scientific american - 1993 10 - is los angeles winning the war on smog

Insulin,like other proteins, is made up of diÝerent amino acids.During the 1940s research-ers were very aware that thechemistry of proteins de-pends on the order in whichthe amino acids

OCTOBER 1993

$3.95

Video proofs, such as this one of a sphere being turned inside out, are transforming mathematics.

Is Los Angeles winning the war on smog?

Computers that mimic damaged brains.

Vast lava ßows that reshaped the earth.

October 1993 Volume 269 Number 4

32

42

50

58

Clearing the Air in Los Angeles

James M Lents and William J Kelly

Large Igneous Provinces

Millard F CoÛn and Olav Eldholm

Evolutionarily Mobile Modules in Proteins

Russell F Doolittle and Peer Bork

The caustic brown smog that often veils the San Bernardino Mountains atteststhat the air quality in Los Angeles is still the worst in the U.S Yet it obscures a remarkable achievement : during the past two decades, pollution has been cutdramaticallyĐeven as the cityÕs population and the number of automobiles clog-ging freeways soared The cleanup is one that other cities might emulate

Periodically in the earthÕs past, massive upwellings of magma have created vast evated plains, both on land and beneath the sea Unlike the comparatively steadyvolcanism at the margins of continental plates, these powerful spasms occurredextremely rapidly in geologic time Studies of these ancient lava ßows indicatethey may have profoundly altered the global climate and the course of evolution

el-Like necklaces strung from beads, many proteins consist of discrete modulesthat have distinct structures and functions Surprisingly, some of these individ-ual domains appear in animal and bacterial cells Does that imply that they areancient relics of their common ancestry? Not always, the authors contend ; some

of them may have jumped across species linesĐand done so fairly recently

With Þckle currents and changing tides, water seems a poor vector to dispersethe pollen of ßowering plants But some aquatic species have developed strate-gies that the mathematics of search theory proves to be quite eÛcient Theseadaptations to exploit the physics of ßuids enabled terrestrial plants to return

to an aquatic environment and are classic examples of convergent evolution

4

Paul Alan Cox

Electrorheological Fluids

Thomas C Halsey and James E Martin

They are liquid until an electric current is applied ; then they ooze like honey

or solidify like gelatinĐall in less time than the blink of an eye The unusualproperties of electrorheological ßuids, Þrst patented in 1947, have suggestedapplications ranging from automotive clutches to adaptive shock absorbers

Only now are technical impediments to commercialization being overcome

Copyright 1993 Scientific American, Inc.

Here is the story of her 30-year-long salvage and painstaking restoration.

Mathematicians have always measured the progress of their search for truth inthe precise language of the proof But computers are putting a new spin on QED

No mere human can verify the accuracy of the enormous calculations in so-calledcomputer proofs Will mathematicians be forced to accept that their assertionsare, at best, only provisionally true, true only until they are proved false?

D E PARTM E N T S

50 and 100 Years Ago

1893: A remarkable experimentthat only a few could repeat

Letters to the Editor

The ease of elaboration Wealthand health Mauled anecdote.Science and the Citizen

Science and Business

Book Reviews

Malthusian musings The ships

of Iberia Following the ßu

Essay :Richard Wassersug

The unstoppable humanpilgrimage to the planets

The Amateur Scientist

Fluids that turn solid

in a magnetic Þeld

The toll of child labor A winningstrategy Gene therapy as a can-cer treatment Nuclear subs foroceanographers Sharks get tu-mors Art that evolves Jurassicviruses PROFILE: Twice NobelistFrederick Sanger

Critical of CRADAs Is magneticresonance really safe? The batterybottleneck Cordless elevators Impoverished elderly A gas meterthat listens to the ßow THE ANA-LYTICAL ECONOMIST: Pondering high-tech Þxes for developing nations

T RENDS IN MATHEMATICS The Death of Proof

John Horgan, senior writer

Simulating Brain Damage

GeoÝrey E Hinton, David C Plaut and Tim Shallice

Certain injuries of the brain produce bizarre patterns of errors in reading Thesame aberrations can be reproduced in computer models by damaging informa-tion pathways Such simulations add support to current ideas about the nature

of dyslexia and the way written language is processed in the brain

rights reserved No part of this issue may be reproduced by any mechanical, photographic or electronic process, or in the form of a phonographic recording, nor may it be stored in a retrieval system, transmitted or otherwise copied for public or private use without written permission of the publisher Second-class postage paid at New York, N.Y., and at additional mailing offices Authorized as second-class mail by the Post Office Department, Ottawa, Canada, and for payment of postage in cash Canadian GST No R 127387652 Subscription rates: one year

$36 (outside U.S and possessions add $11 per year for postage) Subscription inquiries: U.S and Canada 800-333-1199; other 515-247-7631 Postmaster : Send address changes to Scientific American, Box 3187, Harlan, Iowa 51537 Reprints available: write Reprint Department, Scientific American, Inc., 415 Madison Avenue, New York, N.Y 10017-1111, or fax : (212) 355-0408.

¼

Copyright 1993 Scientific American, Inc.

Established 1845

Page Source

Þrst six photographs), Ken

Biggs/ Tony Stone Images

(top, last photograph),

Edwin Maynard/ The

(left), Patricia J Wynne

(right)

Australian Geological

Survey Organization

(top right), Millard F.

CoÛn (bottom right)

Boris Starosta (bottom)

Museum, Stockholm

Vasa Museum (bottom)

Hammarskišld,Vasa Museum

art ), Stephanie Rausser

(photograph)

Star (photograph);

Jean E Taylor, Rutgers

University (computer art )

Edward C Thayer,G.A.N.G., MathematicsDepartment, University

of Massachusetts at

Am-herst (computer art );

Jessica Boyatt (photograph)

THE ILLUSTRATIONS

Cover illustration by the Geometry Center, University of Minnesota

EDITOR: Jonathan Piel

BOARD OF EDITORS: Alan Hall , Executive Editor ; Michelle Press , Managing Editor ; John Rennie, Russell Ruthen, Associate Editors; Timothy M.

Beardsley ; W Wayt Gibbs; Marguerite Holloway ;

John Horgan , Senior Writer ; Philip Morrison ,

Book Editor ; Corey S Powell; Philip E Ross; Ricki

L Rusting; Gary Stix ; Paul Wallich; Philip M Yam

ART: Joan Starwood, Art Director ; Edward Bell,

Art Director , Graphics Systems ; Jessie Nathans, Associate Art Director ; Nisa Geller, Photography Editor ; Johnny Johnson, Assistant Art Director, Graphics Systems

COPY: Maria-Christina Keller, Copy Chief; Nancy

L Freireich; Molly K Frances; Daniel C SchlenoÝ

PRODUCTION: Richard Sasso, Vice President,

Production ; William Sherman, Production ager ; Managers: Carol Albert, Print Production ;

Man-Janet Cermak , Quality Control ; Tanya DeSilva ,

Prepress; Carol Hansen , Composition ; Madelyn

Keyes, Systems; Eric Marquard, Special Projects; Leo J Petruzzi , Manufacturing & Makeup; Ad

TraÛc: Carl Cherebin

CIRCULATION: Lorraine Leib Terlecki, Circulation

Director ; Joanne Guralnick , Circulation tion Manager ; Rosa Davis, FulÞllment Manager ;

Promo-Katherine Robold , Newsstand Manager ADVERTISING: Robert F Gregory, Advertising

Director OFFICES: NEW YORK: Meryle

Lowen-thal, New York Advertising Manager ; William Buchanan, Manager, Corporate Advertising ; Pe-

ter Fisch, Randy James, Elizabeth Ryan Michelle

Larsen, Director, New Business Development.

CHICAGO: 333 N Michigan Avenue, Chicago, IL

60601; Patrick Bachler, Advertising Manager.

DETROIT: 3000 Town Center, Suite 1435,

South-Þeld, MI 48075; Edward A Bartley, Detroit

Man-ager WEST COAST: 1554 S Sepulveda Blvd., Suite 212, Los Angeles, CA 90025; Kate Dobson,

Advertising Manager ; Tonia Wendt Lisa K

Car-den, Lianne Bloomer, San Francisco CANADA: Fenn Company, Inc DALLAS: GriÛth Group

MARKETING SERVICES: Laura Salant, Marketing

Director ; Diane Schube, Promotion Manager ;

Mary Sadlier, Research Manager ; Ethel D Little,

Advertising Coordinator

INTERNATIONAL: EUROPE: Roy Edwards,

Inter-national Advertising Manager, London; Vivienne

Davidson, Linda Kaufman, Intermedia Ltd., is; Karin OhÝ, Groupe Expansion, Frankfurt ;

Par-Barth David Schwartz, Director, Special

Proj-ects, Amsterdam SEOUL: Biscom, Inc TOKYO: Nikkei International Ltd.

ADMINISTRATION: John J Moeling, Jr., Publisher ; Marie M Beaumonte, Business Manager

SCIENTIFIC AMERICAN, INC

415 Madison Avenue New York, NY 10017 (212) 754-0550 PRESIDENT AND CHIEF EXECUTIVE OFFICER: John J Hanley

CHAIRMEN OF THE BOARD: Dr Pierre Gerckens, John J Hanley

CHAIRMAN EMERITUS: Gerard Piel

CORPORATE OFFICERS: Executive Vice President

and Chief Financial Ỏcer, R Vincent

Bar-ger ; Vice Presidents: Jonathan Piel , John J.

Moeling, Jr.

PRINTED IN U.S.A

THE COVER shows a scene from a er-generated Þlm of a sphere being turnedinside out, or everted The Ịvideo proof ,Ĩwhich was produced at the Geometry Cen-ter in Minneapolis, Minn., is based on atopological theorem by William P Thurston

comput-of the Mathematical Sciences Research stitute (see Ị The Death of Proof,Ĩ by JohnHorgan, page 92) The rules of topology al-low the skin of the sphere to be stretchedand twisted and even to pass through it-self, but the eversion must be completedwithout the formation of a kink

In-Simplify, Simplify

Essayist W Brian Arthur [ÒWhy Do

Things Become More Complex?Ó S

CIEN-TIFIC AMERICAN, May] observes that

fun-damental designs tend to accrete

in-creasingly complex hierarchies of

sup-port subsystems as their uses expand,

until they are supplanted by improved

designs Arthur then wonders whether a

general principle underlies this tendency

Yes, a well-known characteristic of

invention: it is much easier to

elabo-rate than to innovate

DAVIDSON CORRY

Seattle, Wash

Arthur writes that ÒCopernicusÕs

daz-zlingly simple astronomical system,

based on a heliocentric universe,

re-placed the hopelessly complicated

Ptol-emaic system.Ó CopernicusÕs

heliocen-tric system was actually more complex

Because Copernicus used circular orbits

to describe the motions of the planets

(as did the Ptolemaic system), he was

forced to use epicycles to account for

their apparent retrograde motions as

viewed from the earth Not until

Kep-ler discovered that the planets move in

elliptical orbits around the sun, some

50 years later, did the heliocentric

sys-tem emerge in its true simplicity

DERICK W OVENALL

Wilmington, Del

Vital Priorities

In ÒThe Economics of Life and DeathÓ

[SCIENTIFIC AMERICAN, May], Amartya

Sen postulates that the problem is not

only production but distribution of food

Unfortunately, he does not touch on the

political conditions necessary for that

distribution The only time he refers to

this issue is when he writes that, in

gen-eral, ÒdemocraticÓ countries such as

In-dia have been able to avoid famine

bet-ter than ÒdictatorshipsÓ like China

The available information, however,

does not support that statement

Con-trary to what Sen indicates, China has

been able to feed its people better than

India has Although China had worse

health and nutritional levels than India

in the 1940s and 1950s, China has been

able to improve its health and

nutrition-al indicators faster, at least until the

1980s, when China started the marketreforms that increased food productionwhile worsening its distribution Simi-larly, Cuba has been able to reduce fam-ine and malnutrition much faster than other Latin American countries, includ-ing Costa Rica and Jamaica, the two Sencited as successes

The root of the problem of trition in the world is the skewed con-centration of wealth that is imperme-able to democratic pressures Only thosecountries that have redistributed theirwealth have been able to solve the mal-nutrition problem

malnu-VICENTE NAVARRODepartments of Health Policy,Sociology and Policy StudiesJohns Hopkins University

We wish to take exception to the cle by Sen The author draws the inac-curate conclusion that life expectancy

arti-in Saudi Arabia is lower than arti-in manypoorer countries because of misman-agement of health care funds You mustremember that major progress towardmodernization in the Kingdom did notbegin until 1970 India, in contrast, has

a long history of modernity, includingBritish colonization Thus, Saudi Arabiahas lacked the necessary infrastructurerequired to achieve the levels attained

by others

ABDUL-MANNAN TURJOMANScientiÞc CounselorRoyal Embassy of Saudi ArabiaCultural Mission to the U.S.A

Washington, D.C

Sen replies:

Navarro is right to argue that unequaldistributions of wealth and income havemuch to do with undernourishment

The need for public services, on which Ifocused, relates to this inequality

His argument about India and Chinaconfuses famine with endemic under-nourishment and ill health They aredistinct problems I had, in fact, not-

ed that Òeven though postrevolutionaryChina has been much more successfulthan India in economic expansion and

in health care, it has not been able tostave oÝ famine.Ó Navarro overlooks thegigantic famine of 1958Ð1961 in China,which is estimated to have killed from

23 million to 30 million people Indiahas not had a famine since its inde-

pendence Any government that had toface opposition parties, free newspapersand regular elections could not aÝord

to have famines During those years offamine, the Chinese government wasunder little political pressure to aban-don the policies that were directly con-tributing to the problem A similar gov-ernmental immunity sustained famines

in Cambodia in the 1970s, in the SovietUnion in the early 1930s and in manynondemocratic regimesÑboth of theÒrightÓ and the ÒleftÓÑin sub-SaharanAfrica More democratic regimes in Bots-wana and Zimbabwe successfully pre-vented famine

I did not attribute the relatively lowlife expectancy in Saudi Arabia to Òmis-management of health care funds.Ó Rath-

er the main issue concerns the overallpriority that is given to health care andeducation (especially female education)compared with other expenditures Toattribute the achievements of the Indianstate of Kerala to the benign inßuence

of the British Empire would be mistaken

on two grounds First, most of Keralaremained outside British India Second,the biggest expansions of education andhealth care in Kerala have come only inrecent decades, much inßuenced by left-wing political movements

What It Was, WasnÕt

I admire presidential science adviserJohn H Gibbons as much as the nextperson [Ò ProÞle,Ó ÒScience and the Citi-zen,Ó by John Horgan; SCIENTIFIC AMER-ICAN, April], but I fervently hopeÑforthe sake of the nationÑthat his expertise

in technology is superior to his edge of popular culture Gibbons quotes

knowl-a supposed Tennessee Ernie Ford storyabout a mountaineer watching a footballgame Alas, it wasnÕt Tennessee ErnieFord, it was Andy GriÛth in the 1953comedy routine that made him famous,

What It Was, Was Football Poor Gibbons

mangled other details, too (although, in

GriÛthÕs original, the tale spinner was

drinking a big orange drink)

HOWARD R COHENLos Angeles, Calif

Because of the volume of mail , letters

to the editor cannot be acknowledged Letters selected for publication may be edited for length and clarity.

LETTERS TO THE EDITORS

50 AND 100 YEARS AGO

OCTOBER 1943

Ò For many years it was believed that

the isolation of a protein in the

chemi-cally pure state was nearly impossible

Recently, however, this situation has

changed quite completely The enzymes

pepsin, trypsin, and urease and the

hor-mone insulin have all been isolated and

appear to be proteins One of the

bright-est chapters in this search for protein in

its chemically pure state is the isolation

of the virus of the tobacco mosaic

dis-ease This virus is described as a

crystal-line protein The signiÞcance of this

dis-covery lies not only in the great advance

in the understanding of protein

struc-tures but in the fact that it connects

proteins with measles, yellow fever, the

common cold, and several other

diseas-es of both plants and animals.Ó

Ò Motors running on alternating

cur-rent have only a few Þxed speeds

de-pending upon the motor construction

and the frequency of the current

sup-plied to them What industry desires is

the ßexibility of the direct-current

mo-tor and the eÛciency of transmission

of the alternating-current system

Elec-tronics now makes this possible The

tubes employed are

thyratronsÑgas-eous or mercury-vapor-Þlled rectiÞers

which convert alternating current into

direct current The portion of the

posi-tive half cycle during which current

ßows through the tube can be varied

by merely changing the voltages on the

grid with respect to the phase of the

alternating voltages placed upon grid

and anode.Ó

Ò When an optical surface of glass is

aluminized, the evaporated molecules,

being in a high vacuum, travel,

with-out bumping into other molecules, from

the hot metal source to the mirrorÕs

cold surface and are deposited in a

non-crystalline metallic Þlm having the same

degree of polish as that of the glass

As soon as air is admitted, the metallic

aluminum begins to oxidize, and this

oxide continues to thicken for about 60

days Why couldnÕt all this be as easily

accomplished by letting a disk of plain

cast aluminum oxidize in the air? Fred

B Ferson, a Biloxi, Mississippi, amateur

telescope maker, states it thus:

ÔAlumi-num is a metal which absorbs gases

readily, and is hard to prevent from

taking up impurities when it is cast

Also in castings it cools into crystallinestructure, the crystals coarse and full

of holesÑpossibly from absorbed

gas-es driven oÝ.Õ Ó

OCTOBER 1893

Ò It now does not seem improbablethat, when by the power of thought animage is evoked, a distant reßex action,

no matter how weak, is exerted uponcertain ends of the visual nerves, and,therefore, upon the retina Helmholtzhas shown that the fundi of the eyesare themselves luminous, and he was

able to see, in total darkness, the

move-ment of his arm by the light of his owneyes This is one of the most remark-able experiments recorded in the histo-

ry of science, and probably only a fewmen could satisfactorily repeat it, for it

is very likely that the luminosity of theeyes is associated with uncommon ac-tivity of the brain and great imaginativepower It is ßuorescence of brain action,

as it were.ÑNikola Tesla , in a paper

read before the Franklin Institute.Ó

ÒJapanese children are suckled untiltheir sixth year, and in language un-mistakable may be heard asking for the lactatious fountain In view of the almost universal use of cowÕs milk inother countries, its exclusion from thediet of the Japanese raises the interest-ing subject of inquiry as to whether or

not the race beneÞts by this custom,and Dr A S Ashmead, of New York,

discusses the question in the Sei-i-Kwai

medical journal In the Þrst place it isassumed that indirectly the absence ofcowÕs milk is most beneÞcial The Jap-anese mother feels the compulsion oflooking after her own health and diet.Japanese mothers chießy live on rice,ÔÞsh, shells, seaweed, and other prod-ucts of the sea,Õ while wine and beerare rigidly excluded The reward of allthis meritorious care of motherhoodand childhood is the absolute freedomfrom rickets Again, the author holdsthat the transmission of tuberculosis isavoided by the exclusion of cowÕs milkfrom the infantÕs dietary.Ó

Ò EÝorts have been made to teach achild how to swim by supporting him inthe water and causing him to eÝect themotions of natation This is the mostpractical process Its inconvenience isthat it necessitates the presence of ateacher with each pupil, and, in a largeclass of children, the teacher cannotoccupy himself with each of them for avery long time Mr Devot has been able

to overcome all the diÛculties of thepreceding method in a very ingenious

manner His apparatus (below) permits

the pupil to learn in conditions entirelyidentical with those that present them-selves when he tries to sustain himselfalone in the water The apparatus is inuse among the pupils of the Michelet Ly-ceum, who have been the Þrst to beneÞtfrom the invention of their master, Mr

