Preservation and Improvement Breeders can have both in their programs

         There are further definitions which do not relate to our animal breeding context.

         The standard text, "Genetics and Animal Breeding" (Johanssen and Rendel, Stockholm 1963; English translation 1968) has this to say: "Very often a breed can be divided into different strains which from a breeding point of view are more or less isolated from each other due to geographic conditions or when in some respects the aim of breeding is different."

         An amazing amount of confusion has been generated on the Arabian scene by the fact that Bedouin breeding has been described in terms of "family strains" when no two speakers seem to have defined "strain" in quite the same sense. For that matter, "Bedouin" seems to have been used in a number of senses and it is not surprising that contradictions have arisen. "The horse breeding tribes" are not and as far as we can tell have never been a monolithic entity with entirely uniform horses or ideas on horse breeding. Since different travelers spoke with different tribes, different ideas as to the importance of the "family strain" concept and totally different ideas as to what the "strains" were like and which were more important or desirable have come down to us.

The only real certainty out of it all seems to be the fact that the Arabian horse was bred in the desert with attention to tail-female descent (this is all "family strain" in the Arabian breeding sense is; it is the eastern equivalent of the Western idea of placing emphasis in breeding on the tail-male line).

         Before considering what the implications of emphasis on matrilineal (just another word of tail-female) descent might be, it could be instructive to consider the background from which the early European travelers were coming when they encountered the Bedouin. In some ways our experience is as foreign to theirs as theirs was to that of the desert raiders, so we can learn by trying to understand the differences.

         The history of Europe is the story of small countries - often of individual tribes - warring among themselves for control of circum- scribed areas which they felt to be especially valuable. The great empires which unified the scene were by comparison short-lived and even during their heydays they did not unify the people in the sense of producing a homogenized culture throughout their areas. This is the background from which the many local varieties - isolated from inter- crossing by wars and war's aftermath, suspicion and rivalry - developed into "nations" of humans and "breeds" of livestock, each closed among themselves.

         Travelers coming from this history were not prepared to under- stand that the Bedouin "tribes" were nomads who wandered over vast areas in the course of a year, their paths crossing and sometimes running together. Even when individual tribes held themselves aloof from their neighbors they were not physically isolated like the citizens of little European countries barricaded behind their rivers and mountain ranges. Each tribe doubtless held that its warriors were the fiercest, its women the loveliest, its horses the swiftest and most enduring - they were nomadic, not inhuman - but they were saying and believing these things in a different context from the European experience.

         When the Europeans heard the Bedouin describing the different lines of horses which they maintained, it did not occur to them to ask whether they could be, or ever were, inter-crossed. Such things were foreign to their ideas of stock breeding which could not conceive of a single breed spread over the Arab countries. Indeed, we are lucky that the terminology did not become set at the earliest stage for it would have us referring to Kehilan, Seglawi, Maneghi etc, as different breeds; even the Blunts made this error at first though they learned better soon enough.

         There may be some significance to the fact that the American, Homer Davenport, when he journeyed with the Arabs buying horses for his 1906 importation, did not come home with the idea of strain separation or of "good" or "bad" strains. He recorded the strains of his horses and the information he was given on them but when reading his accounts one does not get the feeling he thought of this as anything but a source of knowledge of their background. Certain strains are spoken of as being prized in certain areas or by certain tribes but it is not with the feeling of metaphysical superiority. Rather, these became celebrated through the fame of celebrated individuals which happened to belong to them.

         In summary, the Bedouin seem to have used a word which may be translated "strain" in the first sense of the dictionary definition at the beginning of this article. Perhaps it would have been better in a number of ways to have called these entities just "families" rather than "family strains" as we have come to do. European travelers who encountered this idea interpreted it more along the lines of definition 3 and of the animal breeding sense of Johanssen and Rendel. By questioning the Bedouin and sometimes by their own observation of such horses as they saw, the Europeans developed their own concept of "strain" or even "breed" and took it home with them because the Bedouin sense of "female family line" did not make sense to them. Only a few long-term observers carried their ideas beyond this preliminary level.

         Because the casual observers outnumbered the careful ones and because even the careful ones could be misled by thinking one tribe had examples of strains that were like those of all tribes, the descriptions of "the breeds of Arabians" became current in Europe. In fact, what they were describing was not "the Seglawi breed" but "the Seglawis of this tribe" and interpreting this in light of their own experience (in which a breed name would not be used by two different groups for their stock unless the stock were indeed the same in type and by descent).

         Emphasizing tail-female inheritance is foreign to our Western way of doing things but indications are that it used to be rather general among the human family. It is the more primitive system and is based of course upon the fact that even members of groups which have not quite worked out yet how offspring are fathered are pretty clear on the fact that they have mothers (the women are anyway). At a slightly later period of cultural development, it still remains possible to wonder about paternity while maternity, until the era of embryo transplants, was a fixed and certain quantity defined by the legal phrase, "born from the body of".

         In our horse breeding example, it clearly must have appeared to the hard-headed Bedouin that the thing to do was to place emphasis on what you knew for certain. It may be going farther than the evidence warrants to suggest that at an early point in their tradition sires were not known or at least not recorded. Even had this been the case at some time, of course they were too sophisticated not to have come to the realization eventually that emphasis on sires was important in horse breeding. After all, aside from any traditions of maintaining "the right Arabian breed," their success in raiding and at times their lives depended on the horses they bred. There is surely no question of ignoring sires in historical times - strains of both parents are almost always given on desertbreds that have come into our knowledge through being sold to Westerners.

