Anthro Assignment 2

Stephen_Jay_Gould.pdf

Home >History homework help >Anthro Assignment 2

The Evolution of Life on the Earth Author(s): Stephen Jay Gould Source: Scientific American, Vol. 271, No. 4, SPECIAL ISSUE: LIFE IN THE UNIVERSE (OCTOBER 1994), pp. 84-91 Published by: Scientific American, a division of Nature America, Inc. Stable URL: https://www.jstor.org/stable/24942873 Accessed: 21-08-2018 00:39 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide

range of content in a trusted digital archive. We use information technology and tools to increase productivity and

facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

https://about.jstor.org/terms

Scientific American, a division of Nature America, Inc. is collaborating with JSTOR to digitize, preserve and extend access to Scientific American

This content downloaded from 168.16.190.143 on Tue, 21 Aug 2018 00:39:21 UTC All use subject to https://about.jstor.org/terms

All use subject to https://about.jstor.org/terms

SCIENTIFIC AMERICAN October 1994 85

S ome creators announce their in- ventions with grand �clat. God proclaimed, ÒFiat lux,Ó and then

ßooded his new universe with bright- ness. Others bring forth great discov- eries in a modest guise, as did Charles Darwin in deÞning his new mechanism of evolutionary causality in 1859: ÒI have called this principle, by which each slight variation, if useful, is preserved, by the term Natural Selection.Ó

Natural selection is an immensely powerful yet beautifully simple theory that has held up remarkably well, un- der intense and unrelenting scrutiny and testing, for 135 years. In essence, natural selection locates the mechanism of evolutionary change in a ÒstruggleÓ among organisms for reproductive suc- cess, leading to improved Þt of popula- tions to changing environments. (Strug- gle is often a metaphorical description and need not be viewed as overt com- bat, guns blazing. Tactics for reproduc- tive success include a variety of non- martial activities such as earlier and more frequent mating or better cooper- ation with partners in raising oÝspring.) Natural selection is therefore a princi- ple of local adaptation, not of general advance or progress.

Yet powerful though the principle may be, natural selection is not the only

cause of evolutionary change (and may, in many cases, be overshadowed by oth- er forces). This point needs emphasis because the standard misapplication of evolutionary theory assumes that bio- logical explanation may be equated with devising accounts, often speculative and conjectural in practice, about the adap- tive value of any given feature in its original environment (human aggres- sion as good for hunting, music and re- ligion as good for tribal cohesion, for example). Darwin himself strongly em- phasized the multifactorial nature of evolutionary change and warned against too exclusive a reliance on natural se- lection, by placing the following state- ment in a maximally conspicuous place at the very end of his introduction: ÒI am convinced that Natural Selection has been the most important, but not the exclusive, means of modiÞcation.Ó

N atural selection is not fully suf- Þcient to explain evolutionary change for two major reasons.

First, many other causes are powerful, particularly at levels of biological orga- nization both above and below the tra- ditional Darwinian focus on organisms and their struggles for reproductive suc- cess. At the lowest level of substitution in individual base pairs of DNA, change is often eÝectively neutral and therefore random. At higher levels, involving en- tire species or faunas, punctuated equi- librium can produce evolutionary trends by selection of species based on their rates of origin and extirpation, whereas mass extinctions wipe out substantial parts of biotas for reasons unrelated to adaptive struggles of constituent species in ÒnormalÓ times between such events.

Second, and the focus of this article, no matter how adequate our general theory of evolutionary change, we also yearn to document and understand the actual pathway of lifeÕs history. Theory,

of course, is relevant to explaining the pathway (nothing about the pathway can be inconsistent with good theory, and theory can predict certain general aspects of lifeÕs geologic pattern). But the actual pathway is strongly underde- termined by our general theory of lifeÕs evolution. This point needs some bela- boring as a central yet widely misunder- stood aspect of the worldÕs complexity.. Webs and chains of historical events are so intricate, so imbued with random and chaotic elements, so unrepeatable in encompassing such a multitude of unique (and uniquely interacting) ob- jects, that standard models of simple prediction and replication do not apply.

