1997-06-13_REPORT - M1981302 (17)

Home Browse Search

19841 SHEVIAK: SPIRANTHES 9 collaboration with P. M. Calling and W. F. Jennings and will be reported in a subsequent paper. Cytological data have provided the key to understanding the origin of the intermediates and have indicated their true status. In 1981, live plants were collected from the Colorado station by W. G. Gambill and W. F. Jennings and sent to me for study. The following year I collected additional material of the intermediates and of S. romanzo/jaaa in Colorado and at sites in Utah. In S. ' magnicarnporum, the single chromosome number 2n = 30 is present across the species'range(Sheviak, 1982).The situation in .S.rmnanznljana is more complex, with literature reports of 2n = 30 (Taylor & Mulligan, 1968), 2n = 60 (Heslop- Harrison, 1961), and 2n = 44 (Catling, 1980; Sheviak, 1982). During the present study, additional counts of 2n = 44 have been obtained from populations in the northeastern United States and Colorado; this number appears to be widespread in the species and occurs across wide areas of its range. The number 44 is unusual in the genus, and it is known from only one other species, the eastern S. iucida (H. H. Eaton) Ames: all other known numbers are based on 15 (Sheviak, 1982). The rarity of the number 44 is significant and propitious, for it helps to establish the nature of the intermediate plants. Counts of a number of intermediates from three populations in Colorado and Utah have been uniformly 2n = 74. Signifi- cantly, mciosis is generally regular, with the common formation of 37 bivalents. These cytological data indicate that the intermediate morphology of these plants does indeed reflect a hybrid origin; the plant is an amphiploid derived from hybridization of S. magnicamporum (2n = 30) and S. romanznliana (211 = 44) (Fig. 1). Reproduction of the plants appears to be strictly sexual. In cultivation, several plants from Colorado and Utah failed to set seed without pollination.Seeds present in the specimens examined are furthermore wholly monoembryonic, suggesting a sexual origin; polyembryony, which is indicative of adventitious embryony in the S. cernua complex (Swamy, 1948: Catling. 1982). has not been found. The discontinuous distribution of these plants reflects the paucity of suitably moist habitats in a prevailingly arid region. Populations are limited to relatively low elevations in mesic or wet meadows along permanent streams and about springs and the major desert lakes. These sites are commonly subject to inter- mittent and unpredictable inundation, and the plants often emerge front shallow water, even during anihesis. This niche and the plants' distribution suggest an origin in a Pleistocene pluvial period, when the region supported lush grasslands. Apparently S. magnicamporum was present, a conclusion supported by present disjunct populations in New Mexico. Under the cooler and wetter climate. S. romanzofana occurred at lower elevations than today, apparently sym- or para- patrically with S. magnicamporurn. Hybridization resulted in the production of the amphiploid, which successfully colonized extensive areas. As the climate became drier, the parental species and the amphiploid responded differently due to differing habitat requirements. The boreal S. romanzojliana retreated to higher, cooler and wetter areas: S. rnagnicamporum, requiring warm mesic sites, was extirpated front the region: and the amphiploid. combining adaptive features of both parents, persisted in warm wet situations. As aridity increased, the amphi- ploid became progressively more limited to scattered areas of permanent water. The amphiploid condition and associated regular meiosis clearly are responsible for the preservation of the intermediate features of these plants in widely scattered populations over a broad geographic area.This stability, together with an inferred ancient origin, wide distribution, and fertility, dictate that these plants be rec- ognized as a distinct species: