Twice swallowed

A kept seed whose road changes species mid-journey — swallowed by a small animal, carried inside it, then handed, still inside the prey, to a predator’s gut, and delivered only if it survives digestion twice in a row

R879 closed on the mallard: one gut that is store, mill and road at once, where the keeping succeeds only when digestion fails to finish. Its closing turn asked for the passenger whose carrier is itself carried — a seed inside a small animal, that animal inside a predator, so the road changes species, speed and direction at the moment of a death. That doubled journey has a name in the literature, diploendozoochory, and a long pedigree: it was described in the nineteenth century and then studied only a handful of times before a 2017 review gathered the scattered record together. Two corrections first, because this page insists on them. The predator is not a courier hired by the plant; it is eating the prey, and the seed is cargo neither animal chose to deliver. And the evidence comes in two different strengths, which this page keeps apart: seeds found inside prey in a predator’s keeping prove the transfer can begin, while seeds recovered from a predator’s pellets or dung and then germinated prove the doubled passage can actually be survived. Only the second kind closes the loop.

The interrupted transfer, caught in the act. On three small islands north-east of the Canaries, Eleonora’s falcons breed late on purpose: their chicks hatch when the autumn migration is passing overhead, and the adults hunt small birds blown toward the islands on their way from Europe to Africa, storing prey around the nest. Researchers examining the stomachs and intestines of those stored prey — 408 birds of 21 species — found seeds still inside five of them: two pied flycatchers and redstarts carrying seeds of fleshy-fruited shrubs, and a common quail carrying seeds of up to three different plants, 45 seeds in all. The plants those seeds came from mostly grow more than a hundred or two hundred kilometres away, and one of them does not grow in the Canaries at all. So the first gut had already done long-distance work when the falcon struck. But notice what the falcon’s intervention does to the geometry: the small bird’s journey south, at migrant speed on migrant bearings, ends at the island; whatever second journey the seed gets now runs on the falcon’s map instead — a shorter, local circuit between hunting sky and nest cliff — or ends in the chick. The transfer is real and documented; the onward dispersal in this case is the part left unproven, and the honest reading is that predation here mostly truncates a long journey and starts a short one, on ground so arid that the researchers themselves rated colonisation there as unlikely.

The completed passage, counted in pellets. The loop does close elsewhere, and it closes in the least glamorous archive available: raptor pellets. On Fuerteventura, common buzzards and Eurasian kestrels eat two introduced mammals, European rabbits and Barbary ground squirrels, both of which graze weed seeds. Dissecting a few hundred pellets from each raptor turned up on the order of eleven thousand weed seeds, spanning dozens of plant species — 39 in one raptor’s pellets, 62 in the other’s. These are seeds that had already passed through a mammal’s gut, or were sitting in one when the mammal was taken, and then passed through a bird of prey’s gut as well, to be regurgitated in a pellet. When researchers planted them — after an interval between collection and experiment that can only have cost viability — almost one in ten still germinated. A smaller, older North American case gives the same shape at a harsher rate: spotted knapweed seeds eaten incidentally by deer mice, the mice eaten by great horned owls, and about one percent of the seeds recovered from the owls’ pellets germinating. One percent and ten percent are not abundance. They are a toll booth’s receipts: proof that the doubled road is passable, priced in the seeds that did not make it.

Why two guts can be more than twice one. The review that gathered this record points at two asymmetries that keep the doubled journey from being a pure loss. First, scale: predators are generally larger than their prey, range further, and cross barriers — roads, clearings, open water, worked land — that stop the smaller carrier, so the second leg can reach ground the first leg never could, including ground a plant needs to recolonise after disturbance. Second, wear: a thick seed coat that survives one gut barely scratched may be worn by the second passage into germinating more readily than it would have after the prey alone — the same grinding that destroys most seeds finishes scarifying a few. Neither asymmetry rescues the arithmetic. Two digestions multiply two survival fractions, each already well under one, so the doubled road is always a rare road. Its importance is not in the number of seeds it moves but in the seeds it moves uniquely: the ones delivered across a barrier, into an emptied habitat, by the only animal in the system built to cross it. The mallard’s gut was one flesh doing everything. Here the road is a relay neither runner entered, and the baton pass is a killing.

One seed, two keepers, one killing between them. Schematic, compressed from the published record.

Walk it in order.

Schematic only: the four stages compress three separate records — seeds found in falcon-stored prey on Canary islands, weed seeds germinated from buzzard and kestrel pellets on Fuerteventura, and knapweed seeds germinated from great horned owl pellets after passage through deer mice — into a single seed’s journey, and no one study followed a seed through both guts end to end. The falcon record in particular shows seeds inside prey, not seeds surviving the falcon; treating it as completed double dispersal would overstate it, and this page does not. The germination figures are not interchangeable: almost ten percent on Fuerteventura despite a delay before planting, about one percent in the owl–mouse case, each for different plants, prey and predators. The claim this page rests on is narrower and directly documented: seeds are routinely found inside prey animals taken by predators; some seeds recovered from predators’ pellets after two digestive passages remain viable and germinate; and predators’ larger ranges mean that second passage can, in principle, deliver seed across barriers and into ground the first carrier could not reach.