History can be explained, with satis- fying rigor if evidence be adequate, af- ter a sequence of events unfolds, but it cannot be predicted with any precision beforehand. Pierre-Simon Laplace, echo- ing the growing and conÞdent determin- ism of the late 18th century, once said that he could specify all future states if he could know the position and motion of all particles in the cosmos at any mo- ment, but the nature of universal com- plexity shatters this chimerical dream. History includes too much chaos, or ex- tremely sensitive dependence on minute and unmeasurable diÝerences in initial conditions, leading to massively diver- gent outcomes based on tiny and un- knowable disparities in starting points.

The Evolution of Life on the Earth

The history of life is not necessarily progressive; it is certainly not predictable. The earth’s creatures have evolved

through a series of contingent and fortuitous events

by Stephen Jay Gould

STEPHEN JAY GOULD teaches biology, geology and the history of science at Harvard University, where he has been on the faculty since 1967. He received an A.B. from Antioch College and a Ph.D. in paleontology from Columbia University. Well known for his popular scientiÞc writings, in particular his monthly col- umn in Natural History magazine, he is the author of 13 books.

SLAB CONTAINING SPECIMENS of Pteri- dinium from Namibia shows a promi- nent organism from the earthÕs Þrst mul- ticellular fauna, called Ediacaran, which appeared some 600 million years ago. The Ediacaran animals died out before the Cambrian explosion of modern life. These thin, quilted, sheetlike organisms may be ancestral to some modern forms but may also represent a separate and ultimately failed experiment in multi- cellular life. The history of life tends to move in quick and quirky episodes, rath- er than by gradual improvement.

All use subject to https://about.jstor.org/terms

And history includes too much contin- gency, or shaping of present results by long chains of unpredictable anteced- ent states, rather than immediate de- termination by timeless laws of nature.

Homo sapiens did not appear on the earth, just a geologic second ago, be- cause evolutionary theory predicts such an outcome based on themes of prog- ress and increasing neural complexity. Humans arose, rather, as a fortuitous and contingent outcome of thousands of linked events, any one of which could have occurred diÝerently and sent his- tory on an alternative pathway that would not have led to consciousness. To cite just four among a multitude: (1) If our inconspicuous and fragile lineage had not been among the few survivors of the initial radiation of multicellular animal life in the Cambrian explosion 530 million years ago, then no verte- brates would have inhabited the earth at all. ( Only one member of our chor- date phylum, the genus Pikaia, has been found among these earliest fossils. This small and simple swimming creature, showing its allegiance to us by possess- ing a notochord, or dorsal stiÝening rod, is among the rarest fossils of the Burgess Shale, our best preserved Cam- brian fauna.) (2) If a small and unprom- ising group of lobe-Þnned Þshes had not evolved Þn bones with a strong cen- tral axis capable of bearing weight on land, then vertebrates might never have become terrestrial. (3) If a large extra- terrestrial body had not struck the earth 65 million years ago, then dinosaurs

would still be dominant and mammals insigniÞcant (the situation that had pre- vailed for 100 million years previously). (4) If a small lineage of primates had not evolved upright posture on the dry- ing African savannas just two to four million years ago, then our ancestry might have ended in a line of apes that, like the chimpanzee and gorilla today, would have become ecologically mar- ginal and probably doomed to extinc- tion despite their remarkable behavior- al complexity.

Therefore, to understand the events and generalities of lifeÕs pathway, we must go beyond principles of evolution- ary theory to a paleontological exami- nation of the contingent pattern of lifeÕs history on our planetÑthe single actu- alized version among millions of plau- sible alternatives that happened not to occur. Such a view of lifeÕs history is highly contrary both to conventional de- terministic models of Western science and to the deepest social traditions and psychological hopes of Western culture for a history culminating in humans as lifeÕs highest expression and intended planetary steward.

Science can, and does, strive to grasp natureÕs factuality, but all science is so- cially embedded, and all scientists re- cord prevailing Òcertainties,Ó however hard they may be aiming for pure ob- jectivity. Darwin himself, in the closing lines of The Origin of Species, expressed Victorian social preference more than natureÕs record in writing : ÒAs natural selection works solely by and for the

good of each being, all corporeal and mental endowments will tend to prog- ress towards perfection.Ó

LifeÕs pathway certainly includes many features predictable from laws of na- ture, but these aspects are too broad and general to provide the ÒrightnessÓ that we seek for validating evolutionÕs particular resultsÑroses, mushrooms, people and so forth. Organisms adapt to, and are constrained by, physical principles. It is, for example, scarcely surprising, given laws of gravity, that the largest vertebrates in the sea (whales) exceed the heaviest animals on land (ele- phants today, dinosaurs in the past), which, in turn, are far bulkier than the largest vertebrate that ever ßew (extinct pterosaurs of the Mesozoic era ).

