Page 17 - Volume 2,Issue 2, July 2009

© Benaki Phytopathological Institute
Honey bees and bumble bees as dispersers of a biofungicide 61
the three replicates mentioned above. For
each sample group, the means were aver-
aged and the respective standard devia-
tions were computed. The percentage of
flowers or insects in which the presence of
T. harzianum was detected was also calcu-
lated for each analysed sample group. In or-
der to compare the T. harzianum CFU den-
sity per flower between the flowers freely
exposed to pollinator’s visits and those ex-
posed solely to a single visit, non-paramet-
ric Mann-Whitney U tests were performed.
For the bumble bee assay Spearman correla-
tions between the number of CFU per visit-
ing bumble bee and the number of CFU per
flower, and also between the total length
of visits and the number of CFU per flower,
were computed.
The significance level used for all the
tests was 5%. Statistical tests were per-
formed with STATISTICA version 7.0 [Stat-
Soft, Inc. (2004)].
Results
1. Inoculum on pollinators
In both the honey bee and bumble bee
assays, all the individuals that were captured
while exiting the hives-mounted dispensers
carried the inoculum. The mean density of
inoculum transported per insect was 3.92 x
10 3 ± 1.73 x 103 CFU for honey bees and 7.19 x
10 4 ± 2.17 x 104 CFU for bumble bees.
Approximately 40% of the honey bees
that were captured during their foraging ac-
tivity carried remaining traces of T. harzianum,
with a density of 53.33 ± 95.43 CFU per bee.
On all bumble bees that were caught while
visiting flowers the presence of the biofun-
gicide was detected with a mean density of
6.59 x 103 ± 1.92 x 103 CFU per bumble bee.
2. Inoculum deposited on flowers
In both the assays, the presence of T. har-
zianum was not detected in the control flow-
ers. However, the fungus was detected in
other samples of plant material taken from
locations where the experiments took place.
In the honey bee assay, there was no sig-
nificant difference between the mean den-
sity of inoculum found on freely exposed
flowers and on flowers that were exposed to
a single visit (Z= 0.75; P=0.45), although the
latter showed a higher density of CFU (Table
1). The percentage of flowers having detect-
able biofungicide was almost the same for
both flower samples (Table 1).
In the bumble bee assay, the flowers ex-
posed to a single bumble bee visit showed
an average number of CFU per flower signif-
icantly lower than flowers that were freely
exposed to bee visits (Z= -7.28; P < 0.0001)
(Table 2). While 100% of the freely exposed
flowers contained T. harzianum, only 75% of
the flowers exposed to single visits had ves-
Table 1. Percentage of flowers in which the
presence of T. harzianum has been detected,
and means of the number of CFU per freely
exposed flowers and per flower exposed to
a single honey bee visit.
% of flowers
with
T. harzianum
CFU/flower
(mean ± s.d.)
Freely exposed
flowers to honey
bee visits
33.33% 26.27 ± 87.99 a
Flowers exposed
to a single
honey bee visit
34.78% 118.84 ± 222.44 a
Means within a column followed by the same letter do
not differ significantly.
Table 2. Percentage of flowers in which the
presence of T. harzianum has been detected,
and means of the number of CFU per freely
exposed flowers and per flower exposed to
a single bumble bee visit.
% of flowers
with
T. harzianum
CFU/flower
(mean ± s.d.)
Freely exposed
flowers to bum-
ble bee visits
100% 1.25 x 103 ± 8.97 x 102 b
Flowers
exposed to a
single bumble
bee visit
75% 123.44 ± 196.01 a
Means within a column followed by the same letter do
not differ significantly.