scientific american - 1994 12 - nasa's latest view of the earth

Smith, Baylor College of Medicine left , Peter Samek right 86Ð87 Michael Crawford 88Ð92 Michael Goodman 95 Peter Charlesworth/SABA 97 Lisa Burnett left , Associated Press right 109 IB

DECEMBER 1994

$3.95

Sleek and eÛcient, new cars will have fuel-saving features both inside and out.

Fossil hunters in the Gobi.

Future medicines made of DNA.

December 1994 Volume 271 Number 6

52

60

70

76

Improving Automotive EÛciency

John DeCicco and Marc Ross

Fossils of the Flaming CliÝs

Michael J Novacek, Mark Norell, Malcolm C McKenna and James Clark

Berthold-Georg Englert, Marlan O Scully and Herbert Walther

The New Genetic Medicines

Jack S Cohen and Michael E Hogan

The internal-combustion engine is likely to remain the most practical power sourcefor cars and trucks for decades to come Fortunately, modern engineering can stillsigniÞcantly raise the fuel economy of cars without compromising their perfor-mance ModiÞcations in the design of the automobile oÝer substantial savings forcar owners, less dependence on oil and reduced greenhouse gas emissions

In an almost uncharted region of the Gobi Desert, the eerily preserved skeletons ofdinosaurs and other prehistoric beasts lie only half-buried beneath the wind-scarred soil and eroding sandstone cliÝs A team of paleontologistsÑthe Þrst West-erners allowed to visit the area in more than 60 yearsÑdescribes recent discoveriesfrom a series of extraordinarily rich sites

Radar images of the earth, collected from orbit by the space shuttle Endeavour,

re-veal our planet with startling clarity Volcanoes, meteor craters, rain forests andeven a lost city in the Arabian peninsula stand exposed in a new light

A new age in the treatment of diseases may be upon us, these biotechnologists gue ArtiÞcial strings of nucleic acids can pair with RNA or wind around the doublehelix of DNA and in eÝect silence the genes responsible for many illnesses Earlyexperiments, including preliminary clinical trials, are already proving the worth ofsome of these ÒantisenseÓ and Òtriplex DNAÓ strategies

ar-Quantum physics says that electrons, photons and other microscopic objects aresimultaneously waves and particles but that both sets of features cannot be seen atthe same time Many physicists assumed this limitation resulted from the impossi-bility of measuring those properties perfectly Not so: even when those uncertain-ties disappear, the principle behind the duality persists

4

S CIENCE IN PICTURES

Earth from Sky

Diane L Evans, Ellen R Stofan, Thomas D Jones and Linda M Godwin

Copyright 1994 Scientific American, Inc.

100

106

CaulerpaWilliam P Jacobs

Letters to the Editors

More on the origin of the moon Thinking about consciousness.Science and the Citizen

Book Reviews

Readings for children on space,whales, food and more

Essay:Eric J Chaisson

What NASA could learn from the Òdark siders.Ó

The Amateur Scientist

Measuring the friction that hurtsyour carÕs eÛciency

T RENDS IN SCIENTIFIC COMMUNICATION

The Speed of Write

Gary Stix, staÝ writer

Making Environmental Treaties Work

Hilary F French

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 Canada Post International Publications Mail (Canadian Distribution) Sales Agreement No 242764 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; fax: (212) 355-0408 or send E-mail to SCAinquiry@aol.com.

More than 170 international treaties theoretically protect the environment, butmost are too vague or toothless Forging treaties that are both more stringent andwidely acceptable is possible, nonetheless Several novel approaches to negotiationand monitoring show promise for enlisting the compliance of recalcitrant nations

Three feet long and trailing fernlike leaves, this tropical algal plant looks like an

or-dinary clump of seaweed but is actually a single gigantic cell As such, Caulerpa

contradicts the biological tenet that organisms must be multicellular to have greatsize and a complex specialized form

Growing numbers of scientists are abandoning slow, costly printed journals in vor of the Internet Globally linked computers can disseminate research reports in

fa-a ßfa-ash fa-and even fa-allow investigfa-ators to collfa-aborfa-ate or kibitz on experiments whilecontinents apart Now computer scientists, librarians and traditional publishers arescrambling to maintain order and quality in the archives of cyberspace

Breast cancer gene Lake Baikal: asuccess story Uncertaintiesabout hantavirus Lightningabove the clouds A laser lock onliquid helium Who names theheavens? Ig Nobel success

The Analytical Economist HaitiÕs voodoo economics

Technology and BusinessDropping a net on bad cops

Virtual reality: Is anything reallythere? Replicating holy relics

Entrepreneurs hit legal potholes

on the information highway

ProÞleCynthia MossÑher quarter century

of living among elephants Annual Index 1994

William Amarel (bottom

left ) and Amy Davidson

(bottom left ), Ed Heck/

AMNH (bottom right )

70Ð75 Jet Propulsion Laboratory,

California Institute of

Technology and National

Aeronautics and Space

Administration

78Ð79 Tomo Narashima

80 Jared Schneidman/JSD

81 Sean R Smith, Baylor

College of Medicine (left ),

Peter Samek (right )

86Ð87 Michael Crawford

88Ð92 Michael Goodman

95 Peter Charlesworth/SABA

97 Lisa Burnett (left ),

Associated Press (right )

109 IBM Corporation, Research

Division, Almaden ResearchCenter

110 Steve Northup

111 Jared Schneidman/JSD

112Ð115 Kathy Konkle

THE ILLUSTRATIONS

Cover painting by George Retseck

6 SCIENTIFIC AMERICAN December 1994

THE COVER painting of a late-model mobile emphasizes the importance ofsmooth Þt and Þnish to enhancing the eÛ-ciency of cars New painting and laser-weld-ing techniques, as well as gently roundedcorners and a low front end, help to de-crease aerodynamic drag Minimizing otherenergy losses, including those from braking,tire friction and accessories, provides avaluable and often overlooked approach toraising fuel economy (see ỊImproving Auto-motive Ẻciency,Ĩ by John DeCicco andMarc Ross, page 52 )

¨

Established 1845

ACTING EDITOR: Michelle Press

BOARD OF EDITORS: John Rennie, Associate tor; Marguerite Holloway, News Editor; Timothy

Edi-M Beardsley ; W Wayt Gibbs; John Horgan , nior Writer; Kristin Leutwyler; Philip Morrison , Book Editor; Madhusree Mukerjee; Sasha Neme-

Se-cek; Corey S Powell ; 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; Johnny Johnson, Assistant Art Director, Graphics Systems; Nisa Geller, Pho- tography Editor; Lisa Burnett, Production Editor COPY: Maria-Christina Keller, Copy Chief; Nancy

L Freireich; Molly K Frances; Daniel C SchlenoÝ

PRODUCTION: Richard Sasso, Vice President, Production ; William Sherman, Production Man- ager; Managers: Carol Albert, Print Production ; Janet Cermak , Makeup & Quality Control ; Tanya DeSilva , Prepress; Silvia Di Placido, Spe- cial Projects; Carol Hansen , Composition; Mad- elyn Keyes, Systems; Ad TraÛc: Carl Cherebin;

Carey S Ballard; Kelly Ann Mercado

CIRCULATION: Lorraine Leib Terlecki, Associate Publisher/Circulation Director; Katherine Robold, Circulation Manager; Joanne Guralnick , Circula- tion Promotion Manager; Rosa Davis, FulÞllment Manager

ADVERTISING: Kate Dobson, Associate er/Advertising Director OFFICES: NEW YORK:

Publish-Meryle Lowenthal, New York Advertising ager; Randy James, Rick Johnson, Elizabeth

Michi-gan Ave., Chicago, IL 60601; Patrick Bachler, vertising Manager DETROIT: 3000 Town Cen- ter, Suite 1435, SouthÞeld, MI 48075; Edward

S Sepulveda Blvd., Suite 212, Los Angeles, CA

90025; Lisa K Carden, Advertising Manager;

Tonia Wendt 235 Montgomery St., Suite 724, San Francisco, CA 94104; Debra Silver CANADA: Fenn Company, Inc DALLAS: GriÛth Group

MARKETING SERVICES: Laura Salant, Marketing Director ; Diane Schube, Promotion Manager; Susan Spirakis, Research Manager; Nancy Mon- gelli, Assistant Marketing Manager; Ethel D Little, Advertising Coordinator

INTERNATIONAL: EUROPE: Roy Edwards, national Advertising Manager, London; Vivienne

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

Par-Barth David Schwartz, Director, Special ects, Amsterdam SEOUL: Biscom, Inc TOKYO: Nikkei International Ltd.; TAIPEI: Jennifer Wu,

Proj-JR International Ltd.

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

SCIENTIFIC AMERICAN, INC

415 Madison Avenue, New York, NY 10017-1111

(212) 754-0550 CHAIRMAN AND CHIEF EXECUTIVE OFFICER: John J Hanley

CO-CHAIRMAN: Dr Pierre Gerckens

DIRECTOR, ELECTRONIC PUBLISHING: Martin Paul

CORPORATE OFFICERS: President, John J Moeling, Jr.; Chief Financial Ỏcer, R Vincent Barger ; Vice President, Robert L Biewen

PRINTED IN U.S.A

Copyright 1994 Scientific American, Inc.

LETTERS TO THE EDITORS

A Misbegotten Moon

I read ÒThe ScientiÞc Legacy of

Apol-lo,Ó by G JeÝrey Taylor [SCIENTIFIC

AMERICAN, July], with increasing

disbe-lief To consider the moon as the result

of an interplanetary liaison in the

for-mative phase of the earth would only

seem valid if we ignore the satellites in

the remainder of the solar system Are

we to believe that 16 interplanetary

col-lisions resulted in the moons of Jupiter

and that 21 caused the formation of the

satellites of Saturn? Neptune and

Ura-nus also have a large number of moons;

Mercury and Venus are the only

plan-ets not to have at least one The odds

of this number of moons developing

af-ter random hits are minute

If God doesnÕt play dice with the

uni-verse, then itÕs unlikely he would

toler-ate billiards

DIGBY QUESTED

Epsom, England

TaylorÕs excellent article raised as

many questions as it answered He cites

the slow rotation of Venus as evidence

of the low spin acquired by accretion

but conveniently leaves out that of

Mars, with its 24.6-hour period Neither

Mars nor Venus has a sizable moon

Un-less convincing evidence is produced

as to how Mars acquired its high rate

of rotation, the theory that the moon is

an outcome of a collision between the

earth and a Mars-size body remains very

much a conjecture

M H KUBBA

Steinhausen, Switzerland

Taylor replies:

No one really claims that all

solar-system satellites formed in the same

manner The moons that make up

min-iature solar systems around the giant

outer planets almost certainly formed

in fundamentally diÝerent ways than

the earthÕs moon did The gas-giant

planets and their major satellites

prob-ably formed in ways somewhat

analo-gous to that of the solar system as a

whole Furthermore, the huge

gravita-tional pull of the gas giants most likely

captured any debris that would have

been lifted by impacts

The most intriguing problem is why

Mars rotates as fast as it does (almost

a 24-hour day) yet has only two tiny

moons The giant-impact hypothesis gues that the earthÕs rotation is mostlyattributable to the giant impact thatmade the moon Perhaps giant impacts

ar-on smaller bodies ( Mars has ar-only 10percent of the earthÕs mass) completelydisrupted them, leading to the re-accre-tion of a single larger body rather than

to a shaken but intact target body rounded by orbiting raw materials fromwhich its satellite formed Alternatively,the total energy of collision with the pro-to-Mars body may not have been suÛ-cient to cause large amounts of materi-

sur-al to reach orbit Perhaps only scrapsmade it into orbit and are represented

by Phobos and Deimos, the two littlesatellites orbiting Mars Then, too, per-haps we do not fully understand howplanets accrete or how moons form

Raising Consciousness

I enjoyed reading ÒCan Science plain Consciousness?Ó by John Horgan[SCIENTIFIC AMERICAN, July], but youwent overboard in your enthusiasm forÒÞrstsÓ in the Þeld The exhortations byFrancis Crick and Christof Koch toward

Ex-a scientiÞc Ex-assEx-ault on Ex-awEx-areness hEx-avecertainly been beneÞcial to advancingthe Þeld In my less prominent position,

I in fact initiated such scientiÞc mentation in the late 1950s, with the

experi-Þrst major papers out in 1964 ( Journal

of Neurophysiology) and 1965 tives in Biology and Medicine).

(Perspec-The statement that the Society forNeuroscience would host its Þrst sym-posium on consciousness in November

1994 is incorrect I organized andchaired a symposium on ÒCerebral Pro-cesses and Conscious FunctionsÓ held

at the 1985 annual meeting of the ety Also, when Robert W Doty of theUniversity of Rochester was the presi-dent of the society, he held a sympo-sium on ÒConsciousness from NeuronsÓ

soci-at the annual meeting in 1976

BENJAMIN LIBETDepartment of PhysiologyUniversity of California, San FranciscoWhen Koch cautions mysterians byquoting Wittgenstein about thingsÒwhereof one cannot speak,Ó he is onthe right track Another great philoso-pher, Mark Twain, spoke of the samecategory of endeavor as that of Colin

McGinn and David J Chalmers in hisspeech ÒThe Science of OnanismÓ: ÒAs

an amusement it is too ßeeting As anoccupation it is too wearing As a pub-lic exhibition, there is no money in it.Ó

BOB FOSTERTucson, Ariz

Horgan characterized my New

York-er article as raising the possibility that

Gerald M Edelman would win a secondNobel Prize for his work on conscious-ness Actually, the speculation concern-ing a return trip to Sweden centers onEdelmanÕs role in the discovery of cel-lular adhesion molecules

STEVEN LEVYOtis, Mass

Further Fabre

As one of the scientists inspired at

an early age by FabreÕs writings, I wasdelighted to read ÒJean Henri Fabre,Ó

by Georges Pasteur [SCIENTIFIC CAN, July] I was disappointed, however,that the ÒFurther ReadingÓ did not listany of FabreÕs work that has been trans-lated into EnglishÑit would be nice ifthose of us who teach young peoplecould leave a copy lying where some-one might pick it up! Fortunately, there

AMERI-is a nice edition, The Insect World of

J Henri Fabre, edited by Edwin Way

Teale, available in paperback from con Press in Boston

Bea-J E HOLMESPortland, Ore

Letters selected for publication may

be edited for length and clarity licited manuscripts and correspondence will not be returned or acknowledged unless accompanied by a stamped, self- addressed envelope.

Unso-ERRATA

In the timeline illustration on page 48

of ÒLife in the UniverseÓ [October], thedate for Robert HookeÕs microscopeshould be 1665 Also, the vertical scales

on the second and third charts on page

88 of ÒSoftwareÕs Chronic CrisisÓ tember ] should begin at Þve, not at zero

[Sep-10 SCIENTIFIC AMERICAN December 1994

50 AND 100 YEARS AGO

DECEMBER 1944

ÒApproximately $25,000,000 has been

invested in television research and

de-velopment by the radio industry to get

television ready for the public,

accord-ing to James H Carmine, of Philco

Cor-poration ÔAs soon as television

receiv-ers can be made and sold, the public will

eagerly buy them in tremendous

quan-tities,Õ Mr Carmine says.Ó

ÒA new synthetic foam rubber, as soft

and ßuffy as an angel food cake, has

been announced by The Firestone Tire

and Rubber Company Whipped into a

creamy froth, much as a housewife beats

egg whites for her cake, the synthetic

latex traps innumerable interconnected

tiny air bubbles, which give the foam

rubber its softness and permit free

cir-culation of cooling air.Ó

ÒStandard textile machinery adapted

to handling glass textiles is now

allow-ing continuous glass Þlament

and staple Þbers to be twisted,

plied, and woven The Þneness

and strength of the latest glass

Þbers are almost incredible

Fibers with a diameter of

23/100,000 of an inch have a

tensile strength of more than

250,000 pounds per square

inch Experimental Þbers have

been produced with a diameter

of 2/100,000 of an inch and

with a tensile strength

exceed-ing 2,000,000 pounds per

square inch.Ó

ÒDevelopment of a precision

x-ray tube that operates at two

million volts makes it practical

for the Þrst time to inspect by

x-rays exceedingly thick

sec-tions of metal Physicians will

likewise welcome the new tube

as a more effective tool for

re-search in cancer therapy.Ó

ÒIn heavy industry, the main

objection to female labor was

the lack of physical strength for

lifting heavy parts into and out

of machines This was overcome

through the installation of

me-chanical lifting devices such as

hand or electric hoists, or by

overhead traveling cranes It was

shown that once women were

relieved of the physical exertion, theyactually liked machine-tool operationsbetter than did the men.Ó

ÒMore than 16 types of wood go intothe building of the giant Douglas C-54

The woods range from featherweightrattan to heavy mahogany About 30percent more wood is used today inaircraft than just a year ago, largely be-cause of the metal shortage.Ó

DECEMBER 1894ÒOne lady, of whom we read not longago as having reached the age 120 orthereabout, maintained that single bles-sedness is the real elixir vitae She as-cribed the death of a brother at the ten-der age of ninety to the fact that he hadcommitted matrimony in early life.Ó

ÒInvestigations have been

undertak-en to determine the speciÞc action of

a considerable lowering of temperatureupon the brilliancy of bodies whichshine in the dark after having been ex-posed to sunlight Apparently, the pro-duction of phosphorescent light requires

a certain movement of the constituentmolecules of bodies When these arefrozen, the luminous waves are not pro-duced and the phosphorescence disap-pears accordingly.Ó

ÒDonations to the Society of the NewYork Hospital amount to a minor frac-tion of its total income, so that the re-freshing spectacle of a great charityrun on strictly business principles ispresented in perfection by the societyÕsadministration.Ó

ÒBefore the Society of Amateur tographers a few days ago Mr Frederick

Pho-E Ives, of Philadelphia, exhibited hisnew triple-colored lantern slide on thescreen Specimen slides shown

of landscapes had the sky tooblue But several ßower and fruitpictures appeared so accuratelythat one could imagine theywere solid enough to be picked

up or plucked.ÓÒMany persons weigh them-selves frequently and imaginethat they know their weight.Sweet illusion! Nothing is morediÛcult than to know oneÕsweight exactly, even with access

to Þrst class scales For adults,though, it is good to consult thescales, for they are the barome-ter of health Any sudden in-crease of weight, amounting to

a pound or so in a day, cates a tendency to disease.ÓÒOne of the most curioussights recently seen is calledthe EiÝel Tower Bicycle Thismachine is constructed on thesame principle as an ordinaryone, but has a frame which car-ries the rider at a distance ofsome ten feet from terra Þrma.The adventurous spirit whorides this remarkable wheel isusually accompanied by a num-ber of companions who preventvehicles and pedestrians fromobstructing the way.Ó

indi-The EiÝel Tower Bicycle

Copyright 1994 Scientific American, Inc.

