Íslenskar landbúnaðarrannsóknir - 01.03.1970, Blaðsíða 63

COLOUR INHERITANCE IN ICELANDIC SHEEP 61
TABLE 20
Comparisons of x2’s from table 19
Comparison among Obtained from lines DF X2 Probability
ai, a3, a4, 2-3 6 1.900 0.95 > P > 0.90
bi, bs, b4
Cl, C3, c4 5-6 3 0.741 0.90 > P > 0.80
a^, a2, a3, a^ vs. 2-5 4 1.898 0.80 > P > 0.70
bi, b2, b3, b4
a2 vs. b2 3-6 1 0.739 0.50 > P > 0.30
TABLE 21
Estimates of a2, b2 and c2 and their standard errors when matings A2A5 X A5A5
and A5A5 X A2A5 are excluded
Parameter Estimate S.E. | 1.0 —p |
(P) S.E.
a2 0.967 0.115 0.290
b2 0.905 0.127 0.749
c2 0.931 0.082 0.518
viation by itseli' is not significant it shows
the same trencl as the two other matings
mentioned above.
The first explanation of the deíiciency
of grey progeny in the above matings that
comes to mincl is tliat misclassification of
lambs with respect to colour has occurred.
The results fronr the above matings were
therefore re-analysed and this time they
were grouped according to tlie birth-year
of the lambs. The resulting segregation by
year is shown in table 23.
The heterogeneity x2’s in table 23 based
on one degree of íreedom show that no
definite difference can be demonstrated
with respect to the segregation ratios be-
tween the period prior to 1959 and the
period 1959—1961. The grouping of the
data into the two periods shown in table
23 was determined by the number of pro-
geny out of the A2A5 X A2A5 mating, as
this was tlie only groujiing of the data in-
to two year-classes which gave expectecl
numbers above 5 in all classes.
The A2A2 X A5A5 mating shown in
table 9 gave no nongrey progeny. As this