rarely been questioned. Third, general knowledge about genetic diversity of many grass species native to North Little bluestem [Schizachyrium scoparium (Michx.) Nash] is one America is largely lacking (Huff et al., 1998; Larson et of the most widespread native grasses in North America. Little is al., 2001a), as few molecular diversity studies on these known about the genetic diversity of this species in natural populations and in seed collections. The amplified fragment length polymorphism species have been made. Thus, informative diversity stud-(AFLP) technique was applied to assess the comparative genetic ies on native grasses with molecular techniques are wardiversity of six natural populations of little bluestem in Manitoba and ranted. Saskatchewan and their corresponding seed collections. Five AFLP Little bluestem is a native, warm-season, highly outprimer pairs were employed to screen a total of 180 samples representcrossing, tetraploid, perennial bunchgrass (Archer and ing about 15 tillers per population and 15 seeds per collection, and Bunch, 1953; Gould, 1956) that inhabits a wide ecologi-158 polymorphic AFLP bands were scored for each sample. Analyses cal range in North America where it is used for reclamaof these scored bands revealed that Ͼ91% of the total AFLP variation tion and ground cover (Hitchcock, 1950). While breedwas present within the natural populations and within the seed collecing efforts to improve little bluestem seed production tions. The among-population and among-collection variation compohave been made in

Inadequate seed supplies of adapted blue grama [Bouteloua gracilis (H.B.K.) Lag. ex Steud.] and little bluestem [Schizachyrium scoparium (Michx.) Nash.] cultivars limits the use of these two species in the northern Great Plains of Western Canada. This study examined variation in seed yield traits of 11 blue grama and 14 little bluestem plant collections obtained from southern Manitoba, Canada to facilitate development of improved germplasm of these species. Measurements of seed yield traits including harvested air‐dried biomass, culm number, seedhead number, seed yield, caryopsis weight, and kernel index were taken from randomized complete block spaced‐plant nurseries. Tests were conducted for 2 yr at Winnipeg and 1 yr at Carman, MB, Canada. Collections within both species differed significantly for all traits. Significant collection × year (Winnipeg) and collection × location interactions were present for biomass and seed yield, and many of the seed yield‐component traits. While significant, the interactions resulted in few changes in rank among collections. Generally, the most northern collection in both species showed earlier anthesis, produced less biomass, and had lower seed yield than more southern collections. These findings indicate that indigenous plant collections of blue grama and little bluestem show high levels of genetic diversity for biomass, seed yield, and seed yield components, which provide opportunity for development of adapted cultivars with enhanced forage and seed production capability for the region.

tively expensive for large-scale plantings. Releases of a commercial Minnesota ecotype in 1995 and the USDA Blue grama [Bouteloua gracilis (Willd. ex Kunth) Lag. ex Griffiths]ecotype Bad River originating from South Dakota in is one of the most widespread native grasses in western North America.1996 have improved the seed supply of blue grama. As Several blue grama seed sources are currently used for rangeland these ecotypes represented germplasm selected from seeding, but little is known about the genetic diversity of these seed sources. Amplified fragment length polymorphism (AFLP) technique single native sites, however, concern has been raised was applied to compare the genetic diversity among four blue grama that these plant materials may lack the genetic diversity seed sources (a precultivar germplasm [balanced multisite composite, to maintain adaptation in dynamic, nonlocal environ-BMSC], the ecotype Bad River, a Minnesota ecotype, and a native ments (Knapp and Rice, 1996; Roundy, 1999; Larson et Manitoba seed collection) and to assess the genetic shift during two al., 2000). To address this concern, a BMSC was develgenerations of BMSC seed multiplication. Germplasm BMSC was a oped in 2000 from 495 live plants collected from 11 balanced multisite composite of 99 clones selected from 495 live plants sites across Manitoba (Phan, 2000). While the BMSC collected from 11 sites across Manitoba. Six AFLP primer pairs were is expected to maintain higher genetic diversity than the employed to screen a total of 176 individual plants sampled from both other seed sources, no comparison of genetic diversity the first three generations of BMSC and the other three seed sources has been made among them. Even though the BMSC and 167 polymorphic AFLP bands were scored for each plant. Large AFLP variation was observed within the four seed sources. Greater had been shown to capture most of the RAPD variations AFLP variation was detected in the BMSC than Bad River, Minnesota present in its source populations (Phan et al., 2003), it ecotype, and the Manitoba native harvest. No genetic shift in the still remains unknown if genetic shift has occurred dur-BMSC was found across the two seed multiplications. These results ing successive generations of the BMSC seed multipliindicate a balanced composite of multisite blue grama germplasm cation.can maintain high genetic diversity with little genetic shift in a few

used for reclamation and ground cover (Hitchcock, 1950). These adaptive features have stimulated some interest Blue grama (Bouteloua gracilis H.B.K. Lag. ex Steud.) is one of in utilization and cultivar development of this species the most widespread native grasses in western North America. Little (Smith and Phan, 1999). Two blue grama populations, is known about the genetic diversity of this species and the genetic one a balanced multi-site composite (BMSC) and the shift in selected populations. Random amplified polymorphic DNA (RAPD) markers were used to assess the genetic variations of blue other a mass-selected composite (MSC), originated from grama plants in 11 natural source populations (SPs) in southern Mani-11 natural source populations (SPs) in southern Manitoba and two selected populations. Population 'BMSC' was a balanced toba (Phan, 2000). Populations BMSC and MSC were multisite composite of 99 clones selected for higher seed yield from selected for increased seed yield (Phan and Smith, 2000). the 11 SPs. Population 'MSC' was a mass-selected composite of 25 Studies that compare genetic diversity and shift beclones with higher seed yield from eight SPs. Twelve RAPD primers tween the selected and source populations are lacking. were used to assay 108 original propagules from the 11 SPs and 96 Genetic shifts attributed to selection have been reported seedlings from each selected population. A total of 69 polymorphic in morphological studies of oat (Avena sativa L.) (Rod-RAPD bands were detected. No unique RAPD bands were found gers et al., 1983), maize (Zea mays L.) (Coors and Marfor any SPs and 97.8% of the total RAPD variation was detected dones, 1989; Helms et al., 1989; Stojsin and Kannenberg, within SPs. Variation in RAPD markers was not associated with geographical distances. Highly significant changes in RAPD band 1994), and Pensacola bahiagrass (Paspalum notatum frequency from their SPs were detected in both selected populations, Flugge var. saurae Parodi) (Werner and Burton, 1991; but only MSC displayed the fixation of polymorphic bands. The esti- Pedreira and Brown, 1996). mated genetic shifts were small, 0.6% for BMSC and 1.9% for MSC.

No abstract