91av

LOCUSTS aren’t the kind of creatures that inspire much in the way of sympathy. In a bad year locust plagues can wreck the livelihoods of a tenth of the world’s population and cover a fifth of the land surface of the globe. Desert locusts (Schistocerca gregaria) – the most harmful species – have been a problem for millennia, munching their way across Africa, Asia and parts of Europe since the dawn of agriculture.

Yet the insects that make up these mass- migrating swarms may not be as hell-bent on destroying our agriculture as they seem. This month, a team of researchers reported that far from pushing relentlessly forward in search of crops, locusts are just as concerned with avoiding being eaten themselves (Current Biology, ). This insight could help stop them in their tracks even before they take to the skies.

Locusts are the most voracious of the grasshoppers. Unlike other grasshoppers, which lead solitary lives, under certain environmental conditions locusts band together in huge numbers and eat almost anything in their path. In the past hundred years, there have been eight particularly serious plagues of desert locusts, some of which have spanned several years (See Diagram). Yet despite more than 50 years of research, we still don’t know exactly what drives locusts to swarm and, more importantly, how to stop them.

Computer models of swarming, and of similar behaviours such as flocking in birds and shoaling in fish, have shown that individuals within a group adjust their speed and alignment in response to others around them. As their density reaches a critical level, which varies depending on the species, there is a rapid transition from a chaotic mob to a highly focused and ordered group. In the case of real locusts, they spontaneously align with their neighbours and start to march in the same direction, this being decided by those at the front (see “Follow the leader”).

The big question, though, is what drives the locusts to march. Young locusts, or hoppers, are solitary, avoiding others of their species and wandering around harmlessly. They shed their exoskeletons several times and get progressively larger, though at this stage they are still in their flightless, larval form. If all goes well they remain solitary after they undergo their final moult into the adult winged form, and, except for when they mate, will avoid one another.

However, if population densities are particularly high, perhaps because of plentiful food in the preceding years, or because the patchiness of food supplies forces them to compete, the hoppers will come into close contact with each other. This sets off a chain of events that changes not only their behaviour and colour, but all manner of bodily functions, including metabolism and how their nervous system functions. This is their “gregarious” phase, during which they begin to actively seek one another out and start marching, joining with neighbouring groups until they form huge swarms that can be several million animals strong. Then, after a final moult, they become winged adults and set off as one.

In recent years researchers have discovered what triggers the change from solitary to gregarious – actual physical contact. In 2001 Stephen Simpson, then at the University of Oxford, showed that it is possible to change solitary locusts into the gregarious form simply by tickling them with a paintbrush (). “We stroked various body parts of locusts for 5 seconds each minute over several hours. Unless we stimulated hairs on the back legs, the change from solitary to gregarious behaviour failed to occur,” he says.

But this didn’t explain why the gregarious phase leads to marching. Finding out is crucial because this is the time when it might be possible to stop small groups joining together and getting out of hand. Conversely, if a flying swarm has already formed, then “the [appropriate] response will be much more difficult to provide quickly and [the swarm’s] impact on the environment and human health more difficult to reduce,” says Annie Monard of the locust monitoring group at the UN Food and Agriculture Organization (FAO) in Rome.

Because of this, locust control is a matter of predicting when and where small-scale marching bands will appear, then spraying them with insecticide. Locust Watch, set up by the FAO, provides risk forecasts based on local observations, meteorological forecasts and satellite data, but with locusts spread over such large areas, mostly in poor countries, they are difficult to keep track of.

Simpson, along with Iain Couzin, a mathematical biologist now at Princeton University, wondered if a better understanding of locust biology could improve the forecast. They decided to investigate what drives these newly gregarious insects forward. They had a few sinister clues from previous work with Mormon crickets – large, flightless insects that are found along cattle trails in the western US. , which can be up to 10 kilometres long. Each is made up of millions of insects that can march 2 kilometres per day in search of food.

Simpson, Couzin and colleagues found that Mormon crickets deliberately seek out high-protein foods, including seeds, their own shed exoskeletons and even carrion and mammal faeces. The locusts also ate soil soaked in cattle urine, suggesting their diet lacked salt. What’s more, they sought out another prime source of salt and protein – each other. Cannibalism in crickets and locusts is well known, but the researchers wanted to find out whether it had anything to do with their tendency to swarm.

