One Monday, a chemist was researching the properties of a new explosive. He weighed it carefully, ignited it, and then weighed the product. He was astonished to find that the product weighed more than the starting materials.
"I must have missed something," he said. "Certainly this result is not enough to overturn the well established atomic theory of matter."
He soon realized that he had forgotten to account for the mass of the air, and everyone agreed that it was prudent for him to re-examine his work.
The next day, a physicist was studying transmission of light through a new substance. When he completed his experiments, it seemed that the light was coming out of the substance before it had gone in.
"I must have missed something," he said. "Certainly this result is not enough to overturn Maxwell's electromagnetic field theory."
It took him some time, but finally he found an error in an equation. Everyone agreed that it was prudent for him to re-examine his work.
On Wednesday, a biologist was studying the genome of a bacterium. He was amazed to find that the genome had more similarity to a certain species of fungus than it did to other bacteria, even though he had expected it to be a typical bacterium.
"I must have missed something," he said. "Certainly this result is not enough to overturn the well established theory of evolution through natural selection."
"Just-so-stories!" screamed one onlooker. "The bacterium must have been designed!" shouted another.
Why oh why oh why does the study of biology get such special attention? Scientists make mistakes. More importantly, scientists, even smart ones, are not always right. When they predict something based on a well established theory, and that something turns out to be false, it is prudent to re-examine their work to see what might be wrong before chucking the well established theory. Of course, if it isn't possible to explain the evidence in the framework of the well established theory, or if the explanations require twists and turns that can be more simply explained by some other theory, than even a well established theory can be overturned (for example, it was well known in the 19th century that light propagated through a medium called the "luminiferous aether," but then the Michelson-Morley experiment showed that the aether did not exist, and a new theory had to replace aether).
Evolution is supported by piles and piles and piles of evidence, from molecular biology (both DNA comparisons and protein comparisons) to paleontology (the fossil record isn't complete, thanks to the way fossilization works, but everything in it supports descent with modification) to direct experimental observation (not to mention years of artificial selection, which is just a form of natural selection, in which humanity takes the role of the environment). And, as creationists strangely like to use as an argument, elements of the theory of evolution through natural selection are tautologous, or, as m-w.com puts it, "true by virtue of its logical form alone." Of course the organisms that are more likely to pass on their genes pass on their genes more than the organisms that aren't.
So, when we find a single organism that does something weird, of course biologists attempt to explain it--not explain it away as creationists like to say, but explain it--by fitting it into the framework of natural selection, by figuring out what selective advantage its weirdness gives it. Until someone finds something much, much stranger than anything we've found so far, natural selection is by far the best explanation we have for the diversity-yet-clear-relatedness of life.