Professional Documents
Culture Documents
Scientific American - 1995 - 03
Scientific American - 1995 - 03
$3.95
The AIDS crisis has driven home the necessity of making potentially therapeutic
drugs available quickly to the patient population. Yet even desperately needed
medicines must not be brought to market unless their demonstrated beneÞts out-
weigh their risks. The director of the Food and Drug Administration explains the
new review procedures that maintain a balance between these priorities.
Fish, dolphins and other marine creatures maneuver through the water with a
speed and eÛciency that put propeller-driven craft to shame. The secret of their
success is their exploitation of the swirling vortices that their own transit creates in
the surrounding water. Engineers, striving to match that Þnny feat, have developed
a mechanical model to test their ideas. Meet RoboTuna.
Normal cells do not turn malignant instantaneously. Instead they gradually fall vic-
tim to an accumulation of irreversible genetic accidents. Some of these mutations
spur growth or replication; others remove the molecular brakes that normally hold
these activities in check. The good news is that this multistep process oÝers many
opportunities for medical intervention.
None of the great apes is more human in appearance, intelligence and behavior
than the bonobo. Surprisingly, though, none also does more to upset conventional
views of how our evolutionary forebears acted. Bonobos live in a uniquely peaceful
society in which femalesÑnot the physically larger malesÑdominate the hierarchy,
and casual sex soothes all conßict.
4
Scientific American (ISSN 0036-8733), published monthly by Scientific American, Inc., 415 Madison Avenue, New York, N.Y. 10017-1111. Copyright © 1995 by Scientific American, Inc. All
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). 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. Subscription inquiries: U.S. and Canada (800) 333-1199; other (515) 247-7631.
90 Protein-Based Computers
Robert R. Birge
Primitive bacteria and tomorrowÕs most advanced computers might have some-
thing in common: bacteriorhodopsin, a protein that changes shape in response to
light. Switches made from it could be ideal for three-dimensional optical systems
that would hold 300 times more information than todayÕs computer memories.
The 19th-century Romantics pined nostalgically for a bygone ÒGolden AgeÓ when
preindustrial society lived harmoniously with nature. Unfortunately, their Arcadian
Greek paradise seems to have been as mythical as the centaurs: archaeological and
geologic evidence shows that ancient Greek farmers seriously eroded their soil.
DEPARTMENTS
18 Science and the Citizen 10 Letters to the Editors
Objections to origins....
Future of the brain.
Old nuclear waste, new nuclear
Þssion.... Gutting the Endangered
Species Act.... The embargo on nu- 12 50 and 100 Years Ago
trition in Cuba .... Beauty in junk 1945: ÒCopper manÓ warms up.
DNA .... Tom Stoppard puts science 1895: Argon is announced.
center stage.... Flaky prediction.
108 Mathematical Recreations
The Analytical Economist Mapping an escape from
An Indian enigma. a puzzling room.
5
THE COVER painting depicts an artistÕs con- ¨
ception of a free-swimming robotic Þsh.
The forerunners of such devices are under Established 1845
development at M.I.T., where they are help-
ing scientists understand how Þsh instinc- EDITOR IN CHIEF: John Rennie
tively exploit the principles of ßuid mechan- BOARD OF EDITORS: Michelle Press, Managing
ics to achieve enviable speed and agility. Editor; Marguerite Holloway, News Editor; Ricki
L . Rusting, Associate Editor; Timothy M. Beards-
One day robotic craft may use these same
ley ; W. Wayt Gibbs; John Horgan , Senior Writer;
principles as they explore the ocean, main- Kristin Leutwyler; Philip Morrison , Book Editor;
tain oÝshore platforms and perform military Madhusree Mukerjee; Sasha Nemecek; Corey S.
missions (see ÒAn EÛcient Swimming Ma- Powell ; David A. Schneider; Gary Stix ; Paul Wal-
chine,Ó by M. S. Triantafyllou and G. S. Trian- lich ; Philip M. Yam
tafyllou, page 64 ). Painting by Al Kamajian. ART : Edward Bell , Art Director; Jessie Nathans,
Associate Art Director; Johnny Johnson, Assistant
Art Director; Nisa Geller, Photography Editor;
Lisa Burnett, Production Editor
COPY: Maria- Christina Keller, Copy Chief ; Nancy
L . Freireich ; Molly K . Frances; Daniel C. Schlen-
oÝ; Bridget Gerety
PRODUCTION: Richard Sasso, Associate Publish-
er/Vice President, Production ; William Sherman,
Director, Production; Managers: Carol Albert,
Print Production; Janet Cermak, Makeup & Qual-
ity Control ; Tanya DeSilva , Prepress; Silvia Di
Placido, Special Projects; Carol Hansen , Compo-
sition; Madelyn 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 Publish-
er/Advertising Director. OFFICES: NEW YORK:
Meryle Lowenthal, New York Advertising Man-
ager; Randy James, Rick Johnson, Elizabeth
Ryan, Timothy Whiting. CHICAGO: 333 N. Michi-
gan Ave., Suite 912, Chicago, IL 60601; Patrick
Bachler, Advertising Manager. DETROIT: 3000
Town Center, Suite 1435, SouthÞeld, MI 48075;
Edward A . Bartley, Detroit Manager. WEST
COAST: 1554 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 Sil-
ver. 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
INTERNATIONAL: EUROPE: Roy Edwards, Inter-
national Advertising Manager, London; Vivienne
Davidson, Linda Kaufman, Intermedia Ltd., Par-
is; Karin OhÝ, Groupe Expansion, Frankfurt ;
Barth David Schwartz, Director, Special Proj-
ects, Amsterdam. SEOUL: Biscom, Inc. TOKYO:
Nikkei International Ltd.; TAIPEI: Jennifer Wu,
JR International Ltd.
ADMINISTRATION: John J. Moeling, Jr., Publisher;
Marie M . Beaumonte, General Manager; Con-
stance Holmes, Manager, Advertising Account-
ing and Coordination
Endangered: One Endangered Species Act tration had an unstated goal for the
amount of timber to be harvested. ÒIf
Cutting resumes in the forestÑand on the Hill weÕd known that, we would have spent
our time more wisely,Ó Meslow explains.
L
ast December, Federal District Judge the act should compensate property The compromise breaks new ground
William L. Dwyer gave a legal owners who refrain from development by employing Òecosystem management.Ó
thumbs-up to the Clinton admin- because of its provisions. To many en- Rather than catering to every species
istrationÕs compromise plan for logging vironmentalists, the cost of such a that might cause concern, the approach
and conservation in federally owned change would make the act untenable. focuses on preserving entire habitats.
forests of the Northwest. The program Congressional and legal assaults on The pragmatic philosophy acknowledg-
allows logging operationsÑwhich the the act could aÝect the outcome of any es the paucity of hard data on most af-
judge halted in 1991Ñto recommence future challenges to the way in which fected creatures. ÒMy opinion has al-
at reduced rates. It also establishes a the forestry plan is implemented. And, ways been that if we persisted in a spe-
mosaic of six diÝerent forms of logging- according to scientists, there are many cies-by-species approach, society was
controlled areas. The strategy aims to ways its enactment could be less than not going to have enough patience,Ó
guard not only species listed as threat- perfect. E. Charles Meslow of the Wild- says Jerry F. Franklin, a professor at
ened under the Endangered Species Act, life Management Institute in Washing- the University of Washington and a key
such as the northern spotted owl and ton, D.C., contends the Òagencies have player in the drafting of the plan.
the marbled murrelet, but hundreds of never been able to accomplish the mon- Franklin points out that the view of
other life-forms. itoring thatÕs been speciÞed.Ó many biologists that the spotted owl is
The verdict marked a crucial turning Meslow says he can now see ßaws in in accelerated decline Òis not complete-
point in the protracted, bitter struggle the plan that he helped to design. The ly accepted.Ó He thinks environmental-
between environmentalists and timber scientiÞc groups involved drew up 10 ists should be content to have achieved
interests. Yet, in all probability, it will diÝerent options, each allowing various most of their objectives. ÒTaking ex-
not be the last word. Biologists, who amounts of logging. Today Meslow be- treme positions on the amount of pro-
generally accept the agreement as a lieves that from the outset the adminis- tection needed is a pretty dangerous
step in the right direction, dis- thing to do,Ó he states. ÒA num-
agree over how much it protects ber of environmental scientists
imperiled species. In addition, have not got the message yet.Ó
the plan relies on federal agen- The critics remain uncon-
cies to monitor habitatsÑsome- vinced. The plan estimates, for
thing critics say government has example, that the spotted owl
failed to do. and the marbled murrelet have
Moreover, Congress may re- more than a 80 percent chance
write the very laws underlying of keeping a Òwell-distributedÓ
the plan. Representative Don population over the next 100
Young of Alaska, chairman of years. But Daniel Doak, a math-
the Resources Committee in the ematical modeler at the Univer-
House (known until this year as sity of California at Santa Cruz,
the Natural Resources Commit- disagrees. ÒThere is a real ques-
tee), has declared his immedi- tion about whether there will be
ate intention to rework the En- any owls left in the wild in 100
dangered Species Act. Young, yearsÕ time to enjoy the nice
who has considerable inßuence, landscape weÕre making for
openly disdains eÝorts to pro- them,Ó Doak declares.
tect rare plants and animals. Christopher A. Frissell, an
He denies there is good evi- aquatic ecologist at the Univer-
dence that the spotted owl is sity of Montana, complains that
threatened, a view shared by the Forest Service deliberately
JOE MCDONALD Animals Animals
I
n some circles science and art are Arcadia deals with related concepts: in
known as Òthe two cultures,Ó seem- a recent essay, Stoppard described the
ingly separated by an impenetrable piece as Òa seasoning of chaos and a
wall. Members from each side of the di- pinch of thermodynamics following a
vide gathered recently for a roundtable dash of quantum mechanics.Ó
discussion at the Mitzi E. Newhouse The panelists opened their discussion
Theater at Lincoln Center in New York by noting that scientists and artists ben-
City to talk about how the two disci- eÞt from models and metaphors. Biolo-
plines can enrich each otherÑparticu- gists image proteins, chemists use balls
larly as science and technology increas- and sticks to show molecules, and phys-
ingly aÝect peopleÕs lives. icists describe the atom with quantum-
The discussion was catalyzed by the mechanical representation. Writers take
appearance at Lincoln Center of two advantage of metaphors to deepen
plays by Tom Stoppard. Both Hapgood, meaning. ÒI think artists [and scientists]
now Þnishing its run, and Arcadia, use models in very similar kinds of
which opens this month at the Vivian ways,Ó remarked poet-naturalist Diane
Beaumont Theater, draw on images and Ackerman. Both hope the representa-
ideas from physics and mathematics. In tions will let them Òsee some aspect of
Hapgood the title character is a British the human condition that fascinates
intelligence oÛcer who protects scien- them from yet another vantage point.Ó
tiÞc secrets during the cold war; anoth- She compared the poem ÒThirteen Ways
er character, a Soviet scientist and dou- of Looking at a Blackbird,Ó by Wallace
ble agent, employs models from phys- Stevens, to its scientiÞc counterpart on
ics as metaphors for human experience. video, which another speaker dubbed
T
Yale University. ÒModels are built in or- he discovery of new elements can ing other heavy elements more diÛ-
der that they may be discarded when be cause for celebration, but late- cult: number 104 was previously known
there are new data. Metaphors never ly it has become cause for argu- as rutherfordium, except by some Rus-
get discarded,Ó he stated. ment. Researchers at the Center for sian scientists who referred to it as
Indeed, chemist and poet Roald HoÝ- Heavy Ion Research in Darmstadt, Ger- kurchatovium. Now the recommended
mann of Cornell UniversityÑwho rep- many, announced last November that name is dubnium.
resents a convergence of science and they had created element 110. Then, in As for elements 110 and 111, they will
artÑmaintained that diÝerences of lan- December, they presented 111. Choos- have to wait their turn. Albert Ghiorso
guage exist between the two worlds. ing the right name for the new sub- of Lawrence Berkeley Laboratory men-
One of those Òis the positive valuation stances may prove more challenging tions a rumor he heard that the Darm-
of ambiguity in art. A word may have than making them. stadt group might hold element 111
two meanings that conßict with each The Þndings come right on the heels hostage until the other names are set-
other; it may mean the same and the op- of an intense Þght over what to dub 101 tled. And Seaborg has Òno ideaÓ what ele-
posite of itself. ThatÕs what makes po- through 109. Although the elements ment 110 might be christened: ÒThereÕs
etry work, in part,Ó HoÝmann explained. themselves do not endureÑfor instance, tremendous confusion right now.Ó
ÒIn science we try, but we donÕt really 110 lives for about two thousandths of Irrespective of the naming game, sci-
succeed, to clean up the language and to a secondÑsome of the researchers who entists continue to make these short-
get the concepts straight between us.Ó made them would like to. The discover- lived compounds to verify theoretical
Nevertheless, scientiÞc duality persists: ers of 106 provisionally named it sea- calculations and to satisfy basic curios-
a character in Hapgood draws on the borgium, after Glenn T. Seaborg, a lead- ity. ÒYou never know what will happen
theory of lightÕs being both waves and ing U.S. researcher. But the International along the way,Ó Ghiorso says. The just
particles to justify how he can work as a Union of Pure and Applied Chemistry created elements promise to gratify re-
spy for both the British and the Soviets. will vote this August on whether the searchers by demonstrating Òthat super-
Ackerman said she believes artists name should be rutherfordium instead, heavy elements are within our grasp,Ó
like Òthe pure fun of using metaphors claiming that an element should not be he adds. Investigators are now aiming
and structures from science.Ó She gave named after a living person. (British for element 114, which calculations sug-
the example of Paul WestÕs novel Gala, physicist Ernest Rutherford died in gest will be particularly stable. But 112
in which every paragraph begins with a 1937.) Not only does this plan upset the will probably be very diÛcult to makeÑ
letter of the genetic code: ÒFor him, it parents of element 106, it makes nam- as well as to name. ÑSasha Nemecek
was a form of organization and play. I
think writers do that very often.Ó
Although the application of scientiÞc
metaphor is obviously not limited to
writing, some of the speakers cited ex-
amples from their favorite texts. Physi-
cist Melissa Franklin of Harvard Univer-
sity mentioned Thomas PynchonÕs book
V, in which he describes the electronic
circuitry of a stereo system. According
to Franklin, Pynchon Òunderstands it all
perfectly, from the shuddering of the
speakers to the music going into his
[characterÕs] head.Ó Furthermore, Frank-
lin noted that Pynchon Òdescribes it as
if heÕs describing a sunset. It is just one
of the most beautiful things IÕve seen.Ó
The panelists concluded that people
tend to feel intimidated by science but
that artistic treatments might help con-
vince them that science is interesting
and accessible. At the close of the dis-
cussion, HoÝmann answered a question
about how art and science can become
more integrated into daily life. ÒI think
we must get away from that ÔhighÕ
thing,Ó he responded. ÒI think it is im-
portant not to deÞne theater as high
theater. I think Bob Dylan writes poetry.
And a lot of simple, everyday experienc-
es are examples of physics, like cooking
or watching a tire deßate,Ó he said. ÒI
think we can bring that to young people.
I think to do that would be to accom-
plish a great thing.Ó ÑSasha Nemecek
W
hatÕs in a word? Several nucleo- segments, which are interspersed with grouped pairs of nucleotides to create
tides, some researchers might the coding parts. In humans, about 97 words between three and eight base
say. By applying statistical percent of the genome is junk. pairs long (it takes three pairs to speci-
methods developed by linguists, inves- Over the past 10 years biologists be- fy an amino acid). In every case, they
tigators have found that ÒjunkÓ parts of gan to suspect that this feature is not found that noncoding regions followed
the genomes of many organisms may be entirely trivial. ÒItÕs unlikely that every the Zipf relation more closely than did
expressing a language. These regions base pair in noncoding DNA is critical, coding regions, suggesting that junk
have traditionally been regarded as use- but it is also foolish to say that all of it DNA follows the structure of languages.
less accumulations of material from mil- is junk,Ó notes Robert Tjian, a biochem- ÒWe didnÕt expect the coding DNA to
lions of years of evolution. ÒThe feeling ist at the University of California at obey Zipf,Ó Stanley notes. ÒA code is lit-
is,Ó says Boston University physicist H. Berkeley. For instance, studies have eralÑone if by land, two if by sea. You
Eugene Stanley, Òthat thereÕs something found that mutations in certain parts of canÕt have any mistakes in a code.Ó Lan-
going on in the noncoding region.Ó the noncoding regions lead to cancer. guage, in contrast, is a statistical, struc-
Junk DNA got its name because the Physicists backed the suspicions a tured system with built-in redundan-
nucleotides there (the fundamental few years ago, when those studying cies. A few mumbled words or scattered
pieces of DNA, combined into so-called fractals noticed certain patterns in junk typos usually do not render a sentence
base pairs) do not encode instructions DNA. They found that noncoding se- incomprehensible. In fact, the workers
for making proteins, the basis for life. quences display what are termed long- tested this notion of repetition by ap-
In fact, the vast majority of genetic ma- range correlations. That is, the position plying a second analysis, this time from
terial in organisms from bacteria to of a nucleotide depends to some extent information theorist Claude E. Shannon,
mammals consists of noncoding DNA on the placement of other nucleotides. who in the 1950s quantiÞed redundan-
Their patterns follow cies in languages. They found that junk
a fractallike property DNA contains three to four times the
called 1/f noise, redundancies of coding segments.
which is inherent in Because of the statistical nature of the
many physical sys- results, the researchers admit their Þnd-
tems that evolve over ings are unlikely to help biologists iden-
time, such as elec- tify functional aspects of junk DNA.
tronic circuits, period- Rather the work may indicate some-
icity of earthquakes thing about eÛcient information stor-
and even traÛc pat- age. ÒThere has to be some sort of hier-
terns. In the genome, archical arrangement of the information
however, the long- to allow one to use it in an eÛcient
range correlations fashion and to have some adaptability
held only for the non- and ßexibility,Ó Goldberger observes.
coding sequences; the Another speculation is that junk se-
coding parts exhibit- quences may be essential to the way
ed an uncorrelated DNA has to fold to Þt into a nucleus.
pattern. Some researchers question whether
Those signs sug- the group has found anything signiÞ-
gested that junk DNA cant. One of those is Benoit B. Mandel-
might contain some brot of Yale University. In the 1950s the
kind of organized in- mathematician pointed out that ZipfÕs
formation. To deci- law is a statistical numbers game that
pher the message, has little to do with recognizable lan-
Stanley and his col- guage features, such as semantics. More-
leagues Rosario N. over, he claims the group made several
Mantegna, Sergey V. errors. ÒTheir evidence does not estab-
Buldyrev and Shlomo lish ZipfÕs law even remotely,Ó he says.
MICHAEL CRAWFORD
G
etting rid of the long-lived, ra- cess for other people,Ó comments Dan-
CORRESPONDENCE dioactive by-products of nucle- iel Leroy of the French Embassy in Wash-
ar power is a problem that has ington, D.C. He predicts that more coun-
Reprints are available; to order, write stalked the industry for its 40-year his- tries will begin to reprocess their own
Reprint Department, ScientiÞc Ameri- toryÑand the pressure will intensify waste. Indeed, Japan, which has the
can, 415 Madison Avenue, New York, this year. No nation has established a largest nuclear program in the world
NY 10017-1111, or fax inquiries to permanent disposal site for its most after the U.S., the U.K. and France, is
(212) 355-0408. dangerous products: used reactor fuel building a second reprocessing plant.
Back issues: $6 each ($7 outside U.S.) and the high-level waste created when A drive toward nuclear self-suÛcien-
prepaid. Most numbers available. Cred- it is reprocessed. Attempts to locate cy could be bad news for companies
it card (Mastercard / Visa) orders for such sites have ignited public outcry in such as COGEMA and for British Nucle-
two or more issues accepted. To or- several countries. And next month is ar Fuels Limited (BNFL), which also re-
der, fax (212) 355-0408. likely to see controversy over another processes foreign fuel. BNFL is commis-
nuclear by-product, plutonium. sioning a new plant at its site in Sella-
Index of articles since 1948 available What promises to be a stormy con- Þeld, despite doubts that have been
in electronic format. Write SciDex , ference will decide whether, or for how raised about its health eÝects on the lo-
ScientiÞc American, 415 Madison Ave- long, to extend the Nuclear Non-Prolif- cal population and its economic viabili-
nue, New York, NY 10017-1111, fax eration Treaty, which aims to prevent ty. The Þnancial worries were under-
(212) 980-8175 or call (800) 777-0444. the spread of nuclear weapons. Global scored last December when two German
ScientiÞc American-branded products trade in plutonium, which is extracted utilities said they would break contracts
available. For free catalogue, write Sci- from used fuel during reprocessing, with BNFL and pay the penalties.
entiÞc American Selections, P.O. Box makes it harder to enforce the treaty, Russia may also be competing for
11314, Des Moines, IA 50340-1314, or reports the Nuclear Control Institute in COGEMAÕs and BNFLÕs business. The
call (800) 777-0444. Washington, D.C. Reprocessing whets country is expanding reprocessing op-
appetites for plutonium in countries erations as a means of earning foreign
Photocopying rights are hereby grant- lacking nuclear weapons. It also increas- capital. A plant at Krasnoyarsk in Siber-
ed by ScientiÞc American, Inc., to li- es the chances that a few kilograms of ia, which has languished unÞnished for
braries and others registered with the the metal, enough for a crude nuclear several years, will apparently be com-
Copyright Clearance Center (CCC) to bomb, will Þnd their way into the hands pleted, according to Russian press re-
photocopy articles in this issue of Sci- of terrorists. ports. South Korea is among the nations
entiÞc American for the fee of $3.00 Putting aside issues of safety, the Nu- that might reprocess spent fuel there.
per copy of each article plus $0.50 per clear Control Institute argues that re- Opposition by environmentalists is
page. Such clearance does not extend processing spent fuel makes little eco- mounting, however. Russia allowed high-
to the photocopying of articles for pro- nomic sense. The fundamental prob- level waste from reprocessed Finnish
motion or other commercial purposes. lem for countries that reprocess is that fuel to remain in the countryÑtaking the
Correspondence and payment should plutonium is accumulating faster than problem out of Finnish hands. Damon
be addressed to Copyright Clearance it is being used. Nuclear power has not Moglen of Greenpeace International
Center, Inc., 222 Rosewood Drive, Dan- expanded as the industry expected it states that the Czech Republic wants to
vers, MA 01923. Specify CCC Ref- would, and the price of uranium, the establish a similar arrangement. Rus-
erence Number ISSN 0036-8733/95. primary nuclear fuel, is at a historic low. siaÕs preferred method of disposal has
$3.00 + 0.50. Thus, there is little industrial demand been to pump high-level waste in liquid
Editorial correspondence should be for plutonium, which can be used in form into the groundÑbut it has al-
addressed to The Editors, ScientiÞc some reactors as a supplemental fuel. ready started to leach out.