Devot.ÑLa Nature.Ó

Apparatus for teaching swimming

Hard Times

Occupational injuries

among children are increasing

The sepia photographs and

etch-ings lie in the archives of nearly

every industrialized nation:

chil-dren in factories operating or

dodg-ing dangerous machinery, workdodg-ing too

many hours for too little pay Now, for

the Þrst time since U.S labor laws

pro-tecting children were enacted in 1938,

such stark images are being recorded

again in record numbers The number

of children workingĐlegally and

illegal-lyĐhas risen dramatically throughout

the country in the past decade

Accom-panying this increase are numerous,

sometimes grisly, reports of labor law

violations as well as occupational

inju-riesĐamputations, electrocutions,

frac-tures, burns, lacerationsĐand deaths

Federal law prohibits anyone under

the age of 14 years from working, except

in some agricultural jobs, and anyone

younger than 18 years from performing

certain tasks, such as operating heavy

machinery or working on construction

sites or with toxins Nevertheless,

near-ly every industry employs children of all

ages About Þve million teenagers work

in the U.S In 1991 an additional 27,528

children were discovered in illegal jobs,

an increase of about 300 percent since

1983, according to the General

Account-ing Ỏce (GAO) The number of youths

working clandestinely has been

estimat-ed to be as high as 676,000

Few studies chronicle the hazards

ex-perienced by these young people The

U.S Department of Labor keeps

com-prehensive records only of wage and

hour infractions: children working for

less than minimum wage or for

exces-sive hours By this measure alone, the

country is not doing well In 1990, the

year of a major federal eÝort called

Op-eration Child Watch, the department

un-covered 39,000 such violations In 1991

and 1992, years when no special

opera-tion was in place, about 29,000 and

19,-000 violations were reported,

respective-ly Advocacy groups argue that the

Þg-ures are actually much higher ỊWe have

no compunction about saying that there

are over two million child labor

viola-tions each year,Ĩ asserts JeÝrey

New-man of the National Child Labor

Com-mittee in New York City

Estimating the number of injuriesamong these children is even more dif-ficult The statistics that are availablerely on workerÕs compensation data andhospital records that are, at best, in-complete According to child labor ex-perts and pediatricians, children work-ing for family businesses often do notreport injuries to workerÕs compensa-tion boards In most states, agriculturallabor is not even covered by workerÕscompensation

Other youths, particularly those ployed illegally, can be unaware of theprocedure or can be discouraged bytheir employer from making a claim A

em-1988 survey by the New York State partment of Labor, for instance, foundthat more than half of the teenagersquestioned had experienced an injury

De-or a wage De-or hour violation Only onethird of those injured reported the ac-cident Underreporting may also occurbecause physicians generally do not re-cord information related to a young pa-tientÕs occupation, notes Philip J Lan-drigan of the division of environmen-tal and occupational medicine at Mount

Sinai Medical Center in New York City.Despite gaps in the data, the Nation-

al Institute for Occupational Safety andHealth determined that more than 130children died on the job in 1990 Sever-

al other studies have looked at state

cas-es Landrigan and his colleagues

recent-ly reported in the Journal of the

Ameri-can Medical Association that 9,656 New

York State adolescents received erÕs compensation between 1980 and

work-1987 Of those, 43.5 percent experienced

a permanent disabilityĐthat is an age of 525 teenagers a yearĐand 31children died

aver-In a study to be published in the

American Journal of Industrial cine, the Massachusetts Department of

Medi-Public Health found that the rate of cupational injury among teenagers wasabout twice that of adults Of the 17-year-olds whose location of injury wasrecorded when they were treated in sam-pled emergency rooms, 26 percent werethere because of a work-related injury.And researchers at the University ofWashington in Seattle determined thatfarm work accounted for nearly 50 per-

oc-SCIENCE AND THE CITIZEN

SWEATSHOPS in New York CityÕs garment district have been found to employ young people illegally Many such factories have dangerous working conditions.

cent of the serious injury claims among

children 13 years old and younger

Ac-cording to an article in the American

Journal of Public Health, children aged

19 years and younger are involved in

25,000 farm accidents around the

coun-try annually ; 300 are killed

The reasons for the increases in child

labor are varied Landrigan and others

cite poverty and immigration as

con-tributing factors Indeed, according to

the ChildrenÕs Defense Fund, there are

14.3 million poor children in the U.S.,

the highest number since 1965 Many

of them have diÛculty Þnding

above-board jobs, notes Joseph A Kinney of

the National Safe Workplace Institute in

Chicago The GAO has found that

minor-ity and low-income children are more

likely to be employed in hazardous

oc-cupations than are their white or

high-er-income counterparts

In New York City, for example, some

1,500 sweatshops in the garment

indus-try are a large source of illegal workĐ

and child labor violations ỊMost of the

children in these factories are below 18,

some as young as eight,Ĩ notes

Thom-as Glubiak, chief of the cityÕs

20-mem-ber garment district task force These

youths are exposed to myriad dangers,

he notes ỊWe Þnd unguarded

machin-ery, no Þre exits, boilers, wiring

prob-lems, egress probprob-lems, machines too

close together.Ĩ According to the New

York State Department of Labor, the

number of city establishments illegally

employing children rose from 19 in

1987 to 122 in 1988

At the same time as more children

have entered the work force, advocates

argue that enforcement of child labor

laws has diminished Many experts

con-sider the labor laws to be strong,

in-deed, most states have legislation that

is more stringent in certain aspects than

federal law But they perceive follow-up

to be weak ỊActive enforcement is

ab-solutely essential,Ĩ Newman notes ỊThis

is one of the areas where you can

actu-ally stop the crime from happening.Ĩ

Budget cuts and inadequate staÛng

have limited the ability of the federal

De-partment of Labor to conduct sweeps,

Newman says The GAO has reported

that the wage and hour inspectors

fol-low up on repeat oÝenders

infrequent-ly And a National Safe Workplace

In-stitute report stated that only 11

per-cent of the inspectorsÕ time is spent on

child labor; Ịan establishment that

em-ploys adolescents can anticipate a

fed-eral inspection once every 50 years,Ĩ it

estimates The Department of Labor

counters that its 850 inspectors check

for child labor violations constantly

Regardless of the departmentÕs claim,

however, two bills before Congress cite

enforcement as the primary deÞciency

in U.S law and seek to increase sucheÝorts and associated penalties

Internationally, conditions look moredire According to the International La-bor Organization ( ILO) in Geneva, some

200 million children under the age of

15 years work The numbers Ịseem to

be increasing in the Third World, whereyou have worsening economic condi-tions and worsening political situations,Ĩnotes Susan E Gunn of the ILO In Bra-zil, 18 percent of children aged 10 to 14years workĐa total of seven million

ỊBut frankly, it is very hard to generalizefrom country to country,Ĩ Gunn says

For instance, in Thailand the number

of working children rose 34 percent to

an estimated 1.7 million between 1983and 1987 while the economy boomed :exports rose 84 percent

If it is diÛcult to Þnd Þgures on theextent of occupational injuries in theU.S., it is virtually impossible worldwide

Isolated reports hint at the extent of the problem For instance, this spring awidely publicized factory Þre near Bang-kok left 500 people injured and 188,mostly young women, dead The work-ers were making dolls for Kader Indus-trial, a supplier for the manufacturer

of Cabbage Patch Kids Whereas in theU.S more young boys are injured, inother parts of the world girls are morelikely to be abused Employers Ịfavorhiring girlsĐthey feel girls are moreexploitable,Ĩ says Robert Senser of theAsian-American Free Labor Institute inWashington, D.C ỊMost of the workers

in the garment sectors are girls Theyare considered more docile and moredexterous.Ĩ

The recent attention turned on youthemployment is not limited to the U.S.Đinternational concern is rising as well

The ILO reports that it is stepping up itsprograms Although there is no mecha-nism for enforcement, Gunn notes thatmoral pressure can be eÝective In ad-dition, American consumer groups andtwo congressional bills are seeking tohalt imports of products that have beenmade by children under 15 years of age

The Indian government has already sponded to the threat of sanctions withpromises of more crackdowns on childlabor violators and a system of labeling

re-to ensure products were made withoutexploiting children According to theILO, about Þve million Indian childrenwork as bonded laborers

But, as with all aspects of this lem, the impact of trade barriers on thelives of these children has not beenstudied ỊI do not have an opinion yet,ĨGunn says ỊIt is obviously having aneÝect, but positive or negative? We justdonÕt know.Ĩ ĐMarguerite Holloway

prob-Sentries and Saboteurs

Mutating patientsÕ genomes

to suit their medicine

Doctors treating cancer with

chemotherapy play the odds.Too little drug and the diseasedtissue will survive; too much and thedamage to healthy cells will be morethan the patient can bear But some re-searchers believe gene therapy can in-crease the spread In upcoming clinicaltrials they intend to insert into cancerpatientsÕ DNA mutations that eithermake healthy cells more tolerant of ex-isting medication or make tumors morevulnerable to it ỊThis is a totally nov-

el approach, but a natural extension ofgene therapy,Ĩ says Albert B Deisseroth

of the University of Texas M.D son Cancer Center in Houston

Ander-In June the Recombinant DNA

Adviso-ry Committee of the National Institutes

of Health approved two protocols forhuman tests of a strategy that Deis-seroth calls Ịchemoprotection.Ĩ He andothers plan to genetically modify pa-tientsÕ bone marrow cells so they canwithstand more courses and higher dos-

es of chemotherapy Human trials arealso under way to test another genetherapy technique that is essentially amirror image of chemoprotection Rath-

er than inserting ỊsentriesĨ to guardhealthy cells against dangerous drugs, astrategy called molecular surgery sends

in genetic ỊsaboteursĨ to make tumorcells susceptible to a normally harm-less medicine

The Texas group headed by roth plans to enlist 20 to 30 womenwhose late-stage ovarian cancer has al-ready bested standard surgery and drugtreatment The most eÝective medicineavailable for such cases is often Taxol,

Deisse-a toxin derived from the bDeisse-ark of the PDeisse-a-ciÞc yew tree Taxol kills cells caught inthe act of dividing ; since tumor cellsmultiply more rapidly than do normalcells, they are more vulnerable Unfor-tunately, Deisseroth adds, Ịthe majorside eÝect of Taxol is that it also killsbone marrow cells,Ĩ which produce thewhite blood cells of the immune sys-tem Just two or three courses of Taxolare enough to deplete most patientsÕmarrow so severely that they succumb

Pa-to infection or internal bleeding.Two or three courses are often notenough to vanquish the slow-growingtumors of ovarian cancer So Deisserothwants to remove 10 percent of each pa-tientÕs bone marrow before chemothera-

py begins Using an antibody that bindsonly to younger marrow cells, the re-searchers will try to separate out the

stem, or progenitor, cells that have yet

to diÝerentiate into a particular kind of

white blood cell They then plan to

in-fect these stem cells with a mouse

vi-rusÑcrippled so that it cannot

repro-duce and spreadÑinto which has been

inserted the human MDR gene for

mul-tiple-drug resistance

All marrow cells naturally have the

MDR gene When turned on, it

produc-es a protein that acts as a sentry, produc-corting toxic chemicals such as chemo-therapeutic agents out of the cell beforethey can do any damage But for someunknown reason the gene is turned oÝwhen a stem cell matures into a whiteblood cell, so only stem cells are pro-tected and then only weakly ÒThe guard

es-is easily overrun,Ó Dees-isseroth says ÒThehigher the concentration of the drug,

the more sentries you need to protecteach cell.Ó

By infecting stem cells with bearing viruses, the researchers hope

MDR-to secure them against Taxol After eachpatient completes an ordinary round ofchemotherapy, she will be injected withher own mutated stem cells Many ofthose cells should carry multiple copies

of the MDR gene and an extra copy of

Creative Evolution

Like all things cultural, art evolves—a

platitude to most artists, but to a

handful, a description of technique By

ap-propriating scientific principles from

evo-lutionary biology and advanced

technol-ogy from computer graphics, this

avant-garde has fashioned a new medium for

creative expression: virtual evolution

British artist William Latham designed

software to mutate the “genome” of

three-dimensional forms From each

genera-tion he selects and breeds the most

aes-thetically “fit.”

Karl Sims, artist-in-residence at

Think-ing Machines, discovers his artworks

us-ing a self-mutatus-ing program that draws

from a genome of equations and

transfor-mations “You quickly evolve equations

that you couldn’t design or even

under-stand,” he says Nevertheless, the process

is simple enough that museum-goers can

express their taste via interactive exhibits

More dynamically, neural networks add

collective behavior to an ecosystem of

in-telligent paint brushes bred by the

genet-ic algorithms of Mgenet-ichael Tolson, chief

sci-entist of Xaos Tools Make way for virtual

surreality —W Wayt Gibbs

SURVIVAL OF THE MOST INTERESTING is the rule when artists play God and gardener Beginning with simple objects, William Latham breeds myriad generations of computer- generated mutations to arrive at his pseudo-organic Ò White FormÓ (left) His creations

SUPERCOMPUTER and image evolution software installed by Karl Sims at the Pompidou Center in Paris let visitors view his art (left) or interact with the exhibit to create their own collaborative works (above).

the regulatory machinery needed to

pro-duce it After a few weeks of recovery

time, Deisseroth will begin giving

Tax-ol again, in gradually increasing doses

With each course, more of the

unmodi-Þed marrow cellsÑgreater than 90

per-cent of the marrow at ÞrstÑwill die But

if the progeny of the transgenic stem

cells keep making the drug-resistance

protein, they should prosper If all goes

ÒOUR FOUNDER,Ó part of a series by Michael

Tolson awarded the Goldene Nica Award at

the 1993 Prix Ars Electronica in Austria.

can display characteristic behavior and

re-spond to their surroundings, like the

chame-leonic ÒMutation in Red RoomÓ (above).

Never Give a Sucker an Even Break

hall the meek inherit the earth, or is might right? Students of behavior haveexpended much effort analyzing whether evolution should favor individ-uals who cooperate or exploiters who go for short-term gains A computertournament over a decade ago indicated that a cooperative behavior strategyknown as tit-for-tat can beat out exploiters, although it is not stable in morelifelike simulations Recent computer modeling has uncovered a more resilientformula, dubbed Pavlov, that prevails over tit-for-tat in the evolutionary raceand yet is still cooperative—up to a point Its discoverers believe Pavlov could

be a model for many examples of altruism in the natural world

The standard paradigm for analyzing the evolution of cooperation is a nundrum called the Prisoner’s Dilemma This problem considers two sus-pects, imprisoned separately, each contemplating whether to confess andplead for a light sentence (to cooperate) or to blame the other for the crime(to defect) The payoff for each player depends on what the other does Ifboth cooperate, they each do better than if they both defect or exploit the

co-other Yet each could do better by defecting than by cooperating [see table].

Although defecting always pays off in a single round, cooperative gies can be advantageous if the Prisoner’s Dilemma is repeated—as interac-tions between members of a community often are When Robert Axelrod, aprofessor of political science at the University of Michigan, organized a con-test of computer programs to play the Prisoner’s Dilemma in the early 1980s,tit-for-tat was the surprise winner It was a surprise because it was so sim-ple: tit-for-tat starts by cooperating and in subsequent bouts simply repeatswhat its opponent did in the previous round

strate-Martin Nowak of the University of Oxford and Karl Sigmund of Vienna versity showed two years ago that tit-for-tat does not endure in simulations

Uni-in which, as Uni-in the real world, occasional mistakes are made Eventually,they found, it gets bogged down in bouts of backbiting Recently Nowakand Sigmund have conducted simulations that also model the effect of mu-

tations, or small changes in strategies The results, published in Nature,

were every bit as startling as those from Axelrod’s tournaments

Under a wide range of conditions Nowak and Sigmund found that if theywaited long enough, the strategy they call Pavlov usually dominated theirmodeled populations A player following Pavlov cooperates after a contest

in which both parties cooperate—and also after a contest in which both ties defect If only one player cooperated on the previous round, however,Pavlov defects The strategy is so named because, put another way, it re-peats, reflexlike, its previous play if it gets one of the two higher payoffsand switches if it comes out behind

par-A population of Pavlovian individuals does cooperate and reap the dant benefits But a Pavlovian population has no tendency to start cooperat-ing indiscriminately The weakness of tit-for-tat, Nowak and Sigmund say, isthat mutation allows populations to become more and more cooperative,which eventually leads to an invasion by selfish “always-exploiters.”

atten-Nowak and Sigmund note that animals often retaliate against defection

These findings have beentaken as support for theidea that animals follow atit-for-tat strategy But suchresults also corroborate theidea that animals use a Pav-lov-like strategy Pavlov,while cooperative with oth-ers who cooperate, has “noqualms about exploiting asucker,” according to its dis-coverers—it keeps on de-fecting against an oppo-nent foolish enough to go

on cooperating And thatruthlessness is what allowsPavlov, and cooperation, tosurvive —Tim Beardsley

S

THE PRISONERÕS DILEMMA is illustrated by the benefit (numbers of points) each player gains depending on whether he or she cooper- ates or defects In sophisticated simulations, the strategy called Pavlov (shown in green and yel-

low) usually eventually predominates.

