(Eros Hoagland for The New York Times)

The other day, I went to the supermarket to buy a pineapple. I didn’t select the one that smelled the ripest, but the one with the most impressive leaves: tall, bushy and uncrushed by the journey from Costa Rica. When I got it home, I put it in the kitchen sink, turned on the tap, and watched how the water gathered and formed pools in the spaces between the leaves. And I began to imagine that I was not a human in an apartment in London, but a small frog in a tropical forest, climbing up the leaves of a plant like a pineapple, looking for a pool where I could deposit the tadpole I’m carrying on my back.

As I tried to envision the pineapple’s leaves soaring above my tiny frog self, like the shells of the Sydney Opera House, a passage from the wonderful though admittedly specialized book “Mites of Moths and Butterflies” ran through my head:

The magic of the microscope is not that it makes little creatures larger, but that it makes a large one smaller. We are too big for our world. The microscope takes us down from our proud and lonely immensity and makes us, for a time, fellow citizens with the great majority of living things. It lets us share with them the strange and beautiful world where a meter amounts to a mile and yesterday was years ago.

I would love, for a few days, to be shrunk down — perhaps not so small as to be microscopic but, say, about the size of my thumbnail — and splash around in the pineapple pools. I might not come back alive, though: when you’re that small, enemies are everywhere. Even the plants might attack you.

Pineapples are members of a large and diverse group of plants called bromeliads, which are native to Central and South America and the Caribbean (one species is native to west Africa). Some bromeliads live, like regular plants, rooted to the ground. Others are epiphytes: they live on the branches of trees. This isn’t usually a parasitic relationship — they don’t take nutrients from the trees, just support.

But whether they live on the ground or high in the air, the typical bromeliad has, as a pineapple does, a central cone of leaves that collects rain water. At the base of the cone, other leaves open outward; the leaf stems are deep, and rain collects there, too, giving each plant cascading tiers of pools. The biggest bromeliads hold as much as two liters (just over two quarts) of water.

And so it is that the forests of the American tropics are studded with miniature lakes, at a variety of elevations above the forest floor. The density can be prodigious. A study in one forest found 175,000 bromeliads per hectare (2.5 acres); at this density, they may sequester as much as 50,000 liters (more than 13,000 gallons) of water per hectare of land.

Islands and lakes (which are islands of water surrounded by land) are famous for being evolutionary laboratories. Isolation allows the evolution of unique and specialized flora and fauna. Island chains like Hawaii and the Galápagos, and lake systems like the Great Lakes of Africa, are famous for the unusual species that have evolved there. The leaf pools of the bromeliads are islands made tiny.

Like more familiar, large-scale lakes, the number of species in a bromeliad is related to its size. But despite being around 20 quadrillion times smaller in volume than Lake Tanganyika, a bromeliad leaf pool can be home to quite a menagerie.

One study of 209 plants from the lowlands of Ecuador found 11,219 animals from more than 300 species. Of these, some were tourists — just passing through. But many of the others are found only on bromeliads. There are bacteria and flatworms, mosquito larvae, jumping spiders and tadpoles, as well as an array of organisms entirely alien to most of us. Among the most charismatic aliens are ostracods, which have minute hinged shells (perhaps just one millimeter — 1/25th of an inch) that resemble clam shells; but unlike clams, ostracods have two pairs of legs and two pairs of antennae. And just as the finches on different islands of the Galápagos have evolved into closely related, but distinct, species, so have bromeliad ostracods.

There are tiny salamanders, perhaps just 2.5 centimeters (an inch) from the ends of their snouts to the base of their tails: it’s surprising to think that an animal with a backbone can get that small. And though they have simple brains, they are astonishing in other respects. Their tongues, with which they catch insects, can extend almost the length of their bodies, and zoom out with extraordinary speed and power: they can project their tongues 1.5 centimeters (3/5ths of an inch) in 7 milliseconds. Calculations on the ballistics of salamander tongues show that the muscular power of the launch is ten times higher than the instantaneous power output of any other known vertebrate muscle.

And if that’s not wacky enough, how about this: among the inhabitants of bromeliad pools are other species of bromeliad. Sometimes, these bromeliads-within-bromeliads are carnivorous, digesting any insects that get trapped inside. See what I mean about the dangers of being small?

But of all the creatures that make their homes on these plants, here’s the one that particularly captures my imagination: Metopaulias depressus, a reddish-brown crab from the bromeliads of Jamaica.

This is an animal I wouldn’t want to meet if I were tiny. Though they are small — a fully-grown adult has a shell just 2 centimeters (three quarters of an inch) across — they can see off a lizard that intrudes on their plant, and they can kill large centipedes, so they’d make mincemeat of little me.

All the same, I like them because they are unusual in two ways. First: their habitat. Most crabs live in the ocean or, if they’re really adventurous, in burrows they dig on a beach; a few live in (or near) streams and lakes. But M. depressus has evolved to exploit the bromeliad pools, and as far as anyone knows, they do so exclusively.

And here’s the other oddity: in this species, mothers look after their young.

Crabs aren’t famous for paying attention to their offspring. In most species, the female carries her eggs until they are ready to hatch, then releases the larvae into the ocean, where they fend for themselves. The numbers can be enormous — female blue crabs (Callinectes sapidus), for instance, can release two million larvae in one go. (After fertilizing all those eggs, the male needs 15 days to replenish his sperm supplies, poor fellow.)

