The New Scientist started a new essay this week
In 1856, geologist Charles Lyell wrote to Charles Darwin with a question about fossils. Puzzled by types of mollusc that abruptly disappeared from the British fossil record, apparently in response to a glaciation, only to reappear 2 million years later completely unchanged, he asked of Darwin: “Be so good as to explain all this in your next letter.” Darwin never did.
While preparing my new lectures on evolutionary medicine, I also have this problem – evolution is indeed a chaos theory. And I agree by heart
But the neat concept of adaptation to the environment driven by natural selection, as envisaged by Darwin in On the Origin of Species and now a central feature of the theory of evolution, is too simplistic.
A paper in the same issue clarifies it further
The evolution of life has many characteristics that are typical of non-linear systems. First, it is deterministic: changes in one part of the system, such as the mutation of a DNA base, directly cause other changes. However, the change is unpredictable. Just like the weather, changes are inexorable but can only be followed with the benefit of hindsight.
Second, behaviour of the system is sensitive to initial conditions. We see this in responses to glaciations in the Quaternary period. The exact circumstances of the beginning of each interglacial determine the development of the whole period, leading to unpredictable differences between interglacials (Quaternary Science Reviews, vol 14, p 967).
Third, the history of life is fractal. Take away the labelling from any portion of the tree of life and we cannot tell at which scale we are looking (see diagram). This self-similarity also indicates that evolutionary change is a process of continual splitting of the branches of the tree.
Fourth, we cannot rewind, as Stephen Jay Gould argued in Wonderful Life. Were we to turn the evolutionary clock back to any point in the past, and let it run again, the outcome would be different. As in weather systems, the initial conditions can never be specified to sufficient precision to prevent divergence of subsequent trajectories.