TO PLAYER 1

PAYOFF TO PLAYER 2

3 3 0 5

5 0 1 1

COOPERATE AFTER DEFECT AFTER

well, the patientÕs entire marrow and

blood supply will grow to tolerate

Tax-ol while the tumors wither away

Deisseroth admits that much of the

process he hopes to modify remains

unknown The virus inserts MDR and

its replication sequence randomly into

the stem cell genome; the researchers

rely on luck to avoid causing unwanted

mutations No one knows why MDR is

naturally turned oÝ in mature white

blood cells There may be a good

rea-son, and short-circuiting that ÒoÝ Ó

switch may produce side eÝects

Since ovarian cancer does not spread

to the bone marrow, there appears to

be little danger of making tumors

them-selves drug resistant But as a patientÕs

cumulative dose of Taxol builds, it will

take a toll on other tissues Raising the

typical Taxol dosage just 40 percent

has been shown to cause neurological

problems, says Charles HesdorÝer of

Columbia University, whereas other

che-motherapeutic drugs can be increased

Þve- to 10-fold without harm HesdorÝer

sponsored the second, recently approved

chemoprotection protocol, which will

explore other drugs in addition to Taxol

and will also include patients with

wide-spread breast cancer or brain tumors

Deisseroth is conÞdent nonetheless

ÒWeÕve tried this out in mice, and it

works,Ó he claims ÒAs far as we can tell,

the cells are not changed by the MDR

gene except in their ability to repel

tox-ins.Ó But Arthur W Nienhuis, whose

group at the NIH has collected much of

the animal data cited by Deisseroth

and has submitted a third

chemopro-tection protocol, warns that mice may

not be the best model ÒIn primatesÑ

and it seems to be true in humans as

wellÑonly 1 to 5 percent of the

modi-Þed stem cells actually produce MDR,

versus 20 to 50 percent in mice,Ó he

says That may be too little too late ÒI

think we need to make sure that weÕre

getting the gene in, that itÕs safe and

that it produces protein at a reasonable

level,Ó HesdorÝer adds ÒThen we can

go on to determine whether this

sys-tem actually works therapeutically.Ó

Cautious optimism also

character-izes experiments to test the molecular

surgery technique spearheaded by

Ken-neth W Culver of Iowa Methodist

Medi-cal Center, Edward H OldÞeld of the

NIH and Genetic Therapy in

Gaithers-burg, Md In experiments last year

Cul-ver and OldÞeld managed to rescue 11

of 14 rats with terminal brain cancer

The Þrst human trial began last

De-cember at the NIH Culver recently won

approval for two more trials to begin

in November at Iowa Methodist and at

Childrens Hospital in Los Angeles

Molecular surgery sidesteps some of

the special diÛculties brain cancer sents and capitalizes on the unique op-portunities it oÝers Brain tumors oftenlurk too deep for scalpels to reach, andthe blood-brain barrier can Þlter outdrugs before they reach their target.

pre-But once a brain has matured, its cellslargely stop dividing, so anything thatmultiplies is probably malignant

Culver came up with the idea ofsneaking a gene from the herpes sim-plex virus into tumor cells and thenkilling them with the antiherpes drugganciclovir The gene, which codes forthe enzyme thymidine kinase and isthus known as HS-tk, can be carried into

a cell by a retrovirus only when the cell

is dividing, so normal brain cells should

be safe But to get enough of the gineered virus into the brain to do anygood, Culver had to genetically trans-form mouse cells into viral factories

en-Culver and OldÞeld found that whenthey injected these virus-producing cellsinto brain tumors in rats, as much as

60 percent of the tumor cells rated the HS-tk gene, whereas virtually

incorpo-no incorpo-normal cells did When they thengave the rats ganciclovir, they got an-other pleasant surprise The drug killednot only the infected tumor cells butalso many malignant cells that did notcarry HS-tk In some cases, even whenonly 10 percent of the cancerous cellstook up the gene, the whole tumor dis-appeared without causing inßammation

or hemorrhage

This bizarre Òbystander eÝectÓ doesnot seem to extend to normal cells sur-rounding tumors ÒWe still donÕt knowhow this works,Ó Culver admits ÒOnehypothesis is that it involves intercel-lular communication.Ó Destruction ofthe circulatory system within the tumormay also play an important role, eÝec-tively starving to death those tumorcells that do survive the drug

Human trials have yet to prove thatmolecular surgery is as precise andthorough in people as it seems to be inrats And, Culver points out, this tech-nique is not without its problems andrisks The virus-producing cells must

be injected close to the cancer for thetreatment to work, but magnetic reso-nance imaging does not always revealevery tumor If the mouse cells (whichalso contain HS-tk and so also die whenexposed to ganciclovir) manage to es-cape, they could infect other parts ofthe body, such as the bone marrow

For patients who have few optionsand little hope, however, molecular sur-gery and chemoprotection may pro-longÑperhaps preserveÑlife If so, thistwist on gene therapy could give doctors

a new weapon in the frustrating battleagainst cancer ÑW Wayt Gibbs

Sharks Do Get Cancer

Cartilage cure relies

on wishful thinking

ApseudoscientiÞc myth holds that

sharks donÕt get cancer Indeed,that proposition is the title of

a book by I William Lane and Linda Comac promoting shark cartilage as aÒbreakthrough in the prevention andtreatment of cancer and other degen-erative diseases.Ó The CBS televisionprogram Ò60 MinutesÓ enthusiasticallypicked up the idea earlier this year.Unfortunately, John C Harshbarger,director of the registry of tumors inlower animals at the Smithsonian Insti-tution in Washington, D.C., says he hasrecords of at least 20 cases of cancer insharks The registry, which is support-

ed by the National Cancer Institute, cludes shark cancers that originated incartilage, as well as cancers of the kid-neys, liver and blood cells ( Harshbarg-

in-er says data do not exist to detin-erminewhether sharks get cancer more or lessoften than do other creatures.)

In their book Lane and Comac cite

an article about cartilage by Arnold I.Caplan, a biologist at Case Western Re-serve University, published in this mag-azine in October 1984 Caplan had not-

ed that cartilage seemed to contain stances that inhibit the growth of bloodvessels and speculated that such sub-stances might someday be used to pre-vent tumors from establishing the bloodsupply they need to grow

sub-Caplan was right : cartilage does deed contain inhibitors of blood ves-sel growth, or antiangiogenesis factors.Some have been isolated and are now

in-in clin-inical trials for treatment of a ety of cancers as well as stomach ul-cers But Caplan says he is ÒappalledÓ

vari-by Lane and ComacÕs eÝort to promoteshark cartilage as a treatment ÒThis is

an extreme interpretation of the data,ÓCaplan complains Moreover, he adds,ÒThereÕs nothing special about sharks.Why not eat pigsÕ knuckles?Ó

Caplan is not the only investigatortroubled by the shark cartilage craze.Judah Folkman, an angiogenesis re-searcher at Harvard Medical School,states Þrmly that there is no evidencefrom any controlled study that ingest-ing cartilage can treat cancer What ismore, he notes, on the basis of what

is known, Òa patient would have to eathundreds of pounds of cartilageÓ tohave any chance of experiencing aneÝect Folkman says he has been tryingfor years to dissuade Lane from usinghis name to promote a shark cartilageproduct

Carl A Luer, an investigator at the

Mote Marine Laboratory in Sarasota, Fla.,

who appeared on Ò60 Minutes,Ó is also

dismayed that his work on

antiangio-genesis factors is being used to promote

shark cartilage ÒI feel our factor is a

pro-tein and canÕt be absorbed,Ó he states

Some tropical shark populations, he

says, are already depleted by overÞshing

The National Cancer Institute has amined some cases of supposed im-provements in patients treated withshark cartilage, but the data Òdid notshow solid evidence of clinical activity,Óaccording to Mary S McCabe, a clinicaltrial organizer at the institute Still, en-

ex-thusiasts are persevering Charles mone, a physician in private practice inNew Jersey, maintains that a few of hispatients have responded favorably Buteven he warns against self-medication.Much of the shark cartilage now onsale in health food stores, he says, isÒbogus stuÝ.Ó ÑTim Beardsley

Si-Run Silent, Si-Run (Not So) Cheap

or decades, polar oceanographers have wanted to

come in from the cold To do their work, they have had

to lug cumbersome sonar and seismic equipment over

the surface of polar ice floes Meanwhile, a few fathoms

below, the crews of Soviet and U.S submarines have lived

in relative comfort as they played cat-and-mouse games

while practicing for the end of the world

Some of those frostbitten scientists may finally get a

chance to do their work in the environment enjoyed by

their undersea neighbors This past summer an SSN-650

attack submarine, the Pargo, set off from Groton, Conn.,

with five civilian scientists nestled among the more than

100 regular crew members The mission, Submarine Arctic

Science Cruise-93, which is to be funded with $3 million

from this year’s defense budget, constitutes the first

un-classified scientific journey on board a nuclear submarine

On August 11, this Sturgeon-class sub set out on a

jour-ney of more than a month to the Arctic Ocean that would

take the researchers to the North Pole Almost half the

mission, 3,300 nautical miles, was to be below the ice

The “hunter-killer” submarine is most likely still equipped

with a cargo of Harpoon antiship missiles and Mk 48

torpe-does in the unlikely event that hostilities break out But this

expedition could mark a step

to-ward perhaps the ultimate

de-fense-conversion project: an

at-tempt to deploy a nuclear

sub-marine for the sole purpose of

scientific research

Since the first nuclear-powered

submarine was launched in the

1950s, scientists have

recog-nized that these vessels might

be harnessed to compile a

com-prehensive profile of physical

changes in the ice cap, an

espe-cially important task in tracking

the possible warming trend in

the earth’s climate “It would be

difficult to get this information

any other way,” says Marcus G

Langseth, a scientist at

Colum-bia University’s Lamont-Doherty

Earth Observatory and chairman

of the science steering

commit-tee for this program

The current mission was to

use a type of sonar that can

look upward to judge the

vol-ume of the ice cap Meanwhile

a synthetic-aperture radar on a

satellite orbiting over the Arctic

would provide topside images

of the same area The

subma-rine also plans to surface in holes or cracks in the ice Thesestops would allow researchers to deploy buoys with strings

of sensors that hang under the ice to measure water peratures and salinity

tem-Langseth, however, believes that researchers will be able

to take full advantage of a submarine only once it is manently retrofitted with the necessary instrumentation Ahigh-level study panel of oceanographers recommended

per-to the navy in 1991 that magnetic and seismic instrumentsfor charting the Arctic basin could be towed behind thevessel For the moment, this technology would probably beunacceptable to the navy: submarine commanders assidu-ously avoid any devices that emit substantial amounts ofradio energy for fear of revealing submarine location andoperating characteristics

So the oceanographers could run silent But can they runcheap? A decision about a dedicated research vessel, Lang-seth says, should come before the planned decommission-ing of the entire Sturgeon fleet Langseth and other scien-tists have their work cut out: the $8 million to $15 millionneeded to keep a submarine in operation annually couldmake a submarine into the oceanographic equivalent of theSuperconducting Super Collider “The big problem is that

the navy doesn’t want to pay for

it, and the scientific

communi-ty doesn’t want to pay,” remarksGeorge B Newton, a formersubmarine commander whohelped to organize the mission.Are the Russians coming? Ourformer adversaries may ignite

a fare war for scientist gers They have twice offered

passen-to convert one of their nuclearsubmarines into a research ve-hicle for the West

Hitching a ride on a Russiansubmarine, even as part of aU.S foreign aid package, mayprove a hard sell at a timewhen U.S defense companiesare laying off workers Indeed,there may be a way to get atleast some of the informationwith little cost If Congress ul-timately decides that a multi-million-dollar passage on board

a submarine is an

unneed-ed joyride, the U.S research community might pressure thenavy to speed up the declas-sification of Arctic data thatthey have been gathering for

years —Gary Stix

STURGEON-CLASS SUB, like this one, was to have surfaced in the Arctic with Þve scientists on board.

Jurassic Virus?

CanÕt clone a Tyrannosaur?

Then try chicken pox

Molecular archaeologists

immedi-ately cried foul when they saw

the visions of dinosaurs cloned

from the last meals of amber-entombed

mosquitoes in the Þlm version of

Mi-chael CrichtonÕs Jurassic Park Ancient

DNA is too damaged, they argued; it

would take centuries to piece it

togeth-er Moreover, the reconstructed genome

would be riddled with errors In short,

they concluded, extinct organisms must

remain so, forever

Not necessarily, say George O Poinar,

Jr., and Raœl J Cano, the

entomologist-microbiologist team who hold the record

for sequencing the most ancient DNAÑ

that of a Jurassic wood beetle embedded

in amber more than 120 million years

ago They believe it might be possible to

resurrect organisms, albeit ones much

simpler than Tyrannosaurus rexÑa

bac-terium, say, or a virus ÒWeÕre very close

to reconstructing a gene of a bacillus

spe-cies that lived in the gut of a stingless bee

40 million years ago,Ó says Cano, who is

at California Polytechnic State

Univer-sity in San Luis Obispo ÒYouÕd put the

genes into a living species bit by bit,

un-til nothing was left of the modern DNA.Ó

The bacterial gene, known as 16S

rRNA, helps make ribosomes So far, the

researchers say, they have strung

to-gether 1,300 of a possible 1,500 or so

nucleotide bases They concede that

some genes are harder to reconstruct

but contend that the entire genome of abacterium could be built in a few years

Yet bringing an ancient bacteriumback to life might not be so simple Mi-chael A Goldman of San Francisco StateUniversity points out that bacterialDNA forms loops having quasichromo-somal statusÑthat is, it encodes infor-mation that lies outside the sequence

of nucleic acid bases In addition, teria bequeath proteins as well as nu-cleic acids to their daughter cells, andthese proteins may inßuence the ex-pression of certain genes ÒAlready thatmakes the task diÝerent from just mak-ing DNA and throwing it into a hostcellÕs genome,Ó he argues

bac-But Poinar and Cano retort that theywill simply shoot at a smaller target

ÒWe hope to be looking into diÝerentkinds of viruses pretty soon,Ó says Poi-nar, who is at the University of Califor-nia at Berkeley The smallest ones havegenomes no larger than a fair-sized geneand can be made to replicate withouthelper proteins One merely inserts thenaked viral DNA into a host cell, whichthen produces complete viral particles

Experts in viral evolution generally agreethat this procedure can be done ÒJuras-sic virus? Sure,Ó says Stephen S Morse

of the Rockefeller University

The problem, of course, will be ing an ancient virus No virus fossils areknown, although the new techniques ofenzymatic ampliÞcation may yet wringviral genes from ancient remains ÒThe-oretically, resurrecting a virus would bepossible, but it may be hard to get goodenough DNA,Ó says Peter M Palese ofMount Sinai Medical Center in New York

isolat-City Palese and others have been ing a more elusive quarry: the virus thatcaused the massive 1918 inßuenza pan-demic Such work is particularly chal-lenging because the genes are encoded

seek-in RNA, which is less stable than DNA.Virologists want to know more aboutthe 1918 strain because another suchpandemic can, in principle, recur Untilrecently, workers had to proceed indi-rectly, inferring the viral structure fromantibodies produced by survivors ofthe infection or extrapolating from thegenomes of more recent viral samples.Walter M Fitch of the University of Cal-ifornia at Irvine has done so by chart-ing the evolutionary trajectories of in-ßuenza in humans and pigs He foundthat the strains indeed diverged from acommon ancestor around 1918.Although Fitch believes he has re-constructed the genome of the commonancestor, he says it requires correction

by direct sequencing of the 1918 virus.While Palese says his attempts to isolatethe virus from human remains havebeen unsuccessful, Robert G Webster

of St Jude ChildrenÕs Research tal in Memphis says he is sequencingthe 1918 virus He refuses to discussthe results, however

Hospi-Another logical RNA target is the HIVretrovirus, which causes AIDS ÒYou can

go into museum collections and selecthuman tissues preserved in the 19thcentury, when they used spirit ratherthan formaldehyde,Ó Morse says ÒWeknow that some diseases killed quickly

in those days Galloping consumptionwas not unlike the presentation of tu-berculosis in AIDS patients today May-

be these were individuals with AIDSwho died of secondary infection.Ó

So far, however, the only viral RNA

to come from old human remains areendogenous retroviral genes, so calledbecause they became integrated in hu-

man chromosomes long before Homo

sapiens evolved In a paper that is to

appear in Biological Anthropology and

the Study of Ancient Egypt, Jaap

Goud-smit and his colleagues at the AcademicMedical Center in Amsterdam reportthat an Egyptian mummy more than5,000 years old has yielded traces ofhuman endogenous retrovirus type C.More recent studies on mummiÞedmonkeys produced similar results.Microbiologists do not seem overlyworried about disturbing the sleep ofundead germs Most microbes are innoc-uous, and most laboratories are designed

to contain pathogens But the risks arenot zero ÒWhat was a minor microbe inthe time of the mummies might be lethaltoday,Ó Goldman notes ÒI would keep inmind Michael CrichtonÕs earlier novel,

The Andromeda Strain.ÓÑPhilip E Ross

BUGS INSIDE BUGS: Raœl J Cano and an amber-clad, bacteria-laden insect.

One might expect a two-time

No-bel Prize winner to spend his

days raising funds for a

world-class laboratory, giving lectures to

ador-ing colleagues and collectador-ing royalties

on his best-selling novel Yet Frederick

Sanger seeks neither fame nor fortune

Instead the man who built the

founda-tions of modern biochemistry lives

qui-etly in SwaÝham Bulbeck, England,

tend-ing a garden of daÝodils, plum trees and

herbs ÒI think Sanger hasnÕt

been recognized as much as

some, partly because he is

an undemonstrative person,Ó

says Alan R Coulson, who

collaborated with Sanger for

16 years at the MRC

Labora-tory of Molecular Biology in

Cambridge

Forty years ago, Sanger

was the Þrst to reveal the

complete structure of a

pro-tein Then during the 1970s

he developed one of the Þrst

techniques for reading the

genetic code ÒHe deserved

two Nobel Prizes,Ó says

bio-chemist G Nigel Godson of

New York University

Medi-cal Center ÒHe

single-hand-edly engineered two

revo-lutions in biology.Ó In

addi-tion, Coulson boasts, Sanger

deserves much of the credit

for laying the groundwork

for the Human Genome

Proj-ect, the multinational eÝort

to determine the entire

se-quence of nucleotides in

hu-man DNA

Sanger, now 75, chose to

meet me at one of his

favor-ite pubs, the Red Lion, near

Cambridge He has mastered

the art of understatement

in both appearance and

ac-tionÑa no-frills kind of guy Wearing

a red sweater over a plaid shirt, he

orders a chicken sandwich and a

half-pint of lager He speaks in a shy voice

that is barely audible above the noise of

the restaurant Sanger is uncomfortable

talking to journalists, but in the relaxed

atmosphere of the pub he reveals some

of what made him one of the great

sci-entists of this century

Sanger was born into a wealthy home

His grandfather made a fortune in thecotton trade and passed it on to hismother, Cicely As a boy, Frederick wasfascinated by nature, collecting every-thing from rocks to insects His earlyambition was to become a doctor, likehis father, Frederick senior Yet althoughthe younger Sanger enjoyed learningabout the science of medicine, he wasnot interested in the art of diagnosingand treating disease

By all accounts, Sanger was not a liant student In 1936 he was accepted

bril-to the University of Cambridge, but hestruggled with the basic sciences Sang-

er received passing grades in try, but he ÒbombedÓ in physics ÒI nev-

chemis-er won scholarships,Ó he notes ÒI amnot sure I would have been able to at-tend Cambridge if my parents had notbeen fairly rich.Ó

Sanger found his calling early in

his college career An enthusiastic fessor, Ernest Baldwin, enticed him tostudy the new science of biochemistry.Sanger enjoyed the subject so thorough-

pro-ly that he took advanced courses during

a fourth, extra year at Cambridge Twoweeks after the Þnal exams, he says, hewas Òvery surprised to learn that he hadbeen awarded a Þrst-class degree.Ó

In 1940, unlike most of his peers, the22-year-old Sanger did not go oÝ toÞght in World War II ÒI was brought up

as a Quaker, and I felt pretty stronglythat people should not go around kill-

ing others for any reason,Ó

he explains He successfullydefended his position before

a military tribunal and spentthe war years pursuing a doc-torate Graduate studentswere obviously in short sup-ply during the war, and thebiochemistry department atCambridge was all too hap-

py to accept a promisingstudent

After earning his Ph.D in

1943, Sanger joined the oratory of Cambridge pro-fessor A C Chibnall, whowas a pioneer in the Þeld ofprotein chemistry Chibnallasked Sanger to study insu-lin, the pancreatic hormonethat governs the metabolism

lab-of sugar The suggestion led

to a 10-year-long project thatestablished Sanger as theleader of his Þeld Insulin,like other proteins, is made

up of diÝerent amino acids.During the 1940s research-ers were very aware that thechemistry of proteins de-pends on the order in whichthe amino acids are arranged.The problem: no one hadfound a technique for deduc-ing the sequence

Chibnall and Sanger chose

to work on insulin for several reasons

It was available in a pure form, and itwas a small molecule, at least as pro-teins go More important, perhaps, in-vestigators realized that if they couldwork out the structure of insulin, theywould understand how it controlledsugar metabolismÑinsights that hadmany implications for medicine

