Carnivorous Plant - Evolution

Evolution

The evolution of carnivorous plants is obscured by the paucity of their fossil record. Very few fossils have been found, and then usually only as seed or pollen. Carnivorous plants are generally herbs, and their traps primary growth. They generally do not form readily fossilisable structures such as thick bark or wood. The traps themselves would probably not be preserved in any case.

Still, much can be deduced from the structure of current traps. Pitfall traps are quite clearly derived from rolled leaves. The vascular tissues of Sarracenia is a case in point. The keel along the front of the trap contains a mixture of leftward- and rightward-facing vascular bundles, as would be predicted from the fusion of the edges of an adaxial (stem-facing) leaf surface. Flypapers also show a simple evolutionary gradient from sticky, non-carnivorous leaves, through passive flypapers to active forms. Molecular data show the Dionaea–Aldrovanda clade is closely related to Drosera, but the traps are so dissimilar that the theory of their origin—very fast-moving flypapers became less reliant on glue—remains rather speculative.

There are over a quarter of a million species of flowering plants. Of these, only around 630 are known to be carnivorous. True carnivory has probably evolved independently at least six times; however, some of these "independent" groups probably descended from a recent common ancestor with a predisposition to carnivory. Some groups (the Ericales and Caryophyllales) seem particularly fertile ground for carnivorous preadaptation, although in the former case, this may be more to do with the ecology of the group than its morphology, as most of the members of this group grow in low-nutrient habitats such as heath and bog.

It has been suggested that all trap types are modifications of a similar basic structure—the hairy leaf. Hairy (or more specifically, stalked-glandular) leaves can catch and retain drops of rainwater, especially if shield-shaped or peltate, thus promoting bacteria growth. Insects land on the leaf, become mired by the surface tension of the water, and suffocate. Bacteria jumpstart decay, releasing from the corpse nutrients that the plant can absorb through its leaves. This foliar feeding can be observed in most non-carnivorous plants. Plants that were better at retaining insects or water therefore had a selective advantage. Rainwater can be retained by cupping the leaf, leading to pitfall traps. Alternatively, insects can be retained by making the leaf stickier by the production of mucilage, leading to flypaper traps.

The pitfall traps may have evolved simply by selection pressure for the production of more deeply cupped leaves, followed by "zipping up" of the margins and subsequent loss of most of the hairs, except at the bottom, where they help retain prey.

The lobster-pot traps of Genlisea are difficult to interpret. They may have developed from bifurcated pitchers that later specialised on ground-dwelling prey; or, perhaps, the prey-guiding protrusions of bladder traps became more substantial than the net-like funnel found in most aquatic bladderworts. Whatever their origin, the helical shape of the lobster pot is an adaptation that displays as much trapping surface as possible in all directions when buried in moss.

The traps of the bladderworts may have derived from pitchers that specialised in aquatic prey when flooded, like Sarracenia psittacina does today. Escaping prey in terrestrial pitchers have to climb or fly out of a trap, and both of these can be prevented by wax, gravity and narrow tubes. However, a flooded trap can be swum out of, so in Utricularia, a one-way lid may have developed to form the door of a proto-bladder. Later, this may have become active by the evolution of a partial vacuum inside the bladder, tripped by prey brushing against trigger hairs on the door of the bladder.

Flypaper traps include the various true flypapers and the snap traps of Aldrovanda and Dionaea. The production of sticky mucilage is found in many non-carnivorous genera, and the passive glue traps in Byblis and Drosophyllum could easily have evolved.

The active glue traps use rapid plant movements to trap their prey. Rapid plant movement can result from actual growth, or from rapid changes in cell turgor, which allow cells to expand or contract by quickly altering their water content. Slow-moving flypapers like Pinguicula exploit growth, but the Venus flytrap uses such rapid turgor changes that glue became unnecessary. The stalked glands that once made it and which are so evident in Drosera have become the teeth and trigger hairs—an example of natural selection hijacking preexisting structures for new functions.

Recent taxonomic analysis of the relationships within the Caryophyllales indicate that the Droseraceae, Triphyophyllum, Nepenthaceae and Drosophyllum, while closely related, are embedded within a larger clade that includes non-carnivorous groups such as the tamarisks, Ancistrocladaceae, Polygonaceae and Plumbaginaceae. Interestingly, the tamarisks possess specialised salt-excreting glands on their leaves, as do several of the Plumbaginaceae (such as the sea lavender, Limonium), which may have been co-opted for the excretion of other chemicals, such as proteases and mucilage. Some of the Plumbaginaceae (e.g. Ceratostigma) also have stalked, vascularised glands that secrete mucilage on their calyces and aid in seed dispersal and possibly in protecting the flowers from crawling parasitic insects. These are probably homologous with the tentacles of the carnivorous genera. Perhaps carnivory evolved from a protective function, rather than a nutritional one. The balsams (such as Impatiens), which are closely related to the Sarraceniaceae and Roridula, similarly possess stalked glands.

The only traps that are unlikely to have descended from a hairy leaf or sepal are the carnivorous bromeliads (Brocchinia and Catopsis). These plants use the urn—a fundamental part of a bromeliad—for a new purpose and build on it by the production of wax and the other paraphernalia of carnivory.