Variances and covariances between productive and adaptive traits and temperament in a composite breed of tropical beef cattle

Variances and covariances for productive and adaptive traits and temperamen t were estimated in a tropically adapted composite breed of cattle grazed a t pasture in the tropics. Measurements included weights and period weight g ains between birth and 18 months, mature cow weights, repeated tick, worm e gg and buffalo fly counts, rectal temperatures (measures of heat resistance ), flight speed scores (measures of temperament), scrotal circumference at weaning, 12 and 18 months and repeated measures of pregnancy rate and days to calving. Additive heritabilities (h(a)(2)) of all weights except weaning weight a were moderate to high (h(a)(2) = 0.31-0.84), but low to moderate for weaning weight and period weight gains (h(a)(2) = 0.14-0.21). Maternal heritabilities (h(m)(2)) for weights to 18 months and preweaning gain were significant, particularly for weaning, weight (0.34) and preweaning gain (0 .30). Genetic and phenotypic correlations between weights at all ages were positive, but ranged from low (r(g) = 0.07 between yearling weight and matu re cow weight; r(p) = 0.08 between birth weight and mature cow weight) to h igh (r(g) = 0.95; r(p) = 0.87 between weights at 12 and 18 months), with ge netic correlations generally being higher than phenotypic correlations. Her itability of tick and worm counts were moderate to high (0.35-0.44). Herita bility (0.18) and repeatability (0.27) of rectal temperature was low. Fligh t speed scores were moderately to highly heritable (0.40-0.44). Heritabilit y of scrotal size at all ages was moderate to high (0.35-0.46) and maternal genetic effects were also important. Pregnancy rate and days to calving we re lowly heritable (0.04 and 0.07, respectively). Genetic correlations betw een tick and worm counts (r(g) = 0.30) and tick count and rectal temperatur e (r(g) = 0.22) were both favourable, though the genetic relationship betwe en worm count and rectal temperature was close to zero. Relationships betwe en tick and worm counts and growth, male and female fertility and flight sp eed scores were all close to zero, indicating that tick and worm resistance were largely independent traits, except for relationships amongst other re sistance traits. Genetic relationships between rectal temperatures and most weights and period weight gains were favourable (r(g) = - 0.20 to - 0.49) though genetic relationships between rectal temperatures and birth weight, mature cow weight and dry season gain were lower (-0.08 to - 0.12). Low to moderate, favourable genetic relationships existed between rectal temperatu re and pregnancy rate (-0.16) and days to calving (0.16), although phenotyp ic relationships were close to zero. To maximise genetic progress, it is co ncluded that homogeneous rather than heterogeneous variance should be assum ed for males and females to estimate breeding values in zebu derived cattle grazed at pasture in the tropics. For beef breeding programs based in the tropics, where there are more sources of variation for economically importa nt traits, careful consideration needs to be given to use of appropriate an alytical models, to ensure that effects such as sire X year or contemporary group interactions are correctly accounted for when population genetic sta tistics are estimated. (C) 2001 Elsevier Science BY All rights reserved.