Characterizing larval breeding sites: microbial density
We examined the microbial profile in a subset of larval breeding sites,
as previous studies have shown that the microbiome, particularly
bacteria, affect Ae. aegypti oviposition choices (Arbaoui &
Chua, 2014; Ponnusamy et al., 2015) as well as larval development
(Correa et al., 2018; Dickson et al., 2017; Souza et al., 2019). In
brief, we collected 15 mL (in La Lopé) or 50 mL (in Rabai) water samples
from each larval breeding site. To first examine the total microbial
density, we fixed an aliquot of each water sample with formaldehyde and
stained it with DAPI (4′,6-diamidino-2-phenylindole). The number of
microbial cells was counted using hemocytometers under a fluorescence
microscope and back-calculated to its original density.
Densities were log-transformed before statistical analysis. We compared
the microbial density among larval breeding site groups in La Lopé with
the Kruskal–Wallis test and post hoc pairwise Wilcoxon rank-sum tests.
The distribution of data in Rabai samples did not violate parametric
test assumptions, so we performed the comparisons using analysis of
variance (ANOVA) and post hoc Tukey tests.