NABT Convention, Reno, NV, Nov.4, 1998
NABT Convention, Reno, NV, Nov.4, 1998
The study of macroevolution, or evolution above the species level, involves a change in scale and a change in approach from that of microevolution. Macroevolution is concerned with origins, diversifications and extinctions over the sweep of geologic time: what the embryologist C.H. Waddington called "The whole real guts of evolution, how we get horses and tigers and things. " The invasion of land by plants and animals, the derivation of wings in bats, pterosaurs, and birds from the forelimbs of their respective ancestors, and the evolutionary explosion of mammals after 120 million years of life in the nooks and crannies of the dinosaurs' world, are just a few of the dramatic events that fall under this heading. This means that much of macroevolution, like astronomy, is an historical science, a matter of reconstructing history and testing hypotheses of ancient cause-and-effect using evidence from the fossil record, embryology, molecular biology, and other geological and biological sciences. What is sometimes not appreciated is that these different lines of evidence tend to converge on a single picture; their disagreements are almost always in the details, and even the larger remaining points of contention make sense in terms of the weaknesses and biases of one style of research or another. I'm going to review a few major points in order to clarify the outlines of an emerging macroevolutionary consensus (while acknowledging that this a very active field and that consensus does not imply unanimity).
1. Speciation is too slow to be easily observed by the biologist, but from the perspective of the fossil record, speciation is often too fast!
The fossil record contains many beautiful examples of the evolutionary transformation of one -- morphologically defined -- species to another, but these mainly occur in a few environments that are particularly good at recording the fine details of a time sequence: in lakes that accumulate annual beds of silt, for example, or in deep-sea deposits that record the steady rain of plankton from the surface waters. In contrast, the land-surface is more continuously subject to erosion than the seafloor, with catchbasins like lakes and swamps tending to be geologically short-lived, so the terrestrial record will be patchier and less complete than the marine record. The operation of plate tectonics, consuming the Earth's crust in subduction zones and smashing continental edges into mountain ranges during collisions, creates a record that becomes more patchy with age: the vast Precambrian interval is the first 80% of earth history, but it now provides only 10% of the rock record. This is a basic point about the nature of the fossil record: it's undeniably incomplete and imperfect (as is our knowledge of the living biota, of course), but it's imperfect in ways that make sense from what we know about how sediments accumulate and rocks form. Similarly, the nature ofthe record is a logical consequence ofthe biology ofthe organisms that contribute to it. Hardparts more readily resist physical and chemical destruction, so the record of organisms with shells, teeth, or tough pollen grains is more complete than the record for flimsier creatures. Rare species will be less frequently preserved than abundant ones, localized species will be less frequently preserved than widespread ones. Common-sense rules like these (which are now being put into quantitative terms) go a long way towards explaining the unevenness of the fossil record: fabulous for shelled, marine microplankton, not very good for dinosaurs or early hominins, and miserable for earthworms and slugs.
2. "Punctuated equilibrium" is a hypothesis about evolutionary change at the species level.
This most famous of macroevolutionary concepts has been wildly overextended and misused, but it simply states that species tend to be morphologically static over most of their histories, and that most changes in form occur in close association with the geologically-rapid splitting of populations into new species. Everyone now agrees that both stasis and splitting are common, but that gradual change and non-branching evolution also occur and may even dominate in some situations; notice these alternatives can be tested even if the fossil record is bunched into packages separated by gaps of missing or fossil-free rocks. We can recognize stasis in species form even using widely spaced samples (one photo per decade can demonstrate that the Statue of Liberty hasn't changed much since it was installed in New York Harbor), and we can detect lineage splitting whenever the ancestral species outlives the first appearance of its descendent. Note that this doesn't require that all new species be morphologically distinct, only that species-level changes in morphology arise at splitting events, and not by continuous change of the kinds of large, widespread populations that can be studied in the fossil record.
The most exciting research in this area now focuses on testing for regular patterns in the distribution of stasis versus gradual change, and in splitting versus unidirectional evolution, among species in different major groups, life-habits, environments, or regions; on testing hypotheses for the mechanisms of stasis (the punctuations are geologically rapid but slow on biological timescales and thus consistent with an array of speciation mechanisms); and on exploring the macroevolutionary implications of stasis and lineage-splitting (some argue, for example, that if species are static through much of their history, then large-scale trends, e.g. from primitive to modern horses, must arise via differences in speciation and extinction rates among different sublineages within a larger group).
3. The topology of evolution is a bush, not a ladder.
As biologists and paleontologists gain a fuller picture of the large-scale outlines of evolutionary history, it has become abundantly clear that our basic view of evolution at this scale should not be a ladder or any other icon of directional, progressive change, but a bush with many branches, stems and twigs. This in no way undermines the role of natural selection and other forces at the population level, but when we take a step back and look at the broader outlines, evolutionary lineages must diversify if they are to withstand even the relatively low levels of "background" extinction that prevail over much of geologic time. Complex evolutionary transitions, e.g. from dinosaur to bird, or from small, multi-toed ancestral horses to their large, hooved descendents, do not occur as single mutational steps, or as a simple parade of increasingly more modern forms. Instead, these major shifts generally occur within a swarm of related lineages.