Predictable ecological rules govern the structuring of communities by prin- ciples of energy ßow and thermodynam- ics (more biomass in prey than in pred- ators, for example). Evolutionary trends, once started, may have local predict- ability (Òarms races,Ó in which both predators and prey hone their defenses and weapons, for exampleÑa pattern that Geerat J. Vermeij of the University of California at Davis has called Òesca- lationÓ and documented in increasing strength of both crab claws and shells of their gastropod prey through time). But laws of nature do not tell us why we have crabs and snails at all, why in- sects rule the multicellular world and why vertebrates rather than persistent algal mats exist as the most complex forms of life on the earth.

Relative to the conventional view of lifeÕs history as an at least broadly pre- dictable process of gradually advancing complexity through time, three features of the paleontological record stand out in opposition and shall therefore serve as organizing themes for the rest of this article: the constancy of modal com- plexity throughout lifeÕs history; the concentration of major events in short bursts interspersed with long periods of relative stability; and the role of ex- ternal impositions, primarily mass ex- tinctions, in disrupting patterns of Ònor- malÓ times. These three features, com- bined with more general themes of chaos and contingency, require a new framework for conceptualizing and drawing lifeÕs history, and this article therefore closes with suggestions for a diÝerent iconography of evolution.

T he primary paleontological fact about lifeÕs beginnings points to predictability for the onset and

very little for the particular pathways thereafter. The earth is 4.6 billion years old, but the oldest rocks date to about 3.9 billion years because the earthÕs sur-

86 SCIENTIFIC AMERICAN October 1994

PROGRESS DOES NOT RULE (and is not even a primary thrust of ) the evolutionary process. For reasons of chemistry and physics, life arises next to the Ò left wallÓ of its simplest conceivable and preservable complexity. This style of life (bacterial ) has remained most common and most successful. A few creatures occasionally move to the right, thus extending the right tail in the distribution of complexity. Many always move to the left, but they are absorbed within space already occupied. Note that the bacterial mode has never changed in position, but just grown higher.

LEFT WALL OF MINIMAL COMPLEXITY

BACTERIA

COMPLEXITY

PRESENT

PRECAMBRIAN F

R E

Q U

E N

C Y

F O

C C

U R

R E

N C

BACTERIA

All use subject to https://about.jstor.org/terms

face became molten early in its history, a result of bombardment by large amounts of cosmic debris during the solar systemÕs coalescence, and of heat generated by radioactive decay of short- lived isotopes. These oldest rocks are too metamorphosed by subsequent heat and pressure to preserve fossils (though some scientists interpret the propor- tions of carbon isotopes in these rocks as signs of organic production). The old- est rocks suÛciently unaltered to retain cellular fossilsÑAfrican and Australian sediments dated to 3.5 billion years oldÑdo preserve prokaryotic cells (bac- teria and cyanophytes) and stromato- lites (mats of sediment trapped and bound by these cells in shallow marine waters). Thus, life on the earth evolved quickly and is as old as it could be. This fact alone seems to indicate an inevit- ability, or at least a predictability, for lifeÕs origin from the original chemical constituents of atmosphere and ocean.

No one can doubt that more complex creatures arose sequentially after this prokaryotic beginningÑÞrst eukaryotic cells, perhaps about two billion years ago, then multicellular animals about 600 million years ago, with a relay of highest complexity among animals passing from invertebrates, to marine vertebrates and, Þnally ( if we wish, al- beit parochially, to honor neural archi- tecture as a primary criterion), to rep- tiles, mammals and humans. This is the conventional sequence represented in the old charts and texts as an Òage of invertebrates,Ó followed by an Òage of Þshes,Ó Òage of reptiles,Ó Òage of mam- mals,Ó and Òage of manÓ (to add the old gender bias to all the other prejudices implied by this sequence).