To determine the sequence of lin, Sanger began with a simple strate-

insu-PROFILE : FREDERICK SANGER

Revealing the Hidden Sequence

FREDERICK SANGER transformed biochemistry by developing methods for determining the structure of proteins and DNA

gy that chemists often use to analyze

large compounds: trying to break the

molecule into fragments and then

Þg-ure out how the pieces Þt together

Eas-ier said than done To snip insulin into

pieces of a meaningful size and then

sort them, Sanger tried every trick in

the chemistÕs handbook and then came

up with some new schemes

One of SangerÕs important

innova-tions was a method for labeling an end

of a protein fragment These tags made

it easier to deduce what pieces belonged

together For example, if three amino

ac-idsĐcall them A, B and CĐare linked

together in some order in a chain, the

sequence could be determined in the

following way After labeling the chains,

they could be broken down into

individ-ual amino acids, indicating, say, that B

was at one end of the chain Then a new

sample of the same chains could be

cut into pieces consisting of two amino

acids The Þnal step was isolating all

those pieces that had a B at one end If

all the BÕs were linked to Ãs, then the

sequence would have to be BAC

While pursuing the insulin sequence,

Sanger tried hundreds of diÝerent

tech-niques ỊMost experiments go wrong,Ĩ

Sanger sighs ỊI didnÕt spend too much

time trying to Þgure out what went

wrong I just started thinking about the

next experiment It prevented me from

getting depressed.Ĩ

There was one experiment in

par-ticular that Sanger would like to

for-get In 1947 he spent a year in

Uppsa-la, Sweden, working in the laboratory

of the eminent biochemist Arne W K

Tiselius A technique Sanger developed

seemed to suggest that insulin was not

a single chain but rather four

cross-linked chains When he presented the

results to his mentor, Tiselius

suggest-ed that they rush to publish a paper

to-gether ỊI was rather shocked as he had

not really contributed anything.Ĩ

But as a junior member of the

labo-ratory, Sanger gave in ỊThe paper is the

only one of which I am ashamed,Ĩ he

comments Sanger later discovered that

insulin is actually made of two

cross-linked chains, one 30 amino acids long

and the other 21 long The larger chain

was by far the most complex structure

that a protein chemist of that era had

ever struggled with It was not until

1952 that Sanger and his co-workers

Hans Tuppy and E.O.P Thompson

Þg-ured out the complete sequence of

both chains

Sanger then had to decide how the

chains linked together to make an

insu-lin molecule The problem turned out

to be extremely complicated An initial

analysis seemed to show that chains

were linked at nearly every point Then

Sanger realized that his techniques forcutting the insulin molecule were in-troducing new bridges between thestrands By 1955 he found a way to pre-vent the introduction of cross-links andsucceeded in determining the completestructure of insulin

Four years later, Sanger won the bel Prize in Chemistry for the insu-lin work He was immediately besieged

No-by professors inviting him to teach

and administrators asking him for vice But Sanger wanted no part of it ỊI have actively tried to avoid both teach-ing and administrative work,Ĩ he says

ad-ỊThis was partly because I thought Iwould be no good at them but also out

of selfishness.ĨIronically, in the year Sanger became aNobel laureate, his research began tofounder He had taken the insulin study

as far as it could go and was looking fornew questions to grapple with ỊI thinkthese periods occur in most peopleÕsresearch careers and can be depressingand sometimes lead to disillusion,Ĩ heobserves ỊI have found that the bestantidote is to keep looking ahead.Ĩ

In 1961 Sanger joined the Laboratory

of Molecular Biology at the Medical search Council in Cambridge and hiredCoulson as his research assistant ỊHeÕsnot the kind of guy you pal around withright away,Ĩ Coulson says

Re-Soon after his arrival at the tory, Sanger decided his talents might

labora-be useful for analyzing DNA, the cule that stores the genetic code In the1950s James Watson and Francis Crickhad Þgured out that DNA was a long,double helix made of four diÝerent nu-cleotides The arrangement of nucleo-tides determines what proteins an or-ganism can make, that is, what genes itwill express Yet at the time, scientistscould determine the sequence of nu-cleotides for only a very small section

mole-of a DNA strand

Sanger set an ambitious goal for self: sequencing the many thousands

him-of nucleotides in the DNA him-of a virus Yet

he and other biochemists soon foundDNA more diÛcult to analyze than pro-teins for two reasons First, they hadless experience handling DNA than theydid proteins Second, DNA is made fromonly four basic building blocks, whereasproteins are constructed from some 20diÝerent amino acids Just as a puzzle

with many similar pieces is more cult to solve than one with many diÝer-ent pieces is, the sequence of DNA wasmore diÛcult to decode than the se-quence of insulin was

diÛ-For more than 10 years, Sanger tigated techniques for sequencing DNA,competing with several laboratoriesaround the world Then in 1975, Sang-

inves-er and Waltinves-er Gilbinves-ert of Harvard versity, working independently of oneanother, developed methods for rapid-

Uni-ly sequencing DNA Their techniquescould determine the arrangement ofnucleotides in segments of DNA 200 ormore units long in a few days With ear-lier methods, the job would have takenyears The new technique enabled Sang-

er and his collaborators to determinethe sequence of 5,375 nucleotides in avirus called fX174

In 1980 Sanger won a second bel Prize, which he shared with Gilbertand Paul Berg of Stanford Universi-

No-ty Berg had found a way to insert

piec-es of DNA from one organism into theDNA of another His work launched thetechnology of recombinant DNA.During his career, Sanger publishedabout one major scientiÞc paper everyeight years, but his colleagues say eachone is a classic in experimental bio-chemistry ỊI donÕt publish papers unless

I have something to write about andsomething I am sure about,Ĩ he explains.These days, although Sanger still givesadvice to his colleagues at the Laborato-

ry of Molecular Biology, he has retiredfrom his research ỊThe aging processwas not improving my performance inthe laboratory,Ĩ he complains ỊI wouldhave felt guilty about occupying a spacethat could have been available to a youn-ger person.Ĩ

Reßecting on his research, SangerÞnds it hard to recall any moments ofgreat inspiration He does not subscribe

to the Ịpopular idea that scientiÞc ress depends on sudden breakthroughs.ĨInstead he thinks about many eventsthat Ịwere more often associated withsmall and gradual advances.Ĩ

prog-Sanger Þnishes his lunch and invites

me to his home We drive through thecountryside to an unpretentious housesurrounded by an elaborate garden Itseems Sanger is as devoted to garden-ing as he once was to research He andhis wife of 52 years, Joan, grow a doz-

en varieties of ßowers, cultivate

sever-al fruit trees and raise chickens Whenthey are not working in the garden, the Sangers entertain their three chil-dren, grandchildren and friends ỊI have always had a happy and peaceful life,Ĩ

he remarks The key, he asserts, Ịisworking on the right thing at the righttime.Ĩ ĐRussell Ruthen

SangerÕs talent for solving the right problem at the right time earned him two Nobel prizes in chemistry.

On some hot, sunny days, the 14

million residents of the Los

An-geles area inhale a thick,

brown-ish-gray haze, and no one can ignore

its eÝect The smog obscures the San

Bernardino Mountains and the warm

California sun; it irritates the eyes and

nose; it restricts the activities of

ath-letes and people who have breathing

disorders; it injures the lungs of both

young and old

Southern CaliforniaÕs air quality is

the worst in the U.S Air pollution in

the region reaches unhealthful levels

on half the days each year, and it

vio-lates four of the six federal standards

for healthful airĐthose for ozone, Þne

particulates, carbon monoxide and

ni-trogen dioxide In 1991 the South Coast

Air Basin exceeded one or more federal

health standards on 184 days

Yet these statistics hide a

remark-able accomplishment of the citizens ofsouthern California Los Angeles is one

of the few places in the nation whereair quality has improved dramaticallysince the 1970s From 1955 to 1992 thepeak level of ozoneĐone of the best indicators of air pollutionĐdeclinedfrom 680 parts per billion to 300 partsper billion The California Air ResourcesBoard recently documented that popula-tion exposure to unhealthful ozone lev-els has been cut in half in just the pastdecade Furthermore, the smog levelsmeasured during each of the past threeyears have been the lowest on record

All these improvements were achieved

at a time when human activity in the LosAngeles area was increasing at a rapidrate Since the 1950s the populationhas almost tripled, from 4.8 million to

14 million; the number of motor cles on the road has more than quad-rupled, from 2.3 million to 10.6 million;

vehi-and the city has grown into one of themost prosperous regions of the world

Although the residents of southern

Clearing the Air

in Los Angeles

Although Los Angeles has the most polluted skies

in the nation, it is one of the few cities where air quality has improved in recent decades

by James M Lents and William J Kelly

JAMES M LENTS and WILLIAM J

KEL-LY work together at the South Coast Air

Quality Management District ( AQMD ),

the regional air-pollution control agency

for the Greater Los Angeles area Before

becoming executive oÛcer of AQMD in

1986, Lents headed the air-pollution

con-trol program for the state of Colorado

In 1970 he received a Ph.D in physics

from the Space Institute at the

Universi-ty of Tennessee Some 11 years ago

Kel-ly earned an M.A in journalism from

Co-lumbia University, and since then he has

written extensively on the environment

AIR POLLUTION SOURCES have creased in size and number in Los An-geles, yet technical innovation and so-cial policy have led to an improvement

in-in air quality durin-ing the past two cades Some typical sources of air pol-

de-lution include ( from left to right)

indus-trial coatings, barbecues, trash ators, paints, dry cleaners, commercialovens and motor vehicles

inciner-California still face and continue to

tack-le many air pollution probtack-lems, they

have an advantage in that they have

spent 50 years studying the local

at-mosphere and experimenting with

var-ious policies We hope the cities of all

nations will learn from the experiences

of Los Angeles [see ÒThe Changing

At-mosphere,Ó by Thomas E Graedel and

Paul J Crutzen; SCIENTIFIC AMERICAN,

September 1989]

The movement to clean up the air

in southern California began

dur-ing the 1940s, a period of rapid

industrialization At the time, the

re-gion was plagued by sudden Ògas

at-tacksÓ that irritated the eyes,

dimin-ished visibility and produced an

un-pleasant odor Then, as now, the smog

was so obvious and odious to the

pub-lic that elected leaders were compelled

to take meaningful action Yet their

ef-forts provoked strong conßict Some

citizens and industries Þercely resisted

suggestions to clean up sources of

pol-lution But the Los Angeles Times

pub-lished dozens of editorials demanding

that the smog problem be solved Thepaper also put its money behind its ed-itorial mouth In 1947 it retained Ray-mond R Tucker, the former smoke reg-ulation commissioner of St Louis, tostudy air pollution in the area TuckeridentiÞed and investigated several ma-jor sources of air pollution, includingheavy industries, foundries, motor vehi-cles, backyard incinerators and smudgepots for protecting crops from frost

In the same year, the oil industrypaid the Stanford Research Institute( SRI ) to give another perspective on thecauses and control of pollution The or-ganization discovered that the hazydays were caused in part by a naturalweather phenomenon known as an in-version layer The warmest part of theatmosphere is, more often than not,that nearest to the ground, but undercertain conditions a layer of cool air canslip underneath a stratum of warm air

Such inversions often form oÝ the coast

of Los Angeles as the PaciÞc Oceancools the atmosphere just above it Af-ter ocean breezes blow the air mass in-land, the inversion layer traps air pol-

lutants in the cool air near the groundwhere people live and breathe Themountains that surround the regioncompound the problem; they preventthe pollutants from dispersing.SRI pointed out that natural materialssuch as dust, pollen, Þbers and salt wereimportant components of the haze Butthe institute also recognized that in-dustries and motor vehicles contributed

to the problem by adding carbon cles, metallic dust, oil droplets and wa-ter vapor

parti-In the 1950s SRI and Arie J Smit and his colleagues at the Califor-nia Institute of Technology began to ex-amine the chemistry of the atmosphereabove Los Angeles Their work and theresearch of others have revealed thecomplexity of atmospheric chemistry.Automobiles, factories and other sourc-

Haagen-es release such raw pollutants as drocarbons, water vapor, carbon mon-oxide and heavy metals When thesechemicals are exposed to intense sun-shine, they react to yield a vast number

hy-of secondary pollutantsÑfor instance,ozone, nitrogen dioxide, various organ-

ic compounds and acidic particles of

ni-trate and sulfate This concoction then

interacts with plants and animals,

caus-ing a variety of diÝerent eÝects Many

of these phenomena are still not

un-derstood, but 50 years ago even less

in-formation was available

In 1953, with the public fearing that

the Los Angeles haze might become as

bad as LondonÕs ỊkillerĨ fog, Governor

Goodwin J Knight appointed an

air-pollution review committee Chaired by

Arnold O Beckman of Beckman

Instru-ments, the committee proposed Þve key

ideas for reducing pollution over the

short term First, they asked that the

emission of hydrocarbons be reduced

by improving procedures for

transfer-ring petroleum products Second, they

set standards for automobile exhausts

Third, they encouraged the use of trucks

and buses that burned liqueÞed

petro-leum gas instead of diesel fuels Fourth,

they considered whether industries that

polluted the area heavily should be

asked to slow their growth Fifth, they

advocated that the open burning of

trash be banned

The committee also hoped that over

the long term Los Angeles would

devel-op a sustained automotive

pollution-control program, construct a rapid

tran-sit system and start a cooperative

pro-gram to regulate industrial sources of

pollution Ironically, the report was

is-sued while the regionÕs public train

sys-tem was being dismantled Today Los

Angeles is trying to get back on track,

so to speak, by developing an extensive

regional commuter rail network The

Beckman committeeÕs

recommenda-tions eventually grew into a coherent

air-quality management plan for the

gion, but it emerged slowly and was

re-shaped many times

Soon after the Beckman report, the

neighboring counties of Orange,

Riv-erside and San Bernardino began their

own pollution-control programs During

the past 40 years, these regions have

experienced explosive growth in

popula-tion and in vehicular traÛc The control

programs in these regions were, for the

most part, as energetic and innovative

as those in Los Angeles But residents of

these counties soon realized that they

needed to coordinate their eÝorts; smog

does not respect political boundaries

In 1975 the regional governmentstried voluntarily to consolidate theirpollution-control programs Two yearslater little progress had been made Sothe California legislature forged an un-easy alliance between the local pro-grams by creating the South Coast AirQuality Management District ( AQMD)

The AQMD was given jurisdiction overthe counties of Los Angeles, Orange andRiverside and part of San BernardinoĐ

an area of 13,350 square miles

Initially, the AQMD was responsiblefor stationary sources of air pollution,and the California Air Resources Boardwas assigned to regulate mobile sourc-

es such as cars, trucks and buses In itsearly years the AQMD adopted a view-point held by many business leaders

It argued that many industries were

as clean as they could get, the statewas not doing enough to clean up carsand the region might never achieveclean air standards As required by law,the AQMD adopted air-quality manage-ment plans in 1979 and 1982, but thesewere regarded as mostly paper exercis-

es At the time, the federal Clean Air Actrequired all American cities to achievefederal standards by 1987, althoughnearly everyone realized that the taskwould be impossible in Los Angeles Asthat deadline drew near, however, en-vironmentalists and even some busi-ness groups attacked the AQMD for its complacency and alleged lax en-forcement In 1987 the legislature re-structured the AQMD governing board and granted broad powers to the new

WEATHER PATTERN known as an

in-version layer traps air pollution above

Los Angeles and neighboring counties

Air cools over the ocean and then is

blown inland, creating a cold layer near

the ground and a warmer layer above

The warm layer prevents most of the

smog from escaping upward The

sur-rounding mountains keep the polluted

air from moving farther inland

SAN GABRIEL MOUNTAINS

WARM AIR

COOL AIR

OCEAN BREEZE

OTHER POLLUTION SOURCES (RESIDENTIAL, COMMERCIAL)

P A C I F I C O C

board The AQMD now has

responsi-bility for achieving local, state and

fed-eral standards

Despite all the political upheaval,

the state and local governments had

managed to curb many sources of air

pollution in the South Coast Air Basin

before the AQMD was formed in 1977

They gradually developed regulations

that concentrated on reducing the

ma-jor sources of air pollution: particles

from trash incineration, emissions from

industry and pollutants from motor

vehicles

In 1958, at the recommendation of

the Beckman committee, backyard

trash incinerators were banned

de-spite opposition from the public and

some waste-disposal managers More

than 300,000 families owned such cinerators, and many were unwilling

in-to give up the convenience and costsavings But slowly attitudes changed ;today most residents of Los Angeleswould consider it shameful to burntheir trash and subject their neighbors

to the smoke and smell

Waste-disposal managers objected tothe ban for a diÝerent reason These an-alysts realized that the economical al-ternative to burning trash was burying

it, and therefore they correctly

predict-ed that although replacing incineratorswith landÞlls would reduce air pollu-tion, the additional landÞlls would cre-ate other types of environmental prob-lems For example, as rain has seepedthrough the landÞlls and carried awaysoluble materials, it has contaminated

the local groundwater Southern nians are now working to clean up thegroundwater, but the long-term plansare, Þrst, to decrease the volume oftrash through recycling and, second, toreduce further groundwater contami-nation by improving the management

Califor-of landÞlls

The ban of backyard incineratorsgenerated relatively little public resis-tance when compared with attempts toclean up industrial sources of pollu-tion Historically, the control of indus-trial emissions has been challengingbecause of the need to balance the re-gionÕs environmental interests with itseconomic needs The Beckman report of

1953 quickly led to requirements thatindustries use vapor-recovery equip-ment when they transferred petroleumproducts, but service stations were notobliged to install such equipment ongasoline pumps until 1978 This equip-ment, while originally somewhat diÛ-cult for the motorist to handle, has beenstreamlined, so it is eÝective and easy

to use today Yet the measure is still resisted by most communities outside

of California because of its expenseand an undeserved reputation for be-ing cumbersome

Regulations were adopted during the1960s to eliminate industrial solventsthat play a major role in promoting theformation of ozone The rules aÝect-

ed a wide variety of businesses, from construction to auto manufacturing to dry cleaning To meet the requirements,most industries chose to use nonreac-tive solvents instead of installing con-trol equipment because it was the lessexpensive solution Sadly, the nonre-active solvents were later shown to destroy ozone in the stratosphere (Athigh altitudes, ozone serves the criticalfunction of shielding the earth fromharmful solar radiation.) The solution

to this problem is to Þnd truly benignsolvents; for example, Hughes AircraftCompany has recently developed a sol-dering ßux made from citrus juice.Although Californians succeeded inreducing the quantity of pollution gen-erated by certain industrial sources,their eÝorts have been partially oÝset

in recent decades Many sources havebeen introduced as the regionÕs econo-

my has grown To compensate for thegrowth factor, oÛcials instituted, in

1976, Ịnew sourceĨ regulations Theserequired expanding industries to usethe cleanest technology available Therules also speciÞed that if a companyplanned to start a project that increasedemissions, it was required to earn acertain number of credits by reducingemissions from another project.The new source regulations have stim-

SAN BERNARDINO MOUNTAINS

SANTA ANA MOUNTAINS

ON-ROAD VEHICLES

(CARS, BUSES, TRUCKS)

HEAVY INDUSTRY

OFF-ROAD VEHICLES (PLANES, TRAINS, SHIPS)

C E A N

Air Pollution in the Los Angeles Areaý

HYDROCARBON EMISSIONS, 1992

(TOTAL: 1,375 TONS PER DAY)

NITROGEN OXIDE EMISSIONS, 1992

(TOTAL: 1,208 TONS PER DAY)

