Behavioral geneticists frequently make the statement that full siblings, on average, share 50 percent of their genes, or have 50 percent of their genes in common. Following their lead, I will use the same shorthand way of describing the genetic relatedness between siblings. First, though, I will explain what is meant by that statement, which many readers find confusing.

It is confusing because even unrelated people share far more than 50 percent of their genes; in fact, they share about 98 percent of their genes with chimpanzees. But the genes that are essentially the same in all normal humans (many of which we share with other animals -- even fruitflies) are of no interest to behavioral geneticists. The genes that interest them are the ones that vary among humans: genes that determine eye color or nose shape, that play a role in variations in intelligence and personality, or that make one person more susceptible than another to various physical and psychological disorders.

Aside from random mutations, siblings are the same in all the genes that do not vary among humans. They are also the same in all the genes that do not vary within the population to which they belong -- for example, genes for eye color vary in European populations but most people in China carry only brown-eyed genes, so eye-color genes are unlikely to vary between Chinese siblings.

Of the genes that do vary within their population, about 50 percent do not vary between siblings -- two full siblings will have identical versions of these genes. These identical versions are exact copies of their parents' genes. The percentage of variable genes that full siblings have in common averages 50 percent but it can be more or less than that, because each child's genome is selected randomly and independently from the parents' genomes. By chance, two siblings may end up with considerably more than 50 percent in common (these are the siblings who, if they are close in age, are frequently asked, "Are you twins?") or considerably fewer.