IT’S the superstar of the fruit world: delicious, nutritious, conveniently packaged and easy to eat. You can pop it in your bag as a handy snack, or use it as part of all kinds of meals. Even children who turn up their noses at other fruits will happily eat a banana.
The yellow-skinned Cavendish banana that is so popular in Western countries is just one of hundreds of varieties. Bananas are a staple food throughout much of the tropics, with the flowers and stems as well as the fruit being eaten in some regions. In terms of financial value, bananas are the fourth most important food crop in the world, after rice, wheat and maize.
Advertisement
What’s more, the plants have many other uses besides eating. The fruits are used to brew beer in eastern Africa. Fresh banana leaves are still widely used for wrapping and serving food. Dried leaves were once used for writing on. The fibres in banana stems can be used to make everything from mats and ropes to fine cloths. In Japan, .
So where did this extraordinary plant come from? A combination of genetic and linguistic detective work is revealing exactly how some barely edible wild plants were turned into the tasty treats of today. The story of the banana turns out to be the story of the peoples who shaped it. It was their prehistoric movements around the Indian Ocean and tropical Pacific that gave rise to the fruit we love.
The ancestors of cultivated bananas can still be found growing in the rainforests of southeast Asia. The fruits of these wild bananas (pictured) are very different to those we eat today: they are short, squat and full of hard, bitter, pea-sized seeds. They have little flesh, and that doesn’t taste anything like the bananas we are accustomed to. “It’s not really worthwhile eating it,” says Pat Heslop-Harrison, a botanist at the University of Leicester, UK.
“Wild bananas are short, squat and full of hard, bitter, pea-sized seeds. They have little flesh”
Not today, anyway. Thousands of years ago, people living in regions where wild bananas grew may have had little choice, or may have eaten . Or perhaps they valued the plant not for food but for its other uses. Whatever it was grown for initially, banana cultivation began at least 7000 years ago, and possibly as early as 10,000 years ago.
The evidence comes from the highlands of New Guinea. At an ancient site called Kuk Swamp, archaeologist Tim Denham of La Trobe University in Melbourne, Australia, and his colleagues have found many signs of plants being cultivated (). But what was being grown? Although the plants rotted away millennia ago, they left behind telltale signs in the form of microscopic particles called phytoliths.
Phytoliths are tiny silica granules that form within some plants. Why they form is a mystery, but they have some extremely useful properties: they do not rot and they often have distinctive shapes that reveal not only which plant they formed in, but even which part of the plant. Banana phytoliths are found in abundance at Kuk.
Before these discoveries, archaeologists thought that full-blown agriculture was brought to this region by a group they call the Austronesians. These peoples migrated outward from southern China around 4000 years ago and became the dominant ethnic group in much of southeast Asia. Now it seems they were not the first farmers in this part of the world. “In island southeast Asia there were forms of agricultural cultivation going on that predate Austronesian involvement,” says Denham. “The banana opens vistas on worlds we don’t really know much about yet.”
At first, the bananas of Kuk must have been much like their wild relatives. The phytoliths from banana seeds look a bit different to those from leaves, and the earliest deposits at Kuk contain plenty of seed phytoliths, so these early bananas must have been seedy. However, Denham’s colleague Carol Lentfer of the University of Queensland in Brisbane has found that the proportion of seed phytoliths in the mix decreases as the centuries pass. “It would seem that there’s something going on at that time,” says Denham, who thinks the likeliest explanation is that the phytolith record traces a gradual transition from seeded wild bananas to seedless ones.
At the time cultivation began, the most widespread wild banana was a species called Musa acuminata, with many different subspecies growing in different regions (see “Mixing pot”). The genetic differences between these subspecies are large enough that if they are crossed, the resulting hybrids are usually sterile. Most sterile plants would not produce fruits, but bananas are peculiar – fruits still develop, but they are seedless. “It’s like having the placenta without the embryo,” says Heslop-Harrison. Presumably, some sort of genetic quirk encourages the developing fruit to pour its energy into pulp rather than seeds.
Whatever that quirk is, it seems widespread among wild bananas. Edmond De Langhe, one of the world’s leading banana researchers before his recent retirement from the Catholic University of Leuven (KUL), Belgium, has tried crossing different kinds of wild bananas. He found that the first generation had no seeds – but also very little pulp.
But then something interesting happened. Banana stems die after bearing fruit, but are replaced by new stems growing up from an underground corm. To De Langhe’s surprise, after this had happened a few times the fruits contained more pulp.
So the first step in the creation of modern bananas occurred when people began carrying the plants from place to place. Once a foreign subspecies started growing on a new island, it would have pollinated a few of the native bananas. The resulting seeds may have grown into hybrid plants producing seedless fruit with, once they had been growing a while, plenty of pulp. “Once that happens, any person walking through the forest is going to immediately detect that this is something to pick up and eat,” says Heslop-Harrison.
Mouthwatering mongrel
It has long been known that hybridisation played a key role in the domestication of the banana, but over the past few years researchers have started to work out when and where it happened. De Langhe and an international team of colleagues, for instance, collected nearly 550 samples of cultivated bananas and their wild relatives, and used genetic markers to work out how they were related. In virtually every case, they found, edible bananas were the result of hybridisations between two M. acuminata subspecies that were native to different regions. The resulting bananas are known as AA bananas, where A stands for acuminata.