For decades, Lake Baikal has

sym-bolized the threat of economic

de-velopment to RussiaÕs wilderness

The largest, oldest, deepest container

of freshwater on the planet, Baikal has

engaged the passions of SiberiaÕs poets

and the intellect of its scientists, who

have charted its increasing burden of

pollution Now there is reason to hope

that Baikal will come to symbolize

some-thing else: rational compromise between

the economic needs of a people and

the ecological needs of their land

With only a modicum of support

from the U.S and the United Nations, a

team of American scientists and

envi-ronmental advocates has persuaded

Russia and Mongolia to develop

sus-tainable land-use programs for the

Baikal watershed The ambitious plans

aim to save the lake and to propel the

region toward a free-market economy

Although their success is not assured,

the Baikal agreements are already

serv-ing as a model elsewhere In November,

China and Russia began drafting a

sim-ilar plan for the Ussuri River basin The

Altai Republic in Siberia has also agreed

to work on an Ịecological-economic

zone.Ĩ If adopted and enforced, theseprograms will protect a combined areamore than twice the size of California

The projects share a common proach and leader, George D Davis,president of Ecologically Sustainable De-velopment Davis has adapted a strate-

ap-gy he used successfully two decadesago to protect the six-million-acre NewYork State Adirondack Park, for which

he was chief planner Inspired by ing laws that cities use to segregate in-dustrial from residential areas, Davis or-

zon-dered a scientiÞc survey to determinethe carrying capacities of the parkÕs re-sources He then drew up a zoning mapand rules restricting where and how for-estry, farming and construction are al-lowedĐeven in the 58 percent of the re-serve that is privately owned The resultwas the Þrst U.S regional land-use plan

After winning a MacArthur tion grant in 1989, Davis was invited toapply his method to the Baikal water-shed, a 150-million-acre region encom-passing parts of Mongolia and threeprovinces of Russia With foundationfunding, Davis joined forces with 30American and Russian scientists

Founda-Through sometimes heated debateand many public hearingsĐamong theÞrst ever held in Siberia, Davis notesĐthe team forged a consensus It gerry-mandered the watershed into 25 diÝer-ent kinds of zones, ranging from farm-land to industrial parks Each zone hasbeen assigned ỊpreferredĨ and Ịcondi-tionalĨ uses; the latter require permits.Anything unspeciÞed is forbidden Morethan 52 million acres, including the lakeitself, have been set aside as nationalparks, scientiÞc reserves, landscapes,scenic rivers, greenbelts and landmarks.For Baikal, protection arrives none toosoon More than a mile deep and ßushwith oxygen, the lake is home to some1,800 species found nowhere else.Its 5,330 cubic miles of drinkablewater are as pure as rainĐwhich

is unfortunate, because the rainover Baikal has turned acidic, con-taminated by the smokestacks ofIrkutsk to the west Many morepollutants pour in from the Selen-

ga River ỊBoat captains will not gowithin a mile of the Selenga delta,because the pollution is so thick,Ĩreports Gary A Cook, director ofBaikal Watch at the Earth IslandInstitute in San Francisco.The threats to Baikal have di-minished noticeably as Russianindustry has ground to a halt,Cook reports And at least twoprovincial governments are be-ginning to act on the plan But asministers sell oÝ defunct state-owned factories and farms, andnew owners convert them, it isunclear whether the Baikal planswill be enforced Last year, afterthe Buryat Republic and ChitaOblast adopted the zoning strategy aspolicy, Russian president Boris Yeltsinsigned a decree creatingĐbut not fund-ingĐa commission to carry it out

So far, says Sergei G Shapkhaev, rector of the 15-person commission,Ịwe have encountered no organizedopposition The most serious problemseems to be that the actual mechanisms

di-of enforcing the laws in court are not inplace.Ĩ A special court that allows citi-zens to sue polluters is now operating,Shapkhaev reports, but the republic hasnot found money to provide any legalassistance to the public

Davis conÞdently predicts that since

SCIENCE AND THE CITIZEN

No-Polluting Zone

Russia follows Adirondack approach to environmental protection

LAKE BAIKAL, with one Þfth of the planetÕs freshwater, may be saved by zoning.

This past March a team of

scien-tists poring over images from the

Galileo spacecraft made a

remark-able discovery The asteroid IdaĐa

chunk of rock just 50 kilometers across,

orbiting between Mars and JupiterĐhas

a tiny moon For the astronomical

com-munity, the Þnding raised big questions

about the origin of Ida and its satellite

For the Galileo researchers, it posed a

more pressing problem: What shouldthe moon be called?

The task is more diÛcult than onemight suppose Naming planets in thesolar system has proved easy because

only three have been discovered in ern times Comets turn up more fre-quently but beneÞt from a well-estab-lished convention: each bears the name

mod-or names of the astronomers who

spot-ted it For asteroids, however, the cess is rather chaotic Hundreds arefound annually, and the discoverer hasfairly free rein in picking the name Inaddition, robotic spacecraft have sentback images of most of the major bod-ies in the solar system, unleashing aßood of unnamed surface features

pro-To keep a little order in the Wild West

of celestial nomenclature, scientistsfounded the International Astro-nomical Union (IAU), which laysdown the law Features on planetscannot bear the name of a livingpersonĐa restriction that does notapply to asteroids Names of politi-cal and religious Þgures of the past

200 years are a no-no: too versial And planetary satellites andasteroids cannot share names ỊOh,yes, names get rejected,Ĩ says Brian

contro-G Marsden of the Smithsonian trophysical Observatory, who over-sees much of the naming of the solar systemÕs minor players Some-times the names are too silly; some-times they just run contrary to hissensibility ỊI objected to calling afeature on Venus Elizabeth Tudor.Nobody calls her Elizabeth Tudor;sheÕs Queen Elizabeth I.Ĩ

As-ỊIdaĨ comes from the traditionalend of the naming spectrum Found

in 1884 by Austrian astronomer JohannPalisa, the asteroid was named after themythic mountain where the infant Zeushid from his father Honoring PalisaÕs

spirit, the Galileo team proposed

call-Siberians only recently gained the right

to own land, Ịthey wonÕt feel the sting

of restrictions on what they can do with

it.Ĩ Businesses may be harder to placate

The U.S Agency for International

Devel-opment has promised $3.4 million for

12 projects in areas such as ecotourism

and forest management DavisÕs Þrm is

identifying American companies

will-ing to abide by the new rule that all

foreign-owned facilities must meet the

environmental regulations of the

own-erÕs country as well as local standards

The two provinces that have adopted

the land-use program cover 95 percent

of the Russian watershed But Irkutsk,

which has balked at the Baikal plan and

recently accepted German funding for

its own survey, accounts for 40 percent

of the lakeÕs shoreline and much of the

waste that is dumped from it Another

70 million acres of BaikalÕs watershed

lies in Mongolia, which has just begun

reviewing its own, very similar plan

Perhaps more important than the

Bai-kal project itself is the speed with which

it is being copied Davis is now working

with the Khabarovsk and Primorsky

territoriesĐ1,500 miles to the east ofBaikal Nestled against the HeilongjiangProvince of China, with which they sharethe Ussuri River, these Far Eastern Rus-sians worry less about the threat of pol-lution than the temptation to sell oÝrights to their lush woodlands The for-est supports the richest diversity ofplant species in the former Soviet Union

ỊOn the Chinese side, the Ussuri sin contains the most wetlands remain-ing anywhere in the country,Ĩ says JimHarris, deputy director of the Interna-tional Crane Foundation, which hasbeen monitoring wetland destruction

ba-in Chba-ina ỊHundreds of thousands ofacres have already been drained andconverted to farms,Ĩ he explains Whatlittle is left is worth preserving ỊIn thisbasin live the last 250 Siberian tigers,last 30 Amur leopards and two endan-gered species of cranes,Ĩ Davis reports

The Russian, American and ChinesescientiÞc teams will soon present theirrecommendations to ensure that devel-opment in the 60-million-acre area doesnot overburden the ecosystems At thatpoint, says Elizabeth D Knup, program

director of the National Committee onU.S.-China Relations, the real Þreworksmay begin ỊTo have the two sides nowtalking about how to jointly manage thewatershed is pretty extraordinary,Ĩ Knupsays ỊItÕs a very sensitive borderĐtheywere shooting over it until the 1960s.Ĩ Davis notes that the plans have raisedrelatively little opposition because ofthe areaÕs remoteness and the slowgrowth of these economies ỊWeÕre for-tunate in all of these areas that wearenÕt dealing with an overpopulationsituation,Ĩ he concedes ỊBut if we canprove that it can work in these regions,then we can consider other, more chal-lenging areas.Ĩ It appears as though Da-vis may get that chance: he has beenapproached by the Haisla Indian Nation

of British Columbia, by the Miskito dians of Nicaragua and by oÛcials inBolivia and in Chile

In-In time, the ultimate symbolism ofLake Baikal will emerge, and it may well

be the failure of good intentions On theother hand, practice could well makesustainable development, if not perfect,

at least more practical.ĐW Wayt Gibbs

16 SCIENTIFIC AMERICAN December 1994

The Astronomical Naming Game

A quick ßip through the baby book for heavenly bodies

ASTEROID IDA is accompanied by the Þrst known asteroid moon, Dactyl (far right).

Such discoveries test the system for coming up with distinctive but consistent names

Copyright 1994 Scientific American, Inc.

Rise and shine has taken on new

meaning in the physicistÕs

vo-cabulary Investigators at Brown

University have managed to trap

ßoat-ing droplets of liquid helium in midair

by shining laser light on them The feat

should allow researchers to probe for

the Þrst time how the ßuid behaves in

free space

Although helium is most familiar as

a gas that Þlls up balloons and changes

the pitch of the human voice at parties,

it serves in its liquid form as a major

tool in condensed-matter physics That

is because it behaves unlike anythingelse when cooled to near absolute zero

SpeciÞcally, below 2.172 kelvins heliumbecomes a quantum liquid known as

a superßuid It loses all resistance toßow and viscosity, enabling it to seepthrough cracks even a gas could notpenetrate Sloshing a bucketful of itaround in circles produces even strang-

er phenomena The rotation createsnanometer-size whirlpoolsĐcalled quan-tized vorticesĐthroughout the liquid

Researchers have been exploiting theproperties of superßuid helium to study

condensation, turbulence, ßuid ßow andnew forms of matter

But physicists had never looked atisolated drops of superßuid heliumĐinfact, nobody is quite sure how the dropsbehave To help answer that question,Mark A Weilert, Dwight L Whitaker,Humphrey J Maris and George M Sei-del of Brown applied a technique thathas been reÞned to an art during thepast several years: the trapping of par-ticles by laser beams They submerged

a small piezoelectric speaker in a perßuid helium bath kept in a cryostat.Turning on the speaker produced aÞne mist of superßuid helium dropletsabove the surface of the liquid Two la-ser beams shot through windows in

su-the cryostat were aimed

in opposite directions atthe droplets

ỊMost of the dropletssimply fall down,Ĩ Marisexplains, Ịbut we areable to trap one or even

a few at a time.Ĩ The vestigators could tell theyhad succeeded by look-ing at the laser light re-ßected oÝ the surface ofthe drops They deducedthat they had suspendeddrops 10 to 20 microns

in-in size for up to threeminutes, during whichtime the droplets slowlyshrank through evapora-tion Larger drops couldnot be held, because theywould require lasersstronger than those thatcould be provided.The work, to be pub-lished in the January is-

sue of the Journal of Low

Temperature Physics,

rep-resents the Þrst step inexploring a novel realm

Trapped in the Light

Laser beams levitate droplets of superßuid helium

ing IdaÕs moon Dactyl, after the Dactyli,

a group of magicians who inhabited

Mount Ida

At Þrst, the IAU was not sure whether

the moon even merited a moniker of

its own Because asteroids are so

numer-ous, the union approves a name only

after the orbit has been determined;

Galileo did not observe the moon long

enough to describe its motions

Mars-den Þnally decided the discovery of the

Þrst asteroid satellite was important

enough to modify the requirements

At the more free-form end of

solar-system nomenclature is the asteroid

Zappafrank After the death of musician

Frank Zappa, Arizonan John Sciatti led

a campaign to have a celestial body

named after the late guitarist Marsdenrapidly found himself inundated withE-mail Because of ZappaÕs close rela-tionship with V‡clav Havel, president ofthe Czech Republic, Marsden prevailed

on Czech astronomers to ỊproduceĨ anunnamed asteroid to bear ZappaÕs name

An asteroid named Zappala already isted, as did several whose names be-gan with Frank, so the IAU settled onZappafrank

ex-Although the IAU can stomach a tain amount of whimsy, it does draw theline News that three planets had beendiscovered around a pulsar promptedNational Public Radio to solicit sugges-tions for what to call them The winners:

cer-Curly, Moe and Larry ỊI donÕt think the

IAU would go for that,Ĩ Marsden les In addition, he notes, Ịthe IAU doesnot name stars.Ĩ Marsden is particular-

chuck-ly disdainful of the International StarRegistry, an unoÛcial and utterly unre-lated organization that names stars for

a fee ỊItÕs a total racket,Ĩ he hisses.From MarsdenÕs point of view, thewhole naming game is just a pleasantdistraction from the real business ofastronomy ỊI donÕt care about thenamesĐI study the orbits,Ĩ he crustilyjokes But he concedes that the impulse

to name is tough to Þght; the best theIAU can do is try to bring some order

to the process ỊIf the IAU declares Ơnomore names,Õ Ĩ he sighs, Ịsomebody elsewill just do it.Ĩ ĐCorey S Powell

Red sprites and blue flashes were recently found to live above some

thunderstorms—al-though pilots have been reporting the luminous phenomena for many years The red

flashes appear for only a few thousandths of a second and can extend upward for 60

miles; the blue jets also appear atop the storms and can rise for about 20 miles These

first color images of the activity, shown here inside a photograph of a storm, were taken

by researchers at the Geophysical Institute at the University of Alaska at Fairbanks

Electrical Activity above Thunderstorms

If a little knowledge is a dangerous

thing, it might follow that vast

amounts of knowledge

concentrat-ed in one place are downright

hazard-ous Evidence for such a conclusion

could be found at the Massachusetts

Institute of Technology this past

Octo-ber, when a good portion of AmericaÕs

scientiÞc elite, including three bona Þde

Nobel laureates, cringed through an

evening of ear mites, constipation and

threats of eternal damnation Yes, it

was time once again for the awarding

of the Ig Nobel Prizes

Some 1,200 spectators jammed

M.I.T.Õs Kresge

Audi-torium to witness the

ỊFourth First Annual

Ig Nobel Prize

Cere-mony.Ĩ They also

ogled real Nobelists

Unlike the awards

won by these

exem-plary scientists, the

to the oÛcial

pro-gram A joint

produc-tion of the Annals of Improbable

Re-search (described by some as the Mad

magazine of science) and the M.I.T seum, the Igs take their name from theỊlegendary Ignatius (Ig) Nobel, co-in-ventor of soda pop,Ĩ allegedly a distantrelative of TNT inventor Alfred, whofounded those other prizes Whereasproof of IgÕs existence might be hard todocument, the Igs are awarded to realpeople, embarrassed though they mayfeel, for real work, embarrassing though

Ser-U.S Troops,Ĩ which appeared in Military

Medicine in 1993 W Brian Sweeney, one

of the writers, showed up to receive the

Ig, a gold-painted, wax brain hemisphere.ỊIÕd like to acknowledge all of our won-derful U.S servicemen,Ĩ he said, Ịwhowere willing to become constipated forthe country There were various theo-ries as to why constipation occurs, un-til it was pointed out to me by one ofthe marines in the Þeld He said, ƠDoc,let me tell you When weÕre out in theÞeld, weÕre scared sĐless.Õ Ĩ

Patient X, who refused to be named,won the Medicine Ig for his attempt touse electroshock to neutralize venomafter he had been bitten by his pet rat-tlesnake The juice came from a car en-gine revved to 3,000 rpm for Þve min-utes It was applied through sparkplugwires attached to Patient XÕs lip Xshared the award with the authors of amedical report of the incident, ỊFailure

of Electric Shock Treatment for tlesnake Envenomation,Ĩ published in

Rat-the Annals of Emergency Medicine In a

taped message, co-author Richard C.Dart of the Rocky Mountain PoisonCenter said, ỊI was stunned to receivethe 1994 Ig Nobel Prize in Medicine, al-though not as shocked as our patient.ĨVeterinarian Robert A Lopez took theEntomology Ig for his brave and suc-cessful attempts to Þnd out whetherear mites from cats can inßict damage

on humans He did this by insertingmites into his own ears, not once, nottwice but three times LopezÕs chilling

report was published in the Journal of

the American Veterinary Society At a

post-Ig gathering,Lopez elaborated onhis actions: ỊSome-bodyÕs got to be cra-

zy enough to do it.Hey.Ĩ

Former Texas statesenator Bob Glasgowcopped the Ig inChemistry for hissponsorship of a

1989 drug-controllaw that would make

it illegal to purchaselaboratory glasswarewithout a permit.Accepting for himwas one Tim Mitch-ell, a representative

of Corning Ratherthan a total ban onglassware, Mitchellsuggested a ỊÞve-daycooling-oÝ period.Ĩ

He admitted,

howev-er, that beakers and

22 SCIENTIFIC AMERICAN December 1994

for superßuid helium ỊThere are a lot

of things to do with superßuid drops,Ĩ

says Russell J Donnelly, a physicist at

the University of Oregon who has been

attempting to levitate superßuid drops

with electric and magnetic Þelds One

could, he remarks, observe how drops

collide or move about Indeed, MarisÕs

group is primarily interested in seeing

how a superßuid drop rotatesĐỊa

sur-prising thought a couple of years ago,Ĩ

Maris says An ordinary drop of liquid

may rotate in a complicated fashion for

a while, but it eventually settles into a

motion like that of a rigid body, where

each part has the same angular velocity

A superßuid droplet, however, would

not behave that way The liquid has no

viscosity and must obey certain

quan-tum-mechanical conditions that prevent

it from rotating as a rigid body Instead

theorists suggest that the droplet might

become peppered with quantized tices or produce a bulge that circles thedroplet

vor-To see such dynamics, workers willprobably need to suspend larger drops,perhaps several centimeters in size Forthat job, Maris and his colleagues havealready begun redesigning their appa-ratus, using superconducting magnetsrather than lasers Helium is slightly repelled by magnetic Þelds, so dropsshould be able to ßoat on a magneticcushion, sidestepping the practical en-ergy limitations of lasers

In fact, the new magnet should able the physicists to go beyond exoticdrops of ßuid ỊWeÕre thinking aboutlevitating frogs,Ĩ Maris says, because theability to ßoat amphibians oÝers an al-ternative to seeing how they develop inthe absence of gravity Besides, it would

en-make a great party trick ĐPhilip Yam

The Annual Ig Nobel Prizes

This yearÕs winners are, well, just as pathetic as last yearÕs

INTERPRETIVE DANCE of the electrons cast authentic Nobel laureates as atomic nuclei William Lipscomb ( left), winner of the 1976 Prize in Chemistry, notes that his rhythm is good, Ịbut IÕm a lousy dancer.Ĩ

Copyright 1994 Scientific American, Inc.

test tubes can start a habit that might

leave one Òstrung out, begging for grant

money.Ó

The awards were interrupted

period-ically for Heisenberg Certainty Lectures

(named for that pillar of modern

phys-ics, the Heisenberg uncertainty

princi-ple), delivered by the real Nobel

laure-ates and other honored guests The

cer-tainty: no lecture lasts more than 30

seconds, or a black-clad referee

whis-tles the speaker oÝ the stage

ArtiÞcial-intelligence maven Marvin Minsky

bare-ly Þnished his comments, but Lipscomb

wrapped up his address with plenty of

time to spare ÒThe following statement

of the Heisenberg Certainty Principle is

dedicated to the U.S Congress,Ó

Lips-comb began ÒIf your position is

every-where, your momentum is zero,Ó he

concluded

One of last yearÕs winners, Harvard

UniversityÕs John Mack, had been asked

to deliver the keynote address, but he

backed out Mack won the 1993

Psychol-ogy Ig for his theory that people who

believe they were abducted by aliens

probably were ÒWeÕre disappointed and

hurtÓ over MackÕs absence, said Ig

mas-ter of ceremonies Marc Abrahams, Òbut

above all, weÕre concerned.Ó

The eveningÕs Þnal Ig, for

Mathemat-ics, went to the Southern Baptist Church

of Alabama, for Òtheir

county-by-coun-ty estimate of how many Alabama

citi-zens will go to hell if they donÕt repent.ÓThe Honorable Terje Korsnes, consul

of Norway, accepted the Ig on behalf ofthe people of Hell, a little town in Nor-way ÒWe have a special place in Hellfor all of you,Ó Korsnes said

During the apr•s-Ig celebration,

Min-sky summed up his impressions of theceremony ÒItÕs one of my principlesthat if I have a complex experience thatlasts a couple of hours, I can neverthink of any few silly words to describeit,Ó he stated ÒSo I think itÕs bad tosummarize.Ó ÑSteve Mirsky

And the other 1994 Ig Nobel Prize winners are:

Lee Kuan Yew, former prime minister of Singapore Winner of the Ig in

Psy-chology for his 30-year study of the effects of negative reinforcement,

name-ly, the punishing of the citizens of Singapore “whenever they spat, chewedgum, or fed pigeons.”

The Japanese Meteorological Agency Awarded the Physics Ig Nobel “for

its seven-year study of whether earthquakes are caused by catfish wigglingtheir tails.”

L Ron Hubbard Recipient of the Ig in Literature “for his crackling Good Book,

Dianetics, which is highly profitable to mankind or to a portion thereof.”

