Altruism in insect societies and beyond: voluntary or enforced?

1
Altruism in insect societies and beyond
voluntary or enforced?

• Tom Wenseleers
• Department of Biology
• University of Leuven, Belgium
• tom.wenseleers_at_bio.kuleuven.be

2
The origin of social behaviour
advanced social behaviourapparent in many
animals, e.g. social insects, some birds and
mammals, and in humans most advanced form of
social behaviour altruism helping another at a
cost to oneself puzzle how can behaviour that is
individually costly evolve under a regime of
natural selection?
3
How can altruism evolve?
William D. Hamilton (1964) close family ties are
essential helping relatives results in the
propagation of copies of the altruists own
genes in this way the gene for the altruistic
behaviour can spread high relatedness should
cause greater altruism
4
THREE BROTHERS
FIVE NEPHEWS
5
CLASSIC EXAMPLESOCIAL INSECTS
6
Could the textbooks be wrong?

• due to haplodiploidy, relatedness is indeed very
  high in insect societies (¾)
• but is it high enough to explain the workers
  altruism?
• no, based on theoretical models I will show that
  the levels of altruism observed in many
  contemporaneous species can only be explained as
  having evolved in response to social coercion
• altruism is not voluntary, but enforced

7
Altruism in insect societies
in
the context of the origin of social behaviour
decision
for a female to become a worker rather
than
breed solitarily
in advanced social species
becoming a worker rather than a queen
worker sterility (not lay eggs)
I
compared level of altruism you should get in
these last
2 contexts if individuals could behave free from
social
coercion
(voluntary altruism)
versus when coercion is
present
(enforced altruism)
8
First case of altruismbecoming a worker
9
Become a queen ora worker?
female larva
QUEEN EGOISTICOPTION
WORKER ALTRUISTICOPTION
Bourke Ratnieks 2001 Beh. Ecol. Sociob.
Wenseleers et al. 2003 J. Evol. Biol.
10
Theoretical model

• if every individual is able to control its own
  caste development you should get anarchy in the
  colony excess queens
• single mating (stingless bees) 14-20 of all
  larvae selected to develop as queens
• 10 matings (honey bees)56 of all larvae
  selected to develop as queens
• bees do not require so many queens, since mainly
  workers are needed for colony multiplication via
  swarming
• adult workers are selected to try to prevent
  excess queens from developing via social control

Wenseleers et al. 2003 J. Evol. Biol.
11
Honeybee caste fate enforced
99.99 of all larvae forced to develop as
workers even though 56 would like to develop as
queens enforced altruism
Individuals cannot choose their own caste fate.
Only 1 in 10,000 is allowed to become a queen.
12
Most stingless bees caste fate enforced
99.98 of all larvae forced to develop as
workers even though 20 would like to develop as
queens enforced altruism
queen cell
Individuals cannot choose their own caste fate.
Only c. 1 in 5,000 is allowed to become a queen.
13
Absence of social control Melipona stingless
bees
(queen overproduction)
Power to the individual larvae, social control
impossible
14
Yucatan, Mexico
São Paulo, Brazilië
15
Yes, complete anarchy !
Melipona stingless beesca. 10 of the female
larvae develop as queens anarchistic outcome, as
predicted by model
Wenseleers Ratnieks Proc. Roy. Soc. 2004
16
Most excess queens killed...
Wenseleers et al. Ethology 2003
17
...or escape being killed by parasitizing
queenless colonies

• Melipona scutellaris some virgin queens escape
  being killed by leaving the colony and
  parasitizing unrelated queenless hives
• if the mother queen dies in 30 (7/24) of the
  cases it is replaced by an unrelated queen coming
  from other queenright colony

