C1139 Social Insects. Task Partitioning in Insect Societies. Lecture 16
Task Partitioning in Insect Societies
Aims
1. To provide information on the following topics in task partitioning: definition, relationship to division of
labour, benefits and costs, taxonomic occurrence, patterns of material transfer, etc.
Objectives
1. Understand the relationship of task partitioning to division of labour in the organization of work in insect
societies
2. Learn specific examples
Big Picture
Insect societies have sophisticated ways of organising their work. Two of the main principles used in work
organisation are Division of Labour (DoL) and Task Partitioning (TP). DoL has long been studied whereas TP
has been less studied. DoL and TP are not mutually exclusive. In fact, they often go together. Nectar collection
in the honey bee involves both. A forager transfers her nectar to a receiver bee. This is TP as the task of
collecting and storing nectar has been divided into two subtasks of collecting and storing. The receivers are a
different, younger, group of workers than the foragers. This is DoL. Without TP, DoL between receivers and
foragers would not be possible. TP in nectar collection is not essential. Forager bumble bees collect nectar and
store it themselves. The following word equations summarise the difference between TP and DoL.
TP
DoL
=
=
task/workers
workers/tasks
(a task/two of more workers that do the subtasks)
(the workers in the colony/the tasks that need doing)
Basic Information
All examples of TP involve the handling of materials, usually materials moving into the nest (forage) but also
materials moving through (leaves being processed in leafcutter ants) and out (e.g., garbage, excavated soil) of the
nest. The study of TP raises many questions. For example, when does it occur, in what context does it occur, and
in which species does it occur? How is material transferred (directly, indirectly, one transfer, two transfers etc.)
What is the adaptive significance of TP—why do it? Why make things more complicated? How to organize
things so that the sub-tasks are co-ordinated.
Possible benefits
TP occurs widely in insect societies. Presumably, it occurs because it enables the colony to function better. In
Polybia wasps building is a partitioned task. Some forager workers collect wood pulp that they transfer directly
to builders at the nest. A forager can collect a large load of wood pulp, sufficient for several builders. This
reduces the number of foraging trips. Foraging is a risky activity so reducing the number of trips is beneficial.
Honey bee foragers directly transfer their nectar at the nest to receiver bees. What is the advantage of doing this?
No one knows for sure, but it may enable the forager to get back to the flowers more quickly. Flower patches do
not last long, sometimes just a few days, so it may increase the total nectar collected by a forager over her life to
work fewer patches more intensely. This is because when one flower patch finishes the forager must locate
another, and this can take some time. For example, if the forager finds a new patch of flowers by following
dances she will likely have to repeat this several times as there is only about a 25% chance of a dance follower
locating the advertised patch. The lecture slides also show benefits in the collection of honeydew by ants and in
the collection of leaves in Atta ants. By cutting leaves in trees and allowing them to fall to the ground, where
they are cut up into small pieces, much walking is eliminated.
Possible benefits: hygiene in leafcutter ants
Atta colonies face the challenge of keeping their colony and fungus gardens healthy. The garden fungus is
susceptible to the weed fungus Escovopsis. This can kill the fungus garden. This will then kill the colony or
entail very costly remedies such as building a new nest. The ants have a whole series of defences against
Escovopsis. They carefully lick and clean leaves brought into the nest. They weed the fungus garden and even
groom individual strands of mycelium. They can direct more weeding to areas with Escovopsis. The worker ants
1
C1139 Social Insects. Task Partitioning in Insect Societies. Lecture 16
also have an antibiotic bacterium growing on their bodies that kills Escovopsis. It gives the ants a whitish
appearance. The workers also have a special gland, called the metapleural gland, which secretes disinfectant
chemicals. Research has shown that Atta colonies also have hygienic adaptations in the disposal of garbage.
These include several behavioural and organizational features.
Atta cephalotes can have massive colonies with many garden chambers in which their fungus crop is
grown. These are connected to underground waste chambers that receive spent compost from the fungus garden
and other waste. Workers working in the fungus garden take waste towards the garbage chambers. But few of
them enter the garbage chamber. Instead, the waste is placed in a cache in the connecting tunnel. Workers living
in the garbage chamber then take it the rest of the way. This is another example in which TP and DoL are
combined. The use of separate chambers, TP, and DoL isolate the garbage chambers from the garden chambers.
This is important as the waste may contain Escovopis and other pathogenic microorganisms. Ants working in the
garbage chamber are not allowed back into the garden chamber. Ants contaminated with garbage odour are
aggressed or killed if they try to enter the garden chamber. What A. cephalotes do is similar to the barrier
washing machines found in many hospitals, in which the contaminated and clean materials are kept separate.
