Carnivorous plants

Every carnivorous plant photosynthesises. None of them needs an animal for energy, and a well-lit specimen that never catches anything will survive — it will simply grow slowly and flower poorly. What the animal supplies is nitrogen and phosphorus, the two elements a plant cannot make and which are missing from the ground these species grow in.

That constraint is the whole story. Carnivory only pays where light is abundant, water is abundant, and the soil is exhausted — which describes bogs, wet heath and sandy barrens rather precisely. In fertile soil the trade goes the other way: trap leaves photosynthesise badly, so a carnivorous plant in good ground is out-competed by an ordinary one that spent the same carbon on ordinary leaves.

It also explains a common piece of accidental cruelty. Feeding a potted flytrap fertiliser, or planting it in compost, tends to kill it — not because it dislikes food but because its roots are adapted to ground with nothing in it.

Every trap is a modified leaf, and across eleven origins only a handful of designs have appeared. Most carnivorous plants use the simplest of them.

The bladderworts deserve particular attention, because they are both the largest carnivorous genus and by far the fastest. A bladder pumps water out of itself until its walls are under tension; touching a trigger hair breaks the seal, water rushes in, and the prey is carried with it. The whole event takes under a millisecond, which makes it one of the fastest movements known in any plant or animal.

Overwhelmingly small insects: ants, flies, beetles, springtails and spiders. Ants are a much larger part of the diet than the name "flytrap" suggests, and for many species the crawling prey outnumbers the flying kind considerably.

Vertebrate capture happens and is rare and mostly incidental. Large Nepenthes pitchers occasionally drown a small frog, lizard or rodent; those pitchers are dealing with a windfall rather than hunting. One group of Bornean pitchers has gone a stranger route entirely and stopped catching much at all — they provide a comfortable roost for a small bat, or a feeding perch for tree shrews, and collect the droppings. The nitrogen arrives without any trapping.

Digestion is enzymatic, using proteases and other enzymes secreted into the trap, sometimes with bacterial help. What is absorbed is the dissolved nitrogen and phosphorus; the insect’s cuticle is left behind, which is why an old pitfall trap contains a layer of empty shells.

A flytrap requires two trigger touches within about twenty seconds before it closes, and more touches before it begins to digest. This is regularly reported as counting and memory, and it is worth being exact about what the mechanism is: the first touch produces an electrical signal and a rise in calcium inside the cells, which then decays. A second touch arriving before the decay finishes pushes the total over threshold. Nothing is stored as a number, and the "memory" is a chemical concentration falling.

That is not a deflation of the plant. A threshold with a decaying signal is a genuinely elegant solution to a real problem — a raindrop or a windblown grain should not cost the plant a closure, and a closure that catches nothing is expensive. The mechanism filters false alarms with no nervous system and no decision.

• How much of a carnivorous plant’s nitrogen actually comes from prey?

  Why it matters: Isotope studies give a wide range across species and sites, and the whole cost-benefit argument for carnivory depends on the number.

• Why has carnivory evolved so many times and then so rarely diversified?

  Why it matters: Eleven independent origins is a lot for a strategy that has produced only around 800 species. Something is limiting it.

• How do carnivorous plants avoid trapping their own pollinators?

  Why it matters: Several species hold their flowers well above the traps, which looks like a solution, but how well it works has rarely been measured.