This is the english version of a post written in portuguese, which can be found here.

The desert ant Cataglyphis fortis is able to find its nest even after walking for meters like crazy in the featureless Sahara desert. It manages that by integrating two information: the direction it is walking and the distance. Every time it walks towards a direction, it manages to recalculate where its nest is supposed to be with great accuracy.
In order to know to which direction the desert ant walked, it ant uses what we call a celestial compass (seriously), which I will detail in another post. Today we will understand how desert ants measure distances and, better, how mymercologists figured that out.
One of the hypothesis made by the scientists is that the ants can measure their energetic status. Basically, they would be able to estimate how much gas they still have in their tanks and use this information to calculate the distance they already have travelled. However, ants are able to correctly measure the travelled distance even carrying objects with different weight, which would require different amounts of energy to be transported. Another hypothesis was that desert ants use the image flow from their eyes but they also managed to accurately estimate distances in the dark.
What Cataglyphis fortis actually do is to count their steps. As their steps have a well defined distance, this is a good parameter to use as a reference. Now, the question arises: how the scientists managed to test this hypothesis?
The first step was to figure out a way to measure C. fortis notion of distance. To do that, they made the ants walk thought a corridor from their nests to a food source. When the ants reached the food, the scientist took them and put in a second parallel corridor. In this corridor, the ants would walk for a distance equivalent to the distance they walked to the food source before starting a searching behaviour for their nest. Believe or not, they always guessed this distance from the food source to the nest accurately.
The second step was to try to change the step-counting system of the desert ants. How would they do that? Heavy drugs? Lobotomy? That was the million dollar question! What they did was to change the length of the ants’ steps! After all, if their steps where larger, they would cover a greater length with the same number of steps. So, after the ants arrived at the food source, the scientists would get them and either cut a piece of their legs or put small pieces of plastic to extend them! The image bellow shows what was done and, if you pay close attention to the photo above, you will notice the red pieces of plastic used. Surprisingly, the ants were able to walk without problems afterwards!

The last step was to repeat the first experiment with the ants with different leg length. In the figure bellow, we can see the frequency of the search behaviour plotted against the estimated distance to the nest. The behaviour of a normal ant is in purple: they usually intensify the search behaviour exactly where the nest was supposed to be. Ants with legs cut, in green and yellow, start their search behaviour before the place the nest was supposed to be. Indeed, they walked a smaller distance with the same number of steps. The opposite happens with the ants with plastic pieces, in red. The data also coincide with the theoretical predictions.

Moreover, if the ants with stilts or stumps went to the food source from the nest in those conditions, they would estimate their distance to their nest accurately, showing that the whole treatment did not impair the ant’s ability to measure distances.

This set of experiments allowed the scientists to finally conclude that the ants measure distances by counting their steps. In this video, in German, you can see all the parts of this experiment: how ants behaviour in the field, how the experiment with the corridor works, the ants’ search behaviour and how they change their legs length. I did not understand what they are saying, to guarantee its quality but the images are certainly amazing. Enjoy!
Wittlinger, M., Wehner, R. and Wolf, H. (2007). The desert ant odometer: a stride integrator that accounts for stride length and walking speed. J. Exp. Biol. 210, 198-207.
* translator note: the title is a reference to the Portuguese poet Fernando Pessoa, whose famous verses can be understood either as “Navigation is required/ it is not required to live” or “Navigation is accurate/it is not accurate to live”.