Chile’s Juan Pablo Davila, former employee of the state-owned company

Codelco Davila’s Ig in Economics was awarded for instructing his computer

to “buy” when he meant “sell.” The ultimate consequence was the loss of 0.5percent of the gross national product In Chile “davilar” is now a verb mean-ing “to botch things up royally.”

John Hagelin of Maharishi International University and the Institute of

Sci-ence, Technology and Public Policy Winner of the Ig Nobel Peace Prize “for hisexperimental conclusion that 4,000 trained meditators caused an 18 percentdecrease in violent crime in Washington, D.C.” —Mervin Stykes

Mark H Skolnick of the

Universi-ty of Utah and his 44

collabo-rators at Þve research facilities

had good reason to celebrate when they

found BRCA1, a gene whose

malfunc-tion accounts for nearly half of all

in-herited breast cancers, or some 5

per-cent of the total The discovery ended

one of the most widely publicized andpotentially proÞtable gene hunts to

date Once revealed, BRCA1Õs secrets

may eventually lead to better ments for familial breast and ovariancancers

treat-But despite such promise, some vocacy groups and scientists alike are

ad-questioning how knowledge of the sive gene will be applied in the inter-imÑand who stands to gain, by howmuch These ethical and legal issues are

elu-complicated by the fact that BRCA1Ña

stretch of chromosome 17 that is some

10 times longer than the average man geneÑseems far from ordinary.Unlike most other known cancer genes,which play a role in both familial and

hu-nonfamilial cancers, BRCA1 apparently

plays no role in nonfamilial breast and

Toxic Waste and Race:

An Unnatural Association

Hazardous-waste sites are too close for comfort in many

minority communities, concludes a report by the Center

for Policy Alternatives in Washington, D.C The recent

up-date of the well-publicized 1987 study by the United

Church of Christ Commission for Racial Justice, Toxic

Wastes and Race in the United States, found that the

situa-tion has worsened during the past six years People of

col-or—defined by the report as the total population less

non-Hispanic whites—are currently 47 percent more likely than

are whites to live near a commercial toxic-waste facility

The population of neighborhoods changes according to

whether there is

Deciphering the Breast Cancer Gene

Experts grapple with the implications of the Þnding

19801993

no hazardous-waste site nearby,

one such facility,

one landfill,

more than one waste facility or a large landfill,

or three facilities, an incinerator or a large landfill

RESIDENTS WHO ARE PEOPLE OF COLOR (PERCENT)

0 10 20 30 40 50

ovarian cancer And so far the

discover-ers have identiÞed Þve mutations that

occur in diÝerent regions of the gene,

all of which prevent it from producing

whatever protein it normally should,

presumably a tumor suppressor

Because BRCA1 is so complex, it will

be diÛcult to invent a simple test that

accurately predicts a womanÕs risk for

breast cancer, says David E Goldgar, a

member of the team at Utah ÒCertain

mutations seem to confer a higher risk

of ovarian cancer, and some seem to

trigger an earlier onset of the disease,Ó

he explains ÒIt could be random chance

that one woman never develops breast

cancer and that another with the same

mutation does before age 30.Ó

Current estimates suggest that a

wom-an who has inherited a BRCA1

muta-tion faces an 85 percent lifetime risk of

battling the diseaseÑbut that Þgure is

based on studies done before the debut

of BRCA1Õs location In fact, a womanÕs

risk might vary considerably depending

both on which hallmark mutation her

family passes along and on

environ-mental factors, notes Donna

Shattuck-Eidens, a co-discoverer and project

lead-er at Myriad Genetics, a company based

in Salt Lake City founded three years

ago by Skolnick and Nobel laureate

Wal-ter Gilbert of Harvard University The

Þrm is currently seeking patent

protec-tion for BRCA1.

By January 1996 Myriad hopes to

of-fer a blood test (costing about $1,000)

that detects deleterious copies of BRCA1,

Shattuck-Eidens says Hybritech, a

sub-sidiary of Eli Lilly that contributed $1.8

million to the BRCA1 quest, has licensed

the right to market this test Because the

test screens for one speciÞc mutation

at a time, Myriad will need to know

which one a woman might

carryÑprob-ably from having tested a relative with

breast or ovarian cancer ÒThe results

will take some expert interpretation to

assess what risks a woman really

fac-es,Ó Shattuck-Eidens admits

Because women who learn their risks

can, for the moment, do little to change

themÑshort of having their breasts

surgically removed before a tumor

ap-pearsÑsome people question the merit

of MyriadÕs planned service Fran Visco,

president of the National Breast Cancer

Coalition, an advocacy group, points

out that women who show positive

re-sults might forfeit their health and life

insurance A 1993 survey of health

in-surance commissioners in 32 states

found that 44 percent believed a family

history of breast cancer was suÛcient

reason to deny coverage

ÒWhen it comes to health issues, the

more information you have, the better

oÝ you are,Ó Shattuck-Eidens counters

SCIENTIFIC AMERICAN December 1994 27

Copyright 1994 Scientific American, Inc.

Certainly, detecting breast cancer at anearly stage is crucial If it is treated be-fore cancerous cells metastasize to oth-

er sites in the body, the Þve-year vival rate is 95 percent Once it has invaded other systems, that windownarrows to 17 percent Moreover, can-cers that emerge at a young ageĐaquarter of which are genetic in originĐcan spread more quickly

sur-An estimated 600,000 U.S women

harbor bad copies of BRCA1, and

ex-perts all agree that translating the geneÕscode into treatments for them, if possi-ble, will require great eÝort Some worrythat MyriadÕs pending patent will im-pede such progress by discouraging co-operative research A group at the In-stitute of Cancer Research in Englanddecided not to participate in the Utahsearch for another likely breast cancer

gene, BRCA2, citing disagreements over

the ethics of patenting human genes

ỊYou can agree to disagree, but it tainly doesnÕt mean you canÕt work to-gether,Ĩ Goldgar says, observing that re-ports of the split between American and

cer-British teams are overblown (BRCA2

may cause as many cases of inherited

breast cancer as does BRCA1, and its

identity could soon be uncoveredĐworkers now know that it resides some-where along chromosome 13 Evidencesuggests that other such similar genesĐ

including BRCA3 and 4 and perhaps even 5 and 6Đmay exist as well, al-

though taken together, they would count for far fewer cases of cancer than

ac-does BRCA1 or 2 alone.)

In the past, gene hunters have sharedvast amounts of data so that they mightrapidly ferret out the cause of a disease

Patents and the call of proÞts, however,might make some researchers more se-cretive At present, the U.S Patent Ỏcerequires that any discovery or invention

be novel (or unpublished) and ousĐstandards that, if misread, couldlimit free exchange Moreover, the Þndmust be useful and neither an idea nor

nonobvi-a product of nnonobvi-ature Mnonobvi-any resenonobvi-archersmaintain that human genesĐparticular-

ly partial DNA fragments or sequences

of unknown functionĐdo not fully meetthese Þnal criteria

Reid G Adler, a patent attorney atMorrison & Foerster in Washington,D.C., and former director of the Ỏce

of Technology Transfer at the NationalInstitutes of Health, concedes that aspecial system may be needed to pro-tect some gene-related discoveries Nev-

ertheless, BRCA1Õs case seems clear-cut,

he says The geneÕs malformation givesprediagnostic indication of a disease,and although the gene itself is natureÕshandiwork, a diagnostic kit based onthe characterization of its role in cer-

tain cancers is not ỊNo one developscommercial products that are risky andrequire vast sums of money when any-one else could then proÞt from them,ĨAdler says ỊThe main purpose of thepatent system is to encourage people toinvest in research by giving them someeconomic advantage.Ĩ

Without the promise of patent

pro-tection, Goldgar guesses that BRCA1

would not have been located so tiously Rival researchers began chasingdown this gene four years ago, whenMary-Claire King of the University of

expedi-California at Berkeley traced BRCA1 to

the long arm of chromosome 17 ỊPart

of the reason it was found when it was

is because there was a company volved with adequate resources to get

in-a lot of people working on it,Ĩ Goldgin-arsays Shattuck-Eidens concurs: ỊThis un-dertaking was a cooperative eÝort be-tween research, university and industri-

al partnersĐand of course they all havediÝerent weights and measuresĐbutIÕm of the opinion that it works to ev-eryoneÕs advantage in the end.ĨAdler dismisses any fears that

BRCA1Õs medical potential might be

compromised by its tion ỊMyriad and Eli Lilly canÕt monop-olize the entire universe of breast-can-cer test kits,Ĩ he notes Because the NIHhelped to fund the project, the govern-ment could establish sublicensing ar-

commercializa-rangements if knowledge about BRCA1

were not being used in the publicÕs bestinterest The NIH has never sought theseso-called margin rights, though, and itseems in this case the agency hopes toassume an active role in licensing tech-

nology based on BRCA1 On October 6,

the NIH Þled a counterapplication toadd its scientistsÕ names to MyriadÕspatent as coinventors ỊPatents donÕtinterfere with academic science,Ĩ Adlerstates, Ịand they are essential in thecommercial realm.Ĩ

The Ỏce of Technology Assessmentwill produce an investigative report ear-

ly next year But BRCA1 is by no means

the Þrst human gene that scientists havesought to patent Human Genome Sci-ences and SmithKline Beecham hold a

patent on APC, which causes colon

can-cer, and Sequana Therapeutics likewisehas rights to the so-called obesity gene

Still, BRCA1 has stirred up far more

controversy ỊBreast cancer is a muchmore emotional issue for many people,ĨGoldgar says, Ịand incredibly common.ĨOne in eight American women will ac-quire breast cancer during their lives,and the disease claims some 46,000mothers, sisters, wives and daughters

every year In that light, BRCA1 deserves

all the scientiÞc, legal and public tion it can get ĐKristin Leutwyler

atten-32 SCIENTIFIC AMERICAN December 1994

Mathematicians like to think

their truths are as objective as

any we humans are permitted

to know That may be so, yet an

unusu-ally persistent and rancorous dispute

over a famous problem called KeplerÕs

conjecture has revealed just how

sub-jective the process of judging those

truths can be

The controversy began innocently

enough four years ago, when Wu-Yi

Hsiang of the University of California

at Berkeley decided to teach a course in

classical geometry To sharpen his skills

in this old-fashioned Þeld, he took on a

conjecture posed in 1611 by Johannes

Kepler, the same German polymath who

discovered that planets travel in

ellipti-cal rather than circular orbits

Kepler contended that the most

com-pact method of packing spheres is the

one exploited by nature to arrange

atoms into crystals and by grocers to

stack oranges into four-sided pyramids

The easiest way to create this pattern

is to form a layer of spheres consisting

of even vertical and horizontal rows;

spheres in the next layer up nestle in

the niches between each foursome of

spheres in the layer below

Few mathematicians doubt KeplerÕs

conjectureÑwhich is related to

prob-lems in solid-state physics, information

processing and other ÞeldsÑbut they

have had a devilishly diÛcult time

prov-ing it After all, there are inÞnite ways

to arrange spheres in a given volume

Douglas J Muder, until recently of tre Corporation, has established thatspheres can Þll no more than 77.3 per-cent of a volume, but KeplerÕs conjec-ture states that the upper bound is ap-proximately 74 percent (that is, π di-vided by the square root of 18)

Mi-After six months of pondering theproblem, Hsiang became convinced hehad a proof Although his argumentdrew on relatively standard techniquesfrom geometry and calculus, it waslongÑmore than 100 pages in an earlydraftÑand intricate HsiangÕs basic ap-proach was to calculate the ÒlocalÓ den-sity achieved by various Þnite conÞgu-rations of spheres and then to extrapo-late these results to inÞnitely largevolumes

Hsiang began circulating a draft of

his proof and lecturing on it in 1990,

and his work was soon hailed in Science,

New Scientist and this magazine

Mean-while a group of four experts on spherepackingÑMuder, John H Conway ofPrinceton University, Neil J A Sloane ofBell Laboratories and Thomas C Hales

of the University of MichiganÑstartedquestioning the proof The group com-plained that HsiangÕs paper, as long as

it was, was short on details: its jumpsfrom particular cases to generalitieswere insuÛciently justiÞed

The critics wrote several letters toHsiang challenging his proof Far fromretracting his claim, Hsiang submitted

his paper to the International Journal of

Mathematics, which is edited by

anoth-er Banoth-erkeley mathematician, ShoshichiKobayashi After Hsiang had made somerevisions, the journal published HsiangÕs92-page paper in October 1993.This past spring Conway, Hales, Mu-

der and Sloane announced in The

Math-ematical Intelligencer that they Òdo not

consider that HsiangÕs work constitutes

a proof of KeplerÕs conjecture, or can becompleted to one in a reasonable time.Ó

In the summer issue of the

Intelligenc-er, Hales presented a tart, 12-page

sum-mary of the groupÕs main objections toHsiangÕs work He suggested thatHsiangÕs paper was at best a series ofconjectures that, if demonstrated, mightconstitute a proof ÒMathematicians caneasily spot the diÝerence between hand-waving and proof,Ó Hales concluded.Conway predicts that Hsiang will sac-riÞce his ÒdistinguishedÓ reputation if

he persists in claiming to have a proof

ÒI think heÕd be better advised to dropit,Ó he remarks According to Mu-der, the controversy has alreadydiscouraged other mathemati-cians from working on KeplerÕsconjecture, since no one wants topursue a problem that may besolved ÒIt slowed things down alot,Ó he says

Sloane contrasts HsiangÕs havior with that of Andrew J.Wiles of Princeton University In

be-1993 Wiles announced he hadproved FermatÕs Last TheoremÑperhaps the most celebrated co-nundrum in mathematicsÑbut

he promptly withdrew his claimafter colleagues pointed out short-comings Sloane calls HsiangÕsdecision to publish his paper inspite of the objections Òextraordi-nary.Ó ÒYou canÕt regard [ Hsiang ]

as a serious mathematician,ÓSloane sniÝs

Hsiang, whose rebuttal to hiscritics will be published in the

winter 1995 Intelligencer, retorts

that their complaints consist ofÒmisunderstandings, misinterpretations,misaccusations.Ó His proof Ògives allthe crucial understandingÓ and omitsonly Òboring computation,Ó he declares.Hsiang admits only that he may have aÒcommunication problem.Ó In collabo-ration with Karoly Bezdek, a Hungarianmathematician now at Cornell Univer-sity, Hsiang plans to construct a moredetailed version of his proof

Bezdek agrees with Hsiang that Halesand his colleagues Òhave either misun-derstood or by purpose did not want

to followÓ his ideas Yet he also thinksHsiangÕs proof is not complete ÒIÕmoptimisticÓ that at least one crucial com-ponent of the proof can be completed,

Global Politics

Mathematicians collide over a claim about packing spheres

PYRAMIDS OF FRUIT display what Johannes Kepler conjectured in 1611 to be the most

compact arrangement of spheres, the so-called face-centered cubic lattice

Copyright 1994 Scientific American, Inc.

Bezdek notes, Ịbut there are gaps.Ĩ

Moreover, Bezdek acknowledges that he

may be inclined to favor HsiangÕs work

in part because it extends an approach

Þrst developed in Hungary Struggling

to sum up his view of the situation,

Bezdek says, ỊThe picture is at the

mo-ment not so objective.Ĩ

For now, public opinion seems to have

turned against Hsiang In 1992 Ian

Stew-art of Warwick University still thought

Hsiang might have achieved Ịone of themost astonishing successes in the en-tire history of mathematics.Ĩ Stewartcheerfully admits that he is not an ex-pert on sphere packing; his assessmentwas based on secondhand reports aboutHsiangÕs reputation and argument rath-

er than on a rigorous analysis of theproof Now Stewart is inclined to be-lieve HsiangÕs critics, who are equallyeminent

Indeed, some observers fear that thespat over KeplerÕs conjecture points to

a deeper, more pervasive quandary ing mathematics: as the Þeld grows in-creasingly complex and specialized, theevaluation of proofs is becoming morediÛcult ỊIt is harder to check proofsthan it used to be,Ĩ conÞrms ChandlerDavis of the University of Toronto, edi-

fac-tor of the Intelligencer ỊThe process has become unmanageable.ĨĐJohn Horgan

The power of modern medicine

has rarely been demonstrated so

well as it was in the spring of

1993, after physicians near the

junc-tion of New Mexico, Arizona, Utah and

Colorado reported a spate of severe

res-piratory illness resulting in more than

a dozen deaths, primarily among

Nava-jo Indians Within months, researchers

from the Centers for Disease Control

and Prevention and elsewhere had

iden-tiÞed a viral culprit in the so-called Four

Corners outbreak

Yet medical mysteries rarely succumb

to science without a struggle, and

inves-tigators of the incident are still trying

to tie up some signiÞcant loose ends:

Why do some people become infected

while others seemingly exposed to the

same risk factors do not? Why do some

people who show all the signs of the

respiratory illness not test positive for

the virus? ỊI still think lots of questions

need to be answered,Ĩ remarks

Shyh-Ching Lo, a researcher at the Armed

Forces Institute of Pathology

The CDC has tentatively named the

new pathogen the Sin Nombre, or

no-name, virus It is genetically similar to

a family of viruses, called hantaviruses,

known to cause acute kidney-related

illness in Asia and Europe Hantaviruses

take their name from the Hantaan

Riv-er, which runs through an area in

Ko-rea where the disease is endemic

Han-taviruses were Þrst detected in the U.S

more than a decade ago but only in a

nonvirulent form

By October of this year the CDC had

reported 94 cases of the hantavirus

pul-monary syndromeĐmore than half of

them fatalĐin 20 states Investigators

believe victims become ill by inhaling

dried urine or feces of infected deer

mice, which are the primary vectors of

the virus About 30 percent of the deer

mice in the Four Corners region carry

the Sin Nombre agent; infected rodents

have been found in other parts of the

country as well

Yet some victims seem to have

tracted the illness after little or no tact with rodent carriers One Rhode Is-land man who died this past January ofhantavirus pulmonary syndrome wasinitially thought to have contracted thevirus a month earlier while sweepingout a warehouse in Queens, N.Y Yet a

con-recent report in the Journal of the

Amer-ican Medical Association noted that

none of the rodents trapped in that cationĐor any others where the victimhad been during the two months be-fore his deathĐhad positive results forhantaviruses

lo-Studies have also shown that evenpeople seemingly most at risk rarely be-come infected and that the virus doesnot trigger illness in all those it infects

Laurie R Armstrong of the CDC

recent-ly tested more than 900 pest-controlworkers and others who frequently han-dle deer mice and other rodents known

to carry hantaviruses Only eightĐlessthan 1 percentĐwere positive for SinNombre Of these, only one recalledhaving an illness resembling hantavi-rus pulmonary syndrome

The CDC has analyzed blood samplestaken from some 500 Navajos in theFour Corners area before the outbreak

One percent of that group had ies to the Sin Nombre virus, but nonereported having an illness resemblingthe pulmonary syndrome A study ofsouthern CaliforniaÕs Channel Islandshas turned up similar results The is-lands are so infested with deer micethat the animals commonly run over thefeet of hikers; a signiÞcant percentagecarry hantaviruses Yet Michael S Asch-

antibod-er, an investigator for CaliforniaÕs partment of Health Services, says a sur-vey of residents of the islands hasturned up no apparent cases

De-James E Childs of the CDC edges that the link between rodents andvictims remains unclear ỊWe do notknow why some people become infect-

acknowl-ed and others donÕt,Ĩ Childs says Peter

B Jahrling of the U.S Army Medical search Institute of Infectious Diseases

Re-suggests that the Sin Nombre virusmight act in concert with a cofactor tocause the pulmonary syndrome Work-ers at the CDC and elsewhere say theyhave considered such cofactors as

Chlamydia, Mycoplasma and various

environmental toxins but have found

no supporting evidence

Perhaps the most disturbing questionraised by the outbreak concerns peoplewho exhibited symptoms of hantaviruspulmonary syndrome but showed neg-ative results for the virus This issuewas highlighted in a recent letter to the

New England Journal of Medicine by two

workers at the University of California

at San Francisco, Tina Harrach claw, a pharmacologist, and her hus-band, Wilfred F Denetclaw, a cell biolo-gist who grew up in a Navajo family inthe Four Corners region

