Subtribe Spiranthinae - The Pollination Biology of North American Orchids: North of Florida and Mexico - page 41

spaced inflorescences (e.g., Rathcke 1983 ). However, when the inflorescences are
nonclonal, competition between plants could also be higher and the number of
flowers visited per plant might decrease (Klinkhamer and De Jong 1990 ). Duffy and
Stout ( 2008 ), in a study in western Ireland, found that intraspecific competition
among inflorescences of S. romanzoffiana increased and visitation rates per inflo-
rescence decreased at high densities. At the same time, they reported a positive
relationship between the number of pollinator visits to S. romanzoffiana and total
floral density in mixed patches of this and other rewarding plants, the so-called
magnet species effect.
Many Spiranthes species congregate in patches, and outcrossing may be restricted
because pollen transport is usually undirected and occurs chiefly among densely
spaced plants that are likely to be closely related (Ackerman 1975 ; Larson and
Larson 1987 ). However, Larson and Larson ( 1987, 1990 ) found that when bees fin-
ished foraging on S. romanzoffiana at their study site, they consistently moved on to
other scattered patches of this orchid in the adjacent forest. This, along with their
clearly directional flight path within plots, suggests that the bees may trapline through
the forest, effecting some level of gene flow between patches (Heinrich 1976 ).
Some parallels between S. cernua in northeastern North America and G. tessel-
lata are noteworthy. Both are polyploid (apparently of hybrid origin) and sympatric
with two closely related diploid species. Like G. tessellata , S. cernua blooms before
the diploids, and this phenological separation may be a statistically significant
obstacle to hybridization (Catling and Brown 1983 ). Northeastern North American
species of Spiranthes (14 of 15 taxa) also resemble Goodyera in column structure
as well as mode of viscidia attachment, and both genera are protandrous with
sequential acropetal flowering (Darwin 1862 ; Ames 1921 ; Hagerup 1952 ; Kipping
1971 ; Ackerman 1975 ; Catling 1980b, 1983c ).
Fruiting Success and Limiting Factors
Extensive studies are available on the reproduction of Nebraska populations of
S. cernua . According to Schmidt ( 1987 ), artificially crossed plants produced a sig-
nificantly higher proportion of monoembryonic seeds than unmanipulated (control)
plants. This suggests that pollinator limitation may influence the proportion of
agamospermic seeds produced. Schmidt ( 1987 ) also found that the frequency of
polyembryonic and monoembryonic seeds did not differ significantly for emascu-
lated, bagged, and unbagged plants. Presumably, if there was no pollinator limitation,
a higher proportion of monoembryonic seeds would be expected in the unbagged
treatment.
Sheviak ( 1982 ), Catling ( 1980a, 1983a, c ), and Schmidt and Antlfinger ( 1992 )
saw few pollinators visiting the flowers, and few pollinaria were removed. Factors
that might limit pollinator visits have yet to be identified. This species might simply
be less attractive to pollinators than other fall-flowering taxa (Anderson and
Schelfhout 1980 ; Kaul and Rolfsmeier 1987 ) or fragmentation of the prairie may