The table displays the GP map estimated from the linear (L) and multilinear (M) regressions described in Table 1. The standard errors are those calculated from the linear model. The precision of the estimates may vary according to the genotype; frequent genotypes (such as the double heterozygote 22) is estimated more accurately than the rare ones (double homozygotes). The GP map from the general model is very close to the actual map, while the multilinear model is constrained and the fit with the expected values may be loose if the real map is not multilinear.