by self-intolerance.
So turn the whole thing over. Fresh windblown sand is not food. It is distance. Burial gives the plant a few centimetres of ground that nothing has learned to attack yet, and roots pushed up into it are, for a season, in enemy-free space. Marram is not a plant that likes being buried. It is a plant running from its own past, upward, because upward is the only direction the wind offers. The dune is not what the grass built. It is how far the grass has had to climb.
Two honest limits. First, “grows faster when buried” is looser than it sounds: working on the American species, Maun and Lapierre found little change in height or above-ground biomass under burial. The uncontested half is the other half — that it fails without sand. Burial is a reprieve more than a stimulant. Second, the escape is not clean. Marram was brought to California in 1896 and now dominates that coast, and when the enemy-release idea was tested there, the invaded-range soil still bit: sterilising it still helped. The plant had shed its nematodes and not its problem.
People have been managing this for six hundred years without knowing any of it. Culms were being planted in bundles on the Dutch island of Voorne by 1423; British records of planting dune grasses run to the fourteenth or fifteenth century. Every one of those crews was running a scheme to keep a plant permanently unsettled.
And it does not start with a good plant. It starts with enough of them. This March, a Groningen group reported a decade of measurements on a young dune system and found that height was set less by how big each grass patch was than by how many neighbours it had at the beginning — an S-shaped curve, an abrupt switch from isolated tufts trapping sand to clusters acting as one. Below the threshold, sand blows through. Above it, a coastline stands up.
≈
One loop I’m watching
Next: a gap in a coastline that stays a gap. Where this issue’s ridge holds because sand keeps arriving and piling, a tidal inlet holds because water keeps leaving and scouring — the sea pushing sand into the throat with every current along the shore, the tide dragging it back out twice a day, and an opening that survives only while those two rates argue to a draw. Too little tide behind it and the channel silts shut in a season; enough, and it widens until the arithmetic balances. Storms cut new ones overnight. The same coast, the same sand, and a shape made of removal instead of deposit. Next time.