Jim writes: > Sorry, the scenario is to poorly defined to say anything about it, and > there is probably no contradiction. But, there is also no reason to > think that natural selection is always in action. And, certainly, > natural selection CANNOT select for extinction.
In my continuing bid to become the group's curmudgeon, let me say that natural selection can quite easily select for extinction. Natural selection judges only whatever advantages it finds in populations in the moment. What it truly never does is assess the long-term consequences of its preferences. Although there are probably a dozen hypothetical scenarios where natural selection of one attribute or another could drive a lineage to extinction, let me just mention two that are well-known and reasonably well-documented. The first is the reversion of a sexual lineage back to parthenogenesis. Doing this offers the lineage a number of hypothetical advantages, most especially freeing itself from the burden of maintaining males, who often represent substantial ecological costs to the species and who quite frequently do not participate in the economy of deme. A population free of males is also capable of rapid expansions into recently vacated territories. It can also survive in extremely adverse situations where a sexual population would go extinct, simply due to low population numbers and the difficulty in finding a mate. The second condition is the evolution of high-order polyploidy. High-order polyploidy seems on the surface to be an excellent information-assurance mechanism, mitigating the informational corruption of any body of information that is replicated generation after generation indefinitely. While we find both types of populations in nature, their rarity is prima facie evidence that they are not strategies that are successful on the long-term, and that any lineage that adopts them for whatever short-term gain it may accrue also soon disappears. Both mechanisms so evolutionary stabilize a lineage that it cannot adapt to changing conditions. Reversion to parthenogenesis is relatively common in the arthropods, but it also is known to occur in vertebrates as complex as reptiles and birds. It's very rare in these animals, but it does occur. However, the phenomenon is unknown in mammals, and I have long attributed the evolution of differential "imprinting" of the chromosomes that pass through either maternal or paternal gametogenesis to be an evolutionary brake that prevents a reversion to parthenogenesis in mammals. Syngamous chromosomes derived from either gender have been rendered incapable of producing a viable individual because some critical information has been suppressed on one chromosome or the other. Only when the chromosome is matched with the complementary gender's is the library complete and embryogenesis allowed to go forward. On a second. related subject, two people wrote privately and asked if I had a reference for the barn fly story that I told. Unfortunately I don't. I heard the story at the XII International Congress of Entomology at Canberra in 1972, as a contributed talk. Only the keynote and plenary talks were published. If there is any published work on the subject somewhere, it is probably published in an agricultural bulletin somewhere in Australia. Nonetheless, there is another virtually identical story regarding the CCR5 chemokine receptor in human immune systems that is more current and a great deal more readily available. Ordinarily, the CCR5 gene appears to be involved with the inflammatory immune response and thus serves an important purpose, but in some people bearing one particular allele of the gene, the gene is defective for its primary purpose. Very similar to the fly story, this defective allele also cripples one of two receptor molecules that the HIV virus requires when infecting a macrophage, and thus the homozygous bearers of the defective CCR5 appear completely immune to HIV infection, rendering them as completely protected from this plague as were the flies with the longer tarsal hairs. As this article from the CDC states: "At least 23 alleles have been described for the coding region of this gene, and most of them are very rare. The most common and most studied is the 32 allele, a 32 base pair (bp) deletion that confers almost absolute protection from infection with macrophage tropic (M-tropic) viruses in homozygous individuals and provides an average 2 to 3 year delay in the progression to AIDS in those heterozygous for the deletion." --http://www.cdc.gov/genomics/hugenet/factsheets/FS_CCR5.htm If you're interested, you won't have any trouble finding articles on this example, where once again a mild genetic defect (in normal circumstances) proves to be of great benefit in a shifted environment. Indeed, if the selection coefficient were as strong in humans due to HIV as the toxin was to the flies, we too would all soon be carriers of the modified allele and HIV would be reduced to a childhood disease. Wirt Atmar
