Darwin's Doubt
Page 13
Even if we assume that mutation and natural selection (and other similarly undirected evolutionary processes) can account for the emergence of novel proteins and body plans, we cannot also assume that the protein molecular clock ticks at a constant rate. Unlike radiometric dating methods, molecular clocks depend on a host of contingent factors. As Valentine, Jablonski, and Erwin note, “Different genes in different clades evolve at different rates, different parts of genes evolve at different rates and, most importantly, rates within clades have changed over time.”35 So great is this variation that one paper in the journal Molecular Biology and Evolution cautions, “The rate of molecular evolution can vary considerably among different organisms, challenging the concept of the ‘molecular clock.’ ”36
Keep in mind too that molecular clocks are calibrated based on the estimated age of presumably ancestral fossils. If, however, such estimates are incorrect by even a few million years, or if the fossil used to calibrate the mutation rate does not lie at the actual divergence point on the tree of life, the estimated mutation rate may be badly skewed. Calibration of molecular clocks depends on an accurate understanding of the ancestor-descendant relationships between fossils and their presumed descendant taxa. If the fossil used to calibrate the divergence time of two later groups was not actually a true ancestor, then the mutation-rate calculation based on that fossil’s age may be grossly inaccurate. As Andrew Smith and Kevin Peterson note: “Molecular clocks are not error-free and come with their own suite of problems… . The accuracy of the technique depends upon having an accurate calibration point or points, and a reliable phylogeny with correct branching order and branch-length estimates.”37 Because these conditions are rarely met, “the idea that there is a universal molecular clock ticking away has long since been discredited.”38
Applying the molecular clock to dating the alleged Precambrian ancestor of the animals further complicates matters. Because there are so few fossils in the Precambrian and no clear ancestor-descendant lineages, the calibration of the molecular clock must be done on the basis of very different fossil lineages arising hundreds of millions of years later. Indeed, without evidence from the fossil record (older than 550 million years ago) with which to calibrate the molecular clock, any attempt to date the origin of the Cambrian animal phyla becomes highly questionable.39 Perhaps for this reason, Valentine, Jablonski, and Erwin have wondered whether “molecular clock dates can ever be applied reliably to such geologically remote events as Neoproterozoic branchings within the Metazoa.”40 (The Neoproterozic is the last Precambrian era.) These methodological problems may help account for the cacophony of conflicting results.
Smuggling in Darwin
A second crucial assumption behind the deep-divergence hypothesis is the idea of the common descent of all the animal forms—i.e., that all the Cambrian animals evolved from a common Precambrian ancestor. As the textbook Understanding Bioinformatics admits, “The key assumption made when constructing a phylogenetic tree from a set of sequences is that they are all derived from a single ancestral sequence, i.e., they are homologous.”41 Or as the Harvard University Press textbook The Tree of Life states, “We are obliged to assume at first that, for each character, similar states are homologous,” whereby “homologous” the text means characters are similar because they share a common ancestor.42
This assumption (of universal common descent) raises the possibility that the ancestral entities represented by divergence points in these studies are artifacts of the assumptions by which molecular data are analyzed. Indeed, the computer programs that are used to compare molecular sequences have been written to produce trees showing common ancestors and branching relationships regardless of the extent to which the genes analyzed may or may not differ. Phylogenetic studies compare two or more gene sequences and then use degrees of difference to determine divergence points and nodes on a phylogenetic tree. Inherent in that procedure is the assumption that the nodes and divergence points existed in the past.