American, 415 Madison Avenue, New International commercial reprocess- Although the U.S. refrains from re-
York , NY 10017-1111. Manuscripts are ing means that high-level waste and processing, utilities have encountered
submitted at the authorsÕ risk and will spent fuel have to be shipped around legal problems over accumulating spent
not be returned unless accompanied the globe. France is steeling itself for fuel. Most of it is in ÒpoolsÓ where it is
by postage. protests when the state-owned repro- allowed to cool for some years. Some of
cessing company COGEMA ships the these are now full, and local jurisdic-
Advertising correspondence should Þrst in a likely series of cargoes of high- tions are objecting to the huge casks
be addressed to Advertising Manager, level waste from its plant in La Hague that the utilities want to use for ongoing
ScientiÞc American, 415 Madison Ave- to Japan. The material, produced during storage. The Department of Energy re-
nue, New York, NY 10017-1111, or fax the reprocessing of Japanese fuel, has mains under congressional mandate to
(212) 754-1138. been solidiÞed into glass blocks. The decide by the year 2001 whether Yucca
Subscription correspondence should Þrst shipment was scheduled for Febru- Mountain in Nevada is suitable as a lo-
be addressed to Subscription Manager, ary. The obvious route runs through the cation for a permanent repository. If
ScientiÞc American, P.O. Box 3187, Har- Panama Canal, but Caribbean nations certiÞed, the site could commence op-
lan, IA 51537. The date of the last issue have objected to having the waste enter erations around 2010.
of your subscription appears on each their waters. The Philippines, which lies Prospects for the facility, however,
monthÕs mailing label. For change of on an alternative route, has also banned have had a recent setback. The DOE
address notify us at least four weeks in the cargo. found water containing tritium from
advance. Please send your old address The French government seems to aboveground atomic testsÑwhich there-
(mailing label, if possible) and your new have little stomach for a Þght. ÒIt will fore must be younger than 50 yearsÑ
address. E-mail: SCAinquiry@aol.com. in the future be very diÛcult to repro- less than half a mile from the suggest-
Nuclear Empowerment
J
ust as nuclear power seems to be dying a slow, disrep- radioactive for thousands of years or that might leak out of
utable death comes a proposal for making reactors a container—could be transmuted by neutrons. Weapons-
clean, safe and unattractive to those who might want grade plutonium or waste from conventional reactors may
to blow up New York City. If nuclear fission is induced not also be consumed. (But if all the neutrons are used up dur-
by neutrons from chain reactions but by neutrons from a ing cleaning, none would be around to generate more en-
new generation of powerful accelerators, the reactor can- ergy, so burning all the waste might not be cost-effective.
not burn out of control, argue scientists at Los Alamos Na- After about 40 years of running, some kilograms would
tional Laboratory and CERN, the European laboratory for probably have to be consigned to man-made repositories.)
particle physics near Geneva. Radioactive waste could also “Our goal is similar to that of fusion,” says Carlo Rubbia
be eliminated, and bomb makers could not divert the fuel. of CERN. “We want to find an environmentally acceptable
Conventional reactors often use a mixture of uranium fuel. But in this case, there are no significant technological
isotopes. When a neutron hits one such nucleus, it breaks barriers.” Although no accelerator can yet deliver a proton
into two halves, releasing energy and two or three other beam of the required intensity, Charles Bowman of Los
neutrons. These particles either escape or induce further Alamos envisions building a prototype in seven years: “It’s
fissions, producing a chain reaction. The reactor needs to just a question of getting on with it.” That is, if the Depart-
maintain a delicate balance in the number of neutrons. Too ment of Energy or the European Union, as the case may
few, and the reaction dies out. Too many—even by 1 per- be, sees fit to provide the funds. —Madhusree Mukerjee
cent—and there could be a Chernobyl.
(Control rods, which absorb excess neu-
trons, help to tame the reaction.)
Imagine instead that the neutrons
were produced by an accelerator—in
practice, by a beam of protons hitting a
heavy target. Then there would be no
danger of a runaway. “If the lights go
out, the cooler stops circulating, and
nothing works,” says Edward A. Heigh-
way of Los Alamos. “The reaction will
just stop.” The neutrons transmute the
fuel, which consists of thorium 232,
into uranium 233, which fissions, pro-
ducing heat. About 15 percent of the
energy would be fed back to run the ac-
celerator. The rest—1,000 megawatts in
one scenario—can be purchased.
Because thorium 232, which is natu-
rally abundant, is not fissile, the reac-
tor’s inventory is not attractive for mili-
tary uses. And although uranium 233
can fission, it is contaminated with oth-
er uranium isotopes. A dedicated bomb
CHIP SIMONS
N
othing is constantÑnot the the time the upper and lower mantles way to the top of the earthÕs core. Such
world, and certainly not the way convect independently of each other. deep divers may correspond to the cat-
scientists view it. For nearly two But cooler material accumulates at the astrophic ßows that show up on the
decades geophysicists have engaged in bottom of the upper mantle until a computer screens. ÒWe now see that
spirited disagreements about how the threshold is reached. At that point, the the mantle convects in a very complex
earth moves internally. Yet a consensus material crashes down through the style,Ó comments Hua-Wei Zhou of the
Þnally appears to be emerging on University of Houston.
the nature of the deep motions that So is the debate closed? Not quite.
give rise to some of the globeÕs Don L. Anderson of the California
great changes: earthquakes, volca- Institute of Technology, a longtime
noes and the roving of continents. booster of layered convection, re-
T
great pulses of geologic activity oc- he Cuban government has al- physiciansÕ group to ask that the Amer-
curred when the mantle overturned. ways pointed to its health care ican government lift the embargo on
Such episodes could have aÝected life, system as one of the triumphs of sales of food and medical supplies. Last
both by creating dry land and by dump- the 1959 revolution. But the end of fall Jack P. Whisnant, president of the
ing carbon dioxide into the atmosphere, subsidies from the former Soviet Union American Academy of Neurology, wrote
which would cause increased green- and an economy overly reliant on pro- to President Bill Clinton and every con-
house warming. Indeed, rapid crust ducing sugar have placed enormous Þ- gressional representative to request that
formation seems to have been associat- nancial strains on the islandÕs network the ban on these items cease. The ap-
ed with the balmy temperatures di- of clinics and hospitals. Basic supplies, peal is similar to pleas by the American
nosaurs enjoyed 100 million years ago. from antibiotics to sterilizing deter- Public Health Association and by the
ÒThis is an exciting time, because peo- gents, are hard to Þnd. And CubansÕ United Nations General Assembly, which
ple from diÝerent disciplines can all Þ- health has worsened because they have have both called for the full lifting of
nally work together,Ó Zhou exclaims. In had little to eat during what the govern- the embargo.
science as in nature, chaos makes the ment has dubbed the periodo especial. For its part, the academy decided to
world go Õround. ÑCorey S. Powell The crisis has prompted a U.S.-based tread into the mire of Cuban-American
politics when a member documented
one of the worst neurological epidemics
of the 20th century. At the behest of
ScientistsÕ Sense of Snow the Pan American Health Organization
(PAHO), Gustavo C. Roman led a team
of physicians on a visit to Cuba from
May to September 1993. Roman, who
was then chief of neuroepidemiology at
the National Institutes of Health, and his
colleagues conÞrmed the Cuban Þnding
that more than 50,000 of the 11 million
inhabitants were suÝering from such
maladies as optic neuropathy (visual
loss), deafness, sensory neuropathy
(loss of sensation in the hands and
feet) and a spinal cord disorder that
impaired walking and bladder control.
Cuban and PAHO physicians discard-
ed a hypothesis that the illnesses result-
ed from toxins. They determined that a
spare diet, along with great physical ex-
U.S. DEPARTMENT OF AGRICULTURE
S eeing the world in a grain of sand may be passé. Researchers at the U.S.
Department of Agriculture are endeavoring instead to find the future in a
grain of snow. By analyzing the size, structure and water content of snow-
Roman reported his Þndings in Neu-
rology in 1994. Although details of the
scourge had been chronicled elsewhere,
flakes, hydrologists there hope to predict the amount of spring runoff that Roman went beyond his capacity as a
will be available in areas where snow is a major source of agricultural water. neutral statistician: he lay part of the
The approach originated in the USDA’s Scanning Electron Microscopy Labo- blame on a 1992 law that tightens the
ratory in Beltsville, Md. In December 1993, when the lab completed the instal- 30-year-old embargo prohibiting U.S.
lation of a low-temperature specimen holder, scientists scurried outside to find companies from trade with Cuba.
an insect to image. All they found was a fresh blanket of snow. The investiga- The Cuban Democracy Act of 1992
tors collected flakes, dipped them in liquid nitrogen, coated them with a thin blocks foreign subsidiaries of U.S. busi-
layer of platinum and scanned them. The pictures ultimately reached hydrol- nesses from trading with Cuba. Until its
ogist Albert Rango and his colleagues, also at the USDA, who envisioned us- passage, dealings with subsidiaries had
ing them to improve estimates of the amount of water held in drifts. “We use allowed Cuba to import critical medi-
microwave data from satellites to determine the area and number of snow cines and foodstuÝs cheaply. ÒAlthough
grains in a snowpack,” Rango notes. “But there is some confusion about the ac- the U.S. economic embargo may not
tual water content.” The group is sampling snow in the Sierra Nevada Moun- have been the primary cause of the epi-
tains and will combine these data with satellite information. —Steven Vames demic in Cuba, it has contributed to its
development, complicated its investi-
A
lthough the American population cess fat and a host of diseases, the bill part of accumulating scientiÞc evidence
is stabilizing in number, it is bal- for our larger meals is also becoming that brings both good and bad news.
looning in mass. The epidemic clearer: more than $70 billion in health The bad news is that the obeseÑthose
in obesity continues to spread, now af- care costs, according to a recent report whose weight in kilograms is more
ßicting 35 percent of women and 31 by the Institute of Medicine (IOM). than 27 times the square of their height
percent of men, up from 27 and 24 per- Fads, quacks and the proliferation of in metersÑrun a signiÞcantly higher
cent just 10 years ago. As metabolic re- ÒliteÓ foods have failed to reverse the risk of coronary heart disease, stroke,
searchers clarify the links between ex- trend. Medicine, an increasingly vocal high blood pressure, sleep apnea, dia-
F
or the past 30 years the Indian Another set of noneconomic data, and oÝ between 1957 and 1991 nor the
state of Kerala has been one of however, may be calling into question current conservative regime has been
economistsÕ favorite anomalies. the rosy picture that Sen and others able to do much about the situation. In-
People there are very poor, and the paint. Crime is up, the suicide rate is dian industrialists have tended to avoid
economy is growing very slowly, but its three times the national average and Kerala because of powerful unions and
respectable Þgures for child mortality, more than 600,000 of KeralaÕs 29 mil- high wages relative to other states; in
literacy and general life expectancy are lion citizens have left to Þnd employ- spite or, perhaps, because of this pro-
what one might expect from a far rich- ment elsewhere. Indeed, the expatriate worker reputation, local businesses lose
er region. Experts such as Amartya Sen work forceÑabout 300,000 of whom fewer days to strikes than does the av-
of Harvard University have cited the labor in Persian Gulf countries such as erage Þrm elsewhere in India.
stateÕs accomplishments as proof that Saudi Arabia and KuwaitÑsent home KeralaÕs mixed situation seems to call
purely economic yardsticks, such as enough money in 1989 to expand Ker- into question not only any simple mea-
per capita gross national products, are alaÕs total income by 13 percent, says sure of development or economic health
doubtful indicators of an areaÕs true Richard W. Franke of Montclair State but many of the assumptions on which
prosperity [see ÒThe Economics of Life College. Most of these remittances went such measures are based. The volume
of remittances, for example, skews re-
gional income Þgures in ways that are
diÛcult to adjust. When economists
speak of integrating a locality into the
global economy, they usually think in
terms of large, easily measured ßows
of imports and exportsÑnot hundreds
of thousands of accountants, nurses,
maids and construction workers send-
ing cash to their cousins.
Similarly, labor economics predicts
that high unemployment will eventually
cause wages to fall, but this adjustment
does not appear to have happened in
Kerala. The anomaly might be explained
by the strong unions or by government
intervention in the market or even by
the notion that gaining the basic neces-
sities of life makes work at low wages
less attractive. Yet invoking such factors
raises questions about whether such an
RAGHUBIR SINGH
J
im P. Corrigan, the proprietor of squash. Because the gourd passed mus- ico. Few other economically useful
Carrot Top, a premium produce ter, it will be treated like any other fruit plants are so classiÞed. Scientists wor-
store in the Chicago suburbs, says or vegetable. ry that this large and varied gene pool
he talked to hundreds of report- Calgene, maker of the Flavr Savr, vol- needed for crossbreeding might be di-
ers following his Þrst sales last May of untarily slapped a Ògenetically engi- minished if a genetically engineered
the genetically engineered Flavr Savr neeredÓ label on its tomatoes, the trait gets transferred to a domesticated
tomato. ÒI was surprised that a tomato worldÕs Þrst produce with that designa- plantÕs wild relatives. A wild plant con-
could generate that much interest,Ó tion. But farmers who sow seeds this taining the new gene might thrive and
Corrigan notes. spring for the squash, known as Free- drive out close kin lacking the gene.
As the months passed, the number dom II, or more cryptically ZW-20, will The USDA found no merit to any of
of calls dwindledÑand so did the in- probably not label it. (The other six these arguments. The transfer of the
tense national interest in bioengineered crops that passed through the FDA still gene to a wild relative would not pro-
crops. ÒAll of a sudden biotechnology need a stamp from the USDA or the En- vide any selective advantage, the agen-
foods have become a nonissue for the vironmental Protection Agency, or both, cy concluded, since there was no indi-
media,Ó remarks Thomas J. Hoban of but no major obstacles are expected.) cation that the squashÕs cousins were
North Carolina State University, who has Some biotechnology companies hope infected by these viruses. So resistance
conducted consumer surveys on food the quiet will persist. Asgrow Seed Com- to these viruses would not enhance the
biotechnology. pany, parent of ZW-20, was pleased that Þtness of this plant.
Much of the USDA evaluation was
based on an analysis of interactions be-
tween the squash and its relatives that
was performed for the agency by As-
grow, ZW-20Õs creator. ÒWe think the
fact that the mere 14 wild plants they
looked at didnÕt have a virus doesnÕt
tell you anything about selective pres-
sures,Ó says Jane Rissler of the Union
of Concerned Scientists.
The biotechnology industryÕs hopes
for keeping a low proÞle may be short-
lived. The Pure Food CampaignÑa
group that has squashed tomatoes and
dumped milkÑplanned to begin a new
series of protests in February. ÒWe be-
ASSOCIATED PRESS
GIF Us a Break
I
nstead of becoming higher tech with Superconductive designs further suf- from Toshiba, a company that sells both
time, magnetic resonance imaging fer from restrictive geometry. The pa- permanent-magnet and supercon types.
(MRI) appears to be going retro. Al- tient must lie in a narrow tunnel within The key question is whether the use
though clinical use of MRI scanners is a helium-Þlled vacuum dewar, an expe- of lower magnetic-Þeld intensities com-
nearly two decades oldÑand several rienceÑnot unlike being trapped in a promises the technique. JeÝrey C. Wein-
million scans are now done annuallyÑ thermos bottleÑthat can sometimes reb, a radiologist at New York Universi-
some doctors and patients remain frus- provoke claustrophobia. J. Carlos Me- ty Medical Center, cautions that the an-
trated with the machinesÕ low-tempera- lŽndez and Ernest McCrank of the Uni- swer is complex: ÒIf your major interest
ture superconductive magnets and their versity of Western Ontario reported in is in the quality of the images and the
expense. Several manufacturers are an- the August 11, 1993, issue of the Jour- range of capabilities, there may be some
swering these concerns, not with so- nal of the American Medical Associa- advantage to the high-Þeld systems.Ó
phisticated high-temperature supercon- tion that Ò10 percent of patients under- Weinreb adds that ÒmidÞeld machines
ductors but with straightforward solu- going an MRI examination at our insti- maybe donÕt have all the bells and whis-
tions reminiscent of some early versions. tution experience anxiety to the point tles, but they can do a nice job for 90
From the outset, many MRI designers that the procedure has to be modiÞed, percent of the things you need to do.Ó
chose to employ technically elegant su- postponed, or canceled.Ó Raymond Damadian, a pioneer of
perconductive magnet systems; these Another disappointment with high- magnetic resonance imaging who is
ÒsuperconsÓ make up the majority of Þeld designs is the expense: such MRI now president of Fonar, says price is
the real issue. ÒMy perception is
that the diÝerence in quality of
the images is small: they get to
looking a lot diÝerent when you
havenÕt got the money to buy
one,Ó he notes. Catering to cus-
tomers who have not previously
had the money seems to be a
growing trend.
General Electric, long promi-
nent in manufacturing supercon-
ductive MRI systems, will soon
T
o hear the worldÕs telephone tion, battery life, reception within build- networks (PCNs), currently being built
companies tell it, providing basic ings or tracking roaming callers. on both sides of the Atlantic, will use
communications to homes and This spring Ionica will roll out its op- digital, wireless systems for telephones
small businesses requires high-tech no- tions. It promises prices well below cur- that travel everywhere. (Nearly every
blesse oblige. Customers are demand- rent ones and quality at least as high. U.S. telecommunications company put
ing, the technology is uninteresting and Rivals sniÝ a bit at IonicaÕs contention, in a bid at the governmentÕs recent
regulations keep the prices of such ser- arguing radio interference could prove auction of spectrum so they can set up
vices below the cost of providing them. more of a problem than the company PCNs.) Peter W. Huber, a telecommuni-
Who, they muse, could aÝord to shoul- now admitsÑparticularly in the English cations expert in Washington, D.C., has
der this load without a regulatory Þllip rain. Ionica retorts that its system is coined the phrase Ògeodesic networkÓ
here, a cross-subsidy there or a market amenable to error-correction techniques to describe the interconnected network
privilege elsewhere? Enter Ionica, a small that its competitorsÕ analog ones are not. of networks that is beginning to follow
British company that reckons technolo- The interesting twist to IonicaÕs prop- deregulation. These choices for consum-
gy has transformed telephone service osition is that the very technology that ers bring quandaries for regulators.
from burden to opportunity. If Ionica is enables it to compete limits it to basic The Þrst concerns pricing. Regulators
right, many assumptions about telecom- services. IonicaÕs bandwidth of tens or often tax long-distance and business
munications markets are wrong. hundreds of thousands of bits a second customers so as to subsidize low-cost
For now, this experiment is unfold- is great for voice, fax, data and Internet services. Inasmuch as Ionica and its ilk
ing in Britain because Britain is live up to their promise, they will
the only country whose regulators push consumers toward a less
will allow it to happen. But given a than optimal world: one in which
Republican Congress talking about artiÞcially high prices for long dis-
reviving legislation to deregulate tance are used to make the price
U.S. telecommunicationsÑand oth- of perhaps less eÛcient local ser-
er countries moving in this direc- vices artiÞcially competitive.
tion as wellÑthe British case mer- More intriguing problems con-
its attention. Unlike the U.S., which cern diversity. The oÛcial job of
has maintained regulated monop- regulators has been to establish
olies for local services while pro- minimum levels of performance,
moting competition in long-dis- thereby creating common ground
tance and other lucrative arenas, for communications. The assump-
Britain has encouraged rivalry in tion of this role seems to live on in
all telecommunications sectors for the American politiciansÕ promise
a decade. In 1991, when BritainÕs that competition will foster an
market was opened to all, most Òinformation superhighwayÓÑas
entrants avoided local service and if that single result were a fore-
grabbed for higher proÞt margins. gone conclusion. But what if some
Nigel Playford, however, founded people simply opt to travel in the
DAVID LEVENSON Black Star
Ionica to compete solely for low- slow lane? Are they Òinformation
volume, local services. have-nots,Ó to take the phrase of
A key problem for telecommu- Vice President Al Gore, or merely
nications contestants is that the smart consumers?
cost of wiring homes and busi- Diversity is approaching every-
nesses is huge compared with the where. In remote areas of Texas,
amount of revenues collected. Ca- GTE is using wireless technologies
ble television companiesÑthe NIGEL PLAYFORD, founder of Ionica, is tackling to cut the cost of reaching people
assumptions about telecommunications markets.
chief competition in local British far from wires. In the developing
telecommunicationsÑspent more world, many countries are won-
than £600 ($900) for each of the some access. But video requires at least a mil- dering if Þxed-cellular services might
four million homes reached by their ca- lion bits a second. One of the ironies of oÝer a quick, cheap way of creating a
bles. Only the sale of video entertain- the struggle for telecommunications telecommunications infrastructure.