*For Los Angeles, Orange, Riverside and San Bernardino counties

SOURCE: South Coast Air Quality Management District

ORANGE

ulated considerable innovation When

researchers compared the best

con-trol technologies of 1976 with those

available in 1990, they found that the

amount of pollution generated

dur-ing the manufacture of a product had

dropped, on average, by 80 percent

Un-fortunately, many businesses found a

way to avoid the regulations The rules

exempted new projects that would emit

daily less than 75 pounds of

hydrocar-bons or 100 pounds of nitrogen oxides

To exploit this loophole, many

indus-tries expanded by adding several small

projects instead of a few large ones so

that all the undertakings would be

ex-empt Furthermore, most small

busi-nesses were exempt, even though they

have accounted for the fastest-growing

segment of stationary source emissions

during the past few decades

Growth of these small sources,

com-bined with increased population and

motor vehicle use, tended to oÝset

cleanup eÝorts through the 1980s

Con-sequently, in 1990 the new source

reg-ulations were revised to include

proj-ects that introduced small quantities of

air pollution Even then, some

exemp-tions were allowed, and they continue

to hurt the eÝort

Los AngelesÕs diÛculties with

reduc-ing industrial pollutants were

sim-ilar in one respect to its attempts

at diminishing automotive emissions

Although oÛcials identiÞed the

prob-lem and took action, they were slow to

adjust to the scale of the matter From

the start, for example, the Beckman

committee recommended controls on

automotive emissions, but it did not

an-ticipate how quickly auto travel would

grow throughout the region

Neverthe-less, Los Angeles has succeeded in three

important areas: promoting the use of

clean fuels, purifying engine exhaust

and encouraging carpooling and the

use of public transportation

A signiÞcant Þrst step in

regulat-ing automotive pollution was the

reduc-tion of ỊblowbyĨ gases Most car engines

that were designed 30 years ago allowed

the gaseous by-products of combustion

to escape past the piston, through the

crankcase and into the atmosphere In

1960 such blowby gases accounted for

one quarter of the total hydrocarbon

emissions in the region The gases could

be eliminated by installing a $5 device

that routed the crankcase emissions to

the engine intake manifold; in this way,

the hydrocarbon fumes were burned

instead of being released In 1963 the

state mandated that the devices be

add-ed to all motor vehicles, new and old By

the middle of the decade, a special

po-lice force was patrolling the roadways

in southern California, issuing citations

to vehicles that smoked excessively

The crankcase device proved tive on new vehicles, but objectionswere raised to equipping old cars withthem Many auto mechanics were nottrained to install the devices, and ru-mors spread that even properly in-stalled devices caused engine damage

eÝec-That misinformation was the primaryreason for the state legislatureÕs deci-sion in 1965 to stop requiring the in-stallation of crankcase devices in carsmade before 1963 It was not the Þrsttime, nor certainly the last, that inaccu-rate information has become the basis

of political compromise

In addition to installation of case devices, Los Angeles tried to re-duce automotive pollution by encour-aging the production and use of cleangasolines In 1960 oÛcials required thatall gasoline sold in the area have a lowcontent of oleÞns, which are some ofthe most reactive compounds in petro-leum products Unfortunately, as theyears passed by, the clean gasoline pro-gram lost its momentum Then, in theearly 1980s, the state government ofColorado revitalized the concept, hop-ing to reduce air pollution problems inDenver Colorado required the blending

crank-of a methanol derivative into gasoline

In California, meanwhile, the AQMDand the state were pushing for metha-nol and other alternative fuels throughregulatory programs Faced with losinglarge market shares to methanol andother clean fuels, the Atlantic RichÞeldCompany in Los Angeles came up with

a constructive solution by developing atruly clean gasoline These events stimu-lated California and the federal govern-ment to go beyond regulations and in-troduce new legislation for clean fuels

Another development in reducingautomotive emissions was the remov-

al of pollutants from the exhausts ofcar engines through the use of catalyticconverters California required that be-ginning with the 1975 model year, allnew cars have the converters The con-verters and similar devices substan-tially reduced emissions, but for sev-eral reasons the program was not as successful as it might have been First, researchers discovered that the per-formance of a converter decreases as the car accumulates mileage Second, many motorists tampered with or even removed the emissions-control equip-ment Third, many consumers damagedtheir converters by fueling their carswith cheap leaded gasolines instead ofthe more expensive unleaded varieties

Indeed, in 1983 the National ment Investigations Center found that

Enforce-in almost one Þfth of all the vehicles on

the road, the emissions-control ment was either removed by the own-

equip-er or damaged by the use of leaded els The latter problem has abated inrecent years as the price of unleadedfuels has moved closer to that of lead-

fu-ed gasoline

To enforce legislation on automotiveemissions, Los Angeles initiated an in-spection program in 1976 The regula-tions speciÞed that whenever a vehiclewas sold, the new owner was required tobring the car to a special station wherethe emissions could be measured withthe engine of the car running at normalcruising speeds Los Angeles had manydiÛculties managing the program, andthe state government was reluctant tohelp Consequently, the city received in-numerable complaints from car ownerswho were forced to wait in long lines toget their vehicles tested

The California legislature was ally persuaded to develop a new inspec-tion program when the EnvironmentalProtection Agency imposed sanctionsthat reduced highway funding Underthe new Ịsmog checkĨ program, eachcar in California is tested once every twoyears at a repair shop Furthermore, thestate gave local governments the option

eventu-to conduct tests either while the enginewas idling or while it was running atnormal speedsĐthe latter option beingmore expensive but more eÝective Theprogram achieved pollution reductions,but the results were lower than expect-

ed Today the federal government isagain pressuring California to create aprogram that lives up to its potential

OÛcials had much trouble

en-couraging southern ans to share rides, thereby re-ducing congestion and emissions Oneway to promote ride sharing is to re-serve lanes on major highways for theuse of cars transporting two or morepeople When state transportation ad-ministrators proposed restricting one

Californi-of four lanes Californi-of the Santa Monica way, however, motorists opposed theidea so strenuously that the oÛcialsquickly withdrew it Later they foundthat the public was less antagonistic ifnew lanes were added to existing high-ways for the use of high-occupancy ve-hicles California now has an extensivenetwork of such lanes

Free-In 1987 the AQMD required Þrmswith 100 or more employees to oÝerincentives to institute carpooling Al-though the regulation generated com-plaints, it has been successful The ruleapplies to 5,200 work sites, harboring1.2 million employees A study of thatgroup from 1987 to 1992 revealed thatthe number of employees per vehicle

rose from 1.13 to 1.24 The program

has eliminated 90,000 trips a day and

has achieved about half the desired

re-ductions in emissions In fact, southern

California is one of the few places in the

U.S where ride sharing has increased

over the past Þve years

Today the motor vehicles and

indus-tries operating in California are among

the cleanest in the world A new car sold

in California emits just one tenth of the

pollution that a new car did in 1970

Such industries as electric utilities rely

almost exclusively on clean-burning

natural gas Manufacturing plants and

construction companies use advanced

paints, solvents and adhesives that have

been formulated to minimize pollution

For these reasons and others, southern

California has made tremendous

prog-ress in reducing air pollution

But despite all the successes,

pollu-tion in the South Coast Air Basin is still

overwhelming Average daily emissions

total 1,375 tons of hydrocarbons, 1,208

tons of nitrogen oxides, 4,987 tons of

carbon monoxide, 134 tons of sulfur

oxides and 1,075 tons of particulates

Transportation, including cars, airplanes,

trains and ships, adds 47 percent of the

hydrocarbons, 70 percent of the

nitro-gen oxides, 90 percent of the carbon

monoxide, 60 percent of the sulfur

ox-ides and 89 percent of the particulates

Industry contributes 26 percent of the

hydrocarbons, 18 percent of the

nitro-gen oxides and 30 percent of the sulfuroxides and smaller percentages of oth-

er pollutants The remainder of the lutants are emitted by households andservice-oriented businesses, such as res-taurants, dry cleaners, gas stations andoperations in commercial buildings

pol-Los Angeles must reduce pollutioneven further to meet federal healthstandards Computer projections indi-cate that southern Californians mustcut hydrocarbons by 80 percent, nitro-gen oxides by 70 percent, sulfur oxides

by 62 percent and particulates by 20percent The consequences of compla-cency could be disastrous

Air pollution in the South Coast Air

Basin has already taken a gering toll on residents A grow-ing body of evidence reveals that thesmog is a serious health hazard

stag-In 1991 David Abbey, an gist at Loma Linda University, found acorrelation between long-term expo-sure to air pollution and the develop-ment of chronic diseases Abbey stud-ied 6,340 Seventh-Day Adventists, 62percent of whom lived in the basin (theremainder lived throughout California)

epidemiolo-In that group of 6,340, those who

resid-ed in areas that exceresid-edresid-ed governmentstandards for suspended particles on

42 days or more per year had a higherrisk of respiratory disease, including a

33 percent greater risk of bronchitis

and a 74 percent greater risk of

asth-ma In addition, women living in thoseareas had a 37 percent greater risk ofdeveloping some form of cancer.Roger Detels of the University of Cal-ifornia at Los Angeles studied respira-tory disease among residents in threeareas of southern California that diÝer

in air quality During a Þve-year period,

he periodically questioned and ined volunteers, aged seven to 59 years,

exam-to determine the health of their ratory systems In 1987 Detels discov-ered that residents who lived in high-pollution areas had more symptoms ofrespiratory disease, such as bronchitisand asthma, than did people who resid-

respi-ed in low-pollution regions

If the citizens of the South Coast AirBasin succeeded in meeting federalstandards for ozone and particulates, a

1989 study by the AQMD predicted,they would gain $9.4 billion in healthbeneÞts every year If the residentscould reduce just particulates to the fed-eral standard, they would prevent 1,600premature deaths annually among thosewho suÝer from chronic respiratory dis-ease In addition, they would eliminate

15 million person-days on which peoplewith respiratory disease are unable to

go to work, school and other activities,and they would reduce the risk of dyingprematurely from exposure to particu-lates ( The risk of death is about one in10,000Đabout half the risk a Californi-

an faces of dying in a car accident.)

If Los Angeles residents were able toachieve the federal standard for ozone,they would eliminate annually 18 mil-lion person-days of restricted activity,

65 million person-days of chest fort, 100 million person-days of head-ache, 120 million person-days of cough-ing, 180 million person-days of sorethroats and 190 million person-days ofeye irritation

discom-These terrible statistics have capturedthe attention of the federal and stategovernments The federal Clean Air Act,which was amended in 1990, gives LosAngeles until 2010 to achieve feder-

al health standards, but the law also requires the region to make incremen-tal progress toward healthful air TheCalifornia Clean Air Act demands thatthe region reduce emissions by 5 per-cent a year until health standards have been met

Between 1989 and 1991 the AQMD vised an air-quality management plan torespond to the strong mandates of thefederal and state law During the next

de-17 years, the plan seeks to reduce lution from virtually all sources and fos-ter the development of new and clean-

pol-er technologies The AQMD will executethe plan with the cooperation of the

MESSENGERS for the Rapid Blueprint Company in Los Angeles were outÞtted

with gas masks in the fall of 1955 so that they would not suÝer from the eÝects of

the smog Shortly thereafter, local oÛcials began air-pollution control programs

EPA, local governments and state Air

Re-sources Board Each organization has a

well-deÞned role The AQMD is charged

with cleaning up stationary sources and

encouraging carpooling The EPA is to

set standards for airplanes, trains and

ships that travel through the region

Local governments will work to

alle-viate traÛc through improvements in

the transportation infrastructure and

expanding mass transit The Air

Re-sources Board is responsible for motor

vehicle standards It has already taken

aggressive action that should reduce

automotive emissions by 85 percent by

the next decade and should increase

sales of electric cars

The AQMD and its sibling

agen-cies plan to attack the pollution

problem in three stages The Þrst

stage, dubbed Tier I, includes 135

mea-sures that can be accomplished using

ex-isting technologies and are to be

adopt-ed by 1996 The measures limit and

re-duce pollutants from such sources as

electric utilities, motor vehicles, small

businesses and even backyard

barbe-cues The electric companies will be

in-stalling low-polluting burners and

cata-lysts on power plants to reduce nitrogen

oxide emissions Pollution from

back-yard barbecues will diminish as

house-holds use reformulated charcoal

er ßuid and other products for

light-ing grills Automotive emissions will

de-crease because of tailpipe standards as

well as programs encouraging carpools

and use of public transportation

The second stage, or Tier II, will take

the region into the 21st century These

measures rely on technologies that have

just entered the commercial market

The list includes a new house paint,

developed by the Glidden Company,

that does not release hydrocarbons

An-other initiative is developing

automo-bile engines that run on methanol,

nat-ural gas or other alternative fuels The

third stage, Tier III, requires

technolo-gies that have not been fully developed

but are likely to be available in the next

decade or so Researchers are

fabricat-ing paint coatfabricat-ings that dry under

ul-traviolet lamps, without emitting any

signiÞcant quantity of pollutants They

are also working on fuel cells, such as

those used in the space shuttle, which

will power a kind of zero-emission

elec-tric vehicle

To help develop these new

technolo-gies, the AQMD is working with

govern-ment agencies and private corporations

through its technology advancement

of-Þce To date, the oÛce has contributed

some $40 million in seed money to

fa-cilitate almost 250 projects Among the

technologies now entering

commercial-ization are alternative-fuel vehicles, dustrial emissions controls for nitrogenoxides, and cleaner paints, solvents andcoatings

in-After the AQMD released its tious plan to clean up the air, the orga-nization quickly realized it could notsimply play the role of tough pollutioncop and expect companies, especiallysmall ones, to follow the plan The re-gionÕs businesses, which were caught in

ambi-a recession ambi-and fambi-aced cuts in defensespending, were reluctant to invest re-sources in pollution-control equipment

To sustain progress, a diÝerent proach was required, especially to clean

ap-up the growing number of small

sourc-es of air pollution In 1991 the AQMDoutlined a series of reforms, which in-cluded providing free technical assis-tance and starting a guaranteed loanprogram so that small businesses couldpurchase pollution-control equipment

In the long run the task of gettingbusinesses to reduce pollution will hinge

on technological innovation One of theAQMDÕs most powerful incentives fordeveloping these clean technologies isthe Regional Clean Air Incentives MarketProgram Under this program, a busi-ness is asked to meet emission stan-dards not only for individual pieces ofequipment but also for its entire op-erations and facilities Those business-

es that cut pollution below the standardwill be granted emission-reduction cred-its Those companies that generate morepollution than the standard must buyenough credits to make up the diÝer-ence The program gives Þrms ßexibility

to choose how to reduce emissions, and

it also provides an incentive for

innova-tion because Þrms can make a proÞt bycleaning up the air

During the past 40 years, the zens of Los Angeles have been very ac-tive in the Þght for clean air, but from time to time they have neglected the is-sue because they feared that air-quality regulations were incompatible with the regionÕs economic interests Today, de-spite occasional doubts about environ-mental regulation and legitimate con-cerns about the postÐcold war economy

citi-of Los Angeles, residents and businesspeople seem to recognize the need tosolve the serious air-pollution problems

We hope that by 2010 when our dren climb the San Bernardino Moun-tains, they will see not a sea of smogbut rather the PaciÞc Ocean and Cata-lina Island We would like them to live

chil-in a region where technological chil-vation sustains both the economy andthe environment But most important,

inno-we want them to reap the beneÞts ofhealthy, fresh air

L.A MECHANIC at Green Motorworks, a local manufacturer of electric vehicles,converts a gas-fueled car to battery power Such initiatives are receiving strong sup-port from California state agencies as a way to meet ambitious air-quality standards

FURTHER READINGENVIRONMENTAL ETHICS: DUTIES TO AND

Rolston Temple University Press, 1989

CHANG-ING GLOBAL ENVIRONMENT Cheryl S ver and Ruth DeFries National Acade-

Sil-my Press, 1990

FOR GOVERNMENTS, THE OPPORTUNITIESFOR BUSINESS Frances Cairncross Har-vard Business School Press, 1992

THE HUMAN SPIRIT Al Gore HoughtonMiÜin, 1992

Sometimes getting there is not half

the fun of geophysical research

Bearing westward into a rising

Southern Ocean gale, the Australian

re-search vessel Rig Seismic was pounded

so hard that its anchor broke loose and

smashed into the forecastle, causing the

waves breaking over the bow to ßood

parts of the ship Progress toward the

Kerguelen Plateau, an elevated region

of seaßoor just north of Antarctica, was

delayed while the shipÕs crew secured

the anchor and welded the ship to keep

it watertight Several days later the

ex-pedition, including one of us (CoÛn),

arrived at Kerguelen, deployed

instru-ments for analyzing the structure of

the ocean bottom and began collecting

data The aim of the voyage was to

un-derstand the origin and evolution of the

huge underwater plateau

In that same year, 1985, but in the

oth-er hemisphoth-ere, a ßoating earth science

laboratory and drilling vessel, JOIDES

Resolution, steamed north through

stor-my North Atlantic seas to the V¿ring

Pla-teau oÝ the coast of Norway That

ex-pedition, led by one of us (Eldholm) and

Jšrn Thiede of the Research Center for

Maritime Earth Sciences in Kiel, many, had as its purpose the investiga-tion of the geologic structures that formwhen continents tear apart and an ocean

Ger-is born Confounding the expectations

of most of their colleagues, the ers managed to obtain core samples ofigneous rock nearly one kilometer deep,lying under 300 meters of soft sediment

research-Although they lie thousands of meters apart in disparate geologic set-tings, we recognized fundamental sim-ilarities in the two plateaus Our Þnd-ings, merged with seismic and drillingdata gathered by many other workers,have helped demonstrate that the Ker-guelen Plateau and the volcanic conti-nental margin oÝ Norway belong to aclass of huge magmatic features gener-ically known as large igneous provinc-

kilo-es These features cover areas of up tomillions of square kilometers, and yetthey seem to have formed quite swiftly

The recognition of large igneous inces has forced geophysicists to re-think their notions regarding the struc-ture of the earthÕs interior The theory ofplate tectonics tidily accounts for theslow, steady volcanic activity that occurs

prov-at mid-ocean ridges (where new

ocean-ic crust is born) and near subductionzones (where old, dense sections of theocean ßoor sink back into the earthÕshot interior), but it cannot readily ex-plain the abrupt outbursts necessary tocreate large igneous provinces Althoughthe rocks that make up these featuresgenerally resemble the composition ofthe lavas that emerge at mid-oceanridges, igneous provinces diÝer in their

trace element content and mix of

atom-ic isotopes And they sometimes breakthrough the middle of the normallyplacid lithospheric plates