Thus, the deep-divergence studies do not, in any rigorous sense, establish any Precambrian ancestral forms. Did a single, original metazoan or bilaterian ancestor of the Cambrian animals actually exist? The Precambrian–Cambrian fossil record taken on its face certainly doesn’t document such an entity. But neither do deep-divergence studies. Instead, these studies assume the existence of such ancestors, and then merely attempt, given that assumption, to determine how long ago such ancestors might have lived. One could argue that the conflicting divergence points do at least show that some common ancestor existed in the Precambrian, since, despite their conflicting results, all divergence studies indicate at least that. But, again, to invoke molecular studies that assume the existence of a common ancestor as evidence for such an entity only begs the question. Certainly it provides no reason for using molecular evidence to trump fossil evidence. Perhaps the Precambrian rocks do not record ancestors for the Cambrian animals because none existed. To foreclose that possibility, and to resolve the mystery of the missing Precambrian ancestral fossils, evolutionary biologists cannot use studies that assume the existence of the very entity their studies are thought to establish.
The “Shmoo”: A Catch-22 Revisited
The concept of deep divergence raises another issue related to my discussion at the end of the previous chapter about what would be required to document the missing ancestral forms of the Cambrian animals. Recall that I argued there that any plausible postulated common ancestor to all the animal phyla must have necessarily lacked most (or all) of the specific anatomical features that distinguish one phylum from another. The more arthropod-like a hypothetical animal form, the less plausible it would have been as an ancestor to the chordates, mollusks, echinoderms, annelid worms, and vice versa. In each case, the design logic and arrangement of parts necessary to provide the foundation for one mode of animal life preclude it from providing the foundation for other modes of animal life—just as a system of parts providing one mode of transportation (with a bicycle, for example) will typically preclude functioning as another (as with a submarine, for example).43
For this reason, biologists thinking about the characteristics of the earliest ancestor of all the metazoan phyla—the actual animal at the deep-divergence point—have typically postulated an extremely simple form of life—what one evolutionary biologist described to me as a “shmoo,” after the blob-like cartoon character made famous by Al Capp in the 1940s and 1950s. Some have proposed that the ur-animal might have been something like a placozoan, a modern amorphous animal with only four cell types and no bilateral symmetry.44 Other paleontologists have mainly characterized the hypothetical ur-metazoan negatively, by reference to the characteristics that it must not have had to be a plausible common ancestral form to all other metazoa. (This need to characterize the ur-metazoan negatively has led some leading paleontologists to question whether the ur-animal can be described affirmatively with any specificity.45)
In any case, the need to characterize the ur-animal as an extremely simple “shmoo-like” form, lacking the numerous characteristics and anatomical novelties present in the Cambrian animals, highlights a deep dilemma for evolutionary theorists. On the one hand, to be plausible as a common ancestor of all the animal phyla, a hypothetical ur-metazoan must have few characteristics of later metazoan forms. Indeed, the more plausible the hypothetical ancestor, the simpler it must be, meaning it will lack more of the specific distinguishing features of the individual animal phyla. But that means any evolutionary scenario for the origin of the animals that postulates such a “stripped down” animal form as its starting point will need to envision those distinguishing characteristics arising later. And the fewer the number of characteristics in the hypothetical common ancestor, the more such characteristics will need to arise later. This logical requirement implies, in turn, the need for an even deeper divergence point in Precambrian history and the need for more time to produce these specific anatomical novelties—in turn, exacerbating the problem of fossil discontinui
ty. The more plausible the hypothetical common ancestor, the deeper the necessary divergence point and the greater the morphological discontinuity in the fossil record.
On the other hand, proposing a more complex (and more anatomically differentiated) common ancestor closer in its affinities to some Cambrian animal forms, would eliminate the need for so deep a divergence point. Nevertheless, it would also diminish the plausibility of such a hypothetical ancestor as an ur-metazoan common to all the other Cambrian animals. Again, the more a hypothetical form resembles one of the specific animal forms or phyla, the less plausible it will be as an ancestor of all the others. And that is the dilemma. Could there have been an animal form simple enough to serve as a viable ancestor common to all the animal phyla? Perhaps. But positing such a form only deepens the required depth of the divergence point and intensifies the already significant problem of Precambrian–Cambrian fossil discontinuity.