They placed a selection of diets in the path of marching Mormon crickets. The food used was rich in either protein or carbohydrate, or contained equal amounts of both, or had neither. Their showed the crickets had a clear preference for the protein dish. The second most popular was the protein and carbohydrate dish, with no difference in the popularity of the other two.

They then offered the crickets cotton wool soaked in either pure water or salt solution of various strengths. The crickets showed a clear preference for the dish containing 0.25-molar salt solution. Given their propensity to eat each other, this didn’t surprise the scientists at all: Mormon crickets themselves are about this concentration.

“They seemed to be in an environment where salt and protein were the nutrients that they were clamouring for,” says Couzin. “They themselves are perfectly bundled packages of these essential nutritional requirements, and so the insects were constantly trying to bite each other.”

So could marching be a way of avoiding being eaten? To test this, the team captured 80 marching crickets and gave some of them access to as much protein and salt as they could eat, while leaving others hungry or salt-deprived. They found that sated crickets showed significantly less interest in marching than their hungry comrades.

These results suggest that juvenile crickets swarm because they are both protein and salt-deprived. In a situation where the scarcity of resources has brought large numbers of them together, the crickets would be forced to compete for food. Unless they are among those lucky enough to arrive first, they will be unlikely to get enough sustenance from plants or carrion, meaning that the next best source of these nutrients is each other. The crickets at the front move to escape the hungry jaws behind, and the whole group follows. The result is mass migration.

Boldly going forward

The team have recently found that the same is true for desert locusts. In their latest paper, they identify a number of cues the locusts respond to when they start to march. In one experiment, they restricted the vision of locusts already in the gregarious phase by applying black paint to their compound eyes. Some locusts had their vision restricted completely, others either from the front or from behind. When placed with other locusts, blind and back-blind locusts showed significantly less propensity to march than those that could see behind them. This suggested that seeing locusts approaching from behind was a stimulus to head in the opposite direction.

Another stimulus was physical contact with the locusts behind. When the team severed the sensory nerve to the locust abdomen so the locusts couldn’t feel other group members approaching, they marched significantly less than controls. These laid-back hoppers were eaten significantly more often as a result.

Taken together, these results suggest that while most models of locust and other swarm behaviour assume the group is driven by the appetites of those in front, in crickets and locusts at least, there is also considerable pressure from behind.

Nigel Franks, a biologist at the University of Bristol in the UK who specialises in collective behaviour, thinks it’s an important finding. “To understand mass movements in animals, it is vital to work out the mechanisms they use to respond to one another. This is a beautiful example of positive feedback which should help to explain why high densities and critical masses generate marching bands, with wave upon wave of these relentless insects,” he says.

Even so, individually disabling every locust in Africa and the Middle East clearly isn’t practical. However, the example of the Mormon crickets shows that if there is enough salt and protein in the vicinity of marching bands, they will lose the impetus to march. So could protein and salt lures be used to steer locusts away from farmland? “I’m not sure of the practicality of that,” says Couzin. “I’m sure if you put down large amounts of protein you could strongly influence the swarm, but in terms of management the real challenge is to find the swarm. And once you’ve found them you might as well spray them.”

Joyce Magor, a locust-control expert at the FAO, agrees. She says one strategy is to focus on areas which have experienced good rainfall, as this allows the locust population to boom. “We then decide whether numbers seen during surveys [of these areas] will multiply sufficiently for gregarisation and marching to occur. Whether the locusts march when crowded to colonise new habitats or to avoid cannibalism will not alter the forecast.”

Couzin, however, sees the new information feeding into the forecasting process, providing better models to predict where marching bands are likely to occur. “We know that food patchiness and nutrient content are important. This gives us a locust’s eye view of a habitat, so we can tell when and where [grouping] might happen, before an outbreak gets to swarm proportion,” he says.

Bringing locusts under control for good is likely to be a long way off, but as control agencies gear up for another summer of damage control, they can perhaps take heart from the fact that the locusts are having at least as difficult a time as they are.