ment could justify such an outlay. But markets has been that competition has Ionica and Northern Telecom have li-
Ionica Þgures that by using radio-based been made Þercer by the assumption censed their technology to Mexico and
systems it can reach homes for less that most companies will eventually of- Indonesia, and they are talking to,
than £10 apieceÑso it can boldly com- fer similar services. Because information among others, China. Today in the de-
pete where none have ventured before. is increasingly transmitted in digital veloped world the question is how the
Ionica will reach customers with a form, runs conventional wisdom, it will low cost and convenience of wireless
stripped-down version of the technolo- all increasingly ßow together across the will compete with the huge bandwidth
gy used to create digital mobile tele- same lines. Given new investments in of wired services. Tomorrow in the de-
phones. IonicaÕs device, developed joint- telephone and cable networks, the two veloping world the question could be
ly with Northern Telecom in Canada, is will ÒconvergeÓ into a web that carries quite the opposite. In either case, local
not mobile: its transceiver is screwed to voice, data and video side by side. Ioni- markets may yet be transformed from
the side of a buyerÕs house. Thus, Ioni- ca presents another option. Choosing erstwhile public service to competitive
ca need not worry about miniaturiza- Ionica is not merely choosing a diÝer- battleground. ÑJohn Browning
I
n the dead of night, when the de- for sounding reasonable. Arguing that Illinois found the spectral signature of
mons come, a special fear may creep the seeds of life must have come from amino acidsÑthe building blocks of
into the hearts of scientists: What if space, he points out that asteroid im- proteinsÑin interstellar space. New Sci-
Fred Hoyle is right? Then astronomy is pacts rendered the earth uninhabitable entist concluded, rather generously, that
a sham, biology a house of cards and until at least 3.8 billion years ago and the observations lent credence to the
modern medicine an illusion. that cellular life had almost certainly claims of Hoyle and his collaborator
Those who adhere to the paradigms appeared by 3.7 billion years ago. If one N. Chandra Wickramasinghe that space
that be have more reason than usual to thinks of the entire 4.5-billion-year his- is teeming with microbes. Just one year
harbor such worries these days. The me- tory of the planet as a 24-hour day, earlier the British journal had compared
dia, no doubt bored by the glacial pace Hoyle elaborates, then life appeared in one of their books with Erich Von Dan-
of mainstream research, have acquired about half an hour. ikenÕs pseudoscientiÞc best-seller Char-
a sudden fondness for Hoyle, whom ÒYouÕve got to discover DNA; youÕve iots of the Gods.
they had long viliÞed or ig- Hoyle has accepted the
nored. Journalists are treat- recent ßurry of favorable
ing the British iconoclastÕs publicity with a grain of
attacks on the big bang salt. ÒIf a chap lives to 80,
theory and other pillars of he deserves a pat on the
modern science with new- back.Ó (Actually, Hoyle
found respect. ÒA second turns 80 this June.) He
dawn for the universal reb- seems pleased at his con-
el,Ó proclaimed the London tinuing ability to provoke
Times last fall. controversy. ÒWhen I was
Peer-reviewed journals young, the old regarded
have also clasped HoyleÕs me as an outrageous
autobiography, Home Is young fellow,Ó he says
Where the Wind Blows, to with a rakish grin, Òand
their bosoms. ÒWhat good now that IÕm old the
fortune to have this beauti- young regard me as an
fully written autobiography outrageous old fellow.Ó
of one of this centuryÕs In his autobiography
leading scientists,Ó gushed Hoyle depicts himself as a
Nature, which in 1993 had bright but restless boy.
harrumphed that HoyleÕs The son of a teacher and a
recent writings provided cloth salesman, he often
Òfull documentation of the skipped classes in favor of
DAVID LEVENSON Black Star
the steady-state theory, which posits ÒScience today is locked into para-
that the universe is inÞnite both in digms,Ó he declares. ÒEvery avenue is
space and time and constantly gener- blocked by beliefs that are wrong, and
ates new matter through some un- if you try to get anything published by
known mechanism. a journal today, you will run up against
Together with his longtime collabo- a paradigm, and the editors will turn
rator GeoÝrey Burbidge of the Univer- it down.Ó Hoyle emphasizes that he has
sity of California at San Diego and the HOYLE sees purpose everywhere. never asserted, as some reports have
Indian astronomer J. V. Narlikar, Hoyle stated, that the AIDS virus has an ex-
has recently developed a new and im- retical speculations somewhat as a re- traterrestrial source. It Òis such a strange
proved version of the steady-state the- sult of his involvement in CambridgeÕs virus I have to believe itÕs a laboratory
ory. This ÒquasiÐsteady stateÓ theory re- Institute of Theoretical Astronomy. Al- product,Ó he comments. Is Hoyle sug-
places one big bang with many little though he helped to make the institute gesting that the pathogen might have
bangs, perhaps occurring within qua- a potent force in astronomy, he grew been produced by a biological warfare
sars and other so-called active galaxies. tired of Þghting with university oÛcials program that went awry? ÒYes, thatÕs
These little bangs generate light ele- over administrative matters and re- my feeling,Ó he responds.
ments such as helium and lithium as signed from the directorship in 1972. Purpose pervades HoyleÕs universe.
well as the local expansion of galaxies. The hatchet was not buried until 1992, He has long felt that natural selection
Although Hoyle and his colleagues when the institute held a ceremony alone could not account for the appear-
have outlined their quasiÐsteady state honoring Hoyle and unveiled a statue ance and rapid evolution of life on the
theory in such reputable outlets as the of him. earth. Some supernatural intelligence
Astrophysical Journal, they have won Hoyle claims to have no regrets about must be directing the evolution of life
few converts. Astronomers contend that abandoning his lofty position. ÒFrom and indeed of the entire cosmosÑal-
the theory merely substitutes many lit- the point of view of ideas it was a stulti- though to what end Hoyle does not
tle miracles for a single large one. More- fying period,Ó he says. Hoyle soon began know. The universe is an Òobvious Þx,Ó
over, HoyleÕs group oÝers no plausible collaborating with Wickramasinghe, he remarks. ÒThere are too many things
explanation for the microwave radia- now at the University of Wales, on a that look accidental that are not.Ó
tion, discovered in 1965, which most study of complex molecules in space. Sensible scientists will dismiss such
cosmologists believe is the afterglow of They eventually concludedÑbased on talk as preposterous. But every now
the big bang. Hoyle insists that recent their interpretation of data from radio and then, in their inevitable moments
versions of the big bang theoryÑwhich and optical telescopesÑthat space is of doubt, they may wonder: Could Sir
can account for observations only by Þlled not only with organic compounds Fred be right? ÑJohn Horgan
I
n the spring of 1994 a group of ance the urgent needs of desperately ill
AIDS activists gathered on the third patients with the FDAÕs responsibility
ßoor of the Parklawn Building, head- to determine whether the drugs work.
quarters of the Food and Drug Admin- Have these policies been too success-
istration in Rockville, Md. The activists ful in speeding the availability of new
made an extraordinary plea to top agen- drugs? AIDS activists at the FDA last
cy oÛcials: donÕt approve drugs to treat spring feared that answers had begun
disease caused by the human immuno- to take a back seat to access. They were
deÞciency virus ( HIV ) too quickly. No concerned that patients and their phy-
one familiar with events of the previous sicians did not know how to make opti-
decade could have predicted this turn- mal use of the existing antiviral AIDS
about. Only a few years earlier angry drugs. They wanted to discuss ways to
activists had besieged Parklawn, de- learn more about the value of experi-
manding access to compounds that mental therapies and to learn it soon-
had barely moved out of the test tube. erÑwhen each drug should be admin-
Faced with imminent death, people with istered, to whom and in what dose.
AIDS clamored for the right to take ther- Without such information, novel drugs
apeutic risks. Experimental drugs were might be of little use.
a ray of hope on a bleak treatment land- Although these activistsÕ caution re-
scape, and many patients were unwill- ßected only one perspective within the
ing to accept any restraint on access. AIDS communities, there is no doubt
Promising drugs generally take sever- that making a drug widely available
al years to test; an additional 18 months early means that it will initially be used
or more may elapse from the time a without full understanding of its char-
sponsor requests product approval un- acteristics. The FDA stands behind the
til a compound is widely available. In tools that have put therapies into pa-
response to concerns about the length tientsÕ hands more swiftly, but it also
of time needed to develop and evaluate remains committed to the time-hon-
new drugs, the FDA made some dramat- ored legal standards for evaluating new
ic changes in the way it conducts busi- products.
ness. In the late 1980s it introduced
rules that expand access to unapproved The Usual Approval Threshold
DONNA BINDER Impact Visuals
STANDARD
APPROVAL
PROCESS
EXPANDED ACCESS
PARALLEL
TRACK
TREATMENT IND
(INVESTIGATIONAL
NEW DRUG)
ACCELERATED
PROCEDURES
APPROVAL BASED
ON SURROGATE MARKERS
TELESCOPED
TRIALS
DRUG DEVELOPMENT PROCESS can span years as a com- the time it takes to approve new drugs for sale, and it may is-
pound passes through a series of successively larger clinical sue provisional approval for widespread marketing of a com-
studies (top). The FDAÕs expanded-access rules may make a pound on the basis of signiÞcantly fewer data than it once re-
drug for a serious illness available to patients while it is still quired (bottom). A drug may then be removed from the mar-
under investigation (middle). The agency has also reduced ket if later evaluations show that it is not beneÞcial.
a treatment known to improve survival majority of cases the manufacturer based on preliminary or incomplete
or prevent irreversible damage. covers the cost of the trial; the National data can increase the chance that im-
Phase 3 studies, whose design is guid- Institutes of Health also supports stud- portant adverse eÝects will emerge to
ed by the results of earlier trials, may ies, and patients or their insurers may undermine the drugÕs value or that er-
involve as many as several thousand cover the cost of other medical care as- rors in the recommended dosage will
patients and are intended to provide sociated with the trial. In some cases, lead to suboptimal use. Moreover, once
additional deÞnitive data for approval. the availability of a drug or of research a drug has been made widely available,
These controlled trials assess response support may be a crucial factor in de- controlled studies to determine appro-
to diÝerent doses, long-term safety, termining where and how a trial is car- priate dosages, duration of use or inter-
drug interactions and responses that ried out. actions among drugs can become more
diÝer by sex. They also gather other For a product (such as a mild pain- diÛcult to perform. Sponsors have less
data needed to label the drug properly killer ) that provides at most a marginal incentive to conduct further studies,
and deÞne the conditions for its use. clinical advantage or treats a relatively and patients who can get medication
Ideally, phase 3 trials will also investi- innocuous condition, there is seldom without signing up for a trial may be
gate any diÝerences in a compoundÕs any need to depart from the usual pro- less willing to participate. These risks
eÝect on subpopulations that diÝer by cedures. Under certain urgent circum- make it imperative that research contin-
age, race, concurrent illness or other stances, however, the FDA may allow ac- ue until an experimental drug has met
factors. Many of the studies are dou- cess to the therapy before approval or the same standards of safety and eÛ-
ble-blindÑneither the patients nor the approve promising therapies, on the cacy as other therapeutic products. The
investigators know who is getting which basis of more limited data. The agency FDA is responsible for ensuring that
therapyÑto safeguard against bias, and considers many factors, including the sponsors complete those studies.
patients are randomly assigned to dif- severity of the disease and current op-
ferent treatments. Long-term safety tions for treating it, the information Early Access to Therapies
data, however, may come from unblind- available on the new treatment, the ad-
ed, uncontrolled studies.
Trials may be conducted by drug
companies or by researchers aÛliated
equacy of studies designed to secure
more conclusive answers and the risk
that increasing access might make it
F or many years, the agency has per-
mitted cancer patients to receive
certain promising experimental thera-
with hospitals, government laboratories more diÛcult to Þnd out whether the pies. Since 1987 the Treatment Investi-
or universities; depending on the drug drug works or not. Drug companies gational New Drug (IND) regulations al-
being tested, sponsors may advertise must still conduct well-controlled trials low promising drugs for other serious
for patients directly or wait for referrals before Þnal approval. illnesses to be widely distributed on the
from physicians. Sources of funds for Inevitably, there is a trade-oÝ be- basis of considerably fewer data than
trials are similarly variable: in the vast tween speed and certainty. Judgments are required for full marketing approv-
PHASE 2 LIMITED
TRIALS AVAILABILITY
al. Typically the clinical trial process is allowed the drug to be approved sever- idence that they Òmay be eÝective.Ó If
nearing completion; the sponsor must al years earlier. the agency actually approves a drug for
be pursuing marketing approval dili- Exosurf, used to prevent or treat hya- marketing before trials are completed,
gently. More recently Òparallel trackÓ line membrane disease, a leading cause the data must demonstrate a Òreason-
regulations, issued in 1992, have let pa- of death among premature infants, ableÓ likelihood of beneÞt. In every
tients with HIV disease who do not qual- yields a more reassuring example of case, information gathering continues.
ify for clinical trials and have no thera- how expanded access mechanisms No matter who gets to use a drug as it
peutic alternatives receive experimental should be applied. After Burroughs winds through the testing process, it
therapies through their private physi- Wellcome shared interim data from pla- must eventually meet the most rigorous
cians very early in testing. cebo-controlled trials with the FDA, regulatory standard, providing Òsub-
Expanded access can provide treat- agency reviewers concluded that mar- stantial evidenceÓ that it works.
ment to people with otherwise untreat- keting approval was likely, assuming
able, life-threatening diseases. Yet it can that the Þndings were conÞrmed by a Accelerated Approval
also discourage enrollment in controlled closer review. The agency then agreed
studies and so prevent deÞnitive data
from ever being collected.
The use of ganciclovir to prevent cy-
with Burroughs WellcomeÕs decision to
halt the trial and allowed Exosurf to be
distributed while the company wrote
W hat indicators should researchers
measure to determine a drugÕs ef-
Þcacy? In the case of a life-threatening
tomegalovirus ( CMV ) retinitis, a sight- the marketing application. disease, mortality is the most conclu-
threatening opportunistic infection as- From the FDAÕs perspective, partici- sive indicator of a drugÕs impact, but
sociated with AIDS, oÝers a sobering pation in the analysis of interim data when a fatal disease progresses over
example of what can go wrong. The was potentially riskyÑif the deÞnitive the course of years, waiting for a dis-
widespread, uncontrolled distribution review had led to rejection, early termi- cernible eÝect on mortality may lead to
of unapproved ganciclovir convinced nation of the Exosurf study would have unacceptable delays. Instead research-
both patients and the scientiÞc com- resulted in a lost opportunity to secure ers sometimes look at surrogate end
munity that the therapy worked; re- more complete information. Yet the points, laboratory or clinical measure-
searchers thus could not conduct con- drug was important enough that the ments they believe to be correlated with
trolled trials in which patients risked chance seemed worth taking. Fortunate- a real beneÞt, even though they do not
receiving a placebo. After much discus- ly, full analysis resulted in approval just help a patient per se. For example, de-
sion, the sponsor, Syntex, gained ap- Þve months after the company submit- creased blood pressure or cholesterol
proval on the basis of a trial using a ted its application, well before the drug levels do not make a patient feel better
historical control (that is, by comparing would have been available otherwise. but may reduce the risk of a heart at-
the course of disease in patients who The standards of evidence that allow tack or stroke.
used ganciclovir with records of pa- a therapeutic product to be provided on In December 1992 the FDA formal-
tients who had had CMV retinitis before an expanded-access basis become more ized its procedures for accelerating the
ganciclovir became available). In retro- rigorous as a product approaches Þnal approval of drugs for serious diseases
spect, it seems clear that small random- approval. Drugs distributed on a paral- with no good alternative treatment on
ized trials conducted before the drug lel track must be Òpromising.Ó Treat- the basis of surrogate end points that
was widely distributed would have col- ment INDs, named nearer to the time are reasonably likely to be valid. In these
lected more rigorous eÛcacy data and of an approval decision, must show ev- cases, the sponsor must commit to con-
function. The available information was icity of the drugs studied or some oth-
20 unlikely to permit drug approval be- er factor is not known.
cause of the uncertainty of the CD4+ Another risk, not so well appreciated,
end point, but the absence of therapeu- is that researchers may abandon a drug
16
tic alternatives for patients not respond- prematurely because it has no eÝect on
ing to zidovudine argued against wait- a surrogate. Workers initially investi-
12
ing nearly a year for deÞnitive data from gating gamma-interferon expected it to
three controlled trials then under way. aÝect superoxide production and the
8 The agency therefore launched an in- destruction of bacteria by blood cells
LAURIE GRACE
tensive eÝort to obtain more informa- and thus to alleviate the life-threaten-
4 tion about the action of ddI on CD4+ ing infections that characterize chronic
cells and the relation between CD4+ granulomatous disease, a rare immune
0 cell counts and clinical outcome. In ad- disorder seen in children. Although the
1991 1992 1993 1994
dition, investigators took an unusual, drugÕs lack of inßuence on these labo-
unscheduled look at CD4+ data from ratory Þndings was determined within
APPROVAL for drugs to treat some life-
threatening diseases has been speeded patients in an ongoing trial. a few weeks, the trial continued as
up in recent years. The Þve drugs ap- On October 9, 1991, the FDA ap- scheduled. The results were surprising:
proved under accelerated procedures proved ddI for patients with no satis- after a year, gamma-interferon had
during the past four years typically factory options, primarily because the clearly reduced the rate of infection but
passed FDA review in less than half the compound had a positive eÝect on had no detectable eÝect on either sur-
time required of other new therapies. CD4+ cell count, and with the under- rogate. Only the clinical end point al-
standing that the trials would be com- lowed the drugÕs value to be identiÞed.
pleted. These ultimately produced per- Even with their drawbacks, surrogate
ducting controlled trials to verify clin- suasive clinical data that conÞrmed the data can be a valuable way to bridge the
ical beneÞts after approval. ( Usually early approval and also led to labeling gap between access (provided on the
such trials will be under way at the time changes broadening ddIÕs indication. basis of a promising surrogate) and an-
of approval.) If those beneÞts fail to ma- Even so, CD4+ cell counts have not been swers (secured when clinical beneÞt is
terialize, the FDA can remove the drug so closely correlated with the progres- established ). In AIDS, as in other life-
from the market quickly. This regula- sion of HIV disease and survival as threatening diseases, the search contin-
tion accepts a modest risk of approving many had hoped: a reminder that sur- ues for the best surrogates possible.
a drug that eventually proves ineÝec- rogate end points add uncertainty to
tive in return for the gain of making a drug evaluation. Flexible Trial Designs
drug for a serious illness available sig- The National Heart, Lung and Blood
he FDA is trying to encourage eÛ-
niÞcantly faster.
The review and approval of dideoxy-
inosine (ddI, an antiviral drug that acts
InstituteÕs Cardiac Arrhythmia Suppres-
sion Trial ( CAST ) shows how treacher-
ous surrogate end points can be. Re-
T cient drug development strategies
by increasing the range of trial designs
much like zidovudine, also called AZT ) searchers had found that individuals that may be acceptable. Researchers
was the model for the accelerated ap- with ventricular premature contractions may be able to expedite data gathering
proval regulation. In April 1991, Bristol- (an electrically abnormal heartbeat) af- without compromising quality. The
Myers Squibb, ddIÕs sponsor, submitted ter a heart attack were more likely to agency encourages investigators to
a marketing application to the FDA on die than patients without those con- meet with FDA representatives so that
the basis of data from the initial phase tractions. Because it seemed intuitively research plans can be reviewed; it is of-
of clinical studies, which included just reasonable to believe that suppressing ten possible to depart from traditional
170 adults and 98 children. Like most those contractions would reduce mor- practices and still maintain rigor.
early trials, these had not been designed tality, the results of the CAST trials were One approach to speeding data col-
to provide deÞnitive eÛcacy data. There a major surprise. The drugs under study lection is the Òlarge, simple trial.Ó Such
trials have relaxed criteria for eligibility,
enroll patients in a variety of care set-
tings and collect considerably less ex-
tensive baseline and outcome data than
LAURIE GRACE AND JANA BRENNING
A
lthough the science of chemistry can be enormously eÝective in steering Þlament therefore consists of a jumble
has made considerable progress a reaction down a preferred pathway. of waves from individual atoms. In a
in the past century, the main The idea of using lasers to drive reac- way, light from ordinary sources resem-
principles behind the industrial prac- tions is not revolutionaryÑin fact, at- bles a column of soldiers marching out
tice of chemistry have remained basi- tempts to do so began shortly after the of step with one another. Physicists say
cally unchanged. Methods for breaking Þrst laser was invented about 35 years that such light is incoherent, or out of
and re-forming chemical bonds still of- ago. These devices emit radiation of a phase. In contrast, atoms in lasers act
ten rely heavily on altering the temper- precise frequency, or color, and thus together. Consequently, all the waves
ature and pressure of the reaction or can impart a well-deÞned parcel of en- that make up the entire laser beam
adding a catalyst. This approach is of- ergy to a given target. Chemical bonds match perfectly, behaving like soldiers
ten ineÝective because it takes no ac- were seen as individual springs of dif- marching in step.
count of our understanding of the mo- ferent strengths, each of which vibrated Coherent light readily displays an
tions of molecules. As a result, bulk re- when given a certain amount of energy. important propertyÑconstructive and
actions are often ineÛcient, generating The hope was that lasers could be tuned destructive interference. The term Òin-
large quantities of useless by-products to attack a particular bond, weakening terferenceÓ does not necessarily mean
in addition to the desired materials. or breaking it and thereby encouraging that the waves disturb one anotherÕs tra-
Recently investigators have devised one product to form in lieu of another. jectories; it refers to the way all waves
new techniques, based on illuminating But this approach, called mode-selec- ( including those on water ) combine. In
chemical compounds with lasers, that tive chemistry, has seen only limited constructive interference, the crests or
can potentially control the paths taken success. In fact, it is doomed to fail for troughs of two (or more) waves meet
by reactions. These methods promise the vast majority of molecules because so that the wave heights or dips add to-
to alter the yields in selected ways by it assumes that chemical bonds are gether. The wave therefore becomes
exploiting an essential feature of quan- largely independent of one another. bigger and deeper ( in other words, the
tum mechanicsÑnamely, the wavelike Only a few compounds have bonds that amplitude of the wave increases). In
properties of both light and matter. meet this requirement. Rather most destructive interference, crest meets
The latest calculations show that lasers bonds are strongly interdependent: en- trough, and the wave is extinguished.
ergy readily ßows between them. As a
consequence, the energy imparted by Interfering Light Waves
the laser is distributed throughout the
PAUL BRUMER and MOSHE SHAPIRO
have been research collaborators since
they met at Harvard University more
molecule in a fashion similar to that re-
sulting from traditional, and far cheap-
er, sources.
T he phenomena of constructive and
destructive interference become ap-
parent if we shine a laser beam on an
than 20 years ago. Brumer is a professor The latest approach, called coherent opaque plate that has two narrow slits
in the chemical physics theory group at control, has grown out of research be- cut into it. Each slit acts as a new source
the University of Toronto and is current- gun in the mid-1980s, when investiga- of light waves. The waves emanating
ly a Killam Research Fellow. He recently
tors began to take a look at properties from each slit, which fan out, can then
received the Chemical Institute of Cana-
daÕs Palladium Medal, one of the organ- of laser light that had been ignored in interfere with each other. The interfer-
izationÕs highest awards. He earned his earlier considerations of chemical con- ence can be rendered visible if we place
Ph.D. from Harvard. Shapiro obtained his trol. One such property is the coher- a viewing screen just beyond the slits.