Deep Trouble
Comparative genetic analyses do not establish a single deep-divergence point, and thus do not compensate for a lack of fossil evidence for key Cambrian ancestors—such as the ur-bilaterian or ur-metazoan ancestor. The results of different studies diverge too dramatically to be conclusive, or even meaningful; the methods of inferring divergence points are fraught with subjectivity; and the whole enterprise depends upon a question-begging logic. Many leading Cambrian paleontologists, and even some leading evolutionary biologists, now express skepticism about both the results and the significance of deep-divergence studies. For example, Simon Conway Morris has rejected the idea that such studies should trump fossil evidence of a more explosive, shallow and rapid Cambrian radiation. After assessing the inconsistent track record of deep-divergence studies, he concludes, “A deep history extending to an origination in excess of 1,000 Myr [million years] is very unlikely.”46 Conway Morris is one of several leading evolutionary biologists and Cambrian paleontologists who have expressed skepticism about these studies.47 In any case, there is now little reason to regard the deep-divergence hypothesis as a genuine solution to the Cambrian conundrum.
6
The Animal Tree of Life
In 2009, in honor of the bicentennial anniversary of Darwin’s birth, a piece of artwork was created to adorn the ceiling of an exhibit room at the Natural History Museum in London. A paper in the journal Archives of Natural History noted that the inspiration for the artwork, titled “TREE,” came from a diagram that Darwin had sketched in one of his notebooks—a diagram that later came to be known as the “tree of life” (see Fig. 2.11a). One BBC radio program called the TREE exhibit the “Darwinian Sistine Chapel.”1 Another article in Archives of Natural History, a journal published by the University of Edinburgh, remarked that, “TREE celebrates Darwinian evolutionism” and “secular science and reason.”2
For many biologists, the iconic image of Darwin’s tree of life represents perhaps the single best distillation of what the science of evolutionary biology has to teach, namely the “fact of evolution,”3 apart from which “nothing in biology makes sense.”4 Though the fossil record does not directly attest to many of the expected intermediate forms represented on Darwin’s tree, leading authorities assert that other lines of evidence, particularly from genetics, firmly establish Darwin’s tree as the correct picture of the history of life.
In the previous chapter, we saw that there are many good reasons to doubt the deep-divergence hypothesis and its claim to have determined, based upon genetic evidence, the time at which the Cambrian animals began to evolve from specific Precambrian ancestors. Indeed, the idea that these studies can pinpoint when an ur-metazoan, or an ur-bilaterian, arose has engendered increasing skepticism among a growing number of evolutionary biologists and paleontologists.
The tree of life as a whole, however, is another matter. Many evolutionary biologists think the case for universal common descent is something close to unassailable because, they argue, analysis of both anatomical and genetic similarities converges on the same basic pattern of descent from a universal common ancestor. As Richard Dawkins asserts, “when we look comparatively at … genetic sequences in all these different creatures—we find the same kind of hierarchical tree of resemblance. We find the same family tree—albeit much more thoroughly and convincingly laid out—as we did with … the whole pattern of anatomical resemblances throughout all the living kingdoms.”5 Likewise, Jerry Coyne argues that gene sequences independently confirm the same set of evolutionary relationships—the same basic tree—established from the analysis of anatomy.6 Oxford University chemist Peter Atkins is even more emphatic: “There is not a single instance of the molecular traces of change being inconsistent with our observations of whole organisms.”7
As a result of this confidence, evolutionary biologists often dismiss the missing Precambrian fossil precursors and intermediates as a minor anomaly—one awaiting explanation by an otherwise completely adequate theory of the history of life. Because most evolutionary biologists are confident that a single continuous tree, with a single root, best represents the history of life—and explains so many other diverse facts of biology—they continue to think the same treelike pattern also accurately describes the Cambrian explosion and the Precambrian history of animal life. Moreover, when evolutionary biologists reconstruct the phylogenetic history of a group (including animals), they typically do so in a time-independent manner. Their concern is usually to establish a relative order of branching along the tree of life, not to establish or “pinpoint” a series of absolute dates at which divergences occurred. Thus, although deep-divergence studies do not establish the existence of Precambrian animal ancestors for all the reasons argued in the previous chapter, the uncertainty surrounding the dates in these studies has not, for most evolutionary biologists, undermined their confidence in the overall treelike pattern of animal life. Instead, many evolutionary biologists believe the strength of the case for the tree of life as a whole, based on other phylogenetic studies of similar genes and anatomical traits, indirectly establishes the existence of the missing evolutionary precursors of the Cambrian animals. As Coyne explains, “It stands to reason that if the history of life forms a tree, with all species originating from a single trunk, then one can find a common origin for every pair of twigs (existing species) by tracing each twig back through its branches until they intersect at the branch they have in common. This node, as we’ve seen, is their common ancestor.”8
On the basis of similar logic, evolutionary biologists have typically assumed that what they think is true of all other forms of life is true of the Cambrian forms—that there must be a universal animal tree, the absence of fossil evidence, and the conflicting results of deep-divergence studies, notwithstanding.