I do not deny the facts of the preced- ing paragraph but wish to argue that our conventional desire to view history as progressive, and to see humans as predictably dominant, has grossly dis- torted our interpretation of lifeÕs path- way by falsely placing in the center of things a relatively minor phenomenon that arises only as a side consequence of a physically constrained starting point. The most salient feature of life has been the stability of its bacterial mode from the beginning of the fossil record until today and, with little doubt, into all future time so long as the earth endures. This is truly the Òage of bacte- riaÓÑas it was in the beginning, is now and ever shall be.

For reasons related to the chemistry of lifeÕs origin and the physics of self- organization, the Þrst living things arose at the lower limit of lifeÕs conceivable, preservable complexity. Call this lower limit the Òleft wallÓ for an architecture of complexity. Since so little space ex-

ists between the left wall and lifeÕs ini- tial bacterial mode in the fossil record, only one direction for future increment existsÑtoward greater complexity at the right. Thus, every once in a while, a more complex creature evolves and ex- tends the range of lifeÕs diversity in the only available direction. In technical terms, the distribution of complexity becomes more strongly right skewed through these occasional additions.

But the additions are rare and epi- sodic. They do not even constitute an evolutionary series but form a motley sequence of distantly related taxa, usu- ally depicted as eukaryotic cell, jelly- Þsh, trilobite, nautiloid, eurypterid (a large relative of horseshoe crabs), Þsh, an amphibian such as Eryops, a dino- saur, a mammal and a human being. This sequence cannot be construed as the major thrust or trend of lifeÕs histo- ry. Think rather of an occasional crea- ture tumbling into the empty right re- gion of complexityÕs space. Throughout this entire time, the bacterial mode has grown in height and remained constant in position. Bacteria represent the great success story of lifeÕs pathway. They oc- cupy a wider domain of environments and span a broader range of biochem- istries than any other group. They are adaptable, indestructible and astound- ingly diverse. We cannot even imagine how anthropogenic intervention might threaten their extinction, although we worry about our impact on nearly ev- ery other form of life. The number of Escherichia coli cells in the gut of each human being exceeds the number of hu-

mans that has ever lived on this planet. One might grant that complexiÞca-

tion for life as a whole represents a pseudotrend based on constraint at the left wall but still hold that evolution within particular groups diÝerentially favors complexity when the founding lineage begins far enough from the left wall to permit movement in both direc- tions. Empirical tests of this interesting hypothesis are just beginning (as con- cern for the subject mounts among pa- leontologists), and we do not yet have enough cases to advance a generality. But the Þrst two studiesÑby Daniel W. McShea of the University of Michigan on mammalian vertebrae and by George F. Boyajian of the University of Pennsyl- vania on ammonite suture linesÑshow no evolutionary tendencies to favor in- creased complexity.

Moreover, when we consider that for each mode of life involving greater com- plexity, there probably exists an equal- ly advantageous style based on greater simplicity of form (as often found in parasites, for example), then preferen- tial evolution toward complexity seems unlikely a priori. Our impression that life evolves toward greater complexity is probably only a bias inspired by pa- rochial focus on ourselves, and conse- quent overattention to complexifying creatures, while we ignore just as many lineages adapting equally well by be- coming simpler in form. The morpho- logically degenerate parasite, safe with- in its host, has just as much prospect for evolutionary success as its gorgeous- ly elaborate relative coping with the

SCIENTIFIC AMERICAN October 1994 87

NEW ICONOGRAPHY OF LIFEÕS TREE shows that maximal diversity in anatomical forms (not in number of species) is reached very early in lifeÕs multicellular histo- ry. Later times feature extinction of most of these initial experiments and enor- mous success within surviving lines. This success is measured in the proliferation of species but not in the development of new anatomies. Today we have more spe- cies than ever before, although they are restricted to fewer basic anatomies.

ANATOMICAL DIVERSITY

T IM

All use subject to https://about.jstor.org/terms

slings and arrows of outrageous for- tune in a tough external world.

E ven if complexity is only a drift away from a constraining left wall, we might view trends in this

direction as more predictable and char- acteristic of lifeÕs pathway as a whole if increments of complexity accrued in a persistent and gradually accumulating manner through time. But nothing about lifeÕs history is more peculiar with re- spect to this common (and false) expec- tation than the actual pattern of extend- ed stability and rapid episodic move- ment, as revealed by the fossil record.

