Tribe Pogonieae - The Pollination Biology of North American Orchids: North of Florida and Mexico - page 172

In a 2-year study, they observed no fruit production among emasculated and enclosed
flowers, indicating an absence of agamospermy (Table 9.2 ). Fruit set did not differ
among enclosed, emasculated plants that were artificially selfed or cross-pollinated
(both 82%), indicating that selfing resulted in no decrease in fruit set. Moreover, the
difference in fruit set between unmanipulated, enclosed (67%) and unenclosed, con-
trol plants (83%) was not statistically significant; autogamy was therefore sufficient to
explain the level of fruit set observed in natural populations. Autogamy is also reflected
in the development of relatively few, monocolored flowers per genet, small sepals and
small, unembellished petals (Table 9.1 ), an absence of nectar guides and floral scent,
a reduced rostellar flap, low levels of pollen and ovule production, and high levels of
fruit production (Ornduff 1969 ; Mehrhoff 1983 ).
Mehrhoff ( 1983 ) attributes the development of autogamy in I. medeoloides
chiefly to a scarcity of flowers within local populations and an associated reduction
in the probability of effective pollen transfer. The median population of
I. medeoloides produces only one or two widely separated flowers (Mehrhoff 1980 ).
Under such circumstances, autogamy may provide the best chance for seed genera-
tion, insuring some seeds are formed even if only one flower is developed (Baker
1955 ). At the same time, the mechanism of self-pollination in this orchid might not
exclude the possibility of insect visitation. Such visits could, however, prove dis-
ruptive. A vector that removed pollen and failed to transport it to another flower
would seriously deplete or eliminate the reproductive output of a population with
only one or two flowers. Selection might therefore operate to minimize such floral
visits by reducing the visual appeal of the flowers (Mehrhoff 1983 ).
Pollinators and Pollination Mechanisms
I. verticillata is pollinated by solitary, short-tongued bees in the Andrenidae
( Andrena ceanothi Viereck, A. miranda Smith, and A. nasonii Robertson), Apidae
( Nomada sayi Robertson and N. sp.), and Halictidae ( Augochlora pura (Fig. 4.2)
and Lasioglossum cressonii (Robertson) [as Dialictus creasonii (Robertson)]
(Mehrhoff 1983 ). Mehrhoff ( 1983 ) believes that three other morphologically simi-
lar species ( Andrena imitatrix Cresson, A. miserabilis Cresson, and Nomada superba
Cresson), seen on flowers but not observed to remove pollen, may also be pollina-
tors of I. verticillata . All of the species identified as pollinators have a wide geo-
graphic range and are known to visit a large variety of spring and summer flowers
(Mitchell 1960 ). Widespread pollinators in combination with vegetative reproduc-
tion may facilitate the establishment of new populations.
Mehrhoff ( 1983 ) described the mechanics of pollination. The pollinator, search-
ing for nectar, lands on the labellum and enters the floral tube (Fig. 9.1a, c ). It
squeezes beneath the anther and rostellar flap to the stigma (Fig. 9.1b ), which depos-
its a sticky secretion on the dorsal surface of its thorax. As it backs out under the
rostellar flap the anther cap catches on the posterior part of its upper thorax, is
forced open, and dumps a mass of mealy pollen onto the stigmatic secretion. If the
pollen is transported to another flower and the process repeated, the pollen mass
may be deposited on the stigma.