Íslenskar landbúnaðarrannsóknir - 01.03.1977, Síða 28

26 ÍSLENZKAR LANDBÚNASARRANNSÓKNIR
Table 12. Heritability estimates.
Trait Total Ram lambs Ewe larribs
Weaning weight 0,18±0,015 0,21±0,020 0,18±0,023
Carcass weight 0,13±0,013 0,17±0,019 0,11±0,020
Dressing percentage 0,18±0,015 0,21±0,021 0,17±0 j-022
material. Much fewer estimates are found
for carcass weight and dressing percentage
and those are mostly based on limited mate-
rial. Gjedrem (1970) estimated herita-
bility for carcass weight 0.09 and for dressing
percentage 0.24 on data from the ram circles
in Norway.
For a total of 50996 lambs with recorded
weaning weight, in sire groups of 10 and
more progeny the heritability was estimated
0.20±0.015. The estimates in table 12 are
therefore probably biased a little downward
by the selection of lambs kept for replace-
ment. According to Rönningen (1972) such
selection should give a downward bias. The
estimates for ewe lambs are probably more
affected than those for ram lambs. In this
material 4% of the ram lambs and about
30% of the ewe lambs are kept for replac-
ement. The calculated sélection differential
was 0.72 standard deviation units for ram
lambs and 0.47 units for ewe lambs (Jón-
MUNDSSON, 1975).
The selection of older rams on basis of
progeny results could also give a downward
selection bias. For 122 rams that have pro-
geny in this material both in 1970 and 1971,
the progeny in 1970 were 0.017 kg above
flock average. Therefore there seems to be
little selection based on the progeny results.
As mentioned before Hallgrímrson
(1971) found higher heritability for weaning
weight of singles than of twins. The same
was found by Gjedrem (1967) in Norway.
In order to test this effect the heritability
was calculated for both singles and twins
by a one way analysis of variance within and
between sires. For singles the heritability
was found to be 0.26±0.03 and for twins
0.34±0.0l6. The estimate for twins is biased
upward by environmental correlation within
twin pair. However, one must conclude
that no tendency is shown towards lower
heritability for twins in this material.
Assortative mating can affect the herita-
bility estimates. One way to test this is to
study the relationship structure. The only
recorded information for this purpose in these
data is the sire of dam. This is most likély
the main reason for the relationship among
ewes in the flock. The average relationship
(R) was calculated as:
z
Sn - N
_ ----------- 0.25
N (N-1)
where nj is the number of daughters sired by
ram i and N is the total number of ewes
in the flock. The average relationship (R)
among a total of 26095 ewes in 334 flocks
was cálculated 0.034. The ratio of the rela-
tionship between ewes mated with the same
ram to the relationship of all ewes in the