Technical Abstract: A protocol was designed that is simple in design and easy to use and that consistently separates genotypes based on their salt tolerance. Seed of 12 alfalfa cultivars were planted 1.5 cm deep in 3.8 X 21 cm "Cone-tainers" in 2000 and 2001. Bottoms of the Cone-tainers were plugged with capillary matting, and filled with 70 grit silica sand. Matting kept the sand from running out of the cones and slowed the flow of water into the cones preventing sand abrasions on roots of the young seedlings. Screening for salt tolerance was done by immersing 98 cone flats containing 6 week old seedlings in a complete nutrient solution containing NaCl starting at an electrical conductivity of 3.0 dS m-1 and increased 3.0 dS m-1 per week for 10 weeks. NaCl and CaCl2 were used in proportions to maintain a sodium adsorption ration of 3.5. Probit analysis was used to estimate the lethal dose and time required to kill 50% (LD50) and 75% (LD75) of plants for each cultivar. Pearsons and rank correlations were estimated between years (runs) to determine the repeatability of the protocol. The correlation between years at the LD50 level was r=0.83 (P<0.001), and r=0.86 (P<0.001) at the LD75 level. Rank correlations were slightly better with r=0.90 (P<0.001) between years at the LD50 level and r=0.88 (P<0.001) at the LD75 level. High correlations between results from the two years demonstrate the robustness of this protocol and verify that it produces repeatable results.