Crossbreeding Draft Horses
by Karen Gruner

Interesting things happen when we crossbreed animals. We have all seen litters of crossbred puppies with wide variations in appearance. We have also seen litters of crossbred puppies that are remarkably uniform. The same thing happens when crossbreeding horses, but since we get at best one offspring per mare per year, the results are often harder to see than when looking at puppies.

Over the last 70 years considerable change in type has occurred between the Brabant and American Belgian. We can all see this change, and it is this difference that has driven myself and others to choose the Brabant over the American Belgian.

Phenotype vs. Genotype
The difference we can see between these two types of horse is called the phenotype. Underlying what we can see is a genetic base called the genotype. The genotype determines the phenotype. So all animals (ourselves included) have a genetic makeup of specific chromosomes called the genotype and an external appearance called the phenotype.

When crossbreeding, considerable variation in both genotype and phenotype is possible, even between offspring from the same set of parents. I will try to explain what the possible genotype of crossbred animals could be, using Brabants and American Belgians as my example. The genotype of a 75% Brabant horse could vary between 50% and 100%, making the 75% Brabant designation only an average measure of the genetic type of the crossbred animal. We see this all the time. If you have seen a number of Brabant-American Belgian cross horses you know that all 50% or 75% Brabants do not look alike. How can some 50% horses be indistinguishable from 100% horses, while some 75% horses look distinctly crossbred?

Chromosome Pairs
Horses have 64 total chromosomes. The chromosomes come in homologous pairs (one chromosome from the dam and one from the sire with the same function). So horses have 32 pairs of chromosomes. The chromosomes are generally numbered by pairs, so the chromosomes are called chromosome 1 to chromosome 32, each number corresponding to a pair. For the sake of simplicity, let's consider five chromosome pairs, and let's call them Bi for Brabant and bi for American Belgian, where denotes a number 1 through 5. Let's mate our hypothetical 5 chromosome horses, a Brabant stallion to an American Belgian mare. The genotypes of these horses would be:

When the sex cells are formed (meiosis), the pairs split. Each egg the mare produces gets one chromosome from each pair. Similarly, each sperm cell the stallion produces gets one chromosome from each pair. The following egg and sperm cells would be created:

Remember, this discussion is over simplificatied, since the stallion really has two slightly different B1 chromosomes, one from his dam (B1d) and one from his sire (B1s), and on down the line for all the chromosomes. So when the sperm cell is made, it could have B1d, B2s, B3d, B4d, and B5s, or any other combination. The same goes for the mare. For the sake of this discussion, we won't take this into account. Another interesting complicating phenomenon occurs during meiosis called crossover. During crossover, sections of a specific chromosome, called genes, can exchange between the Bis and Bid chromosomes, so the resulting chromosome becomes a blend of sire and dam. Again for this discussion we won't take this into account. When this sperm and egg unite to form the 50% Brabant offspring, the following genotype results:

F1
B1b1
B2b2
B3b3
B4b4
B5b5

First Generation
Every F1 (first generation) Brabant-American Belgian cross has 50% Brabant chromosomes (Bi) and 50% American Belgian chromosomes (bi). These horses tend to look like crossbreds, although the Brabant chromosomes seem to be rather dominant to the American Belgian chromosomes. As a result 50% F1 crosses have a nice Brabant type (phenotype).

What happens when we cross two F1 50% Brabant horses? Now things can get very interesting! During meiosis (formation of egg and sperm cells) the homologous pairs of chromosomes split. Each egg and sperm carries half the genetic information of the parent. Each egg and sperm has one of each of the pairs. A minimum of 2x2x2x2x2 = 32 possible different genotypes can occur for the gametes (eggs or sperm) in our five chromosome pair example. Twelve of the 32 possible egg or sperm types are shown below.

Most of the possible gametes are a combination of Brabant and American Belgian in genotype. Of course pure Brabant gametes (type 1) may be formed, as well as pure American Belgian gametes (type 6). Any combination in between can also be formed. So we see right away that if an egg of type 1 is fertilized by a sperm of type 1, our supposedly 50% second generation (F2) offspring is genotypically a full Brabant. Conversely, two type 6 gametes could fuse to create a 100% American Belgian offspring. But most of the possible gametes carry both Brabant and American Belgian genes, so most of the offspring will be cross-bred. In fact, on the average, we expect the offspring to have 50% Brabant chromosomes and 50% American Belgian chromosomes. But it is clear that there can be a wide variation in appearance between F2 50% cross-breds since there is a wide variation in possible genotype. I fact, there are theoretically 32 x 32 = 1024 different possible genotypes just for our 5 chromosome example!

Now let's look at what can happen with a 50% Brabant crossed with a 100% Brabant. Note that all 75% Brabant horses are necessarily F2 (or higher) crosses on at least one side. For our example we'll let the mare be the full Brabant and the stallion be the crossbred. The stallion can produce the sperm cells shown above plus the other 20 possible combinations not shown. The mare can only produce full Brabant eggs, so each of the offspring is guaranteed to be at least 50% Brabant. If the egg unites with a sperm cell of type 1, the offspring will be genetically full Brabant. If, however, the egg unites with a sperm cell of type 6, the offspring will be genetically 50% Brabant. When we look at 75% Brabant crossbreds we therefore often see a wide variation in appearance (phenotype), even among full sibling offspring. Again, since most of the sperm cells a 50% Brabant stallion produces have both American Belgian and Brabant chromosomes, we expect that on the average the offspring will be 75% Brabant.

So when we quote percentage Brabant for a horse other than an F1 cross, we are reporting only the average genetic makeup. We do not really know if the horse is 50%, 75%, or 100% Brabant.

Karen Gruner is editor of "Trace Chains," the newsletter of the American Brabant Association, from which this article has been reprinted with permission.