Table 4. The statistical cross-fostering design matrix.

Combination	Coefficients
	R	ao	do	io	am	dm	cmo	ddmo
M11O11	1	0	$-\frac{1}{2}$	0	0	$-\frac{1}{2}$	1	$\frac{1}{4}$
M11O21	1	0	$\frac{1}{2}$	0	1	−$\frac{1}{2}$	0	−$\frac{1}{4}$
M12O11	1	1	−$\frac{1}{2}$	0	0	$\frac{1}{2}$	0	−$\frac{1}{4}$
M12O21	1	0	$\frac{1}{2}$	−1	0	$\frac{1}{2}$	0	$\frac{1}{4}$
M12O12	1	0	$\frac{1}{2}$	1	0	$\frac{1}{2}$	0	$\frac{1}{4}$
M12O22	1	−1	−$\frac{1}{2}$	0	0	$\frac{1}{2}$	0	−$\frac{1}{4}$
M22O12	1	0	$\frac{1}{2}$	0	−1	−$\frac{1}{2}$	0	−$\frac{1}{4}$
M22O22	1	0	−$\frac{1}{2}$	0	0	−$\frac{1}{2}$	−1	$\frac{1}{4}$

The rows are labeled to reflect the corresponding maternal–offspring genotype combination, as in Table 3 [given as M_ijO_kl, where M_ij is the unordered genotype of the mother (with i and j being the two alleles she has at the focal locus) and O_kl is the ordered genotype of the offspring (with again, k and l being the two alleles that the offspring has at the focal locus]. The columns labeled “Coefficients” correspond to the index scores used in a regression model to fit the intercept or reference point for the model (R), the three direct genetic effects (a_o, d_o, and i_o), and the two maternal effects (a_m and d_m). The last two columns correspond to terms that we do not interpret here because they do not correspond to independent effects of the maternal and offspring genomes. The first of these, c_mo, is the confounded effect (see text) and the second, dd, is an interaction between the dominance effects in mothers and their offspring (see Discussion)