Ph.D. from the Hebrew University of Je- ence of laser light. ÒCoherenceÓ relates The screen will show a pattern of dark
rusalem. The Jacques Mimran Professor to the way that atoms emit light. In con- and light stripes, or fringes. The light
of Theoretical Chemical Physics at the ventional light sources, such as house- areas correspond to constructive in-
Weizmann Institute of Science in Rehov-
hold bulbs, electricity heats up the Þl- terference; the dark areas, destructive
ot, Israel, he has also held several visit-
ing appointments throughout the U.S.,
ament, exciting tungsten atoms. The interference.
Europe and Canada. The authors grate- atoms spontaneously lose this energy The latest technique to control mole-
fully acknowledge funding and support by giving oÝ a bit of light. Each atom, cules takes advantage of the fact that
from the U.S. OÛce of Naval Research. however, emits light independently of light is not the only entity that displays
the other atoms. The total light from the interference. In accordance with a basic
LASER BEAMS Þred into a vessel Þlled with paired sodium interfere with one another. Each molecule subsequently
atoms, or dimers, can control the breakup of the molecules. breaks up into one sodium atom in its normal state and an-
The beams, each of a diÝerent frequency (appearing as yel- other sodium atom in any of several excited states. The par-
low and red ), place the molecules into quantum states that ticular state formed is governed by the degree of interference.
10 PERCENT A + B C
PHASE-SHIFTED
PHOTON
YIELD
A
B C 10 PERCENT A
B + C
DIMITRY SCHIDLOVSKY
90 PERCENT A +
AMPLITUDE-SHIFTED B C
PHOTON
CONTROLLED CHEMICAL DISSOCIATION can be accomplished third the energy. The wave functions associated with the ex-
if a compound is irradiated simultaneously by two types of citation by each type of photon quantum-mechanically inter-
laser photons, one of which is more energetic than the other. fere with one another. The extent of the interferenceÑand
For instance, the molecule ABC can dissociate if it absorbs a hence, the yieldÑis controlled by adjusting the relative ampli-
photon of a certain energy, or three photons, each with one tude ( the height of the wave) and the phase of the laser light.
TOTAL OUTPUT photons from one laser beam and four the phase of the laser light. Given two
photons, each of half the energy, from arbitrary laser sources, we generally do
the second laser beam. Similarly, we not know the extent to which the light
have shown that using one photon in from one will be in phase with the oth-
conjunction with two photons, each of er. In addition, the phase diÝerence is
half the energy, can be used to control aÝected by any instabilities in the equip-
the direction that molecules take when ment. An unstable phase diÝerence be-
ser beams would provide an enormous they leave the reaction region. Such an tween the two lasers reduces the degree
range of control over the reaction. We ability may render the separation of of interference and control.
would be able to vary the amount of products easier and thereby boost the Sophisticated optical techniques have
energetic bromine produced to ensure eÛciency of the reaction. the potential to eliminate such phase
that it accounted for between 25 and problems. For example, photons may
95 percent of the total product formed. Current Limitations be generated by passing light at one
This degree of control far exceeds the frequency through a particular material
hopes of conventional industrial chem-
istry, which seeks to improve reaction
selectivity by about 10 percent. Experi-
A lthough the concepts behind con-
trolling reactions with coherent
light apply to a wide range of isolated
that is thereby induced to emit light at
another frequency. This process yields
two light Þelds whose phase relations
ments by Daniel S. Elliott of Purdue molecules, there are at least two obsta- are well deÞned. We can further con-
University have demonstrated the suc- cles to their immediate broad applica- trol the phase diÝerences between the
cess of this approach for controlling the tion. One is that the eÛciency of laser two light sources by temporarily slow-
ionization of atoms. Robert J. Gordon control drops substantially when the ing one light beam relative to the other.
MOLECULE A
A C C
A C +
B B
B
C
A +
DIMITRY SCHIDLOVSKY
PULSE 1 PULSE 2
TWO-PULSE LASER STRATEGY can also govern the yield of a the characteristics of the molecule and the laser. The second
chemical reaction. The Þrst pulse places the molecule in a su- pulse causes the molecule to break up. The diÝerences in
perposition stateÑin other words, it sets the molecule vibrat- yield are controlled by varying the time between pulses and
ing and rotating in a particular way. The motion depends on by changing the frequencies that make up the pulses.
P erhaps the most unusual aspect of quantum mechan- nations of maxima and minima meeting at the screen.
ics is that, under certain circumstances, matter be- Now consider replacing the light with particles such as
haves just as waves do. In particular, it displays interfer- electrons, atoms or molecules. Classical intuition would
ence. Because this property is essential to the ability of suggest that the emergent pattern on the screen would re-
lasers to control chemical reactions, some of the mathe- semble two nearly rectangular blobs, resulting from the
matics and more technical quantum-mechanical concepts direct passage of particles through the slits. In fact—and
underlying the phenomenon may be of interest. this is the essential feature of quantum mechanics—the
Interference arises from the way waves add together. observed pattern can clearly display an interference pat-
The rule for combining many different waves is first to tern. That is, the particles can show behavior characteris-
sum their amplitudes (the height of the wave) and then tic of waves (as long as the particles are themselves “co-
square the result. Consider the interference of two waves herent”—that is, prepared so that they have well-defined
whose amplitudes at a given position and time are a and wave functions). There is simply no way to explain these
b. The intensity of each observations on the ba-
wave is a 2 and b 2. The sis of a classical theory
DIMITRY SCHIDLOVSKY
SCREEN
combined amplitude, c, of particles.
is the sum c = a + b, INTERFERENCE For this reason, the
PATTERN
and the combined in- proper description of
tensity is c 2 = (a + b) 2 = particle dynamics in
a 2 + b 2 + 2ab. quantum mechanics en-
Note that the com- tails a wave function
bined intensity is not that, as in the case for
merely the sum of the LASER waves, is described by
intensities of each wave both an amplitude and
(which would be a 2 + a phase. In the double-
b 2 ), but that an addi- slit experiment applied
tional interference term, to particles, quantum
SLITS
2ab, contributes. If a wave functions arriving
and b are both positive at the screen via the
or both negative, this INTERFERENCE FRINGES are made if coherent light passes through slits interfere with one
interference term is two slits. Particles such as molecules also interfere in this way. another. Most signifi-
positive. The resulting cant is another unusual
intensity, c 2, is thus greater than the simple sum of inten- feature of the quantum world: interference arises because
sities of each wave. The interference in this case is said to we do not know through which of the slits the particles
be constructive. If a is positive and b is negative (or vice passed. If we do record such information, the interference
versa), the interference term is negative, and the resulting pattern disappears.
intensity is smaller than the simple sum of the individual The two-slit experiment embodies the fundamental quan-
intensities. Interference in this case is said to be destructive. tum principle that two (or more) phase-preserving routes
The variation in intensity produced by interfering light that a molecule can take to some final state can be made
waves can be seen in the famous double-slit experiment, to interfere. In the case of the one-photon plus three-pho-
in which a beam of coherent light passes through two slits ton experiment mentioned in the main body of the article,
and onto a screen (diagram ). The bright regions on the we do not know which of the two possible excitation
screen arise from the constructive interference between routes leads to the observed final state. Hence, the two
the two beams of light passing through the two slits; dark routes interfere. This optically induced interference forms
regions result from the destructive interference. Intensi- the foundation for coherent radiative control of molecular
ties in between these two extremes arise from the combi- processes.
Another way to control phase prob- trol over the various products formed quencies and, hence, of a collection of
lems relies on intense laser beams, the in the dissociation of paired sodium photons with diÝerent energies. Such
focus of many recent investigations. molecules. light also has a perhaps counterintuitive
With Zhidang Chen of the University of property. The briefer the pulse, the
Toronto, we have shown that such la- Using Pulses broader the range of energies within it.
sers make it possible to bypass the need This property plays a major role in
for radiation having well-deÞned and
carefully controlled phases. In addition,
such strong-Þeld methods have the po-
B ecause interference of molecular
pathways is the key to governing re-
actions, any laser scenario that induces
pulsed-laser methods for controlling
the outcomes of chemical reactions. By
delivering a range of energies, a pulse
tential to increase substantially the ab- such interference may serve as a means can induce motion (such as vibration
solute yield of the reaction and to over- of controlling reactions. Instead of shin- or rotation) in a molecule, which in turn
come unwanted collisional eÝects. A ing two steady beams on a target, one aÝects the way it interacts with other
particular scheme has just been used might use ultrashort pulses of laser light pulses. Ordinarily a molecule such
experimentally at the Weizmann Insti- light. Modern lasers can generate bursts as ABC exists at a speciÞc (that is, quan-
tute of Science in Israel by Irit Sofer, as short as 10 Ð14 second. Unlike contin- tized ) energy value. A system at one of
Alexander Shnitman, Ilya Golub, Am- uous-wave radiation, a light pulse is these Þxed energies resides in a so-called
non Yogev and us to demonstrate con- made up of a collection of distinct fre- stationary state and does not move over
O
ver millions of years in a vast water with impressive grace and appar- Similarly, detailed observations have
and often hostile realm, Þsh ent ease, playfully bursting through the shown that Þsh that depend on aquatic
have evolved swimming capa- waves as they follow ships cruising at agility for their survival can reverse di-
bilities far superior in many ways to 20 nautical miles per hour (knots), or rection without slowing down and with
what has been achieved by nautical sci- about 23 mph. Whereas records of the a turning radius only 10 to 30 percent
ence and technology. Instinctively, they maximum speeds of Þsh are not always of the length of their bodies. For com-
use their superbly streamlined bodies reliable and are often quite contentious, parison, maneuvering ships must re-
to exploit ßuid-mechanical principles marine biologists have reported that yel- duce their speed by more than 50 per-
in ways naval architects today can only lowÞn tuna caught on a Þshing line can cent, and their turning radius is at least
ROBERTO OSTI
dream about, achieving extraordinary swim at speeds of at least 40 knots. The 10 times larger than the corresponding
propulsion eÛciencies, acceleration and aggressive pike overcomes its prey with value for Þsh.
maneuverability. short bursts of acceleration that can ex- Nevertheless, and despite huge po-
Dolphins, for example, dart through ceed that of gravity by about 20 times. tential payoÝs, relatively little work has
Delphine Mystery
COD
IAN WORPOLE
DAVE BARRETT
damental diÝerence between steady cerning the Strouhal number, suggest recapturing kinetic energy from a wake.
airplane ßight and ßapping propulsion. that a properly designed foil could be a Fish instinctively exert precise and
The basic principles of Þxed-wing ßight very attractive propulsor for ships, mo- eÝective control of the ßow around
require that, to avoid a stall, wings gen- tor yachts and underwater vehicles. Giv- their bodies to extract energy from
erally be kept to an angle of attack well en its natural advantages and the fact waves, turbulence and even their own
below 15 degrees. Noticeable stall did that development of the proper motors wakes. They have also evolved ways of
not occur with the foil until the angle and gears is well within todayÕs techno- controlling the ßow so as to enhance
exceeded 30 degrees. logical capabilities, this foil might be their turning and starting. The underly-
These results show that the criteria the Þrst Þsh-inspired technological ap- ing principles are not unique to Þsh or
that indicate a stall for Þxed wings do plication. The use of an even number even to ßapping propulsion. A propel-
not apply to a ßapping foil. True, what of countermoving foils, properly posi- ler mounted on a ship is somewhat
causes a stall in both cases is the sud- tioned, could minimize unpleasant sway- more eÛcient than one tested in a tank,
den formation of uncontrolled vortic- ing or vibration. Of course, future ship- because the moving propeller recovers
esÑabove and behind the wings, in the builders would also have to address some of the energy from the wake. The
case of aircraft, disrupting the normally structural reliability, the hydrodynamic phenomenon is routinely exploited by
smooth ßow over them. With the ßap- shapes of sterns and other variables. ship designers.
ping foil, however, vortices do not in Fish, marine animals and their me-
and of themselves cause a stall. In fact, Trick of the Tail chanical imitators, however, are much
vorticesÑproperly controlled and ar- better suited to this kind of control.
rangedÑare essential to a foilÕs eÛcient
operation, so it should be no surprise
that they can be controlled over a wid-
T he coincidence of high thrust and
eÛciency is not the only advantage
of a ßapping foil. It also oÝers the pos-
Frolicking and leaping in the wakes of
ships for miles on end, dolphins are
clearly recovering energy by position-
er range of angles to produce useful sibilities of more ßexible operation, ing their bodies and ßapping their tails
thrust. more maneuverability and, most in- appropriately, as Gray noted decades
These Þndings, along with those con- triguingly, tempting opportunities for ago and as Neil Bose of Memorial Uni-
ticulated foam and conformal Lycra al- tices. Such capabilities could prove in- dolphin and the tuna are both fast and
low smooth ßexing and keep stray tur- valuable even in oceanographic research, ßex their bodies in similar ways while
bulence to a minimum. We attached the where underwater vehicles must some- swimming, there are signiÞcant diÝer-
entire assembly to a carriage, on which times operate in forbidding or conÞned ences in the details of swimming as
we mounted all the motors and control environments. Near thermal vents, for well. Are they both optimal solutions?
and communication equipment. A sin- example, temperatures can shoot up If one is better than the other, is the su-
gle strut encloses the cables for data to hundreds of degrees Celsius in the periority limited to certain situations?
and power. space of a few feet or in a few seconds. More important, as far as we are con-
Several sensors along the side of In cluttered spaces, too, agility can cerned, is there an even better design
RoboTuna record ßow pressure, just as sometimes stave oÝ a collision or a than either of them for swimming?
Þsh use their Òlateral lineÓ sense organs catastrophic failure. These are among the questions we
to detect pressure variations. Along with hope to address with our next robotic
force and motion transducers, they Nothing Like the Real Thing tuna. The state of the art in mechanical
permit detailed evaluation of swimming systems suggests that it will take our
forces and propulsive eÛciency. Simul-
taneous measurement of forces lets us
directly link ßow features to swimming
T he more sophisticated our robotic-
tuna designs become, the more ad-
miration we have for its ßesh-and-
best eÝorts to approach the breathtak-
ing abilities of its living model, but we
will be patient. After all, in the span of
performance and also control the ßow blood model. Aware that we will never a few years we are learning processes
to enhance the modelÕs propulsion and match the perfection of design of the that took eons to develop.
maneuvering. Soon the side-mounted living creature, we strive instead to un-
pressure transducers will enable us to cover natural, useful mechanisms opti-
experiment with closed-loop control of mized by millions of years of evolution.
FURTHER READING
vorticity, so that RoboTuna, like its nat- Once identiÞed, a kind of reverse engi-
ural counterpart, will be able to move neering may enable us to devise novel ANIMAL LOCOMOTION. James Gray. Wei-
its tail in response to oncoming vortic- ways of using these mechanisms. In denfeld & Nicolson, 1968.
SWIMMING AND FLYING IN NATURE, Vol.
es and the ßow around its body. We time, these biologically inspired cre-
2. Edited by Theodore Y. T. Wu, C. J.
made the ßow around the robotic Þsh ations may even outperform their natu- Brokaw and Christopher Brennen.
visible by using either dyes or a laser ral antecedents in useful waysÑfor in- Plenum Press, 1975.
beam that causes microscopic particles stance, in surveying a stretch of seaßoor. LOCOMOTION BY SCOMBRID FISHES: HY-
in the water to phosphoresce. This goal is the guiding principle of DROMECHANICS MORPHOLOGY AND BE-
In a few months, another generation the emerging science of biomimesis. By HAVIOR. J. J. Magnuson in Fish Physiolo-
of RoboTuna will take shape. We expect focusing research eÝorts and guiding gy. Edited by W. S. Hoar and D. J. Ran-
to begin building a free-swimming mod- the selection of parameters, details of dall. Academic Press, 1978.
OPTIMAL THRUST DEVELOPMENT IN OS-
el, borrowing on the technology devel- the behavior and instincts of highly CILLATING FOILS WITH APPLICATION TO
oped for the existing robot. This suc- adapted, successful creatures can be a FISH PROPULSION. G. S. Triantafyllou, M.
cessor will be used to develop still more great asset in developing certain robots S. Triantafyllou and M. A. Grosenbaugh
advanced technologies, based on our and other useful systems. Our project in Journal of Fluids and Structures, Vol.
growing understanding of ßow-control has required us to pose and answer 7, No. 2, pages 205Ð224; February 1993.
mechanisms, for possible application fundamental questions about the me- ACTIVE VORTICITY CONTROL IN A SHEAR
to commercial and naval vehicles. chanics of swimming. FLOW USING A FLAPPING FOIL. R. Gopal-
krishnan, Michael S. Triantafyllou,
More important, it will be a test bed For example, is Þsh swimming a sim-
George S. Triantafyllou and David Bar-
for improving maneuvering and fast ple perfection of hydrodynamic princi- rett in Journal of Fluid Mechanics, Vol.
starting of vehicles using the fast gen- ples, constrained only by the mechani- 274, pages 1Ð21; September 10, 1994.
eration and manipulation of large vor- cal limitations of muscle? Although the
P
atients stricken with cancer feel and damage healthy tissue in its vicini- velops primarily because cells suÝer ir-
as if they have been invaded by ty. What is worse, it can invade the bar- reversible damage to particular classes
an alien force. Yet malignancies riers that separate one organ from an- of genes. It is also creating opportunities
arise from our own tissue. In fact, the other and can metastasize, establishing for improved diagnosis and therapy.
weight of evidence today indicates that new colonies at distant sites. The emerging view of tumor progres-
cancers generally derive from a single Studies carried out over the past 20 sion reßects a convergence of several
cell that is changed dramatically by a years have begun to identify many of lines of research, the oldest of which
series of genetic alterations. the genes that take part in this progres- still involves painstakingly looking at
A healthy cell has a well-deÞned sion from normalcy to cancer. The on- cells through a microscope. By 1914, for
shape and Þts neatly within the ordered going research is conÞrming and ex- instance, the German cytologist Theo-
array of cells surrounding it. It responds tending early proposals that cancer de- dor Boveri had concluded from such
to the dictates of its environment, giving
rise to daughter cells solely when the
balance of stimulatory and inhibitory
signals from the outside favors cell di-
vision. But the process of replication, or
growth, carries the constant hazard of
genetic mutations: random changes that
can impair the regulatory circuits of a
cell. If a single mutation occurs, the
newly damaged cell, which may look
normal and be slightly less responsive
to external messages, may occasionally
undergo unscheduled cell division.
Eventually, an accumulation of genet-
ic damage can cause a daughter cell to
become quite deaf to external messages
and to display the signs of malignancy.
In particular, it loses its distinctive shape
and boundaries, ceases to respond to
growth-inhibiting signals and gains the
ability to replicate uncontrollably. The
resulting mass, in turn, can compress
TOMO NARASHIMA
STIMULATORY PATHWAY INHIBITORY PATHWAY
GROWTH-
GROWTH INHIBITING
FACTOR FACTOR
RECEPTOR RECEPTOR
INTRACELLULAR INTRACELLULAR
RELAYS RELAYS
DNA
NUCLEUS
NUCLEUS
NORMAL CELL REPRODUCES ITSELF (sequence at top) in re- overproduced or overactive. In one example, mutation of a
sponse to stimulation by external growth factors ( green); it particular ras gene can lead to synthesis of a hyperactive ras
stops dividing in response to inhibitory factors (red, far right ). protein (inset at left ). Many other cancer-related genes code
For either reaction to occur, messages from the factors must for proteins in inhibitory pathways (right panel ) and are of-
be relayed deep into the target cell (large panels). Many can- ten called tumor suppressors. Damage to these genes can
cer-causing genes are abnormal versions of ones that code promote cancer if the defects prevent inhibitory proteins
for proteins in stimulatory pathways (left panel ). The altered from being made or functioning properlyÑas often occurs
genes, called oncogenes, cause stimulatory proteins to be when the p53 gene is mutated (inset at right ).
MALIGNANT CELL
is p53, which resides on chromosome be expected if the added mutations ac- Cancer of the brain is deÞned some-
17 and is now known to be involved in counted for the increased aggressive- what diÝerently than it is in other tis-
many diÝerent cancers. The normal pro- ness of a cancer. For instance, some sues. In that organ, cells do not need to
tein product of this gene functions in polyps carried a ras mutation without invade connective tissue or metastasize
several biochemical pathways, includ- a p53 defect, but the cancers growing in order to be lethal; sadly, proliferation
ing those enabling a cell to repair dam- from the polyps had both mutations. at a site critical to survival can some-
age to DNA. The other is a geneÑprob- As yet, there is no strong evidence that times be enough to kill a patient. Hence,
ably DCC ( for deleted in colorectal can- mutation of ras, p53 and DCC genes most masses in the brain are called can-
cer )Ñthat resides on chromosome 18. must happen in any particular order for cers. CaveneeÕs group examined pro-
DCC codes for a protein that appears a polyp to become cancerous, although gression of astrocytomas from their less
on the cell surface and helps colon cells the ras mutation seems to come Þrst malignant to more malignant stages, as
stick to one another. fairly often. determined by the size and shape of the
The discovery that genetic changes in tumors and by the structure of their
the APC gene occur early and persist, Brain Tumors Reveal Their Secrets constituent cells.
whereas other changes appear only in When the investigators began this
later stages, Þts well with the theory of
clonal evolution. But that conclusion
was initially statistical and based on ex-
I n spite of these encouraging Þndings,
study of colon cancer has a major an-
alytical limitation. To truly demonstrate
work in 1987, they did not have the
blueprint of genetic change that was
emerging for colon cancer. They there-
amining tissues removed from many that a given clone of cells is undergoing fore began by laying the groundwork
diÝerent patients. That approach could progressive changes in its genes, one for future studies of individual patients.
not demonstrate conclusively that mu- needs to examine the same tumor over They obtained tumors from many dif-
tations appearing in one generation of time. In the case of colon cancer, tumors ferent patients, grouped them accord-
cells are passed to later generations of are almost always removed at the earli- ing to stages, or grades, of advancing
those same cells. Another strategy, how- est stage of detection. Such practice disease, and compared the genetic de-
ever, provided more convincing results. makes good clinical sense, but it pre- rangements found in each stage.