In the 1960s the late J Tuzo Wilson

of the University of Toronto, followed

by W Jason Morgan of Princeton sity, developed a hypothesis that helped

Univer-Large Igneous Provinces

These vast fields of lava record powerful but geologically

brief pulses of magmatic activity Their formation may

have triggered significant changes in the global environment

by Millard F CoÛn and Olav Eldholm

MILLARD F COFFIN and OLAV

ELD-HOLM have collaborated for the past

three years on a project to develop a

fuller understanding of giant igneous

formations CoÛn is a research scientist

at the Institute of Geophysics at the

Uni-versity of Texas at Austin He received

his Ph.D in geology in 1985 from

Co-lumbia University ; he met Eldholm while

completing his doctorate work Eldholm

is a professor of marine geophysics at

the University of Oslo in Norway, a

posi-tion he has held since 1981 He

complet-ed his doctoral work in geophysics at the

University of Bergen in 1976 In 1991 he

took a sabbatical at the University of

Tex-as to coordinate his investigations with

CofÞnÕs; a year later CoÛn took a

fellow-ship in Norway Both men have spent

considerable time on board research

ves-sels collecting data on submarine

igne-ous provinces

LARGE IGNEOUS PROVINCES (red

re-gions) appear in widely varied geologic

settings around the globe Dots indicate

HAWAII

COLUMBIA RIVER FLOOD BASALTS

to elucidate the phenomenon The

re-searchers proposed that the earthÕs

mantle, the vast zone lying below the

crust but above the core, circulates in

two diÝerent modes The dominant one

consists of large-scale convection that

nudges plates across the surface and

causes continents to drift But about one

tenth of the heat now escaping from

the mantle does so in the form of

deep-rooted, narrow plumes of warmer than

average material that rise through the

mantle When it reaches the base of the

lithosphere, a plume decompresses and

partially melts, producing an upwelling

of magma and a long-lasting locus of

volcanic activity known as a hot spot

Some of that magma may erupt as a

tremendous ßood of lava Unlike the

slow, steady drifting of continents and

the spreading of mid-ocean ridges, the

surfacing of mantle plumes takes place

in an erratic and episodic manner

All models of the internal structure

of the earth are built on inference Eventhe deepest boreholes extend only about

10 kilometers Direct and indirect ies of large igneous provinces are there-fore vitally important for learning moreabout the nature of mantle plumes andhow they may ultimately aÝect condi-tions on the earthÕs surface

stud-The most fundamental observation

about these provinces is that theyconsist of basalt, a common, iron-and magnesium-rich rock The large ig-neous features that appear in the middle

of continents, where geologists can ily collect samples and determine theircomposition, are therefore referred to

eas-as continental ßood beas-asalts At the

edg-es of continents, igneous provincedg-es arecalled volcanic passive margins; if theyare located in the middle of the ocean,they are dubbed oceanic plateaus

Geologists recognized the existence

of continental ßood basalts late in the

19th century, when they realized thatseveral far-ßung volcanic constructionsactually constitute connected ßows ofbasaltic lava One of the most spectac-ular of these is the Deccan Traps in

west-central India (Trap is Dutch for

Òstaircase,Ó a reference to the steplikeappearance of the eroded lavas.) Sim-ilar structures include the ColumbiaRiver basalts in the northwestern U.S.;the North Atlantic volcanic outpourings

of the British Isles, the Faeroe Islandsand Greenland; and the Karoo basalts

in southern Africa, to name just a few.The layers of lava in a ßood basalt pile

up several kilometers thick Individualßows may contain several thousand cu-bic kilometers of rock and may extendhundreds of kilometers

Because of the inaccessibility of derwater igneous provinces, earth sci-entists only recently became aware oftheir resemblance to continental ßoodbasalts In 1981 Karl Hinz of the Feder-

un-locations where seismic reßection studies seem to reveal

an-cient volcanism along continental margins The largest igneous

provinceÑthe Ontong Java PlateauÑcovers nearly two million

square kilometers Studies of these huge magmatic tions are providing new insight into the internal workings ofthe earth and how they may affect conditions at the surface

forma-ONTONG JAVA PLATEAU

BROKEN RIDGE PLATEAU KERGUELEN

PLATEAU RÉUNION

KAROO BASALTS

DECCAN TRAPS

SIBERIAN TRAPS NORTH ATLANTIC

VOLCANIC

MARGIN

LACCADIVE RIDGE

CHAGOS-MASCARENE PLATEAU

al Institute for Geosciences and Natural

Resources in Hannover, Germany,

sug-gested, based on his analysis of

record-ings of reßected seismic waves, that

the submerged margins of many

con-tinents contain extensive, layered lava

ßows Since then, investigators have

ob-tained improved seismic data that seem

to corroborate HinzÕs conclusion

Igne-ous rocks recovered by scientiÞc drilling

projects on the continental margins oÝ

Ireland and Norway conÞrm that the

structures seen in the seismic images

do in fact have volcanic origins

Analogous seismic studies of the

Ker-guelen Plateau present convincing

evi-dence that oceanic plateaus likewise

con-sist primarily of volcanic rocks More

recently scientists on board the JOIDES

Resolution have recovered direct

sam-ples of rocks from two immense

ocean-ic plateaus: Kerguelen in 1988 and

On-tong Java, situated in the PaciÞc Ocean

northeast of Australia, in 1990 The

sam-ples consist of basaltic rock, similar to

that found in continental ßood basalts

Earth scientists have wondered how

such massive volcanic constructions

arise Unfortunately, the most basic

in-formation needed to answer that

ques-tionĐthe total volumes of lava and trusive rock in igneous provinces andthe rate at which they formedĐis poorlyknown On land, geologists can directlymeasure the area of ßood basalts butcan only infer their depth The crust un-der continents averages 35 kilometers inthickness Only the top parts are acces-sible to exploration Beneath the oceansthe situation is even more diÛcult Sam-pling just the uppermost sections of theÞve- to 25-kilometer-thick crust in whichoceanic plateaus reside requires the ser-vices of an expensive drilling vessel

in-The passage of time also disguisesthe true extent of ancient volcanic struc-tures The older igneous provinces areoften heavily eroded and are thereforeboth disÞgured and diminished in scale

John J Mahoney of the University of waii estimates that the Deccan Traps,for example, originally covered an areathree times larger than at present More-over, in the 65 million years since theformation of the Deccan Traps, seaßoorspreading has apparently dispersed sec-tions of the original lavas to the Sey-chelles, the Mascarene Plateau and theChagos-Laccadive Ridge far to the southand southwest of India

Ha-Undersea volcanic margins and anic plateaus similarly have been re-conÞgured by the forces of change Drill samples and seismic-reßection images of the igneous province along the Norwegian continental margin and

oce-of the Kerguelen Plateau demonstratethat when those formations Þrst erupt-

ed, they stood above sea level and onlygradually subsided to their present po-sition in deep water CoÛn has estimat-

ed that the Kerguelen and Broken Ridgeplateaus, along with most of the otheroceanic plateaus and ridges in the Indi-

an Ocean, formed above sea level andremained there for as long as 50 mil-lion years, an ample span of time inwhich erosion could proceed

Despite the paucity of data and thephysical transformations of large igne-ous provinces, we have managed to de-duce the original areas covered by lavafor Þve of the best studied of these for-mations: the Ontong Java and Kergue-len-Broken Ridge oceanic plateaus, theNorth Atlantic volcanic margins and theDeccan and Columbia River continentalßood basalts The smallest of theseĐtheColumbia River basaltsĐencompasses aregion larger than the state of New York;

JOIDES RESOLUTION research vessel

(left) is the centerpiece of the Ocean

Drilling Project, the most extensive ternational program in earth sciences.The ship incorporates a drill that pene-trates the sediments that accumulate onthe ocean bottom and collects core sam-ples lying as much as two kilometers

in-deep (below) In this way, earth

scien-tists can directly analyze igneous rockfrom underwater continental marginsand from submerged oceanic plateaus

the largest, the Ontong Java, is two

thirds the size of Australia To evaluate

the area and volume of large igneous

provinces, geologists realized that they

needed to examine modern-day

ana-logues of the gigantic igneous provinces

No currently active volcanic

re-gions even begin to approach

the extinct large igneous

prov-inces in magnitude The most active

modern hot spots (the big island of

Ha-waii and RŽunion, an island lying to the

east of Madagascar) cover an area of the

earthÕs crust only about one Þfth the

size of the Columbia River basalts

Nev-ertheless, studies of the crustal

struc-ture below Hawaii contribute to the

un-derstanding of the generic forms that

underlie all volcanic hot spots

In 1982 Anthony B Watts and Uri S

ten Brink and their co-workers at the

La-mont-Doherty Earth Observatory

collect-ed seismic data by recording artiÞcially

generated sound waves that are

reßect-ed and refractreßect-ed underneath the

Hawai-ian Islands and seamounts The speed

of those waves depends on the physical

properties (such as density and elastic

constants) of the rock; where those

pa-rameters change, the seismic waves willalter course Seismic analysis revealsthat the Hawaiian Islands built up as basaltic magma intruded into and col-lected atop the preexisting oceanic crust

Those surface lavas evidently do notcontain all of the material associatedwith the islandsÕ formation, however

Below the islands and seamounts lies ananomalous zone of rock distinguished

by the rapid rate at which sional waves travel through it Geolo-gists think the underlying body of rockderived from the same mantle source asdid the islands and seamounts Exist-ing techniques cannot reveal how muchnew igneous material was added to thepreviously existing oceanic crust as theislands formed Any calculated volumes

compres-of mantle plumeÐderived material tained in the Hawaiian Chain thereforemust be considered minimal

con-Even so, we were able to establish abasic relation between seismic-velocitystructure (as well as other geophysicaldata) and the total volume of materi-

al encompassed within the Hawaiian lands We then used Hawaii as our mod-

Is-el for inferring the distribution of canic material in various ancient, large

vol-igneous provinces By our calculations,

a small province such as Columbia

Riv-er incorporates approximately 1.3 lion cubic kilometers of rock The giantOntong Java province contains at least

mil-36 million cubic kilometers of igneousrock, enough to bury the contiguousU.S under Þve meters of basalt.The next stage in understanding theorigin and signiÞcance of the various ig-neous provinces is to learn how quicklythey formed Did they accumulate slow-

ly and steadily over tens of millions ofyears, in much the way that new oceancrust forms at mid-ocean ridges? Or didthey emerge from a series of volcanicÞrestorms that swiftly pumped gasesand rock fragments into the air and wa-ter and that abruptly transformed thegeology of large areas of the earth?Geologists have expended a great deal

of eÝort attempting to answer thosequestions for the Deccan Traps Thisßood basalt has garnered special atten-tion because it erupted some 65 millionyears ago, at just about the time of themajor extinction at the end of the Cre-taceous period, and it may have con-tributed to that event [see ÒA VolcanicEruption,Ó by Vincent E Courtillot; SCI-

RIG SEISMIC vessel (top right), operated

by the Australian Geological Survey

Or-ganization, collects seismic information

about the ocean bottom An air-gun

ar-ray creates sound waves that reßect oÝ

the oceanic crust (below) Underwater

mi-crophones towed behind the ship sense

the reßected waves, which are then

corded on board the ship One such

re-cording (bottom right) reveals the

igne-ous structure (area below red line)

un-derlying the Kerguelen Plateau

ENTIFIC AMERICAN, October 1990]

Rob-ert A Duncan of Oregon State

Universi-ty, working with Vincent E Courtillot

and Didier Vandamme of the Institute

of Physics of the Earth in Paris and

sev-eral other colleagues, has performed

radioactive dating and magnetic

analy-sis on samples from the Deccan Traps

The scientistsÕ results indicate that most

of the lavas erupted within a span of

less than one million years Using

simi-lar methods, Ajoy K Baksi of Louisiana

State University found that the

Colum-bia River ßood basalts mostly erupted

over one and a half million years

Geologists have few direct samples

on which to judge the ages and rates of

formation of underwater volcanic

mar-gins and oceanic plateaus The scant

available evidence hints that submarine

igneous provinces accumulated about

as rapidly as those on land Eldholm

es-timates that the bulk of the North

Atlan-tic volcanic province formed in three

million years or less John A Tarduno

and his collaborators at the Scripps

In-stitution of Oceanography, in

conjunc-tion with Mahoney, conclude that the

Ontong Java province was constructed

in less than three million years Our

analysis of rock dating, conducted by

Hugh Davies, formerly of the Australian

Geological Survey Organization, and

Hu-bert Whitechurch, formerly of the

Uni-versity of Strasbourg, indicates that theKerguelen Plateau mostly formed with-

in a span of 4.5 million years

From a geologic perspective, the est igneous provinces emerged remark-ably fast In comparison, the RockyMountains have been rising for morethan 40 million years, and the chain ofHawaiian Islands and the Emperor Sea-mounts have been building for at least

larg-70 million years Moreover, much of thevolcanic activity associated with theprovinces may have occurred in short,violent episodes separated by long pe-riods of relative quiescence

Once we had collected data on the tal volume of the major igneous prov-inces and on how quickly they formed,

to-we could Þnally deduce the magnitude

of the volcanic forces that created them

The eruptions that built up the OntongJava province unleashed between 12and 15 cubic kilometers of igneous rockeach year; Deccan volcanism producedbetween two and eight cubic kilometersannually Assuming that the creation ofigneous provinces, like other eruptiveprocesses, occurs in Þts and starts, thepace of crustal production may havebeen far greater in some years For ref-erence, Roger L Larson of the Universi-

ty of Rhode Island estimates that theglobal network of mid-ocean ridges hasyielded between 16 and 26 cubic kilo-

meters of ocean crust a year over thepast 150 million years In other words,individual igneous provinces have gen-erated new crust at rates comparable to

or greater than that of the global ßoor spreading system

sea-Large igneous provinces build up so

swiftly compared with the ing of the earthÕs deep interiorthat they must have derived from single,discrete sources If one knows the vol-ume of basaltic rock contained in thoseprovinces, one can evaluate the dimen-sions of the hot plumes in the mantlethat led to their genesis Only a fraction

churn-of the plume material actually meltsand reaches the crust, and this fractionpresumably is smaller at greater depths,where increased pressures tend to keepthe mantle rocks in a solid state Hence,less melting should take place underthick continental lithosphere than belowthin oceanic lithosphere Stephen M Eg-gins of the Australian National Univer-sity and Shen-su Sun of the AustralianGeological Survey Organization estimatethat the molten, basaltic portion of theplume (the part that gives rise to the sur-face volcanism and to subsurface igne-ous intrusions) accounts for from 5 to

30 percent of the plumeÕs total volume

We used those numbers to calculatediameters for the thermal anomalies in

RISING PLUMES of hot material migrate through the earthÕs

mantle; where the head of the plume reaches the surface, a

large igneous province forms (left) Plumes probably originate

at the boundary layers between the core and mantle (the D′′)

and between the upper and lower mantle The parent plumes

of the most voluminous igneous provinces were so huge thatthey must have originated at least in part in the lower mantle,most likely at the D′′ The spheres on the right depict the mini-

mum (white) and maximum (dark orange) inferred diameters

of the plumes associated with Þve major igneous provinces

MID-OCEANRIDGE

INNERCORE

D′′

ONTONGJAVA PLATEAU

DECCANTRAPS

NORTH ATLANTICVOLCANIC MARGIN

COLUMBIA RIVERFLOOD BASALTS

KERGUELENPLATEAU

the mantle associated with the Þve

best-studied igneous provinces For

simplic-ity, we assumed the rising plumes to

be spherical because such a shape

rep-resents the most eÛcient way to

trans-port material and so permits the plume

to move at a plausibly slow velocity

through the earthÕs interior Our analysis

suggests that the Ontong Java province

must be derived from a mantle plume

at least 600 kilometers, and possibly as

much as 1,400 kilometers, in diameter

That size carries special signiÞcance

for geologists It indicates that large

plumes must contain at least some

ma-terial from the lower mantle, more

than 670 kilometers below the surface

At that depth, the velocity of seismic

waves changes abruptly, probably

be-cause of a change in the mineral

struc-ture of the mantle rocks The issue

of whether the whole mantle mixes or

whether the upper and lower parts of

the mantle behave as independent

sys-tems that circulate separately strongly

divides geophysicists

Our work favors models that allow

for at least some interaction between

the upper and lower mantles In our

view, the largest plumes originate in

the lower mantle, most likely at the D′′,

a region having unusual seismic

proper-ties that lies just outside the core

Small-er plumes may arise at the

670-kilome-ter-deep transition zone between the

upper and lower mantle

Regardless of where mantle plumes

originate, their attributes and eÝects

at the surface depend strongly on the

temperature, composition and physical

state of the material they encounter just

beneath the lithosphere These factors,

combined with the local strength of the

lithosphere, determine the volume,

tim-ing and position of the eruption at the

surface When plumes rise under

conti-nental masses, they may help force the

continents apart at a weak spot and

then induce the formation of extensive

volcanic features along the margins of

the rifted landmass [see ÒVolcanism at

Rifts,Ó by Robert S White and Dan P

McKenzie; SCIENTIFIC AMERICAN, July

1989] Under certain circumstances, the

plume may penetrate the thick central

regions of continental blocks and give

birth to a continental ßood basalt

If the upwelling mantle plume

surfac-es underneath the seaßoor, it may give

rise to an oceanic plateau Laboratory

experiments suggest that a drawn-out

tail of hot material should lag behind

the spherical head of the plume, yielding

a long-lived, focused source of magma

Over millions of years, plate motions

cause the ocean ßoor to migrate over the

site of the hot spot Where lava erupts

at the surface, it gradually constructs a

linear submarine ridge or a sequence ofislands and seamounts The Hawaiian-Emperor chain presumably developed

in this manner, although how a plumecould persist for more than 70 millionyears remains a puzzle

As the 1991 eruption of Mount

Pin-atubo in the Philippines ments, even moderate volcan-

docu-ic outbursts can severely damage thelocal environment Yet such geologicevents are incidental twitches comparedwith the convulsions of magmatic ac-tivity during the formation of large ig-neous provinces One can thus begin toimagine that those ancient eruptionshad profound consequences In 1972Peter R Vogt of the Naval ResearchLaboratory Þrst proposed that the sur-facing of a mantle plume would lead tophysical and chemical changes aroundthe globe; the environmental eÝects

associated with these changes couldstrongly inßuence the evolution of life.The current fascination with thegreenhouse eÝect and global change hasrenewed interest in VogtÕs ideas Ste-phen Self of the University of Hawaii andMichael R Rampino of New York Univer-sity note that the environmental impact

of a large igneous province depends inpart on whether it forms on land or un-derwater Ocean plateaus and volcanicpassive margins modify the geometry ofthe ocean basins and alter the global sealevel We estimate that accumulation ofthe Ontong Java plateau material elevat-

ed sea level by about 10 meters

Ocean-ic plateaus and volcanOcean-ic margins mayblock or constrict circulation in waysthat inßuence large-scale circulation,erosion and sedimentation, especially

if the volcanic activity occurs in a sitive location between ocean basins.Because they are much denser and

sen-ENVIRONMENTAL EFFECTS from the eruption of large igneous provinces include tuations in the global sea level Powerful volcanic activity could alter the chemistryand circulation of the atmosphere and oceans, inßuencing the evolution of life Volca-noes release carbon dioxide, which may contribute to greenhouse warming ; warmertemperatures promote the formation of black shale, in part by enhancing global bio-mass production In contrast, several mass extinctions of the past 150 million yearscoincide with the appearance of igneous provinces, hinting at a causal relationship

EXTINCTIONS

BLACK SHALE FORMATION

100 150 200 250

COLUMBIA RIVER FLOOD BASALTS

NORTH ATLANTIC VOLCANIC MARGIN

DECCAN TRAPS

KERGUELEN PLATEAU

ONTONG JAVA PLATEAU

INFERRED GLOBAL TEMPERATURE (DEGREES CELSIUS)