Life remained almost exclusively uni- cellular for the Þrst Þve sixths of its historyÑfrom the Þrst recorded fossils at 3.5 billion years to the Þrst well-doc- umented multicellular animals less than 600 million years ago. ( Some simple multicellular algae evolved more than a billion years ago, but these organisms belong to the plant kingdom and have no genealogical connection with ani- mals.) This long period of unicellular life does include, to be sure, the vitally

important transition from simple pro- karyotic cells without organelles to eu- karyotic cells with nuclei, mitochondria and other complexities of intracellular architectureÑbut no recorded attain- ment of multicellular animal organiza- tion for a full three billion years. If com- plexity is such a good thing, and multi- cellularity represents its initial phase in our usual view, then life certainly took its time in making this crucial step. Such delays speak strongly against general progress as the major theme of lifeÕs history, even if they can be plausibly ex- plained by lack of suÛcient atmospher- ic oxygen for most of Precambrian time or by failure of unicellular life to achieve some structural threshold acting as a prerequisite to multicellularity.

More curiously, all major stages in organizing animal lifeÕs multicellular architecture then occurred in a short period beginning less than 600 million years ago and ending by about 530 mil- lion years agoÑand the steps within this sequence are also discontinuous and episodic, not gradually accumula- tive. The Þrst fauna, called Ediacaran

34. Sidneyia 35. Odaraia 36. Eiffelia 37. Mackenzia 38. Odontogriphus 39. Hallucigenia 40. Elrathia 41. Anomalocaris 42. Lingulella 43. Scenella 44. Canadaspis 45. Marrella 46. Olenoides

22. Emeraldella 23. Burgessia 24. Leanchoilia 25. Sanctacaris 26. Ottoia 27. Louisella 28. Actaeus 29. Yohoia 30. Peronochaeta 31. Selkirkia 32. Ancalagon 33. Burgessochaeta

11. Micromitra 12. Echmatocrinus 13. Chancelloria 14. Pirania 15. Choia 16. Leptomitus 17. Dinomischus 18. Wiwaxia 19. Naraoia 20. Hyolithes 21. Habelia

1. Vauxia (gracile) 2. Branchiocaris 3. Opabinia 4. Amiskwia 5. Vauxia (robust) 6. Molaria 7. Aysheaia 8. Sarotrocercus 9. Nectocaris

10. Pikaia

7 8

11 12

14 15

19 20

28 29

31 32

39 38

All use subject to https://about.jstor.org/terms

to honor the Australian locality of its initial discovery but now known from rocks on all continents, consists of high- ly ßattened fronds, sheets and circlets composed of numerous slender seg- ments quilted together. The nature of the Ediacaran fauna is now a subject of intense discussion. These creatures do not seem to be simple precursors of lat- er forms. They may constitute a sepa- rate and failed experiment in animal life, or they may represent a full range of diploblastic (two-layered ) organiza- tion, of which the modern phylum Cnid- aria (corals, jellyÞshes and their allies) remains as a small and much altered remnant.

In any case, they apparently died out well before the Cambrian biota evolved. The Cambrian then began with an as- semblage of bits and pieces, frustrat- ingly diÛcult to interpret, called the Òsmall shelly fauna.Ó The subsequent main pulse, starting about 530 million years ago, constitutes the famous Cam- brian explosion, during which all but one modern phylum of animal life made a Þrst appearance in the fossil record. ( Geologists had previously allowed up to 40 million years for this event, but an elegant study, published in 1993, clearly restricts this period of phyletic ßowering to a mere Þve million years.) The Bryozoa, a group of sessile and co- lonial marine organisms, do not arise until the beginning of the subsequent, Ordovician period, but this apparent

delay may be an artifact of failure to discover Cambrian representatives.

Although interesting and portentous events have occurred since, from the ßowering of dinosaurs to the origin of human consciousness, we do not exag- gerate greatly in stating that the subse- quent history of animal life amounts to little more than variations on anatomi- cal themes established during the Cam- brian explosion within Þve million years. Three billion years of unicellularity, fol- lowed by Þve million years of intense creativity and then capped by more than 500 million years of variation on set anatomical themes can scarcely be read as a predictable, inexorable or con- tinuous trend toward progress or in- creasing complexity.