Sometimes the polyp from which a vents sequential observations. This con- Over the next four years they made
cancer has emerged can be identiÞed at sideration led Cavenee to seek out a good headway. They learned, for in-
the edge of a cancer. By comparing the disease in which removal of a tumor is stance, that tumors of every grade had
DNA in a polyp with that in its adjacent sometimes followed by the reappear- inactivating alterations in chromosome
cancer, Vogelstein showed that every ance of the tumor in a more aggressive 17, in a gene they had not yet identiÞed.
mutational hit found in a polyp also ap- form at the same site. In 1987, while he Moreover, the proportion of tumors dis-
peared in the corresponding cancer, as was at the Ludwig Institute for Cancer playing the mutation in the lowest stage
would be expected if the tumor formed Research at McGill University, he and his was equal to that in all other stages;
by clonal evolution. Further, the cancer co-workers settled on cancers known as this pattern is a sign that the mutation
invariably included mutations that were astrocytomasÑthe most common tu- came early and was retained. If a muta-
not found in the polyp, as would also mors that originate in the brain. tion generally occurred later in disease,
the frequency would rise in the later some as well. Middle- and late-stage as- tein products are involved in regulating
stages. By the end of the 1980s Vogel- trocytomas, but not early ones, often cell division. Disappearance of any of
steinÕs laboratory established that mu- showed a loss in both copies of this these genes could potentially contrib-
tations in the p53 gene, on chromosome chromosome. Thus, the deletion proba- ute to a variety of cancers.
17, were among the most common al- bly encouraged progression to middle- The tissue studies also extended re-
terations in human cancer. Subsequent stage tumors from a lesser stage. The ports by Axel Ullrich of Genentech,
analysis of CaveneeÕs tissue samples lost region contains a cluster of genes Michael D. WaterÞeld of the Ludwig In-
conÞrmed his growing suspicion that that code for proteins known as inter- stitute in London and Joseph Schlessin-
the chromosome 17 mutation was ac- ferons. Such proteins can draw the at- ger of the Weizmann Institute of Science
tually a defect in the p53 gene. tention of the immune system to dis- in Israel that chromosomes in astrocy-
Aware that a particular region of eased cells, and so elimination of their tomas often carry more than one copy
chromosome 9 was deleted in other genes presumably helps cancer cells of the gene specifying the receptor for
kinds of brain tumors, C. David James evade immune destruction. The missing epidermal growth factor. Because each
on CaveneeÕs team, in conjunction with region may additionally include two copy can be used to make the protein,
V. Peter Collins of the Ludwig Institute newly discovered genes, called multiple cells will carry extra receptors on their
in Stockholm, examined this chromo- tumor suppressors 1 and 2, whose pro- surface. That abundance, in turn, can
cause cells to overreact to the presence
of the growth factor. This alteration
seems to participate in bringing tumors
from a middle to a late stage of disease.
Finally, CaveneeÕs group found that
virtually all the end-stage tumors exam-
ined were missing one copy of chromo-
some 10 and that the loss was rare in
earlier stages. This pattern says the loss
is probably involved in advancement to
the most virulent stage. Regrettably,
though, we do not yet know which gene
or genes on the lost chromosome are
most important to the progression.
These results suggested by 1991 that
formation of brain tumors involves, at
a minimum, inactivation of the p53
PETER C. NOWELL University of Pennsylvania
NORMAL TISSUE LOW-GRADE TUMOR HIGHER-GRADE TUMOR MOST AGGRESSIVE FORM OF TUMOR
form (right ). Other genes not listed here play roles as well. that both copies of a tumor suppressor gene have been dam-
Unless otherwise indicated, the term Ògene lossÓ indicates aged or deleted. The images show magniÞed slices of tissue.
cause matched pairs of tumors are hard tered cells almost certainly enhance When the mutations follow in a fairly
to obtain. A patient seen initially at one their genetically deÞned deregulation. set sequence, their identiÞcation in a
institution may be cared for elsewhere Questions remain. Why do cell types patientÕs tumor should be of value for
when the cancer returns. Also, physi- diÝer in the mix of mutations they re- clarifying the stage of disease and thus
cians do not remove tumors that reap- quire in order to become cancerous? for tailoring therapy to the individualÕs
pear if it is thought that surgery is un- And how is it possible for Þve or more needs. In addition, knowledge of the
likely to extend survival. Luckily, howev- mutations to accumulate in cells? After genes that are mutated in a primary tu-
er, two distinguished cliniciansÑMark L. all, the probability is actually quite mor may make it possible to detect re-
Rosenblum of the University of Califor- small that any given cell bearing a per- currences of some cancers earlier than
nia at San Francisco and Karl Schwech- manent mutation in a cancer-related is now possibleÑby spotting mutations
heimer of Albert Ludwigs University in gene will independently gain another that have occurred in tissues not yet
Freiberg, GermanyÑhad come forward mutation in such a gene. displaying detectable masses.
with collections of frozen tissue that Newly discovered genetic aberrations Expanded understanding of the genet-
included a few matched sets. found in a second form of inherited co- ic bases of cancer can also be expected
To CaveneeÕs satisfaction and delight, lon tumors (hereditary nonpolyposis co- to lead to the introduction of drugs that
the genetic analysis of these tissuesÑ lon cancer ) may oÝer a partial answer will counteract the eÝects of selected
done in collaboration with David Sid- to the last question. The aÝected genes mutations and thereby slow tumor de-
ransky in VogelsteinÕs groupÑfulÞlled specify proteins responsible for identi- velopment or halt it altogether. Some
the predictions of the theory of clonal fying and repairing mistakes made evidence suggests it may not be neces-
evolution. The initial tumors possessed when DNA in a replicating cell is cop- sary to correct the eÝects of every mu-
fewer mutations than did the recurrenc- ied. If these repair genes themselves tation; doing so for one or two genes
es. These alterations included one or are damaged, the number of mutations may well prove to be suÛcient for tam-
more of the genetic hits (such as dam- passed to daughter cells will go up dra- ing renegade cells.
age to chromosome 17) that had been matically. The daughter cells may then The process by which normal cells
identiÞed in the low-grade tumors ana- deliver DNA carrying still more muta- become cancerous and grow ever more
lyzed previously. And, most signiÞcant, tions to their progeny. Defects in repair dangerous is undoubtedly even more
the corresponding high-grade versions genes may thus play a role in making complicated than has been discovered
possessed each alteration found in the late-stage tumors highly aggressive. so far. But continued investigation of
primary tumor as well as additional de- They may even account for the aston- the genetic changes underlying speciÞc
fects (of the kinds identiÞed in the ear- ishingly fast rate at which some tumors cancers seems a rational way to tease
lier studies). For reasons that are not arise and become killers. apart many of those complexitiesÑand
obvious, progression of astrocytomas Mutations in certain genes can also be to gain new leads for treatment.
seems to follow a more deÞned se- especially devastating if the mutations
quence of genetic changes than is ap- have multiple eÝects. As a case in point,
parent in colon cancer. damage to the p53 gene can apparently
FURTHER READING
do more than release a brake on prolif-
Next on the Agenda eration. Certain mutations seem to re- THE CLONAL EVOLUTION OF TUMOR CELL
duce the ability of cells to limit blood POPULATIONS. Peter C. Nowell in Science,
Vol. 194, pages 23Ð28; October 1, 1976.
T he collected results we have de-
scribed oÝer strong support for the
idea that cancer develops and becomes
vessel formation. As extra vessels grow
in a tumor, they help to nourish the
mass and to serve as conduits through
GENETIC AND EPIGENETIC LOSSES OF HET-
EROZYGOSITY IN CANCER PREDISPOSI-
TION AND PROGRESSION. Heidi J. Scra-
more dangerous primarily because cells which malignant cells can spread to dis- ble, Carmen Sapienza and Webster K.
in a single lineage accumulate defects tant sites. In parallel, the abnormal pro- Cavenee in Advances in Cancer Re-
in genes that normally regulate cell pro- teins yielded by the altered gene may search, Vol. 54, pages 25Ð62; 1990.
liferation. Changes in other kinds of aid tumor cells in resisting the destruc- A GENETIC MODEL FOR COLORECTAL TU-
genes, many of which have not yet been tive eÝects of radiation. MORIGENESIS. Eric R. Fearon and Bert
Vogelstein in Cell, Vol. 61, No. 5, pages
identiÞed, presumably facilitate the As investigators gain clarity on the
759Ð767; June 1, 1990.
ability of tumors to grow, invade local speciÞc groups of genetic changes that TUMOR SUPPRESSOR GENES. Robert A.
tissue and establish distant metastases. lead to and exacerbate particular forms Weinberg in Science, Vol. 254, pages
Hormones and other factors in the en- of cancer, their insights should point the 1138Ð1146; November 22, 1991.
vironment around the genetically al- way to practical beneÞts for patients.
by Frans B. M. de Waal
A
t a juncture in history during The species is best characterized as logical continuity is true, the bonobo
which women are seeking equal- female-centered and egalitarian and as may have undergone less transforma-
ity with men, science arrives with one that substitutes sex for aggression. tion than either humans or chimpan-
a belated gift to the feminist movement. Whereas in most other species sexual zees. It could most closely resemble the
Male-biased evolutionary scenariosÑ behavior is a fairly distinct category, in common ancestor of all three modern
Man the Hunter, Man the Toolmaker the bonobo it is part and parcel of social species. Indeed, in the 1930s Harold J.
and so onÑare being challenged by the relationsÑand not just between males CoolidgeÑthe American anatomist who
discovery that females play a central, and females. Bonobos engage in sex in gave the bonobo its eventual taxonom-
perhaps even dominant, role in the so- virtually every partner combination (al- ic statusÑsuggested that the animal
cial life of one of our nearest relatives. though such contact among close fami- might be most similar to the primogen-
In the past few years many strands of ly members may be suppressed ). And itor, since its anatomy is less specialized
knowledge have come together con- sexual interactions occur more often than is the chimpanzeeÕs. Bonobo body
cerning a relatively unknown ape with among bonobos than among other pri- proportions have been compared with
an unorthodox repertoire of behavior: mates. Despite the frequency of sex, the those of the australopithecines, a form
the bonobo. bonoboÕs rate of reproduction in the of prehuman. When the apes stand or
The bonobo is one of the last large wild is about the same as that of the walk upright, they look as if they
mammals to be found by science. The chimpanzee. A female gives birth to a stepped straight out of an artistÕs im-
creature was discovered in 1929 in a single infant at intervals of between Þve pression of early hominids.
Belgian colonial museum, far from its and six years. So bonobos share at least Not too long ago the savanna baboon
lush African habitat. A German anato- one very important characteristic with was regarded as the best living model
mist, Ernst Schwarz, was scrutinizing a our own species, namely, a partial sep- of the human ancestor. That primate is
skull that had been ascribed to a juve- aration between sex and reproduction. adapted to the kinds of ecological con-
nile chimpanzee because of its small ditions that prehumans may have faced
size, when he realized that it belonged A Near Relative after descending from the trees. But in
to an adult. Schwarz declared that he the late 1970s, chimpanzees, which are
had stumbled on a new subspecies of
chimpanzee. But soon the animal was
assigned the status of an entirely dis-
T his Þnding commands attention be-
cause the bonobo shares more than
98 percent of our genetic proÞle, mak-
much more closely related to humans,
became the model of choice. Traits that
are observed in chimpanzeesÑinclud-
tinct species within the same genus as ing it as close to a human as, say, a fox ing cooperative hunting, food sharing,
the chimpanzee, Pan. is to a dog. The split between the hu- tool use, power politics and primitive
The bonobo was oÛcially classiÞed man line of ancestry and the line of the warfareÑwere absent or not as devel-
as Pan paniscus, or the diminutive Pan. chimpanzee and the bonobo is believed oped in baboons. In the laboratory the
But I believe a diÝerent label might to have occurred a mere eight million apes have been able to learn sign lan-
have been selected had the discoverers years ago. The subsequent divergence guage and to recognize themselves in a
known then what we know now. The old of the chimpanzee and the bonobo lines mirror, a sign of self-awareness not yet
taxonomic name of the chimpanzee, P. came much later, perhaps prompted by demonstrated in monkeys.
satyrusÑwhich refers to the myth of the chimpanzeeÕs need to adapt to rela- Although selecting the chimpanzee
apes as lustful satyrsÑwould have been tively open, dry habitats [see ÒEast Side as the touchstone of hominid evolution
perfect for the bonobo. Story: The Origin of Humankind,Ó by represented a great improvement, at
Yves Coppens; SCIENTIFIC AMERICAN, least one aspect of the former model
May 1994] . did not need to be revised: male supe-
FRANS B. M. DE WAAL was trained as In contrast, bonobos probably nev- riority remained the natural state of af-
an ethologist in the European tradition, er left the protection of the trees. Their fairs. In both baboons and chimpan-
receiving his Ph.D. from the University present range lies in humid forests zees, males are conspicuously dominant
of Utrecht in 1977. After a six-year study south of the Zaire River, where perhaps over females; they reign supremely and
of the chimpanzee colony at the Arn- fewer than 10,000 bonobos survive. often brutally. It is highly unusual for a
hem Zoo, he moved to the U.S. in 1981 to ( Given the speciesÕ slow rate of repro- fully grown male chimpanzee to be
work on other primate species, includ- duction, the rapid destruction of its dominated by any female.
ing bonobos. He is now a research pro-
tropical habitat and the political insta- Enter the bonobo. Despite their com-
fessor at the Yerkes Regional Primate Re-
search Center in Atlanta and professor bility of central Africa, there is reason mon nameÑthe pygmy chimpanzeeÑ
of psychology at Emory University. for much concern about its future. ) bonobos cannot be distinguished from
If this evolutionary scenario of eco- the chimpanzee by size. Adult males of
BONOBO FEMALE interacts with an infant. Juvenile bonobos years. They are extremely well tolerated by adults, who have
depend on their mothers for milk and transport for up to Þve rarely been seen to attack or threaten them.
the smallest subspecies of chimpanzee zee it may be older than 40 in the wild ers or almost falling. She seems to be
weigh some 43 kilograms (95 pounds ) and close to 60 in captivity. imposing a rule on herself: ÒI cannot
and females 33 kilograms (73 pounds), Fruit is central to the diets of both look until I lose my balance.Ó Other apes
about the same as bonobos. Although wild bonobos and chimpanzees. The and monkeys also indulge in this game,
female bonobos are much smaller than former supplement with more pith but I have never seen it performed with
the males, they seem to rule. from herbaceous plants, and the latter such dedication and concentration as
add meat. Although bonobos do eat in- by bonobos.
Graceful Apes vertebrates and occasionally capture Juvenile bonobos are incurably play-
and eat small vertebrates, including ful and like to make funny faces, some-
LAURIE GRACE
AGO their genital swellings laterally together,
HUMANS OLD WORLD emitting grins and squeals that proba-
AND APES MONKEYS bly reßect orgasmic experiences. ( Labo-
ratory experiments on stump-tailed ma-
EVOLUTIONARY TREE of primates, based on DNA analysis, shows that humans di- caques have demonstrated that women
verged from bonobos and chimpanzees a mere eight million years ago. The three are not the only female primates capa-
species share more than 98 percent of their genetic makeup. ble of physiological orgasm. )
Male bonobos, too, may engage in
pseudocopulation but generally per-
hand (or, sometimes, a foot ) to a pos- time. As soon as a caretaker approached form a variation. Standing back to back,
sessor of food and will pout their lips the enclosure with food, the males one male brießy rubs his scrotum
and make whimpering sounds if the ef- would develop erections. Even before against the buttocks of another. They
fort is unsuccessful. But bonobos make the food was thrown into the area, the also practice so-called penis-fencing, in
diÝerent sounds than chimpanzees do. bonobos would be inviting each other which two males hang face to face from
The renowned low-pitched, extended for sex: males would invite females, a branch while rubbing their erect penis-
Òhuuu-huuuÓ pant-hooting of the latter and females would invite males and es together.
contrasts with the rather sharp, high- other females. The diversity of erotic contacts in bo-
pitched barking sounds of the bonobo. Sex, it turned out, is the key to the nobos includes sporadic oral sex, mas-
social life of the bonobo. The Þrst sug- sage of another individualÕs genitals and
Love, Not War gestion that the sexual behavior of bono- intense tongue-kissing. Lest this leave
bos is diÝerent had come from obser- the impression of a pathologically over-
ORANGUTAN
CHIMPANZEE
bonded to other females or to any gamy, polygamy and polyandry Each female has her own, sepa-
one male. are all in evidence. rate home range.
Yet another motivation is probably the food. A jealous male might chase an- Then Leslie rubbed her vulva against his
real cause: competition. There are two other away from a female, after which shoulder. This gesture calmed Kako, and
reasons to believe sexual activity is the the two males reunite and engage in he moved along the branch. It seemed
bonoboÕs answer to avoiding conßict. scrotal rubbing. Or after a female hits a that Leslie had been very close to using
First, anything, not just food, that juvenile, the latterÕs mother may lunge force but instead had reassured both
arouses the interest of more than one at the aggressor, an action that is im- herself and Kako with sexual contact.
bonobo at a time tends to result in sex- mediately followed by genital rubbing During reconciliations, bonobos use
ual contact. If two bonobos approach a between the two adults. the same sexual repertoire as they do
cardboard box thrown into their enclo- I once observed a young male, Kako, during feeding time. Based on an analy-
sure, they will brießy mount each other inadvertently blocking an older, female sis of many such incidents, my study
before playing with the box. Such situa- juvenile, Leslie, from moving along a yielded the Þrst solid evidence for sex-
tions lead to squabbles in most other branch. First, Leslie pushed him; Kako, ual behavior as a mechanism to over-
species. But bonobos are quite tolerant, who was not very conÞdent in trees, come aggression. Not that this function
perhaps because they use sex to divert tightened his grip, grinning nervously. is absent in other animalsÑor in hu-
attention and to diÝuse tension. Next Leslie gnawed on one of his hands, mans, for that matterÑbut the art of
Second, bonobo sex often occurs in presumably to loosen his grasp. Kako sexual reconciliation may well have
aggressive contexts totally unrelated to uttered a sharp peep and stayed put. reached its evolutionary peak in the
a b c
CHIMPANZEE
SUZANNE BARNES
d e f
DOMINANCE BY BONDING is evinced by female bonobos, dominated chimpanzee society the male eats Þrst (d ), while
who engage in genito-genital ( GG ) rubbing before eating sug- the females wait at a safe distance. After he leaves (e), carrying
arcane (a), while a bigger male displays to no avail. The fe- as many bananas as he can, the dominant female gets what is
males then share the food without competition (b). Only when left (f ). ( Small amounts of sugarcane and bananas are provid-
they leave can the male get to the sugarcane (c). In male- ed at some research sites in Zaire.)
Bonobo society is, however, not only In the bonobo group, it was the fe- ologist Takeshi FuruichiÕs summary of
female-centered but also appears to be males that approached the honey Þrst. the relation between the sexes at Wam-
female-dominated. Bonobo specialists, After having engaged in some GG rub- ba, where bonobos are enticed out of
while long suspecting such a reality, bing, they would feed together, taking the forest with sugarcane. ÒMales usual-
have been reluctant to make the contro- turns with virtually no competition be- ly appeared at the feeding site Þrst, but
versial claim. But in 1992, at the 14th tween them. The male might make as they surrendered preferred positions
Congress of the International Primato- many charging displays as he wanted; when the females appeared. It seemed
logical Society in Strasbourg, investiga- the females were not intimidated and that males appeared Þrst not because
tors of both captive and wild bonobos ignored the commotion. they were dominant, but because they
presented data that left little doubt Observers at the Belgian animal park had to feed before the arrival of fe-
about the issue. of Planckendael, which currently has males,Ó Furuichi reported at Strasbourg.
Amy R. Parish of the University of the most naturalistic bonobo colony,
California at Davis reported on food reported similar Þndings. If a male Sex for Food
competition in identical groups (one bonobo tried to harass a female, all fe-
adult male and two adult females ) of
chimpanzees and bonobos at the
Stuttgart Zoo. Honey was provided in a
males would band together to chase
him oÝ. Because females appeared more
successful in dominating males when
O ccasionally, the role of sex in rela-
tion to food is taken one step fur-
ther, bringing bonobos very close to
Òtermite hillÓ from which it could be they were together than on their own, humans in their behavior. It has been
extracted by dipping sticks into a small their close association and frequent speculated by anthropologistsÑinclud-
hole. As soon as honey was made avail- genital rubbing may represent an al- ing C. Owen Lovejoy of Kent State Uni-
able, the male chimpanzee would make liance. Females may bond so as to out- versity and Helen Fisher of Rutgers Uni-
a charging display through the enclo- compete members of the individually versityÑthat sex is partially separated
sure and claim everything for himself. stronger sex. from reproduction in our species be-
Only when his appetite was satisÞed The fact that they manage to do so cause it serves to cement mutually prof-
would he let the females Þsh for honey. not only in captivity is evident from zo- itable relationships between men and
women. The human femaleÕs capacity Despite such quid pro quo between cations for models of human society.
to mate throughout her cycle and her the sexes, there are no indications that Just imagine that we had never heard
strong sex drive allow her to exchange bonobos form humanlike nuclear fami- of chimpanzees or baboons and had
sex for male commitment and paternal lies. The burden of raising oÝspring ap- known bonobos Þrst. We would at pres-
care, thus giving rise to the nuclear pears to rest entirely on the femaleÕs ent most likely believe that early homin-
family. shoulders. In fact, nuclear families are ids lived in female-centered societies, in
This arrangement is thought to be fa- probably incompatible with the diverse which sex served important social func-
vored by natural selection because it al- use of sex found in bonobos. If our an- tions and in which warfare was rare or
lows women to raise more oÝspring cestors started out with a sex life simi- absent. In the end, perhaps the most
than they could if they were on their lar to that of bonobos, the evolution of successful reconstruction of our past
own. Although bonobos clearly do not the family would have required dramatic will be based not on chimpanzees or
establish the exclusive heterosexual change. even on bonobos but on a three-way
bonds characteristic of our species, Human family life implies paternal comparison of chimpanzees, bonobos
their behavior does Þt important ele- investment, which is unlikely to devel- and humans.
ments of this model. A female bonobo op unless males can be reasonably cer-
shows extended receptivity and uses tain that they are caring for their own,
sex to obtain a maleÕs favors whenÑ not someone elseÕs, oÝspring. Bonobo FURTHER READING
usually because of youthÑshe is too society lacks any such guarantee, but
THE PYGMY CHIMPANZEE: EVOLUTIONARY
low in social status to dominate him. humans protect the integrity of their BIOLOGY AND BEHAVIOR. Edited by Ran-
At the San Diego Zoo, I observed that family units through all kinds of moral dall L. Susman. Plenum Press, 1984.
if Loretta was in a sexually attractive restrictions and taboos. Thus, although THE COMMUNICATIVE REPERTOIRE OF
state, she would not hesitate to ap- our species is characterized by an ex- CAPTIVE BONOBOS (PAN PANISCUS )
proach the adult male, Vernon, if he traordinary interest in sex, there are no COMPARED TO THAT OF CHIMPANZEES.
had food. Presenting herself to Vernon, societies in which people engage in it F.B.M. de Waal in Behaviour, Vol. 106,
Nos. 3Ð4, pages 183Ð251; September
she would mate with him and make at the drop of a hat (or a cardboard box,
1988.
high-pitched food calls while taking over as the case may be). A sense of shame PEACEMAKING AMONG PRIMATES. F.B.M.
his entire bundle of branches and leaves. and a desire for domestic privacy are de Waal. Harvard University Press, 1989.