D.A. Alves, V.L. Imperatriz-Fonseca, T. Francoy,
P. Nogueira-Neto T. Wenseleers, unpublished
18
Effect of social control
M. quinquefasciata
M. seminigra
M. pseudocentris
M. beecheii
M. interrupta
M. bicolor
NO SOCIAL CONTROL Queens reared in worker
cellsExcess queens reared anarchy
M. melanoventer
M. quadrifasciata
M. subnitida
M. marginata
M. scutellaris
M. fuliginosa
M. asilvae
M. rufiventris
M. favosa
M. trinitatis
M. compressipes
SOCIAL CONTROL Queens reared in queen
cellsOptimal of queens rearedfemales forced
to become workers enforced altruism
Trigona amalthea
Trigona ventralis
Trigona ruficrus
S. postica
S. bipunctata
Tetragonisca angustula
Apis mellifera
0.01
0.10
1.00
10.00
100.00
of females reared as queens
D.A. Alves, V.L. Imperatriz-Fonseca, P.
Santos-Filho T. Wenseleers, unpublished
19
Anarchy in termites
lower termites (Kalotermitidae, Termopsidae)
all individuals except soldiers totipotent when
royal pair is lostexcess of individuals develop
asreplacement reproductivesup to half of all
individuals developas replacement
reproductivesfight until only a single pair
remains
Cryptotermes sp.
photo J. Korb
20
Evasion of social control
Schwarziana quadripunctata
21
Evasion of social controldwarf queens in
Schwarziana bees
some females reared in worker cells develop as
small dwarf queens rather than
workersstrategy to evade an intended worker
fate89 of all queens produced are dwarf
queenssame weight as workers, so meant to
become workers22 of colonies headed by these
small queens
q
Q
w
Q
q
q
Q
Wenseleers et al. 2005 Biol. Lett. Wenseleers et
al. 2004 Am. Nat.
22
Conclusion

• caste development generally under tight social
  control
• females usually forced to become workers against
  their own evolutionary interests
• altruism is not voluntary, but enforced
• absence of social control causes anarchy

23
Second case of altruismworker sterility
24
Workers can reproduce but often they dont
Workers can lay unfertilised male eggs but
usually only few do so.Why are workers so
altruistic?
worker
queen
25
Reproduce or remain sterile?
LAY EGGS
EGOISTICOPTION
REMAIN STERILE
ALTRUISTICOPTION
Wenseleers, Helantera Ratnieks 2004 J. Evol.
Biol. Wenseleers et al. 2004 Am. Nat.
26
Theoretical model

• Hamiltonian prediction high relatedness should
  favour greater voluntary altruism(fewer egg
  laying workers)
• But also an influence of sociale pressure in
  many species, eggs laid by workers are
  cannibalized or policed by the queen or by
  other workers More effective policing selects
  for fewer workers to lay eggs in the first
  place enforced altruism

Wenseleers, Helantera Ratnieks 2004 J. Evol.
Biol. Wenseleers et al. 2004 Am. Nat.
27
Queen policing
Common bumblebee Courtesy of the BBC series Life
in the Undergrowth
28
Queen policing
29
(No Transcript)
30
Worker policing
Ratnieks Visscher Nature 1989
31
Worker policing
German wasp Vespula germanica
Bonckaert et al. Beh. Ecol. 2008
32
Effectiveness of the police system
W
W
Worker policing many
against many most effective
W
W
Queen policing
one against many less
effective
W
W
Q
W
W
33
What causes worker sterility?
34
Which factor is the most important relatedness
or social pressure?

• comparative study of 10 species (9
  waspshoneybee)
• effectiveness of the policing and of
  reproductive workers determined
• mother queen mates with a variable of males ?
  variation in relatedness

35
Altruism is enforced
Asian paper wasp
30
saxon wasp
degree of altruism
red wesp
of egg-laying workers
10
tree wasp
Norwegian wesp
median wesp
5
hornet
German wasp
common wasp
honeybee
0
shows social pressure is the cause of workers
altruism !
Wenseleers Ratnieks Nature 2006
100
99
98
95
90
80
70
50
30
effectiveness of the policing
36
and not voluntary
Asian paper wasp
Polistes chinensis
25
25
saxon wasp
Dolichovespula saxonica
tree wasp
red wasp
D. sylvestris
Vespula rufa
7.5
7.5
Norwegian wasp
D. norwegica
5
5
median wasp
D. media
2.5
2.5
German wasp
Vespula germanica
degree of altruism
hornet
Vespa crabro
of egg-laying workers
common wasp
0.75
0.75
Vespula vulgaris
0.5
0.5
0.25
0.25
0.075
0.075
honeybee
Apis mellifera
opposite to Hamiltonian scenario !
Wenseleers Ratnieks Nature 2006
0.3
0.4
0.5
0.6
0.7
0.3
0.4
0.5
0.6
0.7
genetic relatedness
37
Why does low relatedness lead to more altruism?