We do not know why the Cambrian explosion could establish all major anatomical designs so quickly. An Òex- ternalÓ explanation based on ecology seems attractive: the Cambrian explo- sion represents an initial Þlling of the Òecological barrelÓ of niches for multi- cellular organisms, and any experiment found a space. The barrel has never emptied since; even the great mass ex- tinctions left a few species in each prin- cipal role, and their occupation of eco- logical space forecloses opportunity for fundamental novelties. But an Òinter- nalÓ explanation based on genetics and development also seems necessary as a complement : the earliest multicellular animals may have maintained a ßexibil- ity for genetic change and embryologi- cal transformation that became greatly reduced as organisms Òlocked inÓ to a set of stable and successful designs.

In any case, this initial period of both internal and external ßexibility yielded a range of invertebrate anatomies that may have exceeded ( in just a few mil- lion years of production) the full scope of animal form in all the earthÕs envi- ronments today (after more than 500 million years of additional time for fur- ther expansion). Scientists are divided on this question. Some claim that the anatomical range of this initial explo- sion exceeded that of modern life, as many early experiments died out and no new phyla have ever arisen. But sci-

entists most strongly opposed to this view allow that Cambrian diversity at least equaled the modern rangeÑso even the most cautious opinion holds that 500 million subsequent years of opportunity have not expanded the Cambrian range, achieved in just Þve million years. The Cambrian explosion was the most remarkable and puzzling event in the history of life.

Moreover, we do not know why most of the early experiments died, while a few survived to become our modern phyla. It is tempting to say that the vic- tors won by virtue of greater anatomi- cal complexity, better ecological Þt or some other predictable feature of con- ventional Darwinian struggle. But no recognized traits unite the victors, and the radical alternative must be enter- tained that each early experiment re- ceived little more than the equivalent of a ticket in the largest lottery ever played out on our planetÑand that each surviving lineage, including our own phylum of vertebrates, inhabits the earth today more by the luck of the draw than by any predictable struggle for existence. The history of multicellu- lar animal life may be more a story of great reduction in initial possibilities, with stabilization of lucky survivors, than a conventional tale of steady eco- logical expansion and morphological progress in complexity.

Finally, this pattern of long stasis, with change concentrated in rapid epi- sodes that establish new equilibria, may be quite general at several scales of time and magnitude, forming a kind of frac- tal pattern in self-similarity. According to the punctuated equilibrium model of speciation, trends within lineages occur by accumulated episodes of geological- ly instantaneous speciation, rather than by gradual change within continuous populations (like climbing a staircase rather than rolling a ball up an inclined plane).

E ven if evolutionary theory implied a potential internal direction for lifeÕs pathway (although previous

facts and arguments in this article cast doubt on such a claim), the occasional imposition of a rapid and substantial, perhaps even truly catastrophic, change in environment would have intervened to stymie the pattern. These environ- mental changes trigger mass extinction of a high percentage of the earthÕs spe-

GREAT DIVERSITY quickly evolved at the dawn of multicellular animal life dur- ing the Cambrian period (530 million years ago). The creatures shown here are all found in the Middle Cambrian Burgess Shale fauna of Canada. They in- clude some familiar forms (sponges, bra- chiopods) that have survived. But many creatures (such as the giant Anomaloca- ris, at the lower right, largest of all the Cambrian animals) did not live for long and are so anatomically peculiar (rela- tive to survivors) that we cannot classi- fy them among known phyla.

45 46

All use subject to https://about.jstor.org/terms

cies and may so derail any internal di- rection and so reset the pathway that the net pattern of lifeÕs history looks more capricious and concentrated in episodes than steady and directional. Mass extinctions have been recognized since the dawn of paleontology; the ma- jor divisions of the geologic time scale were established at boundaries marked by such events. But until the revival of interest that began in the late 1970s, most paleontologists treated mass ex- tinctions only as intensiÞcations of or- dinary events, leading (at most) to a speeding up of tendencies that pervad- ed normal times. In this gradualistic theory of mass extinction, these events really took a few million years to unfold (with the appearance of suddenness in- terpreted as an artifact of an imperfect fossil record ), and they only made the ordinary occur faster (more intense Dar- winian competition in tough times, for example, leading to even more eÛcient replacement of less adapted by superi- or forms).