When Loretta had no genital swelling, typical human concepts related to the UNDERSTANDING CHIMPANZEES. Edited
she would wait until Vernon was ready evolution and cultural bolstering of the by Paul Heltne and Linda A. Marquardt.
to share. Primatologist Suehisa Kuroda family. Harvard University Press, 1989.
reports similar exchanges at Wamba: Yet no degree of moralizing can make THE LAST APE: PYGMY CHIMPANZEE BE-
ÒA young female approached a male, sex disappear from every realm of hu- HAVIOR AND ECOLOGY. Takayoshi Kano.
Stanford University Press, 1992.
who was eating sugarcane. They copu- man life that does not relate to the nu-
CHIMPANZEE CULTURES. R. Wrangham,
lated in short order, whereupon she clear family. The bonoboÕs behavioral W. C. McGrew, F.B.M. de Waal and P.
took one of the two canes held by him peculiarities may help us understand the Heltne. Harvard University Press, 1994.
and left.Ó role of sex and may have serious impli-
by Robert R. Birge
T
he worldÕs most advanced super- of switches, usually of the kind known ly generate at least two discrete states
computer does not require a sin- as logic gates, that ßip between two in a molecule, we can use each state to
gle semiconductor chip. The hu- statesÑdesignated as 0 and 1Ñin re- represent either 0 or 1. Such switches
man brain consists of organic mole- sponse to changes in the electric cur- oÝer reductions in the size of hardware
cules that combine to form a highly rent passing through them. ( Comput- because they are themselves smallÑ
sophisticated network able to calculate, ers typically represent all information about one thousandth the size of the
perceive, manipulate, self-repair, think in terms of such binary digits, or bits. ) semiconductor transistors used today
and feel. Digital computers can certain- If the trend toward miniaturization con- as gates (which measure about one mi-
ly perform calculations much faster tinues, the size of a single logic gate cron, or a millionth of a meter, across).
and more precisely than humans can, will approach the size of molecules by Indeed, a biomolecular computer could
but even simple organisms are superior about the year 2030. in principle be one Þftieth the size of a
to computers in the other Þve domains. But there is a serious roadblock. Each present-day semiconductor computer
Computer designers may never be able factor of two in miniaturization increas- composed of a similar number of logic
to make machines having all the facul- es the cost of manufacturing a chip by a elements. In the computer business,
ties of a natural brain, but many of us factor of Þve. At some point the search smaller gate size generally makes for a
think we can exploit some special prop- for ever smaller electronic devices may faster device, and protein-based com-
erties of biological moleculesÑparticu- be limited by economics rather than puters could theoretically operate 1,000
larly proteinsÑto build computer com- physics [ see ÒThe Wall,Ó by Gary Stix, times faster than modern computers.
ponents that are smaller, faster and ÒScience and Business,Ó SCIENTIFIC At this stage no one is seriously pro-
more powerful than any electronic de- AMERICAN, July 1994 ] . On the other posing a purely biomolecular computer.
vices on the drawing boards thus far. hand, the use of biological molecules Far more likely, at least for the near fu-
The size issue is especially pressing. as the active components in computer ture, is the use of hybrid technology in
Since the 1960s the computer industry circuitry may offer an alternative ap- which molecules and semiconductors
has been compelled to make the indi- proach that is more economical. are used in combination. Such an ap-
vidual components on semiconductor Molecules can potentially serve as proach should provide computers that
chips smaller and smaller in order to computer switches because their atoms are one Þftieth the size and as much as
manufacture larger memories and more are mobile and change position in a 100 times faster than current ones.
powerful processors economically. predictable way. If we can direct that Biological molecules also appeal be-
These chips essentially consist of arrays atomic motion and thereby consistent- cause they can be designed one atom
BACTERIORHODOPSIN
BARRIE ROKEACH; RICK GROSS AND DESHAN GOVENDER ( inset )
TOMO NARASHIMA
CHROMOPHORE
CARBON
HYDROGEN
NITROGEN
OXYGEN
TOMO NARASHIMA
SALTY WATERS in Owens Lake, Calif., shown in an aerial view (left ), have a purple
hue caused by the presence of bacteria (inset ) containing a colorful protein called
bacteriorhodopsin. This protein, depicted here as a ribbon (center), includes a seg-
ment known as a chromophore (shown as balls and sticks) that absorbs light. After
this chromophore is excited by light, its structure changes (right ) and thereby al-
ters the conformation of the rest of the protein. Because bacteriorhodopsin adopts
diÝerent, readily detectable states in response to light, it can serve as logic gates,
or switches, in protein-based optical computers.
TOMO NARASHIMA
M quence of structural changes induced by
lightÑallows for the storage of data in memo-
O ry. Green light transforms the initial resting
state, known as bR, to the intermediate K. Next
K relaxes, forming M and then O. If the O inter-
mediate is exposed to red light, a so-called
branching reaction occurs. Structure O con-
verts to the P state, which quickly relaxes to
the Q stateÑa form that remains stable almost
K indeÞnitely. Blue light, however, will convert Q
P back to bR. Any two long-lasting states can be
assigned the binary value 0 or 1, making it pos-
BLUE sible to store information as a series of bacteri-
LIGHT
SOURCE orhodopsin molecules in one state or the other.
Q
nent known as a chromophore. The
chromophore absorbs energy from
light, triggering a complex series of in-
ternal motions that result in dramatic
changes in the structure of the larger
protein. These changes alter the pro-
teinÕs optical and electrical characteris-
GREEN RELAXATION
bR tics. For example, when rhodopsin ab-
LIGHT SOURCE
LIGHT- sorbs light in the human eye, the change
INDUCED in structure releases energy that serves
TRANSMISSION
as an electrical signal able to convey vi-
sual information to the brain.
for such research because he had the optical data processors using protein
ear of Soviet military leaders and was technology. The details of their most Computer Applications
able to convince them that by exploring impressive accomplishment, a proces-
bioelectronics, Soviet science could leap-
frog the West in computer technology.
Many aspects of this ambitious proj-
sor for military radar, remain obscure.
I became interested in bacteriorho-
dopsin in the 1970s, while I was study-
A t Þrst I was concerned purely with
understanding how such light-acti-
vated changes to rhodopsin occurred.
ect are still considered military secrets ing the biochemical basis of vision at During the late 1970s, however, I be-
and may never be revealed. We do know the University of California at Riverside. came interested in bacteriorhodopsin
that the Soviet military made microÞche My work had initially focused on a re- as well. I had also decided to apply my
Þlms, called Biochrome, out of bacteri- lated protein, rhodopsin, present in the knowledge of its properties to the de-
orhodopsin. Informal reports from for- retina of mammals. Both rhodopsin and sign of computer memories and pro-
mer Soviet scientists now in the U.S. in- bacteriorhodopsin are complex proteins cessors based on the protein. Albert F.
dicate that researchers there also made that include a light-absorbing compo- Lawrence, then at Hughes Aircraft Com-
pany, played an important role in con-
WRITING DATA vincing me that bioelectronics had po-
tential. He joined my lab for one year
1
LASER 2 3 P STATE
ARRAYS DETECTOR
WRITING OF INFORMATION into cubes of bacteriorhodopsin the proteinÕs photocycle. Then, red lasers are Þred (2) at the
( purple), and reading out of that information, is accomplished particular set of molecules in the plane ( green) to be convert-
with laser beams. The writing process is begun by Þring green ed to the binary 1 state; the remaining molecules represent
laser beams through a plane of the cube (1); this step begins binary 0. The targeted molecules Þrst form the P state (3 ),
4 6
MICHAEL GOODMAN
then relax to the Q structure (4 ). Reading from the protein- light, and molecules in the P or Q state allow the low light
based memory is begun by again activating the plane with levels to pass through. Hence, the resulting pattern of dark
green light (5 ). Then, red lasers of low intensity are Þred. and lightÑthat is, 0Õs and 1ÕsÑcan be picked up by a detector
Molecules that were originally in the bR state absorb the red placed directly opposite from the red laser array (6 ).
MICHAEL GOODMAN
COMPUTERS OF THE FUTURE might be hybrids, consisting of cards shown here, which have not yet been built, could pro-
cards with both proteins ( purple) and semiconductors. The vide associative memory (a) and three-dimensional memo-
green paging beam Þres at the square be possible. Indeed, I anticipate that the tems based on bacteriorhodopsin are
of protein to be read, starting the nor- major near-term impact of bioelectron- being investigated. For example, biolog-
mal photocycle of the molecules in the ics on computer hardware will be in the ical molecules seem to hold promise as
bR state. After two milliseconds, the en- area of volumetric memory. components of the associative memo-
tire laser array is turned on at a very low Speed is also an important beneÞt of ries needed for neural networks and,
intensity of red light. The molecules volumetric memories. The combination eventually, for artiÞcial intelligence.
that are in the binary 1 state (P or Q in- of three-dimensional storage with the
termediates) do not absorb these red use of parallel architectures enhances Neural Networks
beams or change their state. the speed of such memories, just as
But the molecules that started out in
the original binary 0 state (bR ) do ab-
sorb the beams (but do not change their
parallel processing in the human brain
overcomes relatively slow neural pro-
cesses and allows the brain to be a
A ssociative memories operate rather
diÝerently from the memories that
dominate current computer architec-
structure), because they have cycled to thinking machine with fast reßexes and tures. This type of architecture takes a
the red-absorbing O intermediate. A de- rapid decision-making capability. The set of data, often in the form of an im-
tector images the light passing through entire writing process described above age, and scans the entire memory bank
the cube of memory and records the lo- takes place in about 10 milliseconds. If until it Þnds a data set that matches it.
cation of O and of P or Q structuresÑ we illuminate a square measuring 1,024 In some cases, the computer will Þnd
or in terms of binary code, the detector bits by 1,024 bits within a larger cube the closest match if it cannot Þnd a per-
reads 0Õs and 1Õs. The process is com- of protein, we can write 1,048,576 bits fect match, in a sense taking an educat-
plete in approximately 10 milliseconds, of data, or about 105 kilobytes, into ed guess at an answer. Because the hu-
a rate of 10 megabytes per second for memory in a 10-millisecond cycle. These man brain operates in a neural, asso-
each page of memory. values represent an overall write speed ciative mode, many computer scientists
of 10 million characters per second, believe large-capacity associative mem-
Three-Dimensional Memories comparable to slow semiconductor ories will be required if we are to achieve
memory. Yet each memory device can artiÞcial intelligence.
MICHAEL GOODMAN
ryÑ32 gigabytes of permanent memory (b) and eight giga- conductor central processing unit (d ), these cards form a com-
bytes of removable memory (c). When combined with a semi- plete computer system with enhanced capabilities.
from many more times than can crys- tral processing unit of this computer less, we are conÞdent that hybrid com-
tals, which suÝer from fatigue after re- will consist of traditional semiconduc- puters of some type will be available
peated read-write cycles. tor technology. Two cards will contain within the next eight years.
As studies of natural bacteriorho- protein-based volumetric memory with We further expect that during the
dopsin continue, many laboratories are a total capacity of roughly 40 gigabytes. next two decades, they will evolve into
also exploring the value of modiÞed One of these cards will be a fast, perma- the dominant architectures for certain
forms of the protein in computer de- nent, random-access memory using no types of computing, such as for scien-
vices. SpeciÞcally, they are studying ge- moving parts; the other will oÝer less tiÞc calculations and multimedia appli-
netically engineered versions of the expensive, removable, long-term data cations. Personal computer users will
protein, in which one amino acid re- storage. The fourth card will contain an beneÞt by having large and inexpensive
places another in order to enhance the associative memory based on Þlms of memory boards that have many giga-
properties needed for particular appli- bacteriorhodopsin. bytes of data storage and removable
cations. For example, the lifetime of the memory components that contain a few
M state in the photocycle can be length- The Future of Computers gigabytes of data storage in a small
ened by removal of an internal amino cube. Imagine the advantage of carry-
acid from the protein, as shown by Nor-
bert Hampp and Christoph BrŠuchle of
the University of Munich, in collabora-
T he hybrid computer we envision
would be highly ßexible. By taking
advantage of particular combinations
ing in your pocket a small cube storing
the equivalent of a comprehensive en-
cyclopedia and all the words you have
tion with Oesterhelt. of the memory cards described above, written in the past 10 years.
Of course, biomolecular computers the computer should be able to handle But the most dramatic application
represent the ultimate goal. As I men- large pools of data, carry out complex may well be found in yet another realm.
tioned earlier, however, most scientists scientiÞc simulations or serve as a With terabytes of data storage, neural
believe the Þrst step in the development unique platform for investigations of associative capabilities and a high ca-
of protein-based computers will be the artiÞcial intelligence. With close to a ter- pacity for parallel processing, hybrid
generation of hybrid systems that com- abyte (10 12 bytes) of memory in cubes computers will, for the Þrst time, incor-
bine the best features of semiconduc- of bacteriorhodopsin, this machine porate the three crucial hardware re-
tor and molecular architectures. In par- would handle large databases with alac- quirements for artiÞcial intelligence. We
ticular, hybrid technology, composed rity. Associative memory processing are indeed at the threshold of an excit-
in part of high-density, protein-based coupled with volumetric memory would ing new era in computing.
memory, may help solve the lingering make database searches many orders
problem of memory capacity. of magnitude faster than is currently
During the past decade, the speed of possible. Because this hybrid computer FURTHER READING
computer processors increased almost can be designed to function as a neural THREE-DIMENSIONAL OPTICAL STORAGE
1,000 times, whereas external data stor- associative computer capable of learn- MEMORY. Dimitri A. Parthenopoulos
age capacities increased by only a fac- ing and of analyzing data and images and Peter M. Rentzepis in Science, Vol.
tor of 50. Also, the transfer of data with- in much the same way as the human 245, pages 843Ð845; August 25, 1989.
in the computer remains the principal brain, the likely importance of hybrid BACTERIORHODOPSIN: A BIOLOGICAL MA-
bottleneck that limits performance. Par- computers to studies in artiÞcial intelli- TERIAL FOR INFORMATION PROCESSING.
Dieter Oesterhelt, Christoph BrŠuchle
allel processing and light-based inter- gence cannot be underestimated. and Norbert Hampp in Quarterly Re-
connections, both made faster with hy- Although my group and others have views of Biophysics, Vol. 24, No. 4, pag-
brid computers that exploit the eÛcient had remarkable success developing vol- es 425Ð478; November 1991.
switching of biological molecules, allow umetric memories and associative pro- PROTEIN-BASED OPTICAL COMPUTING AND
for the storage, transfer and manipula- cessors, more work is needed before a MEMORIES. Robert R. Birge in Computer,
tion of massive amounts of data. fully operational hybrid computer can Vol. 25, No. 11, pages 56Ð67; Novem-
To explore the possible value of hy- be built. Along the way toward develop- ber 1992.
PROTEIN-BASED THREE-DIMENSIONAL
brid computers, my laboratory is cur- ing a powerful yet reasonably priced
MEMORY. Robert R. Birge in American
rently designing one that contains four design, other competing architectures Scientist, Vol. 82, No. 4, pages 348Ð355;
types of memory units or processors, may replace many of the hardware com- JulyÐAugust 1994.
known as cards. The card with the cen- ponents we have described. Neverthe-
by Curtis N. Runnels
T
he stark Greek landscape charms anced and reverential regard they sup- took place in antiquity, but the evidence
everyone who sees it. For thou- pose our ancestors had for the natural allows one also to place the responsibil-
sands of years, Greeks and visi- world. The Garden of Eden is a primal ity for this destruction primarily on the
tors alike have sung the praises of this myth of Western civilization, and it was inhabitants of that time.
small country, famous for its Bronze preceded in classical antiquity by the
Age civilizations and the cultural belief in the Golden AgeÑa time, alas A New Kind of Archaeology
achievements of its people in classical now lost, when human beings were said
times. But is the countryside one sees
today the result of climate acting alone,
or have humans played a part by clear-
to have lived in innocent harmony with
their natural environment.
Sir Peter B. Medawar ( in The Limits of
O ver the past two decades, the in-
troduction of multidisciplinary and
intensive regional survey techniques in
ing forests and causing soil erosion? Science, a book of his essays published Greece has revolutionized Greek archae-
Recent archaeological work is chang- in 1984) described this kind of thinking ology. One aspect of this revolution is a
ing a long-standing view of the impact as ÒArcadian.Ó He compared the con- shift away from the investigation of sin-
of agriculture on the land in Greece. cepts of Utopia and Arcadia and con- gle sites to studies of the natural and
The evidence mounts for episodes of de- cluded that Arcadia is closer to the ide- cultural history of entire regions. This
forestation and catastrophic soil ero- al of a Golden Age than are the Utopias approach is achieved in part by using
sion over the past 8,000 years. Many of Thomas More and Francis Bacon. new methods of Þeldwork, particularly
scholars believe they resulted from a Their Utopias were places where sci- searching large areas ( typically more
long history of human land use and ence-based technology was employed than 100 square kilometers) with teams
abuse. This new perspective on human for the melioration of society, whereas of people who record every artifact, fea-
settlement and its impact on the natu- Arcadia Òis the conception farthest re- ture and site that can be detected. An-
ral environment stands in sharp con- moved from Utopia, for one of its prin- other innovation is the large-scale use of
trast to the views of the 19th-century cipal virtues is to be pastoral, prescien- remote sensing, ranging from ground-
Romantics, who saw the ancient Greeks tiÞc and pretechnological. In Arcadia, penetrating radar to satellite images, to
as careful stewards of a land they held mankind lives in happiness, ignorance assist the archaeologist in detecting
to be Þlled with gods. and innocence, free from the diseases past human activity. Further aid has
Indeed, it remains a widely espoused and psychic disquiet that civilization come from the new discipline of geoar-
opinion today that the destructive eco- brings with itÑliving indeed in that chaeology, which combines the tech-
logical practices of modern civilization state of inner spiritual tranquillity which niques of geology and archaeology to
are a new development. The popular comes today only from having a sub- improve the interpretation of the natu-
press frequently carries reports of peo- stantial private income derived from ral contexts of ancient cultures.
ple who advocate returning to the bal- trustee securities.Ó An example of the collaboration be-
The eÝects of Arcadian thinking can tween archaeologists and geoarchaeol-
be seen in the debate about the degree ogists can be seen in the study of land-
to which the ancient Greeks were re- scape change in southern Greece. Two
CURTIS N. RUNNELS earned degrees
in archaeology from the University of sponsible for the deforestation and ero- major projects in which I have partici-
Kansas and Indiana University and has sion that have reduced much of Greece pated since 1979 in the Argolid (the
taught at Stanford University. He is asso- to a barren, stonyÑif picturesqueÑ northeastern part of the Peloponnese
ciate professor of archaeology at Boston wasteland. To be sure, environmental peninsula ) combined walking tours de-
University. Runnels has worked for 20 degradation ( if not its causes ) was not- signed to identify archaeological sites
years in Greece, specializing in the study ed in ancient times; many references to with geologic surveys that attempted to
of early prehistory and the relation be- it appear in the writings of the ancients, reconstruct the history of the landscape
tween human settlement and landscape
through time. His most recent book is A
particularly Plato and Aristotle. They during the Pleistocene and Holocene
Greek Countryside: The Southern Argolid give accurate and apparently eyewit- epochs of the past 50,000 years.
from Prehistory to the Present Day, writ- ness accounts of deforestation and soil One of these projects targeted the
ten with Michael H. Jameson and Tjeerd erosion in the fourth century B.C. Ar- Argive plain. According to Aristotle, it
H. van Andel. chaeology can now conÞrm that de- had undergone considerable alteration
spoilment of the natural environment in the Bronze Age, from about 3000 to
GOLDEN AGE of Greece is represented in Claude LorrainÕs desses. The persistent view of the ancient Greeks as careful
painting The Judgment of Paris (1646), in which the goddess stewards of a land that they saw as guided by gods and god-
of strife throws down a golden apple inscribed Òto the fair- desses is belied by recent archaeological evidence of soil ero-
estÓ and says that Paris must award it to one of three god- sion caused by human land abuse.
1000 B.C. To investigate ancient land on the edges of the Berbati Valley near less, they demonstrate the more inten-
use, I joined archaeologist Berit Wells, the best soils and supplies of water. Af- sive use of the landscape at this time be-
director of the Swedish Institute in Ath- ter this initial colonization, the size of cause they are located on steep slopes,
ens, and Eberhard A. W. Zangger, a geo- the original Neolithic settlements began at high elevations and in areas where
archaeologist now at the University of to grow, and new ones were founded. only soils of marginal productivity ever
Heidelberg. Our goal was to survey two By the Late Neolithic period (between existed.
valleys, Berbati and Limnes, which lie on 4000 and 3000 B.C.), more than 20 set- In addition to sites where past hu-
the northern edge of the Argive plain tlements and smaller sites were scat- man activity left concentrations of arti-
and have a combined area of 60 square tered over the region. This expansion facts, our searching technique revealed
kilometers. continued for the Þrst few centuries of tens of thousands of individual Neo-
This project, sponsored by the Swed- the third millenniumÑthe Early Bronze lithic items scattered over the country-
ish Institute and headed by Wells, had Age in cultural terms. side. These thin scatters of artifacts are
two parts. One was a geoarchaeological found in areas now uninhabited and in
study of the two valleys by Zangger, Ancient Land Use some cases uninhabitable because they
who had done a similar study of the are devoid of soil and vegetation.
