Using a 2-step extension methodology to freeze ram semen, 2 freezing protocols (P1 and P2) and 3 extenders were evaluated in a split-sample experiment. The freezing protocols were tested in combination with Extenders A and B (Experiment 1), and B and C (Experiment 2). Protocol 1 included centrifugation before filling the straws to reconcentrate the diluted semen to a calculated sperm concentration of 800x10(6) cells/mL. Protocol 2 involved appropriate ejaculate extension to yield 800x10(6) cells/mL as in P1, albeit avoiding centrifugation. Extenders A and B were milk-based and were supplemented with 5% egg yolk and fructose. Extender B was clarified by centrifugation (twice at 3310g/20 min). Extender C was based on TRIS-citrate-fructose supplemented with 20% egg yolk and clarified as described for Extender B. Final glycerol concentration was 7% for all 3 extenders. Post-thaw parameters studied were subjective motility, computer assisted sperm motility analysis (CASA), membrane integrity (SYBR-14/PI), and capacitation status (chlortetracycline assay, CTC). The overall sperm concentration (x10(6)/straw) differed (P<0.001) between P1 (mean +/- SD, 138.1+/-14.8) and P2 (216.5+/-13.9). Despite centrifugation, P1 appeared to be less harmful for spermatozoa than P2, yielding higher percentages of subjective motility, linearity, membrane integrity and uncapacitated spermatozoa. Due to the difference in concentrations obtained between P1 and P2, the total calculated numbers of spermatozoa having desirable characteristics were higher in samples processed as P2. In Experiment 1, P1 resulted in lower calculated numbers x10(6) in the AIdose of subjective motility (87.2+/-5.1 vs 125.3+/-5.1; P<0.05), linearity (70.6+/-4.3 vs 79.8+/-4.3; NS), intact-membrane (77.4+/-5 vs 108.5+/-5.1; P<0.001), and uncapacitated (36.5+/-2.5 vs 46.5+/-2.5; P<0.05) spermatozoa, than P2. In Experiment 2, calculated sperm numbers (x10(6)/straw) were lower in P1 than in P2 for subjective motility (80.8+/-5.4 vs 92.0+/-5.4; NS), linearity (63.3+/-5.6 vs 73.1+/-5.6; NS), membrane integrity (77.7+/-3.6 vs 101.0+/-3.6; P<0.001), and uncapacitated spermatozoa (28.3+/-3.24 vs. 4.1+/-3.2; P<0.01). Extender B (clarified milk extender) was consistently better than Extender A (nonclarified milk extender) for all parameters studied, but the difference was only statistically significant for linearity after 1 h of incubation at 38 degrees C (44.0+/-2.4 vs 36.2+/-2.4; P<0.05). Extender B was also better than Extender C (TRIS-citrate-fructose) for percentage of uncapacitated (49.7+/-2.2 vs 34.4+/-2.3; P<0.001), subjective motile (57.5+/-2.7 vs 43.8+/-2.7; P<0.01), and linear motile (46.5+/-2.8 vs 33.7+/-2.8; P<0.01) spermatozoa, but not for membrane integrity (51.6+/-1.5 vs 51.7+/-1.5). It was concluded that exclusion of centrifugation, as in P2, yielded higher sperm numbers with desirable characteristics per straw. Clarification of milk-based extender (B) resulted in better post-thaw sperm quality, especially compared with TRIS-based extender (C). (C) 2000 by Elsevier Science Inc.