Figure 1.
Study site and nest locations.
A, Locations of the two ecological zones (+) and of the eight populations collected (−). YM: Ministère de la Recherche Scientifique et de l'Innovation; YN: North of Yaoundé; YS: South of Yaoundé; YU: Université de Yaoundé I; CA, CB and CC: populations A, B and C in the Campo Forest Reserve; AK: Akok. B, Locations of the nests within each of the eight populations. Populations YM, YN, YS and YU correspond to P. megacephala var. 1 and populations AK, CA, CB and CC to P. megacephala var. 2. Nests belonging to the same supercolony (cf. Results section) are indicated by the same symbol.
Figure 2.
Neighbour-joining tree of Pheidole megacephala from different localities, the sister species Pheidole sexspinosa and P. xerophila, and the outgroup Aphaenogaster senilis based on COI.
The percentage bootstrap supports are shown above the branches for the major groups. GeneBank accession numbers, population or supercolonies are indicated between brackets. Australian, South African and Malagasy samples refer to [34], [40] and [39], respectively. Mauritian samples come from personal collection (DF). Sequences for P. sexspinosa, P. xerophyla and Aphaenogaster senilis are taken from [38].
Figure 3.
Average intra- and inter-group aggressiveness between P. megacephala workers.
Box-plots indicate median (horizontal line), mean (diamond), interquartile range (box) and minimum and maximum values (whiskers). Nests are assigned to their population (A) or supercolony (B). (AK: Akok; CA, CB and CC: Campo Forest Reserve; YM, YN, YS and YU: Yaoundé and its periphery).
Table 1.
Number of nests sampled, location and descriptive population statistics (mean ± se and [95% confidence interval]) for genetic and chemical analyses of eight Cameroonian populations of the invasive ant P. megacephala.
Figure 4.
A: ΔK (a measurement of the rate of change in the structure likelihood function) values as a function of K, the number of putative supercolonies. In this case, K = 2. B: Graphical representation of the data set for the most likely K = 2, where each colour corresponds to a suggested cluster and each individual is represented by a vertical bar. The populations are indicated in the X-axis (AK: Akok; CA, CB and CC: Campo Forest Reserve; YM, YN, YS and YU: Yaoundé and its periphery). The Y-axis represents the probability for which an individual will be assigned to each cluster.
Figure 5.
Inference of the number of genetic clusters (K) in populations of P. megacephala var. 1 (i.e. urban) (top) and P. megacephala var. 2 (i.e. rainforest) (bottom).
A. ΔK (the standardized second order rate of change of ln P(X|K)) is plotted as a function of K. B. Proportional membership of P. megacephala workers to genetic clusters (K) for K = 2. Each vertical bar represents an individual whose estimated proportion of membership to either cluster (Y-axis) is indicated by the two different shades. Black lines separate individuals from different nests. Colours correspond to genetic clusters.
Table 2.
Statistics (mean ± se and [95% confidence interval]) for genetic and chemical analyses of the four supercolonies defined from behavioural assays.
Figure 6.
Chemical differentiation between the seven populations (A) and the four supercolonies (B).
The differentiation is illustrated by a factor map of the two first axes of the canonical discriminant analysis on the relative proportions of cuticular lipids. Functions one and two account for 98.7% and 99.7% of the total variability among populations and supercolonies, respectively. Only group centroids are plotted on figure A. P. megacephala var. 1: populations YM, YN, YS, YU and supercolonies SC-1, SC2; P. megacephala var. 2: populations AK, CA, CB, CC and supercolonies SC-3, SC-4.
Table 3.
Results of the hierarchical analyses of molecular variance (AMOVAs) evaluating the amount of genetic and chemical variance between and within groups.
Table 4.
Results of the hierarchical analyses of molecular variance (AMOVAs) evaluating the amount of genetic and chemical variance between and within supercolonies.
Figure 7.
Box-plot of the allelic richness estimated at eight microsatellites loci for samples collected from Cameroon (P. megacephala var. 1), South Africa [40] and Australia [34].
Lower case letters link groups that are not statistically distinguishable using post-hoc tests (Tukey's HSD) at α = 0.05.
Table 5.
Hierarchical analysis of molecular variance (AMOVA) with country as grouping factor (Cameroon, n = 304; Australia, n = 419; South Africa, n = 20).