Argive plain while working on his doc-
toral dissertation at Stanford Universi-
ty. The other was an intensive archaeo-
F rom the larger settlements came ev-
idence of agriculture and many arti-
facts such as pottery, querns for grind-
The pattern of settlements and arti-
facts reveals the structure of what ar-
chaeologists call fossil cultural land-
logical survey. ing grain and simple stone tools of ßint scapes. Particularly interesting are the
The survey brought to light evidence and obsidian. Some of the smaller sites numerous Þnds of ground and polished
of human activity from the Middle Pa- doubtless served special purposes (as hard-stone axes and ßint blades, which
leolithic period, some 50,000 years ago, huts for shepherds and storehouses for have edges coated with silica deposited
and showed that agriculturists entered tools, for example) and were not neces- when the blades were used to cut grass-
the area about 7,000 years ago, settling sarily inhabited permanently. Neverthe- es and other plants. These common agri-
17TH TO 20TH
CENTURY
MIDDLE
BYZANTINE
300–50 B.C.
b
2500–2000 B.C.
45,000–
32,000 B.P.
c 60,000 B.P.
JANA BRENNING
250,000
YEARS
BEFORE
PRESENT
LAND ABUSE by ancient Greeks followed a consistent pattern result, a typical geologic proÞle (right ) from the southern Ar-
(left ). At first (a ), a mixed forest held the soil in place on a golid region of Greece shows a sequence of erosional depos-
slope. When the forest was cleared for farming (b ), the soil its and intervening soils (with their approximate ages). In each
stayed in place for a time, but eventually erosion carried it period the soil that gradually forms above the deposits is thin-
oÝ the slope, depositing alluvium in the valley bottom (c). As a ner and less developed and is less able to support vegetation.
I
magine you are the father of an might revolutionize our criminal justice
eight-year-old boy,Ó says psycholo- system,Ó asserts Roger D. Masters, a po-
gist Adrian Raine, explaining where litical scientist at Dartmouth College.
he believes his 17 years of research on ÒWith the expected advances, weÕre
the biological basis of crime is leading. going to be able to diagnose many peo-
ÒThe ethical dilemma is this: I could ple who are biologically brain-prone to
say to you, ÔWell, we have taken a wide violence,Ó claims Stuart C. Yudofsky,
variety of measurements, and we can chair of the psychiatry department at
predict with 80 percent accuracy that Baylor College of Medicine and editor
your son is going to become seriously of the Journal of Neuropsychiatry and
violent within 20 years. We can oÝer Clinical Neurosciences. ÒIÕm not worried
you a series of biological, social and about the downside as much as I am en-
cognitive intervention programs that couraged by the opportunity to prevent
will greatly reduce the chance of his be- tragediesÑto screen people who might
coming a violent oÝender.Õ have high risk and to prevent them
ÒWhat do you do? Do you place your from harming someone else.Ó Raine, Yu-
boy in those programs and risk stigma- dofsky and others argue that in order
tizing him as a violent criminal even to control violence, Americans should
though there is a real possibility that trade their traditional concept of justice
he is innocent? Or do you say no to the based on guilt and punishment for a
treatment and run an 80 percent chance Òmedical modelÓ based on prevention,
that your child will grow up to (a) de- diagnosis and treatment.
stroy his life, (b) destroy your life, (c) But many scientists and observers do
destroy the lives of his brothers and worry about a downside. They are con-
sisters and, most important, (d) de- cerned that some researchers underplay
stroy the lives of the innocent victims the enormous complexity of individual
who suÝer at his hands?Ó behavior and overstate scientistsÕ abili-
For now, such a HobsonÕs choice is ty to understand and predict it. They
purely hypothetical. Scientists cannot also fear that a society desperate to re-
yet predict which children will become duce crime might Þnd the temptation to
dangerously aggressive with anything make premature or inappropriate use
like 80 percent accuracy. But increas- of such knowledge irresistible.
ingly, those who study the causes of Indeed, the history of scienceÕs as-
criminal and violent behavior are look- sault on crime is blemished by instanc-
ing beyond broad demographic charac- es in which incorrect conclusions were
teristics such as age, race and income used to justify cruel and unusual pun-
level to factors in individualsÕ personal- ishments. In the early 1930s, when the
ity, history, environment and physiolo- homicide rate was even higher than it is
gy that seem to put themÑand soci- today, eugenics was in fashion. ÒThe eu-
BRUCE DAVIDSON Magnum Photos
etyÑat risk. As sociologists reap the genics movement was based on the idea
beneÞts of rigorous long-term studies that certain mental illness and criminal
and neuroscientists tug at the tangled traits were all inherited,Ó says Ronald L.
web of relations between behavior and Akers, director of the Center for Studies
brain chemistry, many are optimistic in Criminology and Law at the Universi-
that science will identify markers of ty of Florida. ÒIt was based on bad sci-
maleÞcence. ÒThis research might not ence, but they thought it was good sci-
pay oÝ for 10 years, but in 10 years it ence at the time.Ó By 1931, 27 states
had passed laws allowing compulsory
sterilization of Òthe feeble-minded,Ó
the insane and the habitually criminal.
TEXAN TEENS playing with gang signs
Studies in the late 1960sÑwhen crime
and loaded guns are acting their ageÑ
most adolescents dabble in delinquency
was again high and risingÑrevealed
for several years. But a small fraction that many violent criminals had an ex-
grow into the chronic felons that com- tra Y chromosome and thus an extra set
mit the majority of violent crimes. Can of ÒmaleÓ genes. ÒIt was a dark day for
scientists identify the dangerous few be- science in Boston when they started
fore they attackÑand if so, what then? screening babies for it,Ó recalls Xandra
Copyright 1995 Scientific American, Inc. SCIENTIFIC AMERICAN March 1995 101
140 were 14.3 times more likely to commit
130 violent oÝenses. ÒThis is a good way of
120 predicting,Ó she says.
110
ALL PERSONAL CRIMES Good is a relative term. For if one
100 were to predict that every boy in her
90 study who was arrested early would go
80 on to commit violent crimes, one would
70 be wrong more than 65 percent of the
time. To statisticians, those so misiden-
60 BURGLARY
tiÞed are known as false positives. ÒAll
RATE PER 1,000 PEOPLE
50
of these predictors have a lot of false
40
positivesÑabout 50 percent on aver-
30 ASSAULT
age,Ó says Akers, who recently complet-
20 ed a survey of delinquency prediction
10 ROBBERY models. Their total accuracy is even
0 lower, because the models also fail to
INCARCERATION
identify some future criminals.
4 The risk factors that Akers says re-
3 searchers have found to be most close-
2
RAPE ly associated with delinquency are hard-
1 ly surprising. Drug use tops the list, fol-
0 lowed by family dysfunction, childhood
behavior problems, deviant peers, poor
0.11 school performance, inconsistent par-
MURDER
0.10 ental supervision and discipline, sepa-
0.09 ration from parents, and poverty. Nu-
JOHNNY JOHNSON
0.08 merous other controlled studies have
0.07 found that alcoholism, childhood abuse,
1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 low verbal IQ and witnessing violent
YEAR acts are also significant risk factors.
SOURCE: Bureau of Justice Statistics, U.S. Department of Justice Compared with violent behavior, how-
CRIME RATES have not responded consistently to Òget toughÓ approaches to incar- ever, all these experiences are exceed-
ceration. Since the early 1970s the proportion of Americans behind bars has more ingly common. The disparity makes it
than tripled. Property crime (including burglary, robbery and personal larceny) has very diÛcult to determine which factors
dropped about 30 percent, but violent crime remains high. are causes and which merely correlates.
Preventive Intervention
O. BreakeÞeld, a geneticist at Massa- make a sizable impact on the violent
chusetts General Hospital. Subsequent
studies revealed that although XYY men
tend to score lower on IQ tests, they are
crime rate, which has remained persis-
tently high since 1973 despite a sub-
stantial decline in property crime. (Fe-
T he diÝerence is important, notes
Mark W. Lipsey of Vanderbilt Uni-
versity, because Òchanging a risk factor
not unusually aggressive. males accounted for only 12.5 percent if it is not causal may have no impact,Ó
of violent crime in 1992.) ÒFor every 1 and the ultimate goal of prediction is
False Positive ID percent that we reduce violence, we save to stop violence by intervening before
the country $1.2 billion,Ó Raine asserts. it begins. Unfortunately, improvements
102 SCIENTIFIC AMERICAN March 1995 Copyright 1995 Scientific American, Inc.
Patrick H. Tolan of the University of cated analyses suggest that
Illinois at Chicago has also recently pub- both of these assumptions
POOR PARENTAL SUPERVISION is a major risk factor for later crack cocaine vials. Parent training programs have been among
delinquency. These children in Philadelphia play with empty the most successful in reducing kidsÕ antisocial behavior.
Copyright 1995 Scientific American, Inc. SCIENTIFIC AMERICAN March 1995 103
The Tangled Roots of Violence
any problem in society using just one tance of that role is still very much in hormone important for building muscle
discipline,Ó says Diana Fishbein, a pro- doubt, however. mass and strength, among other func-
fessor of criminology at the University As with social risk factors, no biolog- tions. James M. Dabbs, Jr., of Georgia
of Baltimore. ÒSociological factors play ical abnormality has been shown to State University has found in his exper-
a role. But they have not been able to cause violent aggressionÑnor is that iments with prison inmates that men
explain why one person becomes vio- likely except in cases of extreme psy- with the highest testosterone concen-
lent and another doesnÕt.Ó chiatric disorder. But researchers have trations are more likely to have com-
Some social scientists are looking to spotted several unusual features, too mitted violent crimes. But Dabbs em-
psychiatrists, neurologists and geneti- subtle even to be considered medical phasizes that the link is indirect and
cists to provide answers to that ques- problems, that tend to appear in the Òmediated by numerous social factors,Ó
tion, ready or not. ÒScience must tell us bodies and brains of physically aggres- such as higher rates of divorce and
what individuals will or will not become sive men. On average, for example, they substance abuse.
criminals, what individuals will or will have higher levels of testosterone, a sex ÒLow resting heart rate probably rep-
not become victims, and what law en-
forcement strategies will or will not
work,Ó wrote C. Ray JeÝery, a criminol-
ogist at Florida State University, last
year in the Journal of Research in Crime
and Delinquency.
As medical researchers have teased
out a few tantalizing links between
EUGENE RICHARDS Magnum Photos
104 SCIENTIFIC AMERICAN March 1995 Copyright 1995 Scientific American, Inc.
in the blood or the brain. “We know so little about man- ferences and on television talk shows (left ), much to the
ganese’s role in the body that we haven’t even set an RDA scientist’s annoyance. “We have asked that all reports on the
[recommended daily allowance] for it.” study be embargoed until the final paper goes through the
Hodges remains convinced he is on the right track. “Vio- peer-review process,” Schoenthaler says, “but he [Hodges]
lence can be detected and treated,” he argues. In 1987 a continues to make an example of it.”
mugger fractured the skull of another of Hodges’s sons. With two studies in hand, Hodges has redirected his cru-
That year Hodges founded the Violence Research Founda- sade to Washington. “What we need is a leap of faith from
tion ( VRF) to lobby public officials to experiment with treat- the Justice Department,” he says. So he has begun search-
ment programs that use what he calls “the power of nutri- ing for converts. Hodges claims to have the support of for-
tion” to pacify violent criminals. mer attorney general Edwin Meese, with whom he has met
The VRF found an ally in Senator Robert Presley of Cali- several times, and Senator Tom Harkin of Iowa. “I have an
fornia, who pushed through a bill in 1989 authorizing a invitation from [Utah senator] Orrin G. Hatch to come up to
study of male prisoners by Stephen Schoenthaler of Califor- Washington as soon as he becomes chairman of the Judicia-
nia State University at Stanislaus. In the first part of the ry Committee,” Hodges stated in December. A member of
study, 402 offenders were divided randomly into three Hatch’s staff says that Hodges’s “material is under review, but
groups and given vitamin-mineral supplements equivalent no agreements have been made yet.”
to the RDA, three times the RDA or a placebo. Preliminary Trace element deficiencies are just one of many frequent-
results showed that rule violations among the first group ly cited but poorly demonstrated claims that nutritional
dropped 38 percent during the study. Strangely, the behav- problems can cause criminal and violent behavior. A 1992
ior of inmates getting the higher dose did not improve sig- report by the Federal Bureau of Prisons stated that correc-
nificantly, and violations rose 20 percent among the place- tional facilities in 46 states have incorporated a wide array
bo group. of dietary intervention and testing programs, even though
Although encouraging, the equivocal results are so incon- “such programs are perceived by many physicians, scientif-
clusive that Schoenthaler has decided not to publish them ic researchers, registered dietitians, and other health care
until he completes further studies with more controls. professionals as an incorporation of food faddism into pub-
Hodges, however, has publicized the results widely at con- lic policy.” —Steven Vames and W. Wayt Gibbs
resents the best replicated biological abnormal heart rates in psychopaths. derstanding and Preventing ViolenceÓ
correlate of antisocial behavior,Ó Raine Jerome Kagan, a Harvard University convened by the National Research
observes, pointing to 14 studies that psychologist, has suggested that an in- Council suggested in its 1993 report
have found that problem children and hibited ÒtemperamentÓ may explain that inhibited children may be protect-
petty criminals tend to have signiÞcant- why the great majority of children from ed by their fearfulness from becoming
ly lower pulses than do well-behaved high-risk homes grow up to become aggressive, whereas uninhibited chil-
counterparts. A slower heartbeat Òprob- law-abiding citizens. One study tested dren may be prone to later violence. The
ably reßects fearlessness and under- pulse, pupil dilation, vocal tension and panel concluded that Òalthough such
arousal,Ó Raine theorizes. ÒIf we lack blood levels of the neurotransmitter factors in isolation may not be expect-
the fear of getting hurt, it may lead to a norepinephrine and the stress-regulat- ed to be strong predictors of violence,
predisposition to engage in violence.Ó ing hormone cortisol to distinguish in- in conjunction with other early family
But that hypothesis fails to explain why hibited from uninhibited, underaroused and cognitive measures, the degree of
at least 15 studies have failed to Þnd two-year-olds. An expert panel on ÒUn- prediction may be considerable.Ó
Perhaps the most frequently cited bi-
ological correlate of violent behavior is
a low level of serotonin, a chemical that
in the body inhibits the secretion of
stomach acid and stimulates smooth
muscle and in the brain functions as a
neurotransmitter. A large body of ani-
mal evidence links low levels of sero-
tonin to impulsive aggression. Its role
in humans is often oversimpliÞed, how-
STEPHEN SHAMES Matrix
Copyright 1995 Scientific American, Inc. SCIENTIFIC AMERICAN March 1995 105
KENNETH JARECKE Contact Press Images
olence.Ó A front-page article in the Chi- than a dozen diÝerent neural receptors, sume a dominant status. The numerous
cago Tribune in December 1993 ex- each of which seems to perform a dis- pathways through which serotonin can
plained that Òwhen serotonin declines... tinct function. The low levels of 5-HIAA inßuence mood and behavior confound
impulsive aggression is unleashed.Ó repeatedly seen in violent oÝenders attempts to simply Òreduce the level of
Such explanations do violence to the may indicate a shortage of serotonin in violenceÓ by administering serotonin
science. In human experiments, re- the brain or simply a dearth of MAOÑ boosters such as Prozac, a widely pre-
searchers do not generally have access in which case their serotonin levels may scribed antidepressant.
to the serotonin inside their subjectÕs actually be high. Moreover, serotonin Nevertheless, the link between 5-HIAA
braincase. Instead they tap cerebrospi- can rise or drop in diÝerent regions of and impulsive aggression has led to a
nal ßuid from the spinal column and the brain at diÝerent times, with mar- concerted hunt for the genes that con-
measure the concentration of 5-hydrox- kedly diÝerent eÝects. trol the production and activity of sero-
yindoleacetic acid (5-HIAA), which is Environment, too, plays a role: non- tonin and several other neurotransmit-
produced when serotonin is used up human primate studies show that sero- ters. ÒRight now we have in our hand
and broken down by the enzyme tonin often ßuctuates with pecking or- many of the genes that aÝect brain
monoamine oxidase (MAO). Serotonin der, dropping in animals when they are function,Ó says David Goldman, chief of
does its job by binding to any of more threatened and rising when they as- neurogenetics at the National Institute
106 SCIENTIFIC AMERICAN March 1995 Copyright 1995 Scientific American, Inc.
maintains is ÒÞnally getting interested.Ó
But some worry that voluntary screen-
ing for the good of the child might lead
to mandatory screening for the protec-
tion of society. ÒIt is one thing to con-
vict someone of an oÝense and compel
them to do something. It is another
thing to go to someone who has not
done anything wrong and say, ÔYou
look like a high risk, so you have to do
this,Õ Ó Akers observes. ÒThere is a very
clear ethical diÝerence, but that is a
very thin line that people, especially
politicians, might cross over.Ó
Copyright 1995 Scientific American, Inc. SCIENTIFIC AMERICAN March 1995 107
MATHEMATICAL RECREATIONS by Ian Stewart
Turning the Tables Around bles inside some well-deÞned area [see
bottom illustration on page 109]. Hmm.
ThereÕs one, a bit more complicated,
U
p on the 67th ßoor of RuÝ Tow- bles,Ó Max said. where you have a rectangular region
ers, two employees from We- Dan stooped to peer under a table. containing only two rectangular tables
Haulit-4U Moving dragged the ÒYouÕre rightÑthereÕs plenty of room.Ó and two square ones.Ó
last of nine tables into a storeroom. Thirty minutes later they had suc- ÒSo you could assume that the posi-
The door clicked shut. cessfully moved the antique table to tions that diÝer from one another by
ÒDone,Ó said Dan, breathing heavily. the middle of the right-hand wall, more shuÜing tables around inside one of
ÒFinal check, then IÕll buy us lunch at or less out of harmÕs way, but now one these subpuzzles are eÝectively the
the Plushy Pink Pizza Palace. Okay, two of the pine tables was getting wet, and same,Ó Dan said. ÒThat must cut the
oak square tables, six rectangular ta- the door was blocked [see top illustra- list of positions down quite a bit.Ó
blesÑfour pine and two formicaÑand tion on page 109]. ÒYeah. And sometimes thereÕs only
one giant antique mahogany square.Ó ÒWhat we need,Ó Max mused, Òis a one sensible way to continue moving
ÒRight,Ó said Max, crossing oÝ each map.Ó the tables, if you donÕt want to undo
item on his clipboard. ÒSay, itÕs a bit ÒMax, we can see where the tables what youÕve already done.Ó
crowded in here.Ó are.Ó ÒSo provided you know where youÕre
ÒJam-packed. Wall-to-wall tables, ex- ÒNot a map of the room. A map of starting from and where youÕre trying
cept where weÕre standing.Ó the puzzle.Ó to go, sequences like that can be left
ÒWe were lucky to Þt them in so well.Ó Dan stared at him. ÒHave you gone oÝ the map?Ó
ÒSure were,Ó said Dan, glancing around crazy? Puzzles donÕt have maps.Ó ÒPrecisely. Hand me that clipboard.Ó
the room. ÒOh, no. Look! ThereÕs a leak ÒI hate to contradict you, but puzzles In no time, Max had drawn a map show-
in the ceilingÑdripping right onto the do have conceptual maps, imaginary ing some of the possible positions and
antique. WeÕll have to move it. It will be maps in the brain. Maps that show you moves [see upper illustration on page
ruined if we donÕt.Ó all the positions in the puzzle and how 110].
ÒThis is not going to be easy,Ó Max to get from one to another.Ó ÒIÕve marked the start and the Þnish
observed. Dan nodded. ÒItÕs going to be a pret- positions,Ó Max said. ÒThen there are
ÒCanÕt we pile it on top of the other ty complicated map, Max. There are an six diÝerent ways to place key tables,
tables?Ó awful lot of positionsÑand moves.Ó labeled A, B, C, D, E and F.Ó
ÒNot a chance. The ceiling is too low.Ó ÒTrue. So weÕd better Þnd some way ÒIÕd have expected more than six.Ó
ÒWeÕd be okay provided we could to break the problem into simpler piec- ÒThere are more. This is just part of
move the two formica tables under the es. Hey! ThatÕs it. First of all, letÕs Þnd the map. But six are enough to solve
leak and the antique table to the oppo- out what we can do easily. Then we can the puzzle. Now shut up and listen.
site corner, just in case the leak spreads string those together somehow.Ó
[see diagram on this page]. We can slide ÒWell, if youÕve got a square hole
the tables into the space thatÕs left, one with just the two smallest tables in
by one, thereby making new spaces to it, you can move them around pretty
slide more of them into.Ó freely,Ó Dan said.
JOHNNY JOHNSON
ÒWonÕt we get trapped?Ó ÒThatÕs the idea. A sort of subpuz-
ÒNo. We can crawl underneath the ta- zle, where you move only a few ta- ?
ROBIN BRICKMAN
MOVERS Dan and Max, shown as two circles on the diagram, tables against water damage, they must move the tables until
are surrounded by nine tables. To safeguard the seven wood the two formica ones Þt in the top left corner under the leak.
108 SCIENTIFIC AMERICAN March 1995 Copyright 1995 Scientific American, Inc.
1 2 3 4 5 6
7 8 9 10 11 12
13 14 15 16 17 18
?