Denet-The Denetclaws pointed out that aminority of the cases investigated bythe CDC had shown signs of infection

by the Sin Nombre virus ỊRegardless

of whether hantavirus is the etiologicagent of the hantavirus pulmonary syn-drome, a large number of cases in theoutbreak were not associated with han-tavirus and remain unexplained,Ĩ theDenetclaws stated

Indeed, Bruce Tempest of the IndianHealth Service notes that at least onesuch case has occurred recently in NewMexico In California, Ascher has uncov-ered half a dozen incidents in whichrelatively young and healthy peopledied suddenly of acute respiratory fail-ure yet did not test positive for hanta-virus or any other pathogen The vic-tims had all the classic symptoms ofhantavirus syndrome, including expo-sure to rodents, Ascher says Similarcases have turned up in Nevada, ac-cording to Arthur F DiSalvo, director ofthe stateÕs public health laboratory.Clarence J Peters, chief of the CDCÕshantavirus task force, conÞrms thatonly 25 percent of the cases of suspect-

ed hantavirus pulmonary syndrome ported in the Four Corners area by thispast January had been linked to the SinNombre virus The percentage may bemuch smaller when cases from otherparts of the country are taken into ac-The No-Name Virus

re-Questions linger after the Four Corners outbreak

36 SCIENTIFIC AMERICAN December 1994

Athin layer of soot falls

continu-ously on Port-au-Prince, a visitor

reports This black dust is not

so much the fallout of two years of

in-ternational oil embargo as of two

cen-turies of economic myopia Made

indis-criminately from any wood, charcoal

powers dry-cleaning plants, bakeries

and the cookstoves of the richĐthe

poor burn their wood only once More

than 90 percent of Haiti has reportedly

been denuded, leaving the country

be-reft of natural resources crucial to its

economic survival

Wallace Turnbull, a missionary who

has spent most of his life in Haiti, once

asked oÛcials why they did not

im-port cheap kerosene to reduce

de-forestation They replied, he says,

that such largesse would endanger

the proÞtable tax on diesel fuel

be-cause people might adulterate the

kerosene to run trucks Other

oÛ-cials, meanwhile, were exporting

charcoal to the neighboring

Domin-ican Republic, where environmental

regulations forbid its production

This kind of bizarre subtext

seems typical of the Haitian

land-scape Barbara Lynch of the Ford

Foundation recounts how the

Hai-tian army destroyed tree seedlings

that were part of a reforestation

project Although the trees might

have been good for the

environ-ment and hence the country as a

whole, she explains, the rural

devel-opment program that planted them

brought peasants together The

re-sulting coalition diminished the

army and the Tonton MacoutesÕ

control over the countryside,

threat-ening the long-standing

arrange-ments by which they Ịextracted

re-sources upward.Ĩ

Political and economic power are

often closely linked, but in Haiti the

two became almost indistinguishable

The government, according to Rolph Trouillot of Johns Hopkins Uni-versity, had Ịno role other than as apredatory mechanism for the elite.ĨViewed in this light, many counterpro-ductive aspects of the Haitian economycan be understood not simply as short-sighted individualism run amok butrather as trade-oÝs between the per-ceived utility of higher proÞts for theislandÕs owners and the risk that anymoney trickling down might upset theestablished imbalance Lynch notes thelack of investment in the simplest ofinfrastructure: roads, schools and pub-

Michel-lic utilities, even in rich neighborhoods Some of the elite have made a proÞtfrom both the instigation and the after-math of environmental destruction.Government lands near Port-au-Princeserved as free quarries for concrete tobuild mansions, Trouillot says, but theresulting erosion loaded nearby riverswith sediment The delvers then trucked

in potable water from more distant ers and sold it to those whose supplythey had rendered undrinkable.Yves Renard, director of the Carib-bean Natural Resources Institute, re-ports malign neglect throughout thecountryside, where hoe-based farmingmethods have not changed substantial-

riv-ly since the earriv-ly 19th century Wealthylandowners had little incentive to raisetheir opponentsÕ standard of living, andpeasants saw no reason to improvetheir husbandry as long as those abovethem stood ready to extract whateversurplus they might produce Turnbullrecalls how the annual harvest of man-goes from the village of Marmont, near

St Michel in central Haiti, dwindledfrom $60,000 to nothing in two yearsduring the late 1980s, as farmers cutthe trees to make perhaps $15,000worth of charcoal

The current challenge for Haiti is toset such a self-destructive system on asustainable path The U.S occupationhas mitigated the traditional means ofenforcing distinctions of wealth andpower, but most of the perverse incen-tives are still in place Initial U.S.plans for funneling half a billiondollars of aid to the island havecalled for the standard measuresthat the International MonetaryFund and the World Bank impose

on supplicants: cutbacks in ernment spending (a near oxy-moron here) and removal of tariÝs

gov-on imported goods

This ỊurbanĨ plan, drafted out input from the incoming Min-istry of Agriculture, could be disas-trous, Trouillot warns If importedfoodstuÝs undercut local produc-tion, the poor will become evenpoorer, and the last nonÐÞrewooduse of the land will disappear

with-As Renard points out, however,simply injecting capital into the ru-ral economy could easily do moreharm than goodĐeither reinforc-ing existing inequalities or creatingnew ones Wise investments maydepend on getting people forciblysilenced for nearly 200 years tospeak up and on having a govern-mentĐcurrently as bereft of inde-pendent power as any of its prede-cessors ever wereĐin a position tolisten to them ĐPaul Wallich

DEFORESTATION, carried out by manual ers, reßects the eliteÕs economic shortsightedness

count Peters says the tests cannot be

blamed: they are highly sensitive

In-stead he argues that most of the

nega-tive cases, if investigated, would be

found to stem from bacterial

pneumo-nia and other known causes of

respira-tory distressĐafter all, oÛcials have

es-timated that some 50,000 cases of

res-piratory failure occur in the U.S every

year, and many of these cases are

nev-er adequately explained

But the undiagnosed cases of ent pulmonary syndrome from the FourCorners region and elsewhere intriguePetersÕs co-worker Sherif R Zaki ỊAfter

appar-I get out from under [the Sin Nombreinvestigation],Ĩ he says, Ịmy Þrst plan

is to go back and see what caused thedeaths of these other patients.Ĩ Afterall, it is always possible that yet anoth-

er unknownĐand deadlyĐvirus is onthe loose ĐJohn Horgan

THE ANALYTICAL ECONOMIST

The Wages of HaitiÕs Dictatorship

Copyright 1994 Scientific American, Inc.

Proselytes of virtual reality have

promised a technology that can

immerse participants in

synthet-ic worlds of compelling illusion It

would seem from recent headlinesÑ

witness ÒVR Revolution Looms Larger

Each DayÓ (Business Times), ÒVirtual

Re-ality Finally Getting RealÓ (Orlando

Sen-tinel ) and ÒVR Arrives

HomeÓ (Financial Times)Ñ

that these boosters have

persuaded an initially

skep-tical public that virtual

re-ality has cleared its major

technical hurdles and will

soon hit the mass market

Yet in a report released

this autumn by the

Nation-al Research Council (NRC)

a panel of computer

scien-tists, engineers and

psy-chologists reached quite a

diÝerent conclusion

De-spite the hype

surround-ing the Þeld, the experts

wrote, Òthere is a

substan-tial gap between the

tech-nology that is available

and the technology that is

needed to realize the

po-tential of [VR] systems.Ó

Henry A Sowizral, who

leads a VR research

proj-ect at Boeing Computer

Services in Bellevue, Wash.,

agrees ÒThe three biggest

problems in VR are

perfor-mance, performance and

performance,Ó he quips,

referring to persistent

in-adequacies in the state of

the art for virtual-reality

displays, computers and

software

Fooling a human brain

into believing it is

some-where itÕs not is a tricky

task So far most research has focused

on deceiving the eyes High-resolution,

wide-angle, three-dimensional displays

are one obvious prerequisite; devices

that track the direction of your gaze are

another Yet current VR helmets that

place a miniature liquid-crystal screen

in front of each eye are grainy and

ex-pensive The military spends up to $1

million each for the best, which oÝer the

resolution of a typical desktop

comput-er monitorÑviewed at a distance of

about four inches ÒMost affordableheadsets render you legally blind,Ó Sow-izral says ÒYou canÕt make out the big

E on an eye chart at a virtual 20 feet.ÓAlthough screens will quickly getsharper, it will not be so easy to makelighter helmets, and that bodes ill for

VR explorers At several pounds,

head-mounted displays make it hard to turnyour head Combined with a strictly vi-sual illusion of movement, the weightinduces motion sickness in many wear-ers Nausea and headaches are just thebeginning, the NRC report notes ÒAmore severe problem is the sopitesyndrome This refers to the chronic fa-tigue, lack of initiative, drowsiness, leth-argy, apathy and irritability that can per-sist for prolonged periodsÓ even aftershort gambols through virtual worlds

Current VR tracking systems areeven clumsier than are helmets ÒTrack-ing is the stepchild that nobody talksabout,Ó Sowizral says Mechanical boomsattached to the face and hands are fastand accurate but tend to get in thewayÑespecially when the eyes are cov-ered Magnetic systems that use com-passlike sensors are also popular, Sow-izral reports ÒBut they are susceptible

to interference from anything metalÑlike computers, for example I once put

a Coke can down next to the Þeldsource, and I must have jumped 50 feet

in the virtual environment,Ó he chuckles.Vision aside, virtual environmentswonÕt feel real until you can reach out

and touch them Variouscomputer-controlled de-vices for simulating forceand texture have beentried, but, Sowizral warns,Òunless a few problems aresolved, they may be verydangerous.Ó To create theillusion of a solid wherethere is none requiresbrawny robotic arms thatfollow your hand and re-sist where appropriate.ÒBut if you slam your handdown on a virtual table,the device needs multiplehorsepower motors tomake it feel like youÕve hit

a tabletop,Ó Sowizral serves ÒWell, multiplehorsepower is enough tobreak your arm if some-one has written the pro-gram wrong So people arewimping out and usingmuch smaller forces,Ó mak-ing apparently solid ob-jects actually feel soft andsquishy

ob-While they may feelspongy, these virtual ob-jects will look unrealisti-cally angular and will reactstrangely to touch untilcomputers become morepowerful To the comput-ers that draw virtualworlds, three-dimensionalobjects are composed ofmany two-dimensional polygons Ex-perts estimate that each frame of a VRanimation must contain about 80 mil-lion polygons to appear photorealistic

At least 10 frames per second are

need-ed to sustain the illusion of continuousmotion (Cinematic Þlms run at 24frames per second; television uses 30.)

So any VR system that aspires to visualrealism must be able to compute anddraw at least 800 million polygons persecond

TECHNOLOGY AND BUSINESS

Virtual Reality Check

Imaginary environments are still far from real

IMAGINARY WORLDS may be the province of virtual reality elers But a National Research Council report concludes that the technology cannot yet meet public expectations.

42 SCIENTIFIC AMERICAN December 1994

For about $200,000, Silicon Graphics

in Mountain View, Calif., oÝers a

graph-ics supercomputer, called the

RealityEn-gine2, that can render two million

poly-gons per second under ideal

condi-tionsÑ0.4 percent of the speed needed

for verisimilitude The PixelFlow, a more

costly experimental system under

con-struction at the University of North

Car-olina at Chapel Hill, is expected to reach

about 30 million polygons per second

Virtuality, a video arcade game billed as

virtual reality by W Industries in

Leices-ter, England, renders scenes of just a

few hundred polygons with all the

com-plexity of LegoLand

ÒIf you want photorealism, then for

many environments, a RealityEngine2

is woefully inadequate,Ó states Joshua

Larson-Mogal of the Advanced

Graph-ics Division at Silicon GraphGraph-ics ÒBut

while realism may have something to

do with VR, it is not a necessary

condi-tion by any means.Ó The NRC

commit-tee agreed but added in its report that

drawing is just one part of the work

re-quired of a VR computer

Useful VR applications need more

than just pretty moving pictures Virtual

objects must also mimic the behavior

of their real counterparts, which meansmaking millions of additional calcula-tions each second to ensure that theyact like massive solids rather than mass-less surfaces Add a sense of touch, asmany programs strive to, and the work-load again increases dramatically, sincetextures must be updated hundreds oftimes a second to feel lifelike

The NRC report warns that while search proceeds apace on VR displayhardware, equally important eÝorts onsoftware lag behind Cognitive studieshave shown that separating the sight,sound and touch of an event by a fewtens of milliseconds can cause confu-sion VR researchers have yet to writeoperating software that can guaranteesimultaneous responses from visual,auditory and tactile displays And thejob of Creator is a tough one: ÒIt takesmonths or yearsÓ to create these envi-ronments, Sowizral says The commit-tee recommended that the federal gov-ernment fund Òa major uniÞed researchprogramÓ to develop VR software

re-So why all the hype that VR has rived, when even Larson-Mogal esti-mates that it will be eight to 10 yearsbefore the marginal VR capabilities of a

ar-RealityEngine2reach the consumer ket? Perhaps because researchers tend

mar-to focus on future improvements

rath-er than current limitations and seem tothrive on publicity ÒVirtual reality cur-rently has an extremely high talk-to-work ratio,Ó the NRC report admonish-

es The study also suggests that most

VR researchers are interested primarily

in the graphics software ÒThus, the portance of adequate hardware, with-out which the VR Þeld will never comeclose to realizing its potential, tends to

im-be underplayed by the VR community.Ó More disturbing, the study notes thatscientists in the Þeld seem to have aban-doned their scientiÞc objectivity ÒTheextent to which the usefulness of virtu-

al reality has actually been seriouslyevaluated is vanishingly small,Ó thecommittee concluded Rather than com-paring the cost-eÝectiveness of a virtu-al-reality system with a more tradition-

al approach, the higher-tech solution istoo often simply pronounced better.Letting students swim with virtual dol-phins sounds cool, but taking them to

a real aquarium may be both cheaperand more valuable Sometimes realitydoesnÕt bite ÑW Wayt Gibbs

A cross section of a young beech tree won the 1994

Ni-kon International Small World Competition The

photomi-crographer, Jean Rüegger-Deschenaux of Zurich, colored

the specimen with chrysoldine and astral blue beforeshooting it at a magnification of 40-fold The competitionwas established 20 years ago

1994 Nikon International Small World Competition

Copyright 1994 Scientific American, Inc.

Law and order is coming to the

elec-tronic frontier, by Þts and starts

But hardly anyone, even lawyers,

seems pleased by the prospect One

deeply unsettled, and unsettling,

ques-tion is Ịwhich law?Ĩ Cyberspace

perme-ates nearly every corner of the physical

world, so people who enter it at one

keyboard and monitor may Þnd

them-selves dragged out through another

ter-minal halfway around the country or

the globe to face charges for crimes

they have no idea theyÕve committed

Statutes diÝer from country to country

or even city to city, so cybernauts may

be considered wrongdoers even if their

acts are perfectly legal in their home

jurisdiction

Last year, for instance, a Canadian

sued several American universities for

libel because their computers

transmit-ted derogatory messages about him that

had been broadcast by a British

gradu-ate student The universitiesĐwhich

owned property in England, where libel

laws are stricterĐreportedly settled

in-stead of Þghting

And in July a San Jose couple who ran

an adult bulletin-board system called

Amateur Action found themselves victed for obscenity according to thestraitlaced standards of Memphis, Tenn

con-Law-enforcement oÛcials there dialed

up the system in California,

download-ed pornographic images and had thepair extradited to stand trial The twoface additional charges in Utah

The Amateur Action case will be pealed, but in the meantime bulletin-board operators have already begunpurging their Þles Some Internet ac-cess systems have dropped discussiongroups that might get them in trouble

ap-Karl Denninger of MCS in Chicago says

he has probably lost customers since hestopped carrying Ịalt.binaries.pictures

tastelessĨ and Ịalt.binaries.pictures

erotica,Ĩ but he does not consider thelegal risk to be one worth taking

Denninger and others are more cerned by proposals to regulate text aswell as pictures Senator J James Exon

con-of Nebraska introduced a measure thispast summer that would have given theFederal Communications Commissionauthority to regulate ỊindecencyĨ on thenet, just as it now does for radio, tele-vision and telephone-sex lines The leg-

islation to which it was attached died

in the Senate in October, but observersexpect it to return ỊNobody wants topass the Exon AmendmentĨ because it

is unworkable and probably tutional, says Michael Godwin of theElectronic Frontier Foundation ỊBut if

unconsti-it goes to a vote, theyÕll pass unconsti-itĨ to avoidappearing in favor of pornography.Mikki Barry, an attorney at InterConSystems Corporation and one of thefounders of the Internet Business Asso-ciation, asserts that network access pro-viders should not be responsible forpolicing every Þle that passes throughtheir computers She notes that courtshave long held that booksellers cannotgenerally be prosecuted for libelous orobscene material on their shelves andadvocates similar protection for elec-tronic purveyors

There are, however, some kinds offree speech that net users are Þghting toeliminate An entire Usenet discussiongroup (Ịalt.current-events.net-abuseĨ)

is now devoted to complaints aboutỊspams,Ĩ material posted to dozens orhundreds of unrelated news groups ormailing lists Advertisements for any-thing from software tools to herbalweight loss regularly clutter mailboxes.Often, spammers have paid for ac-cess to the Internet and so cannot easi-Watch Your Electronic Mouth

Cyberspatial speech runs into legal quagmires

44 SCIENTIFIC AMERICAN December 1994

If reputation is everything, Chicago

and its police department may

nev-er completely live down the summnev-er

of 1968, when violent clashes erupted

between oÛcers and demonstrators

Chicago Mayor Richard M Daley, a son

of the Richard J Daley who was mayor

26 years ago, knows that When the

Democrats announced this past summer

that the convention would return to

Chicago in 1996, the younger Daley had

to rush to the defense of the city that

had been the site of what a presidential

commission termed a Ịpolice riot.Ĩ

A few weeks before DaleyÕs remarks,

the Internal Affairs Division of the

Chi-cago police had made public an attempt

at preserving its oÛcersÕ reputations

The division, which looks into

allega-tions of wrongdoing by its own

oÛ-cers, enlisted the help of a software

package that purports to emulate the

way brain cells operate Every 90 days

the division intends to come up with a

list of oÛcers that the software

sug-gests may be headed for trouble

To produce these names, the software

uses a predictive model called a neural

network The program forecasts

wheth-er each of the 12,500 oÛcwheth-ers on the

force is likely to behave in a manner

similar to their nearly 200 colleagues

who were dismissed or resigned under

investigation during the past Þve years

for actions ranging from

insubordina-tion to criminal misconduct The Þrst

list, with 91 names, was to have beendelivered to the personnel oÛce in lateOctober Those oÛcers were to haveenrolled in a counseling program

The seeming ability of neural-networksoftware to extract meaningful conclu-sions from disparate data has resulted

in its use for everything from predictingrecidivism by criminals on probation torecognizing mosquitoes from the sound

of their beating wings The Chicago lice may be the Þrst to employ the tech-nology to anticipate misconduct by law-enforcement oÛcers

po-The network consists of a softwaresimulation of a grid of interconnectedprocessors The processing elementsand the connections among them cor-respond roughly to neurons and syn-apses in the brain Like the brain, thenetwork must undergo a ỊtrainingĨ pro-cess In the police network, input pro-cessors accept personnel informationabout an individual oÛcerĐsuch as cit-izen complaints and traÛc accidentsĐthat have been translated into a series

of numeric values These variables alterthe strength of signals, or synapticweights, that move from one processor

to another The change in weights sets

up a chain of eventsĐfor example, thesignal strengths are multiplied withand added to other values at each pro-cessor The process continues until thenetwork yields values that estimate thelikelihood or not of dismissal The val-

ues are then compared with anothernumber, a zero or a one, that signiÞeswhether the oÛcer being consideredhas, in fact, been Þred or is in goodstanding

