A STUDY OF GENOTYPE

By Roger E. Lyons
Originally appeared in Owner Breeder journal, December, 1997

        All of the sons of Teddy who stood in America, including full brothers Bull Dog and Sir Gallahad III, though they were aptitudinally quite different from one another, were able to exploit a broodmare population heavily laden with Domino influence. One son, Case Ace, was out of a mare by Ultimus, inbred 2x2 to Domino, and sired Raise You, the dam of Raise a Native. Teddy’s most important daughter, La Troienne, had but one significant offspring who was not a tail-male descendant of Domino. That one, Baby League, was by Bubbling Over, who was out of a mare inbred 3x4x4 to Domino. Regardless of differences in phenotype, genotype tends to assert itself in sometimes strikingly regular ways.
        With this principle in mind, Taylor Made Farm, where Saint Ballado, a full brother of Devil’s Bag, stood beginning with the 1998 breeding season, commissioned me to do an empirical study of the offspring of these two stallions as part of its stallion development program. The study, began with a survey of the ancestries of the dams of all progeny, aged three and up as of 1997, by both Devil’s Bag and Saint Ballado. The superior runners among these foals--SWs or stakes-placed in at least a G2 race--were separated from the main group for the purpose of comparing them with it as to the frequency of occurrence of each ancestor within six generations of the dams.
        No distinction was made between the Devil’s Bag and Saint Ballado groups. They were simply assumed to have the same genotype. The high-performance group had 39 cases, the low-performance group 372. A confidence interval was constructed around the proportion of each ancestor in order to test for the statistical significance of comparisons at the .10 level.
        If a 99 percent certainty is required for engineering and medical research and a 95 percent certainty for social, economic, and educational policy research, then a 90 percent certainty seems appropriate for thoroughbred pedigree research. In my experience that is the extent to which it can be an exact science. The claim that it is an inherently inexact science presupposes the possibility of a science that is absolutely exact, but, in truth, no such science exists. So, the study conducted for Taylor Made Farm did not measure up to any idealized expectations, but it had just the right amount of exactness--no more, no less.
        Obviously, the factual yield is of interest to those who would consider applying for a season to Saint Ballado, but the larger principles that emerged from these facts are of even broader interest from the standpoint of pedigree interpretation.

VIABILITY OR COMPATIBILITY?
        Mares with Mr. Prospector in their pedigrees produced five SWs in this group, and mares with Nijinsky produced seven. Note that both of these sires stood at Claiborne Farm, as does Devil’s Bag. No reference to the question of compatibility is necessary in order to explain Devil’s Bag’s success with mares by either of these great sires. Mr. Prospector and Nijinsky are both highly viable broodmare sires, and there can be no doubt that the viability of the mares played a large part. The factual detail of our study, however, is sufficient for us to infer that Mr. Prospector has a level of compatibility with our subject genotype that Nijinsky does not have.
        The first clue emerges from the numbers of Mr. Prospector-line and Nijinsky-line mares bred to the DB/SB genotype. Five SWs came from 16 mares with Mr. Prospector influence while seven came from 42 mares with Nijinsky influence. Tests of statistical significance showed that Mr. Prospector had a favorable effect while Nijinsky did not. Unless Mr. Prospector is that much more highly viable than Nijinsky as a broodmare sire, then the difference in their effects must be attributable to something else.
        Note the large number of mares with Nijinsky influence that were bred to the subject genotype and the relatively small number of mares with Mr. Prospector influence. Nine of the former mares also had Round Table in their pedigrees, three of them SWs, but the Nijinsky-Round Table combination had no significant effect. Three other mares had Round Table’s full sister Monarchy in their pedigrees, and two of them were SWs, but the Princequillo-Knight’s Daughter cross, which covers both siblings, showed no statistically significant effect (although there may be something to the fact that the average generational distance of Knight’s Daughter in the non-stakes-winners was 4.75 and only 3.86 in the SWs).
        Devil’s Bag was an undefeated Champion Two-Year-Old with lots of speed. Based on his phenotype, it would have seemed perfectly appropriate to return stamina to him through mares by Nijinsky and to reinforce this with the soundness and heart conferred by Round Table. But it would appear that Devil’s Bag has sired more towards his genotype than to his phenotype, which is not an unusual occurrence in the annals of breeding.
        The next clue that Mr. Prospector’s effect is in some relatively large measure attributable to compatibility with the subject genotype is that certain ancestors in his pedigree also showed a favorable effect. The most notable one was Nashua, who occurred in the pedigree of 41 mares, ten of which produced SWs. Clearly, Nashua is not just a trailer of Mr. Prospector’s effect. Marshua’s Dancer, a three-quarters relative of Mr. Prospector and out of a Nashua mare, sired the dam of one SW by Devil’s Bag; and Roberto, also out of a Nashua mare, was the broodmare sire of another. Nashua was also the broodmare sire of a SW and the only strain of Nasrullah who had a favorable effect when combined with Princequillo.
        Native Dancer, Polynesian, and Case Ace, who occur in many other ways than through Mr. Prospector, also showed a significantly favorable effect. This suggests that Mr. Prospector’s ancestry was pertinent in a way independent of the viability of the dams.
        Further confirmation that compatability is at play in the cross of Mr. Prospector with our subject genotype has to do with the dominant make-up of the pedigree of Devil’s Bag and Saint Ballado. The most salient pattern in this pedigree is evident in the two occurrences of the Blue Larkspur-Teddy cross through Nothirdchance (Blue Swords-Galla Colors, by Sir Gallahad III) and Belle of Troy (Blue Larkspur-La Troienne, by Teddy). The reverse of this cross (Teddy-Domino) happens to be the most common element among the dams of SWs sired by this genotype.
        Our test of statistical significance was sufficiently rigorous that it proved only 12 ancestors to have a favorable effect--Asterus, Betty Beall, Case Ace, Helene de Troie, Mr. Prospector, Nashua, Native Dancer, Roman, Sir Cosmo, Sardanapale, Teddy, and Stimulus. Among these, only Sardanapale does not have obvious Teddy and/or Domino connections. In fact, the Teddy-Ultimus cross actually proved to have a significantly favorable effect.

