California Institute

of Technology

E N GI NEE

Volume LX, Nllmber I,

I 9 9 7

IN THIS ISSUE

Taxi Driving

Land-Speed Racing

Unleaded Gasoline At the Bonneville Salt Flats, all sorts of strange and wonderful machines wait in line to break the speed limit and perhaps even a world record. A motorcycle whose stream­ lined body shape was designed and tested at (altech did just that last fall. To see this remark­ able bike, turn to page 32. California Institute

o f Technology

2 Random Wa lk

10 Tax i Drivers and Bea uty Contests - by C o lin F. Camerer Real people don't always behave the way classic economic theory says they should.

The emerging field of experimental economics is beginning to find out why.

20 Duck Soup and Lea d - by Clair C. Patterson In this excerpt from his oral history, the late professor of geochemistry talks about meteorites, lead, the age of the earth, lead, the oceans, lead, polar ice, and lead.

32 I'm Not Driving Fast, I' m Just Flying Low - by Douglas L. Smith A group of aeronautics grad students and Caltech's 10-foot wind tunnel are enlist­ ed in pursuit of a land-speed record at the Bonneville Salt Flats.

42 Books

44 Faculty File

On the cover: The taxi Engineering & Science (fSSN OOJ 3-7812) is published Edward M. Lambert quarterly at the California fnstitute of Technology, 1200 Pmident o/Ihe Alumni AJJociatioll stand outside Grand East California Boulevard, Pasadena, California 91125. J. Ernest Nunnally Central Station in midtown Annual subscription $10.00 domestic, $20.00 foreign Vice President jor Imlilllle Relatiolls Manhattan is really a air mail; single copies $3.00. Third-class postage paid at Robert L. O 'Rourke Pasadena, California. All rights reserved. Reprodunion of Associate Vice PresidelJl for Imlilllte Reia/iom laboratory for some material contained herein forbidden without authorization. experimental economists. © Copyright 1997, California Institute of Technology. Published by the California Institute of Technology and Their findings blend the Alumni Association. Telephone: 818-395-3630. psychology and economics PICTURE CREDITS: Cover - Jane Dietrich; inside front cover, STAFF: Editor - Jane Dietrich to explain human behavior 36-41 - Doug Smith; 2 - Beth Fernandez; 3, 7 - JPL; 4,6 - Managing Editor - Douglas Smith much better than standard Bob Paz; 5 - Laurie Spauldingj 8 - Alycia Weinberger, STSd; COlltrib1ftill}!, W'rilel1 - Rebecca economic theories do. The 9 - Heidi Aspaturianj 10 - NYSE; 12 - Chris Anderson; 13 - Rothenberg, Robert Tindol Linda Babcock; 15,16 - Colin Camerer; 17 - Teck-H ua Ho; Copy Editors - Michael Farquhar, story begins on page 10. 18 - Charles Plan; 20 - Yigal Ercl; 21,22,25,31 - lorna Danielle Gladding, J ulie Hakewill Patterson; 23,29 - Cahech Archives; 27, 30 - Floyd Clark; Business Mal/agel' - Debbie Bradbury 32 - E. John Tbawley III; 34 - Jerry Landry; 35 - Scan eiratlation Manager - Susao Lee Rathburn; 42,43 - Tom Willard Phologl'dpher - Robert Paz It can't leap school buses the way Evel Knievel did in his prime, but an eight-wheel ed prototype suspension system strutted its stuff by climbing over Brownie Troop S I of Monrovia, California, during JPL's Kid's Day last summe r.

Random Walk

blooming. The lander rights How's My D RIVING ? itself (the petals open with C ALL 1-800-NASA-JPL enough fotce that they can actually lever the lander out from between rocks , should Tbe only likely di s a ppo int ~ If everything goes accord­ only Likely disappointment it be unlucky enough to get ing to plan, a six-wheeled is Sojourner's top speed-a wedged), and its stereo ment is Sojourner's top all-terrain vehicle named So­ blistering 16 inches per min­ camera makes a sweep of its journer will stan exploring uee. And speaking of toys, speed-a blistering surroundings. At 6:25 and the surface of Mars this sum­ Cooper has the world's coolest 6: 30, tWO metal ramps unroll 16 inches per minute. mer. Brian Cooper, a member sandbox, roo--while the real in opposite ditections from ' of the technical staff at Cal­ Sojourner is en route to Mars, one of the petals, like ted tech's J et Propulsion Labora­ Cooper practices weekly wi th carpets preceding the queen's tory, will be sitting in the an identical copy, which he entrance. The queen in this driver's seat. Well, SOrt of. drives around in the Space case is Sojourner, which spent Sojourner is the size of a Flight Operations Facility, the trip bolted to the petal. microwave oven, and Cooper in a large room that has been Meanwhile, back at the Lab, will be sitting at a computer converted into a Mars-scape Cooper and the flight team in his 1FL office. Cooper has of sand and rocks. The other will be srudying the pano­ created the user interfaces team members rearrange the rama to decide which ramp for all the robotic vehicles rocks our of Cooper's view, Sojourner should drive down. leading up to Sojourner since and he then has co negotiate Sojourner will do so at 8:20 1985, and he drove them dur­ the terrain using the same p.m., posing for a press photo ing testing, so he was the data he'd get from Mats. at 9:30. Later, the rover will natural choice for this job. So here's rhe plan: ]pr;s use its own stereo camera to Sojourner is any kid's Mars Pathfinder mission hirs see how the lander fared. Christmas wish. It's solar Mars at 10:00 a.m. Pasadena These photos, and the othet powered, so it never needs time on the Fourth of July, images the mission sends batteries (although it has and "hits" is the operative back, will be posred on the some for nighttime opera­ word. Cushioned by its air Web at http://mpfwww.jpl. tion). And jt's a killer on the bags and given some exrra nasa.gov as soon as they are obstacle course. Each wheel spring by the red planet's low processed. has an independent motor in atmospheric density, Path­ Cooper will have a joystick the hub, and anyone wheel finder will make several and 3-D goggles, but that's can lift the rover's full weight. multi-stoty bounces , perhaps where the sim il arity to an The sllspension system is rolling a kilometer or more arcade game ends. For one flexible enough that a wheel before coming to rest. By thing, it will take 11 minutes can drive over things larger I :00 p.m., rhe ai r bags will for his radioed commands to than its own diameter­ have deflated, with three of reach Mars. So instead of Sojourner can even climb rhe tetrahedral spacecraft's steering in teal time, Cooper stairs! (Not that NASA fOllr faces then opening like will move a cursor the size expects to find any.) The a time-lapse movie of a rose and shape of the rover over

1 ENGINEERING & S(lfNCE NO. 1997 the 3-D landscape he sees. glass." Cooper and the sci­ mounted on a flexible neck (This landscape is a mosaic ence team study the images because the APXS has to of image pairs, corrected for and negotiate where So­ remain pressed tightly camera distortions and per­ journer should go next-the against the sample for the up spective, from ei ther the geologists decide which rocks to 10 hou rs it takes to get a Below:The 60- by 120-mile oval in lander or the rover.) Starting to sample, and he tells them complete profile. from where Sojourner is in if Sojourner can get there. Pathfinder's landing site this Viking Orbiter photograph the picture, he clicks the Once a consensus has been (l9.SoN, 32.S0W) was marks Pathfinder's planned landing cursor on an itinerary of reached, Cooper assembles the chosen because it's a geologic site in the Ares Valles outwash points to be visited . Sojour­ command sequence, which souvenir store. Eons ago, a plain. The mouth of Ares Valles is ner navigates to those "way­ gets uplinked back to the wall of water hundreds of feet points" on its own, using an lander. He figures he'll be at high and 10-15 miles across at the bottom right of the image, on board gyroscope to deter­ it for 12 or more hours a day, tore through the Martian and north is toward the top. mine its heading and taking at least ro begin with; there's highlands and ridged plains, the average of the odometer a backup driver, Jack Morri­ gouging out a 700-mile-Iong readings from each wheel to son, should Cooper get an arroyo named Ares Valles. calculate the distance trav­ inopportune flu bug. The Although nobody knows for eled. "It's like a blind person primary mission only lasts a sure, the water is believed to with a cane," says Cooper. "It week, bur with luck and have come from the sudden, knows where it wants to go, TLC, Sojourner might keep catastrophic melting of and feels its way there." The going for months. Sojourner roughly enough ice to cover cane is a set of laser beams will stay within 10 meters of California's Central Valley that draw a five-by-five grid the lander to start with, but one mile deep from Bakers­ on the ground in front of the in a few days Cooper hopes to field to Redding. The water, rover, starting at the front go out of its view-perhaps and the collection of rocks wheels and extending out­ as far as 300 meters away­ from all the terrain it had ward in a fan shape. Measur­ and rely on the rover's scoured clean, eventuall y ing how the grid points camera. spilled out into the lowlands deviate from where they'd be The point of all this wan­ of Chryse Planitia. And it's if the ground were flat tells dering is to study the rocks there, in the floodplain of Sojourner the size of the rocks and soil. Sojourner carries an Ares Valles, that Sojourner ' in its path. On board tilt­ Alpha Proton X-Ray Spec­ can sample rocks from thou­ meters give warning of the trometer (APXS), built by sands of sq uare miles of Mars potential for toppling over. If the Max Planck Institute in without straying far from the the hazard-avoidance software Mainz, Germany, and by rhe lander. Of course, the geolo­ chickens out (it can be made University of Chicago. To gists won't know where any more gutsy as Cooper gains determine the surface's ele­ one rock came from , but they confidence in the rover's mental composition, the can get a lot of information abilities), Sojourner backs off APXS bombards a sample about the region's overall and tries another route to the wirh alpha particles (helium mineralogy. next waypoint. Sojourner nuclei) and measures the Oh, and by the way, 1-800- prefers to drive in smooth alpha particles, protons, and NASA-JPL tings at a paging arcs, like a car, but it can turn X rays that come back. The service; don't call that num­ Right: A view of the Mars Room as in its own length, like a tank, sensor head, which is about ber. Sojourner is unlisted. if it has to. the size of a demitasse cup, is D -DS Cooper sees it. The colored rover The rover and the lander in the center of this mosaic of six talk to each other all the images from the lander's camera is time, and once per Earth the cursor that Cooper manipu- day (the Martian day is 41 minutes longer than a ter­ lates to drive the real rover, a restrial day) the lander down­ corner of which can be seen in the links the accumulated conver­ lowe r right frame. The darts mark sations to Earth, along with all the images and data from the waypoints the rover will visit. the lander's other experi­ Superimposing the cursor exactly ments. To ease the competi­ on the real rover's image tells the tion for transmission time, computer where the rover is and the images Cooper uses are compressed, but there's a fine which way it's facing. The comput­ line here. "On some tests, it er then tells the rover the distance was like we were looking and heading to each waypoint. through Vaseline-covered

19 9 7 ENGINEERING & SCIENCE NO. 3 MAN OF THE YEAR

AJDS researcher David D. infection. theorizing, accord­ Ho '74, was named Time ing co the Time article, that if magazine's Man of the Year the virus could be knocked fo r 1996 becallse his (a nd out at this paine, it could be Others') work brought real completely eradicated and the hope this past year that AIDS immune system would still ca n be beaten. The tide be healthy enough co re­ aga inst AIDS has turned, and bound. H e announced at the those who accomplished this 11 th International Confer­ feat, Time claims, history will ence on AIDS in Vancouver anoint as true heroes of the last summer that none of his age. subjeCts showed a (face of the Time devotes 16 pages of its virus in their blood after a December 30, 1996, issue co year of treatment, although H o, his career, and the assault this does not necessarily on AIDS. After graduating assure that the virus has been from Caltech in biology (not eradicated in their bodies. phys ics. as the magazine Remnants of the virus could states) Ho went on to Har­ still be hiding, infection may vard Medical Sc hool and, as rec ur if treatment is stopped , a virologist, began work on and the drugs are very AIDS immediately after the expensive. "But the worst d isease was discovered in fea r-the one that seeded a 1981. Acco rding to Time, he decade with despair, the was among those who, from fo reboding sense that the the very beginning, suspected AIDS virus might be invin- · that a virus was the cause. cible- has finally been W hile most researchers con­ subdued," says Time. The Sherma n Fairchild Library of Engineering and Applied centrated on the later stages Science, featuring the latest in "t echnological information of the disease, Ho's most im­ delivery," opened for business January 2. Made possible by a po rtant insights ca me from $9.6 million grant from the Sherman Fairchild Foundation, the hi s focus on the early stages of infection with HIV, character­ new library arose on the sit e of the old steam plant, between ized by a long period after Thomas and Spalding Laboratories. Besides providing a central initial infection during home for the collections from the department libraries in which, it had been presumed, the virus was dormant. Ho aeronautics, applied physics, electrical engineering, chemical showed that HIV is ex treme­ e ngineering, computer science, engin eering. earthquake ly active during this symp­ engineering, and environmental engineering. t he building is tomless petiod, replicating computerized, wired, and networked up to its gutters. But itself billions of tim es over and causing the immune it's also user.friendly, with comfortable furniture and cozy system to wear itself OUt in reading spaces (yes, there are still a co unteraHack. books and journals). New therapies with combi­ nations of drugs including protease inhibitors have shown promise for holding Al DS in check in its later stages, although not for curing ie. Ho, who has been director of the Aaron Dia­ mond AIDS Resea rch Center in New York since 1991, used these drug combinations co trea t 24 subjects in the very ea rly smges of H IV