If the network has guessed

incorrect-ly, and it usually does initialincorrect-ly, a ematical formula makes a correction tothe weights By exposing the network

math-to hundreds of examples of dismissedoÛcers and those with a clean record,the network continuously adjusts theweights for about half an hour Hence,

it ỊlearnsĨ to make accurate predictionsconsistently

At least that is how things are posed to work Neural networks havetrue disbelievers The police union, forone Relations between internal investi-gators and the union are uneasy even atthe best of times When the union heardabout a computerized brain quietlymulling through personnel Þles to Þndproblem cops, it experienced the insti-tutional equivalent of an aneurysm ỊItÕsabsolutely ludicrousĐit stinks,Ĩ fumesBill Nolan, president of the FraternalOrder of Police in Chicago Nolan saysthe neural network, which he has called

sup-a Ịcrystsup-al-bsup-all thing,Ĩ is merely sup-a tsup-actic

by the department to avoid managingtheir oÛcers ỊYou got a guy slackingoÝ? Supervise him, correct him,Ĩ Nolandemands And he adds: ỊI told them ifthis thing is so good, we should give it

to all the detectives so they can solveall the murders and robberies.ĨNolanÕs impressions do not diÝermarkedly from those of some cognitivepsychologists and computer scientists.ỊVoodoo,Ĩ remarks Zenon Pylyshyn, aprofessor of cognitive science at Rut-gers University ỊPeople are fascinated

by the prospect of getting intelligence bymysterious Frankenstein-like meansĐ

by voodoo! And there have been few tempts to do this as successful as neu-ral nets.Ĩ

at-The criticsÕ main objection is that ral networks are a form of black box:they do not indicate how they arrive at

neu-a conclusion Unlike expert systemsĐanother kind of artiÞcial-intelligencetechnique that makes recommendationsbased on an explicit set of rulesĐneu-ral networks operate by complex non-linear processes ỊA neural networkÕsabilities, as such, reside in connectionweights, a vast numerical table that de-Þes eÝective analysis,Ĩ write Charles X.Ling and A K Dewdney of the Univer-sity of Western Ontario ỊIt is next toimpossible to interpret and understandwhat neural networks of a moderatelylarge size have learned As technology,the art may have promise, but as sci-ence, it fails on this count alone.ĨLing and Dewdney represent one po-

ly be squelched In mid-September, for

instance, a southern California

compa-ny carpet-bombed all Internet mailing

lists beginning with the letters A and

B with a missive that opened, ỊDear

Friend, Since you are someone who

reads E-Mail Ĩ Complaints to Delphi,

the large on-line service where the spam

originated, went unanswered

More devious approaches to on-line

marketing have met with negative

pub-lic responses as well In the jazz

discus-sion group Ịrec.music.bluenote,Ĩ a

con-sultant for Atlantic Records used

sever-al diÝerent names to write a series of

glowing reviews of new releases by the

companyÕs artists The scheme

back-Þred when another net denizen

ex-posed the connection, but the vigilante

briefly lost access to the net because of

his Ịharassment.Ĩ

Rather than broadcasting their

adver-tisements at othersÕ expense, some

com-panies have begun taking advantage of

the World Wide Web, a distributed

hy-pertext system, to let potential

custom-ers come to them With programs such

as Mosaic, users can browse through formation from all over the world; Mo-saic Communications Corporation hasannounced a version that can encryptcommercial information such as a cred-it-card number so it can safely traversethe Internet Net surfers may then beable to buy products as well as justscan on-line catalogues

in-When that day comes, however, Barryforesees more legal headaches No oneknows what jurisdiction these transac-tions will take place in: that of the buy-

er, the seller or the Internet site wherethe productÕs ỊpageĨ is located Al-though sellers may attempt to attachcontract terms to network sales (simi-lar to the Ịshrink-wrap licensesĨ includ-

ed in most commercial software ages), there is no guarantee that courtswill enforce them Says Barry: ỊJudgesand juries have no clue whatÕs goingonĐthey still think the information su-perhighway is about 500-channel cable

pack-TV systems.Ĩ ĐPaul Wallich

Bad Apple Picker

Can a neural network help Þnd problem cops?

Copyright 1994 Scientific American, Inc.

sition in a debate that has continued for

decades Leaving aside hyperbole about

similarities with the human brain,

pro-ponents of neural networks maintain

that the technology is nothing more

than a complicated twist on the

stan-dard statistical method of deducing a

pattern from numeric values by

draw-ing a curve over a set of data points

Moreover, the underlying

methodolo-gies are open to analysis ỊMost of the

time, users donÕt need to know what the

software is doingĐthey only need to

know whether it works,Ĩ says Michael

Mittmann of California ScientiÞc

Soft-ware, which sold the Chicago Police

De-partment its $795 BrainMaker sional software package

Profes-The Chicago police, in fact, found thatabout half of the 91 individuals identi-Þed by the software had already beenplaced in a program set up by the per-sonnel department to counsel oÛcerswho had experienced misconduct prob-lems The software is now intended tocomplement that program, letting theInternal AÝairs Division Þnd oÛcerswho may run into diÛculties before asupervisor does ỊIn departments of 150people or less this software wouldnÕt benecessary,Ĩ says Raymond Risley, the as-sistant deputy superintendent in charge

of the Internal AÝairs Division ỊBut forthe Chicago police, it is pretty muchimpossible for all at-risk individuals to

be identiÞed.ĨCompanies that sell neural-networksoftware may inadvertently add fuel toskepticsÕ arguments California Scien-tiÞc Software cites a number of highlyspeculative uses for the product Cus-tomers claim to have achieved betterthan average results in forecasting win-ners of horse and dog races

The dog track is one thing But

wheth-er BrainMakwheth-er or any othwheth-er neural work can outpoint a grizzled line ser-geant remains to be seen ĐGary Stix

net-The Stone Age literally meets the

space age in John KappelmanÕs

laboratory at the University of

Texas at Austin, where laser beams

bounce oÝ skulls and blasts of x-rays

penetrate ancient bones A computer

monitors the results and compiles

in-formation on the exact,

three-dimen-sional shape of specimens such as the

skeletal remains of long-dead Native

Americans or fossils of even

longer-dead hominid ancestors In a matter of

hours these ancient objects are

trans-formed into data Þles that can be stored

on a CD-ROM or restored into precise

replicas of the original ỊItÕs a very new

and very untried technology,Ĩ man explains eagerly The anthropolo-gist is convinced that the process willtransform his Þeld and resolve some ofthe bitter conßicts that have arisen overissues of ownership and access to relics

Kappel-Such an embrace of high technology

is unusual in a discipline more closelyassociated with notebooks and calipers

ỊArchaeologists and anthropologists ways get to the technology about 20

al-years after everyone else,Ĩ sighs mas R Hester, director of the TexasArcheological Research Laboratory Afew years ago, however, Kappelmanlatched on to the idea that electronics

Tho-might oÝer an easier and more ough way to analyze fossils He did nothave to look far to follow up on the no-tion ỊWeÕre right in the middle of ƠSil-icon Hills,Õ Ĩ he says, referring to thegathering of high-technology compa-nies around Austin

thor-Sensing an opportunity to showcasetheir products and explore new mar-kets, three companiesĐDigibotics, Sci-entiÞc Measurement Systems and DTMCorporationĐare working with Kappel-manÕs group Each Þrm provides a com-plementary piece of equipment Digibot-ics manufactures 3-D laser scanners,

which record the contours of

a specimen by running alaser beam across its surface.The scanners can capture de-tails smaller than a millime-ter across

The second company, entiÞc Measurement Systems,builds computed tomographymachines Computed tomog-raphy is a 3-D x-ray imagingtechnique widely used inmedicine The newest tomog-raphy devices provide enoughresolution to measure suchdiverse details as the wornenamel on a hominid tooth,healed injuries in an ancientbone or subtle aspects of themethods used to make ashard of pottery

Sci-Laser scanning and puted tomography both pro-duce digital data Þles that de-scribe the form of an object.ỊBut weÕre still tactile ani-mals,Ĩ Kappelman reßects.ỊMillions of years of evolu-tion have taught us to learn

com-by touching.Ĩ To satisfy thatneed, he turned to DTM Corporationand a new process known as laser sin-teringĐa way to do 3-D photocopying.Sintering essentially reverses the result

of scanning: a computer-guided laser

Relinquishing Relics

3-D copies of artifacts could stand in for the real thing

LASER SCANNER captures a 3-D image of a pelvic bone from ỊLucy,Ĩ an early hominid The

scanned bone can then be analyzed, animated or replicated by computer.

marks out the shape of the sample and

etches it into nylon or polycarbonate

powder The laser then fuses the

pow-der and builds up a replica

Computer scanning and replication

technologies are still costly and

unfa-miliar to most anthropologists Stephen

Koch, president of Digibotics, says that

even given a discount for universities, a

laser scanner would cost about $30,000

A complete scanning, tomography and

sintering setup might run close to $1

million

So far the university has purchased

only the laser scanner Learning how to

apply engineering analysis techniques

to anthropological research may take

some time In Òthe next six months weÕll

start to see where we can go with this

technology,Ó says Samuel Wilson, who

collaborates with Kappelman Or, in the

more skeptical words of anthropologist

Ralph L Holloway of Columbia

Univer-sity, ÒItÕs not something that makes

you think, ÔOh, God, IÕve got to do this

right away.Õ You donÕt want to end up

with a system that forces you to fritter

your time away.Ó

If the new technologies do realize

their promise, they could assist

muse-ums and universities wrestling with the

need to repatriate Native American

rel-ics With the passage in 1990 of the

Na-tive American Graves Protection andRepatriation Act, all institutions thataccept federal funds are required tohonor requests for the return and re-burial of bones and artifacts The lawhas led to a massive project to cata-logue such relics by the November 1995deadline; it has also engendered con-cern among anthropologists about thetremendous amount of knowledge thatwill disappear into the ground ÒOnceitÕs buried, itÕs gone forever,Ó as Kap-pelman puts it

Laser scanning could oÝer a way toreturn artifacts to their rightful ownerswhile maintaining an electronic simu-lacrum (or an actual model) for futurestudy ÒIn no way is the process destruc-tive,Ó Kappelman insists ÒIt just in-volves shining light on the specimen.Ó

At least one tribe, which has requestedanonymity, agrees: they consented tolet KappelmanÕs group scan bones be-fore reburial The Smithsonian Institu-tion is also experimenting with laserscanning But again the issue of costarises, as many museums express dis-may that the repatriation act is alreadystraining their Þnances Martha Graham

of the American Museum of Natural tory reports that the museum ÒdoesnÕthave the resources to do anything be-yond compliance with the law.Ó

His-Repatriation is only one potential plication The techniques could revolu-tionize education and research bychanging the rules of who gains access

ap-to primary materials A CD-ROM base now being compiled at the Univer-sity of Texas, with the aid of a grantfrom the National Science Foundation,will make images of rare artifacts, frag-ile fossils and extinct primates avail-able throughout the university Studentswill help with the time-consuming laserscanning ÒWeÕve got lots of graduateand undergraduate labor here; weÕll bescanning nearly 24 hours a day,Ó Kap-pelman says

data-And after the CD-ROM project? tually there will be an Internet archive.ThatÕs a few years down the road, butitÕs inevitable.Ó If the price of sinteringdevices falls, even relatively poor insti-tutions could aÝord to buy the devicesand hook them up with a computertied to the Internet An anthropologistcould then call up a Þle over the mo-dem, download it and then print out aperfect replica of a rare fossil

ÒEven-Such is the irony of the forwardmarch of computer technology Even as

it pushes humans steadily into a worldour ancestors would hardly recognize,

it provides a new wayÑliterallyÑto get

a feel for the past ÑCorey S Powell

SCIENTIFIC AMERICAN December 1994 47

Copyright 1994 Scientific American, Inc.

Like grounded albatrosses, the great

bleached jawbones of elephants

past encircle one of Cynthia MossÕs

tents While the veteran researcher sits

in a chair in the shade, two students

examine the shape of each heavy relic

and count its massive molars, looking

for dental conßuence, a sign of age

Moss checks her records, using the

wealth of information about each bone

to describe why one could not

but be that of a 30-year-old

fe-male and why another is clearly

from a 15-year-old male

Sexing and aging such ghostly

jaws can be tricky, but making

such determinations about

free-roving elephants is even harder

Which is why the two young

members of EthiopiaÕs wildlife

service have traveled to Kenya

to visit MossÕs camp in

Ambose-li National Park For more than

20 years, Moss has studied some

1,300 elephants, identifying

ev-ery individual and family Her

Þndings about the social

struc-ture of the community as well as

about communication and

be-havior have changed how many

people perceive the creatures

These insights have, in turn,

posed questions about how

ele-phants should be protected in a

world that increasingly has less

room for them ỊElephants have

a really complex

problem-solv-ing intelligence, like a primate

might have,Ĩ Moss explains in

her characteristic even tones, as

the students poke around and

the wind picks up, cooling the

marshy campsite

Moss herself could be compared to

any one of the ỊtrimatesĨĐresearchers

Jane Goodall, Dian Fossey and BirutŽ

Galdikas, who lived in the rough African

and Asian Þeld for decades observing

chimpanzees, gorillas and orangutans,

respectively MossÕs up-close

behavior-al observations are part of the same

tradition, as is her familiarity with the

elephants and, perhaps more important,

theirs with her This intimacy has made

possible some surprising insights

Moss hastens to point out that her

powers of perception do not come from

a classical scientiÞc education Instead

they come from training at Smith

Col-lege in philosophy and experience in a

world that science sometimes shuns:

journalism Moss, who grew up near theHudson River just north of New York

City, was a researcher at Newsweek

when she decided in 1967 to take aleave to travel in Africa ỊI was alwaysinterested in animals, but I was not awildlife person I was a wilderness per-son,Ĩ she recalls ỊAnd I wanted to go

to one of the last wilderness areas.Ĩ

During her trip, Moss visited the camp

of Iain Douglas-Hamilton, a renownedelephant researcher who had pioneered

a means of using photographs to ognize individual elephants The storyhas it that Douglas-Hamilton was struck

rec-by MossÕs remarks about the elephantsÕbehavior and her attention to detail;

she agreed to work as his assistant forabout a year in Tanzania and, in theend, moved to Kenya Moss continued

to write as she raised money for herown elephant project in AmboseliĐfor

14 years, in fact, she edited the zine of the African Wildlife Foundation,the organization for which she nowworks In 1972 she started her study

maga-Amboseli proved to be an ideal site.Just across the border from Tanzaniaand in full view of Mount Kilimanjaro,the 390-square-kilometer (150-square-mile) park has been relatively untouched

by the poachers who have more thanhalved AfricaÕs elephant population inthe past two decades Between 1973and 1988 the number of elephants inKenya fell from 135,000 to 22,000; inAfrica at large, the population fell from1.3 million to 600,000 between 1979and 1989 Amboseli, which now has apopulation of 830 elephants in 50 fam-ilies, may have been secure because ofthe Maasai The tribe herds cattle on thesurrounding land and has no patiencewith hunters

The continuity of elephantfamilies has permitted Moss tocollect extensive demographicdata ỊThe information on age atsexual maturity and inter-calf in-tervals and other reproductiveparameters is just so valuable forpeople working on other stud-iesĐjust to give them an idea ofwhat is more or less a baseline,ĨMoss notes In the case of thejawbones, the most completedescription has been a 1966 pa-per based on a few elephants,some of unknown age Moss isgradually building up a collec-tion of jaws of elephants whosebirths and histories have beenrecorded Every elephant has sixsets of four teeth that grow for-ward as the previous assemblage

is worn down; when the seriesruns out, often after more than

50 years of grazing, the animalcan no longer forage and dies.The droughts that have oc-curred over the past two de-cadesĐparticularly a severe one

in 1976Đled Moss to anotherdiscovery: elephants can ceasebreeding in response to chang-ing environmental conditions.This Þnding could have implica-tions for wildlife managers seeking toanticipate the population dynamics oftheir herds According to Moss, mostprevious reports about fertility hadcome from culled animalsĐthat is, ele-phants that are killed, often by the hun-dreds, in order to stabilize populations,

a common practice in some protectedareas in southern Africa Informationfrom culled creatures, however, Ịis onlyfrom one point in time,Ĩ Moss says, so

it is not helpful for recognizing terns of fertility and correlating them

pat-to external shifts

Working with other cluding Joyce Poole, formerly of theKenya Wildlife Service, and, currently,

researchersĐin-On the Trail of Wild Elephants

PROFILE: CYNTHIA MOSS

ELEPHANT MAVEN Cynthia Moss has lived in the Þeld for more than 20 years.

Karen McComb of

the University of

SussexÑMoss has

also begun to

inves-tigate the

compo-nents that are out

of the range of

hu-man ears One day,

for instance, Moss

watched a family

crossing a channel

A week-old calf

be-came stuck and

emitted a

low-fre-quency distress

call ÒSuddenly, all

the adults turned

aroundÑthe

moth-er was already in the watmoth-er with itÑ

and ran back to the channel Two of

them started digging the bank outÓ and

rescued the calf, she describes

Some of MossÕs records cannot be

quantiÞed in the way that sounds, teeth

and fertility patterns can be Moss

can-not explain why the animals sometimes

recognize, touch and carry the

desiccat-ed bones of their relatives Or how

pre-cisely older females, such as Echo,

whom Moss recently followed

closely for a Þlm, direct their

families ÒIt is a very diÛcult

concept to be able to describe

scientiÞcally: leadership,Ó Moss

says Elephant families are led

by females and tend, in

Ambo-seli, to have about 11 members;

bulls usually remain loners The

families have very diÝerent

char-acters, which reßect the

person-ality of the matriarch In the

case of Echo, the family is

low-key and nonaggressive

How the 50-year-long

rela-tionships between elephants in

a family or between elephant

families are established and

maintained remains a mysteryÑ

a long-standing one In West

with the Night, pilot Beryl

Mark-ham describes ßying over herds

trying to Þnd a suitable male for

hunters to track She sees one

huge elephant with its head

im-mersed in foliage and circles it

until the others have dispersed

Finally, the elephant moves

away from the tree, only to

re-veal itself as a small-tusked

fe-male And Markham wonders if

the matriarch deceived her expressly

ÒThese animals are diÝerent,Ó Mossconcurs ÒThey are incredibly intelligentand long-lived, and they have complexsocial lives.Ó They are incredibly threat-ened as well In 1988, when it becameapparent that the African populationwas plummeting because of ivory hunt-ers, Moss turned to conservation work

The eÝorts of researchers such as Mossand of Richard Leakey, then head of

the Kenya WildlifeService, resulted in aban in ivory tradethat has dramatical-

ly reduced poachingand hunting.Today, althoughseveral countries areseeking to overturnthe ban, Moss saysthe most pressingthreats to the ani-mals are diÝerent.Elephants are in-creasingly invadingagricultural land that

is right up againstthe borders of many

of the national parks.ÒThere are more peo-ple being killed byelephants now thanbefore,Ó Moss pointsout ÒIt seems virtu-ally every week youhear of someone be-ing killed, and it is not because of re-porting, but because people are right

up against the elephantsÕ ranges.Ó

In MossÕs opinion, the choices arelimited: ÒThose areas obviously cannothave elephants It is not a question ofchoosing elephants over people; youhave to choose people over elephants

So you just have to conÞne elephants

to places where there is going to be noconßict.Ó Such sites would include re-mote areas of the bush, placeswhere people cannot easily thrivebecause of, say, tsetse-ßy infes-tation ÒAnd maybe we shouldjust work on having elephants insome of the parks where thepopulation can be self-regulatingand we do not have to interfere

so much,Ó she adds

For now, Amboseli is tuckedaway from encroaching popula-tions of people, and Moss con-tinues to follow the families, not-ing their patterns of movement,their mating and birthing cycles.Driving out one morning, shespots a family of nine making itsway back into the park after anight of foraging closer to Kili-manjaro The elephants seem lessoverwhelming up close than they

do on the horizon, where theyassume majestic proportions.Like the great birds of the seathat Charles Baudelaire described

in his poem LÕAlbatros, elephants

need that very distance from mans to survive: ÒExiled on earthamid the shouting crowds/ Hecannot walk, for he has giantÕswings.Ó ÑMarguerite Holloway

hu-50 SCIENTIFIC AMERICAN December 1994

FAMILY OF THREE feeds in a lush marsh near MossÕs campsite The mother, Esmeralda, has two calves: Eartha and one not yet named.

JAWBONES of elephants collected in Amboseli

Nation-al Park form an unprecedented database.

