Íslenskar landbúnaðarrannsóknir - 01.09.1976, Síða 54

52 ÍSLENZKAR LANDBÚNABARRANNSÓKNIR
Table 4. Constants for the effect of age of dam on autumn weight of lambs ± s.e.
Sexes combined Male lambs Female lambs
Age of dam ------------------- ------------------------ ---------------------
years Number Constant Number Constant Number Constant
1 ....................... 226 -í-6.08 + 0.24 108 +6.89±0.37 118 +5.28±0.31
2 ...................... 1767 +0.71±0.09 888 +0.76±0.14 879 +0.66±0.12
3 ........................ 1969 0.90 ±0.09 939 1.19±0.14 1030 0.64±0.12
4 ...................... 1984 1.33 ±0.09 947 1.28±0.14 1037 1.36±0.12
5 ...................... 2029 1.51±0.09 984 1.68±0.14 1045 1.36±0.12
6 ........................ 1848 1.47 ±0.09 900 1.75±0.14 948 1.20±0.12
7 ........................ 1485 0.87 ±0.10 737 0.89±0.15 748 0.85±0.13
8 ...................... 1018 0.54±0.12 503 0.69±0.18 515 0.36±0.15
9 ........................ 891 0.17±0.13 432 0.18±0.19 459 0.16±0.16
golt 1969), but the lambs in the present
investigation are on the average younger, as
already mentioned.
Age of dam.
All ewes 9 years and older have been com-
bined into one age group. Most of the one
year old ewes in the present investigation
were on only one of the experimental farms.
The constants for age effect of dam are
shown in table 4. These show that the autumn
weights increase up to an age of dam of 5
years, whereafter they decrease. This in in
agreement with several comparables investi-
gations (Blackwell and Henderson
1955, Eikje 1971, Sangolt 1969).
Correction of data.
When correcting data for systematic environ-
mental effects, the conditions must be full-
filled that means and variances are the same
in all subgroups after correction. In the
present investigation, no attempt was made
at studying what type of correction would
be most suitable. Other investigations have
shown that sex difference should be corrected
by multiplicative factors while the same aut-
hors show somewhat diverging results with
regard to correction for type of birth and
rearing (Eikje 1971, Sangolt 1969).
In he present investigation the type of
birth and rearing is by far the most im-
portant source of variation. A within sex
correction of the data described here was
carried out by using the additive correction
factors obtained from the present study for
type of birth and rearing, age of dam and
the regression coefficient of weight on age
of lamb.
Averages and variances for corrected
autumn weights were then calculated within
farm, year and sex for singles and twins
separately. It seemed as if the assumptions
for additive correction held reasonably well
in most cases but failed in some instances.