Acacia erioloba in Namibian savanna Heather Campbell Mark Fellowes amp James Cook School of Biological Sciences University of Reading Reading RG6 6AS Ants and Acacias Antplant associations are widespread in over 100 genera of tropical angiosperms and at least 20 plant families and ID: 199556
Download Presentation The PPT/PDF document "Interactions between domatia-dwelling an..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Interactions between domatia-dwelling ants &
Acacia erioloba in Namibian savanna
Heather Campbell, Mark Fellowes & James Cook
School of Biological Sciences, University of Reading, Reading, RG6 6ASSlide2
Ants and Acacias
Ant–plant associations are widespread (in over 100 genera of tropical angiosperms and at least 20 plant families and in five of 12 ant subfamilies)Research focus on 3 plant genera: Acacia;
Cecropia
; and
Macaranga
Ant benefits:
Shelter Food - extrafloral nectar, food bodies, third party insects i.e. aphids, scale insects
Acacia
benefits:
Reduced encroachment
– competition & fire break
Pollination
Protection against herbivoresSlide3
Acacia
-dominated savanna, Kuzikus Wildlife Reserve, Namibian Central Kalahari
Study site
Kuzikus is approximately 180km south east of Windhoek, situated in the Namibian Central Kalahari.
The study habitat is
Acacia
-dominated savannah; however the site also incorporates dunes, saltpans,
bushveldt
and grassland.Slide4
Insects e.g.
Gonometa postica
caterpillars
Acacia erioloba
Species
Pollinators
Thorn-dwelling ants
Herbivores
Mammals
Tetraponera
Crematogaster
Cataulacus
Tapinoma
Sap sucking insectsSlide5
Regarding herbivory on A. erioloba we tested:1. Do patrolling worker ants act as a deterrent to insect herbivores?
2. How important are G. postica larvae in the defoliation of trees?3. Which ant species are responsible for antiherbivore activity?
QuestionsSlide6
32 trees sampled (16 trees per trial) – two adjacent branches selected; ants excluded from experimental branch and control left with natural ant fauna After 11 days branches removed with secateurs and sealed in plastic bags – this captured any herbivores present on the branch
Estimates of leaf damage on each branch made by recording the number of leaflets with herbivore damage out of the total number of leaflets (for 15 leaves per branch) Herbivores collected from branches were counted and identified
Experiment repeated twice – before and after hatching of
Gonometa
silk moth larvae
MethodsSlide7
Results1. Do patrolling worker ants act as a deterrent to insect herbivores?
The mean (
+
SE) percentage of leaflets damaged per branch, according to treatment, for the first and second trials of the ant exclusion experiment.
First trial; damage was 13% on ant excluded branches, nearly double (7%) the damage on branches with patrolling ants (t
16
= -3.52,
p
<0.005). Second trial ; branches without ants had damage to 33% of their leaves as opposed to 11% damage in branches with ants (t
16
=-6.73,
p
<0.001).
N
= 16 trees for both trials. Error bars show standard errors.
Branches without patrolling ants showed herbivore damage to 22% of leaflets, whereas branches with ants had only 9% leaflet damage (t
32
= -6.83,
p
< 0.001). Slide8
Results1. Do patrolling worker ants act as a deterrent to insect herbivores?
Herbivorous insects were found in greater numbers on branches without ants than on those with ants (t32 = -3.87, p < 0.001), although the relationship was more pronounced in the second experiment than in the first (first experiment; t16 = -2.42, p < 0.05; second experiment t
16
= -3.04,
p
< 0.01).
The mean (
+
SE) number of insects per branch, according to treatment, for the first and second trials collected at the end of the ant exclusion experiment.
N
= 16 trees for both trials. Error bars show standard errors.Slide9
Results
2. How important are G. postica larvae in the defoliation of A. erioloba trees?
The number of hemipterans and coleopterans are significantly lower on ant patrolled branches than branches without ants (Hemiptera: t
32
= -3.33,
p
< 0.005, Coleoptera: t32= 2.31, p<0.05).
The number of lepidopteran larvae are not affected by the presence or absence of patrolling ants (t32
= 0.1817,
p
= 0.8564).
There was no increase in lepidopteran numbers between trials on control or experimental branches (control: t
16
= 1.57, ns, experimental: t
16
=1.31, ns).
So differences in levels of herbivory cannot be attributed to an increase in presence of
Gonometa
larvae.Slide10
Results3. Which ant species on A. erioloba are responsible for antiherbivore activity?
The average percentage of leaflets damaged per branch, for branches with patrolling ants, according to ant species present. The category of multiple ants represents any combination of ant species where more than one species was present. Damage was recorded as number of herbivore damaged leaflets out of total number of leaflets per pinna. Pinna data was pooled for all leaves measured for damage on each branch, giving an average of leaflet damage per branch.
N
= 16 trees.
Analysis showed no significant association between herbivore damage to leaflets and different ant species present on branches (ANOVA, n=32,
d.f
.=4, ns).
Behavioural experiments suggest only
Crematogaster
behave aggressively, which supports data Slide11
Conclusion
Ants DO defend
A. erioloba
but are mainly effective against hemipteran herbivores
Further research needed to identify which species are responsible for defence
Tetraponera
Cataulacus
Crematogaster
TapinomaSlide12
BBSRC for funding my PhD Ministry of Environment and Tourism, Namibia for permission to carry out this research (permit
# 1614/2011
)
My supervisors Dr Mark Fellowes and Professor James Cook
My field assistant Ian Townsend
Staff at Kuzikus for logistical support
Dr
Graham Stone, University of Edinburgh for advice and loan of equipment
Brian
Taylor for advice on ant identification
Acknowledgements