4 ENGINEERING & SCI E NCE NO. , 9 9 7 homemade banner strung from Millikan Library TWENTY-FIVE YEARS OF WOMEN AT CALTECH - BUT WHO'S COUNTING? reading "Welcome Co-Techs," which she thought more clever than malicious. A few end-of-year events Board of Trustees that the Otherwise, she says , she was casually marked the 25th Institute proceed with all nOt treated any differently anniversary of the admission deliberate speed to the ad­ [han the male students, was of female undergraduates to mission of women to under­ not subjected to different Caltech: some concerts graduate work at Caltech"­ pressures, and was made [Q alluded [0 the date, a academic bureaucracy's ver­ feel "welcomed, successful, Women's Club luncheon sion of a call for revolution. and valued." Kim Fisher celebrated it. Not exactly Evidently the Board of Thomas, an environmental hoopla. But then, 1996 was Trustees was persuaded but economist at Whittier not exactly the 25th anniver­ cautious, because in Novem­ College, says she passed her sary; it was probably the ber 1968 it passed a reso­ whole first year (she entered 26th, depending on how lution approving "in prin­ in 1971) with a knot in her you count. ciple" the admission of stomach, but that can prob­ Certainly people at Caltech women both as entering ably be said of many Caltech can do rhe math. Maybe rhe freshmen and as upperclass freshman. anniversary just slipped by Woman of the Year transfers. Caltech's first truly The speakers emphasized, coeducational class entered before anyone noticed because £&5, October 1970 however, that they had been a it's now so hard to imagine two years later, in 1970, and diverse group whose experi­ Caltech without women. But included 29 first-year women ences differed. One alumna­ the change to coeducation and two sophomore transfers. not at the luncheon-has was neither an easy battle nor 10th year, or the first) (The present freshm an class another freslunan-year a sure thing. At a Women's described the chronology. includes 67 women). A few memory: a professor privately Club luncheon in November Owen began lobbying for the more women came as upper­ advising her [Q drop his class, 1996, Professot of Biology, admission of women as early class transfers in 1972, so because she wasn't going to Emeritus, Ray Owen (with­ as 1962, citing the pervasive 1973 marked the first grad­ be up to it. out whose crusading efforts soc.ial changes of the time; by uation of women from Cal­ For those committed to we might be celebrating the 1967 he was calling the ex­ tech, and 1974 the gradua­ celebrating 25 years of clusion of women from under­ tion of women who had women at Cal tech, several graduate work at Caltech "an entered as freshmen. authentic milestones loom on inrolerable anachronism." And how did that pioneer­ the horizon: 1998, the 25th By 1967 [Owen] was calling And in Novembet 1967 the ing class of women fare? anniversary of the first female the exclusion of women from Faculty Board minutes record Surprisingly well, if the sev­ undergraduates to graduate; that "Professor Owen moved. eral alumnae who spoke to and 1999, the anual 25th undergraduate work at .. that the Faculty Board the Women's Club are a rep­ anniversary of the graduation recommend to the Faculty resentative sample. Attorney of the first genuine class of Cal tech "an intolerable that the Faculty recommend Debbie Dison Hall remem­ women at Cal tech. Mark your to the Administration and the bers being greeted by a calendats. D- RR anachronism."

MORE THINGS To WORRY ABOUT

A popular treatment for doesn't involve surgery, and '9 1, PhD '96) decided to find kidney stones may not be as can be done even in a mobile our what was going on. They safe as has been thought. The clinic. But there can be used a laboratory lithotripter, procedure, called extracorpo­ short-term side effects: inter­ as the stone-crushing real shockwave lithotripsy nal bleeding, blood in the machine is called, buil( by by the doctors and plain old urine, and bruises where the Dr. Bruce Hartenbaum of lithotripsy by the rest of us, shock wave enters and exits H-Tech Laborarories in Santa uses tightly focused shock the body. This wou ld seem Monica, California. waves to pulverize the stones to indicate tissue damage and Sturtevant and Howard (which really are rocks­ possible long-term effects, so suspected that the shock they're mineral deposits, Bradford Sturtevant (MS '56, waves weren't so tightly usually calcium oxalate) while PhD '60), Liepmann Professor focused after all. The speed they're still inside the kidney. of Aeronautics, and grad of sound in blood is about 5 The procedure is quick, student Danny Howard (MS percent higher than in tissue,

I 9 9 7 ENGINEERING & SCIENCE NO. • and the kidneys, which filter mm); and kidney (30 mm) [he blood, ace filled with between the shock source and blood vessels. Some simple the pressure sensor. Again, ca lculations showed that the the peak dispersed and lost porrion of a shock wave most of its punch. passing through a blood To assess the potencial for vessel can get far enough collateral damage. Howard ahead of the rest of the wave repeated the bead and tissue to distort the overall wave­ tests, replacing the pressure front. As the wave travels sensor with a nitrocellulose through the maze of blood membrane about 10 microns vessels en route to conver­ thick. This membrane, gence on the kidney scone, which represented cell walls the wavefront can get hope­ in the vic inity of the kid ney lessly folded. The pulse's scone, emerged looking like energy can blur, forming mi­ i[ had been hie by buckshot. croscopic regions of intense Small holes and tears could shear that can tear the sur­ appear after as few as 20 rounding tissue and cause pulses; after 100 shots there additional damage with generally wasn't much left. each pulse. Since several thousand pulses are generally administered in The first set of experiments "A Conversation with Walt er Cronkite" launched the Lee DuBridge established a baseline by a lithotripsy treatment, this finding out what a properly is clearly cause for concern. Distingui shed Lecture Series last November. Playing to a capacity crowd in focused pulse looks like. By "When tbese machines were Beckman Auditorium, the veteran newscaster recounted stories of using a pressure sensor im­ originally approved," says presidents he has known (from Hoover onwa rd) and events he had Sturtevant, "basicall y none mersed in four test fluids witnessed in his career-most of t he major ones of this century since (wace[, e[hylene glycol, of the right questions were glycerin, and castor oil), it asked." Sturtevant is now World War II. Jess Marlow, news a nchor of KNBC, was the other half of the was establis hed that a pulse collaborating with D r. J ames conversat ion, I

• ENGINEERING & !OCIENCf NO • '99 7 DEC OMPRESSION CHILI HEATS UP STUDENTS

Wan, scaring, night­ superm arket- look for pre­ dwelling, they stagger [arch chopped onions, she's learned; from whatever dark place much easier on the eyes­ they've nested. It's the smell then the whole crew m ixes u p that draws them, char and the the concoction (see recipe warm red liq uid. They need it below) in the kitchen of to replenish vital energy, to Chandler Dining Hall, and rebuild neural circuits, ro go doles out steami ng bowls on. They need it badly. And on the sidewalk between they need a lot of it. About Chandler and Winnett Stu­ 3,000 bowls, in fact. dent Center. The Caltech Y's Decom­ The fi nal, and most crucial, p ression Chili has been a step of the process is the cusrom-"permanent if addition of spices, and here, sporadic," says Sue Borrego, Borrego admits, she was a executive director of the Y­ failure: her chili was too since the 19705. But in the bland. The students muti­ past few years it's become a nied, dumping in three times regular part of undergradu­ more chili powder. ''I'm from ates' Decompression, an event the Midwest," she says apolo­ designed to lighten chose getically, "where ketchup is a hellish weekends before fall, spIce. D -RR winter, and spring finals. (D ecompression was D ecem­ ber 7-8 last term .) When DECOMPRESSION C HILl Borrego becam e director of the Y several years ago and Saute: heard about the chili, she • 90 Ibs ground beef asked innocently, "Who cooks (s ubstitute carrots, broccoli, it?" "You do," she was told. mushroom s, other vegeta­ So she became a chef on an bles in season fo r meatless industrial scale. recipe) She has a lot of he lp: rhree • 30 Ibs onions, chopped to five undergraduates each • 20 lbs (one case, or 25 time, and sometimes alumni. bunches) celery, chopped (In Decem ber, Greg Steiert '96, nostalgic for all-nighters, Si mmer with: came down from his job at • 24 32-oz cans each Intel in O regon to saute and tomatoes, beans stir). And they have it down • 60 oz chili powder (to to a science, she claim s; she taste) procures the vanful of ingre­ d ients fo r both beef and Yield: abour 3,000 bowls vegetarian batches at her local

Not the tracks left over from a dirt-bike convention (or a lost robotic vehicle), these complex faults and ridges were created by tectonic processes in Europa's surface. jupiter's icy moon was captured in this very high-resolution mosaic image made during Galileo's first close pass (2,119 miles) by Europa in December, during the spacecraft 's fourth orbit around Jupiter. The resolution of the mosaic (the section shown here spans about 10 miles by 30 miles) is SO times better than the best Voyager images and 500 times better than Voyager pictures of Europa. The sun is illuminating the surface from the right.

199 7 fNGINEERING & SCIENCE NO. 7 The new infrared image of the nucleus of NGC 1068 (far left). made at a wavelength of 2.2 microns. shows the same level of detail as the Hubble Space Telescope image (at ultraviolet and visible wavelengths) of a much wider view of the galaxy's nuclear region (belOW). Weinberger's image is 160 light-years on a side and sits just at the bottom apex of the central

bright cone in the HST picture, which spans 1,200 light ~ years by 1,700 light-years. The two pictures are oriented in the same direction here, but the close-up infrared image shows an elongated structure. probably made of stars, which does not point in the same direction as the cone in the larger-scale picture. A distribution of stars li ke this could help force gas into the black hole at the center. In both pictures, the colors from red to white represent increasing brightness.

CLOSE-UP VIEW OF A GALACTIC NUCLEUS

The most detailed infrared which are only 12 lighe­ image of the environmenc years across. T his is an shape of a "bar" are a way of of an active black hole has extremely small d istance funneling material to an emerged from work at the 10- by galactic standards, as active nucleus. As gas moves meter W. M. Keck Telescope small as about three times in a noncircular distribution on Mauna Kea, Hawaii. G rad the distance between the sun of stars, such as what may be student Al ycia Weinberger and its nearest stellar neigh­ emiSSIOn. T he near-infrared seen in Weinberger's image, and her collaborators have bors. Alehough eaken from ligh t used to make this it is forced into orbits likely used rhe computer-intensive a ground-based observatory, picture typicall y traces the ro take it near rhe ceocral technique of speckle imaging this image has resolution distribution of hot dus t and black hole. This provides a at the Keck to image rhe as fi ne as the H ubble Space cool stars. continuous mechanism for nucleus of NGC 1068 and Telescope ac hieves in the In NGC 1068, however, ie "feeding" the central engine. have uncovered a new struc­ visual part of the spectrum. is very unlikely that there "The significance of rh is ture in rhis nearby active The space telescope does not could be duSt 100 lighe-years research is that it finds a galaxy. T he galaxy, found in currently have an infrared from the cen tral bl ack hole brand-new fea ture in this the constellation Cecus at a camera, but is sc heduled to that would be hot enough to galaxy," Weinberger says . distance of about 50 m illion receive one as E&S goes to produce rhe observed emis­ "And even more, this new lighe-years, reveals a a bright press. The elongated feature sion. Rather, Weinberger feature may provide observa­ active nucleus at infrared discovered by ehe Caleech says, it is likely that the ob­ tional evidence fo r a theoreti­ wavelengths. This nucleus group has not been see n in served extended near- infrared ca ll y predicted means of has long been thought co Hubble's optical images. light is from stars. Further­ channeling material to rhe harbor a black hole as irs cen­ T here are twO very inter­ more, since it points in a black hole on very small tral engine and, because it is esting aspects of this image. different direc tion, this newly scales. " The image is by no brighe and n·earby, has been First, the image does not resolved infrared emission is means detailed enough to intensely studied by as rro­ show a circular d istribution of likely to come from a sOllfce show the in-fall of the matter physicists. light, but rather an elongated entirely different from that of itself, Weinberger stresses. Made at a wavelength of one, and second, the axis of the visual emission. For this, one would need a 2.2 m icrons, Weinberger's the emission points in a It has long been proposed resolution of less than a light­ near-infrared image has the d irection different from that that elongated, noncircular year, and there is currently capabil ity to reveal structures of previously observed visual distributi ons of Stars in the no way to make such finely

