Íslenskar landbúnaðarrannsóknir - 01.03.1970, Qupperneq 95

COLOUR INHERITANCE IN ICELANDIC SHEEP 93
TABLE 53
Number of progeny showing alleles Ax—A& out of matings of A5A5-sires
to white dams, and gene frequency
Progeny No. of Gene S.E. of
showing allefe progeny frequency gene frequency
Ai 371 0.679 0.020
A2 35 0.064 0.010
A3 4 0.007 0.004
A4 17 0.031 0.007
A5 120 0.219 0.018
A 6 0 0.000
Sum 547 1.000
would be expected to give white and non-
white progeny in the ratio 1:1, i.e. one
would on tlie average expect one nonwhite
progeny to be produced for each two pro-
geny obtained from heterozygous white
ewes. Tlius one obtains the expected num-
ber of heterozygous ewes by doubling the
number of nonwhite progeny obtained.
This gives tlie proportion of heterozygous
white ewes as 0.642, and that of homo-
zygous wliite ewes as 0.358.
In table 54 is shown the frequency of all
lambs with colour description by farm and
main colour of lamb.
It is obvious from table 54 that there is
a large difference between farms with re-
spect to the frequency of individual cofours.
This is to a large extent a result of the
use of certain nonwhite or lieterozygous
white rams on certain farms for experi-
mental purposes. Nonwhite ewes mainly
were selected for the experiments on these
farms and the remaincfer of the ewes were
not included in the records. The data in
table 54 are therefore not suited for estima-
tion of gene frequencies for this reason.
Another reason for a bias is the fact that
even on the farms where all sheep have
been recorded, the matings have been far
from random witli respect to colour. On
several of the farms the nonwhite ewes
have been mated to nonwhite rams mainfy,
and the white ewes to white rams. This
deviation from random mating woukl tend
to decrease an estimate of tlie frequency
for allele Ax. In other cases records have
been available about lieterozygous AjA5-
ewes, which have then been mated to homo-
zygous grey rams in order to produce dark-
grey lambs, ancl this procedure also has in-
troduced a bias.
It is possible that some bias has been
introduced in the data of table 53 through
selective mating. l’his is believed to be of
minor importance, except for ewes of geno-
type Ax A2, which in some cases were select-
ed for matings with A5A5-rams in oider
to produce darkgrey lambs. As the observed
frequency of A2 is low, this type of selec-
tion can hardly have affected the results
to any marked extent. In other respects
A5A5-rams were mainfy usecl alongside
with white rams in the ewe flock, their use
not being determined by their colour but