• explanationwhen relatedness is low (r lt 0.5)
  workers are morehighly related to queens sons
  (r 0.25) than toother workers sons (r lt 0.25)
• this selects for workers to police each others
  eggs
• worker policing is more effective than queen
  policing
• meta-analysis of 90 species also shows that
  worker reproduction is more effectively inhibited
  in species with low relatedness

Ratnieks 1988 Am. Nat.
38
RELATEDNESSLOW HIGH
100
100
MIEREN
ANTS
BIJEN
BEES
WESPEN
WASPS
Workers most related tothe sons of other workers
werksters meest verwantmet zonen koningin?
worker policing
adult males producedby workers
Workers most relatedto the sons of the queen
10
10
1
1
t-test, p0.0000000001 n90 species
0
0
-
0.15
-
0.10
-
0.05
0.00
0.05
0.10
0.15
-
0.15
-
0.10
-
0.05
0.00
0.05
0.10
0.15
relatedness difference between workers and
queens sons
Wenseleers Ratnieks Am. Nat. 2006
39
In queenless coloniesHamiltonian prediction
recovered
40
honeybee
35
german wasp
30
common wasp
Asian paper wasp
25
degree of altruism
of egg-laying workers
red wasp
20
tree wasp
15
median wasp
saxon wasp
10
hornet
norwegian wasp
5
0
in queenless coloniesno policing/enforcementHam
iltonian predictionrecovered
0.3
0.4
0.5
0.6
0.7
Wenseleers Ratnieks Nature 2006
genetic relatedness
40
What about the variation in specieswith
single-matedqueens?
RELATEDNESSLOW HIGH
100
100
MIEREN
ANTS
BIJEN
BEES
WESPEN
WASPS
werksters meest verwantmet zonen koningin?
worker policing
adult males producedby workers
10
10
1
1
0
0
-
0.15
-
0.10
-
0.05
0.00
0.05
0.10
0.15
-
0.15
-
0.10
-
0.05
0.00
0.05
0.10
0.15
relatedness difference between workers and
queens sons
41
Variation in male parentage not linked to
policing

• e.g. stingless bees worker reproduction usually
  not policed, no variation in relatedness (r
  0.75)
• yet worker reproduction varies massively 0-98
  of all males workers' sons
• possible explanation colony-level cost if
  workers deposit a male egg in a cell it will
  reduce the number of workers produced since
  worker-laid eggs will also compete with female
  eggs laid by the queen
• prediction worker reproduction should be more
  common if colony produces a lot of workers, i.e.
  if the queen lays mostly female eggs (smaller
  colony-level cost)

42
Prediction supported
43
Conclusion

• social pressure is often the true cause of the
  workers altruism (worker sterility), and close
  family ties usually not required
• altruism is usually not voluntary, but enforced
• low relatedness actually correlated with greater
  cooperation because it is associated with tighter
  social control

44
Anti-social bee-behaviour. BBC News Magazine Egg
police crack down on broody bees. New
Scientist Cops with six legs. Law and order
among insects. Science News A bug's life -
Orwell style. Imprint Bijenpolitie.
Knack Bijen leven in politiestaat. Quest
Magazine Sociale sancties werken. De
Standaard Diktatur im Bienenstock. n-tv.de
(Duitsland) (No reproductive rights in insect
police states). Kisti (Korea) ?????????
???????????? ????????? ??????????????
???????????? ???????? (Social insect altruism is
maintained by policing methods). Elementy
(Rusland) Insectos decretan pena de muerte
(Insects declare the death penalty). El
Colombiano (Colombia) Insectos reciben órdenes
(Insects receive orders). TVN (Chili)
45
What about the origin of eusociality?

• Richard Alexander (1974) parental manipulation
  theoryparents force offspring to take on a
  worker role, e.g. by unferfeeding them
• But little supporting evidence, e.g. in Polistes
  annularis even the smallest females can leave the
  nest and become a foundress the next year
• Eusociality should be more common if it had
  evolved via a route of parental manipulation

46
Enforced cooperation in social vertebrates
Cooperatively breeding cichlid Neolamprologus
pulcher Subordinates that dont help are evicted
(Balshine-Earn et al. 1998)
47
Enforced cooperation in social vertebrates
Meerkats dominant females suppress breeding by
subordinates (Young et al. 2006)
48
Enforced cooperation in social vertebrates
Rhesus monkeys who do not share food are
punished(Hauser 1992)
49
Enforced cooperation in mutualisms
Soybean plants sanction root nodule bacteria that
do not fix nitrogen (Kiers et al. 2003)
50
Enforced cooperation in humans
51
the level of altruism displayed by 15 small-scale
societies is correlated with degree to which
defectors are punished
52
Acknowledgements
F.LW. Ratnieks
D.A. Alves
V.L. Imperatriz-Fonseca
wasp work F.L.W. Ratnieks, F. Nascimento, A.
Tofilski, M. Archer, N. Badcock, W. Bonckaert, T.
Burke, K. Erven, H. Helantera, L. Holman, K.
Vuerinckx stingless bee work V.L.
Imperatriz-Fonseca, D. Alves, T. Francoy, M.
Ribeiro, J. Quezada