Scientific Bloopers
All things considered, humanity is doing pretty well. We are living in an era where virtual idols hold holographic concerts and mechanized artisans craft flawless sculptures from chunks of metal -- and even as I write this column, a mobile laboratory is roaming Martian soil in search of life, and a giant eye in space is scouring the depths of the universe for potentially habitable planets (the latter -- the Kepler space telescope -- acquires such immense amounts of data that it has to discard most, and even its trash heap has led to incredible discoveries). We might not have flying cars or biological immortality yet, but we're getting there.
But this was not always so. Heroin was once a popular cough syrup marketed by Bayer (a pharmaceutical giant and former holder of the patent for aspirin) and commonly administered to children; hysteria was thought to be a disease where a woman's uterus went rogue, leaped around her innards like a frog and forced her into illogical acts (I don't know about you, but it seems to me that a berserking womb would cause more dire complications than just psychological instability); and mental illness was often treated by scraping away at the brain (a particularly effective method of cerebral mutilation, which had an unfortunate tendency to turn the recipient into a bumbling idiot, even earned its discoverer a Nobel Prize). The road to mankind's success is paved with mistakes that today might seem outright insane, and scientists are doing their darnedest to pave the rest of the way -- which, obviously, means more well-intentioned madness.
Researchers, you see, are still human. No matter how acclaimed they may be, they are prone to making errors, and their highly publicized position means that their errors are often spectacular (much like the misstatements of politicians and football commentators, in a way -- as an aside, those interested in football ought to check out Colemanballs, ridiculous football comments made in the heat of the moment and featuring such gems as "Batistuta gets most of his goals with the ball," "Victor Hernandez, like an orchestral conductor directing his troops..." or "Michael Owen -- he's got the legs of a salmon"). Add to that the human tendency to defend one's opinions even in the presence of evidence to the contrary, and you get some very interesting claims -- such as how evolution is driven by interspecies mating.
One corner of this claim is horizontal gene transfer, exchange of genetic information between unrelated species. This is a common event, having led to aphids that derive energy from the sun via not-quite-photosynthesis, wasps with symbiotic, stinger-delivered viruses and even our acquisition of mitochondria. In the other corner are various zoological oddballs, such as tunicates (tadpole-like animals that digest much of their nervous system to become sponge-like, sessile water pumps -- they are close relatives of larvaceans, which also begin as tadpoles, but instead encase themselves in snot houses and never mature), rhizocephalans (parasitic barnacles that spread roots inside crabs and force them to take care of the fleshy sacs that appear where the crustacean's own egg sac would normally be carried) and insect larvae, all sharing a marked metamorphosis during the transition to adult form. The idea is that such drastic changes during growth resulted from the mating of two unrelated species, one of which became the larva while the other remained as the adult form.
This notion culminated in a stunning (in the physical sense -- I think the proper word is "bloodcurdling") paper, published in the prestigious journal Proceedings of the National Academy of Sciences, declaring that caterpillars evolved when a butterfly mated with a velvet worm (I don't have the heart to describe velvet worms, as I wish to spare you the mental image) and asking the scientific community to test the author's claims by trying to breed cockroaches with velvet worms -- given how a far simpler interbreeding experiment gave us killer bees, I fear to think what could result from that union (the fact that it is quite impossible offers only mild comfort). Needless to say, scientific outrage soon followed (the claims were untenable for several reasons, many of them physical), though I must also note that the very same author also reported successfully breeding sea urchins with tunicates (which is not so far off from breeding humans with sharks), so I suppose we should be getting ready for hippogriffs and myrmecoleons.
Such mishaps aren't endemic to biology, either -- ArXiv, the premier source of cutting-edge physics and mathematics research, features large doses of preposterous claims (ArXiv is not peer reviewed, and is used as a proof of authorship prior to publication in a journal), and one of the most notable is the suggestion by two eminent physicists that the Higgs boson is acting from the future to sabotage attempts at its discovery. Citing various incidents that befell particle accelerators with the potential to produce the Higgs, the authors suggest that the Large Hadron Collider should be operated by randomly choosing among run parameters written on a large number of cards -- if the card with "do not operate LHC" keeps coming up, it would mean that the Higgs is averse to discovery, and has the power to retaliate from the future (I would have suggested the use of a Ouija board, so that the boson can explicitly state its wishes).
Of course, many scientists are quite genial when confronting their mistakes. A geology paper, written in 1963, takes the cake when it comes to author responses: when attention was called to a minor error that had no bearing on their conclusions, the authors accepted the objection without delay (the norm nowadays is fighting to the last drop of blood to defend your point) and replied with a simple, "Oh, well, nobody is perfect."