GENERATIONAL DISTANCE
        Betty Beall, the third dam of Better Self, has close Teddy-Domino associations in the following sense. Since we counted six generations of the dams (seven of the foals), Betty Beall was counted only when Better Self, who has the Teddy-Domino influence, was in the fourth generation of the foal or closer. Better Self was present in 21 mares, five of whom produced SWs, with no significant effect. But Betty Beall was counted in only 16 cases, which all had Better Self within four generations of the foal. His average generational distance in the non-stakes producers was 4.63 while in the stakes producers it was 3.80.
        The effect of Helene de Troie, La Troienne’s dam, is likewise indicative of La Troienne’s optimal generational range. SWs resulted from 15 of the 118 mares in whose ancestries La Troienne was counted, but only 75 foals were out of mares with La Troienne in their fifth generation or closer, from which group came 12 of the 15 SWs. Helene de Troie is in this case a signifier, not the thing signified.
        Sir Cosmo’s effect might have similar meaning as regards Round Table and Monarchy since he is the sire of their dam, Knight’s Daughter, who has connections to the ancestries of Teddy and Domino. When we tested Knight’s Daughter with Sir Cosmo’s sire present, we still got no effect, but there was an effect in pedigrees which included Orby, his grandsire, within seven generations of the foal.

RESIDUAL INFLUENCE
        The significantly favorable effect of Sardanapale tends to confirm that a very distant influence can have value even when not reinforced by an intense accumulation of strains. Skylarking II, the fourth dam of Saint Ballado, is out of a mare by a son of Sardanapale. That seems a thin connection, but Devil’s Bag has crossed well with mares who have a strain or two of this relatively rare ancestor. Sardanapale is the sire of Polynesian’s fourth dam, and he is the sire of Nashua’s second dam. To Market’s grandsire is out of a Sardanapale mare, and Devil’s Bag sired Abaginone out of a mare by Spectacular Bid, who is inbred to To Market (like Nashua, out of a Johnstown mare). Bold Bidder, with one strain of Sardanapale, figured in the dams of two other SWs. One SW resulted from four mares with Roberto in their pedigrees, and Bramalea, his dam, is inbred to Sardanapale. Interesting that Skylarking’s strain of Sardanapale is male while all of these other strains are female.
        A number of ancestors proved to be significantly unfavorable to the genotype surveyed, but most fell within the confidence interval established for comparison, which means that they were neither favorable nor unfavorable on their own. In the successful cases these ancestors were redeemed, some by the viability of the mare and others by fortunate combination with other influences. There is yet much to be gathered from the data, but about one thing we can have somewhat more than a 90 percent certainty: mares selected for Saint Ballado will meet a higher standard of compatibility than has ever been established for a stallion.