in the past [176, 177]. In Brassica rapa the heritability of floral scent is correlated
with the pleiotropic responses of various plant traits [178]. These latter authors
observed the alteration of the entire floral scent bouquet after only three generations
in response to artificial selection of a single compound.
A study of the divergent evolution in plants of Brassica rapa compelled by
pollinators was conducted, in which the plants were developed in a phytotron
under standardized conditions [179]. The replicates were preserved as isolated
lines during 11 generations to be able to evaluate independent, repeatable evolutionary changes. In this study they used three pollinator treatments: bumblebees,
hoverflies, and hand pollination. Pollination was performed 23 days after seeding
out in a flight cage in the greenhouse under standardized conditions, with
bumblebees and hoverflies. Pollinators were let to forage on fast cycling B. rapa
plants of the control group of the respective generation. Before pollination, pollinators were hungry. Five pollinators were added individually and sequentially; each
insect was allowed to visit a maximum of three different plants before removing
them from the cage. Floral scent was measured before pollination 19–21 days
after sowing out, and the quantification of volatiles was conducted by gas
chromatography with mass-selective detection. Hoverfly-pollinated plants showed
a significant decrease in the emission of four scent compounds: methyl salicylate,
p-anisaldehyde and indole, and benzyl nitrile. In bumblebee-pollinated plants, the
total amount of scent emission per flower almost doubled, as more than half of
the analyzed volatiles showed increased emission.
In addition, plants with floral scent may have different pollination syndromes
within the same group or may be related to a larger type of syndrome. Studies
addressing these relationships are very informative, providing excellent opportunities to determine whether pollinator changes are correlated with parallel variations in
the chemistry of floral bouquets [152].
3.5
Floral Scent Production in Deceptive Plants
Just as there are animals that collect resources in flowers without an
effective pollination, there are also plants that attract pollinators without offering
any sort of resource, a phenomenon known as deceptive pollination [180–183].
Floral deception occurs in more than 30 families of angiosperms (e.g.,
Aristolochiaceae, Apocynaceae, Araceae, Berberidaceae, Bignoniaceae, Iridaceae);
however, Orchidaceae is the family with the highest number of species with
this condition [182, 184, 185]. In deceptive pollination the flowers mimic floral
signals of rewarding plants (food deception) or mating signals of receptive females
(sexual deception) to attract pollinators [186]. These signals may include visual
and olfactory cues. Therefore, these flowers do not synthesize nectar, oil, or resins,
while scents may still be produced [183].
The production of floral scents by osmophores in deceptive plants may be
essential to an efficient pollination, especially when other visual and/or morphological cues to attract floral visitors are lacking [187–191]. Several experiments
28 Diversity of Floral Glands and Their Secretions in Pollinator Attraction
727
with the pleiotropic responses of various plant traits [178]. These latter authors
observed the alteration of the entire floral scent bouquet after only three generations
in response to artificial selection of a single compound.
A study of the divergent evolution in plants of Brassica rapa compelled by
pollinators was conducted, in which the plants were developed in a phytotron
under standardized conditions [179]. The replicates were preserved as isolated
lines during 11 generations to be able to evaluate independent, repeatable evolutionary changes. In this study they used three pollinator treatments: bumblebees,
hoverflies, and hand pollination. Pollination was performed 23 days after seeding
out in a flight cage in the greenhouse under standardized conditions, with
bumblebees and hoverflies. Pollinators were let to forage on fast cycling B. rapa
plants of the control group of the respective generation. Before pollination, pollinators were hungry. Five pollinators were added individually and sequentially; each
insect was allowed to visit a maximum of three different plants before removing
them from the cage. Floral scent was measured before pollination 19–21 days
after sowing out, and the quantification of volatiles was conducted by gas
chromatography with mass-selective detection. Hoverfly-pollinated plants showed
a significant decrease in the emission of four scent compounds: methyl salicylate,
p-anisaldehyde and indole, and benzyl nitrile. In bumblebee-pollinated plants, the
total amount of scent emission per flower almost doubled, as more than half of
the analyzed volatiles showed increased emission.
In addition, plants with floral scent may have different pollination syndromes
within the same group or may be related to a larger type of syndrome. Studies
addressing these relationships are very informative, providing excellent opportunities to determine whether pollinator changes are correlated with parallel variations in
the chemistry of floral bouquets [152].
3.5
Floral Scent Production in Deceptive Plants
Just as there are animals that collect resources in flowers without an
effective pollination, there are also plants that attract pollinators without offering
any sort of resource, a phenomenon known as deceptive pollination [180–183].
Floral deception occurs in more than 30 families of angiosperms (e.g.,
Aristolochiaceae, Apocynaceae, Araceae, Berberidaceae, Bignoniaceae, Iridaceae);
however, Orchidaceae is the family with the highest number of species with
this condition [182, 184, 185]. In deceptive pollination the flowers mimic floral
signals of rewarding plants (food deception) or mating signals of receptive females
(sexual deception) to attract pollinators [186]. These signals may include visual
and olfactory cues. Therefore, these flowers do not synthesize nectar, oil, or resins,
while scents may still be produced [183].
The production of floral scents by osmophores in deceptive plants may be
essential to an efficient pollination, especially when other visual and/or morphological cues to attract floral visitors are lacking [187–191]. Several experiments
28 Diversity of Floral Glands and Their Secretions in Pollinator Attraction
727