MACROEVOLUTION: TEMPO AND MODE. I

We now consider the tempo (i.e., rate, and any modulation thereof) and the mode (i.e., the particular form or manner) of evolution. The use of these two words to focus the study of evolution is attributed to George Gaylord Simpson who's book, Tempo and Mode in Evolution, brought a paleontological perspective into the Modern Synthesis and applied the thinking of population variation and genetics the patterns of the fossil record. Thus, Simpson attempted to show that macroevolution (evolution above the species level) could be accounted for by familiar mechanisms of microevolution (genetic changes within population). He begins with the question: "How fast, as a matter of fact, do animals evolve in nature?"

To answer this question we consider evolutionary rates. Simpson identified two general classes of evolutionary rates: Phylogenetic (or morphological) rates measure the rates of evolution of characters (single or complexes) within phylogenetic lineages and are quantified as measured change in a character(s) per unit time; taxonomic rates measure the rates at which taxa replace one another in the fossil record and are quantified by two reciprocal methods: 1) number of taxa originating and going extinct during a span of time, or reciprocally 2) the average number of years a taxon remains extant (more below).

Potential problems when taxonomy is based on morphology: high rate of morphological evolution may lead to the division of organisms into many chronospecies. This can result in a high rate of taxonomic evolution: with pseudoextinction and pseudoorignination, taxonomic rates of evolution can be correlated with morphological rates of evolution (graph A). But, the reverse (morphological rates correlated with taxonomic rates) will not necessarily hold in cases where real extinction of taxa, and replacement by newly evolved taxa, occurs at a high rate. The latter does not assume there is continuous change specific morphologies, thus we could have high rates of turnover of taxa without extensive morphological change (graph B).

The relationship between morphological and taxonomic rates also depends on which characters are used to determine taxonomic status. One character may evolve rapidly in a lineage but the lineage may not be split into chrono "genera" because "important" characters have not changed, i.e., there can be mosaic evolution.

Haldane proposed a unit of measure: 1 darwin = change by a factor of e per million years (e = base of natural logs = 2.71828). See page 553 and figure 20.1, pg.554.

Horse teeth: 0.04 darwins = 4% per million years

Triceratops lineage of dinosaurs: 0.06 darwins

Rates appear to vary between different groups. There are apparent living fossils that have changed very little, or not at all, in millions of years: the coelacanth, the horseshoe crab (Limulus, see fig. 20.11, pg. 575) and the tadpole shrimp (Triops) are good examples. Other species or groups have evolved relatively rapidly (Horses, see figure).

Rates of change also can vary during the evolution of a lineage. In lungfish a "score" was tabulated for each taxon as to whether it possessed ancestral or derived traits. This score was plotted against the age of the taxonomic group for which the score was tabulated. Resulting graphs (See fig. 20.10, pg. 575) show a rapid loss of "primitive" characters (=acquisition of "new" characters) through time and the slope of this curve shows a peak early in the lineage.

The point is that morphological rates of evolution can be very different in both tempo and mode in different lineages of organisms.

Some of the between-group differences are real and some are an artifact of temporal scale. Gingerich recorded rates of change in selection experiments, colonization events, post-Pleistocene changes and long-term changes (domains I, II, III, IV in table figure below) and plotted them against the measurement interval in years. The clear relationship indicates that changes measured over short time spans exaggerate the changes one might predict if carried out for a long time. There are reversals of morphological trends and periods of no change (="stasis"; next lecture) that reduce the rate of change when averaged over a long time period. See table and fig. pp. 556-558.

Taxonomic rates can be quantified as number of genera / time span = average genera per million years, or one can take the reciprocal of this and ask: How long does a genus last? and thus quantify the duration of a taxon. By tabulating the period of first appearance of a taxon (genus, lets say) and the period of its last appearance in the fossil record one can obtain such taxonomic rates (see table below for example of data). In Pelecypod mollusks (bivalves) 13 genera appeared in the Ordovician and disappeared in the Ordovician; 4 genera appeared in the Ordovician and disappeared in the Triassic. Compiled over the entire data set the "average" genus of Pelecypod lasts 78 million years. Similar calculations for Carnivores show that in this group a genus lasts about 8 million years. See data below and compare with fig. 20.13, pg. 578.

These data can be plotted on survivorship curves which tabulate the percent of taxa alive today that originated at the time indicated on the horizontal axis (broken line). For extinct taxa the graphs show the percent of taxa with a known duration > the value indicated on the horizontal axis. Thus 100% of the extinct genera have a duration greater than or equal to 0 years and 0% of the Pelecypod genera have a duration greater than or equal to 275 million years; other genera have intermediate durations. Note the shorter range of the horizontal axis in the two graphs and the different shapes of the extant vs. extinct survivorship curves in the two graphs.

The point is that the taxonomic rates of evolution can be very different in both tempo and mode in different lineages. While the data are sound the interpretation of the data as a general phenomenon holds to the extent that the unit of a genus is comparable in the two lineages. Since the duration of genera in the two groups differs by about a factor of 10, this large a difference suggests that there is a real difference. In other comparisons with more subtle differences between lineages, the unit of measure (genus, family?) could become an important consideration.