Íslenskar landbúnaðarrannsóknir - 01.03.1977, Page 20

18 ÍSLENZKAR LANDBÚNAÐARRANNSÓKNIR
Table 3 . Analysis of varianee. Variation in weaning weight of lambs.
Source of variation D.F. M.S. E(M.S.)
Between flocks 369 840,33^ c2 + 156,5c2 e
Within flock 57626 26,26 c2
e
registered for lambs that are slaughtered. The
mean age of lambs on 1. October was 133-8
days with a standard deviation of 6.55 days.
Analysis of variance between and within
flocks is shown in table 3. On this basis it
can be estimated that 83.5% of the total
variation in the weaning weight of lambs is
found within flocks. It is mostly this variation
that is used when selecting for weaning
weight and this variation is therefore of
greatest interest in the breeding work.
Yi jkl
SOURCE OF ENVIRONMENTAL
VARIATION WITHIN FLOCKS
It is found that factors such as type of birth
and rearing of lamb, age of lamb, age of
dam and the sex of the lamb affect the
weight of lamb.
Because of the uneqal number of observa-
tions in subclasses in this material least
squares procedures (Harvey, 1960) have
to be used to estimate the effects of these
factors on the weight of lamb. The following
model was chosen:
y+ki+aj+rk+sik+b1(Xlijk.L-X1)+b2(X2ijk;i-X2)+eijkle
Here Y..,, denotes weight of
l]Kl
lamb
y = the overall mean
k. =the effect of the ith sex, i=l,2
í
a. -the effect of the jth age of dam, j -2,..., 9.
3
r =the effect of the kth type of birth and rearing group, k-l,..5.
k
s =the effect of interaction between k.. and r. .
ík + ^
=the partial regression coefficient of weight on age of lamb.
b^ =the partial regression coefficient of weight on age of lamb squared
=the mean age of lamb.
X2 =the mean age of lamb squared.
e. . .= the error term in the model.
íjkl >;