• EHGI N EERING & HIENCE N O . 1997 detai led pi ctures. Nonetheless, the quali ty A BRIEF HISTORY OF NAKEDNESS of this image is unparalleled because it reli es on the unique resolving power of Caltech's la-meter Keck It's not every day that a Telescope and the techn ique theoretical phys icist pays off of speckle interferometry to a bet on nakedness, bur remove the distorting effects Stephen Hawking did so of Earth's atmosphere. With February 5. Hawking, the this technique, a series of very Lucasian Professo r of Math­ rapid exposures are made of ematics at Cambridge Uni­ the object, freezing the atmo­ versity, had been at Cal tech sp heric distortions that cause for the previous six weeks as stars to twinkle. Then the a Sherman Fairchild Distin­ distortions are removed in guished Scholar. He con ­ computer post-processing. ceded, juSt before leaving As the largest infrared tele­ for home in England, a 1991 scope in the world, the Keck bet he had made with Kip Telescope provid es the bes t Thorne and J ohn Preskill that obtainable resolution. a phenomenon kn ow n as the Weinberger is currently naked singularity is possible. completing work on her Thome and Preskill, both doctorate. She wi ll cominue professo rs of theo reti cal phys­ doi ng observations to suppOrt ics at Caltech, think that this resea rch, a part of her naked singularities are al ­ thesis. "It will be exciting lowed by nature. Hawking to look at NGC 1068 with does not, bur conceded the similar resolution in other bet "on a technicality," he From left, John Preskill and Kip Thorne assist Stephen Hawldng in sealing infrared wave lengths," she said. with hi s thumbprint a new bet on whether general conditions for the says. "The more information In accepting Hawking's we have across the spectrum payoff, Preski ll sa id that creation of a naked singularity will be found . the more we 'll understand "We're much more tolerant about the nature of this of nakedness" than the British extended emission." physicist. their nakedness from the singulari ties could exist Also collaborating in rhis "It comes from li ving in vulgar view. Specifically, the outside of black ho les and research are her thes is super­ Southern California," Thorne goods consisted of two T­ therefore be observed. viso r, Gerry Neugebauer. and added. sh irts, which the bettors Hawki ng has always Keith Matthews, both of rhe H awking paid off rhe bet would only say were inscribed rejected the idea of the naked Caltech physics department. by presenting his tWO with "an appropriate mes­ singularity, but admits that D - RT American colleagues with sage" ftom Hawki ng. Choptuik's computer simula­ adequate raiments to shield The big bteakthrough tion shows how one could came when another physicist conceivably exist. There's named Matthew Choptuik virtually no possibility that (now at the University of Choptuik's naked singularity POLITICAL SC I ENTISTS RANKED SIXTH Texas at Austi n) developed a wou.ld ever arise in a real supercomputer simulation universe, however. Even with onl y nine faculty members, Caltech's political showing how a naked singu­ "Basically, it could exist science group ranks sixth (after adjusting for size) in the la rity could occur. To back only in a computer," Preskill nation, according to a report iss ued in December by Political up a bit, a singularity is a said. "Bur it's the sort of Science and Politics. The top 10 with numbers of faculty are: place where matter or light event that would be allowed 1. Rochester Institute of Technology (18) crashes in by force of its own to happen, and that's what 2. Stanford (28) weight to form a region of the bet was all about." 3. Harvatd (48) exceedingly high density. For his part, H awking said 4. Yal e (29) No one disputes that that he's still a betting man 5. University of Michigan (44) si ngularities can exist, but when it comes to theoretical 6. Caltech (9) Hawking believes that a physics, eve n though he is 7. UC Berkeley (41) singularity can occur only now 0-2. In fact, a new bet 8. University of Iowa (22) inside a black hole, where it is al ready on (see above). 9. University of Indiana (27) cannot be seen. According to ''I'm going to win th.is 10. University of Minnesota (30) Thorne and Preskill, there time, but I don't know should be situations in which when," he sa id. Il-RT

! 9 9 7 EN GINEERI N G & HIE N CE NO. • " It is noc a case of choosing [bose [faces] whi ch , to {be best of one's judgment, are rea ll y the prettiest, nor

even chose wh ich ave rage opinion genu inely thinks the preuiest. We have reached rhe third degree where

we devote our intellig en ces to anticipating what average opinion expects the ave rage opinion to be. And

t here are some, I believe, who practise the fourth , fifth and higher degrees."

10 ENGINHRING & S{IENCE NO . 1997 Taxi Drivers and Beauty Contests by Colin F. Camerer

r spent a year in New York City not long ago, to have been familiar with it as well, as the open­ and I took a lot of cabs. Most cabdrivers in New ing couplet of his ballad "Taxi" bears witness: "It York are independent contractors. They rent the was raining hard on a Sa turday / I needed one cab from a taxi company for $76-paid in more fare to make my night. ") Many of the driv­ advance-for 12 hours. They keep all rhe fares ers we talked to said they decided how long to The trading noor of the they collect, and they can call it quits and retmll work by setting themselves an income target every New York Stock Exchange. the cab at any time before rhe 12 hours are up. day-for example, they might want to earn $150 Because Manhattan is so crowded, drivers usually in cash in order to clear $75 beyond the rental The British economist John just cruise the streets waiting for someone to hail fee-and when they reach that target, they guit. Maynard Keynes likened them. Some days are especially good-when it Target setting can be very motivating in unpleas­ playing the market to rains or snows, during the holidays, or when a ant or tedious activities, like exercise. There's aiso voting for the prettiest convention is in town, for example. Other days substantial psychological evidence that people d is­ are bad-weekends, when fewer businesspeople like losing a lor more than they like corresponding face in a beauty contest; are around; and the summer is slow because people amounts of winning. This implies that drivers hence the second part of leave Manhattan to escape the heat, humidity, and hate to guit before they reach the target, but once this article's title. gunfire. So Linda Babcock (whose father, Charles they reach it, they aren't very enthusiastic about Babcock [MS '58, PhD '62), was a professor of trying to go beyond. So income targeting per­ aeronautics and applied mechanics at Cal tech until versely prediCts that cabbies are going to quit his untimely death in 1987) and George Loewen­ earlier on good days. If YOll want to make $150 stein of Carnegie Mellon University, Richard and you're earning $25 an hour (which would be Thaler of the University of Chicago, and I became a pretty good day for these guys), you can go home curious about a simple question-how does the after six hours. But on a bad day, when you're amount of hours a cabbie works vary wirh that earning $ 15 an hour, you've got to drive ten hours. day's average hourly earnings? There are two basic The labor-supply curve will be a hyperbola, which theories that might apply. One is called the law is also shown on the following page. of supply. The other, which we crafted from bits These two theories thus give very different and pieces of psychology, we call "daily income predictions, which we tested in our study. We targetmg. analyzed 3,000 observations of cabdrivers' behav­ The law of supply is the twin sister of the law ior from [he years 1988, 1990, and 1994. The of demand-people should want to sell more of data came from the New York City Taxi and something when the price is high than when the Limousine Commission, ironically known as the price is low, assuming everything else is constant. TIC, which had collected it for other studies. So you'll sell more of your labor hours when wages These data were in the form of taximeter readings, are high, and the so-called labor-supply curve and the TLC was kind enough to give them to us slopes upward, as shown on the page after this (for free!) on floppy disks- bureaucracy has ItS one. The law of supply says that YOll should work moments. When you get into a cab, the driver a lot when it pays to do so, and when it doesn't punches a button and a meter autOmatically pay, go home! Take time off. records the number of miles driven and the The other theory, daily income targeting, was amount of time spent sitting in traffic. From the taught to us by the cabdrivers, many of whom meter records we could compute a driver's earn­ are amateur philosophers, political scientists, and ings, except for tips. Tips aren't recorded any­ labor economists. (The late Harry Chapin appears where, so we left them out of our analysis, but

19 9 7 ENGINEERING & SCIENCE NO. II on average they're probably 10-15 percent of the • Above: The labor-supply d ri ver's income and do not vary much from day to cu rve . Wages are plotted day. The meter data should represent the bulk of the driver's income accurately. on the vertical axis; hours A scatter plot of some of our data is shown at IF \'OtIIlAVE All ACClDEIiT tAU. FROII lCllIE m·. ·:wIII worked on the hori lontal the tOp of the opposite page. I should warn the one. The law of supply faint of heart that it looks very messy, but as eco­ nomic data go-particularly in areas like labor =d:H':4::'I"~~I~'Eii';· I says that when the hou rly economics, where there are lots of external factors "... . " ;'- fl1 "~,,, Fh> 1<...& =-- I wage goes up, people will that influence the data-this is actually a pretty work longer hours (left). strong correlation. I hope you can see that the line " , The income-targeting of best fi t through this cloud of data slopes down­ , s ward. In fact, the slope is significantly negative >-u theory predicts the to a confidence level of more than 99.9 percent. , ~ , opposite: people will work 0 ~ ,"...... -- - So the data clearly suppOrt the targeting theory " ..J,L _ , 1 'u ." . , ",' r<- less as their hourly wage rather than the law of supply. ..? ~ -.... - I There's an objection that can be raised here­ U rises (right). If hours and .... ", <> in order to follow the law of supply, you've got , ~ .,- wages we re plotted on t>- , , to have a certain level of economic seeuri ty. These ~ logarithmi c scales instead guys may have to keep driving until they make ~ J

~ I"• of linear ones, t his $150 because [hey need [he cash-they don't have ,~ enough savings to buy groceries and pay the rent " hyperbola would plot as a " A • jf they quit early on slow days. The reason we • o,u downward-sloping line. don't think this explains our findings is because some drivers in our samples own their own cabs. In order to legally operate a cab in New York City, you have to own a taxi medallion-an ugly, plastic-metal thing that's pasted on the hood of the cab. These medallions are restricted in supply (there are only 11,387 of them, and [hat number has been fixed foe 60 years), so they're quite Above: Some of the meter valuable. They' re worrh about $150,000, yee 10 percent of the drivers in our samples own one data were verified by personally. If we assume that the drivers who can examining "trip sheets" afford to own a medallion have some cash in the such as this one. in which bank, we might predict that they would behave drivers log (from left) differently than the renters. Bur borh groups seem to behave about the same way. pickup point, pickup time, Another important considerarion is that cab­ destination, dropoff time, drivers vary in experience. Happily for us, New number of passengers, York City cabdriving licenses are numbered chro­ nologically by date of first issue, so the person and fare.

12 ENGINEERING & S(IE NCE NO. 1997 Top: If you look at a log­ arithmic plot of the labor­ supply curve (or a sample of taxi drivers, you (an see that the line of best fit slopes downward, contrary to the law of supply. Bottom: But if you analyze o 0 Cl., o ~ 'l50 the sample's inexperienced o o o ~o (upper) and experienced \', 0 0" (lower) drivers separately, 00 0 0 o you will find that the o 00 cf experienced drivers' curve o o g is more nearly horizontal. o o o o with li cense number 14,682 gO[ it just after the person with number 14 ,681. Therefore, we can sort drivers inca high- and low-experience groups by their li ce nse numbers. Look at the difference 1.5 t,---- -,------,-- in their labor-supply curves, as shown at left. 2.5 Again, the data are noisy, but the low-experience drivers on the left have a slope very close to - 1, which is what the income-targeting theory pre­ dicts. (A slope of - 1 means that if your wage goes up by 10 percent, you Cut your hours back by 10 2.' percent to keep you r income constant.) The high­ expetience drivers on the right still don 't look 0 0 much like they're obeyi ng the law of supply, but 0<8 0 it does appear as if experience is teaching them 0 0 to make hay while the sun shines- to drive longer 0 0 o'b hours on good days. 2.2 This distinction between new and old drivers 0 0 is important because about half of the cabbies in New York have been drivi ng cabs for less than a year. In 1991, over 40 percent of all New York cabdrivers were born on the Indian subco ntinent, o 11 percent were from Africa, and another seven ... :,.------::c,------:o:- percent each were from the Caribbean, the Middle 2<1 2.8 ) .2 East, and the former Soviet Union. Only about 10 percent were born in the United States. The point 2.' is that driving a cab is an entry-level job for many immigrants, so there's a constant inflow and out­ fl ow of new drivers. These inexperienced drivers 0 may be using the income- targeting rule because 0 0 0 they haven't yet figured out that they can do better 0 CD &80 0 BO by obeyi ng the law of supply. 2.2 We learned twO bas ic lessons from this study. 0 0 0 '" 0 0:> The first is that cabbies would get an automatic 0 raise of 8 percent if they drove the same number 0 0 of hours every day, rather than knocking off early 0 on the good days and wo rking late on the bad 0 0 days. If rhey obeyed the law of supply, they could ... 0 earn 15 percent more inco me. The median annual ,. 2.' 1.2 wage of these dri vers in 1995 was about $22,000 a year, so they co uld have made about $2,000 more per year by simply changing their driving habits. The second, more important lesso n is that