Public concerns about health and

safety, the environment and

pe-troleum dependence create

pres-sure to build a better car Although

con-gestion and accidents result from

driv-ing itself rather than from fuel use,

much of urban air pollution, greenhouse

gas emissions and the economic burden

of oil imports can all be tied directly to

fuel consumption Automobile use

con-tinues to grow in the U.S and

world-wide Fuel eÛciency must increase at

least as fast just to prevent fuel-related

problems from worsening EÛciency

must improve even more rapidly to

be-gin to solve these problems

In September 1993 the U.S auto

in-dustry and the Clinton administration

announced a historic partnership to

de-velop vehicles having three times thefuel economy of todayÕs ßeet while pro-viding the same comfort, safety and per-formance Prominent options includeelectric vehicles powered by batteries orfuel cells and hybrid vehicles combining

an electric drivetrain with a combustionengine that might use a variety of fuels

While such alternatives are being ied and tested, however, gasoline anddiesel cars and trucks will most likelydominate the roads for decades to come

stud-They oÝer remarkable reliability, fort and utility at an affordable cost

com-Moreover, they are sustained by an mous economic infrastructure: facto-ries, petroleum reÞneries, service sta-tions and all the people, from auto work-ers to garage mechanics, trained to makethe system work

enor-The vibrant state of automotive neering also contributes to the longevity

engi-of cars powered by the bustion engine Although pioneers likeCarl F Benz and Rudolph C K Dieselenvisioned almost all its potential re-Þnements a century ago, only recentlyhave many of them become practical,

internal-com-as new techniques liberate design andproduction engineers Microprocessors,sensors and electronic controls nowpermit optimization of many opera-tions; materials have become stronger,lighter and more adaptable Computersenable designers to create and improvevehicle models rapidly Many advancesuseful for reÞning conventional carsand light trucks are, in fact, essentialfor alternative vehicles Radically diÝer-

ent approaches may be needed in thelong run, but breakthroughs are notnecessary, because late 20th-century en-gineering capabilities can deliver sub-stantial environmental and economicbeneÞts over the next decade

The eÝort to improve fuel

eÛcien-cy begins by examining how andwhere a car uses energy [see ÒTheAmateur Scientist,Ó page 112] Fuel usedepends on the type of driving as well

as on vehicle characteristics For ple, fuel economy is worse in congestedstreets because of more frequent start-ing and stopping Engineers use theterm Òend-use loadÓ to refer to any as-pect of vehicle operation that consumespower provided by the engine Loads in-clude braking loss, tire resistance, aero-dynamic drag and accessories, such asair conditioning and power steering.The energy needed to meet these loads

exam-is greatly multiplied by the need toovercome losses throughout the drive-train Consisting of the engine, trans-mission and associated components,the drivetrain converts fuel energy intouseful mechanical energy that propelsthe car and runs its accessories Afterthe thermodynamics of combustionand the friction have been accountedfor, only about one sixth of the energyavailable in gasoline remains for theend-use loads Put another way, todayÕsdrivetrains are only 17 percent eÛcient

conventional internal-combustion-powered vehicles

by John DeCicco and Marc Ross

JOHN DECICCO and MARC ROSS have

collaborated for several years on

analyz-ing ways to improve motor vehicle fuel

economy DeCicco is a senior associate

with the American Council for an

Ener-gy-EÛcient Economy, where his research

has focused on the technical

opportuni-ties for reducing energy use and

emis-sions in the U.S transportation system

He received his Ph.D in mechanical

en-gineering at Princeton UniversityÕs

Cen-ter for Energy and Environmental

Stud-ies in 1988 Ross is professor of physics

at the University of Michigan His current

research includes investigating energy

use and emissions of conventional and

alternative vehicle systems Ross received

his Ph.D in physics from the University

of Wisconsin in 1952

SCIENTIFIC AMERICAN December 1994 53

FUEL NEEDS arise in stages (top illustration), starting with the

energy lost by tires, air drag, braking and accessories Energy

requirements are multiplied by 1.11 to overcome

transmis-sion friction and again by a factor of 2.2 to oÝset engine

fric-tion Finally, combustion losses increase the energy demand

by another 2.5 New materials, designs and technologies tom illustration) that minimize losses in the early stages of

(bot-the multiplication process will raise fuel eÛciency

Careful control of engine friction

and of mixture flow through

valves and cylinders decreases

energy losses

New transmission technologyallows for smaller engineswhile maintaining today’sperformance levels

Streamlined shapes andsmooth fits and finishesreduce aerodynamic drag

New materials and designs reduce friction losses in the tires

An electric fan that

runs only when needed

requires less energy

FUEL ENERGY NEEDED

COMBUSTION LOSSES

ENGINE FRICTION LOSSES ACCESSORIES

4 GALLONS

1.86 GALLONS

1.48 GALLONS

0.52 GALLON

0.49 GALLON

0.47 GALLON GALLON

10 GALLONS

Copyright 1994 Scientific American, Inc.

cost design changes, most of which are

found in some models already on the

road Improving the drivetrain by

re-ducing friction oÝers one clear path to

greater eÛciency Reducing end-use

loads presents another Even without

any tinkering with the drivetrain,

modi-Þcations to tires, aerodynamics and

ve-hicle mass will trim a carÕs energy

re-quirements Each unit of energy

sav-ings achieved by lower loads yields six

units of energy savings overall Thus,

load reduction is fundamental

Cutting vehicle mass provides

impor-tant leverage on eÛciency because it

exerts a ripple eÝect A lighter vehicle

requires less power, and so it can be

equipped with smaller drivetrain

com-ponents Consequently, mass drops even

further The current weight of a new car

with the gas tank and radiator Þlled but

without passengers averages just under

3,000 pounds Although downsizing is

an obvious way to reduce mass, we

ex-cluded this option from our analysis

Instead we considered the use of

light-er, stronger materials combined with

reÞned design and manufacturing

tech-niques New materials and better use of

space can reduce mass without

sacriÞc-ing vehicle size and carrysacriÞc-ing capacity

We estimated the degree to which cars

could be made lighter based on these

approaches, adjusting for the weight

added by airbags and strengthened

door panels needed for safety On

bal-ance, applying the best designs available

and adopting new materials can cut as

much as 25 percent from a carÕs weight

Some opponents of fuel economy

reg-ulation assert that decreasing mass

de-creases safety But the protective

bene-Þt of heavier automobiles comes at the

expense of greater damage to people

Cars built with lightweight but strong

materials can shield passengers more

eÝectively than can many heavier

vehi-cles of today yet pose less risk to the

occupants of other cars during collision

Safety is assured largely through better

restraint systems and improvements to

vehicle structure and interior surfaces

that minimize the crash energy

trans-ferred to people in the car Better

crash-worthiness comes not from vehicle size

or mass itself but from features thatsafeguard passengers, regardless of ve-hicle size

Whether a vehicle is massive or light,drivetrain ineÛciencies hurt fuel econ-omy The best opportunities for improv-ing the drivetrain lie in reducing enginefriction, which accounts for about onehalf of fuel use In a carÕs motor, pistonsmove through the cylinders, each dis-placing a certain volume Expanding gas-

es pushing on the piston produce

pow-er The combined volume for all the inders is termed engine displacement

cyl-A larger engine can deliver more powerbut entails greater friction

Rubbing friction occurs among ing parts such as valves, pistons, con-necting rods and the crankshaft Thereare losses in ancillary parts such as theradiator fan and water pump Pumpingfriction occurs when the air and fuelmixture is drawn into the cylinders andthe exhaust is expelled A particularsite of pumping friction is the throttlevalve that controls air intake

mov-ReÞnements in design, ing technique, materials and lubricationminimize rubbing friction Ancillarylosses can be reduced through modi-Þcations such as replacing a belt-drivenfan with an electric fan that runs onlywhen needed Pumping friction can becut by intelligent control of intake andexhaust processes And all these fric-tional losses can be lessened with asmaller engine

manufactur-Studying how frictional work relates

to engine power reveals important ways

to enhance drivetrain eÛciency Poweroutput is reduced by internal friction;

it must meet the needs of the end-useloads plus the transmission Enginefriction is proportional to engine speedand displacement Output, however,does not necessarily depend on thesefactors Technologies that provide need-

ed power while reducing average enginespeed or displacementÑor that eventurn the engine oÝ when power is notrequiredÑoÝer opportunities to cutengine friction while meeting outputrequirements

The value of many eÛciency ments lies in their eÝect on speciÞcpower: the ratio of maximum poweroutput to engine displacement Technol-ogies that enhance speciÞc power per-mit reduced displacement while satisfy-ing vehicle loads Increasing the num-ber of valves improves ßow through thecylinders For example, the speciÞc pow-

enhance-er of four-valve engines avenhance-erages 40percent higher than that of two-valveengines Similarly, overhead camshaftdesigns boost average speciÞc power by

at least 20 percent There are trade-oÝs,such as increased rubbing friction withadded valves Motors with four valvesper cylinder and overhead camshaftsachieve peak power at high enginespeeds, so that compensating changes

in gearing are needed for good ability Successful designs take into ac-count such considerations to yield moremiles per gallon at acceptable cost

drive-Perhaps the most profound engine

reÞnement now being cialized aÝects the control of in-take and exhaust processes Fuel igni-tion takes place within a motorÕs cylin-ders Carefully manipulating the ßow ofthe fuel mixture and exhaust productsthrough the cylinders can boost me-chanical eÛciency In conventional en-gines, when and how far a valve opensdepends on the position of the piston,not on engine speed or load Electronicsensing and control capabilities, togeth-

commer-er with precision manufacturing ods, have made it possible to use vari-able valve control This technique opti-mizes cylinder ßows over a broad range

meth-of conditions Greater valve opening creases maximum power, allowing en-gine displacement reduction Under lowloads, reduced valve opening time canlargely replace throttle operation, there-

in-by decreasing pumping friction

In the past, high cost limited tion of variable valve control mecha-nisms Advanced design and assemblytechniques now permit widespread ap-plication Since the late 1980s Japaneseautomakers have increased their use ofvariable valve control in both Japanand the U.S In 1992 Honda introduced

installa-a notinstalla-able improvement in vinstalla-alve controlthat brought a lean-burn engine to theU.S market

Most contemporary gasoline motors

STYLING IMPROVEMENTS have lowered aerodynamic drag The Rumpler Teardrop

was an early attempt at streamlining Future designs, most likely based on ones

sim-ilar to the Opel Calibra or the Impact, can lower drag by 25 percent or more

normally operate with precisely the

amount of oxygen needed for complete

combustion Lean-burn engines run on

mixtures containing excess air

Advan-tages include reduced pumping losses

and better thermal eÛciency But the

emission of nitrogen oxides ( NOx) from

such engines creates a problem:

cata-lytic reduction of NOx compounds is

diÛcult under lean conditions

Devel-opment of an appropriate catalyst is

an active area of research, because

suc-cess would lead to more general use of

lean-burn technology

Another possible reÞnement, the

ad-vanced two-stroke engine, is also

cap-turing industry attention Two-strokes

accomplish compression and ignition of

the fuel and air mixture in fewer strokes

than do the more conventional

four-stroke engines Fewer piston four-strokes

lead to less frictional loss Lighter and

potentially less expensive than

four-strokes, two-strokes also burn lean

air-fuel mixtures

ModiÞcations to the transmission

along with the engine can bring

impres-sive energy savings Although a carÕs

wheels must cover a wide range of road

speeds, the engine operates most

qui-etly and eÛciently in a relatively

nar-row range of revolutions per minute

The transmission has a range of gearratios to couple the motor to the wheels

so that the motor can run eÝectively atall road speeds To take full advantage

of the beneÞts of engine downsizing,one must design the transmission tomaximize the amount of time the mo-tor operates at high eÛciency

Microprocessors permit engineers toprogram a transmission to optimallymatch engine speed to power require-ments Adding gears to the transmis-sion accommodates more gear ratios,

so that a narrow band of engine speedscan better cover the driving range With

a smaller engine, more frequent gearshifting will be required, and driving intraÛc might feel diÝerent Alternatively,

a continuously variable transmissioncan replace discrete gears with a devicefor smoothly varying the gear ratio Ineither case, careful attention to designand electronic control will help smoothshift transitions and avoid compromis-ing driveability

Using the 1990 new-car ßeet as a

base, we developed a range ofestimates for the feasibility ofincreasing miles per gallon The analy-sis examined the extent to which avail-able technology can be applied to reach

this goal Our mid-range projections donot include lean-burn or two-stroke en-gines, as common use of them is lesscertain because of emissions constraints.After screening technologies for theircost-eÝectiveness, we estimate that by

2005 average new-car fuel economy can

be raised by 65 percent, from 28 to 46miles per gallon A comparable increasecan be made for light trucks, becausetheir energy losses are similar to those

of cars

Raising gas mileage to 46 miles pergallon would add about $800 to the re-tail price of a car Compared with to-dayÕs new-car average of 28 miles pergallon, the higher fuel economy wouldsave 2,100 gallons of fuel over a typical12-year vehicle lifetime, worth $2,500even if fuel prices do not go up Phasingthese improvements into U.S cars andlight trucks over the next 10 years wouldsave 2.8 million barrels a day of gaso-line by 2010 The yearly fuel cost sav-ings to all consumers would be $71 bil-lion, far exceeding the estimated $12billion added annually for the technol-ogy reÞnements

Because we import a growing tion of our oil, the 2.8 million barrels aday of gasoline conserved imply thatU.S oil imports could be cut by at least

frac-SCIENTIFIC AMERICAN December 1994 55

FOUR-STROKE CYCLE powers most of todayÕs cars Advanced

designs enable control of air and fuel that ßow into the

cylin-der Decreasing the work to pump gases in and out of thecylinder provides further opportunity to conserve energy

3

POWER

MIXTURE IGNITES,AND GAS EXPANDS

4

EXHAUST

EXHAUST GASESARE EXPELLED

CRANKSHAFT

EXHAUSTVALVESPARK

Copyright 1994 Scientific American, Inc.

two million barrels a day in 2010 These

savings are much larger than the

sup-plies that might be obtained by

exploit-ing reserves oÝshore or in the Arctic

National Wildlife Refuge Moreover, such

oil savings would be achieved with

re-duced rather than increased

environ-mental damage

Reduced fuel consumption brings

additional environmental beneÞts

Car-bon dioxide emissions are proportional

to fuel consumption, so higher fuel

economy means lower greenhouse gas

emissions The amount of hydrocarbon

vapors released into the air is also tied

to gasoline use, so increased eÛciency

reduces their impact as well

Hydrocar-bons react with nitrogen oxides to form

ground-level ozone, a major air

pollu-tant that aggravates asthma and causes

other respiratory problems Because

higher eÛciency pays for itself through

fuel savings, there is no added cost for

the associated reductions in carbon

dioxide and hydrocarbon releases

Better emissions-control technology,

apart from advances in fuel economy,

can lead to further large reductions in

air pollution Extensive industry and

regulatory eÝorts are under way in this

area Unfortunately, progress has been

much slower than expected because of

a lack of real-world data analysis Wewould be more optimistic if pollution-control eÝorts were more solidly based

on fundamental science and signed observations

well-de-Higher fuel economy for cars and

trucks yields broad economicbeneÞts as well Money spent

on oil imports is mostly lost to the U.S

economy, and gasoline purchases vide relatively few jobs per dollar spent

Because enhanced fuel economy duces savings for consumers, they havemore money to spend on goods and ser-vices other than gasoline That stimu-lates domestic industries, including autoproduction, resulting in employmentgains During congressional delibera-tions, U.S auto manufacturers claimedthat raising mileage standards wouldlead to employment loss Although thatmight be conceivable if higher fuel econ-omy were obtained by rapidly mandat-ing smaller vehicles, it is not true for aphased-in, technology-based approach

pro-For a scenario similar to that describedhere, our economic modeling shows anet increase of 100,000 to 250,000 U.S

jobs by 2010

Most of the technologies we haveconsidered appear in cars already onthe road Although higher fuel economy

is clearly cost-eÝective in the long run,there is little market interest in apply-ing better technologies for cutting en-ergy consumption Gasoline prices are

at an all-time low So manufacturers stead concentrate on applying engineer-ing advances to enhance vehicle perfor-mance or luxury, through increased sizeand weight, rather than to provide bet-ter mileage High-performance and lux-ury models dominate the more proÞt-able segments of the market Amongthe models offered for sale in a givenyear, the more fuel-eÛcient ones tend

in-to be the smaller, slower, botin-tom-of-the-line vehicles

bottom-of-the-More fuel-eÛcient cars and truckswould sell well under diÝerent condi-tions, which could be brought about bysuch factors as national policies ( fueleconomy regulation, vehicle-pricing in-centives or dramatically higher fuel tax-es) or international events (wars or car-tel decisions to limit the oil supply) Thewidespread beneÞts of reducing gaso-line consumption justify public policiesdesigned to put more eÛcient vehicles

on the road

Estimating the cost of higher fuel economy is difficult

because information on manufacturing is not

general-ly made public The authors developed an economic

mod-el using published reports that examined prices of

avail-able and useful technology Assessing how extensively

each potential refinement could be used in new cars and

light trucks, the researchers ranked the technologies

ac-cording to their costs as amortized over the average

vehi-cle lifetime of 12 years Although this treatment does not

apply to any particular car, it provides a reasonable idea

of the average expense of improving cars in general Themid-range, or moderate, cost-effective level of 46 miles

per gallon (left ) was estimated by determining the curve’s

intersection with expected gas prices in 2010

Without a change in U.S policy, auto fuel use is casted to rise along the projection shown in the graph at

fore-the right (light brown) The shaded band predicts gas use

if the technologies for increased fuel economy are phased

in over the next 10 years The moderate estimate (dark brown) corresponds to 46 miles per gallon.

The Cost of Improving Fuel Economy

9

1970 1975 1980 1985 1990 1995 2000 2005 2010

PROJECTION BASED

ON CURRENT TRENDS

PROJECTIONS BASED ON MORE FUEL-EFFICIENT CARS

MODERATEREDESIGN

RADICALREDESIGN

EXPECTED GAS

PRICE IN 2010

25

AVERAGE FUEL ECONOMY OF NEW CARS

(MILES PER GALLON)

3.0

Consumers and automakers both face

a dilemma when it comes to increasing

mileage In todayÕs market, individual

customers have little interest in

forgo-ing better performance or luxury for

higher fuel economy because their

di-rect fuel savings are not compelling As

concerned citizens, however, many want

to see energy and environmental

prob-lems solved Approaches in which

ev-eryone participates, rather than those

that rely only on the choices of

individ-ual new-car buyers, can eÝectively

re-spond to public concerns despite

mar-ket disinterest

Similarly, a manufacturer that applies

engineering advances to increase

mile-age might risk losing customers,

where-as another that uses the same

reÞne-ments to boost performance would

probably fare better in todayÕs market

Regulations that give all automakers an

incentive to raise fuel economy can

overcome the risk faced by an

individu-al manufacturer acting individu-alone

Strength-ening the ßeet-average fuel economy

standards would give such an incentive

while also oÝering design ßexibility The

industry can use diÝerent approaches

for each vehicle line and ensure that the

overall goal of reducing fuel

consump-tion is met

To enhance market interest, standards

can be usefully complemented by

spe-cial incentives, such as an expanded

gas-guzzler tax and rebates on vehicles that

are more eÛcient than average

Stan-dards and incentives must apply

equi-tably to manufacturers, so that all face

similar pressures to increase fuel

econ-omy The risk to any one Þrm would

then be minimized

Some economists point out adverse

side eÝects of fuel economy regulation

Higher eÛciency lowers the cost of

driv-ing, so people drive more and partly

oÝset the savings Some therefore

con-clude that raising gasoline taxes is a

preferred approach for reducing

gaso-line consumption Empirical evidence

in-dicates, however, that such eÝects only

fractionally oÝset the beneÞts of

regu-lation The price of gasoline aÝects the

amount of driving far less than might

be expected Parking pricesÑor lack

thereofÑand road building have much

more inßuence Thus, a higher gasoline

tax can be helpful and may be justiÞed

for a number of reasons (current taxes

do not even fully cover highway costs),

but taxation alone is a weak lever for

controlling fuel consumption

The time may come when

convention-al gasoline cars and light trucks will

have to be replaced by fundamentally

new designs This eventuality justiÞes

research and development eÝorts

to-day But more eÛcient conventional

ve-hicles offerÑsooner rather than laterÑlarge, tangible beneÞts Many of theseadvancements, especially load-reduc-tion measures, are essential steps onthe way to the next generation of vehi-cles that will use electric drivetrainsand fuel cells

The average mileage of new light hicles has been stagnant for a dozenyears Lack of market interest, not lack

ve-of technology, is the most serious stacle to tackling automotive fuel use

ob-Enacting stronger standards and otherincentives for higher fuel economy calls

for public policy leadership Comparedwith todayÕs new cars, reÞned autoswould be the same size, with the samecarrying capacity and acceleration abil-ity; they would be lighter, more aero-dynamic and have greater crashworthi-ness They would also have lower emis-sions and better mileage The bene-ÞtsÑdirect consumer savings, lower oilimports, reduced hydrocarbon andgreenhouse gas emissions and higheremploymentÑindicate that increasingfuel economy is one of the best invest-ments the country can make

SCIENTIFIC AMERICAN December 1994 57

FURTHER READING

AUTOMATED WELDING SYSTEMS are crucial to the assembly of vehicle bodies novations in the manufacturing process, such as use of laser welding that leavesjoints smooth, will help raise energy eÛciency of new models

In-DESIGN, EFFICIENCY, AND THE PEACOCKÕS

TAIL Paul B MacCready in Automotive Engineering, Vol 100, No 10, pages 19Ð

21; October 1992

COSTS AND BENEFITS OF AUTOMOTIVE

FUEL ECONOMY IMPROVEMENT: A TIAL ANALYSIS David L Greene and K G

PAR-Duleep in Transportation Research, Vol.

