When I was involved with our software development, SunShow POWER Index software (which never made it to market), we computed the Inbreeding Coefficient by an independently-derived version of the Stevens (1975)-Boyce (1983) adapation of the Sewall Wright Inbreeding Coefficient (1922). An alternative method of merit which we considered adding to the software was the Quass-Henderson (1976) approach. ref 3.

The paths referred to in this discussion are the total generational distances to each occurence of the Ancestor-in-Common, as recursed along the pedigree tree.

        "..there is no evidence for a true increase in inbreeding

        through time. On the contrary, for horses with the same

        degree of pedigree completeness, mean "F" (the Inbreeding

        Coefficient) appears to decrease through time." (p.397) ref 1.

        "As mentioned earlier, breeders tend to avoid very close

        inbreeding. In one treatise on breeding, horses with

        paths of length 4 or less are considered "incestuously

        inbred" (Close Consanguineous Matings); those with minimum

        paths of lengths 5 or 6, "inbred"; 7 or 8, "linebred" (Remote

        Consanguineous); and those whose minimum path lengths are 9

        or greater, "outcrossed". (p.397) ref 1.

        "Analysis of extended (deep) pedigrees can reveal aspects of

        population structure that might otherwise go undetected."

        (p.398) ref 1.

        "..failure to consider more remote generations leads to a

        considerable underestimate of inbreeding." (p.398) ref 1.

        "A striking indication of the extent of shared ancestry among

        Standardbreds is that in spite of the depth and completeness

        of the pedigrees of our 1103 base population horses, we were

        able to identify only 4104 distinct ancestors (yielding a

        total population of 5207)." (p.398) ref 1.

The following supports the reasoning that properly planned inbreeding can reinforce desirable characteristics without reinforcing deleterious side effects.

        "The relative importance of close, as opposed to remote inbreeding, 

        in terms of effects on Identity by Descent and Homozygosity, depends 

        on whether or not the characteristics of interest have a deleterious

        effect on fitness." ref 1.

        For example, in the absence of selection, 2 ^18 (to the 18th power, or 262,144) 

        paths of length 27 (14 generations) makes the same contribution to F (Inbreeding 

        Coefficient), as a single path of length 9. However, if a particular allele has a selective 

        disadvantage of as little as 0.01 in each generation, 314,128 paths will be required 

        to produce the same probability of identity by descent as that produced by a single 

        path of length nine. ref 1.

            "In other words, remote inbreeding will have a smaller effect on identity by descent 

        and homozygosity of such mildly deleterious alleles than of neutral ones. It also follows 

        that alleles contributing to reproductive success may be more likely to be identical by 

        descent than are neutral alleles, and that remote inbreeding can be relatively more 

        important in contributing to homozygosity for such favorable alleles."

        ref 1.

Restated, Remote Consanguineous Inbreedings (commonly referred to as "inbred" and "linebred")in a pedigree tend to have a more positve effect on successful alleles, little effect on neutral alleles, and a smaller (although not zero) effect on deleterious alleles.

Both Thoroughbreds and Standardbreds descend from a relatively small number of ancestors.

- Loren Bolinger