1997 ENGINEERING & SC I ENCE NO. " Mm tb.. b Ioo."r ...... rt "'_ )1 .. It ...... ' ,.,n. no- II)' AlI t.ilnb ... ~ ... I .... ,.boot. .,.... "''--,...... No­ .. ,.. I by_ ~.-...d rI.oI.... Sot. C••• i.p . h ue ...... t ..I ~ .... IW'_ 0.-. "- r ...... h:l. t.. NtIOIT .....1 t.I,. pl_. Ito! d Willi•• \\llith sum ...... S. ~ .. ,., dIo,.ooiaa. _ ...".,...... C'.-.,r J-" s...... ),f.. ~~,...tdM .... yM-.- ...... will \llU elect ..... -"lI , .. ___ cot ,..... , _ ---.. tJw. .... 'If,...... " .Id. ,lIl_ 1M ",","01. """'dI Wpdrrdll .. JUo.a...... dod.... "-b. ..,.,w lUl. 'lis.'; Hheingllld " ... ~ .... _.w...... ~ 0.. '-' )1 _ M...... I n.pt-- ...... il ...... ,-t.o.U·•• lw.y...... ~ the tax rates on cabdrivers to give chern a higher _ Inon ....,... tJoOAOG, ~ .....u1Ml... \ ... ~ . JIfIfflII I ,,# ,_ tn, .. to IW,...,...... r. .. 1957? It ~ lUI .. . ..,. .... -'e;., after-tax wage, it looks as if they would drive fewer """ ...... a ... f __ I'.... ., Uw~"". I a..L.tI hours, not more. ..u.rn.. .. -.t ,w. n...-...w ...... PkIc the II,t who'll win Let me move on to beauty contests. Here n...... "'--' • contrad worth $50,0001 \0.,.. ..,.. ..owiW,..... I don't mean the Miss America pageant or the Vote at on, ~ngoW .." .J ...... ),f.. MIoria- tryoutS for Rose Parade queen, but you'll see in a .... F.JrortiIIII ...... I_' ttor. or tav.m I R ~ Mft .. ta_ A'" moment where the term comes from. Imagine the no4 ,.._ ftltr- I .....' _ HJ' lAy M_...... /w_ .... ' ,' ..... following game: Everybody picks a number from ~s....a-_ ... -,----~ --.... o to 100. I compute rhe average of all your picks, and whoever's number is closest to two-thirds of that average wins. (We actually do this in experi~ ments on students. The winner gets $20, so they think carefully before they choose.) Everyone wants to be at two-thirds of the average, but everyone else does, toO, so the real goal of the contest is to guess what everyone else will guess. This is like playing the stock market. The economist John Maynard Keynes remarked in the 1930s that the stock market is like a beamy contest. He had in mind contests that were popu­ lar in England at the time, where a newspaper Phm.. roo.","y of ,Ilt John W. Han man urn.,.' /"or Salt'S. Ad...,,,is;ns aoo Marh.;ng History. [)uk.,. University would prim 100 photographs, and people would write in and say which six faces they liked most. perhaps we should be skeptical about simple Everyone who picked the most popular face was The "M iss Rheingold" economic principles like the law of supply. Most automatically enrered in a came, where they could previous studies were inconclusive about wherher win a prize. Keynes wrote, "Ie is not a case of campaign, run by the J. the supply curve eve n went up or down, because choosing those [faces] which, to the best of one's Walter Thompson (0. for most people's salaries change relatively rarely­ judgment, are really the prettiest, nor even those Liebmann Breweries, Inc. once a year, perhaps. Bur cabdrivers earn a differ­ which average opinion genuinely thinks the pret­ for over 25 years, is the ent hourly wage every day, and they can adjust the tiest. We have reached the third degree where numbers of hours they drive, so there's enough we devote our intelligences to anricipating what best~known American variation in the data to see trends. That's why our average opinion expects the average opinion to be. example of a Keynesian scudy shows more clearly than ever before that for And there are some, I believe, who practise the beauty contest. At the taxi drivers, the labor-supply curve slopes down, fourth, fifth and higher degrees." not up. During the Reagan years, supply-side If you played this game repeatedly, your height of it.s popularity, economists argued that if income taxes were cut, thoughts might run as follows. You'd assume between 15 and 20 million the after-tax wage would rise. That's just simple that the starring average would probably be 50, votes were cast per year­ arithmetic. And then, the argument went, people so you'd guess 33. But then you'd say, hmmmm, a turnout second only to could earn more spending money by working an if other people are as clever as I am, they will all extra hour, so people u)(}/tld work extra hours, and pick 33, so I should pick 22. Bm if everyone else the Presidential elections. everyone would be bener off. Very logical. Our does that, coo, I should pick two-thirds of 22. results suggest the opposi te-if you were to lower And if you carry this through infinitely many

14 ENGINEERING & HIENCE NO. 199 7 How real people behave in a one-round beauty contest. The two graphs show the same four sets of data, but in two different fro nt-to-hack orderings to mini mize t he number of short bars t hat are obscured by tall er bars in front. (The colors, however, don't travel wit h the bars.)

levels of reasoning co the logical end, you'll wind up picking zero. If 1 were speaking to a game­ theory audience, people would nod profoundly, because zero is what game cheory predicts for chis situation. Game cheory is the branch of social science char analyzes strategic interactions in mathematical terms. It was founded quite a long time ago, but it's had a slow fuse-only in the las t 10 or 15 years has it come to the fore in reasoning about economics and political science. (In fact, people here at Cal tech helped establish the use of game theory in political science, and still do quite a lot of it.) So how do people accually behave? Do they pick zero? The data at left are from experiments ... ," on undergrads from Singapore, Germany, the Whanon School of Business at the University of Pennsylvania, and Caltech. The German dam were collected by Rosematie Nagel of the Univer­ sity ofPompeu Fabra. The Singaporean data were collected by Teck-Hua Ho and Keith Weigelt; they also collaborated with me on the W hanon data. (Ho and Weigelt, who are now on the fac ul­ ty at Whanon, were both students of mine when I was there.) The average pick across all these experiments was around 40, so if you guessed j about two-thirds of 40, or 27, you'd probably win. Notice that 40 is somewhat less than 50, so if we use these data to gauge how many steps of reason­ ing people are doing about other people's·reason­ ing, some number from one to three seems reason­ able. It's clearly not the game-theory prediction of infinity. but it also clearly demonstrates the performance of at least one step of reasoning. We're now trying to refine this estimate of how many steps of reasoning seem natural, and how it varies with education and other factors. For exam­ ple, no Caltech student chose above 40. Most Techers picked numbers between 30 and 40. Several picked in the neighbothood of lOot 20, ... ," and 10 percent of them did, in fact, actually pick zero. The Cal tech students and the German stu-

1997 ENGINtEAING & SCIENCE NO. IS Four more sets of data, again presented in two different orders from front to back. The University of Chicago PhD data is cour­ tesy of Richard Thaler.

dencs appear to have been reasoning one or two steps more deep! y than the Wharton scudenrs and [he students in Singapore. We've also conduned this experiment, more informally, with other groups of subjects. (Repli­ cation with different groups of people is, of course, esse ntial if we want [Q generali ze our findings to all human beings.) The plot at left shows four more groups. The front tWO groups in the top figure are PhD students in economics (none from Cal tech), who may have had some exposure to game theory. And, in face, compared to the under­ grads in the previous plot (except for the Techers), these PhD students do choose lower numbers. The average pick here is around 25-one step beyond the undergrads. The additional education is doing something. The gtoup labeled "Caltech Board" is from an experiment I conducted when I gave a talk at a meeting of Cal tech's Board of Trustees in the fall of 1995. Thete were about 80 or 90 people there, including spouses and some people from the faculty and administration, and I just couldn't res ist the opportunity to see how they would behave. The Caltech Board is a truly amazing group that includes many extremely successful business men, some billionaires, several brilliant scientists, and twO former judges. Notice that they act pretty much like the college stu­ dents-the average pick is about 40. Bur a few people do choose very low numbers, like zeta. And several people, who may have been confused because I didn't explain the procedure as carefully and thoroughly as I would have in a real experi­ ment, picked very high numbers. This was not a well-run experiment, but the subject pool is so unusual that I'll show it nonetheless. The sample labeled "CEOs" is teally temarkable. We' ve seen that college students do not obey game theory, which ass umes that people are perfectly rational. (This is hardly surprising co anybody with teenagers in college.) So it's easy to criticize our expetiments by saying that what really matters

16 ENG I NEERING & SCI ENCE NO. 1997 80.000 I 10000 This strategy is sometimes called the "greater-fool theory," because even

though you're a fool to pay as much as you did, you ' re betting that there's a

greater fool just down the road. And if you're right, then of course you aren't 30.000 20.000 being foolish. 16000 '

0.000

Cbole ..

is not what a bunch of college kids do, bur wheth­ sometimes called the "greater-fool theory," because er rhe people who run large businesses behave even though you're a fool to pay as much as you according [0 economic theory. Well, the Calrech did, you're betting that there's a greater fool just Board includes 20 chief executive officers, presi­ down the road. And if you're right, then of course dents, and corporate-board chairmen. These titans you aren't being foolish. of industry are the "CEOs" sample. As you can Economists call this a bubble. Prices rise see, none of them picked zero; and if anyone of simply because people expect them ro rise, and it's When people play the them had, that person would have lost. So they a se lf-fulfilling prophecy right up ro the moment beauty-con test game for obviously knew who they were playing with. A when the bubble bursts. One fa mous example is several rounds against the few of them picked surprisingly high numbers, that of tulip bulbs in H olland during the 1600s. same group of opponents, but the tallest spike is between 30 and 39, and People were paying several months' income for there's another tall spike between 20 and 29. rare tulip bulbs. Thoroughbred horses in the the behavior quickly If you do the math, it turns out that they were 1970s, and L.A. real estate in the 1980s are other converges to what game reasoning about one step further than the other examples, as are booms in works by dead artists theory predicts will people at the meeting. The numbers they chose (who can't produce any mote supply). The Japan­ are statistically indistinguishable from the num­ ese economy in the 1980s might be the most happen. bers the Caitech undergraduates and the econ PhD spectacular example in world history. However, students chose. a business-school professor who teaches about the The game-theory prediction was flat-our wrong. srock market would probably be reluctant to The same parrern emerged across three continents, admit that these episodes are bubbles, in the both genders, and a tremendous variation in age, sense that I've defined the term. I'm asserting wealth, and educational background. that people are consciously paying more than Bur what happens if we allow people co learn the intrinsic value of the asset, but the professor by announcing the winning number and repeating would probably say chat we don't know its intrin­ the game? Then we see a steady, slow convergence sic value. How do you measure the intrinsic value roward the game-theory prediction. The graph of, say, Van Gogh 's Sun/loUlers? Maybe it was a above shows what happened when [he Singaporean bargain at $50 million. Inscead, most of the students played a multi-round version of the exper ts believe in the so-call ed efficient-market game. After 10 rounds, abour 50 or 60 percent of theory, which says that information about a srock's the students were choosing numbers between zero worth will quickly be refleered in its price. and 10. So game theory, which seemed so laugh­ It would be nice if we had an example to con­ able at first, does predier what people will do wi th vince the experts who believe thac markets are repetition. Again, psychology helps us understand efficient. Until a couple of decades ago, people what happens at first, and game theory tells us thought that economics, like astronomy, was not what will happen eventuall y as people learn. an experimental science-all you could do was We need both ro understand the entire picture. study the data that the market provided. Bur in This brings me to the srock marker. That fact, many of the most interesting propositions passage from Keynes describes a market in which in economics can be tested experimentally. About investors care about what other investors will buy 10 years ago, Charles Platt, the H arkness Profes­ in the future. Here, you often pay more than a sor of Economics and Political Science, founded firm is worth, because you think that somebody the Experimental Economics and Political Science else will pay even more later on. This strategy is Laboratory at Caltech. The whole thing is run

1997 ENGINEERING & SCIEN[E NO. 17 $ 5.00 In this plot of an experimental marl(et, the horizontal axis is time and $4.00 the vertical axis is the

pl'ice per share. Every dot $ 3.00 is a proffered transaction; the actual transactions are $2.00 connected by the red line. The vertical green lines denote the end of each $1.00 five-minute period, at wh ich point dividends $0.00 are paid.

by computer and functions very much like a real below the line are buyers offering tOO little. marker. (You ca n also scudy elections and other Notice that the price remains flat at around processes in ic.) Each participant is isolated in a $3.50--even close to the end, where the efficient­ bomh, and cannot communicare with other par­ market theory says the shares are worth less than ticipants in any way except through the compucer. a buck. (This is like those of you who bought a People type in offers to sell x number of shares at house in L.A. a few years ago, and refused to sell such-and-such a price, or bids CO buy. an d all rhe as the market collapsed. ) The traders are trying to offers and bids are displayed on everyone's com­ forecast whether rhe market will crash, or whether puter screens. Players co nsummate trades with some nut will buy shares that are about to expire. the push of a button. The computer records all And finally, of course, the market collapses. offers, bids, and transactions sequentially; keeps We know that everyone knew a share's intrinsic track of who owns what; and calcul ates everyone's value because we gave them a quiz before the first earnings. (Again, the students get paid real mon­ trading period began, so this is the clearest exam­ ey, so we can be sure that they're taking this seri­ ple of a bubble that you could possibly have. ously and are giving it their besr effore.) Every­ When we asked the subjects how it came about, thing is recorded as jt happens, and software de­ they'd tell us a story that sounded very much like veloped by Plott enables us to make a "movie" of the greater-fool theory. They'd say, sure I knew how the market behaves, and analyze it in detail. the prices were way roo high, bur I saw other In these experiments, we created a market for people buying and selling at high prices. I figured an asset we invented whose value we chose. The I could buy, collect a dividend or two, and then students traded a share- a bond, if you will-for sell at the same price ro some other idiot. And, 15 five-minute periods. Each share paid a divi­ of course, some of them were right. As long as dend of 24 cents at the end of each period, so if they got our before the crash, they earned a lot of you held on to a share for all 15 periods, you'd money at the expense of the poor folks who were earn $3 .60. Everyone had a couple of shares to left holding the bag. stare with, and some money to buy more shares We can see harbingers of the crash in what we've if they wanted to. The ques tion we wanted to come to call nervousness in the market. Near the answer was, what would the price of the shares end, some people who chink that the market has be? The efficient-market theory is very clear on lost its mind will make extremely low bids. These this. It says that since everyo ne knew the share people probably know that a lowball bid of a paid a total of $3.60 in dividends (we told them dollar won't be accepted when the going rate is thar, by the way-we gave them a table of divi­ three times that, so we think this is their way of dends versus periods remaining), then the price expressing their surprise and warning everybody. of the share should be $3.60 in the firs t period. It's the same as when so mebody offers you In the second period, the price should d rop by $350,000 for the house you're desperately tryi ng 24 cents to $3.36, and so on. to sell for the half million you paid for it a few Shown above is what real traders did in a typical years ago. This is their way of politely saying experiment. The slanting purple line shows the you're nuts- your house isn't worth half a million. shares' declining dividend value. Each dot is an After doing a number of such experiments, atrempt to sell or buy; all the completed transac­ we've learned how to turn these bubbles on and tions are connected by the red line. Dots above off. To turn the bubble off, we bring the same the red line are sellers asking tOO much, and dots group of subjects back and run the entire 15-