27A, pages 217Ð235; May 1993

ACHIEVING ACCEPTABLE AIR QUALITY:

SOME REFLECTIONS ON CONTROLLING HICLE EMISSIONS Jack G Calvert, John B

VE-Heywood, Robert F Sawyer and John H

Seinfeld in Science, Vol 261, pages

37Ð45; July 2, 1993

TRANSPORTATION AND GLOBAL CLIMATE

CHANGE Edited by David L Greene andDanilo J Santini American Council for

an Energy-Efficient Economy, Berkeley,1993

AN UPDATED ASSESSMENT OF THE TERM POTENTIAL FOR IMPROVING AUTO-MOTIVE FUEL ECONOMY John DeCiccoand Marc Ross American Council for anEnergy-Efficient Economy, Berkeley, 1993.AUTOMOBILE FUEL CONSUMPTION AND

NEAR-EMISSIONS: EFFECTS OF VEHICLE ANDDRIVING CHARACTERISTICS Marc Ross in

Annual Review of Energy and the ronment, Vol 19, pages 75Ð112; 1994.

Envi-Copyright 1994 Scientific American, Inc.

The Gobi Desert of Central Asia is

one of the earthÕs desolate

plac-es Its million square miles of

sand dunes, sculpted badlands and

saw-toothed mountains are alternately

scorched by summerÕs high-latitude sun

and frozen by winterÕs Siberian winds

It is not a place to explore unprepared:

crossing vast, uninhabited areas

be-tween a sprinkling of oases requires

careful planning akin to the siege tactics

for scaling a Himalayan peak or

travers-ing the Antarctic continent There are

few maps, and modern satellite

naviga-tion is of limited help to a traveler

try-ing to choose among deeply rutted,

wildly crisscrossing roads that wander

as unpredictably as the nomadic ments they connect Even a modern ex-pedition runs the risk of water, fuel andfood shortages Getting lost is not mere-

settle-ly frustrating but a matter of seriousdanger

Yet the Gobi is a paradise for tologists Its eroding terrain exposesnearly complete skeletons of creatureshitherto known only through painstak-ing reconstructions from a few scat-tered bones Our expeditions, jointlysponsored by the Mongolian Academy

paleon-of Sciences and the American Museum

of Natural History, have excavated nosaurs, lizards and small mammals in

di-an unprecedented state of preservation

Freshly exposed skeletons sometimeslook more like the recent remains of acarcass than like an 80-million-year-oldfossil The skeletons and skulls we havefound are often complete or nearly com-plete, in sharp contrast to the ÒsparepartsÓÑfragmentary jaws, teeth and iso-lated bonesÑthat paleontologists typi-cally recover elsewhere

No one knows why fossils in the Gobiare so well preserved In other speci-men-rich areas, such as the one thatbecame the Rocky Mountains, streams

or rivers carried animal remains to sil sites, disarranging them along theway The late Cretaceous environment

fos-in the Gobi, however, may have beenmuch as it is today: open valleys ofsand dunes and cliÝs, sparsely watered

by small, seasonal lakes or streams deed, indications of ancient sand dunescan be observed in rock sections there

In-It is also apparent that the animalswere buried very soon after their death,before scavengers or weather had muchtime to get at them Poorly sorted layers

of sandstone in the Cretaceous rock mations suggest deposits of the kindone would expect in violent sandstorms;Tomasz Jerzykiewicz of the GeologicalSurvey of Canada in Calgary and hiscolleagues have studied fossil beds inChinese Inner Mongolia and found thatvertebrate fossils are often embedded

for-in these layers Such storms might notmerely have buried carcasses but killedanimals as well Entombed in a matter

of minutes or hours, their remainsemerge some 80 million years later, al-most undisturbed

Mongolia was not always

recog-nized for its bounty of toric material During the late19th and early 20th centuries, the RockyMountain region of western North Amer-ica was the mecca for vertebrate pale-ontologists Then, in 1922, Roy Chap-man Andrews, a scientist from theAmerican Museum of Natural History,

prehis-MICHAEL J NOVACEK, MARK NORELL,

MALCOLM C MCKENNA and JAMES

CLARK have together explored fossil

sites in the Gobi Desert under the

aus-pices of the American Museum of

Natu-ral History and the Mongolian Academy

of Sciences Novacek, Norell and

McKen-na are curators in the department of

ver-tebrate paleontology at the museum;

No-vacek is also vice president and dean of

science there, and McKenna holds a

pro-fessorship in geology at Columbia

Uni-versity Clark, who worked at the

muse-um for three years, is now an assistant

professor of biology at George

Washing-ton University

Fossils of the Flaming CliÝs

Mongolia’s Gobi Desert contains one of the richest assemblages of dinosaur remains ever found Paleontologists

are uncovering much of the region’s history

by Michael J Novacek, Mark Norell, Malcolm C McKenna and James Clark

led an expedition into the heart of the

Gobi and changed the geography of the

fossil world He never attained his

pri-mary objectiveÑa search for the fossil

origins of humans in Central AsiaÑbut

a series of spectacular, more ancient

discoveries soon diverted the interests

of the scientiÞc team The Gobi held an

extraordinary treasure of dinosaurs,

mammals and other vertebrates whose

richness is undiminished even today

Andrews chronicled his Þve

expedi-tions in a remarkable narrative entitled

The New Conquest of Central Asia The

romance and excitement of the

enter-prise foreshadowed the exploits of the

movie character Indiana Jones En route,

the explorers were challenged by

track-less dune Þelds, raging sandstorms and

marauding bandits AndrewsÕs caravan

of camels and spindly-wheeled Dodge

motorcars was a logistic nightmare,

ex-tending like the Algoy horkhi horkhi,

the legendary (and probably mythical )

Mongolian sandworm, across the

moon-scape of the Gobi

One of the most important

discover-ies in the history of scientiÞc

explora-tion came in the midst of such

diÛcul-ties Late in the Þrst Þeld season of

1922, the expedition got oÝ track on a

vast plain just north of the Gurvan

Sai-chan Mountains Hopelessly lost,

An-drews ordered the party to stop near a

ger (the domed-shaped tent of Central

Asian nomads, also known as a yurt)

While Andrews sought directions from

the frontier soldiers occupying the ger,

team photographer J B Shackleford

wandered toward an unassuming rock

rim at the edge of a Þeld There he was

startled to Þnd a fantasy of red cliÝs

and spiresÑand fossils

Within 10 minutes he had uncovered

the Þrst known skull of the

Protocer-atops, a parrot-beaked, shield-headed

dinosaur that has since become a

refer-ence fossil of the late Cretaceous of

Central Asia Lingering for the rest of a

warm afternoon, the crew recovered

more bones and even a small egg, whichthey mistook for that of a bird Theyreturned the next summer to Þnd an ex-travagance of dinosaurs, ancient mam-mals and other vertebrates, as well asthe Þrst known cluster of dinosaur eggs

Their Þndings, particularly the eggs, came front-page news from New York

be-to New Caledonia Andrews named theplace the Flaming CliÝs, inspired by themagniÞcent red-orange glow as the sandcliÝs blazed in the late afternoon sun

By the beginning of the 1930s drews, frustrated by a volatile, less sym-pathetic political scene in Mongolia,gave up his exploration The Gobi wasinaccessible to Western interests formore than 60 years, leaving Soviet-blocscientists to extend the work Andrewshad begun Between 1946 and 1949,joint Russian-Mongolian expeditionspenetrated the Nemegt basin, a region

An-of awesome desert beauty whose moteness had deÞed AndrewsÕs at-tempts to explore it They uncoveredrich badlands of Cretaceous and Ceno-zoic fossils there

re-ZoÞa Kielan-Jaworowska, a nowned fossil mammal specialist, now

world-re-at the Paleontological Museum of theUniversity of Oslo, led a highly skilledand energetic Polish-Mongolian team tothe Nemegt and other areas between

1963 and 1971 She and her colleaguesproduced a series of classic scientiÞcmonographs and a magniÞcent display

of dinosaurs and other fossil vertebrates

at the Natural History Museum in theMongolian capital of Ulan Bator Sincethe 1960s Mongolian paleontologistshave conducted extensive Þeldwork bothindependently and in collaboration withSoviet (now Russian) scientists

Westerners Þrst returned to the try after the development of Mongoliandemocracy in 1990 That summer ourcolleagues at the Mongolian Academy

coun-of Sciences invited us for a sance that paved the way for four moreambitious expeditions during succeed-

reconnais-ing years Japanese, German and otherAmerican parties have also begun Þeldprojects Soon Mongolia may be tram-pled by a stampede of bone hunters,but meanwhile we feel fortunate to bethe Þrst Westerners to resume the ad-venture that Andrews inaugurated

If anything, the contrast between

the Gobi and other, more accessiblefossil areas has increased since An-drewsÕs time A century ago, in the glo-

ry days of dinosaur hunting in theAmerican West, prospectors encoun-tered valleys and canyons where skele-tons were exposed like corpses on adeserted battleÞeld, but today manyprime dinosaur hunting grounds appearnearly exhausted Modern paleontolo-gists look back with envy at predeces-sors who roamed that virgin territory.The cumulative activities in Mongoliaover the past 70 years, in comparison,

do not approach those in the Americas.Erosion is still exposing a wealth offossils even at sites well mined by An-drews and others Moreover, the verydiÛculty and unexplored nature of theGobi increases the chance that paleon-tologists may yet stumble onto whollyunexplored pockets of badlands.Early in the 1993 season, with ourMongolian colleague, Demberelyin Dash-zeveg of the Mongolian Academy ofSciences, our Þeld party struck out for

an undistinguished set of red-brownsandstones on the north side of the Ne-megt Valley, near the base of a jaggedmountain range called Gilbent Uul Pre-vious expeditions, Dashzeveg said, hadignored this region in their rush toreach the more dramatic badlands of

SCIENTIFIC AMERICAN December 1994 61

SANDSTONE ESCARPMENT in southernMongolia made headlines in the early1920s, when paleontologists found dino-saur eggs there Seventy years later theFlaming CliÝs continue to yield a richlode of well-preserved fossils

Copyright 1994 Scientific American, Inc.

the western Nemegt Valley We arrived

at the area, struggled for a few miles

along a wash and established a bivouac

where our heavy gasoline tanker and

trailer buried itself in the sand

The next morning we started

pros-pecting the hills and gullies nearest

camp Within hours it was clear that we

had come across one of the richest

concentrations of fossils ever

recov-ered from the age of the dinosaurs In

a basin less than two kilometers across,

we found scores of dinosaur skeletons

and egg nest sites weathering on gentle

slopes Intermixed with the dinosaur

fossils were abundant smaller bratesÑlizards and mammalsÑthatwere also key members of the ancientCretaceous ecosystem

verte-The local name for the site of this nanza is Ukhaa Tolgod (ÒBrown HillsÓ)

bo-Its natural amphitheater containedroughly 100 readily visible dinosaurskeletons, many of them in nearly pris-tine condition During the past two Þeldseasons, we have selected the most de-sirable specimens Among them are 25skeletons of theropod dinosaurs Thisgroup of agile carnivores runs the gam-

ut from the enormous Tyrannosaurus

and Allosaurus through fast-running dromaeosaurs such as Velociraptor (the villainous predator of Jurassic Park, a

title some 60 million years out-of-date)

to smaller birdlike creatures such as theoviraptorids We also gathered an un-precedentedly rich collection of smallvertebrates: more than 200 skulls ofmammalsÑmany with their associatedskeletonsÑand an even greater number

of lizard skulls and skeletons

As the variety of our specimens

makes clear, the ßowering of restrial life during the Creta-ceous of Central Asia was not limited todinosaurs The Gobi of 80 million yearsago supported a wide variety of lizards,crocodilians and mammals We havefound specimens representing morethan 30 species of lizards; some are ex-tremely well preserved and display ana-tomical features that oÝer clues to therelations among major lizard families.Probably the most spectacular of

ter-these is Estesia Early one morning

dur-ing our reconnaissance in 1990, wecame on an exquisite, eight-inch-longskull with knife-edged teeth half em-bedded, like a bas-relief, in a vertical slab

of sandstone At the time we thought itbelonged to a small carnivorous dino-saur, but later examination determinedthat the skull was that of a wholly newkind of large predatory lizard, closelyresembling the Komodo dragon alivetoday We named the species after the

OVIRAPTORID, a large predatory dinosaur, stands near its nest

by the bodies of two young velociraptors (another

ßeet-foot-ed prßeet-foot-edator ) The authors found an oviraptorid nest that tained two skulls of infant dinosaurs of the same family as

con-ADULT OVIRAPTORID SKULL was found at Ukhaa Tolgod in the western Gobi

This birdlike family of dinosaurs bore a resemblance to modern ostriches Some

oviraptorids (perhaps only of one sex) grew a large bony crest after they matured

late Richard Estes of San Diego State

University, the worldÕs foremost

au-thority on fossil lizards

Estesia is a very primitive animal and

as such is signiÞcant for understanding

the family tree of the varanoid lizards

(the group that includes the Komodo)

The skull has an unusual series of

ca-nals at the base of the teeth that

sug-gests Estesia injected poison into its

prey This lethal weapon is not

com-mon to living varanoids but is found in

the Gila monster of the southwestern

U.S and northern Mexico

We have since found fragments of

Estesia in other sites where smaller

liz-ards, tiny mammals and dinosaur

egg-shells are common Modern varanoids

are noted for their voracious and

wide-ranging appetites It is likely that

Este-sia ate smaller vertebrates, small

dino-saurs and possibly dinosaur eggs

Although much of the Cretaceous

Gobi was dry, water must have been

abundant in at least a few places and

times We found occasional fossils of

turtles, usually associated with aquatic

habitats ( Turtle shells, skeletons and

bone fragments are abundant in rock

formations from the North American

Cretaceous, where most evidence

sug-gests an ancient environment of ponds,

streams, mudßats and deltas.) At one

site, in a chromatic badlands west of

the Nemegt Valley, a small depression

roughly the size of a wading pool held

shells and skeletal parts of more than

50 individuals representing two turtlegenera

Some of the greatest treasures of

the Gobi Cretaceous are easy tomiss when scanning the slopesand gullies: the tiny skulls and skele-tons of mammals These fossils repre-sent important precursors of the greatmammalian radiation that followed theextinction of the dinosaurs at the end

of the Mesozoic

The bulk of scientiÞc information onthese earlier mammals comes fromNorth American fossils, which are most-

ly fragmentary jaws and teeth In fact,there are virtually no complete skulls ofthese Cretaceous mammals from NorthAmerica As a result, the Gobi assem-blage, including our Þnds and those ofearlier expeditions, surely representsthe worldÕs reference collection for lateCretaceous mammals

A small block recovered from UkhaaTolgod in 1994 revealed six shrewlikeplacental mammals, each only a fewinches long Amazingly, the fossils con-sist of complete skulls attached toskeletons; such tiny bones are usuallyhighly vulnerable to disarticulation and

SCIENTIFIC AMERICAN December 1994 63

Velociraptor; the interlopers might have been raiding the nest,

they might have been brought by a parent oviraptorid to feed

its young, or they might even have been laid in the nest reptitiously (as the cuckoo does today) and incubated there

sur-OVIRAPTORID EGG (left ) contains an almost perfectly preserved embryo The skull (right ) of a young dromaeosaur (the family of predators that includes Veloci- raptor) was unearthed in the same nest; it is not clear how the interloper arrived.

Copyright 1994 Scientific American, Inc.

breakage These small creatures were

probably buried and preserved rapidly

after they had died

We have found two basic groups of

mammals The Þrst is the

multituber-culates, or Òmultis,Ó as paleontologists

call them They are a curious array of

animals with long front incisors and

molars with a complex of bumps

(tu-bercles) on the tooth crowns The

Mon-golian Cretaceous multis oÝer by far

the best accumulation of skeletal

mate-rial for examining the relations of these

creatures with other mammal lineages

Multis can be thought of as the

ro-dents of their times, even though they

are in fact only distantly related to

mod-ern groups of mammals Their

rodent-like adaptations are a sign of convergent

evolution with the rats, mice and

squir-rels familiar today The multis thrived

through the Þrst several million years of

the Tertiary period, after the dinosaurs

had died out They then dwindled in

number and disappeared, replaced by

more recent groups of similar habits

The second group is the therians,

an-cestors of both marsupials and modern

placental mammals (a category ranging

from whales to bats, aardvarks and

hu-mans) These early therians consist of

half a dozen shrewlike forms whose

traits oÝer clues to the origins of later

members of the group Fossils from

the genus Deltatheridium, for example,

seem to straddle the line between

mar-supials and placentals

Other species point to a more

primi-tive age of placental mammals Modern

species have at most four premolarteeth on each side of the jaw, but cer-tain Mongolian specimens of placentalmammals, such as juvenile individuals

of the genus Kennalestes, have at least Þve Another group, Zalambdalestes, is

interesting because it has rabbitlike

or rodentlike incisors and a skeletonadapted for running and hopping, alsolike that of living rabbits Paleontolo-

gists are divided on whether

Zalamb-dalestes might be an early rabbit

ances-tor or simply an example of convergentevolution

One of the most spectacular prizes ofour expeditions is the best-preserved

skull of Zalambdalestes yet recovered.

