Getting
a buzz from outbreeding
For the second time this month I am
discussing mechanisms that prevent inbreeding in self-fertile species. The species
in question here is
Paraboea rufescens,
a tropical Chinese member of the Gesneriaceae and studied by
Jaing-Yun Gao et
al.
(Yunnan, PR China, pp.
371–376). The plant shows two fascinating floral characters. First, it is
one of those species that possess mirror-image flowers (i.e. are
enantiostylic); in this species the two forms (left-styled, anthers deflected
to the right; right-styled, anthers deflected to the left) are borne within the
same inflorescence. Secondly, it is buzz-pollinated: the visiting bumblebees
rapidly vibrate their indirect flight muscles to induce pollen release from the
tip of the anther. The authors have carried out a detailed study of flower
morphology and pollination in this species. They showed first that pollination
by hand was much more efficient than open-pollination, possibly suggesting
sub-optimal numbers of pollinators. However, open-pollinated flowers, although
yielding fewer fruit per inflorescence, produced more seeds per fruit. Whether
this is related to resource allocation is not clear. Secondly, they confirmed
that
P. rufescens is self-fertile but
that autonomous self-fertilization in the absence of pollinators does not
occur. When flowers were self-pollinated by hand, there was clear evidence of
inbreeding depression in respect of the number of seeds set. Outbreeding is
therefore advantageous to this species. The relative positions of anthers and
style within an individual flower actually make self-pollination within one
flower very unlikely. Indeed, if the pollinator visits another flower of the
same enantiomorph, pollen transfer remains very unlikely because pollen is
present only on one side of the bee, the side away from the style. However,
inbreeding is not completely eliminated in
P.
rufescens because both flower enantiomorphs are present in the same
inflorescence. Because of the geometry, a bee moving from one enantiomorph to
another may well effect a self-pollination.
Professor J. A. Bryant
University of Exeter, UK
j.a.bryant{at}exeter.ac.uk