18 EN GI NEER I NG & SCIENC E NO. 1997 rheir colleagues, what are you guys buying? They want to buy what the other guys buy. so they don't finish last. That, again, is very much like a Keynesian beauty contest, and I think the prevail­ ing theories need to address it. Peter Bossaercs, an assoc iate professor of finance here at Cal tech, is actually working on this now. I should also point The traders are crying to forecast whether the market will crash, o r whether out that nothing I've sa id addresses the issue of stocks that haven't paid dividends yet, but may some nut will buy shares that are abour to expire. at some time in the fueure. This is a very common situation with growth stocks, such as those of startup companies in biotechnology, software, and other high-tech fi elds. The closest we've come to studying those was a couple of experiments where the dividend wasn't guaranteed-there was a large chance you'd get nothing, and a small chance you'd win big. We did see some things that looked like bubbles, but we haven't done much work in that atea ye t. In conclusion, ca bdrivers, beauty-contest games, period market again. We usually see a smaller and srock-market experiments have a common rise rhat crashes much earlier. And if we bring theme. Inexperi enced cabdrivers, novice beaury­ that same group back a third time, we hardly contest players, and traders participating in an get any bubble at all. The marker-price line experimental stock market for the first time don't now follows the intrinsic-value line very closely, seem to conform ro standard economic theory, so experienced traders do obey the effi cient­ which assumes complete rationality by all parcici­ markets theory. We ca n turn a bubble on by pants. However. their actions are reasonably well having had our subjecrs parricipate in a previous explained by psychological theories that allow experiment in which we created inflation by people to have normal, limited reasoning ability, adding money [0 the economy, just the way and limited faith in others. The subjects of these the government does. If prices rose in that earlier experiments aren't dumb, but they're not perfectly market-if they've lived through an inflationary brilliant, either, and they're nOt willing to bet a experience- then we've planted a belief in their lot of money that other people are. But the minds that prices will rise, like seeding clouds behavior of experienced drivers, players who play [0 make rain. Then, when we put them in the the beauty comest over and over again, and readers bubble experiment, prices do rise, because of this who rerurn to the srock market is often explained self-fulfilling prophecy based on their common quite well by economic theori es. Experimental experience. We don't always see bubbles­ observations help us figure Out which theories are sometimes we see juSt what the efficient-market true, and which are false, and under what cond.i ­ theory predicts, with prices sliding down along rions. So we think thar combining the best ideas the intrinsic-value line. But bubbles are very in psychology and economics will make for the common-the several of us doing this kind of best social science of all. 0 research have observed about a hundred/of them. This research is very new, and there are many things we have yet [0 learn. We need help from cognitive psychology to understand what the Colin F. Call1ere1i the Axline Professor of Business people in our experiments are thinking. We need Economics, sil(tiies cOlporate strategy, decision sciences, better pattern-recognition and data-analysis [Ools and experimental ecol1omil·s. Camerer earned his BA in [0 help us look at the data and forecast when quantitative studies from J ohns Hopkins University in bubbles will starr and crashes occur. Compared to 1976. He got all MBA from the Univmity..jChicago other experimental sciences like physics, chemis­ in 1979, followed by a PhD ill behavioral decision try, and biology, the amount of work that's been theory in 1981. He arrived at Caltech as the Axline done in experimental economics is relativel y Professor in 1994. He is now taking advantage of modest. Caltech's proximity to Santa Anita to research an What does all rhis mean in the teal world? upcoming article to be titled IfAmbigltity in Betting on Perhaps one-third of the market's trading volume Umoaced Thoroughbred Hones, n but his interest is purely is due ro a handful of mutual funds and other large academic- he does all his retirement investing through institutions. These portfolio managers may not TlAA-CREF behave rationally, either, aJrhough for other The bea1lty-colltest and b1lbble work was s1lpported by reasons. For example, they operate in a world the National Science FOlmdation. The taxi-driver study where if rhey have one bad quarter-worse than was s1Ipported by the Rmsell Sage Fonndation. This everyone else-they may get fired. So they ask article is adapted from a \Vats011 lecture.

1997 eNG I NEHIN G & S(I ENCE NO. I.

"We'll get the lead out of the iron meteorite. You measure its isotopic

composition and you stick it into the equation. And you'll be famous, because

you will have measured the age of the E arth." - H arrison B ,'own

I said, "Good, I will do that." And he said, "It will be duck soup, Patterson."

Duck Soup and Lead

Oral History - Clair C. Patterson

Clair C "Pat" Patterson, who fint measured the age Clair Patterson: I got a master's degree in nine of the Earth (4.6 billion years) and whose work on the months. Then there was a chemistry professor ubiquitous high levels 0/ lead played a significant role in at the Univetsity of Iowa who said, "Patterson, the passage of the Clean Air Act of 1970 and in the you've got to go to the University of Chicago removal of lead from a number of suhstances including and work on the atomic bomb." gasoline, died December 5, 1995. He had come to Caltech as a research feliow in 1952, Shirley Cohen: Bur he wouldn't have known eventuaJJy becoming a senior research associate, and about the atomic bomb then? /ina"y accepting a professorship (he did,/t believe tenure should be a scientist's goal, just one of his many icono­ CP: Yes, because that's where he went. clastic views) only in 1989, three years before he retired, He was awarded the $750,000 Tyler Prize for Envi­ SC: I thought it was such a big secret. ronmental Achievement in 1995. The foliowing article is adapted from interviews con­ CP: Oh, it was only a secret to chimpanzees who ducted in March 1995 by Shirley Cohen for the Caltech didn't know what they were doing. H e was going Archives Oral History Project. It has been edited and to the University of Chicago to work on the condensed, and some portions, for want of space, have atomic bomb, and he wanted to take me along. I had to be omitted completely. Patterson begins his story bad become a spectroscopist. I had done research with his childhood in Mitchellville, Iowa, in the farm ­ in molecular spectra at the University of Iowa for land outside Des Moines, where he was horn in 1922. my nine-months, whiz-bang, master's degree. His father was a rural mail carrierj his mothel~ who While I was there I got into atomic spectra a little was a member of the school hoard; gave him his first bit. So now, at the University of Chicago I was chemistry set. He earned his AB in chemistry in 1943 doing atomic-emission spectroscopy. They were at Grinnell CoJJege, where he also met his wife Lorna analyzing the various products of the uranium (Laurie), A year later he got his MS, also in chemistry, when it disintegrated. from the University of Iowa, which is where we join the story. (Here Laurie Patterson fills in some gaps: "Pat and I left Iowa City to work 011 the Manhattan Project in the spring of 1944, soon after we were married, \Ve lived in Left: Clair Patterson was photographed at Caltech's 1993 an apartment hotel an'OSJ tbe street from tbe Museum of Commencement by Yigal Erel, his last graduate student, Science and Industry in Chicago. Pat became more and who was awarded his PhD at that ceremony. more unhappy at bejng in the city, doing work we both felt would let the 'genie out of the bottle' much too soon. Right: Laurie and Clair Patterson at their own graduation We went to Iowa for a weekend late in the summer of from Grinnell College in 1943. 1944 in order for Pat to enlist in them'IIIY. (He had bee1l1'ejected because of nearsightedness during our senior year in coiiege, whm aii but 19 men of our graduating class left Grinnell to go into the services.) The physical requirements had been lowered, and he felt he would be accepted at this time. He set this in motion with his dmft board; I enlisted in the \Vaves. Three days later his draft board reported that they could not draft him

1997 ENGINEERING & S(IEN(E NO. 11 becClme of his high seCltrity ,'ating and he 1Itltst return to the University of Chicago. FortimateLYJ I had not tllmed in the finaL papers and was 110t yet formaLLy a \Vave. \Ve returned and were asked to meet with the coLoneL in charge of the Manhattan Project at 5th Army Headquarters. He suggested that he send liS to Oak Ridge, where there were many YONng peopLe. Pat jeLt he was the onLy yotlng male on the streets of Chicago and The Pattersons enjoyed that he was a 'draft dodge/: "'} escaping to the Cumberlands from nearby So we wem down to Oak Ridge, and that's where we spent another year and a half or tWO Oak Ridge, Tennessee years, working at the uranium 235 elecrromag­ ( 1945). netic separation plane. At Oak Ridge I got into mass specrrometers. You see, the isocope of uranium that they wamed was uranium 235, which is what you made the nuclear bomb out of. But 99.9 percent of the original uranium was uranium 238, and you couldn't make a bomb Out of that. But the little tiny bit that was V H5 had a different mass, and you could separate them using insight that enabled him to develop the concept a mass spectrometer. that led to what we call paleotemperatures-the measurement of temperatures 200 million years sc: Is this where YOll met Harrison Brown? ago. This is Sam Epstein's stuff. {Epstein is noUJ the IVilham E. Leonhard Profmor of Geology, EmeritllJ.} CP; No, he was there, bm I didn't meet him until These guys during the war developed these after the war. H arold Vrey wasn'r there but he concepts, you see, and they kept them on the shelf. had done the theoretical work at Columbia that They knew that they were working as engineers on the Oak Ridge diffusion plant was based on. a hideous weapon of warfare. They were the same (There were both electromagnetic se paration and types as my mentor at the University of Iowa who diffusion separation at Oak Ridge.) That's where told me, "Patterson, we are saving democracy for' Ueey got his ideas about isocopic fractionation the world against fascism." These professor­ being a funcrion of temperatures. This was the mentors, who were no longer at the university but working on the bomb projeCt, told you ng people like me that this was the thing to do-this hide­ The University of Chicago ous crime we were committing was a necessary wasn't all work and no thing. play. Here Patterson sails I didn't think this way then, only later. But on l ake Michigan during that time I learned a lot of new ideas and concepts and patterns of thinking. So when the in 1947. war ended, I said, "I want to go back to the uni­ versity. I love the University of Chicago. I'm going to go there and I'm going to get my PhD in sc ience and study some of this important stuff." Harold Urey, H arrison Brown, and all these guys flocked back to the University of Chicago. And all these ideas that had been cooking around in their minds during the war then ca me to fruition as goals. This had nothing to do wi th making the atomic bomb. These were scientific concepts that dealt with acornic physics and chemistry. So when I went to the University of Chicago, Brown found out about me (BrOW1J became PattersonJs research adviser) and said , "Hey Pat, you're familiar with mass spectrometers. Now, there's this other youngster, George Tilton. What we're going to do is learn how to measure the geologic ages of a common mineral that's about the size of a head of a pin. It's called zi rcon." Little tiny zi rcon crystals occur as a minute trace constituent of common ordinary igneous rocks.

12 ENGINE£RING & ~(IENCE NO. 1997 I

The laboratory proved too X ONTII small to contain Harrison Brown, who eventually ATI. I .\ T IC became professor of science and government as well as professor of geochemistry.

When those rocks crystallize and form from because when other parts of the solar disk of rhe magma, a lot of differenc crystals form in there, planets were forming- for example, the Earth­ and among them are tiny bits and pieces of zircon. they tOok in both lead and uranium. Therefore They have uranium bue no lead. And as they si t the lead in the Earrh today is a mixture of two there, the uranium decays to lead, and you can things: the primordial lead that was there in the do uranium-lead measurements. However, the beginning, and the lead thar has been created by amounts of uranium in there are onl y a few parts uranium decay since the Earth was formed. per million, and that's decayed co even smaller There are twO isotOpes of uranium that decayed parts of lead. to two different isotOpes of lead, and there's also What Brown wanted Tilton and me CO do was CO thorium, which decays to another isotope of lead. develop mass spectrometric techniques to measure So you have three different isotopes of lead. Arid amounts of uranium and study rhe isotopic com­ the whole thing gets mixed up. You've got all positions of amounts of lead that were a thousand these separate age equations for the different iso­ times smaller than anything anyone had ever topes of uranium and different isotopes of lead looked at before. that were formed. And it was not known what the Three different methods had been developed isotopic composition oflead was in propottion to for measuring ages: the uranium-lead age, the these different isotopes in the Earth when it was potassium-argon age, and the strontium-rubidium first formed. If we only knew what the isotopic age. I didn't work on the latter tWO, only on the composition of primordial lead was in the Earth at the time it was formed, we could take that num­ ber and stick in into this marvelous equation that If we only knew what the isotopic composition of pri mord ial lead was in the the atomic physicists had worked out. And you could turn the crank and blip-out would come Earth at the time it was formed, we could take that number and stick it into the age of the Earth. So Brown said, "Pat, after you figure out how to this marvelous equation that the atomic physicists had worked out. And you do the isotopic composition of these zircons , you could turn the crank and blip-oLit would come the age of the Earth. will then know how to get the lead-you will have it all se t up. You just go in and get an iron meteorite-I'll get it for you. We'll get the lead out of the iron meteorite. You measure its isotopic uranium-lead. I was the lead man, and Tilton was composition and you stick it into the equation. the urani urn man. And you'll be famous, because you will have Tilton only had to measure concentrations. I measured the age of the Earth." had to measure isotopic compositions, and that is I said, "Good, I will do that." And he said, "It different. And Brown said, "Well , Pat, here's the will be duck soup, Parterson." deal. Once you do that, then here's what you do." Brown had worked Out this concept that the lead SC: Did he ever work in the lab with you? in iron meteorites was the kind of lead that was in the solar system when it was first formed, and that CP: No. He came in one time. He was trying to it was preserved in iron meteorites without change show us something. It blew up in his face. H e from the uranium decay, because there is no had to stick his head under the water faucet. He uranium in iron meteorites. Now, this is crucial was betrer Out of the lab.