JOHNNY JOHNSON
The lines show sequences of forced ÒWhere things are and how to go SEQUENCE OF MOVES illustrates how
movesÑin the sense that if you know from one to another. Well, clues to Dan and Max began rearranging the
where to start and where to Þnish, the those things.Ó tables and managed to slide the large
moves in between are fairly obvious be- ÒIt tells us even more than that. It square antique over to the right wall.
cause thereÕs really only one choice you tells us that one way to solve the puz-
can make at each step, right?Ó zle is to go along the route START-C-A-
ÒOkay, I see that.Ó B-FINISH.Ó ÒIÕll settle for the simplest one.Ó
ÒGood. IÕve shaded in rectangular re- DanÕs face lit up in admiration. ÒYou ÒFine by me. LetÕs get these tables
gions where thereÕs a subpuzzle to could go START-C-D-B-FINISH instead?Ó moving!Ó
solve. To show which one, IÕve drawn ÒSure. Or even START-C-E-F-D-B-FIN- In a little while they had moved the
little pictures of the start and Þnish po- ISHÑbut that would be an unnecessari- tables so that the antique sat far from
sitions within the rectangle, at the ap- ly complicated route.Ó the leak, the two formica tables were
propriate ends of the connecting lines.Ó Dan was getting into it now. ÒOr under the leak, and the door stood free
DanÕs mouth opened like a goldÞshÕs. START-C-D-F-E-C-D-B-A-B-D-C-EÑÓ and clear. Having missed both lunch
ÒSorry, I donÕt quite follow.Ó ÒYes,Ó Max interrupted, Òbut that and dinner, Dan and Max were starved.
ÒWell, suppose you want to know would be an even more unnecessarily They raced to the Þrst ßoor and set oÝ
how to move from C to E. Look at the complicated route.Ó for the Plushy Pink Pizza Palace, which
horizontal line that joins them; it pass-
es through two little diagrams. If you
replace the shaded area in C with the
left diagram and the shaded area in E
with the right one, that will give you
the start and Þnish positions. Because
the moves in between are Ôforced,Õ it
JOHNNY JOHNSON
Copyright 1995 Scientific American, Inc. SCIENTIFIC AMERICAN March 1995 109
PARTIAL MAP of the puzzle shows where START
certain key tables can be placed and how
to move from one position to another. The
lines indicate sequences of moves that are
essentially forced if you know where you
want to go. The small diagrams depict how
the subpuzzles must be arranged at the
start and Þnish of these sequences.
,,,,
A
,,,,,
C
,,,,,
E
JOHNNY JOHNSON
FINISH DEAD
END
complicated maps, even when you use The waitress had included a whole tuna
every trick you can think of.Ó [For addi- and set the licorice on Þre.
tional puzzles, see Winning Ways, Vol. ÒEnjoy your puzzle, sir,Ó she said over
2: Games in Particular, by Elwyn R. her shoulder.
Berlekamp, John H. Conway and Rich- ÒSend it back,Ó Max suggested.
ard K. Guy. Academic Press, 1982.] ÒNo, no, you heard what she said. I
DONKEY PUZZLE involves moving the
ÒLike what?Ó canÕt resist a challenge.Ó He straight- blocks from the position shown in dia-
ÒWell, thereÕs one called the Donkey ened his back, squared his shoulders gram A to that in diagram B. The Cen-
Puzzle, probably from 19th-century and reached for the clipboard. tury Puzzle requires 100 moves to re-
France. ThatÕs a good bit harder. The ÒWhat are you doing?Ó Max asked. conÞgure the blocks from C to B. The
Century Puzzle, invented around 1980, ÒJust wait until IÕve made a map of Century and a Half Puzzle, from C to D,
is harder still. It takes 100 moves to this pizza.Ó takes 151 moves.
110 SCIENTIFIC AMERICAN March 1995 Copyright 1995 Scientific American, Inc.
BOOK REVIEWS by Philip Morrison
The Big Picture modifying the expertsÕ cool economic moved into camera range: many pigs,
and social criteria to emphasize inter- chickens, Þsh ponds, rice storage bins,
MATERIAL WORLD: A GLOBAL FAMILY esting and timely visitsÑthose Òwe can laden freezers, groves of orange and
PORTRAIT, by Peter Menzel. Text by learn from and that I wanted to see.Ó banana trees, and Òtwo acrobatic air-
Charles C. Mann. Sierra Club Books, Finally, he and a respected local com- planes.Ó The Big Picture, a documented
distributed by Random House ($30). panion knocked on doors of typical day of hard work for all, was preceded
CD-ROM from StarPress Multimedia houses in typical neighborhoods to seek by a few days of candid snapshotsÑ
(1-800-782-7944; $59.95). one cooperative family. Each family had 2,000 rolls totalÑand lots of video as
to be willing and able to take part in the well. A long list of questions about dai-
O
ur world now counts about a Big Picture, one that would display the ly living and cherished values was put
billion households. This remark- whole family outside their home sur- to each family as the visitors and hosts
able book, well supported by rounded by their material goods, moved came to know one another. These
the United Nations, vividly presents its out for the day to allow the camera vi- friendly visits took place between late
own cross section of families today, sual access. In crowded Tel Aviv the en- 1992 and early 1994.
with the energy and intimacy of a work terprising photographer hired a con- Whom will you meet? Rice farming is
of art for our times, choosing just one struction crane to lift high a platform the single occupation most represent-
family from each of 30 lands. It carries that bore the entire contents of the ed, just as the world works today. In
a quiet challenge: it is the prosperous small apartment with its tenants, their northern India the young parents and
minority in this world who can most red Alfa Romeo, computer, menorah and four kids are seen on the family string
easily move toward an equitable and all, held in front of their large apart- bed next to three heavy bags from the
stable future. ment house for picture taking. last rice harvest. All they had was in
Californian Peter Menzel, creator of Not everyone wants to put such a the picture, and that is not much. Their
the work, is a globe-trotting photojour- show before the neighbors. But it windowless house of stucco or plaster
nalist of wide experience and acclaim. worked out wonderfully. We even get a measures 12 feet by 30. They named
He selected the Þnal list of countries by full list of those holdings that were not their most valued possessions: four or
PETER GINTER
ISRAELI FAMILY poses outside their apartment (end unit on top ßoor ), surrounded by their worldly possessions.
Copyright 1995 Scientific American, Inc. SCIENTIFIC AMERICAN March 1995 111
Þve prints and Þgurines of gods and billion years, until some local crustal
goddesses; their fondest hopes were for weakness admits a sudden explosion.
a cow or two. One of the most endear- The high-pressure material cools as it
ing spreads here shows the wifeÕs feet expands violently upward, to punch out
among those of her friends, all being a narrow shaft or pipe and to exit with
painted with the traditional red dyes and fearful force, spalling a funnel-shaped
newer nail polish that would proudly set cavity in the surface rocks. The over-
oÝ their silver anklets and toe rings. burden is ßung high, to fall back as a
In Haiti the peasant family we meet jumbled mass of debris across the dia-
is poorer still, growing some sugarcane mond-bearing plug. Such is the celebrat-
and millet to sell. Their breakfast con- ed site at Kimberley, South Africa, the
sisted of potatoes, smoked Þsh, coÝee; Þrst understood among diamond pipes.
potatoes alone make a typical lunch; The rocky outrush must be explosive-
for a typical dinnerÑnothing at all. A ly rapid, for hot diamonds cannot sur-
toy VW Bug, its four wheels missing, vive long at reduced pressure; they turn
adds a touching note. In January 1994 within a minute to stable graphite. The
fear of armed authority was real in Hai- geologists have found within some
ti; the photographer, too, Ògot an adren- pipes the telltale crystal shapes of dia-
aline rushÓ from the militia search at ri- mond in Òincredibly richÓ masses, now
ße point on his way back to the city. A gone to graphite. The natural diamonds
Thai rice-farming household is by com- we know are the rare survivors that
parison rich, its appliances many in reached the surface before Þrst revert-
their two-story frame house with its four ing. Only cooling quenches the atomic
windowed rooms, two water buÝalo in rearrangement.
front of the paddy, and the motor scoot- Diamonds in your hand are metasta-
er most valued among their goods. ble; they may endure for a long time in
We meet about 10 diverse families human terms, but not forever. At white-
from the developed world, among them hot temperatures under pressures of
those of an Icelandic airline pilot, a ca- 50,000 or 75,000 atmospheres, it is the
ble splicer near Houston, a salaried Jap- graphite that is metastable and soon
anese warehouseman and an aÜuent turns to small diamond crystals, most
and worried professor of political sci- easily if dissolved in iron or nickel.
ence at Kuwait University, whose four ÒKimberlite taught scientists what they
cars and a sofa 45 feet long appear in had to do to make diamonds:... squeeze
his familyÕs Big Picture. carbon and heat it unmercifully. . . .
Three instructive color spreads oÝer Those were tricks worth learning.Ó
a dozen or two color photographs each, Not all the tricks were taught by na-
to sample the current televisions, meals ture. This book is a chronicle, often at
and toilets of the world. Most of these Þrsthand, of the livelier tricks of dia-
families are in daily touch with outside mond making, now become a multibil-
events; only four out of the 30 own nei- lion-dollar industry, indispensable in
ther a radio nor a television. The work metalworking and in drilling for oil, not
week is a long one everywhere outside to mention one-hour eyeglass produc-
the developed world; even in most de- tion or thermonuclear bombs. Here is a
veloped countries the toil of women re- lively tangle of the physics of materi-
mains long. als, its experimental ingenuity and the-
This is a record of striking value, in oretical insight, with lots of intrigueÑ
particular for the schools. It should ap- personal, corporate and governmental.
pear anew every 20 years. No time trav- Even the circumstances of the Þrst
eler could bring back a more valuable true synthesis of diamond are still not
souvenir! beyond doubt: it might or might not
have been done and recognized in
Stockholm in 1953, although the claim
Grace under Pressure made for Schenectady late in 1954 has
its strengths. The Þrst patents for dia-
THE NEW ALCHEMISTS: BREAKING mond synthesis were certainly issued to
THROUGH THE BARRIERS OF HIGH PRES- the General Electric Company, although
SURE, by Robert M. Hazen. Times Books, delayed from publication for a few years
1993 ($23). by U.S. government classiÞcation as
strategic. Once classiÞcation was lifted
W
herever you live on the earth in 1959, GE lawyers Þled patent papers
there is a region nearby where across the globe.
diamonds aboundÑbut it is One week later De Beers, Pretoria
out of sight 100 miles underground. sovereign over the diamond trade, sub-
The geologists have inferred this from mitted full diamond-making patents of
indirect evidence, although we have nev- its own. How much was leakage, how
er seen diamonds form in nature. They much convergence, is still clouded and
sit stably in the hot depths for several still fascinating. The eÝorts at patent
112 SCIENTIFIC AMERICAN March 1995 Copyright 1995 Scientific American, Inc.
breaking led to a decade of subsequent high as 100,000 atmospheres. There is
combat between the well-documented still ample room for pinhead samples
GE defenders and the ingenious assail- and probing photons in all variety. It is
ants mustered by De Beers. Would you the high density of energy that makes
believe the audacious witness George high pressure, even when the total en-
Kennedy, a U.C.L.A. diamond physicist ergy is minor.
known both for his originality and his Faraday could have done this; our
precision? He was versatile enough to author thinks even Isaac Newton might
have lectured thrice in one single day have. Heating is available now by care-
at OxfordÑon high-pressure physics, ful design, and x-rays, laser light and
on new exotic orchids and on the dating meticulous spectroscopy are welcome.
of pre-Columbian pottery. On his own Special diamond shaping distributes
oÝ-campus time, the nonpareil Kennedy stress over curved faces and can push
married a Hollywood heiress and scan- the pressure up 10-fold: such desktop
dalously burned down her house while diamond vises now outperform all the
smoking in bed, but he soon made heavy steel and carbide presses in the
amends, Þnding them a magniÞcent world, be they Þve stories high, as is
joint residence in the Santa Monica one in Moscow.
Mountains. Of course, the little cells cannot put
By now patent licenses abound, and out much new material; their product
diamonds grow for GE, De Beers and is knowledge. Many minerals expected
many others. GE operates its scores of in the earthÕs interior are now opened
massive 1,000-ton presses around the to pressure probing. It looks as if the
clock in its plant in the suburbs of Co- earthÕs mantle might converge to one
lumbus, Ohio, where the well-run Òpres- optimal crystal form, whatever its mix
sure cookersÓ synthesize some 30 tons of materials. That old grail of the high-
of tiny diamonds a year. Smaller, newer pressure lab, hydrogen atoms squeezed
presses, in which six cleverly aligned to merge into an electron-sharing metal,
tungsten carbide anvils press hard on is also strongly hinted at but still eludes.
the faces of a cube-shaped sample Plenty of novelty is ahead at atomic
chamber, make a number of larger dia- energy densities, even though the far
monds, fully competitive in smaller grander energy densities between col-
niches of the trade. liding protons will not be studied in
Synthetic diamonds appear now at the echoing tunnels of Texas.
about 100 tons a year, most for indus-
try. The big mines in three continents
yield about 10 tons a year, half for the Local FluÝ and Other StuÝ
gem trade. Meanwhile Du Pont makes
diamond abrasives in noisy caverns by THE GUIDE TO THE GALAXY, by Nigel
passing high-explosive shocks through Henbest and Heather Couper. Cam-
graphite-copper mixes. Whole Þlmy lay- bridge University Press, 1994 ($49.95;
ers of diamond can be deposited from paperbound, $24.95).
vapor, as the winds from carbon stars
I
made tiny diamond plates that drifted t is galaxies that shine out from the
into the early solar system. More re- gravitational potholes in our cos-
cently diamonds have been made from mos. We dwell right inside one
the new molecular buckyballs (C 60 ) well showy specimen so that surely we are
squeezed at room temperature. obliged to map our own backyard. This
But the most unexpected innovation book is a comprehensive tour guide,
is not diamond synthesis. It is the dia- with many square, colorful pages (al-
mond-anvil pressure cell developed in though often the color is computer-
1959 by Alvin Van Valkenburg and contrived, as conventional as the red
Charles Weir of the National Bureau of that once marked maps of the British
Standards. ( An independent Chicago Empire). Dust is both pierced and dis-
team shares credit.) ÒAn elaborate nut- closed by powerful infrared arrays, x-
crackerÓ of two gem diamonds, the little rays single out dilute pools of Þercely
device reproduces the pressures of plan- hot gas and the radio dishes analyze
etary depths right there on the desktop both for cold molecules and for mag-
under the experimenterÕs hand. Turn a netized relativistic plasma. Even the
small crank to wind a sturdy screw that coded colors are packed with meaning.
presses together two ßat diamond faces The chapters open with a useful his-
(nothing is stronger or more transpar- tory of what we know. The black river
ent) that squeeze between them a soft- of the Milky Way that dominates a me-
metal gasket ring. High pressure ap- dieval Korean map of the visible sky is
pears on whatever is placed in the ring. a splendid icon. Like Mount Everest, the
Ten or 20 pounds of push can bring a Milky Way is there. ( Its current loss to
sample volume that is only one tenth of the eye within the wasteful skyglow of
a millimeter across up to pressures as our cities is a major deprivation, com-
Copyright 1995 Scientific American, Inc. SCIENTIFIC AMERICAN March 1995 113
parable with and more widespread than
the scarceness of the blue whale.)
Our knowing guides begin the tour in
the suburbs of the Milky Way. We gain
perspective from the galaxies that com-
prise its near neighbors, called the Lo-
cal Group. Our own view of the star for-
est is lost in trees. By such comparisons
we came to expect and to Þnd how the
stars circle the carousel of the disk
(turning clockwise as viewed from the
north), locally bobbing and even weav-
ing under complicated Newtonian pulls.
Then we map the sights in three arms
of this, our spiral. The Local Arm (some-
times called after Orion, where a near-
by starburst glows) lies between the Per-
seus Arm, just outward, and the arm
next inward, named for Sagittarius. It is
in the Perseus Arm, for instance, that we
Þnd the Crab Nebula and the brightest
radio source, Cassiopeia A, two relics of
very recent supernova explosions. Our
own Arm might as well have been
named for any bright constellation,
since all the bright nearby stars are
within it. Hundreds of important land-
marksÑskymarks?Ñthat bear familiar
names (the Hyades, the Jewel Box, Kep-
lerÕs supernova ) are colorfully mapped
and indexed at diÝering scales. It be-
comes pretty clear how easily we over-
simplify the Þrst maps we make. Ages
and motions allow inferences that no
static map can make clear, especially be-
cause without a deeply physical study
we view the remote sky only in two and
not in all four dimensions of space-time.
Magnify two small regions of special
interest. First, inspect our precincts on
the small scale of 50 light-years. We re-
side within a gas cloud of modest hy-
drogen density, dubbed the Local FluÝ,
not extending out as far as Sirius or
Arcturus. Radio astronomy maps its
physical bounds. But that ßuÝ is itself
within a hotter, more dilute gas bubble
a couple of hundred light-years in di-
ameter. The implied sequence is quite
general; newly formed hot stars do
blow bubbles in the same hydrogenous
medium from which they formed, and
those blowing winds, sometimes the
shock waves of full supernova explo-
sions, then compress the gas they reach,
often to initiate star formation anew.
The many infrared images here are
essential, but they so lack the crispness
of the optical photographs that they
impelled this viewer over and over again
to wipe his spectacles. This is no fault in
the wonderful observations. We badly
miss a page or so to explain how these
infrared images are made and their cur-
rent limitations. Still, any readers who
like to see and to know just where they
are will enjoy this textually and visually
nourishing volume by two deft writers.
114 SCIENTIFIC AMERICAN March 1995 Copyright 1995 Scientific American, Inc.
ESSAY by Steven Shackley
W
hat a wonderful idea it seems, to compel museums and universities to ed from the prehistoric inhabitants of
especially in these politically serve as intermediaries between the the same area. Many groups, particu-
correct days: force museums federal government and unoÛcial na- larly early hunter-gatherers, moved fre-
and universities to return sacred relics tive groups that are attempting to use quently among many regions. The fed-
and excavated burials that scientists claims Þled under NAGPRA as a way to erally recognized Native Americans liv-
stole from Native Americans and Ha- gain federal recognition. The goal, of ing in an area today may have no more
waiians years ago. In November 1990 course, is to get the attendant rights and lineal connection to the makers of the
that logic led the U.S. Congress and funding. NAGPRA has turned into an local archaeological remains than do the
President George Bush to enact a new unintended piece of civil rights legisla- excavators themselves. One solution fa-
law, the Native Americans Graves Pro- tion, one that may require Congress to vored by certain native groups is to give
tection and Repatriation Act (NAGPRA). grant new entitlements even as the old the remains to the local tribe regardless
Most institutions openly embrace the ones become prime targets for cutting. of any empirical connection between
need to repatriate Native American ar- modern and ancient dwellers. Some Na-
tifacts, but NAGPRA represents a quin-
tessential example of good intentions
gone badly awry. In typical fashion, the
I do not mean to say that all the lawÕs
eÝects have been bad. Museums and
universities are now communicating
tive AmericansÑand some scientists,
including meÑoppose such action on
ritual and rational grounds. Allowing
lawmakers gave little forethought to with native groups to an extent unheard one set of people to dictate the treat-
the social and economic consequences of a few years ago. In the course of ne- ment of relics that belong to another
of the act, not only for museums but gotiations, many of those groups are re- runs directly contrary to the ideal that
also for tribal organizations, which must alizing that they have no proper way to motivated NAGPRA in the Þrst place.
now sift through reams of inventory catalogue, store and preserve the sacred I have deliberately not argued for the
lists describing objects and skeletal re- artifacts and burial remains covered by scientiÞc value of keeping collections
mains that may or may not be associ- NAGPRA. In addition, such groups often by universities; this topic has been ar-
ated with their ancestors. The law is no longer practice the original rituals in gued to death. It is only fair that native
straining museumsÕ meager resources which the items were used, or if they do, groups should be able to regain pos-
and may foster legal battles that will do they may produce objects more useful session of artifacts that legitimately be-
little for scientists or for native groups than the old, fragile relics. Native Amer- long to them. But it is worth noting that
trying to rebuild their ethnic heritage. icans and scientists are thus Þnding that genetic research related to skeletal re-
Under NAGPRA, institutions have had their interests often dovetail. mains has reached a critical threshold;
to prepare complete, lengthy lists of One example of the new spirit of col- future work would have great import
holdings that might properly belong to laboration is taking place at the Santa for science and for native groups alike.
Native American groups. The deadline Barbara campus of the University of Cal- Archaeological research into the geog-
for the summaries of this material has ifornia. Phillip L. Walker, a professor raphy of tribal territories, patterns of
already passed; full inventories of ar- there and a member of the NAGPRA Re- migration and interactions between
chaeological artifacts are due in Novem- view Committee, entered into an agree- groups could actually aid Native Amer-
ber 1995. But the Department of the In- ment with the Chumash tribe to create icans in their land claims.
terior, which oversees the enforcement an on-campus ossuary that is controlled The greater problem here is the bu-
of NAGPRA, has yet to generate the fed- by the Chumash but open to Walker and reaucracy in charge of the process. At a
erally mandated guidelines to aid the to the university. This particular solu- 1992 meeting of the American Anthro-
short-staÝed museums in their tasks. tion was possible in part because of the pological Association, C. Timothy Mc-
Recently Congress allocated $2.3 mil- unambiguous history of the Chumash. Keown of the Department of the Interi-
lion to assist both museums and native There is good archaeological and his- or stated that he would feel the depart-
groups in implementing parts of the law. toric evidence that the modern Chu- ment had done its job if all parties were
That sum could not fund the work at mash are direct descendants of the na- dissatisÞed. No wonder the electorate
my museum, let alone the hundreds of tives whose remains have been exca- is so angry at government. Universities,
aÝected institutions around the country. vated on the California mainland and museums and native groups appear,
In the meantime some members of nearby Channel Islands. It therefore despite the many obstacles, to be Þnd-
the NAGPRA Review Committee, an or- seemed sensible to grant the Chumash ing resolutions to the problem of repa-
ganization of Native Americans and control over most of the artifacts found triation. In the end, I am sure the NAG-
nonnative Òprofessionals,Ó have been in this region. As long as the current PRA Review Committee and the Depart-
interpreting the law to suit their ends. Chumash group continues in power, the ment of the Interior will take credit for
Several sections of the original text of agreement worked out by Walker and the beneÞcial solutions that result.
the law state that it applies only to the university will stand. If, as is often
those ÒtribesÓ that have received feder- the case in politics, the next Chumash
al recognition through due process. The leadership disagrees with its predeces- STEVEN SHACKLEY is an assistant re-
review committee now suggests that sorÕs decisions, Walker could lose ac- search archaeologist at the P. A. Hearst
groups that are not federally recognized cess to the collection. Museum of Anthropology at the Univer-
should be consulted, too. The eÝect is In much of North America, working sity of California, Berkeley.
Copyright 1995 Scientific American, Inc. SCIENTIFIC AMERICAN March 1995 115