HMC Bee Lab: Ant Skyscrapers: New Experiments With Turtle Ants

Figure 1: The Future of Bee Lab?

Imagine you’re designing a new subway system in a city. Ideally, you would want to ensure people can easily travel from one location to another in a short amount of time. But you also have a lot of limits; after all there are places you can’t construct subway lines and you can only build so many. Surprisingly, ants have to solve similar problems when choosing nests and finding food. They also must deal with trying to optimize the routes they take and nests they inhabit, while dealing with physical obstacles and limited resources in their environment. Here in Bee Lab, we have been researching exactly how turtle ants make these decisions, since unlike the imaginary subway designer, a colony of ants is making these decisions collectively. Over the last few semesters we have been exploring how the idea of “connectivity” affects this decision making, where ants choose between nesting in an area where all the nests are interconnected versus one where travel between nests is more restricted. The first experiment back in summer 2017 on this topic seemed to give very clear results, where every colony of ants chose to nest in the more interconnected section of the arena. Although this provided evidence that the difficulty of traveling between nests influences ant behavior, the next semester similar experiments were conducted looking at the same question, but gave very different results. Instead of all preferring more connected nests, only half the colonies ended up in these nests, while the other half nested in the less connected section of the arena.

Figure 2: Setup of turtle ant experiment in Fall 2017. The upper right corner is the connected section, while the lower half is the unconnected section

Unfortunately it is very difficult to figure out why these results were so different only using these two experiments. To be sure, there were definitely differences between these experiments that could have affected results: anything from general laboratory conditions to the season the ants were tested could have changed ant behavior. While repeating the same experiments may give a larger sample size and a better sense of the data, is there a better experiment we could conduct?

Figure 3: The natural habitat of turtle ants, mangrove trees, allows them to utilize intertwined branches to travel between nests and food sources

One of the main things we considered in brainstorming new experiments was the idea of distance. Each box in an arena is only a few inches wide, which means an ant might only have to travel a few feet or less to travel through all the different boxes in the arena. However, wild turtle ants live in mangroves and are able to travel much greater distances within a single tree, or even by traveling between trees via intertwined branches. So perhaps when choosing between different nests in the laboratory, ants are not very concerned with how connected nests are when they are only a few inches away from one another.

Figure 4: A hypothetical setup for experiments during the summer. The upper boxes would contain nests, while the lower is just for ants to transfer from one tube to another

Returning to ideas for the summer, simply expanding our arenas to a bigger size was one solution, but we also wanted to look at more novel ways to solve this problem such as making part of the arena vertical. In this new model, instead of traveling in a 2D structure, ants will be able to travel both horizontally to new boxes and vertically upwards. For this setup our preliminary plan would be to have plastic tubing that connects each of the nests to a central box. This would increase the distance ants need to travel between certain nests, while still creating one section of the arena that is interconnected and one that is not. In addition, this set-up is perhaps closer to the natural habitat of these ants. As mentioned before this species of turtle ants nests in mangrove trees in Florida, which often have vines and branches that create a complex network within and between trees. This system of tubes in the new setup not only increases the distance ants travel, but may closer resemble the twisting and turning branches and vines they are used to climbing. Ideally, if this setup is closer to the ants’ natural habitat, they will behave more naturally, ensuring that the results of any experiment are indicative of actual ant behavior and not just laboratory conditions.

However, there are still some concerns that would need to be fully addressed with this model before it could be used for summer research. Because turtle ants nest in tree branches, potentially they could see the inside of the plastic tubes more as potential nests than as pathways to travel. Similarly, we would need to ensure that ants do not prefer the larger and more open space of the central box to that of the smaller boxes with nests. Despite these potential problems, this new design has the potential to help us determine how connections between nests matter to turtle ants. In this new set-up, choosing a less connected nest would mean considerably more effort and time to get to other nests or food sources. If ants were to nest in a more interconnected section, it would support the idea that effort or travel time between nests is an important factor in ant decision making.

HMC Bee Lab

Pages

Wednesday, May 9, 2018

Ant Skyscrapers: New Experiments With Turtle Ants

No comments:

Post a Comment