C1139 Social Insects. Sophistication in the foraging trail networks of Pharaoh’s ants, Monomorium pharaonis. Lecture 15
Sophistication in the foraging trail networks of Pharaoh’s ants, Monomorium pharaonis
Aims
1. To describe experiments investigating foraging trail networks in Pharaoh’s ants
2. To present hypotheses for the existence of multiple trail pheromones in Pharaoh’s ants
Objectives
1 Understand what the experiments found out, why they were designed the way they were, and the implications
of the results
2. Understand the broader picture and be able to compare Pharaoh’s ants to honey bees in how foragers are
directed to where the food is.
Big Picture
In a previous lecture we saw how an ant colony could establish pheromone trails leading to food via self
organization if individual foragers react to local conditions in simple ways: foragers are walking around outside
the nest; if a forager finds food, she lays trail back to the nest; if a forager finds a trail, she follows it to the food.
The global pattern arises from agent behaviour and leads to an adaptive pattern with trails leading to food
sources. However, real ants seem to have greater sophistication that this. This lecture will describe recent
research carried out on the Pharaoh’s ant, Monomorium pharaonis, which uses at least three different trail
pheromones.
Pharaoh’s ants as a study species
The Pharaoh’s ant, Monomorium pharaonis, is a good species to study pheromone trails and foraging. The
worker ants are only 2mm long, and even short foraging distances, 50cm, are realistic. In this way foraging can
be studied in the lab. An ant colony is kept in a small wooden box or tube within a larger plastic box that acts as
a foraging territory. Workers do not seem to be guided by their own memory or landmarks when foraging.
Most social insects are very hard to breed in the lab, but not Pharaoh’s ants. Nestmate males and young queens
mate readily. In addition, colonies have multiple queens. To make two colonies you just split a colony, making
sure that each part has a few queens. Pharaoh’s ants are “unicolonial” meaning that colonies are not well defined.
Ants from different nests can be combined without fighting. It is easy to make up experimental colonies to
whatever size you want simply by adding or removing ants. Pharaoh’s ants have been introduced to the UK
where they can be pests inside buildings. It is thought that they come originally from Africa.
Measuring trail longevity: short-lived attractive pheromone
This experiment used one colony per trial. The colony was given a Y-shaped foraging trail with a syrup feeder at
the end of one branch. The other branch had no food. The branch with no food had a piece of paper over the
substrate. After c. 20-30 minutes a trail had been established. The ants were brushed off and the masking paper
on the non-feeder side was removed. Ants that reached the bifurcation were brushed off after they had made
their choice between the branch that had led to food and the other branch. The proportion that walked towards
the food side was initially c. 80% when the trail had been laid on polycarbonate plastic and 70% when laid on
newspaper. The proportion decayed to 50% within 10-15 minutes. This shows that the attractive pheromone
decays rapidly. In addition, the fact that only 70-80% initially chose the rewarding branch shows that the
attractive pheromone alone, versus a branch with no attractive pheromone, is not enough to eliminate errors in
trail choice.
Jeanson, R., Ratnieks, F. L. W., Deneubourg, J.-L. 2003. Pheromone trail decay rates on different substrates in the Pharaoh's
ant, Monomorium pharaonis (L.) Physiological Entomology 28: 192-198.
Measuring trail longevity: long-lived attractive pheromone
In the first experiment in this study a colony was allowed to form a straight trail on a piece of photocopier paper
to a syrup feeder by using two strips of plastic to confine the ants to a narrow corridor. The paper was then
stored away from any ants for up to 4 days. The paper was then given to another colony, with a syrup feeder in
the same position as before. The ants were observed to see if they formed a trail to the feeder in the same path as
1
C1139 Social Insects. Sophistication in the foraging trail networks of Pharaoh’s ants, Monomorium pharaonis. Lecture 15
before. The results show that a colony could re-establish a trail in the same location as before even two days
after the original trail had been established. The strength of the original trail depended on the number of ants that
had walked the trail when establishing it. These results are strikingly different to the previous experiment, which
showed that an attractive trail decays in 10-15 minutes.