In collaboration with Timothy Rowe ofthe University of Texas, we examined itusing a very high resolution computedtomography scanner The three-dimen-sional x-ray images allowed us to re-construct the paths of arteries, veinsand even nerves The CT images con-Þrm an earlier hypotheses by Kielan-Ja-worowska: the carotid arteries, the mainchannels supplying blood to the brainand the eye, enter the skull along themidline rather than at the sides, as they

do in most living mammals

Mammals, lizards and other brates are crucial to reconstructing thepast environment of the Gobi and totracing the main lines of evolution Butdinosaurs still occupy center stage inthe public eye The Cretaceous Gobi isunquestionably one of the worldÕs greatdinosaur hunting grounds The fossils

verte-range from complete skeletons of

Tar-bosaurus, a Þerce carnivore closely

re-lated to the North American

Tyranno-saurus, to giant sauropods, duck-billed

dinosaurs, armored ankylosaurs, frilled

ceratopsian dinosaurs such as

Protocer-atops and a magniÞcent assemblage of

smaller carnivores Birdlike oviraptorids

and dromaeosaurs such as Velociraptor

are better represented in the stratiÞedrocks of the Gobi than anywhere else inthe world

These remains have given rise to troversies but also to some deÞnitive

con-conclusions Artists often depict

Veloci-raptor hunting in packs like African

wild dogs, for example, but there is nodeÞnitive proof that it was capable ofsuch cooperative behavior The preda-

torÕs taste for Protoceratops, however,

is more than a matter of speculation Inthe late 1960s a group of Polish andMongolian scientists at Tugrugeen, awhite sandstone escarpment about 50miles west of the Flaming CliÝs, exca-vated one of the most remarkable pair

of specimens in the history of ogy Two nearly complete skeletonsÑa

paleontol-Protoceratops and a VelociraptorÑare

preserved locked in mortal combat

Ve-lociraptor desperately clutches the

low-ered head of Protoceratops with its

forelimbs and raises the killing hooks

of its hind claws high against its preyÕsßanks The ÒÞghting dinosaurs,Ó whichmay have met their end in one of theGobiÕs sandstorms, are one of the greatexhibits of the Natural History Museum

in Ulan Bator

Velociraptor skeletons are not only

ROY CHAPMAN ANDREWS (left ) led the Þrst fossil-hunting

ex-pedition to the Gobi Desert in 1922 His group, which included

both camels and primitive cars, became lost on several sions; one such episode sparked the discovery of the Flaming

occa-fascinating for the image they convey

of intelligent, swift and lethal terrors

They oÝer clues to the evolutionary

con-nection between birds and dinosaurs

Velociraptor and its relatives have many

very birdlike features, including the

construction of the bony case of the

brain and the design of the elongated

limbs and digits A nearly complete

skeleton of Velociraptor unearthed at

Tugrugeen in 1991 has a more

com-plete braincase than any other

speci-men; in its details the architecture of

the braincase is surprisingly similar to

that of modern birds

An unexpected discovery at

Tugru-geen in 1992 further ampliÞed the

pro-posed connection between dinosaurs

and birds We found a delicate skeleton

that was identical, except for its

small-er size, to one discovsmall-ered by Mongolian

scientists some years earlier The

ani-mal, roughly the size of a turkey, has a

remarkably gracile frame with long

legs In addition, the keel of the

breast-bone is very well developed In modern

birds, strong pectoral muscles that

pow-er the downstroke of the beating wing

attach to this keel Instead of long wing

bones, however, this creature has

stub-by, massive forelimbs somewhat like

those of a digging mole The end of the

arm and hand is appointed with a

sin-gle very large claw; hence, the scientiÞc

name bestowed on the animal is

Mono-nykusÑliterally, Òsingle claw.Ó (The

orig-inal spelling was the more

etymologi-cally correct Mononychus, but it turned

out that a beetle had Þrst claim.)

Mononykus is a bizarre creature

Al-though it has no wings, it has severalfeatures that suggest a closer relation

to modern birds than the famous

prim-itive bird Archaeopteryx In addition to

the enlarged sternum, these featuresinclude an antitrochanter, a small pro-trusion on the pelvis at the hip joint thatserves as a muscle attachment point, acontinuous crest on the femur for theattachment of the limb muscles, and agreatly shortened Þbula, the thinner ofthe lower limb bones A detailed analy-

sis of Mononykus favors the view that

this creature was a ßightless relative ofmodern birds

That argument has drawn some

criti-cism Certain specialists claim

Monony-kus is simply a small dinosaur whose

birdlike features are a product of vergent evolution The weight of the ev-idence, however, does not favor conver-gence The history of birds is marked

con-by species (such as the ostriches, emusand kiwis) that have lost ßight Our

Mononykus fossils do not show evidence

of feathers, but it is only by some cle of preservation that the Þne Juras-

mira-sic limestone entombing Archaeopteryx

leaves impressions of tiny feathers

Mononykus, like most fossils, is not

pre-served in such unusual rock

We have detected remains of this imal at many localities Among the sev-eral skeletons most recently recoveredfrom Ukhaa Tolgod is a nearly com-plete specimen that includes for theÞrst time a well-preserved skull Thisfossil, though not yet fully exposed and

an-prepared in the laboratory, shows dence of an elongated head What wecan see is much diÝerent from previousreconstructions, which were extrapolat-

evi-ed from partial fragments of the case New information from this skullshould have critical bearing on the cur-

brain-rent debate over Mononykus and the

dinosaur-bird connection

Eggs of both dinosaurs and birds,

found in many parts of the Gobi,add another dimension to thefossil record Some of the eggs containsmall embryonic skeletons of the bird

Gobipteryx, and others preserve the

skeletal remains of a small embryonicdinosaur In some places, several nestsmay be clustered on a hillside, and weinfer that these nests mark a congrega-tion of dinosaurs, much like a colony

of seabirds today

At Tugrugeen Shireh, we found 12

jumbled skeletons of Protoceratops on

a ßat not much larger than a puttinggreen The Sino-Canadian team has also

reported such accumulations of

Proto-ceratops in Cretaceous rocks of

north-ern China

The Protoceratops sample also

in-cludes several growth stages, providing

a glimpse of this largely unknown pect of dinosaur biology Adults typi-cally measure two meters long; in 1994

as-our team recovered some Protoceratops

less than nine centimeters long Theseskeletons are obviously those of veryyoung individuals, possibly newborns

As we make such new discoveries,

SCIENTIFIC AMERICAN December 1994 67

CliÝs Current expeditions (center ) can rely on satellite

navi-gation aids, but they face no less arduous conditions Although

paleontologists have mapped many rich fossil territories,

much of the Gobi remains unexplored (map at right ).

TUGRUGEENSHIREH

FLAMING CLIFFS

KHERMEENTSAV

MONGOLIAN-AMERICAN ROY CHAPMAN ANDREWSMONGOLIAN-POLISH

EXPEDITION ROUTES

G O B I D E S E R T

Copyright 1994 Scientific American, Inc.

however, the picture of dinosaur life

that emerges becomes more complex

Because Protoceratops is the most

com-mon dinosaur fossil in the region,

pale-ontologists have long assumed that the

many shells and egg aggregates found

at the Flaming CliÝs and elsewhere

be-long to it Yet evidence for this

suppo-sition has been unsatisfactory None of

the hundreds of dinosaur eggs

collect-ed have clearly identiÞable

Protocer-atops embryos within them Even the

tiny skulls of Protoceratops we recently

discovered cannot be positively linked

with an egg of a particular type

A new Þnd from Ukhaa Tolgod

sug-gests that this assumption may have

been wrong The examination of a

clutch of eggs containing dinosaur

em-bryos found on our Þrst day there

re-vealed that an oblong, somewhat

wrin-kly egg usually attributed to

Protocer-atops held a nearly perfect oviraptorid

skeleton It appears likely that many of

the eggs found at Ukhaa Tolgod (and

possibly elsewhere) belong to these

small carnivores rather than to the

par-rot-beaked, herbivorous Protos

The Ukhaa Tolgod ỊnestĨ contains

other fossils of great intrigue Two tiny

skulls of a dromaeosaur (possibly

Ve-lociraptor) were found in the clump of

eggs; bits of oviraptorid eggshell were

associated with their bones This

curi-ous coincidence of eggs, an oviraptorid

embryo and two very young or

new-born dromaeosaurs has several

plausi-ble explanations

Perhaps the young dromaeosaurs

were honing their skills at a very early

age by raiding dinosaur nests

Alterna-tively, the parent oviraptorid may havebeen feeding the dromaeosaurs to heroÝspring The dromaeosaurs might alsohave been interlopers, their eggs placed

in the oviraptorid nest in much thesame way that cuckoo birds place theireggs in the nests of other bird species

Although the mystery cannot be solved, these fossils suggest ways oflife and nesting behaviors for theropoddinosaurs that had thus far not beentied to hard paleontological data

re-This discovery also puts an ironictwist on nomenclatural history The An-drews expeditions applied the name

Oviraptor to a skeleton at the Flaming

CliÝs because it was found atop a clutch

of eggs They assumed that the eggs

belonged to the common Protoceratops and that Oviraptor (literally, Ịegg hunt-

erĨ) was raiding a nest Our Þnd

dem-onstrates that Oviraptor may not have

been devouring eggs but rather ing them The name will stick because

incubat-of nomenclatural rules, but it hardly Þts the true circumstances behind thediscovery of the Þrst known skeleton

be-Nesting sites and skeletons of

birds, dinosaurs, mammals andother vertebrates all make up afairly detailed picture of life in the Gobiduring the late Cretaceous The new ev-idence contributed by the MongolianAcademyÐAmerican Museum expedi-tions has been gathered by loggingthousands of kilometers over a widestretch of the Gobi rather than concen-trating for a prolonged period on a sin-gle or a few sites This method not onlyincreases the chance of Þnding new fos-sil sites, it conveys a better sense of therock sequence through comparison offossil-bearing strata over a broad area

Thus, we can try to determine whetherassemblages of animals and sedimentsrepresenting a particular environmentand time interval are widespread orconÞned to isolated outcrops

For example, paleontologists havegenerally believed that the community

of fossils in the Djadokhta Formation(a Central Gobi bed of brilliant red sand-stones named for the Flaming CliÝs) isslightly older that of the Barun GoyotFormation (which gets its name from

an ancient settlement in the NemegtValley) in the western Nemegt Both ournew Þndings at Ukhaa Tolgod and ourbroad survey, however, suggest that thetwo formations preserve contemporary,virtually identical fauna We found anextension of this community in themagniÞcent red and vermilion beds ofKhermeen Tsav, an isolated set of bad-lands in the arid desert west of the Ne-megt region that strongly resemblesthe canyon lands of southern Utah

We have also found fossils from theDjadokhta community, including the fa-

miliar Protoceratops, in an area called

Khugene Tsavkhlant, near the easternrailway These discoveries are particu-larly signiÞcant because the sandstonesthere appear to be the result of stream

or river action, a situation more typical

of North American sites than of theGobi It is slowly becoming clear thatthe animal community once thought to

be localized at the Flaming CliÝs mayhave occupied a range of habitats.The wide geographic separation ofmany sites, however, impedes thesecomparisons of fossil localities In ad-dition, Gobi rock sequences are entirelysedimentary, without even traces of vol-canic rocks Thus, geologists cannot de-termine the age of these strata by ana-

DINOSAUR AND MAMMAL FOSSILS from the Gobi are remarkably well preserved

The newly discovered troodontid (life-size photograph of skull and sketch at left )

was a small carnivore closely related to birds It has yet to be given an oÛcial

name The multituberculate skeleton ( at center and rotated at right ) is almost

com-lyzing their proportions of radioactive

isotopes Estimates of the age of

vari-ous formations must rely on the

simi-larity of the vertebrates to those of

ref-erence faunas on other continents and

on correlations with invertebrate

fos-sils from Cretaceous marine rocks in

Central and East Asia

We have sampled representative rock

sequences in the hope of obtaining

pa-leomagnetic signals, but results have

not yet come in These signals track the

orientation of the earthÕs magnetic Þeld

at the time the rock and its minerals are

deposited The ÒfrozenÓ paleomagnetic

signals can then be matched against a

chronology of reversals in the earthÕs

Þeld Paleomagnetic data would

there-fore provide an independent source for

estimating the age of the Gobi rocks

In yet another ironic twist, the rocks

of the Gobi appear to be missing

pre-cisely those strata that currently hold

the greatest public interest: no sections

found thus far include the

Cretaceous-Tertiary ( K-T ) boundary, when the

di-nosaurs became extinct Although the

Gobi is richly endowed with early

Ter-tiary mammal faunas, there seems to

be a gap of at least several million years

between these and the late Cretaceous

dinosaur faunas Whatever cataclysm

wiped out the dinosaurs (and many

other species then on the earth), its

mark on Central Asia seems to have

been erased If a continuous sequence

could be found somewhere in the

des-ertÕs vastness, it would make a

formid-able contribution to our knowledge

con-cerning the dinosaur extinction and the

subsequent rise of mammals

The notion of Þnding the K-T

bound-ary in the Gobi is not just wishful

think-ing Satellite navigation has already

made a tremendous diÝerence in theeÝectiveness of our work We can plotthe precise location of fossil sites andthe routes that lead to them We have

also used LANDSAT and SPOT satellite

images as a prospecting tool After wereturned from Ukhaa Tolgod in 1993,Evan Smith of the Yale University Cen-ter for Earth Observation enhanced redand brown spectral bands on comput-er-based satellite images by matchingcolors from photographs of the rocksthere The result is a map that showswith high precision the extent and con-tours of fossil-bearing strata

During the 1994 season, we usedthese images as a Þeld guide and sim-ply drove to the latitude and longitude

of a telltale cluster of red pixels Some

of these computer-targeted spots provedproductive Satellite and computer tech-nology have provided us with a usefulpaleontological atlas in a region wheredetailed topographic or geologic mapsare virtually lacking We now also havesomething that might have cost An-drews his most important but seren-dipitous discoveries: a fairly decentroad map of the Gobi

Despite our new technology and thedecades of insights into the evolution

of vertebrates, exploration of the Gobihas much the same quality that An-drews and his colleagues experiencednearly 70 years ago The Flaming CliÝs

we encountered on that Þrst joyful day

in 1990 were as Andrews describedthemÑimposing, brilliant red in color

and replete with fossils Nearby are gers

much like the one Andrews visited toask for directions on his day of discov-ery Sandstorms that engulfed the 1920sexpeditions returned to wreak havocwith our fragile campsites

When the sandstorms clear, one cansee from the top of the cliÝs the mauve,furrowed mountains of the Gurvan Sai-chan Beyond the mountains are hun-dreds of square miles of fossil-rich bad-lands whose existence Andrews couldonly have imagined The Gobi is and will

be for some time a great wilderness Itwill continue to hold many secrets ofprehistory, of the rise and fall of dino-saurs and other biological empires

SCIENTIFIC AMERICAN December 1994 69

THE NEW CONQUEST OF CENTRAL ASIA

Roy Chapman Andrews American

Muse-um of Natural History, 1932

LATE CRETACEOUS MAMMALS AND SAURS FROM THE GOBI DESERT ZoÞa Kie-

DINO-lan-Jaworowska in American Scientist,

Vol 63, No 2, pages 150Ð159; MarchÐ April 1975

NEW LIMB ON THE AVIAN FAMILY TREE

Mark Norell, Luis Chiappe and James

Clark in Natural History, Vol 102, No 9,

pages 38Ð43; September 1993

EARLY RELATIVES OF FLOPSY, MOPSY, AND

COTTONTAIL Malcolm C McKenna in

Natural History, Vol 103, No 4, pages

56Ð58; April 1994

A POCKETFUL OF FOSSILS Michael J

Nova-cek in Natural History, Vol 103, No 4,

pages 40Ð43; April 1994

SKELETAL MORPHOLOGY OF MONONYKUSOLECRANUS (THEROPODA: AVIALAE) FROMTHE LATE CRETACEOUS OF MONGOLIA

Perle Altangarel et al in American

Muse-um Novitates, No 3105; June 24, 1994.

pletely intact even though some of its bones are barely half a millimeter thick; the

skull is about 2.5 centimeters long Multis (sketch at right ), recognizable by the

many bumps (tubercles) on the crowns of their teeth, were small mammals whose

habits presaged those of rodents such as squirrels and mice

FURTHER READING

Copyright 1994 Scientific American, Inc.

Soon after launch on

Fri-day, September 30, the

crew of the space shuttle

Endeavour were treated to a

dramatic sight The

Kliuchev-skoi volcano in Kamchatka,

Russia, was spewing plumes

of ash 65,000 feet into the air

After 49 years, the volcano had

chosen to erupt that very day

Endeavour was carrying a

so-phisticated radar system

de-signed to study the earthÕs

ge-ology and environment The

equipment had ßown only once

before, the preceding April

The radar planning team

quick-ly changed the schedule

dur-ing the next few orbits to

cap-ture the Þery peak on Þlm

Nor was that the only

natu-ral event to coincide with

En-deavourÕs 12-day ßight On

Tuesday, October 4, an

earth-quake struck near the island

of Hokkaido in Japan Some

hours later Endeavour was able

to scan the coastline for

dam-age from tsunamis

Radar systems, such as the

one on board Endeavour, emit

radiation of relatively long

wavelengthsÑranging from a

few to tens of centimetersÑ

and record the echo returned

by a surface Comparison of

the original and the reßected

ray tells researchers the

dis-tance, size, orientation,

rough-ness and other characteristics

of the reßector For example, an object tends to reßect that

wavelength of radar that matches its own size

If the surface is oriented so that it reßects the radar right

back to the source the way a mirror does, it will look bright

If oriented at some other angle, it will look dark Features

that are rough on the same scale as the wavelength scatter

the radiation in all directions, rather than reßecting it back

Thus, plowed Þelds look bright with shorter-wavelength

ra-dar and ra-dark with longer wavelengths, whereas forests look

bright at most wavelengths.And whereas a long wave-length can pass right through

a hurricane, a short one mightdivulge details of a storm sys-temÕs core

American and European entists worked together foryears to build the radar sys-

sci-tems on board Endeavour We

used three wavelengthsÑofthree, six and 24 centimetersÑcalled X-band, C-band and L-band, respectively The twolonger wavelengths of radiationwere emitted by the Space-borne Imaging Radar-C (SIR-C,pronounced ÒsirseeÓ) instru-ment Scientists at the Nation-

al Aeronautics and Space ministrationÕs Jet PropulsionLaboratory ( JPL) developedthis equipment and the dataprocessor in Pasadena, Calif.,where the information is re-trieved and studied

Ad-The radiation from SIR-C ispolarized, so that its electricÞeld vibrates either in the ver-tical or in the horizontal direc-tion The reßected rays may bereceived either vertically orhorizontally polarized, givingscientists another means ofdiscrimination For example,vertical tree branches may re-ßect one polarization betterthan the other, allowing inves-tigators to distinguish betweendiÝerent types of vegetation.The other radar, X-Band Synthetic Aperture Radar (X-SAR,pronounced ÒexsaarÓ), operates at three centimeters, emit-ting and detecting only vertically polarized light It was de-veloped by the German Aerospace Establishment and thecompanies of Dornier in Germany with Alenia Spazio inItaly, for the Italian and German space agencies

On its April and October ßights, EndeavourÕs orbit was

in-clined at 57 degrees to the equator Drifting slightly on

suc-S CIENCE IN PICTURES

Earth from Sky

Radar systems carried aloft by the space shuttle Endeavour provide a new perspective of the earth’s environment

by Diane L Evans, Ellen R Stofan, Thomas D Jones and Linda M Godwin

SPACE SHUTTLE ENDEAVOUR views the earth with radar.

Visible on the shuttle are ßat antenna panels that emitthree wavelengths of radar, accompanying electronics

( marked ÒJPLÓ ) and an apparatus for measuring spheric pollution from satellites (marked ÒLaRCÓ ).

atmo-Continued on page 75

SCIENTIFIC AMERICAN December 1994 71

DIANE L EVANS, ELLEN R STOFAN,

THOMAS D JONES and LINDA M GODWIN

worked on diÝerent aspects of EndeavourÕs

April and October ßights Evans, of the Jet

Propulsion Laboratory, is the project

scien-tist for the SIR-C radar She has worked on

space-borne radar systems since earning

her Ph.D in geology from the University of

Washington in 1981 Evans also conducts

research on geologic remote sensing fan is the experimental scientist for SIR-C

Sto-She studied the geology of Venus for herPh.D at Brown University in 1988 and isalso the deputy project scientist on theMagellan mission Jones ßew on both ßights

of Endeavour A 1977 graduate of the U.S.

Air Force Academy, he served in the AirForce for six years To pursue an interest in

planetary science, Jones obtained a Ph.D.from the University of Arizona in 1988 In

1991 he became an astronaut Godwin, uty chief of NASÃs Astronaut Ỏce, served

dep-as a mission specialist on EndeavourÕs

April ßight, her second visit to space Acondensed matter physicist who received

a Ph.D from the University of Missouri in

1980, Godwin enjoys ßying small planes

KLIUCHEVSKOI VOLCANO (red area) in Kamchatka, Russia,

erupted on September 30 Its last major outbursts had been

in 1737 and 1945 The Kamchatka River (top) ßows across

this volatile region where the PaciÞc plate is sinking under

the Eurasian plate North of the river are dormant volcanoes

(green); south of it are agricultural settlements (lines) Streaks

( yellow-green) on KliuchevskoiÕs slopes indicate new lava

ßows For this 18.5- by 37-mile image, the transmitted tion was polarized horizontally L-band radiation (24-cen-timeter wavelength) was received horizontally and vertically

radia-polarized, called LHH (red ) and LHV ( green), respectively.

Also displayed is the vertically polarized component, called

CHV ( blue), of the reßected C-band radiation (six-centimeter

wavelength)

Copyright 1994 Scientific American, Inc.