1997 ENGINEERING & SCIENCE NO. It was a very fruitful t ype of p ioneering work. But I hadn't yet gotten to the

measurement of the age of the Earth and s tudying the lead in iron meteorites ,

which Harrison had told me was duc k soup five years earlier.

I became aware of the contamination problem because I kept getting the wrong answer for lead in these zircons. We knew what the amount of lead should be because we knew the age of the rock from which it came and because of George Tilton's measurement of the amount of tiny bi ts of uranium in there. We could calculate how much lead there should be and what its isotopic compo­ sition should be. And it kept coming up with the wrong number, so I had to figure out why- to go to all these sources for different possibilities. . That's how J found out all about this contamina­ tion-that there was lead coming from every­ where. And in the process of finding this out, I learned how to analyze very low concentrations of lead in everything- in common ordinary things like hair, clothing- that people had never thought about.

SC: I'd like to talk a little bit more about your interaction with Harrison Brown over the years that you were his student.

CP: Well, I got my PhD at the University of Chicago after five or six years of work there. George Tilton and I published a paper on how to determine the ages of the tiny zircon crystals in the rocks. This was very important because it is one of the three major methods that were subse­ quently used to delineate the geologic history of the Earth. It was a very fruitful type of pioneering work. Bur J hadn't yet gotten to the measurement of the age of the Earth and studying the lead in iron meteorites, which Harrison had told me,was duck soup five years earlier. When I was finishing my PhD, I said, "Well, Harrison, I really would like to continue this work Patterson uses a distillation apparatus to purify reagents in and measure the age of the Earth and get the lead Out of the meteorite, but I need ro work as a post­ his Caltech lab in 1957. He didn't trust the purity of doc here at Chicago to do that." And he said, commercial chemicals, so he redistilled them. "OK, Pat, go ahead." The Atomic Energy Com­ mission had financed George's and my work for

24 EN G INEE R ING & SCIENCE NO. 1997 Right: Patterson and his family in 1955- Charles, Cameron, Claire, and, in front, Susan. Below: 1957- Patterson loads a sample for analysis in the mass spectrometer.

1953. It was the fi rst measurement of the age of the Earth to be published. It was not understood by the geological scien­ tific community at all-how this was done or any­ thing about its meaning or signifi cance. It was a dozen years befo re this munber got into the geol­ ogy textbooks. Before then the age of the Earth was very vague-some billions of years. Then when the correct number began co appear in the geology textbooks, they never said how it was determined, only that it was due co uranium-lead geochronological measurements. Bur what was said was in correct, of course. It the fi ve previous years, and since they had financed was n't unci l maybe 10 or 15 years later that a few us fo r work that led up ro this, I wrote a new of my colleagues were able to really do this cor­ proposal. They turned it down. They said they rectly. You must recognize that chis number that I weren't interested in measuring the age of the had measured related to the ti me of the coalescing Earth. of this planet Out of the solar disk. Now, that is a 1 cried on H arrison's shoulder. He said , "Pat, finite period of time. But do you have a billion that's all right. I'll rewrite you r proposal in my years for thi s to take place-whi ch is a substantial name." And YO ll know, he 's very good at explain­ fraction of the time that's passed since the Earch ing things to people in a nonscientific way. So he was for med-or was it just a very short time? rewrote the darn thing. Boom. I was award ed a Well, they were working on that. And I didn't posrdoc fellowship. J did about half the wo rk in give tWO hoots for that. My attitude was: "1 don't one year in getting lead Out of meteorites ro do want to work on that stuff anymore. What I wane this, and then Harrison got offered a big job here to work on is the evolution of the Earth- what at Cal tech in the geology department. And he happened to the Earth itself du ri ng the time it was brought me along with him. coalesci ng." Now, the reason I could do that was that we SC: And you JU St said yes? could use lead isotopes as measures. But the iso­ topic composition of the lead was changing-it CP: Yes, of course. Because I wanted to continue was dynamic, because uran ium was decaying all this wo rk. After he ca me, he got more money out the time, and there were three radioacti ve progeni­ of the AtOmic Energy Commission to build a mass tors of three different isotopes in this lead that spectrometer here, to build me a laboratory to were being added all the time the Earth was there. work in. And in that laboratory I isolated iron­ These parts were moving around all over the place, meteorite lead. But 1 didn't have a mass spec­ and uranium and thorium were being separated trometer built yet, so I flew back to rhe Unive rsity from each ocher and from lead, due co the differem of Chicago and used their new one to meas ure this chemical ptoperties in these different components stuff. It was wo nderfu l. And that was the data that were moving around. Some had sulfur, some that I used to publish this paper delineating the had oxygen, some had silicon. And these di ffe rent measurement of the age of the Earth. This was in components would grab ontO d ifferent chemica l

1997 (N GI NEHING & 1CIEHCE NO. strengths of the lead, uranium, and thorium, and segregate them in different parts, so that the pro­ porrions of lead and uranium and rhorium would change for millions and hundreds of millions of years at different areas. The lead within would have a different isotopic composition. And you could track this. So, today, you could look at the lead in rocks and begin to put together a picture of how they had been chemically related in past times. And you could get other people's work to help YOLI interpret what that chemistry meant in terms of position in the Earth. And then we could get times. So that's what I was interested in, what I started out doing. I said to hell with this damn stuff about cosmology of the sun to the planets. There were other people who were very interested in that, so they worked their hearts out to prove I was wrong.

sc: They didn't like your number' SC: You have to know how to do it.

CP: No, no. They wanted to make discoveries. CP: Yes, but that's the way I would write my And in order to make discoveries, you can't just proposals. And I never got funded. Harrison prove somebody's right. You want to prove that would get them funded for me. I stopped somebody's wrong. interacting with him when he moved over to Baxter Hall {as of 1967 BrOlvn held a joint appoint­ SC: But meanwhile, you were getting grants to ment as professor of science and government and professor do this. of geochemistry} and I had to start getting my own money, which I failed at, of course. But Brown CP: No, Harrison was getting them. Let's get protected me. back to Harrison. In all this time, I was trying to There was an important phase here in Brown's shift back to using lead isotopes. Now, in order to helping me get money. He shifted from the do that, believe it or nor, Harrison got money Atomic Energy Commission after about four from the Atomic Energy Commission to do this or five years; he had a new idea. I was studying kind of work at Caltech. He was talking about, sediments. In order to figure out what was hap­ oh, how my work was related to uranium, of pening in the past, I would have to get oceanic course. He went through all these calculations, sediments. You see, the rocks would erode (they'd and he told the Atomic Energy Commission how have lead in them), and then they would form there was enough uranium in ordinary igneous sediments, and you could measure the age of the rock, that if you ground that rock up and then sediments. I wanted to sample all the continents leached it with hydrochloric acid, you would get of different times, and the oceans were a mixing enough uranium to use in an atomic generator reservoir. And we would look at this mixture in that would be equivalent in energy to 10,000 tons the sediments as a function of time. of coal. In other words, 10,000 tons of coal would And Harrison said-he was a brilliant guy equal the amount of energy of the uranium in one politically-"Oh, heck, the oil compani es should ton of granite. be interested in this." Why? "Well, the isotopic composition of the lead is a tracer that helps SC: And rhey bought that' identify the stage, or the age, to characterize the time or the type of the sediment that you have." CP: They bought that' It was that kind of sales So he convinced the oil companies that they should pitch he used. Now listen, you know what I finance my research because it would assist them would say. I would say, "Well, I want to know in identifying oil deposits. You know, when you how this chunk of North America evolved and drill a core , you're looking at bands in a rock. And then got thrown around and came over here, and if you measure the lead isotopes in there, it can how this other chunk came up later. And we want give you more information than you had before. It to know when this chunk came up and when that could help characterize the type of sediment, so it chunk came up, and how they were related to each could help you locate and identify oil deposits and other, what their ancestry was." And the Atomic reservoirs here and there. So they started. It was a Energy Commission would say to me, "To hell national consortium of oil companies that had this with you, Patterson! We don't care about that big research fund that they doled out to help them stuff at all." do this stuff. Harrison got money from them

2. ENGINEERING & SCIENCE NO. 1997 scuff over a few thousand years into lattices of minerals. You could get the lead out of those minerals by JUSt taking a piece of sediment and treating it very gently with a li ttle diluted acid. But you wouldn't get the lead in the clay particles that had migrated out from the rivers and then fallen down and formed the bulk of the sediments. I want to know how this chunk of North America evo lved and then gOt thrown You wouldn't touch that lead. You 'd only get this around and came over here, and how this other chunk came up later. little tiny amount of lead that had been in the zooplankton, because that soluble lead collects on the outside of their little bodies. The drilling project was a Scripps {Institlttion of Ocea11Ography} operation, a gigantic one. I don't know whether this was paid for by the govern­ ment at by the oil companies, Ot both. They would lower drills down through fonr miles of water, and then they would bring these cores back up. The cores were down at Scripps, and we got segments of that stuff They came from all over the Pacific. There was a pattern scattered throughout the basin of the Pacific, so we could see what was coming from China, what was coming from North America, what was coming fro m South America at different times. Different scuff came at different times, depending on the climate. Now, this isn't over thousands of years, or even hundreds of thousands of years. I mean, we're talking about millions of years . In addition to dating this stuff, Ed Goldbetg at Scripps and a guy by the name of Gustaf Arrhenius (who's the grandson of the famous Swedish chemist Arrhenius) had worked out this autogenic mineral busin ess. They didn't study lead, but they knew that this mineral was made of reconstituted plankton poop. They had identified this mineral, which nobody had done before . I knew that when I leached it with acid, I'd been taking their mineral and getting the lead out, 1971 which came from the soluble lead. When we measured that, using the ages of the sediments, we had a measurement of the rate of every year, huge amounts, to fund the operation of the past flow of lead through the oceans all over­ my laboratory, which had nothing whatsoever to and this was millions of years ago. So we knew do with oil in any way, shape, or form. the quantity, the rate per square centimeter of sediment surface (the bottom of the ocean), grams SC: That's called basic research. per square centimeter per year, We knew how much was flowing through there. CP: And then a vety bad thing happened. We Then I gOt some data from the rivers. There found from measuring lead in these sediments how were these idiots who were measuring lead in river much lead had been passing through the oceans waters who didn't know anything about how to and depositing in these sediments. Now, there are measure lead. And I knew that because 1 had rwo kinds of lead: there's a soluble lead that's in previously worked out how to do the measure­ the water, and then there's lead in particles. These ments for meteorites. But I took their data for panicles are what the sedimencs are made Out of. river water, and 1 multiplied by all the rivers how However, a small fraction of sedimencs are made much river water there is in the oceans each year. our of residues of organisms that are living in the And I came OUt with a numbet fot lead that was a watet-like zooplanktOn poop. And they fall hundred times greater than the amount that we down through four miles of water! had measured that was flowing through the oceans When the zooplankron residues got inco the in the past. I thought, something is wrong here. sediments down there, there'd be a chemical re­ Are these guys wrong? Or is there really that constitution, what we call "formation of autogenic much lead coming intO the oceans today? minerals." It would be a rearrangement of that At that time we were working out methods fot

1997 ENGINEERING & !(IEN(f NO, 27 polar regions. It comes OUt of the air that has lead in it. Lead is in the snowflakes. It goes down and you have a layer there. Next year you have another layer. r started working at the Noreh Pole. We had to colleer the concentrations of lead in the snow there; it was a thousand times lower than the We collecred rhe snow and brought it back here and we analyzed ir and found concentrations of lead in the "pure" laborarory water in most laboratories. 1 had to measure huge concenrrations of lead increasing since the 1700s until now-a 200- or concentrations that they couldn't measure. In other words, these scientists were using the purest 300-fold increase in the cOllcenrracions of lead. water in the laboratories as a baseline, and they couldn't go below that level. I had to see varia­ tions with time, and the techniques for doing that at these levels were not at all developed. I required a two-foot cubic block of snow. And then we had to dig shafts down 200 or 300 meters to get these blocks of ice over a period of time to see what was happening. We sawed even larger blocks out of the walls of a runnel, while wea ting raking what we call a profile. A ship would sit acid-clean plastic gloves and suits and using clean there and it would lower something and grab saws. We collected the snow and brought it back some water. hoist it up, and then it would lower it here and we analyzed it and found huge concentra­ deeper and hoist it up. We'd collect water up and tions of lead increasing since the 1700s uncil down for vertical miles. and we'd measure it. now-a 200- or 300-fold increase in the concen­ trations of lead. SC: Did you have somebody on the sh ip doing About the same time 1 had proposed a concept this? You weren't on the boat yourself? that I call biopurification, which has to do with what the natural level of lead should be in people. CP: Yes I was! And I gOt sicker than a dog! You start Out by looking at the calcium in our Now, in these profiles, you look at how the bodies and asking where it comes from. You track concentration changes with depth. We found a it back- you look at the food that we eat, you huge increase in the upper portions of the oceans, look at the organisms that made that food. and which decreased to lower concentrations with you keep going down the food chain until you depth. Now, why is that? Why should the lead come to plants. Then you go down to the soil, and be so high? The waters don't mix that rapidly. you go from there to the rocks that the soil came And the waters up near the top are much younger from. It so happens that there are calcium-like than the waters lower down. It takes a long. long trace metals such as barium but in very small time for them to mix. abundances-a tiny, tiny, infinitesimal trace. So I made some calculations. What about the Barium is like calcium but it's chemically differ­ lead in gasoline? We extrapolated from our pro­ ent. For one thing, it's a very massive atom. It files how much lead was in the upper part of all has different chemical properties than calcium­ the world's oceans, and it could easily be ac­ nor grossly different, JUSt enough different so that counted for by the amount of lead that was put barium is poisonous as hell. When we evolved into gasoline and burned and PUt in the atmo­ through all these millions of years of evolution, sphere. We had more tons put in the atmosphere from the tetraethyllead added to gasoline than we could see in the upper part of the world's oceans. And that's what caused the problem. The oil companies wete financing my work. We'te in serious trouble.