5 10 15 20 25 0

more massive than the atmosphere, the

oceans are far more able to absorb and

dilute gases and heat, a factor that

tends to ameliorate the consequences of

volcanic eruptions On the other hand,

submarine volcanism and associated

hydrothermal activity may leach trace

metals such as arsenic, which are

poi-sonous to marine life Heat produced

by submarine eruptions induces

ocean-bottom waters to rise to the top, altering

the circulation of the surface waters and

disrupting the organisms living there

Carbon dioxide emitted at an

under-sea volcanic site spreads through the

ocean In so doing, the compound could

increase the alkalinity of the seawater,

thereby aÝecting both marine life and

climate If elevated levels of carbon

di-oxide lead to global warming, ocean

cir-culation should turn sluggish, in which

case it would carry less dissolved

oxy-gen Oxygen-deprived waters may

con-tribute to the formation of carbon-rich

black shale, as documented by the late

Seymour O Schlanger of Northwestern

University, Michael A Arthur of

Penn-sylvania State University and Hugh C

Jenkyns of the University of Oxford

The atmosphere must have absorbed

an extensive outpouring of carbon

diox-ide around 120 million years ago, when

the tempo of volcanism peaked Larson

speculates that a massive thermal

insta-bility in the D′′layer initiated so-calledsuperplumes that ultimately suppliedthe Ontong Java and Kerguelen oceanicplateaus, along with several other small-

er plateaus in the PaciÞc Ocean neth G Caldeira of Penn State and Ram-pino have run computer models of thegeochemical cycle based on their as-sumptions about the quantity of car-bon dioxide in the atmosphere duringthat era From the models, the workersinfer that worldwide temperatures av-eraged from 7.6 to 12.5 degrees Celsiushigher than todayÕs mean, if one takesinto account the diÝerent geographyand higher sea level at the time

Ken-Where volcanic eruptions occur ondry land, they directly alter the physicsand chemistry of the atmosphere Alan

R HuÝman of Exxon Exploration pany in Houston calculates that a singleßood basalt event that generates 1,000cubic kilometers of lava (the volume of

Com-a typicCom-al ßow in the ColumbiCom-a Riverprovince) emits 16 trillion kilograms ofcarbon dioxide, three trillion kilograms

of sulfur and 30 billion kilograms ofhalogens The thousands of such epi-sodes that must occur in the accumula-tion of an individual large igneous prov-ince would modify the atmosphere inways that would dwarf the eÝects ofmodern, human-generated pollutants

Explosive eruptions of silica-rich rock

often carry sulfurous particles into thestratosphere, where they are converted

to tiny droplets of sulfuric acid, ing to Rampino and Self Basaltic lavasemit about 10 times as much sulfurper unit volume as do silica-rich lavas;Charles B Ỏcer of Dartmouth Collegeand his co-workers deduce that if thegases and particles produced duringbasaltic eruptions were shot into the stratosphere, they could cause short-term plagues of acid rain, worldwidedarkening and global cooling Richard B.Stothers of the National Aeronautics andSpace Administration Goddard SpaceFlight Center, along with several others,hypothesizes that massive lava foun-tains and vigorous atmospheric con-vection taking place over eruptive vents

accord-in contaccord-inental ßood basalts could ply the suÛcient upward momentum

sup-to inject material insup-to the strasup-tosphere

The powerful Lakagigar eruption

in Iceland in 1783Ð84 illustratesthe potential disruptive eÝects ofßood basalt volcanism Although the La-kagigar eruption poured out only about

15 cubic kilometers of lava, local peratures declined noticeably in the fol-lowing years About three quarters of alllivestock in Iceland died, probably theresult both of the deterioration of theclimate and of the emissions of acid gas;

tem-PHYSICAL AND CHEMICAL EFFECTS accompany the

appear-ance of large igneous provinces Volcanoes emit carbon

diox-ide, which can elevate the global temperature Sulfurous

parti-cles and dust in the stratosphere could induce acid rain and

block out sunlight Heavy metals and other chemicals

emit-ted during eruptions would alter the composition of the land,air and water Topographic changes associated with exten-sive volcanism would modify oceanic circulation and changesea level ; heat and hydrothermal activity cause upwelling,further transforming underwater conditions

THERMALVENT

HYDRO-CHANGE INSEA LEVEL

HOT SPOT

ACID RAIN

UPWELLING

HEAT, GAS,PARTICLES ANDCHEMICALS

HOT SPOT

GAS, PARTICLESAND CHEMICALS

SULFUROUS PARTICLES ANDDUST IN STRATOSPHERE

the resulting famine killed about one

fourth of IcelandÕs human population

Dust veils, fog and haze appeared over

most of Europe and adjacent parts of

Asia and Africa for many months after

the eruption

Scientists are uncovering clear

evi-dence that the environmental

conse-quences of ßood basalt volcanism have

in fact contributed to mass extinctions

The most severe extinction in

terrestri-al history occurred 248 million years

ago, when the Siberian Traps formed

At that time, roughly 95 percent of all

marine species perished; the ensuing

evolutionary free-for-all marked the Þrst

appearance of the dinosaurs

The biological repercussions of the

eruption of large igneous provinces

may depend in part on the state of the

global environment at the time When

the environment is already stressed by

other factors, such volcanism may

trig-ger rapid climatic, oceanographic and

biotic changes When the environment

is robust, few eÝects may show up in

the geologic record

Oddly enough, the most voluminous

igneous province, the Ontong Java,

pro-duced virtually no detectable

extinc-tions On the contrary, its formation

coincides with the deposition of black

shale, which is suggestive of an interval

of enhanced biological activity We

pro-pose that the deleterious eÝects

associ-ated with the Ontong Java were

mini-mal in part because it erupted

under-water, as indicated by Ocean Drilling

Program studies led by Loren Kroenke

of the University of Hawaii and

Wolf-gang H Berger of the Scripps

Institu-tion of Oceanography We also suspect

that the global environment was in a

resilient state 120 million years ago,

although such an assertion is diÛcult

to quantify and prove In contrast, the

emplacement of the smaller Kerguelen

and Broken Ridge provinces, which took

place about 110 million years ago,

co-incided with a mass extinction as well

as with major deposits of black shale

The eruptions that built up the

Dec-can Traps, and simultaneous volDec-canism

along the margins of western India and

along the Seychelles and the Mascarene

Plateau, took place approximately 65

million years agoĐjust about the time

that nearly half of all species of life,

in-cluding the dinosaurs, went extinct

Sci-entists continue to debate whether to

attribute those changes to the impact

of a sizable asteroid or to more earthly

explanations Even if an asteroid was

the primary agent of the extinction, the

Deccan eruptions may have

contribut-ed to an environmental deterioration

that could have magniÞed the

reper-cussions of the impact

Another signiÞcant but less

celebrat-ed change in the global environmenttook place 10 million years after thedemise of the dinosaurs, during the em-placement of the North Atlantic volcanicmargins At that time, many deep-oceanforaminifera and land mammals be-came extinct, and hydrothermal activitywas high Moreover, David K Rea andhis co-workers at the University of Mich-igan, along with Ellen Thomas of YaleUniversity, Þnd evidence of major trans-formations in deep water and in atmo-spheric circulation

Analysis of the oxygen isotopes taken

in by foraminifera indicates that oceantemperatures 55 million years ago, dur-ing the early Eocene epoch, were warm-

er than at any other time during thepast 70 million years Eldholm, workingwith Thomas, recently postulated thatthe balmy ocean surface of the early Eo-cene could have resulted from the out-gassing of carbon dioxide during theeruptions along the North Atlantic vol-canic margins Ash layers dating from

55 million years ago cover large regions

of northwestern Europe, aÛrming theviolence of the volcanism Higher atmo-spheric temperatures could have led tothe formation of a layer of warm wateratop the oceans at high latitudes Such

a warm surface layer would tend to sist mixing with underlying, cooler wa-ters The consequent changes in deep-water circulation could have been fatalfor many ocean-bottom species

re-Even the relatively small Columbia

Riv-er ßood basalts coincided with a massextinction 16 million years ago At aboutthat time, the earth began to experience

an ongoing cycle of ice ages, as Maureen

E Raymo of the Massachusetts Institute

of Technology points out Perhaps theglobal environment was already so frailthat even a moderate eruption couldhave had a substantial impact

Obviously, geologists

haveĐliteral-ly and ÞgurativehaveĐliteral-lyĐonhaveĐliteral-ly just scratchedthe surface of large igneous provinces

Current knowledge of these formationsamply demonstrates that they containcrucial information about the internalworkings of the earth and about the

natural causes of global change Earthscientists are now intensifying their ef-forts to produce better seismic images,

to undertake more Þeld and laboratorystudies, to perform additional model-ing and to conduct more scientiÞc drill-ing Those labors promise even betterinsight into the intimate links betweenthe earthÕs inner and outer realms

LAVA FLOW from Kilauea Volcano inHawaii only hints at the magnitude andpotential environmental impact of theßows incorporated into large igneousprovinces Individual sheets of lava inthe Columbia River ßood basalts, for ex-ample, extend for hundreds of kilome-ters and stand several meters thick

FURTHER READING

CONTINENTAL FLOOD BASALTS Edited by

J D Macdougall Kluwer Academic lishers, 1988

Pub-HOTSPOTS, MANTLE PLUMES, FLOOD SALTS, AND TRUE POLAR WANDER R A

BA-Duncan and M A Richards in Reviews of Geophysics, Vol 29, No 1, pages 31Ð50;

February 1991

GEOLOGICAL CONSEQUENCES OF

SUPER-PLUMES Roger L Larson in Geology, Vol.

19, No 10, pages 963Ð966; October 1991.MAGMATISM AND THE CAUSES OF CONTI-NENTAL BREAK-UP Edited by B C Storey,

T Alabaster and R J Pankhurst SpecialPublication, No 68 Geological Society ofLondon, 1992

LARGE IGNEOUS PROVINCES: CRUSTALSTRUCTURE, DIMENSIONS, AND EXTERNAL

CONSEQUENCES M F CoÛn and O

Eld-holm in Reviews of Geophysics ( in press).

Evolutionarily Mobile Modules in Proteins

Many proteins consist of a fairly small set of modular elements.

How these units spread and multiplied during evolution

is not altogether clear, but a pattern may be emerging

by Russell F Doolittle and Peer Bork

Molecular biologists and

bio-chemists have learned in

re-cent decades that many

pro-teins consist of domains, or discrete

blocks of amino acids Many of these

domains have well-deÞned functions

that contribute to the overall activity of

a protein Furthermore, some of these

modular units frequently move about

within and between proteins during

evolution The evolutionary mobility of

these modules is not restricted to hops

within the genetic material of a single

species: in some cases, the modules are

apparently able to travel laterally across

species linesÑeven moving from

ani-mal cells to bacteria, for example

Because a compartmental delineation

has also been found in the regions of

genes that encode proteins, many

biol-ogists have been convinced that these

structural features are reßections of the

same underlying phenomenon They

be-lieve that each genetic coding region

cor-responds to a speciÞc structural feature

in a protein We and our colleagues take

a somewhat diÝerent point of view The

weight of evidence, in our opinion, falls

to the argument that the subdivision ofgenes into separate coding parts is a farmore recent development by evolution-ary standards

Proteins are long chains of small ecules called amino acids Twenty dif-ferent kinds of amino acids, each withits own shape and chemical character,make up all the proteins found in na-ture All the properties of a protein de-pend on which of the 20 amino acidsare used in its construction and, partic-ularly, on the order in which they arestrung together Most notably, the ami-

mol-no acid sequence determines how theprotein will fold up into an active, three-dimensional body

The length of a sequence plays animportant part in that determination

Chains of amino acidsÑusually calledpolypeptidesÑmay be up to many thou-sands of amino acid units long ( Therecord to date is for titin, a muscle pro-tein that contains upward of 30,000amino acid residues.) Short chains, how-ever, are not large enough to have suf-Þcient intramolecular attachments tolock themselves into a single conforma-tion; they tend to ßop from one form toanother Ordinarily, when a polypeptidechain contains 30 or 40 residues, it be-gins to have enough internally cohesiveforces to give it a predominant shape,although it may still need additional sta-bilization from bound metal ions or di-sulÞde bonds between pairs of the ami-

no acid cysteine

In a constant environment, any tein containing more than some mini-mum number of amino acid units willalways fold itself in the same way Thatenvironment may be the dilute salt so-lution that constitutes many biologicalßuids or the greasy conÞnes of a biolog-ical membrane; it can also include near-

pro-by proteins or even other parts of thesame long polypeptide chain

A sequence that folds spontaneously

into a characteristic shape under a Þned set of circumstances is called a do-main, but that formal deÞnition is sel-dom applied with rigor More often thannot, the term is applied to any part of aprotein that can be deÞned as structur-ally distinct from the rest Some smallproteins are completely embodied in asingle domain; many others are com-posed of two or more domains; andsome are made up of many domains,the shapes of which may be very simi-lar or very diÝerent

de-RUSSELL F DOOLITTLE and PEER

BORK analyze the structures of proteins

to gain insights into evolution

Doolit-tle is professor of biology at the Center

for Molecular Genetics of the

Univer-sity of California, San Diego He began

his studies of protein evolution while he

was a graduate student in biochemistry

at Harvard University in the late 1950s

Doolittle is a member of the National

Academy of Sciences and other

scientif-ic associations This artscientif-icle is his third

for ScientiÞc American Bork is a project

leader at the Max DelbrŸck Center for

Molecular Medicine in Berlin and a

visit-ing scientist at the European Molecular

Biology Laboratory in Heidelberg He also

holds a teaching appointment at

Hum-boldt University in Berlin

CD2 DOMAIN 2

The most straightforward way to

iden-tify a domain is to determine its

struc-ture through x-ray diÝraction of protein

crystals or nuclear magnetic resonance

( NMR ) studies Once the amino acid

se-quence of a domain has been

identi-Þed, researchers can Þnd other related

domains without recourse to structural

studies: they can simply look for amino

acid sequences that are similar to those

of familiar structures That shortcut is

extremely helpful because many more

amino acid sequences are known than

are structures from x-ray diÝraction or

NMR Even in the absence of

experimen-tal determinations, it is often possible

to infer the existence of a domain on

the basis of sequence alone By looking

at the structural and sequence

similari-ties of proteins in this way, one canlearn much about their evolution

Until the early 1970s, the

conven-tional wisdom about how teins evolved centered mostly

pro-on Òduplicatipro-on and modiÞcatipro-on.Ó Thegene for a given protein is occasional-

ly duplicated by various recombinationprocesses during which genetic infor-mation is exchanged between strands

of DNA Sometimes the duplication sults in a second gene that can thenundergo further modiÞcation or muta-tion to produce a new protein with anovel function Alternatively, the dupli-cated DNA can be in tandem, in whichcase the original protein is elongatedand may assume novel properties as aresult As comparisons of amino acidsequences make obvious, these kinds ofinternal duplications have clearly givenrise to many extended proteins, rang-ing from small ones, such as the bacte-rial ferredoxins that have only 56 ami-

re-no acid residues, to large ones, such asbacterial beta-galactosidase, which hasmore than 1,000

A hidden aspect of protein evolutioncame to light about 20 years ago, how-ever, when Michael G Rossmann of Pur-due University determined the three-di-mensional structure of the enzyme lac-tate dehydrogenase by x-ray diÝraction

One part of that molecule, he noticed,

closely resembled features of some

oth-er proteins he had seen SpeciÞcally, apart of the enzyme that bound to a co-factor had obvious counterparts in oth-

er dehydrogenases

What made those structural ities noteworthy was that they did notalways occur in the same relative parts

similar-of the molecules It seemed as thoughduring evolution the unit had movedaround within the linear amino acid se-quence without losing its function ofbinding to a cofactor Rossmann sug-gested that proteins were constructed

of modulesÑwhat we would now sider domainsÑthat had appeared ear-

con-ly in lifeÕs history and had been bled in diÝerent combinations.His observation presented a possibili-

assem-ty for protein evolution that greatly

aug-GENETICALLY MOBILE MODULES havebeen found in many proteins Two types

of these modules, or domains, are shownhere The Fn3 and the GHR domains areexamples of Þbronectin type III modules.The PapD and CD2 domains are immu-noglobulin domains These modules arelinear sequences of amino acids that canfold themselves into consistent, recog-nizable structures with speciÞc biochem-ical properties During evolution, thesedomains can move as discrete unitsfrom one protein to another, which helpsnew types of proteins to appear

PapD DOMAIN 1

mented what could be accomplished

purely through duplication and

modiÞ-cation If new proteins could be created

by recombining the components of

oth-er ones, then protein divoth-ersity could

grow explosively

RossmannÕs suggestion proved

pre-scient The amino acid sequences for

numerous large proteins have been

de-termined, and many have the kind of

grossly repetitive structure that one

might expect of a string of mobile

mod-ules For instance, the protein

Þbronec-tin is made up of two long chains, each

of which contains more than 2,000

amino acid residues Casual inspection

revealed that the chains of this large

extracellular protein consist of several

runs of three diÝerent types of

repeat-ed sequences The repeats, which are

referred to as Fn1, Fn2 and Fn3, have

lengths on the order of 45, 60 and 100

amino acids, respectively ( The repeats

are Òimperfect,Ó which means that all

the repeats of a given type are not

ex-actly identical.) Presumably, each type

of repeat can fold up independently as

a true domain, and the overall protein

must be like a long necklace made from

three types of beads

Surprisingly, sequences similar to

Fn1, Fn2 and Fn3 were subsequently

observed in a huge assortment of other

animal proteins Much the same is also

true for a number of other identiÞed

domains The small protein called

epi-dermal growth factor consists of a

sin-gle domain (53 amino acids long in the

human version) that is tightly foldedand pinned together by three disulÞdebonds Similar domains have been iden-tiÞed by their sequences in more than

100 proteins, in which they occur fromone to more than 30 copies

The functions of many of these

modules are not altogether clear,but many of them do bind to orrecognize particular substances There

is a family of lectins that bind to variouscarbohydrates Similarly, the immuno-globulin domain, a feature of antibod-ies and other molecules in the immunesystem, is well known for its binding ca-pabilities Some domains may serve asrecognition tags that identify a protein

as ÒbelongingÓ to a certain tissue Manyappear to be mere linkers or spacers, in-nocuous connecting units Finally, somemay have no function at all It appears,therefore, that many domains can and

do move within and between proteinsduring evolution As long as no damage

or loss of function results from a

shuf-ße, the evolutionary cost of maintaining

a domain in a new setting is cant That result is a corollary of the the-ory of neutral evolution, but it might bephrased as a rule borrowed from profes-sional basketball : Ò No harm, no foul.ÓWhen Rossmann Þrst suggested thatmodular units might move within andbetween proteins, no one pondered veryseriously the genetic mechanisms thatmight account for those rearrangements

insigniÞ-Not long afterward, however, molecular

biologists happened on an unexpectedfeature of genes that seemed to oÝer anexplanation As James Watson and Fran-cis Crick had learned in 1953, genetic information is inscribed in the double helices of deoxyribonucleic acid ( DNA )molecules Just as proteins are strings ofamino acids, DNA molecules are strings

of nucleotide bases The DNA nucleotidesequences are copied, or transcribed, ascomplementary strands of messengerribonucleic acid (RNA) Inside cells, min-iature factories called ribosomes as-semble proteins from the instructions

in the RNA: each codon, or three-basesequence in the RNA, corresponds to

an amino acid

The surprising observation made inthe mid-1970s was that the DNA cod-ing for a polypeptide can be interrupt-

ed by noncoding sequencesÑarrays ofbases that do not correspond to the sequence of amino acids found in theultimate protein product The noncod-ing sequences are excised by a splicingmechanism before the messenger RNAstrand is translated into a polypeptide.The discovery of those interruptions

in genes prompted Walter Gilbert ofHarvard University to suggest that thenoncoding sequences, which he namedintrons (for intervening sequences), fa-cilitated the exchange of the codingparts of the genes, which he termed exons (for expressed sequences) Thethought was that the additional distancebetween coding segments would lead

to proportionately more opportunities

LIKE BEADS IN A NECKLACE , domains appear as distinct

sub-units of modular proteins Some proteins, such as

Þbronec-tin, collagen XII and the muscle protein tiÞbronec-tin, contain manyrepeats of the same few domains

NO KNOWN HOMOLOGIES TOOTHER PROTEINS

RECEPTOR TYROSINE KINASE EPH

TYR-KINIg

TSPOVWA

Fn3Fn2

for recombinations, which depend on

random breaks in the DNA It was

sug-gested that similarities between the

se-quences of introns might promote

mis-alignments and unequal crossovers of

DNA during recombination, which would

make the rearrangement of genes

easi-er Although there was no basis at that

time for supposing that introns would

have similar sequences, subsequent

re-search has shown that introns are a

ha-ven for large numbers of mobile

genet-ic elements The similar sequences in

those mobile elements can contribute

to genetic delinquencies during

meio-sis, the cell division process that gives

rise to eggs and sperm

Of course, many creatures do not

engage in meiosis; have those

organisms missed out on a

splendid way to assemble new proteins?