The first experiment worked at the global level to determine if the colony could re-establish a trail. The
second study worked at the local level by studying the behaviour of single ants. A piece of trail is cut out and
placed on a larger sheet of paper. A forager ant is placed on the paper in such a way as to cross the trail at right
angles. When it does this it may carry on straight over the trail or turn 90 degrees to join the trail. It turns out that
only ants walking with their antennae touching the substrate, that comprise about 20% of the foragers, can detect
the trail. This result indicates that the trail is marked with a non-volatile chemical.
Jackson, D. E., Martin, S. J., Holcombe, M. Ratnieks, F. L. W. 2006. Longevity and detection of persistent foraging trails in
Pharaoh's ants, Monomorium pharaonis (L.). Animal Behaviour 71:351-359.
“No entry”: discovery of repellent trail pheromone
Repellent trail pheromones had been predicted because computer simulation studies showed that they could be
effective. This study was the first to discover a repellent trail pheromone. The first experiment used two
colonies. The first colony was given a Y-shaped foraging trail with a syrup feeder at the end of one branch. The
other branch had no food. The second colony had a similar Y-shaped foraging trail but with feeders at both ends
so that the ants would visit both branches. After c. 20-30 minutes trails had been established. Paper substrate on
the non-feeder side was then cut out from the trail of colony 1 and placed on the trail of colony 2. The other side
received a piece of control substrate from colony 1. This had been in the colony’s foraging territory so that ants
could walk all over it, but had not been part of a trail. In the modified trail system of colony 2, 69% of ants that
reached the branch continued on. Of these 72% chose the control side. These results show that the substrate from
the non-rewarding branch of the trail in colony 1 repelled ants, because it was chosen less than a neutral control.
The results could not be explained by differing amounts of attractive pheromone.
In the second experiment a Y-shaped foraging trail was set up as before with a syrup feeder at the end of
one branch. The other branch had no food. After c. 20-30 minutes trails had been established. Paper substrate on
the non-feeder side and feeder sides were then cut out and tested on a straight trail leading to a feeder. Control
substrate and substrate from the trail stem were also tested. On the straight trail two things were observed. Did
an ant make a U-turn? Did an ant zigzag (move from side to side) or walk straight (not touch the side)? Uturning was less on substrate from the stem and both ends of the feeder branch than on controls. U-turning was
more than controls for substrate from the branch area of the non-feeder branch. U-turning was not-significantly
different than controls for substrate from the end of the non-feeder branch. These data show that the “no-entry”
pheromone is concentrated just after the bifurcation on the non-feeder side. Zigzagging was greater when ants
were approaching substrate from the non-feeder branch. These results support the U-turn data above and also
show that the no entry pheromone is volatile as ants zigzag before they reach it. Just like a human road sign it
tells traffic what to do before the junction is reached.
Robinson, E. J. H., Jackson, D. E., Holcombe, M. Ratnieks, F. L. W. 2005. “No entry” signal in ant foraging. Nature 438:
442.
Why so many pheromones?
The experiments described previously indicate the existence of three trail pheromones. Why so many? What is
their purpose? We don’t know but what follows is one idea based on complementary roles.
1. Attractive: volatile (decay 10 mins)
2. Attractive: non-volatile (decay 2 days)
3. Repellent: volatile
The following ideas are just a hypothesis for the moment. They make sense, but must be tested by experiments.
The trail network is marked out with attractive non-volatile pheromone. This acts as a long-term memory, c. 2
days, of where the colony has been foraging and walking. It leads to all locations where food has been collected
in the past few days and where food may perhaps occur again. If food is found at a particular location the path is
2
C1139 Social Insects. Sophistication in the foraging trail networks of Pharaoh’s ants, Monomorium pharaonis. Lecture 15
marked with the volatile attractive pheromone. This guides ants along the correct path and helps them chose the
correct branch to take at trail bifurcations. At trail bifurcations the non-rewarding branch is marked with the noentry pheromone. This helps ants chose the correct branch at a trail bifurcation in which only one branch leads to
food. When the food at one location is all gone, this branch becomes less attractive by the decay of the volatile
attractive pheromone and the laying of the no entry pheromone.