         It seems that as far back as we have any record, the Arabs used and emphasized the mares; stallions were a noisy but necessary encumbrance and the great majority of colts was sold. This implies that, with few stallions in each tribe, most of the young stock of any generation would be shared out among relatively few male parents. And it follows necessarily that much of the visible variation among the youngsters would be attributable to their dams. This would tend to reinforce the matrilineal emphasis.

         We are told in the records of the Abbas Pasha purchases that certain strains (in particular one Seglawi family) were uniform when the mares were bred to stallions of the same strain but varied more in shape when the sires were of other strains. This is often quoted to show that the Bedouin crossed strains and as often used to show that they bred them within themselves to fix type. I think a much more interesting implication emerges if you consider the scarcity of stallions maintained for breeding in the desert along with this description of strain behavior when outcrossed or not. If much of the breeding of a tribe's mares was done within the tribe, then a small choice of sires was available. If out of this small number a Seglawi was to be picked for the Seglawi mares, it was highly likely that all the Seglawi mares would be bred to one and the same horse. Naturally, if the mares were related by female line and they were bred to the same sire, the offspring should have been uniform. Since, further ,the Seglawi stallions) of a tribe must have come from that tribe's Seglawi mares, it suggests that mares were bred to their own near relations in female line if they were bred within strain within the tribe.

         I sometimes get the feeling that modern Arab breeders think of "strain" almost in the sense of definition 2 of this article, as a mystical or metaphysical quality. I think it is important to keep in mind that if a stain type were fixed in any given situation, it was done so by the straightforward and comprehensible action of inbreeding and selection.

We have considered mammalian sex determination any number of times. Recall that sex is determined by chromosomal constitution. Normally XX individuals are female and XY individuals are male (where X and Y refer to the sex determining chromosomes). Recall too that chromosomal segregation is random. Genes from the other chromosomes of the individual do not travel with any particular sex chromosome. It is also completely a matter of chance whether a fertilized egg is XX and will be a female or XY and will be a male.

Figure 1 shows the consequences of this mode of sex determination on the sex chromosome make-up of sons and daughters of sires and dams. Only the sex chromosome are indicated as the others all assort at random compared with this pair. Note that the Y chromosome follows a patrilineal mode of inheritance; the Y chromosome of any male came from his sire, his sire's sire, and right on back.

Note that a male offspring always receives his sire's Y chromosome and never receives the sire's X chromosome. The female, of course, must receive one x from each parent. This means that the Y chromosome, because it is male-determining, always follows the "tail male" line. There is no such necessary pattern with the X chromosome; a female must receive an X from her own dam, but she need not receive one from her maternal granddam. (The numerical subscripts serve to distinguish one chromosome of the same type from another - they are not meant to have genetic significance.)

There is no comparable matrilineal pattern. Since each individual has at least one X chromosome, it is possible in as few as two generations to lose both X chromosomes of the original female founder. (NB: mitochondrial DNA is not mentioned because this was written about 20 years ago.)

In other words, any tail- male Skowronek stallion has Skowronek's Y chromosome. A tail- female Bint Helwa mare is no more likely to have Bint Helwa's X chromosome than she is any other chromosomes.

         Every modern Arabian has a strain except for a few whose strains were lost because early-day records were not kept as we might have liked before the founding of the various Studbooks. Of course those, the knowledge of whose strains is lost, still have them; we just don't know what they are beyond the generic "Kehilan Ajuz" or "Old Thoroughbred". It is interesting to speculate about the significance of strain names today, especially when there are relatively few sources of a particular strain name (as the Kehilan Dajani which seems to trace back in all cases to just two 19th century foundation mares, Dajania in England and Mlecha in Poland). As we understand the family strain system, this must mean that a Kehilan Dajani of one country is related to an individual of the same strain in another country. The question of course is, "How closely related?" and the answer is, "Probably not very."

         "Strain breeding" in a more specialized sense is practiced by those who attempt, by working within a limited group, to reconstitute separate strains by close breeding among the descendants of each foundation mare, or small group of mares of the same strain. This certainly is "strain breeding" according to the sense of Johanssen and Rendel - "the aim of breeding is different," in this case in meaning to separate the stains - and it also agrees with all senses of the dictionary definition 1. We would like to hope that definition 3 would also be applicable here but of course the key is that the "high standard" is "maintained.... by selection," and that differs with the individual breeders involved.

         Whether or to what extent modern "strain bred" Arabians resemble the original Bedouin versions of their named strains is a trickier question. We have seen that it is at least open to discussion whether the strains ever were uniform and bred to a certain general type in the desert. It is certainly difficult to accept that all the characteristics of the members of a given strain, as they existed 200 years ago in the desert, can be recaptured by inbreeding one family deriving the strain name from one or two mares and containing contributions from many other strains along the way. To risk being repetitious, this absolutely is "breeding a strain" or "strain breeding." The questions are whether the Bedouin practiced "strain breeding" in this sense - and if they did how closely modern horses bred within a named strain resemble their desert progenitors.

I'm for it, every time, in the 3rd dictionary sense. If you aren't trying to develop "a special line of individuals .... maintained at a high standard of perfection by selection" then I don't want you breeding Arabian horses. If thinking in terms of strains helps you to reach this goal, then go to it. On the other hand, of course, if confusion over "family strains" gets in the way of emphasis on the "selection" aspect, then give up family strains by all means.

The Bedouin seem to have done just fine without them until at least the 14th century when the Arab type was already numbering its age in the thousands of years.