The reinterpretation of mass extinc- tions as central to lifeÕs pathway and radically diÝerent in eÝect began with the presentation of data by Luis and Walter Alvarez in 1979, indicating that the impact of a large extraterrestrial object (they suggested an asteroid sev- en to 10 kilometers in diameter ) set oÝ the last great extinction at the Creta- ceous-Tertiary boundary 65 million years ago. Although the Alvarez hypoth-

esis initially received very skeptical treatment from scientists (a proper ap- proach to highly unconventional expla- nations), the case now seems virtually proved by discovery of the Òsmoking gun,Ó a crater of appropriate size and age located oÝ the Yucat�n peninsula in Mexico.

This reawakening of interest also in- spired paleontologists to tabulate the data of mass extinction more rigorous- ly. Work by David M. Raup, J. J. Sepkos- ki, Jr., and David Jablonski of the Uni- versity of Chicago has established that multicellular animal life experienced Þve major (end of Ordovician, late De- vonian, end of Permian, end of Triassic and end of Cretaceous) and many mi- nor mass extinctions during its 530- million-year history. We have no clear evidence that any but the last of these events was triggered by catastrophic impact, but such careful study leads to the general conclusion that mass ex- tinctions were more frequent, more ra- pid, more extensive in magnitude and more diÝerent in eÝect than paleontol- ogists had previously realized. These four properties encompass the radical implications of mass extinction for un- derstanding lifeÕs pathway as more con- tingent and chancy than predictable and directional.

Mass extinctions are not random in their impact on life. Some lineages suc- cumb and others survive as sensible outcomes based on presence or absence

of evolved features. But especially if the triggering cause of extinction be sud- den and catastrophic, the reasons for life or death may be random with re- spect to the original value of key fea- tures when Þrst evolved in Darwinian struggles of normal times. This ÒdiÝer- ent rulesÓ model of mass extinction im- parts a quirky and unpredictable char- acter to lifeÕs pathway based on the evident claim that lineages cannot an- ticipate future contingencies of such magnitude and diÝerent operation.

To cite two examples from the im- pact-triggered Cretaceous-Tertiary ex- tinction 65 million years ago: First, an important study published in 1986 not- ed that diatoms survived the extinction far better than other single-celled plank- ton (primarily coccoliths and radiolar- ia). This study found that many diatoms had evolved a strategy of dormancy by encystment, perhaps to survive through seasonal periods of unfavorable condi- tions (months of darkness in polar spe- cies as otherwise fatal to these photo- synthesizing cells; sporadic availability of silica needed to construct their skele- tons). Other planktonic cells had not evolved any mechanisms for dormancy. If the terminal Cretaceous impact pro- duced a dust cloud that blocked light for several months or longer (one pop- ular idea for a Òkilling scenarioÓ in the extinction), then diatoms may have sur- vived as a fortuitous result of dorman- cy mechanisms evolved for the entirely diÝerent function of weathering sea- sonal droughts in ordinary times. Di- atoms are not superior to radiolaria or other plankton that succumbed in far greater numbers; they were simply for- tunate to possess a favorable feature, evolved for other reasons, that fostered passage through the impact and its sequelae.

Second, we all know that dinosaurs perished in the end Cretaceous event and that mammals therefore rule the

90 SCIENTIFIC AMERICAN October 1994

CLASSICAL REPRESENTATIONS OF LIFEÕS HISTORY reveal the severe biases of viewing evolution as embodying a central principle of progress and complexiÞ- cation. In these paintings by Charles R. Knight from a 1942 issue of National Geo- graphic, the Þrst panel shows invertebrates of the Burgess Shale. But as soon as Þshes evolve ( panel 2), no subsequent scene ever shows another invertebrate, al- though they did not go away or stop evolving. When land vertebrates arise ( panel 3), we never see another Þsh, even though return of land vertebrate lineages to the sea may be depicted ( panel 4). The sequence always ends with mammals ( panel 5)Ñeven though Þshes, invertebrates and reptiles are still thrivingÑand, of course, humans ( panel 6 ).