Ratnieks, F. L. W. 2008. Biomimicry: further insights from ant colonies? Pages 58-66 in: Liò, P., Yoneki, E., Crowcroft, J.,
Verma, D.C. (editors) Bio-Inspired Computing and Communication. Springer Verlag, Berlin & Heidelberg.
Ratnieks, F. L. W. 2006. Can humans learn from insect societies? Nova Acta Leopoldina 345: 97-116.
Robinson, E. J. H., Green, K. E., Jenner, E. A., Holcombe, M., Ratnieks, F. L. W. 2008. Decay rates of attractive and
repellent pheromones in an ant foraging trail network. Insectes Sociaux 55: 246-251.
Jackson, D. E., Ratnieks, F. L. W. 2006. Primer: Communication in ants. Current Biology 16 (15): R570-R574
U-Turning/Trail-laying specialists (note: not covered in lecture slides)
The previous studies showed only some of the sophistication in the trail systems of Pharaoh’s ants. In this
experiment a long trail is set up inside a bowl. The trail curves and is in a bowl so that the ants cannot get
information about their orientation from landmarks. Individual ants are observed to see if they make U-turns
while walking along the trail. When an ant reaches the end of the trail it can be turned around on the turntable.
Ants can also be moved from one part of the trail to another by ferrying them over on a small piece of paper.
The results show that about 37% of the ants make a U-turn. Ants that make a U-turn typically make many Uturns and also walk with their sting extended. These results indicate another behavioural specialization. A
proportion of the ants on a trail are walking backwards and forwards laying pheromone. Presumably this is to
help guide the other ants.
Jackson, D. J. Hart, A. G. 2006. U-turns on ant pheromone trails. Current Biology 16: R42-R43.
Using trail geometry to determine trail polarity (note: not covered in lecture slides)
The physicist Richard Feynman made some observations on ants and suggested that they might polarize their
trails using two pheromones and a space marker. If an ant lays two pheromones in an ABABAB pattern then the
trail is not polarized as it reads the same in both directions. Feynman suggested that a space marker would
polarize the trail, as AB-AB-AB- reads differently in opposite directions. However, this is not a very practical
method as many ants laying pheromone would result in the pattern becoming confused.
The trail network of a Pharaoh’s ant colony is basically a bifurcating network in which the angle
between outward branches is approximately 60 degrees. This allows the branch leading to the nest to be
determined.
The experiment gave individual Pharaoh’s ant foragers trail bifurcations of different angles from 0
(straight) to 120 degrees. When the angle is 0 degrees (a straight trail) or 120 degrees (a perfectly symmetrical
bifurcation), the Correct/Incorrect ratio is 1. This shows that the ants had not done any better than guessing.
They had not been able to extract any information from the trail about polarity. But at intermediate angles the C/I
ratio is greater than 1, reaching 5.8 at 60 degrees. This high ratio means that the ants can obtain good
information from the trail geometry at intermediate angles. The actual angle is not critical although 60 degrees
appears to be the best.
This experiment shows that Pharaoh’s ant workers can use the geometry of a trail bifurcation to
determine polarity, but it does not show how they do this. That is a topic for future experiments.
Jackson, D. E., Holcombe, M., Ratnieks, F. L. W. 2004. Geometry gives polarity to ant pheromone trails. Nature 432: 907909.
Ratnieks, F. L. W. 2005. Outsmarted by ants. Nature 436: 465.
