Indian Society of Genetics & Plant Breeding
FISHER (1928) has shown how to partition total X2variation (with three degrees of freedom in the case of two characters) into its three components (each with single d.f.),
two of them testing the homogeneity of the individual segregations for the two characters
under study arid the third their linkage. These formulae are applicable to two characters segregating in any ratios. In case of three or more characters, the problem
can always be reduced to the above mentioned simple case by summing over the third
or the remaining characters and the linkage between any two characters may be
examined. The fact that this method becomes tedious when there are three or more
characters and the detection of linkage regarding each pair of characters is required,
is evident. Mather (1938) and Kempthorne (1957) have given orthogonal partitions
for the case of three characters each segregating in a 1:1 ratio (the case of backcrosses).
This is the simplest case we can come across for three characters. In this note, an
attempt has been made to bring out the analogy between the partitioning of X2in segregation data involving three or more characters and the partition of the total (treatment)
variation in a 2n factorial experiment into main effects and interactions, by using
Yate's (1937) technique. The advantage here is that the detection of association
between two characters (not genes), independent of the genes for a third character
likely to disturb the phenotypic frequencies of the associated characters, is possible
by this method.
Year: 1962
Volume: 22
Issue: 3
Article DOI: NA
Print ISSN: 0019-5200
Online ISSN: 0975-6906
M. V. PAVATE info_circle
Download PDF