Tribe Calypsoeae - The Pollination Biology of North American Orchids: North of Florida and Mexico - page 114

Banff area were visited, and only ten (1%) were pollinated. These results correspond
closely to the levels that Mosquin ( 1970 ) reported for the same area. He found that
200 (12%) of 1,654 examined flowers had pollinaria missing, disturbed, or depos-
ited on the stigma. Proctor and Harder ( 1995 ) obtained somewhat different results
in a study of this variety from the Rocky Mountain foothills of Alberta. Here, 212
(41%) of 520 examined flowers were visited, 72 (14%) had only pollinaria removed,
11 (2%) had only pollinia deposited, and 129 (25%) had pollinia both removed and
deposited. In Idaho, Lehmberg ( 2002 ) found pollinia removed from 14 (56%) of 25
plants. In variety occidentalis , Ackerman ( 1981 ) reported that 671 (53%) of 1,273
flowers at one site in Humboldt County were visited and 141 (11%) were pollinated,
and Krell ( 1977 ) observed that 117 (41%) of 287 flowers were visited in northwest-
ern Idaho and 63 (22%) were pollinated. At the latter site, 9% of the total number of
flowers produced fruits as compared to 11-34% over five sites in Humboldt County
(Ackerman 1981 ) and 34% in Marin County, California (Kipping 1971 ). Curiously,
there is very little comparable fruiting data for variety americana , although Mousley
( 1924 ) reported that 6% or 3 of 50 plants developed capsules near Hartley, Quebec,
and Lehmberg ( 2002 ) found enlarged ovaries in 8% or 2 of 25 plants in Idaho.
Average fruit set in variety bulbosa from northern Sweden ranged from 21 to 48%,
relatively high compared to the North American varieties, a difference that
Alexandersson and Agren ( 1996 ) believe may be at least partly related to differ-
ences in when and how the data were collected.
In addition to quickly acquired ability of the bee to recognize unrewarding flow-
ers, the levels of male and female success are reduced because pollinaria are not
always removed from the flower or transported pollinia deposited on the stigma
when bees do visit the flowers (Ackerman 1981 ) . This is true despite the adaptation
of the flower for a specific type of pollinator. Boyden ( 1982 ), for example, found
that only 2 out of 60 bees that carried something other than just one pollinarium in
the Banff area carried one pollinarium and one viscidium, a ratio that would reflect
the possibility of a highly efficient pollen transfer mechanism. Forty-four bees bore
only viscida. If it is assumed that the pollinia, once acquired, were removed during
ensuing flower visits, all 44 of these bees deposited pollinia without removing a
second pollinarium.
Moreover, as already suggested by the ratio of pollinaria removed to pollinia
deposited, Boyden ( 1982 ) found that once acquired the transfer of pollinia from bee
to flower is not highly efficient. A dozen bees bore two intact pollinaria indicating
visits to at least one flower with no pollinia deposition. Two bees bore three intact
pollinaria. Mosquin ( 1970 ), Ackerman (in Krell 1977 ), and Krell ( 1977 ) also
observed bees with more than one pollinarium attached to their thorax. Kipping
( 1971 ) captured specimens of Bombus vosnesenskii with attached pollinaria at his
study site in Marin County and placed them in a terrarium with newly opened flow-
ers of variety occidentalis . The bees successfully extracted new pollinaria, but the
original pollinia were not removed by the stigmas. He suggested that the stigmas
might not be very receptive prior to the removal of the pollinaria, a mechanism that
might prevent self-pollination. Krell ( 1977 ), however, found no evidence in support
of protandry in this variety.