Localization of A. citrulli in watermelon seed. The expected
amplicon was produced with 100% of the extracts from pericarpand
pistil inoculated seed samples that were spiked with A. citrulli
DNA (data not shown) indicating no PCR inhibition. Overall, the
tissue colonization results obtained by the real-time PCR assay
were not significantly different to those obtained by plating on
semiselective agar media (P > 0.05) (Table 2). For pericarpinoculated
seed lots, the real-time PCR assay indicated that 5.3%
of the testae samples were contaminated with A. citrulli while
6.3% of testae samples from pistil-inoculated lots were positive
(Table 2). This difference was not statistically significant (P =
0.64). With regards to PE layer tissues, 97% of the samples from
pistil-inoculated seeds were contaminated with A. citrulli as compared
with 80.3% for seeds from pericarp-inoculated seeds, as
determined by the real time PCR assay (Table 2). For both seed
infestation types, the percentage of A. citrulli-positive PE layer
samples was significantly greater than the percentage of contaminated
testa samples (P = 0.002). Additionally, the percentage
of A. citrulli-positive PE layer samples was significantly greater
for seeds fropericarp-inoculated lots (80.3%) (P = 0.0001). The percentage of
A. citrulli-positive embryo samples for seeds from pistil-inoculated
lots (94%) was significantly greater than for seeds from
pericarp-inoculated lots (8.8%) (P = 0.0001). While there was no
significant difference (P = 0.43) between the percentage of A.
citrulli-positive embryos (94%) and A. citrulli-positive PE layer
samples (97%) for pistil-inoculated, for pericarp-infested seedsm pistil-inoculated lots (97%) than for those from