Micro- or macro-
Darwin argued that species evolved from pre-existing ancestors. Changes in genes provided a mechanism for that to happen. The theory is good, but how strong is the evidence?
There is little doubt about 'micro-evolution' – adaptation within a species driven by genetic change. Mussels and snakes are good examples of micro-evolution in action.
But 'macro-evolution' – formation of new species by DNA change – has been questioned. Despite generations of selective breeding, all dog breeds may still be members of one species, even if practical difficulties sometimes inhibit successful mating...
Slow but sure
Evolution is a slow process, so witnessing it would be difficult. Generally, it is an inference: two organisms look alike and their genes are similar, so it is inferred that they are related and shared a common ancestor. Mutation creates genetic variation and natural selection (or drift) does the rest.
Sometimes, though, it is possible to see signs of speciation taking place. One example is the Heliconius butterfly family. More generally, detailed studies of habitats have revealed genetic divergence between organisms of the same species in different locations. A variety of selective pressures seem to be promoting a split into different species.
An alternative explanation is that every species was created independently, and can evolve but within limits. A mussel always stays a mussel. Similar organisms look alike because they were created with similar sets of genes.
Scientifically, that theory has little going for it. It would involve separate species-creation events – by currently unknown mechanisms – for every species that ever lived.
Razor sharp
In such cases a rule of thumb known as Occam's (or Ockham's) razor is often applied. This was a method put forward by 14th-century thinker William of Ockham. When presented with two possible explanations for a problem, choose the more simple – that which makes the fewest assumptions ('shave off' the more contorted explanations).
Occam's razor is not intended to be a 'golden rule' that can be used in every situation, however. More evidence supporting or contradicting a model would be preferable. Identification of the first 'speciation genes' adds further weight to 'natural' macro-evolution.