SC: Even Harrison Brown would have trouble with that one.

CP: Oh. he did! And that's when he disassociated himself from me. H e stopped getting money from the oil companies, and I had to starr getring it myself. I needed money because si nce I got this idea about lead coming from gasol ine, I wanted ro look at the record. Where do you see that record? You see it in the snow that never melts in the -.'

28 ENGINHRING & SCIENCE NO. 1997 On their Antarctic expedition in 1966, Patterson and his crew sawed large blocks of snow from shafts 300 feet deep and from tunnels, then shipped them back to Cal tech to analyze. They found an enormous increase in concentrations of lead since the Industrial Revolution.

nature devised an exclusion mechanism fo r ratio in rocks dropped in our food and in us by a handling this. Each one of the organisms in this factor of a hundred. The barium ratio shows that food chain has to have calci mTI , bur nature evolved lead should be a hundred times less than it actually a process for the exclusion of barium when they is in us today. We are being poisoned by lead. cake up calcium . There are cerrain kinds of And guess where it is coming from? Look at the proteins that grab hold of the calcium and pull it, ocean-thar's coming from tetraethyl lead. and they don't do thar efficiently for barium. So Others thought th.is was a pile of crap! They there's an enormous reduCtion of the barium-to­ said, "Panerson, would you please starr worrying calcium rario. about science instead of this hea lth crap. What a Lead comes in with the calcium too. And it's waste! Here you are; you measured the age of the distributed in the body much like calc ium (not on Earth, and you're worrying about tetraethyl lead a molecular basis-there are different proteins) but and rhis stupid scuff about lead in bones." But I in a general, morphological distribution. Ninety­ was right. When we finally actuall y measured the nine percent of it is in our bones. lead ratio-it took about 25 yea rs to do this I knew what the lead/calcium ratio is in average accurately- it curned OUt to be a factor of a people today. (The people who measured rhis thousand instead of a hundred. I was off-the couldn't measure lead properly in bones, bur I used wrong way. their data anyway.) So I had the ratio going from rocks to food to people. And do you know, the (111 tbe LeISt 15 )Iears Patterson came b

1997 ENGINE£RING & HIENCE NO. Right: In 1972 Patterson examines an ancient copper knife as part of his study of the history of metallurgy. Below: Patterson and Dorothy Settle, member of the professional staff and a longtime colleague of Patterson's, point out the lead solder on tuna cans in 1980.

SC: So it was a lot of cooperation with a lot of people.

CP: Yes. Then they could use my findings as a glowing example of what was being done by these cooperative research projects. Because other peo­ ple couldn't measure these numbers , they had to come to my laboratory from all over the world to find Out how to do that. In the meantime I started going to the poles to We continued measuring the lead, not just in get ice, and measuring lead coming out of volca­ oceans, volcanoes, and the poles, but also the land noes, and developing complicated devices for areas-plants, animals, higb mountains. And getting seawater. finally we got a picture tbat confirmed and clearly showed that Earth's entire biosphere was heavily SC: What was your mocivation at this point? contaminated with industrial lead emitted into Were you thinking in an environmental sense? the atmosphere from smelters and from automo­ bile exhaust. Urban areas were further polluted by CP: No, I was noc! Science, science, science! I other sources of lead being moved around- solder wanted to know what [his natural level of lead is. in tin cans and so on. I didn't care two hoots about verifying what the At the same time, I had been investigating the contamination was. I was forced to measure [he history of the production of lead. I went back contamination in order to arrive at the natural 9,000 years to when metallurgy began; lead came level. But there were friends and colleagues who into rhe picture two-thirds of the way along. And were environmentalists, and they used my work. then I showed how lead was related to [he devel­ My work was used to get the lead out of gasoline. opment of coinage in metal. I made quantitative As a matter of fact, I wrote a paper on this bio­ calculations based upon data that were available purification concept where I said that we have a from historical records of production of lead. I hundred times more lead than we should have. worked out ways to estimate from ancient data And that's when I really got shot down by the oil how much lead had been mined by the Greeks and companies. Romans. I had to figure out the rate of production I got some money from the National Science indirectly, about five different ways, including Foundation for quite a while. And then I shifted figuring Out the half-life of coins. to the National Institutes of Health and the I got this curve for the production of lead, Department of H ealth, Education, and Welfare which I published about 20 years ago. Some and that SOrt of stuff. And then there was the of my colleagues that I'd been teaching how to International Geophysical Year. I had colleagues analyze lead in polar snows gOt together to work who were nOt working with lead, but who were on this. I had dug shafts in the snow back to working with other things that were related to around the Industrial Revolution but I wasn't able lead. And we put in proposals together to get to go back 2,000 years to the time of the Romans support for sample-collection procedutes, the costs and Greeks, so I couldn't measure the lead concen­ afield, and then I could add my laboratory and my trations then. But these French and Australian salary and my visiting colleagues. colleagues used new techniques that we had devel-

'0 ENGINEERING & S(IENCE NO. 1997 The Institute has a whole lot of things wrong with it- everything has some-

thing wrong with it. But because this Institute existed, I existed.

did, to reduce the effects of lead on people and the environment today. But I my-self asked- what is the meaning of this? How did we think? What led us to poison the Earth's biosphere with lead? I therefore shifted to trying to figure out how we thought, and this is where I evolved this new concept of human consciousness in terms of pathways. neuronal circuitries, that are used within the brain to thiok in two different modes.

{It was, in particulat; his work on Roman lead production -((the relationships between social interac­ tions and the metallurgical technologies"- that led Patterson in this direction. His concept involved what he called a utilitarian, or problem-solving, type of thinking, and nonutilitarian thinking, which includes religion, art, philosophy, history, and science, in which ([you're trying to understand; the individual brain 1995 becomes aware of something and it asks why?" He viewed the latter as less hard-wired in the brain than utilitarian thinking, although it would have had an evolutionary role in conferring advantage on tribes who oped in my laboratory to do that. {This work was thought this way.) published in 1994,} Guess what their curve was as a function of time? It fell right on top of my lead­ sc: You must have some good feeling for Cal tech production curve. for allowing you to live this way? Do you rhink I was proud? No! You know what I said? "This proves that for 2,000 years we CP: Oh yes. The Institute has a whole lot of have been unable to understand the evil that we things wrong with it---everything has something are doing to ourselves and the biosphere." Be­ wrong with it. But because this Institute existed, cause, you see, this lead was coming out of the I existed. Greek and Roman smelters iota the atmosphere, going around the Earth, part of it working its way SC: You don't think there would have been up, and was incorporated io the snow that fell at anywhere else where you could have lived this life? the North Pole. Many of my colleagues immediately jumped on CP: No. I would have been a molecular biologist {the evidence of the ina'eased level of lead in human at the University of Ohio, fighting, unhappy, hones}-not the people working with me but quarreling, and not being able to accomplish any­ scientists who were concerned about the environ­ thing. Cal tech provided this environment inad­ ment and about people being hurt by all this. vertently-it didn't do it intentionally. It was They wanted to use this information, and they just there. It was a magnificent opportunity. D

1997 fMGINEERING & HlfNU NO. The machine that resulted from all this effo rt is a motorcycle only in the sense that it has two wheels, one

in front of the other, and an eng ine that sends power to the rea r wheel t hrough a drive chain. O therwise,

the 18-foot , 4 -inch veh icle looks like a wingless 1950s-vintage jet fighter-complete with a scoop-nosed

air intake, a cockpit canopy of clear, shatterptoof plasric, and a fuselage tapering to a vertical-finned tail.

------I fJfJfj S- G

Et/GINEEkIHG & HIEHCE NO. t 9 9 7 I'm Not Driving Fast, I'm Just Flying Low by Douglas L. Smith

On August 19, 1996, a streamlined motorcycle whose aerodynamic shell was designed by a bunch of Caltech grad students as a class project set a land­ speed record at the Bonneville Salt Flats International Speedway. The flats, about 100 square miles in extent, lie entirely in Utah, but the closest hotels­ or habitation in any form!-are in Wendover, Nevada, just over [he srate line. (Nevada's more liberal approach to gambling and liquor may have influenced this.) The flats are part of the Great Salt Lake Desert, the desiccated corpse of Lake Bonneville, which in the late Pleistocene epoch covered most of Utah In September 1996, Sam west of what is now 1-15 and spilled over into Nevada and Idaho. The drying Wheeler returned to the lake·s bequest of dissolved minerals that had washed down from the surround­ Bonneville Salt Flats ing mountains left a dead-level salt pan that, after baking under the summer sun, is as hard as concrete and ideally suited for going way too fast-there's to break the record he had absolutely nothing to run into for miles and miles. Numerous speed and set only a month before. endurance records have been set there since 1935. And there are records galore-in the interest of equity, all competing vehicles are classified by body configuration, engine size and type, and fuel type. Each class has its own record, and there are also overall records. Back in 1970, a gent named Sam Wheeler set a class record there of 208 miles per hour in a streamlined motorcycle. (To be classified as a streamliner, the vehicle must have the front wheel or wheels covered by an aerodynamic skirt, and must be powered by a mechanical drive train (as opposed to jet and rocket cars, whose wheels merely support the vehicle's weight).) This histori­ cal footnore would have nothing whatsoever to do with Caltech, except that some 15 years later, Wheeler would once again catch record fever. He had long since retired from racing by then, and had become chief engineer for E-Z-Hook, a manufacturer of test probes and leads in Arcadia, California. His boss, company owner and Caltech Associate Phelps Wood, quickly agreed ro underwrite him. Wheeler designed and built the motorcycle's frame in E-Z-Hook's machine shop on evenings and weekends over rhe next three years. Meanwhile, Wheeler and Wood became concerned about the possibility of going airborne, and nor in a good way. Wheeler's earlier bike had handled well enough in its record run, but subsequent wind-tunnel tests had shown that its front end was generating a lot of lift-perhaps enough to flip the bike on its back at higher speeds. As Wheeler says, "At 300 miles an hour, you can make a rock fly." In the spring of'89, Wood mentioned this concern to another motorcycle enthusiast, Elliott Andrews, Caltech's division administra­ tor for Engineering and Applied Science. Recalls then-graduate student C. Dennis Moore (PhD '96), "Upside-down and backwards is not really a good way to set a record. So Andrews called Professor {Hans] Hornung Uohnson Professor of Aeronautics and director of the Graduate Aeronautical Laborato­ ries (GALCIT)). Professor Hornung said, 'Hmmmm, Dennis rides a motor­ cycle.' I came in one morning and found a note on my desk saying call Sam

1997 ENGINEERING & SCIENCE NO. increase the drag. A more hori zontal onc would have little drag bur might be hard to see through. The students sc ulpted a variety of tiS-scale clay models aro und a wooden armature that represent­ ed the frame members and sllch other givens as the engine, whee l, and drive-chain clearances, and tested them in the low-speed water channel in Karman Lab. The tests were run by hanging the model upside-down in the channel, so that the wheels' bottoms just touched the water's surface, which stood in for the ground. The class tested six bodies "based on airfoils, previous streamliner •

