This post is part of a series on The Origin Of Species. It was originally posted on the old blog in feb 2009, during the Darwin 200 celebrations.
In chapter nine, Darwin takes a long look at hybrids. And I mean a long look: he can really go off on one when he’s enthusiastic about a subject. After all of those chapters on natural selection, I could not at first understand why the topic of hybridisation came up — thirty pages just to explain why hybrid sterility can not be an adaptation? — before I remembered that Darwin’s remit is to explain the origin of species, not just the process of natural selection.
A major point of the chapter is to explore the proposed idea, which Darwin had once found attractive, that the sterility of hybrid species is an adaptation. It was proposed that the sterility of hybrids was beneficial as it prevents different varieties blending together, diluting incipient species. Darwin puts across the evidence for hybrid fertility being a much more complicated situation, and one not suggestive of sterility being an adaptation. But he could have skipped the dry details and gone straight to the bit where he points out that, since evolution requires reproduction, deliberately reducing one’s legacy is not likely to be adaptive: “… now, what is there that could favour those individuals which happened to be endowed in a slightly higher degree with mutual infertility?” The argument for hybrid sterility being adaptive is a group- or species-selectionist one. (Actually, it might be possible to imagine that, under certain very restricted ranges of conditions, kin selection could favour hybrid sterility — nevertheless, in general Darwin’s argument stands.)
More interestingly, Darwin goes on to hint at the actual mechanism of and “reason” for hybrid infertility. It is not adaptive, but, as we now know, it is merely a constraint of the way that genomes are organised and development progresses. As Darwin notes, first (F1) generation hybrids are quite capable of producing gametes (egg and sperm), but the second (F2) generation produced from F1 gametes tend to fail somewhen during development. We now know that the process of recombination, which occurs during the production of gametes is one key to understanding this.
Our genomes consist of one set of chromosomes inherited from each of our parents. When gametes are produced, these need to be cut down to a single set that we pass on (otherwise the genome would double in each generation!). In recombination, the chromosome you inherited from your mother is lined up with its counterpart inherited from your father. The two contain pretty much the same genes in the same order, though there may be some within-gene variation between them. Sections of the chromosomes are then “crossed over” and shuffled between them, and the result is a new chromosome containing the same genes, in the same order, but in a novel combination of varieties.
Recombination causes a problem in hybrids, because the chromosomes of, say, a horse and a donkey, may have diverged, and might no longer contain the same genes in the same order. The F1 mule develops fine — it has a complete copy of both genomes — but a chromosome produced during the production of gametes in that mule might be missing genes crucial for development, or contain too many copies of some genes. Development is achieved through complex and sensitive systems of gene expression, and when genes go missing, or their numbers get upset, these systems are liable to fail, and, in the case of animals, the developing F2 embryo will miscarry.