chromosome numbers ranging from 2n ϭ 2x ϭ 18 to 2n ϭ 12x ϭ 108 (Nielson, 1944;McMillan and Weiler, Greater knowledge of the magnitude of genetic variability for bio- Henry and Taylor, 1989). Switchgrass is classified mass yield and yield components in switchgrass, Panicum virgatum L., and relationships among the biomass yield component traits would into upland and lowland ecotypes based on morphology facilitate the breeding improvement of the species. Accordingly, we and habitat preference (Porter, 1966). All confirmed conducted two replicated experiments to assess genetic variation for lowland ecotypes have been tetraploids and most upland biomass yield and yield components and quantify relationships among ecotypes are octoploids (Hopkins et al., 1996). those traits in different switchgrass populations. In Exp. 1, 228 half-sib Information on the variability and associations of biofamilies from SU C 3 and 261 from NU C 3 populations were evaluated at mass yield and yield components in switchgrass is lim-Perkins, OK, while 278 half-sib families from the SL C 0 population ited. Talbert et al. (1983) reported narrow-sense heriwere evaluated at Stillwater, OK. Exp. 2 comprised 11 lowland switchtability estimates of 0.25 and 0.59 based on individual grass populations tested at Chickasha and Perkins, OK, in 1998. Sub-half-sib progeny and half-sib progeny means, respecstantial differences (P Ͻ 0.01) in biomass yield per plant existed among half-sib families within the respective SU C 3 , NU C 3 , and SL C 0 popula-

Campbell and Lipps, 1998;Yang et al., 1999). Resistance expression often differs among envi-The development of wheat (Triticum aestivum L.) cultivars resis-

No information is available on the effects of different biomass yield environments on selection efficiency in switchgrass (Panicum virgatum L.) breeding improvement. This study was conducted to assess the effects of high-and lowbiomass yield environments (HYE and LYE, respectively) on recurrent selection for general combining ability (RSGCA) in a lowland population of switchgrass (NL-94). The top 14 of 65 NL-94 C 0 parent plants were selected based on biomass yield of half-sib (HS) progeny tested for one post-establishment year under HYE and LYE conditions. Nine of the 14 C 0 parent plants were the same based on HS performance under HYE and LYE. Selected plants were intercrossed to produce NL-94 HYE and NL-94 LYE C 1 populations. One hundred and twenty-five HS C 1 progeny families (60 NL-94 HYE and 65 NL-94 LYE) were evaluated for biomass yield for 3 years (2002)(2003)(2004) under HYE and LYE conditions. The HYE produced about 2.5 times higher biomass yields than the LYE in both C 0 and C 1 HS progeny tests. Estimated additive genetic variance and predicted gains from selection (DG) were high in the C 1 populations indicating that RSGCA should achieve higher biomass yields. Mean biomass yields of C 1 HS families originating from the LYE protocol were significantly higher than those of families originating from the HYE protocol in both HYE and LYE performance tests, suggesting greater selection response under LYE in the C 0 population. The estimates of narrow-sense heritability ( h 2 n ) and DG from the C 1 populations indicate that positive response to selection for biomass yield is possible in subsequent cycles of selection under either HYE or LYE, with a possible small advantage for HYE.