M. depressus is different. The female lavishes attention on her young. She chooses her plant carefully — she prefers plants with larger volumes of water — and then prepares the pool that will be the nursery. She fishes out any dead leaves that may have fallen in, and drops them onto the ground. (If a sneaky experimenter puts leaves back in, she’ll remove them.) And she drops empty snail shells into the water, often after capturing and feasting on the owners.

These behaviors have two effects. Removing the leaves increases the amount of oxygen in the water; crab larvae need high levels of oxygen in order to breathe. The added snail shells increase the levels of calcium, a mineral without which baby crabs can’t make shells of their own. Unimproved pools can’t sustain baby crabs.

And the hard work doesn’t stop there. For several weeks, the mother feeds her young — perhaps as many as 90 of them (which sounds a lot — but is a lot less than two million) — on cockroaches and millipedes that she catches. And she protects them from being eaten by predators, especially hungry damselfly larvae.

Damselfly larvae generally live in streams, ponds, and lakes; but some have evolved to inhabit bromeliad pools. Among them: Diceratobasis macrogaster. Given a chance, one of these larvae will eat as many as five baby crabs a day. The mother crab does not give them that chance; but an orphaned brood will perish quickly.

Even more unusual, the young crabs don’t disperse immediately, but remain with mom; sometimes you’ll find a couple of generations living together. This is probably because small crabs are more vulnerable to attack as they search for plants of their own, and so it makes sense to grow up before leaving. But whatever the reason, living in family groups is the first evolutionary step towards complicated social arrangements, such as those common among termites and the ants, bees and wasps, but rare for other insects or crustaceans. Perhaps one day, if the evolutionary pressures are right, crabs might join the list of highly social creatures.

Next time you buy a pineapple, therefore, pick one with good leaves, pour on some water, shut your eyes, imagine yourself in a bromeliad pool dodging a mother crab or a salamander’s tongue, and marvel at the richness of island life.

**********

NOTES:

For frogs carrying tadpoles to bromeliad pools, see Weygoldt, P. 1980. “Complex brood care and reproductive behavior in captive poison-arrow frogs, Dendrobates pumilio O. Schmidt.” Behavioral Ecology and Sociobiology 7: 329-332 and Summers, K. 1992. “Mating strategies in two species of dart-poison frogs: a comparative study.” Animal Behaviour 43: 907-919.

The quotation about the wonders of the microscope can be found on p. 141 of Treat, A. E. 1975. “Mites of Moths and Butterflies.” Cornell University Press.

The density of bromeliads, and the volume of water they sequester is cited in Richardson, R. A. 1999. “The bromeliad microcosm and the assessment of faunal diversity in a Neotropical forest.” Biotropica 31: 321-336; this paper also considers bromeliads as islands, and shows that the number of species a bromeliad contains depends on its size. For 11,219 animals in the bromeliads of lowland Ecuador, see Armbruster, P., Hutchinson, R. A., and Cotgreave, P. 2002. “Factors influencing community structure in a South American tank bromeliad fauna.” Oikos 96: 225-234. For a radiation of bromeliad ostracods, see Little, T. J. and Hebert, P. D. N. 1996. “Endemism and ecological islands: the ostracods from Jamaican bromeliads.” Freshwater Biology 36:327-338.

For salamanders living in bromeliads, see Wake, D. B. 1987. “Adaptive radiation of salamanders in middle American cloud forests.” Annals of the Missouri Botanical Garden 74: 242-264. For the ballistics of salamander tongues, see Deban, S. M., O’Reilly, J. C., Dicke, U., and van Leeuwen, J. L. 2007. “Extremely high-power tongue projection in plethodontid salamanders.” Journal of Experimental Biology 210: 655-667; I have given the projection distances and lengths for the Bolitoglossa salamanders (see table 1). For carnivorous bromeliads-within-bromeliads, see p. 261 of Juniper, B. E., Robins, R. J., and Joel, D. M. 1989. “The Carnivorous Plants.” Academic Press.

For egg numbers in the blue crab, see Churchill, E. P. J. 1917-1918. “Life history of the blue crab.” Bulletin of the United States Bureau of Fisheries 36:95-128; for males needing 15 days to recover from sex, see Jivoff, P. 1997 “Sexual competition among male blue crab, Callinectes sapidus.” Biological Bulletin 193:368-380.

For a general account of the biology of Metopaulias depressus, including the existence of family groups and overlapping generations, see Diesel, R. 1989. “Parental care in an unusual environment: Metopaulias depressus (Decapoda: Grapsidae), a crab that lives in epiphytic bromeliads.” Animal Behaviour 38: 561-575. For leaf removal and snail shell additions, see Diesel, R. and Schuh, M. 1993. “Maternal care in the bromeliad crab Metopaulias depressus (Decapoda): maintaining oxygen, pH and calcium levels optimal for the larvae.” Behavioral Ecology and Sociobiology 32: 11-15. For protection from damselfly larvae, and the size of damselfly appetites, see Diesel, R. 1992. “Maternal care in the bromeliad crab, Metopaulias depressus: protection of larvae from predation by damselfly nymphs.” Animal Behaviour 43: 803-812.

Many thanks to Dan Haydon and Gideon Lichfield for insights, comments and suggestions.