All use subject to https://about.jstor.org/terms

vertebrate world today. Most people as- sume that mammals prevailed in these tough times for some reason of general superiority over dinosaurs. But such a conclusion seems most unlikely. Mam- mals and dinosaurs had coexisted for 100 million years, and mammals had remained rat-sized or smaller, making no evolutionary ÒmoveÓ to oust dino- saurs. No good argument for mammal- ian prevalence by general superiority has ever been advanced, and fortuity seems far more likely. As one plausible argument, mammals may have survived partly as a result of their small size (with much larger, and therefore extinc- tion-resistant, populations as a conse- quence, and less ecological specializa- tion with more places to hide, so to speak ). Small size may not have been a positive mammalian adaptation at all, but more a sign of inability ever to pen- etrate the dominant domain of dino- saurs. Yet this ÒnegativeÓ feature of nor- mal times may be the key reason for mammalian survival and a prerequisite to my writing and your reading this ar- ticle today.

S igmund Freud often remarked that great revolutions in the his- tory of science have but one com-

mon, and ironic, feature: they knock human arrogance oÝ one pedestal after another of our previous conviction about our own self-importance. In FreudÕs three examples, Copernicus moved our home from center to periphery; Darwin then relegated us to Òdescent from an animal worldÓ; and, Þnally ( in one of the least modest statements of intellec- tual history), Freud himself discovered the unconscious and exploded the myth of a fully rational mind.

In this wise and crucial sense, the Darwinian revolution remains woefully incomplete because, even though think- ing humanity accepts the fact of evolu- tion, most of us are still unwilling to

abandon the comforting view that evo- lution means (or at least embodies a central principle of ) progress deÞned to render the appearance of something like human consciousness either virtu- ally inevitable or at least predictable. The pedestal is not smashed until we abandon progress or complexiÞcation as a central principle and come to en- tertain the strong possibility that H. sapiens is but a tiny, late-arising twig on lifeÕs enormously arborescent bushÑa small bud that would almost surely not appear a second time if we could re- plant the bush from seed and let it grow again.

Primates are visual animals, and the pictures we draw betray our deepest convictions and display our current conceptual limitations. Artists have al- ways painted the history of fossil life as a sequence from invertebrates, to Þshes, to early terrestrial amphibians and reptiles, to dinosaurs, to mammals and, Þnally, to humans. There are no exceptions; all sequences painted since the inception of this genre in the 1850s follow the convention.

Yet we never stop to recognize the al- most absurd biases coded into this uni- versal mode. No scene ever shows an- other invertebrate after Þshes evolved, but invertebrates did not go away or stop evolving! After terrestrial reptiles emerge, no subsequent scene ever shows a Þsh (later oceanic tableaux de- pict only such returning reptiles as ich- thyosaurs and plesiosaurs). But Þshes did not stop evolving after one small lineage managed to invade the land. In fact, the major event in the evolution of Þshes, the origin and rise to domi- nance of the teleosts, or modern bony Þshes, occurred during the time of the dinosaurs and is therefore never shown at all in any of these sequencesÑeven though teleosts include more than half of all species of vertebrates. Why should humans appear at the end of all se-

quences? Our order of primates is an- cient among mammals, and many oth- er successful lineages arose later than we did.

We will not smash FreudÕs pedestal and complete DarwinÕs revolution until we Þnd, grasp and accept another way of drawing lifeÕs history. J.B.S. Haldane proclaimed nature Òqueerer than we can suppose,Ó but these limits may only be socially imposed conceptual locks rath- er then inherent restrictions of our neu- rology. New icons might break the locks. TreesÑor rather copiously and luxuri- antly branching bushesÑrather than ladders and sequences hold the key to this conceptual transition.

We must learn to depict the full range of variation, not just our parochial per- ception of the tiny right tail of most complex creatures. We must recognize that this tree may have contained a maximal number of branches near the beginning of multicellular life and that subsequent history is for the most part a process of elimination and lucky sur- vivorship of a few, rather than continu- ous ßowering, progress and expansion of a growing multitude. We must under- stand that little twigs are contingent nubbins, not predictable goals of the massive bush beneath. We must remem- ber the greatest of all Biblical state- ments about wisdom: ÒShe is a tree of life to them that lay hold upon her; and happy is every one that retaineth her.Ó

SCIENTIFIC AMERICAN October 1994 91