The experiment was carried out to evaluate saponins and tannins contents, the proteins solubility (PS) of 28 alfalfa cultivates: Crioula, Monarca, BR 4, Alto Great, MH 4, SW 9210 A, 5929, BR 1, El Grande, 5715, MH 15, Valley Plus, BR 2, Rio, SW 8210, Maricopa, ICI 990, 5888, P 30, Alfa-200, WL 516, SW 8112 A, BR 3, Florida 77, Araucana, Falcon, Semit 921 and Sutter. The analyzed material came from the 10th and 14th harvests (which occurred on 12/08/97 and 04/16/98, respectively) from an experiment conducted at the College of veterinary medicine and Animal Science, UNESP, Botucatu. Samples from each cultivate were weighed and oven-dried at 52ºC. Saponins were extracted with a hydrophilic and lipophilic solvent, and their content was calculated by means of the following equation: y = 0.8121x - 1.4759 (R² = 1.00). Total and condensed tannins were ultrasound-extracted for 12 min. Total tannin content was calculated by means of the following equation: y = 44.978 + 0.5644 (R² = 0.9977); condensed tannin content was determined by multiplying the absorbency by 78.26 and dividing it by the dry matter content. PS was determined through the KOH method, using the following formula: PS (%) = (100) (soluble protein) / crude protein. No differences were detected (P>0.05) among cultivates in saponin content, total tannin and condensed content of PS. Harvest time effect was observed (P<0.05) in total tannin content only. The low average saponin content (1.0%) and protein solubility (34.11%) do not constitute a limiting factor for the use of the above alfalfa cultivars.
content; condensed tannins; harvest time; ultrasound
