The Time Scale of Speciation

As we think about the evolutionary transition between major groups, say between dinosaurs and birds, we can agree, along with Darwin, that a major change can only occur by natural selection if small changes are accumulated in small steps along the way. In some real sense, that transition is going to be slow in time (biological time), and slow in rate of morphological change. And given that small changes are going to accumulate, the transition will be gradual.

How does that square with the data presented in lecture 10, that showed that speciation events could sometimes be very rapid? Part of the answer is that speciation is rapid in geological time, slow in "real" time. No matter how you slice it, many generations are needed for a new species to emerge.

[ASIDE. (Hundreds? that would imply at least 20,000 years for a new human species to separate. And we have documented evidence that a small population of humans was isolated on the island of Tasmania for 10,000 years (prime situation for a speciation event) without losing the ability to interbreed with other humans. Darwin suggested a ballpark figure [though he would never have used such a phrase] of 1000 generations, which would mean 200,000 years to produce a new human species. That's probably too much.]

Darwin's "slow, gradual" transition to a new major group has been misused to imply that he thought speciation was not only slow, but represented gradual and uniform change. This is nonsense, and he clearly stated in Chapter 4 of "The Origin of Species" that speciation might sometimes be rapid and sometimes slow.

I must here remark that I do not suppose that the process [evolution by natural selection] ever goes on regularly... nor that it goes on continuously; it is far more probable that each form remains for long period unaltered, and then again undergoes modification.
Stephen J. Gould and Niles Eldredge pushed for nearly 30 years the idea of "punctuated equilibrium", which says that species typically do not change over time. They may split to produce new species, but by the extreme form of allopatric speciation, where a small isolated peripheral population gives rise to a new species very quickly. They imply that the gradual evolution of a new species is so rare that it can be ignored. They love the idea that the small peripheral new species will extend back over the range of its ancestor, displacing it in what will look like a revolutionary event (since the speciation event took place outside the normal range of the species).

Actually, there's no reason to deny the possibility that a small isolated peripheral population could evolve superior adaptations, and migrate back to interbreed and pass them all across the parent population. It's not that situation, but imagine what the fossil record of California will look like, with successive waves of immigrants beginning with the first native Americans, then Spanish, then Anglos, then Asians..., but all members of the same species.

How fast is evolution?

In his book on macroevolution, Stanley cited several historical and ecological studies which he used to argue that evolutionary change could be rapid enough to generate new species very quickly. Data from fossil lineages are more equivocal, because one person's species may be another person's population or genus. If a paleontologist has a narrow concept of a species, evolution may appear to be very rapid, and a broad species concept will give the impression of slow evolution. The only way to avoid this problem is to try to be as uniform as possible when classifying within any clade: for example, within mammals. Then even if the taxonomic categories have only a certain grade of reliability (between 0 and 1), at least they can be used to assess evolutionary radiations, extinctions, diversity changes through time, and so on, with some uniformity within that clade.

Paleontological data have been used to argue that some groups of animals have evolved at a characteristically faster rate than others. Stanley and others have suggested that mammals evolve faster than bivalve molluscs, for example, and that mammals evolved faster during the Pleistocene than they did earlier.

Stanley argued that adaptive radiations happened too quickly to have resulted from "normal", "slow", phyletic or gradual evolution. Evolutionary rate is measured in darwins, which are units of morphological change in unit time. Gingerich (1983) compiled over 500 studies that measured change, ranging from genetic experiments in bottles to 350-million-year fossil lineages. He found that there was an almost exact inverse correlation between the measured maximum rate of evolution and the time over which it was measured. And when measurements are corrected for this factor, invertebrates have evolved two or three times faster than vertebrates; Pleistocene mammals evolved at about the same rate as earlier ones. Episodes in which a population colonizes a new environment typically have large rates of evolution which are not maintained for long and apparently can't be maintained: such rates would change a mouse into an elephant in 10,000 years.

Gingerich's data suggest that organisms can evolve over short periods at a much faster rate than they normally do. Classical theory suggests that the variability in any population is constrained at every generation by differential selection on the less fit individuals. If the less fit individuals are at extremes of the variability, then selection will act to maintain the population at equilibrium with its current morphology, with the characters of the population close to a mean. Stasis in morphology simply implies that there is no change in the pattern of differential selection that is sufficient to cause observable morphological change. The difference between this sort of stable situation, and the evolutionary stasis visible on a geological time scale, is only a matter of degree: long-term stasis implies either long-term stability in the environmental parameters that would select for changes in morphology, or a morphology adapted to be tolerant of the environmental changes that did occur.

Major revision begun, January 16, 2000

Return to Evolution topics

Return to Geology Department home page