Violaceae - Violets and Pansies
Violas often
produce two types of flower: the chasmogamous flowers, as shown
above, meaning flowers that open to disperse pollen and then
later in the season it may produce cleistogamous flowers, flowers which
do not open but instead pollinate themselves. The cleistogamous
flowers set the majority of seed, but the effort put into
chasmogamy is not wasted as even a small amount of
cross-pollination benefits the gene pool. The chasmogamous
flowers rely on insect pollinators, the intention being for
pollen collected from one plant to arrive at a receptive stigma
on another plant, thereby affecting cross-pollination.
The chasmogamous flowers actually have a dual strategy. Five to
six days before anthesis (opening of the
petals) the sepals open out and the enlarging petals take over
their role in protecting the developing reproductive parts in
the bud. There are five sepals: two anterior (upper) sepals, one
posterior (lower) and two posteriolateral sepals. The corolla, or flower tube,
consists of five distinct petals: two posterior (forming the
roof and sides of the bud), two lateral (forming the floor of
the bud) and one anterior petal (rolled-up like a scroll inside
the bud). Pollinating bumblebees have good ultraviolet (UV)
vision and see flowers quite differently to humans. The petals
reflect UV light, accept at
their bases which absorb UV light, so the middle of the corolla
probably appears as a black tunnel to the bumblebee. Anterior
and lateral petals have an array of hairs which close the
entrance to the pollen. The lateral hairs reflect UV and so
appear bright. The large bumblebee can force its way past the
hairs and can access a flower which is even in the early stages
of anthesis.
The spur is an extension of the
petals at the back of the corolla which stores nectar. The
intention is that insects pushing deep into the flower to reach
the nectar brush past the anthers, picking up pollen, or past
the stigma, depositing pollen. The adaxial surfaces of the
petals (the upper surfaces nearest the floral axis) are covered
in conical epithelial cells. This is a characteristic of many
petals and the conical nature of the cells performs several
functions. It gives the petals a matt texture and possibly makes
it easier for landing insects to grip the surface. It also gives
the petals anti-wetting properties -
rainwater rolls off, cleaning the petal without obscuring its
visual display which is designed to attract pollinators (a
property called superhydrophobicity). In many flowers, the
conical cells also secrete aromatics. The cuticle over the
papillae is typically ridged and this pattern can cause
diffraction of the light, giving the petals iridescence, enhancing their
visual display.
The female organ, the carpel, consists of the ovary containing
ovules, each ovule containing a developing egg cell, and the
style and stigma. The style is a stalk holding the
stigma in position. The stigma is the receptive surface which
must contact pollen if the flower is to be pollinated. In Viola, the style is hinged
at the base, at a kink or 'knee' joint. The stigma in Viola is cup-shaped and
contains an opening (the stigmatal
aperture).
This aperture opens into a stigma cavity which extends into the
style where it is filled with mucilage. When an insect brushes
past the style and deflects it, the canal is squeezed shut and a
blob of mucilage protrudes from the stigma aperture. The knee
joint ensures the style springs back into shape and the mucilage
goes back inside. There are two hypothesised functions of this
mechanism: it may touch the insect's back and make it sticky
ready to pick-up pollen; alternatively it may collect pollen
from the insect's back and then draw it into the stigma where
pollination can commence. This first mechanism is a specialised
mechanism intended for specialised pollinators, such as the
large bumblebee.
The 5 stamens are essentially filamentless (or have very short
filaments) and each contains two microsporangia each of which
dehisces (splits open when ripe and dry) along its own suture
from top to bottom. Each bears a distal appendage (on top of the
anther) in the form of a triangular projection, coloured yellow
or bright orange. These are sometimes called connective
appendages, I shall call them conal
appendages,
since the five close together in a cone around the style. The
two anterior stamens bear am additional appendage at their base
- a nectariferous
appendage
(filament
appendage) which projects into the spur into which they secrete
nectar (which apparently oozes out of modified stomata on the
ends of the nectariferous appendages).
Anthesis proceeds over several days. The petals curl apart
throughout. Initially the only structures visible from outside
the flower are the style apex and the conal appendages. One day
after anthesis the posterior anther dehisces releasing pollen.
On the second day, the two posterio-lateral anthers dehisce. The
two remaining anterior anthers dehisce around day 4 to 6. The
stigma is receptive throughout. The cone of conal appendages
serves to trap pollen ready for a specialist pollinator to brush
past. After day 4, however, the conal appendages separate and
pollen falls onto the anterior petal where it can be reached by
a wider variety of insects. Thus, Viola adopts a three-part
strategy:
first its chasmogamous flowers are primed for specialist
pollinators, then later they open up for more general
pollinators, thus increasing their odds of pollination, and
thirdly they produce cleistogamous flowers which self-pollinate.
The petals continue to open-out wider throughout anthesis and
the distal part of the pedicel (flower stalk) slowly
curves over into a hook-shape. This ensures that the opening
petals present the maximum surface facing forwards to attract
pollinators. Additionally, it positions the flower by day four
in such a way that pollen will fall out onto the anterior petal
when the anther cone opens out. Eventually the flower senesces. This is not triggered
by pollination in Viola, and usually occurs
after day 7 post-anthesis (sometimes day 14 to 15 in bright
sunlight).
On the various floral parts there may be various types of hairs
and papillae, which help to characterise the various species,
but which have largely unknown functions. For example, a lower
lip of papillae is generally present on the stigma beneath the
aperture. Clearly there is scope for further research in
understanding these remarkable flowers!
Above and below: Common Dog-Violet (Viola riviniana) is perennial and can be distinguished by its spur, which is curved upward, blunt tipped and notched at the tip and paler than the petals. it is found in woods, hedge-banks, grasslands and among rocks. each of he 5 sepals of Viola bears an appendage or lobe beneath its point of insertion. These squarish appendages can be seen in the photograph below and their length can aid species identification. The main blade of the sepal is pointed. This is one of the early Spring flowers, flowering from March to May in the British Isles (and may flower again from August to October, but often without petals).
Another useful taxonomic feature concerns the stipules. In V. riviniana they have hair-like teeth along their margins. In V. reichenbachiana the stipules are narrower, as are the leaves. Below: leaves of V. riviniana.
Below: Early Dog-violet (Viola reichenbachiana). This specimen occurred in the same wood as the Common Dog-Violet above, which is not unusual. However, this species prefers dry more chalky soils in woodlands and on hedge-banks. The upper petals tend to resemble 'rabbit ears' and the spur is straighter, darker than the petals and not notched. The nectar-guide veins on the lower petal are also much less branched than in Viola riviniana. The sepal appendages of V. reichenbachiana are shorter (< 1.5 mm) than those of V. riviniana (> 1.5 mm) and are generally very short, as in the specimen below. V. rivinana is found across Eurasia and nw Africa and has also become established on the pacific coast of the USA and has also been introduced in Australia.
The specimen below is somewhat inconsistent.The spur is darker than the petals, straight but notched at the tip. The nectar-guide veins are not very branched, but the broad spreading upper petals suggest V. riviniana. This could quite possibly be the hybrid V. x bavarica = V. riviniana x V. reichenbachiana which is largely sterile. The sepal appendages of the hybrid are said to be intermediate in length between the two parents (i.e. about 1.5 mm long). The hybrid is, however, not common despite the frequent cohabitation of both parents (See Stace: New Flora of the British Isles) and one has to be cautious when determining it. Since the hybrid is partially fertile, I would not rule out back-crosses with either parent.
Further Reading