Recruitment to foraging
The experiments described in this lecture investigated how Pharaoh’s ant foragers are directed to food in their
environment but not how foragers are caused to leave the nest. In the honey bee, the waggle dance and the
shaking signal can both function to increase the amount of foragers. Pharaoh’s ants do have mechanisms to send
3
C1139 Social Insects. Sophistication in the foraging trail networks of Pharaoh’s ants, Monomorium pharaonis. Lecture 15
foragers out of the nest. Sudd (1960) claims that a trail does not recruit unless activation of workers in the nest
also occurs. Sudd (1957) describes four types of foraging communication in the nest:
1.
2.
3.
4.
sharing of food brought in
activation by disturbance leading to a propagation wave
direct stimulation from “excited ants”
scent of food may also contribute
Once foraging is in action, the number of foragers recruiting to a trail stays high, even if no fed workers return to
recruit more.
Sudd, J. H. 1957. Communication and recruitment in Pharaoh's ant, Monomorium pharaonis (L.). British Journal of Animal
Behaviour 5: 104-109
Sudd, J. H. 1960). The foraging method of Pharaoh's ant, Monomorium pharaonis (L.). Animal Behaviour 8: 67-75
Other ant species
The common garden ant, Lasius niger, also uses trail pheromones. In this species the trail seems to last longer
than in Pharaoh’s ants. Experienced foragers seem to rely mainly on route memory not pheromone trails in
relocating a feeding site. Possibly, the role of trail pheromone is more to help naïve ants in locating food.
Evison, S. E. F., Petchey, O. L., Beckerman, A. P., Ratnieks, F. L. W. 2008. Combined use of pheromone trails and visual
landmarks by the common garden ant Lasius niger. Behavioral Ecology & Sociobiology 62: 261-267.
Grüter, C., Czaczkes, T. J., Ratnieks, F. L. W. Decision-making in Lasius niger foragers facing conflicting private and
social information. (submitted).
Dussutour et al. (2009) find evidence for long lived and short lived trail pheromone in Pheidole megacephala,
and invasive ant species in Australia. The summary of this study says “In this paper we investigate the foraging
activity of an invasive ant species, the big headed ant Pheidole megacephala. We establish that the ants’
behavior is consistent with the use of two different pheromone signals, both of which recruit nestmates. Our
experiments suggest that during exploration the ants deposit a long-lasting pheromone that elicits a weak
recruitment of nestmates, while when exploiting food the ants deposit a shorter lasting pheromone eliciting a
much stronger recruitment. We further investigate experimentally the role of these pheromones under both static
and dynamic conditions and develop a mathematical model based on the hypothesis that exploration locally
enhances exploitation, while exploitation locally suppresses exploration. The model and the experiments indicate
that exploratory pheromone allows the colony to more quickly mobilize foragers when food is discovered.
Furthermore, the combination of two pheromones allows colonies to track changing foraging conditions more
effectively than would a single pheromone. In addition to the already known causes for the ecological success of
invasive ant species, our study suggests that their opportunistic strategy of rapid food discovery and ability to
react to changes in the environment may have strongly contributed to their dominance over native species.
Dussutour, A. Nicolis, S. C., Shephard, G., Beekman, M., Sumpter, D. J. T. 2009. The role of multiple pheromones in food
recruitment by ants. The Journal of Experimental Biology 212, 2337-2348.
Not all ants use trail pheromones, or even communicate where food is
The experiments described in this lecture show remarkable sophistication in the foraging trail network of
Pharaoh’s ants and their use of pheromones. However, not all ants use trail pheromones and in some species
there is no communication at all among foragers. In many ants a successful forager will directly lead nestmates
to the food. When one forager leads one other ant this is called tandem running.
Cataglyphis desert ants forage for dead insects killed by the heat. The ants forage individually and have
sophisticated navigational abilities. A worker wanders around looking for food. When it finds food it can return
directly to the nest entrance. It knows where it is at all times by summing up the distances it walks at different
angles to the sun. From this it can determine how to walk straight back home. These ants forage individually and
do not share information because their food does not occur in patches worth recruiting nestmates to.
4