Life is precarious for Cataglyphis fortis ants in the baking Tunisian desert. Preyed on by large insects and lizards, the ants are also at risk from the searing midday heat, but none of this deters them from foraging. Harald Wolf from the University of Ulm, Germany, explains that having successfully located fertile foraging ground, the venturers find their way home by taking the most direct path back, regardless of how circuitous the outbound journey was. But how do they assess whether it is worth returning to a location, or whether it is more productive to strike out in a fresh direction? Travelling to Maharès, the Tunisian village that he has been returning to for 16 years, Wolf, Siegfried Bolek and Matthias Wittlinger prepared to find out which factors the ants consider when coming to a decision (p. 3218).

Attaching two parallel channels to an ants' nest with a Y-shaped gate, the team placed a single cookie crumb 10 m along the first channel and waited for an ant to find it and return home. Then, when the ant re-emerged from the nest, the team closed the entrance to the training channel, directing it along the second channel. ‘The animal runs 10 m along the channel and says, “Food was here last time but it is not now, so let's look”’, explains Wolf. Monitoring the ant as it searched to and fro, the team recorded the locations where the ant switched direction and took the median of the turning points as the site where the ant expected to find another crumb. Ants that were sure of the location would concentrate their searches over a small range while ants that were less certain would search over much wider ranges.

Repeating the test on many more ants to determine how certain they were about finding more food, the team then tested other foragers with piles of five or 25 crumbs and eventually a mountain of over 800 crumbs. Then, the team gave the ants longer to become familiar with the location of the food, allowing them to visit the crumb piles five or more times before directing them to the test channel.

After 3 months of painstakingly recording the ants' movements, the team plotted the range over which the ants searched and the position where they expected to find the crumb stashes. ‘In the worst case (one crumb and one visit) then the animals go way past, the point is well past 10 m, and they have a large spread of points’, says Wolf. And the ants that had visited the small piles of five and 25 crumbs were also equally vague about the pile's location. ‘On the other hand, if you have 800 crumbs, then it really concentrates their search almost exactly on 10 m and the spread is really small’, recalls Wolf.

Next, the team compared the ants' behaviour after one training run with their behaviour after five or more training runs, and were amazed to see that the ants that had been provided with a single crumb on five repeat visits were as certain about its location as the ants that had hit the jackpot and located 800 crumbs on one occasion: ‘If the animal has visited the feeding site at least five times, they regard it as very reliable and concentrate their search’, says Wolf.

Finally, the team tested how the insects reacted when they were presented with a densely packed pile and a loose pile of 25 crumbs and found that the ants that had been presented with the densely packed pile searched with more precision. So the ants seem to value sites where they are certain that they are going to find food – and lots of it – but a reliable site is as good as a well-stocked one.

What counts for ants? How return behaviour and food search of Cataglyphis ants are modified by variations in food quantity and experience
J. Exp. Biol.