Atta colombica also has large colonies with many garden chambers in which their fungus crop is grown.
But the waste is dumped externally above ground. There is division of labour between above ground workers.
Foraging and garbage dumping are separate careers. Workers do not switch between these two activities. In
addition, the garbage trail never leads in the same direction as the main foraging trail. The workers who work on
the garbage heap itself are former garbage dumpers. The garbage dump is often below a convenient tree trunk or
log to dump off. In this way the garbage dumping ants do not have to walk on the garbage dump itself. The
garbage dump is always downhill to the nest entrance. Garbage is also dumped into a stream if one is nearby.
Costs
TP involves material transfer. Therefore, time is wasted in transferring material, looking for a transfer partner,
queuing for partner, and sometimes in signalling to balance the system (e.g., tremble dance in honey bees).
Delays occur even if the balance in the work capacities of the various groups of workers is perfect because of
stochastic variation in arrival rates. There may be other costs, such as loss of material during transfer. Possibly,
the increased inter-individual contact that occurs with TP may result in increased disease transmission.
Transfer delay costs might be reduced by caching because this eliminates the need to be served directly
by another worker. Instead, material can be put down by one worker and later picked up by another worker. In
this way, caches can even out the “supply” and “demand” problem of direct transfer. But indirect transfer via a
cache does not eliminate all the difficulties associated with TP. Liquids are never cached even though it might be
possible to have a wax nectar trough for honey bee nectar foragers to cache their nectar in. Cached materials
might also degrade, be lost, or get stolen, and there is still the need to balance “cachers” and receivers.
Task partitioning results in a more complex arrangement of work and introduces the problem of
organizing the system to balance the work capacities of the various sub-tasks. The fact that task partitioning
occurs indicates that it must have substantial benefits, that are large enough to overcome these costs and more.
Occurence
TP occurs in ants, bees, wasps, and termites. It must have evolved many times independently. TP can vary
within a species for different forage materials. For example, honeybees use TP in the collection of nectar but not
in the collection of pollen. TP can vary between species for the same material. For example, nectar collection
involves TP in honey bees but not in bumble bees.
TP occurs in bringing material into the nest. These include liquids such as nectar, honeydew, and water,
and solids such as seeds, leaves, prey, wood pulp, and propolis. TP occurs in bringing various solids out of the
nest, especially garbage and excavated soil. TP also probably occurs inside the nest but has been little studied.
One example is in the processing of leaf fragments in leafcutter ants.
It is likely that many new examples of task partitioning will be found. While studying Atta leafcutter
ants in Brazil, we discovered task partitioning in fruit cutting. Large ants cut the fruit but medium-sized workers
transfer the cut pieces back to the nest.
Types and patterns of transfer
2
C1139 Social Insects. Task Partitioning in Insect Societies. Lecture 16
Direct transfer can be used for both solids and liquids but only solids are transferred indirectly via a cache.
Transfer is always at the nest in bees and wasps, which have flying workers. But transfer may take place at the
food collecting location or on the foraging trail in ants and termites, which have non-flying workers. Only nonflying social insects have foraging trails and much opportunity to meet each other outside the nest along a trail.
There are many different patterns of transfer. The most common is two interlocking cycles. When there
are three interlocking cycles these can be in a line or in a clover leaf arrangement of three interlocking cycles.
This occurs in Polybia occidentalis wasps (Vespidae: Polistinae: Epiponini) and probably in other epiponines. It
also occurs in the British ant Lasius fuliginosus (which lives on the Sussex campus) which builds a nest out of
material it has collected. There are cycles of honeydew foragers, building material foragers, and builders. There
can be considerable diversity in the transfer methods used, with various combinations of direct transfer, indirect
transfer, and no transfer.
Honey bee, Apis mellifera, foragers collect water, nectar, propolis (= plant resin), and pollen. Only
pollen does not involve transfer. The other three are all transferred directly in the nest. Pollen foragers collect
pollen in their pollen baskets. Pollen foragers are not unloaded by another bee but put the pollen directly into a
cell by placing their legs into the cell and kicking the pollen off. Transfer of pollen would probably be inefficient
as it would take two receivers to unload one forager. That is, one to take each of the two pollen loads. This
would presumably be carried in the receiver’s mandibles after transfer. It is not possible to transfer a pollen load
from one pollen basket to another. A pollen load is fragile and could easily break apart or be dropped during
transfer. This is a second possible reason why pollen is not transferred.