And it appears the creation of seedless hybrids happened not just once or twice, but on hundreds of occasions. The mixing occurred in three broad areas. In the south, the subspecies from New Guinea mixed with one from Java. In the east, the subspecies in the Philippines mingled with the one from New Guinea. And in the north, several mainland subspecies crossed with island species (). Many of the resulting AA bananas are still grown for food today, often far from where they originated.
Remarkably, linguistic patterns match the genetic patterns “quite stunningly well”, says Mark Donohue, a linguist at the Australian National University in Canberra. He gathered over 1100 words for “banana” from indigenous languages throughout the islands of southeast Asia. He found that most of these words could be traced back to one of three ancestral words for banana – muku, qaRutayand baRat – that correspond with the three mixing areas. The word muku originated in New Guinea and it, or words derived from it, spread to Java, while qaRutay originated in the Philippines, and variants spread to some Indonesian islands and across to the Indian subcontinent. And baRat appeared in the Philippines, and spread to mainland southeast Asia.
These banana words carried over even when cultures underwent major changes, Donohue found. For example, linguistic reconstructions show that the term muku was used in the languages of the peoples that lived across Indonesia before the Austronesians arrived about 4000 years ago. But even as the Austronesian languages replaced the indigenous ones, people continued to call bananas muku, or variants thereon.
To Donohue, this continuity implies that bananas must have been an important food crop by the time the Austronesians arrived. Why would they adopt the name muku if it referred to a wild species, Donohue asks. “We have no story for that. But if bananas were important – if they were a domesticated species – then we have a reason why the terms would have survived the complete language shift.” A parallel pattern, he notes, can be seen in northern Europe, where ancient words for “apple” predated the appearance of, and were adopted into, the Indo-European languages of later arrivals.
The Austronesians may have adopted the local name for the edible bananas they discovered as they moved from China to Taiwan to the Philippines and spread out among the islands. But it seems they also brought with them another banana species called Musa balbisiana, native to southern China, which was probably grown for its leaves or stem, not its fruit. So by 4000 years ago, the stage was set for the next big step.
While hybrids are usually seedless, viable seeds are occasionally produced after pollination. Pollen from hybrids can also fertilise other strains. So the intermingling of different varieties continued after farmers started growing seedless hybrids – and occasionally something special happened.
Plants normally have two sets of chromosomes, one set from the pollen grain and one set from the egg. But if things go awry – which may be particularly likely in the AA hybrids – a pollen grain or egg can end up with two sets of chromosomes instead of one. If fertilisation occurs, the result can be a banana with three sets of chromosomes. These “triploid” plants are often more vigorous, with bigger fruit, and thus more desirable as crop plants and more likely to be propagated by a sharp-eyed farmer. Most of the banana varieties grown commercially are triploid.
Here, too, detailed genetic analyses have begun to flesh out the pattern. “Genetics allows you to fingerprint what the ancestry is of the major cultivar groups,” says Denham, who was also a member of De Langhe’s international team. The results show that triploids formed independently on a few dozen occasions, but that just a handful of those events produced the most important varieties grown today.
Some produced triploids with two sets of chromosomes from M. acuminata and one set from M. balbisiana (AAB triploids). One such event gave rise to Pacific plantains, which were grown throughout Polynesia and the Pacific islands long before western explorers arrived. Genetically, these form a distinct group with a common origin but many local varieties.
This pattern suggests that the original Polynesian settlers brought them along when they spread eastward through the Pacific islands about 3500 years ago. If so, that means these AAB triploids must have become established as crop plants very early on, before the Polynesians set out on their long ocean voyages.
A second major group of AAB triploid plantains originated at about the same time. The genetic analyses of De Langhe’s group point to a separate triploidisation event involving M. balbisiana and a different AA parent from the eastern contact area around the Philippines. Instead of heading eastward into Polynesia, however, these bananas – and the people who cultivated them – went west, to Africa. Now known as African plantains, this group soon spread clear across the continent.
At an archaeological site called Nkang in Cameroon, on the west coast of Africa, banana phytoliths begin to appear from about 2800 years ago, says Luc Vrydaghs, an archaeobotanist at the Free University of Brussels in Belgium. The phytoliths in cooking residues, in particular, imply that people were wrapping food in banana leaves to cook in their fires, as many still do today. “If they’re already using banana leaves as a cooking tool, that implies that bananas were already well integrated into their life,” says Vrydaghs.
Other triploidisation events gave rise to AAA groups. East Africa is now home to a wide range of AAA bananas used for brewing and cooking, all of which bear genetic fingerprints of a single, ancient origin in the southern mixing area around New Guinea. The Cavendish banana found in western supermarkets is also an AAA triploid, but it probably arose more recently in the Vietnam region from a cross between a local banana and an AA hybrid carried north into that region by people who used the qaRutay word.
So it seems that bananas were, er, ripe for domestication. Whereas other key crops probably derived from single key mutations or hybridisations, edible bananas were created on hundreds of occasions. And their history is not just of academic interest. A fungus called Panama disease wiped out what used to be the main commercial variety, the Gros Michel, in the mid-20th century. Another fungus, called black Sigatoka, is now threatening to do the same to the Cavendish. To make it resistant to the fungus, there are two main options: genetic engineering, or recreating it from its wild ancestors with resistance bred in.
Attempts to recreate it have produced fruits with disappointing flavours. “It has not been satisfactory,” says De Langhe. “More and more, it was realised that you need to know what the exact parents were, and exactly how the crosses led to the cultivated varieties.” Despite all we have learned about the creation of the banana in recent years, those parents remain elusive. Let’s hope they turn up soon.
This article appeared in print under the headline “Go, bananas”