4. At higher levels, the fossil record is rich in " missing links."
We can hardly expect the fossil record to capture every single species within those swarms of transitional lineages, but we have a spectacular sample of intermediate forms between the many of the major groups of plants and animals. For example, an esquisite sample of intermediate forms are now known for the transitions from "fish" to tetrapods, from "reptiles" to mammals, dinosaurs to birds (or, strictly speaking, non-avian dinosaurs to birds). Intermediate fossils for many other lineages within these groups, for example whales, horses, and for that matter humans, are also being increasingly well-documented as exploration and analysis continues. Mosaic evolution is clearly the rule, that is different anatomical features (and presumably behavioral, physiological...) evolve at different rates and different times. Thus, the earliest known bird, Archaeopteryx, has features typical of modern birds like feathers and a wishbone but retains the dinosaurian tail, clawed forelimbs pelvis and teeth of its ancestors; slightly younger forms evolved a more birdlike, perching hindlimb and pelvis, but still retain those teeth and forelimb claws, and so on. Human evolution operated in same way, with upright posture evolving before the short face and large cranial capacity of modern humans.
5. Major changes in morphology don't always require major genetic changes
Remarkable advances in developmental biology are beginning to merge with the study of macroevolution. We now understand that strilkng differences in morphology and behavior need not be correspond to massive genetic differences, but instead may arise by relatively modest changes in the timing, duration, or location of gene expression. Development of fertilized egg into complex, multicelluar adult is orchestrated by a complex hierarchy of genes, which is now being probed by molecular developmental biologists. Many of the major control genes, such as those that establish the body axes of the embryo or that, like the Hox genes, that provide positional assignments along the anterior-posterior axis, are conserved throughout all animals. Changes in the expression of those control genes must have been involved in some of the major evolution transitions during the initial radiation of animals. Such changes have also been implicated in more modest changes as well, such as changes in limb morphology in arthropod evolution. These control genes result in an embryo that develops in modular fashion, with semi-independent regions, so that limb development, for example, can proceed even if jaw development has changed. This helps to explain the mosaic evolution pattern mentioned above. The role of gene regulation in shaping major evolutionary changes also helps to explain how apes and humans can be genetically so similar (differing by only 50 genes out of ca 80,000 by a recent estimate): our differences with our closest living relatives must reside almost entirely in changes in the expression pattern of genes.
Despite this modularity and the existence of major regulatory genes, development is not infinitely flexible: an embryo must retain viability at every step and must give rise to an integrated adult body. This constrains the direction of feasible evolutionary changes.
Further, because some aspects of development can become deeply entrenched as precursors to later developmental steps, the traces of past evolutionary history are almost always retained. Thus, when reptiles went back to the sea and evolved into ichthyosaurs, and when mammals evolved into dolphins, they converged on a shark-like morphology owing to the stringent adaptive requirements of their predatory marine lifestyles, but they retained many tell-tale reptilian and mammalian characters. Developmental programs are too complex, and evolutionary history too long and peculiar for each lineage, to permit true evolutionary irreversibility at this scale.
6. Major evolutionary events are pulses in geologic time
One of the triumphs of evolutionary paleobiology has been the clear documentation that the history of life has not been a steady increase in numbers or a simple linear trajectory towards the modern world. Instead, the story is one of rapid radiations, long plateaus in biodiversity, and mass extinctions. The Cambrian Explosion marks the appearance of most multicellular animal designs (all but one of the living phyla having preservable skeletons plus a number of problematic forms) within a 10-million-year window starting about 530 million years ago (Ma). This event, which lasted less than 0.5% ofthe history of the Earth to that point, and less than 2% of the time from the base of the Cambrian period to the present day, certainly represents a geologically explosive appearance of animal body plans, recognizable not only in tallies of phylum-level taxa but in analyses that directly quantify morphological variety. The relation of those first appearances to their actual time of evolutionary origination is a hot topic, but the simultaneous increase in the number, size and complexity of tracks and burrows in and on the sediment is strong evidence that a major part of the evolutionary action was in fact within the Cambrian explosion interval; the trigger mechanism is a focus of intensive research. We have an increasingly good picture of the run-up to the explosion, with evidence of early metazoans in the form of minute trails from rocks as old at 600 Ma in the late Precambrian (Neoproterozoic), astonishingly well-preserved eggs and embyros exhibiting well-defined cleavage stages at around 570 Ma, more elaborate trails and burrows at the re-defined base of the Cambrian Period at 543 Ma, and a steadily expanding diversity of small shelly forms from 543 to the explosion proper at 530 Ma. One minor lineage recorded within the great range of new body forms were the primitive chordates, which eventually gave rise to the major vertebrate diversifications.
Additional pulses of diversification -- though none so dramatic at the Cambrian explosion -- represent events that opened major ecological opportunities to evolutionary lineages. The invasion of land by plants, invertebrates, and finally vertebrates, is followed by waves of evolutionary experimentation and diversification. We are increasingly coming to realize that mass extinctions play an important role in evolution, by removing dominant forms and providing opportunities for the survivors to diversify in their place. The exuberant radiation of mammals after the demise of the dinosaurs and related forms at the end of the Mesozoic Era (at the end of the Cretaceous Period) is the most famous example' but similar patterns are seen after each ofthe "Big Five" mass extinctions. The marine fauna familiar to today's beachcomber and skin diver was profoundly shaped by the great end-Paleozoic mass extinction (at the end of the Permian Period), which removed roughly 95% of marine species and permanently altered the balance of life in the seas. This is not to say that the accelerating extinctions of the present-day faunas and floras under pressure from human activities can be seen in a natural or positive light: recoveries from mass extinctions are painfully slow on human timescales (5-10 million years for reef systems for example), and the most persistent survivors need have no relation to human needs.