Cynipid gall abundance was sampled across the 20 selected provenances in 2008 and 2009.
Therefore, the following variables were selected: geographical provenance (Italy versus other countries), age, calendar year at pregnancy, CD4+ T-cell count at pregnancy, and presence/type of cART at the time of HIV RNA testing (no therapy versus two antiretroviral drugs versus at least three drugs in combination, i.e., HAART regimens).
For long-lived species like most trees, where strong GE interactions exist, and in areas where large changes in climate are predicted, provenances likely to be the most suitable in the future should be selected (predictive provenancing, B in Fig. 2).
We selected 20 provenances whose origins span 15° of latitude and 34° of longitude, incorporating the greatest available geographic separation and source environmental variation (Fig. 1, Table S1 in Data S1), and included the closest provenance to the trial site (Forêt de Bercé, approximately 50 km from La Petite Charnie).
They were rather selected within provenances that also included families with the highest breeding values for height, because the objective was to set up a control population for the small case population with the same geographic background, so as to neutralize the possible confounding effect of different geographic backgrounds between small case and large case populations.
The select provenances can then be the starting point for vegetative propagation, leading to the establishment of clones (Akinnifesi et al. 2004).
The hypothesis also predicts that provenances selected to match conditions furthest in the future should be least well matched to current conditions and so perform least well in the present.
We test this hypothesis using multiple performance measures for 20 provenances planted at the La Petite Charnie provenance trial, selected to maximize the available range in source latitude and longitude as well as the mismatch in climate between the provenance trial and the provenance origin.
Four sites for the pre-Andean provenance and five sites for the coastal provenance (i.e. intra-provenance level) were selected for seed collection in March 2000.
In the present study, the individuals in the small control population were not selected randomly among the provenances of the large control population, as expected under genetic drift.