Water is a liquid that is held in the forager’s crop and is transferred like nectar. A receiver extends her
tongue and drinks from the forager’s mouth. Propolis is a solid and is held in the pollen basket. However, unlike
pollen, it is subject to TP. A builder unloads a forager by taking the forager’s propolis in her mandibles. Why are
pollen and propolis handled differently whereas water and nectar are not? It is suggested that propolis is directly
transferred because a forager cannot unload herself because the propolis is sticky. Propolis foragers can
sometimes wait hours to be unloaded if there is little demand for propolis.
Simulation modelling of 2-cycle TP
A computer simulation can be set up to investigate queuing delays. It is possible to vary the work capacities of
the foragers and receivers, and the mean and variance in the durations of the foraging and receiving trips. The
simulation generates the delays in being unloaded for both receivers and foragers for either “serve in random
order” or “first come first served” protocols. (British people queue first come first served but in some other
counties, they serve in random order, to the annoyance of any visiting British.)
Even when the work capacities are balanced there are still delays. This is because of stochastic variation
in the arrival rates of foragers and receivers into the transfer area. The average stochastic delay in the larger
colonies is smaller because large size lessens the effect. This suggests that TP is more likely in larger colonies
because the delay cost will be lower.
The data support this prediction. Swarm-founding wasps and honey bees have TP in the handling of
wood pulp and nectar, but Vespinae wasps and bumble bees do not. These species have colonies founded by a
lone queen, so their colonies start small and often do not grow to large sizes. In Vespula wasps there are also
data to show that nectar is handled via TP in larger population colonies. Research carried out in Yucatan, Mexico
on stingless bees provided additional support for the hypothesis. Five species were studied and all had TP in
nectar collection. Stingless bees have swarm-founded colonies with colony sizes typically in excess of 1000
workers.
References
Anderson, C., Ratnieks, F. L. W. 2000. Task partitioning in insect societies: novel situations. Insectes Sociaux 47: 198-199.
Anderson, C., Ratnieks, F. L. W. 1999. Task partitioning in insect societies. 1. Effect of colony size on queueing delay and
colony ergonomic efficiency. American Naturalist 154: 521-535.
Anderson, C, Ratnieks, F. L. W. 1999. Worker allocation in insect societies: coordination of nectar foragers and nectar
receivers in the honey bee. Behavioral Ecology and Sociobiology 46: 73-81.
Anderson, C., Ratnieks, F. L. W. 1999. Task partitioning in foraging: effect of colony size on queueing times and
information reliability. Pages 31-50 in: Detrain C, Deneubourg J L, Pasteels J M (eds.). Information Processing in
Social Insects. Birkhauser, Basel.
Evison, S., Ratnieks, F. L. W. 2007. New role for majors in Atta leafcutter ants. Ecological Entomology 32: 451-454.
3
C1139 Social Insects. Task Partitioning in Insect Societies. Lecture 16
Hart, A. G., Anderson, C., Ratnieks, F. L. W. 2002. Task partitioning in leafcutting ants. Acta Ethologica 5: 1-11.
Hart, A. G., Ratnieks, F. L. W. 2002. Task partitioned nectar transfer in stingless bees (Meliponini): work organisation in a
phylogenetic context. Ecological Entomology 27: 163-168.
Hart, A. G., Ratnieks, F. L. W. 2002. Waste management in the leafcutting ant Atta colombica. Behavioral Ecology 13:
224-231.69.
Hart, A. G., Ratnieks, F. L. W. 2001. Leaf caching in the leafcutting ant Atta colombica: organisational shift, task
partitioning and making the best of a bad job. Animal Behaviour 62: 227-234.
Hart, A. G., Ratnieks, F. L. W. 2001. Task partitioning, division of labour and nest compartmentalisation collectively isolate
hazardous waste in the leafcutting ant Atta cephalotes. Behavioral Ecology and Sociobiology 49: 387-397
Hart, A. G., Ratnieks, F. L. W. 2000. Leaf caching in Atta leafcutting ants: discrete cache formation through positive
feedback. Animal Behaviour 59: 587-591.
Hart, A. G., Ratnieks, F. L. W. 2001. Why do honeybee (Apis mellifera) foragers transfer nectar to several receivers?
Information improvement through multiple sampling in a biological system. Behavioral Ecology and Sociobiology
49: 244-250.
Helanterä, H., Ratnieks, F. L. W. 2008. Geometry explains the benefits of division of labour in a leafcutter ant. Proceedings
of the Royal Society B 275: 1255-1260.
Ratnieks, F. L. W., Anderson, C. 1999. Task partitioning in insect societies (II): Use of queueing delay information in
recruitment. American Naturalist 58: 536-548.
Ratnieks, F. L. W., Anderson, C. 1999. Task partitioning in insect societies. Insectes Sociaux 46: 95-108.
4