L. Alfonso et al., ESTIMATES OF GENETIC-PARAMETERS FOR LITTER SIZE AT DIFFERENT PARITIESIN PIGS, Livestock production science, 47(2), 1997, pp. 149-156
Different genetic models for litter size in pigs were compared. (Co)va
riance components at different parities were estimated by REML in two
Landrace populations. Maternal genetic effects were also assessed. Dat
a consisted of 34417 and 16750 litters records collected on each popul
ation. Different analyses for the number of pigs born in total and ali
ve using an animal model were performed: univariate analyses with dire
ct and maternal genetic effects (using a repeatability animal model wh
en more than one parity was involved) and series of bivariate and mult
ivariate analyses with each of first five parities treated as a differ
ent trait (without maternal genetic effects). The estimates of direct
heritability obtained by a repeatability animal model ranged from 0.05
to 0.07. Maternal genetic effects were not significant in any populat
ions, neither in the first parity nor in all parities. The estimates o
f direct heritability in each of the first five parities using series
of bivariate or multivariate analyses were not different from each oth
er within population. Nor were these estimates different from the heri
tability estimates obtained by a repeatability model. Estimates of gen
etic correlations between parities were not significantly different fr
om unity, except for the estimate between the first and the third pari
ty in one population (r(g) = 0.5). These results confirmed the adequac
y of a simple repeatability model without maternal effects in the sele
ction for litter size in the populations analysed. (C) 1997 Elsevier S
cience B.V.