Tribe Calypsoeae - The Pollination Biology of North American Orchids: North of Florida and Mexico - page 131

and self-pollination occurred in only 4.7% of the flowers. In another 34.3%, the
pollinia on the stigma were not attached to the rostellum, and the flowers were con-
sidered insect pollinated. These observations in conjunction with the morphology of
the flower imply that this species is chiefly adapted to reproduction by outcrossing.
No pollinators have been identified for chasmogamous plants of C. odontorhiza.
However, the absence of seed production in bagging experiments on both unma-
nipulated and emasculated flowers indicates that pollinators are necessary.
Fruiting Success and Limiting Factors
Reported high levels of natural capsule production in populations of C. trifida and
cleistogamous C. odontorhiza correlate with the establishment of nearly obligate to
obligate autogamy in these taxa (von Kirchner 1922a, b ; Summerhayes 1951 ; Luer
1975 ; Catling 1983 ; Freudenstein 1997 ).
Although levels of self-compatibility in C. maculata have not been directly
examined experimentally, Catling ( 1983 ) reported that seed set was abundant among
naturally self-pollinating flowers in southern Ontario. Kipping ( 1971 ) found natural
fruit set in 50.5% of the flowers in El Dorado County and 59% in Marin County.
This is very close to his experimental results with unmanipulated, caged plants
noted above (51% and 64%).
Although chasmogamous plants of C. odontorhiza are fully self-compatible, it is
not unusual to find individuals in natural populations in which 75% of the capsules
remain undeveloped (Catling 1983 ). Consistent with an absence of autogamy, this
result, when compared with the 100% seed set obtained in experimental crosses,
also suggests that pollinator service may be a limiting factor in the reproduction of
this orchid.
Experimental crosses were not performed, but Freudenstein's ( 1997 ) report of
natural insect pollination in about 34% of the flowers in a Michigan population of
C. striata also suggests the possibility of pollinator limitation. Additional research
is needed to further estimate levels of fruit set and seed viability in all Corallorhiza
species and to evaluate the principal factors affecting these levels.
Additional Species and Varieties of Corallorhiza
No data are available on the reproductive biology of C. striata var. vreelandii
(Rydberg) L. O. Williams, a taxon scattered from the Dakotas and New Mexico to
Washington and California. However, the perianth segments, with the exception of
the lip, tend to be connivent suggesting either adaptation to a specific pollinator or
autogamy (Freudenstein 1997 ).
There are also no published accounts of pollination for C. wisteriana S. W.
Conrad and C. mertensiana Bongard. The former is distributed in the east from
Nebraska and Texas to Pennsylvania and Florida and in the west from Montana to
New Mexico and Arizona, the latter, from southeastern Alaska and northern