To assess the other evidence from genetics that supports this conclusion, it’s useful to review how the case for the universal animal tree is similar to the deep-divergence hypothesis and also how it is different. To establish both the fact and shape of the Darwinian animal tree of life, evolutionary biologists have long used methods that assume that both molecular sequences and anatomical similarities provide an accurate historical signal about the past. Like deep-divergence studies, these methods of “phylogenetic” reconstruction assume that the species or larger groupings (taxa) are related by descent from a common ancestor. (The term phylogeny, again, refers to the evolutionary history of a group of organisms. Thus, a “phylogenetic reconstruction” is an attempt to determine that history.) Such studies assume that the degree of difference between molecular or anatomical features in pairs of organisms indicates how long ago they diverged from a common ancestor. They also use independent calibrations of the molecular clock to calculate the exact divergence times.9
Unlike deep-divergence studies, however, which attempt to establish just a single divergence time—such as that of the common ancestor of all the animal phyla—these more detailed phylogenetic studies seek to establish the contours of the Precambrian tree of animal life. This involves assessing degrees of relatedness a
mong representatives of all the Cambrian phyla to establish multiple divergence points and times (nodes on the tree of life) as well as the relationships of the major Cambrian groups.
Investigators employ these methods even in the absence of corroborating fossil evidence. In his textbook on fossils and evolution, following a full-page depiction of the discontinuous appearance of the Cambrian animals in the fossil record, Occidental College geologist Donald Prothero explains, “If the fossil record is poor in one particular group, we look to other sources of data.” He concludes that two such sources of data, anatomical and molecular data, now “converge on a common answer”—one “that is almost certainly ‘the truth’ (as much as we can use that term in science).”10
But is all of this true? Does analysis of the genetic and anatomical similarity of the Cambrian animals really establish that the history of animal life is best depicted as a continuously branching tree? Does the pattern of a branching tree accurately depict the history of Precambrian and Cambrian animal life, and in so doing establish the existence of Precambrian forms that the fossil record fails to document?
The Precambrian and Cambrian Tree of Animal Life
History happened once. And if Richard Dawkins is correct that “there is, after all, one true tree of life, the unique pattern of evolutionary branchings that actually happened,”11 then evolutionary history also happened once. Consequently, if we think of evolutionary trees describing the relationships of animal groups as hypotheses about an unobserved history (which is what they are), then having two or more conflicting hypotheses about only one history—the history that actually happened—means that we haven’t figured out what did happen. A widely used textbook on phylogenetic methods explains this: “The fact that there is only one true tree … provides the basis for testing alternative hypotheses. If two hypotheses are generated for the same group of species, then we can conclude that at least one of these hypotheses is false. Of course, it is possible that both are false and some other tree is true.”12