MATERIALS AND METHODS Grazing animals prefer some plants to others. These choices are Cultivars and Experimental Design likely related to physical and chemical factors such as energy-dense The eight endophyte-free tall fescue cultivars included in carbohydrates contained in plants. This study quantified the nonstructhis study were 'Barcel', 'Kenhy', 'Kentucky-31' (KY-31), tural carbohydrate fractions in each of eight vegetatively growing, 'Missouri-96' (MO-96), 'Mozark', 'Stargrazer', and the experiendophyte-free, tall fescue cultivars (Festuca arundinacea Schreb.) mental selections C-1 and HiMag (Mayland and Sleper, 1993). and relates their sugar concentrations to cattle grazing preferences. On 20 Sept. 1991, these grasses were seeded in rows into The experimental area consisted of eight cultivar plots replicated three an irrigated, Portneuf silt loam loess soil (Durinodic Xeric times in each of three pastures. Within each pasture, forage was Haplocalcid) near Kimberly, in south-central Idaho (42Њ30Ј sampled between 0830 and 1000 h mountain daylight time (MDT) N, 114Њ08Ј W, elevation 1200 m). The experimental design during each of four seasons and 2 yr. Freeze-dried forage samples was a randomized complete block with three pastures (blocks), were extracted with hot water and an amylase (Clarase) solution. three replications nested within blocks, eight entries (main Sugars were quantified colorimetrically using potassium-ferricyanide plots), and six rows per main plot. Rows were 0.56 m apart and glucose-oxidase methods. Cattle grazing preferences among these and main plots were 3.35 by 6.7 m. tall fescue cultivars were related to the concentrations of total nonstructural carbohydrates (TNC) (r 2 ϭ 0.49, P Ͻ 0.05). Other sugar Harvesting and Sample Preparation fractions were not significantly related to grazing preference in this study. The nonstructural carbohydrate concentrations averaged over Vegetatively growing forage was clipped at a stubble height the entire study were glucose, 14; fructose, 5; sucrose, 40; fructan, 23; of 8 cm from randomly located 0.6-m sections of Rows 3 and insoluble starch, 24; and TNC, 129 g kg Ϫ1 . A forage selection criterion 4 in each plot. The forage in Block 1 was clipped on d 131, should include measures of the TNC because of their close relationship 165, 221, and 256 of 1993 and d 129, 164, 220, and 262 of 1994. to animal grazing preference.