Introns that interrupt protein coding are

found only in the DNA of eukaryotes,

or-ganisms with discrete nuclei The genes

for bacterial proteins do not contain

in-trons: every set of three bases

corre-sponds to an amino acid in the protein

( A few types of introns that do not

in-terrupt proteins have been found in

bacteria, but they are not properly part

of this discussion.)

The absence of introns from bacterial

protein genes led Ford Doolittle of

Dal-housie University and James Darnell of

the Rockefeller University to suggest

independently that bacteria may have

possessed introns in the past but had

lost them Presumably, their genomes

had been streamlined during evolution

to make their replication more eÛcient

In short, the introns had been around

since the beginning of life; it was the

short coding sequences that were

en-gendered separately

Doolittle and DarnellÕs proposal led to

a still unsettled debate about whether

introns appeared ÒearlyÓ and are

fun-damental to the origin of all proteins

or whether they came Òlate.Ó The latter

proposition was advanced by Thomas

Cavalier-Smith, now at the University of

British Columbia, who theorized that

in-trons might be invasive bits of nucleic

acids, referred to as transposable

ele-ments, that originated in the symbiotic

organisms that eventually became the

mitochondria and other organelles of

eukaryotic cells His idea has been

ex-tended by Donal Hickey of Ottawa

Uni-versity, among others

As it turns out, the stretches of DNA

that encode the evolutionarily mobile

modules in proteins are frequently, but

not always, ßanked by introns In other

words, the structural units in many

pro-teins are encoded by exons That

obser-vation fostered a widespread belief that

all exons are evolutionarily mobile andcorrespond to potential modular build-ing blocks in proteins

In our view, this notion is

mistak-en on two counts One is that, as L‡zl—

Patthy of the Institute of Enzymology

in Budapest Þrst pointed out, all exonscan be shuÜed, but only a fraction ofsuch shuÜed units will be geneticallycompatibleÑthat is, many of them can-not be sensibly translated in some newpositions When an intron falls in a cod-ing sequence, it will occupy one of threetypes of positions: squarely between twocodons (type 0), between the Þrst andsecond positions of a codon (type 1) orbetween the second and third positions(type 2) If that intron and the codingsequence adjacent to it are shufßed into

a new location, the intron must adoptthe same type of positionÑotherwisethe shifted codons will be translated im-properly and a nonsensical amino acidsequence will result If introns were dis-tributed randomly, we would expect toÞnd that after a shuÜe, only one third

of the new exon combinations would be

in phase Curiously, the overwhelmingmajority of the genes for the most fre-quently shuÜed modules are ßanked

by introns that are of type 1

The other fundamental reason whyonly some exons are evolutionarily mo-

bile is that only true domainsÑthosethat can fold completely and indepen-dentlyÑwill be able to survive in newprotein settings Smaller, less self-suÛ-cient sequences would be unable to foldand would lose their identity Moreover,

if a shuÜed unit were to land betweentwo exons that were not themselves truedomains, then the product of the genereceiving the addition might not be able

to fold itself properly either

These two factors, one genetic and the other structural, contribute great-

ly to why mobile domains are so often found in one anotherÕs company Notinfrequently, proteins that contain one mobile domain contain others; some proteins are mosaics that contain as many as Þve diÝerent commonly shuf-ßed modules These types of proteinsare both genetically and structurally tol-erant of the shuÜing process

The observation that many modulesare encoded by exons has been inter-preted as support for the idea that theprimeval organisms assembled all theirproteins from an inventory of exon-en-coded primitive structural components.Several points argue against such an in-terpretation, however For one, simplearithmetic proves that the hypotheticalearly exons would have been too small

to produce protein components that

Sizes of Some Mobile Domains

NUMBER OF CYSTEINES

866444686646

4 or 6

APPROXIMATE NUMBER

OF AMINO ACID UNITS

40404545606168728080

~100110250

18253550100

~100100100

DOMAINS CONTAINING DISULFIDE BONDS

Somatomedin BComplement C9EGF

Fn1Fn2

“Apple” (Sushi) (ßGP26)Ovomucoid

VWF-CKringleKunitzLinkScavenger receptorFibrinogen-related domain

NO DISULFIDE BONDS

CollagenLeucine-rich

“Gla”

Collagen-bindingLectinlikeFn3SH2SH3

Evolutionarily mobile domains vary in size Some hold their shape in partbecause of disulfide bonds between pairs of their cysteine residues.Other domains are stable without such bonds

could fold on their own The average

size of the known exons in vertebrate

genomes today is 135 nucleotide bases,

which corresponds to a polypeptide of a

mere 45 amino acids A sequence that

short usually needs auxiliary

stabiliza-tion to fold into a stable conformastabiliza-tion

Keep in mind, too, that the

propo-nents of the Ò intron earlyÓ theory

con-tend that introns are constantly being

lost over time They have been forced

to that conclusion by the sporadic

occur-rence of introns in diÝerent species This

inconstant pattern could be the result

of either gain or loss of introns, but if

one is wedded to the notion that introns

were there from the beginning, then

the only explanation is loss

According-ly, the earliest exons would have been

even smaller, encoding polypeptides

that could not reasonably be expected

to fold into domains on their own

Another argument against the use of

modern mosaic proteins as examples of

early building blocks concerns the

dis-tribution of domains among proteins

Far and away, the majority of mobile

modules known to date are found

exclu-sively in animal proteins At this point,

we do not really know when or where

most of them arose initially Perhaps the

trail of their evolution has been partly

obscured by extensive sequence

chang-es in the related domains of plants,

fun-gi and protozoa As we shall discuss,

the fact that three-dimensional

struc-tures are more persistent than

sequenc-es in an evolutionary sense may allow

this puzzle to be solved

Besides all the evidence that most

ex-ons are not evolutionarily mobile, some

mobile domains are also clearly not the

products of single exons One large

do-main, Þrst observed in the Þbrinogen

molecule of vertebrate animals, is

com-posed of 250 amino acids In some

pro-teins, the gene for that domain contains

multiple introns Yet none of these

indi-vidual exons has ever been found

with-out all the others None of the exons,

it seems, has ever shuÜed itself out ofthe domain Thus, the mere presence

of introns in a gene is apparently notenough to cause exons to be mobile

That the vast majority of identiÞed ons has never been found in more thanone setting argues against simple andindiscriminate mobility

ex-There are other examples of

mov-able units that contain intronswithin their coding sequences

One of the Þrst movable modules to beidentiÞed is referred to as a kringle (be-cause it resembles a type of Danish pas-try by that name) It is made of about

80 amino acid units and contains threecharacteristic disulÞde bonds It is quitesimilar to the Fn2 domain, diÝering only

in the number of residues between itscysteine residues, and some workers

do not distinguish between the two Insome of its settings, the gene for thekringle is split by an intron, but so far

no one has found half a kringle in anyprotein

A further point in favor of the ÒintronlateÓ theory is that the introns that in-terrupt coding regions are much morecommon in plants and animals thanamong the earliest diverging eukary-otes No introns at all have been found

in primitive eukaryotes, such as Giardia

lamblia Furthermore, modular proteins

have been identiÞed in plants for which

no recognizable counterpart is known

in animals, and vice versa Finally, there

is indirect evidence that the modularassembly of some bacterial proteins oc-curred so recently that they must haveevolved without the aid of introns Allthis evidence suggests that protein-in-terrupting introns appeared after theevolution of eukaryotes

So some exons encode domains, butmost exons do not; those that do can of-ten be genetically duplicated and shuf-ßed The issue of cause and eÝect be-

tween these phenomena raises a thornyproblem Perhaps the evolution of theseintrons did facilitate exon shuÜing Onthe other hand, it is not impossible thatthese introns often delineate the cod-ing regions for domains because thatplacement is advantageous for intronpropagation If an intron interrupted asequence coding for a domain, it mightsurvive in that situation (if it did not vi-olate the phase rule noted earlier) Itwould not spread further, however, be-cause the exons on its boundaries couldnot stand alone and thus could not bemoved independently Conversely, if theintron lands between regions codingfor independently folding units, it canspread to other locations along with theexons So exon shuÜing may be onlyincidental to the survival of introns.One way to learn more about the evo-lution of movable modules is to look

at the structure and dispersal of one inparticular Our favorite is Fn3, the Þbro-nectin type III domain Like kringles,Fn3 units are sometimes split by sin-gle introns, but they have never been observed with fewer than their full complement of 90 to 100 amino acids The two of us independently followed the discovery of Fn3 in various proteinsfor several years In early reports the sequence was turning up only in ani-mal proteins, so we were both surprisedwhen, in 1990, workers at Niigata Uni-versity in Japan reported the existence

of an Fn3 domain in a bacterial protein.Our common interest came to light at ameeting in Italy in 1991, whereupon wedecided to join forces in a comprehen-sive inventory of Fn3 occurrences

To that end, we screened a data base

of protein sequences by various means,including a pattern-searching algorithmthat one of us (Bork ) had devised withChristian Grunwald when both were atthe Central Institute for Molecular Biolo-

gy in East Berlin We uncovered well over

300 unique occurrences of the Fn3

se-INTERRUPTED STRUCTURE of some genes that encode

pro-teins divides them into exons and introns (expressed and

in-tervening sequences) Triplets of bases in exons are

translat-ed as amino acids in proteins Introns can interrupt coding

sequences in three diÝerent positions If exons and intronsmoved randomly, transpositions between diÝerent types of in-trons would cause frameshift mutations That problem countsagainst the idea that all exons encode movable modules

C C C G G G G A A A C C C

TRANSPOSITION

OF EXONS

quence motif, which established that it

was indeed a true domain, should there

have been any doubt The 300

occur-rences actually represented 67 diÝerent

proteins, not counting the same proteins

from diÝerent species Sixty of these

were from animals and seven from

bac-teria None of the sequences identiÞed

were from plants, fungi or unicellular

eukaryotes

The obvious questions to be asked

were: Did bacteria and animals both

in-herit the domain from some mutual

an-cestor, or did one of the groups

some-how acquire it from the other? If the

domain was present in the common

an-cestor of prokaryotes and eukaryotes,

then why has it not been found in

fun-gi and plants? With the aid of a

com-puter, we aligned all the Fn3 sequences

that we could Þnd and constructed a

crude phylogenetic tree based on their

similarities Because trying to

accom-plish that with all 300 sequences was

cumbersome, we began by using

repre-sentative sets of sequences from all the

bacterial proteins and from only the

most diÝerent of the animal ones

It was soon clear that something was

amiss The bacterial sequences were

simply too similar to the animal ones to

have descended from a common

ances-tor two billion years ago Instead the

evi-denceÑincluding the

computer-generat-ed phylogenetic treesÑfavorcomputer-generat-ed the view

that somehow the bacteria had acquired

the Fn3 domain from an animal source

There were several reasons to think

so First, it was often the case that an

enzyme from one bacterial species

car-ried the domain but that the same

en-zyme from other species did not, which

implied that the domain was

structural-ly and functionalstructural-ly expendable It must

therefore have been a late addition to

speciÞc bacteria Moreover, the Fn3

se-quences appeared sporadically and in

clusters but always in a characteristic

set of extracellular enzymes; if bacteria

were losing copies of Fn3 over time, one

might expect to Þnd surviving copies in

a more varied set of proteins

Lastly, although the bacteria that had

Fn3 domains were of diverse types, they

did have certain features in common

All are soil bacteria that obtain their

food from commonly found polymers,

such as cellulose and chitin, released

during decomposition Vast numbers of

other types of bacteria have been

ex-amined, but none of them have the Fn3

domain in any of their proteins More

than half the genome sequence of the

intestinal bacterium Escherichia coli is

known, and not one hint of an Fn3

se-quence has surfaced The same can be

said for the large number of studied

fungal and plant sequences If the Fn3

domain had been present in a commonancestor of prokaryotes and eukaryotes,

we would have expected its radiation tohave continued along the major lines

of descent and to be represented in allthese groups

The idea that the gene for a domain

can migrate between distantly lated organisms may seem out-landish at Þrst Common experienceshows that genes are transmitted verti-cally, from one generation of organisms

re-to the next It is nonetheless sometimespossible for genes to be transmitted hor-izontally as well, not only between spe-cies but across distant lineages Some vi-ruses can pick up small genes from onehost and carry them to another host ;

in rare instances, the transposed gene

may assimilate itself into the new hostÕsDNA Bacteria can be transformed bypicking up DNA from their surround-ings, which may include decaying ani-mal cells Many bacteria also carry genes

on small rings of DNA, or plasmids,that they can exchange with other bac-teria In theory, all these mechanismsrepresent opportunities for the hori-zontal transmission of genes

Allowing that some bacterium did tain the gene for an Fn3 domain from

ob-an ob-animal source, when did it happen?All that the phylogenetic tree can indi-cate is that it happened within the pastbillion years, after the divergence ofanimals from plants and fungi To pindown the date, we need to know theaverage rates of sequence change alongboth the bacterial and animal lines of

EVOLUTIONARY RELATIONS of 39 Fn3 domains from various sources appear inthis phylogenetic tree Using a computer, the authors created the tree by compar-ing the sequences of the domains from animals and bacteria Surprisingly, the bac-terial domains are closely related to some from animals, which suggests that bac-teria obtained the genes for those domains from animals

TIR3MKC1TWE4TWD1TWD2PCC2NCHUIBH2DPA1

CLCFCHB2CHB1AME3GLC2GLC3GLC1AML2AML1AME2

AME1

PXE1LRH1

TNH1FNH10UNH1UNH2FNH9FNH1TNH7TNH2

45TWEPH1AXH1CDH1CMM5I2M1I7H1GRBFKLH2

divergence For the animal proteins, we

can estimate the rate by comparing

se-quences from various creatures whose

divergence times are shown in the

fos-sil record Unfortunately, we do not have

comparable information about the

bac-terial sequences ( Whereas some

micro-fossils corresponding to bacteria have

been reported, there is certainly not yet

an interpretable evolutionary hierarchy

for them as there is for animals.) We

can see, however, in both the animal

and bacterial proteins a tendency for

tandem duplication of Fn3 sequencesÑ

that is, within any one protein

contain-ing more than one Fn3 domain, the

sequences of those domains are often

adjacent and usually very similar That

observation implies that the duplication

of the DNA for the Fn3 domain must

be relatively recent

The timing of the horizontal

trans-mission and of the genetic duplications

is critical to our understanding of how

these genetic units spread So far as is

known, contemporary bacteria do not

have introns in their protein-coding

gene sequences If they ever did have

introns in their coding sequences, how

long has it been since those introns

dis-appeared? Unless it was fairly

recent-ly, the genes for Fn3 must have been

spreading without the aid of introns

One likely possibility is that the Fn3

domain is being spread among soil

bac-teria by a promiscuous bacteriophage

( bacterial virus) or plasmid We hope

eventually to catch the act of sion in progress by Þnding a phage thatcarries a gene for an Fn3 domain Nowthat a number of bacterial Fn3 sequenc-

transmis-es have been found, it may be possiblefor us to synthesize short tags of DNAthat will bind to the Fn3 genetic units

Used in conjunction with the DNA pliÞcation technique called the poly-merase chain reaction, those tags couldhelp identify the genes in bacteriophag-

am-es or other vectors

Of course, where the Fn3 domainoriginated is still a mystery Did it Þrstappear in animals? Or are we just un-able to identify its ancestral forms by se-quence comparison? Three-dimensionaldeterminations show that the Fn3 struc-ture is suspiciously similar to immuno-globulin domains Three-dimensionalanalyses by x-ray crystallography andNMR have made it possible to trace theimmunoglobulin domain back to pro-teins in prokaryotes, including PapD,

a ÒchaperoneÓ that helps other proteinsfold, and also to a bacterial enzyme thatdigests cellulose It is interesting thatthe immunoglobulin domain, as origi-nally deÞned, contains a disulÞde bondthat pins its two sides together, but themore primitive forms, some of whichpersist in vertebrate animals today, lackthat bond It is these primitive formsthat are most similar to the Fn3 domain

We can anticipate that other

instanc-es of pirated modulinstanc-es will be uncovered

By our census, the Fn3 domain occurs in

about one out of every 50 animal teins (50 out of 2,500 known animalsequences, independent of species re-dundancies) We estimate that about 25modules are very commonly strewnthroughout animal proteins, much asFn3 is More than 100 others are found

pro-in more than one settpro-ing but less quently than the Þrst group Tracing thepedigree and diaspora of these modularunits is a major challenge that shouldreveal much about every aspect of theevolution of all living things

fre-FURTHER READINGTHE GENEALOGY OF SOME RECENTLY

EVOLVED VERTEBRATE PROTEINS

Rus-sell F Doolittle in Trends in Biochemical Sciences, Vol 10, No 6, pages 233Ð237;

June 1985

PROTEIN MODULES Martin Baron, David

G Norman and Iain D Campbell in

Trends in Biochemical Sciences, Vol 16,

No 1, pages 13Ð17; January 1991

MODULAR EXCHANGE PRINCIPLE IN

PRO-TEINS L‡zl— Patthy in Current Opinion

in Structural Biology, Vol 1, No 3,

pag-es 351Ð361; June 1991

MOBILE MODULES AND MOTIFS Peer Bork

in Current Opinion in Structural Biology,

Vol 2, No 3, pages 413Ð421; June 1992.PROPOSED ACQUISITION OF AN ANIMAL

PROTEIN DOMAIN BY BACTERIA Peer

Bork and Russell F Doolittle in ings of the National Academy of Scienc-

Proceed-es, Vol 89, No 19, pages 8990Ð8994;

October 1, 1992

HORIZONTAL TRANSMISSION OF GENES could explain how

bacteria obtained domains from animals Genes pass vertically

between generations of cells Bacteria can also be transformed

by the absorption of DNA from their environment If a terium picked up the DNA for a domain from a dead cell , itmight have transmitted the gene to its progeny

bac-VERTICAL TRANSMISSION

DAUGHTER

CELLS

DEADCELL

HORIZONTAL TRANSMISSION

DAUGHTERBACTERIA

DAUGHTER BACTERIAWITH ANIMAL GENE

TRANSFORMEDBACTERIUM

FREE FRAGMENTS OF DNA

BACTERIUMCELL GENE