--_~~. ~... J;. - ""Jf~ ~~ W heele r. Who is thi s Sam Wheeler guy and why do I have to call him?" But call he did, and quick­ .~ , ly got hooked. Determining how the bike would handle was Above: (from left) Grad clearly a wind-tunnel exercise. But the bike's students Roderick stability would depend on what sort of body was desig ns. and wi ld ideas" says Moore, beginning put on it. Since the body hadn 't been built yet, with rhe variant with the largest nose and tai l­ Daebelliehn, Michael Andrews proposed to Hornung and Gerald Landry, jt's a lot easier to shave clay off accurately than to Dominick, Elizabeth manager of GAL CIT's 10-foot wind tunnel, that stick it on. The winner was then scaled up to 114 McKenney, Yvan Maciel, designing the body and determining its hand ling size for further tests in GALCIT's lO-foot wind characteristics would make a good ass ignment fo r runnel. In the 1I4-size tests, the students even and Dennis Moore with the Aeronautics 104 (Experimental Methods). At Ae worried about the engine's intake and exhaust. 1/4-scale model in the 10 4 's first meeti ng, the professors hawk a bazaar of Plastic tubes scaled to the diameter of the air IO-foot wind tunnel's term-length projects, fro m which five or six actu­ duces ran from the intake scoop through the test section. ally get chosen by the students. This project made model's imeri or. Recalls Moore, "We'd dig the for a nice confluence of interests- the students got tubes Out of the clay, move them around, and Right: In the water-channel a real-world project of manageable scope (most of check the effec t of the exhaust's loca tion on drag." experiments, a lIB-scale the offerings tend to be academic), and Wheeler Yvan Maciel did the mathematical modeling of clay model of each pro­ co uldn't afford to use the 10-foot tunnel at the stability that guided the clay modeling and test­ posed body shape was sub­ outside rate. "E-Z-Hook only had $2,500 to ing done by the res( of the group. For example, spend," recalls Landry. "Normally, $2,500 gets a big fin would lend stability bur increase drag. merged upside-down with you one day." But interdepartmental charges for The fin al design was a compromise, reached after its wheels just touching tunnel time at the student rate are much lower, so seveml iterations. "We'd give him the data out the water's surface. the company donated the money to the class. The of our wind runnel. He'd go plug it in to his class kicked in some of its operating budget as computer simulation and come back and say, 'The well , and the project wound up in the 10-foot bike's not stabl e. You have to do so mething else.'" runnel for two and a half weeks . Moore TA' ed Designers of land-speed vehicles have histori­ the assignment, which was offered in the spring cally favored brute horsepower over fin esse. Says of 1990, and sold it to the class. Moore, "People don't worry about aerodynamics. The students who bought into the project­ They'd much rather spend $5,000 on a new· set of Yvan Maciel (MS '90), Roderick Daebell iehn (MS pistons and stuff for their engines, than to spend '90), Elizabeth McKenney (Ph D '95), and Michael $5,000 to reduce the drag." One exception is Dominick (MS '90)- were given carte blanche to Goldenrod, a streamlined car whose des ign was try any shape they liked, provided that it fit also tes ted in the lO-foot tunnel. The creation of around Wheeler's frame. The goal was to find the Waiter Korff, a Lockheed aerodynamicist, Goldell­ shape with the lowest possible drag that was still rod held rhe overall land-speed record of 409 miles aerodynamica ll y stable- no lifting, no pitching, an hour for over 25 years. But most streamliners and no flopping over sideways. The choices were are laid out by eye, based on a gut feeling for what endless-rounded nose or blunt nose? Are tail fi ns looks fast, leavened wi th experience and imitation. good? How many? How big? How oriented? Streamliners tend to have needle or bullet noses And what about the windshield? A near-vertica l and stubby rear ends, because using a blunt tail so one would give better visibili ty but significantly the thing wi ll fit in the garage is of greater practi-

• 4 ENGIHEfRINCi & SCIENC E NO • 1997 where he ca n slip the clutch and power away from the starting line.) H owever, the laws of aerodynamics say that it really doesn't matter what you do to the front (as long as you round the corners)-ir's what's out back that counts. The airflow off the rear of the vehicle tends to keep going straighr, even though the body isn't there any more, JUSt as Wile E. Coyote can run for some disrance off the edge of a cliff as long as he doesn't look down. Wherever the flow separates from the body, a partial vacuum forms, creating drag. But a properly tapered tail guides the flow and prevents separation. A typica l streamliner has a drag coefficient of 0.15-0.20. This is pretty good, actually-for comparison, a VW bus has a drag coefficient of about 0.45, while an 'S7 Camaro scores a sleeker 0.30. A sheet of plywood held perpendicular to the wind has a drag coefficient of approximarely 1.0. The Ae 104 design has a mean drag coefficiene ofO.103-the lowest ever measured for a stable land vehicle in the lO-foot tunnel-and no visible separation. (The rear tire did generate a wake, but it reattached itself to the underside of the bike fur­ ther downstream; there was no observed separation off [he fcone wheel or [he body.) And what of the stability, which was why Cal­ tech had gotten involved in the first place? With lift and pitch zeroed out, there remain three basic ways ro tip a mororcycle. In rhe "wobble" mode, a wheel oscillates ever more violently around its axis until the bike falls over si deways. Then there's the "capsize" mode, in which rhe front tire kicks to onc side as rhe bike leans in the same direction. The bi ke lies down (on your leg), and you slide ro a srop. This is a good way CO break a leg if you don't have a roll cage, but otherwise it's a comparatively safe crash. And finally, in the "weave" mode, which makes for rhe most spec­ mcular video footage, the bike's frone wheel kicks one way as the bike lea ns the opposite way. Cen­ trifugal force brings YOll back but overshoots, and the sideways swi ng gets wider each time. Eveneu­ ally the front wheel digs in and the bike vaults skyward, rear wheel first, and tumbles end-over­ cal importance than the aerodynamic virtues of a end for some distance before coming ro rest. Wheeler, t he streamliner, tapered tail. A cut-off rear end also makes it eas ier Maciel predicted, and tests in the 10-foot tunnel and t he street bike t hat to install things like exhaust pipes, parachute confirmed, that the final design was extremely t he streamliner's engine rubes, and push bars. (Twin parachutes are stable against wobbling and capsizing. It was un­ required at Bonneville if you' ll be running over stable in weave mode, bur onl y under 40 mph­ norma ll y comes wit h. 250 miles per hour; between 175 and 250, you above that speed, it was rock-steady. only need one. A push bar is a stubby projection The machine that resulted from all this effort bolted ro the rear of the chassis-remarkably, very is a motorcycle only in the sense that it has two few of these speed machines can get under way wheels, one in fronc of the other, and an engine unassisted because of their extraordinarily high that sends power to the rear wheel through a drive gear ratios. Most streamliners come with over­ chain. Otherwise, the IS-foot, 4-in ch vehicle sized, testosterone-oozing pickup ttucks, some­ looks like a wingless 1950s-vincage jet fighter­ times with matching custom paine jobs, to push complete with a scoop-nosed air intake, a cockpit them until they get up enough speed to engage canopy of clear, sharrerproof plas tic, and a fuselage first gear. But the GALCIT bike is light enough tapering to a vertical-finned rai l. The wheel has that a couple of people running alongside and turned full circle, in a way, because the body for pushing suffi ces to get Wheeler up to 10 mph, Wheeler's previous record-breaking bike was

1997 ENGINEER IN G & SCI ENCE N O. ing 10-inch-thick Styrofoam blocks with a hot wire. The Styrofoam shapes, glued together, Right: Phelps Wood and became the full-sized pattern. R . D. Boatworks the streamlined body pose in Dana Point created a mold from the pattern, in his backyard. and built the body up layer by painstaking layer. Recalls Moore, "Pour months later, we got back a Below: The frame hanging shell. We then had to make all the wheel cutouts around in the pit area on and stuff, which took a significant amount of the salt. From left: the time." Furthermore, because of the custom parachute tubes (only the design, virtually every part that went into the motOrcycle had to be machined by Wheeler by top one is loaded), the gas hand. Commercial steering linkages wouldn't tank (with red cap), the fit, and you couldn't buy drive sprockets with the radiator enclosure (yellow), right nwnber of teeth for the gearing ratios the bike needed-the list goes on and on. Moore, the rear wheel and drive who by now was spending almost as much time chain, the engine (between f.1shioned from a drop tank off an old fighter jet. at E-Z-Hook as he was at GALCIT, designed and the silver air box and the The new body, however, is a sandwich of carbon built the bike's electrical, pneumatic, and coolant orange air-intake hoses), fiber, foam, and carbon fiber-the same material systems. Meanwhile, Andrews, who used to they make surfboards and racing-yacht hulls out and the firewall. The of. Within that body, the frame, which doubles The bil

,. ENGINEERING & SCl!NCE NO. 19 9 7 Above: The pit area just goes on and on. racing. and the pit area consists of rank upon rank of moror homes and ueility trailers, incerspersed Below: Here's something with carps pitched by homegrown mechanical you don't see every day-a geniuses ro shade the products of their ingenuity six-door pickup truck. It's from the blistering sun. (Try picking up a socket pushing a much-altered wrench that's been si tting oue in the lOO-degree glare of midday some time. Sunglasses are de 19305 coupe. rigueur, and it's imporeanc co wear sunscreen on the underside of your chin and ears, or the reflec­ ted light off the salt will burn some really sensi­ tive skin really fast. "Many people are surprised when their tan lines don't Stop at the botrom edge of their shores," says Andrews.) People spend more time working on their machines, cruisi ng around the pit area to see how everybody else is fixed, swapping lies with old pals, and offering sage advice to newcomers than they do actually racing. You can see every sort of rig imaginable­ from stock Trans Ams and tricked-out, chromed­ to-the-max street rods with snappy paint jobs that have no business being on the highly corrosive salt, ro the semi with a rug-boat engine that set a record for diesel trucks at 238 miles per hour. Sue the streamliner is king. The racecoutse itself, which is located several miles from the pit area for safe ty reasons, used ro be 11 miles long, but the salt (originally aboue Speed Week is Woodstock for gearheads. Bonneville is the last bastion of three feet thick) has been getting very thin oflate. mom-and-pop racing, and the pit area consists of rank upon rank of motor In some places, the dirt beneath is starring to show through. For the past 50 years, a mining homes and utiliry trailers, interspersed with tarps pitched by homegrown concession on the east side of the flats has been sucking up the salt as brine, extracting the potash mechanical gen iuses to shade the products of their ingenuity from and some other minerals-all told, only a few per­ cent of the salt by volume-and dumping the rest the b listering sun. in great heaps on the flats south of 1-80. Despite protests by racers and environmentalists alike, the BLM has JUSt renewed the mine's lease for another decade. So in order to stay safely on the good salt, the course is now only seven miles long. This gives you two miles to get up to speed, followed by three "flying miles" that are timed, and two more miles ro stop and clear the course for the

1997 ENGINEERING & ICIEN([ NO. Left: Brute force personified-the only concession to aerodynam­ ics is the small spoiler on the hood. And this guy don't need no stinkin' push truck, either!

Right: Racing apparel can next guy. You break a beam of light at each end get awfully hot, as this of each flying mile, and this informacion is relayed rider of a partially through a wireless system to the timing boOth, streamlined motorcycle which is actually a couple of fo lding [abies under a sunshade pitched at the three-mile mark. The can tell you. fascest of the three flying miles is you r official time. The electronic timing system is accurate to one one-thousandth of a mile per bour. h used to be that in order CO set a record, you had co cover the same piece of real estate twice, in opposite directions, and take the average speed. But in 1996, for the first time, all runs were made heading west, away from the potash works-the salt had shrunk so much that the company's dikes were getting uncomfortably close ro the course's eastern end. However, an average speed over twO runs is still required fo r a record. To run the course, you register once your vehicle has undergone an elaborate safety and technical inspection-a ritual that often takes an entire day-and you join the staging line by the timing booch. (On che firs< couple of days of Speed Week, you can wait in line all day fo r one run. The line shortens considerably over the week, as people break down, crash, or give up.) The stager sends the vehicles, a dozen or so at a time, out to the starting line to wait their turn. Going fast enough to qualify for a record attempt sends you to an impound area to await a second run. Otherwise, it's back to the tail of the staging line---or; most likely, back to the pits to see what went wrong. And lots can go wrong. The maiden run in Augusc 1991 was licerally a flop. The problem wasn't stability, but visibility. The cockpit canopy slopes a mere 10 degrees from the horizontal when lowered into posi tion, and ripples in its plastic that were imperceptible when it was upright suddenly caused the horizon to waver worse than a heat-shimmery parking lot on the Fourth of July. Undaunted, Wheeler tried to balance the bike by feel bur wiped our ar abour 60 mph. He was unhurt, which is a tribute to the SeTA's stringent

•• EHGINEERING & SCIENCE NO • 1997 On the first couple of days of Speed Week, you can wait in line all day for one

run. The line shonens considerably over the week, as people break down,

crash, or give up.

Above: Besides a fireproof safety rules-cirivers roucinely walk away from suit, t he SCTA requires crashes at much higher speeds-and the bike suffered less chan $300 worch of damage. drivers to wear fireproof "What did do damage, however, was when we boots and gloves, a hel­ were inspecting for damage back [at E-Z-Hook] met, a neck brace (here in Arcadia," says Moore. "We slung the bike up on scraps that were rated co 3,000 pounds apiece; being adjusted by Joe Von the bike, with Sam in it, weighs 1,000 pounds. Harten). and wrist re­ The straps, which admittedly were old, broke, straints to I

1997 ENGINEERING & SCIENCE NO. •• Below: The reason for a ll that safety gear-the driver didn't suffer so much as a scratch. That's his front suspension system sitting on the salt behind hi s door. His other door got so thoroughly folded that it would easily fit in his trunk. Right: Whenever somebody crashes, or the salt gets a little rough, they groom it with this saline zamboni: a grid of welded-together I-beams weighted down with drums of salt.

leave them alone. It took most of '92 to solve the alignment problem, at which point "I found that we hadn't been getting any stiction between the salt and the front wheel," says Wheeler. "But before I discovered this, 1 probably solved a whole lot of problems that didn't need fixing." In the final analysis, that inch-and-a-half-wide aluminum wheel just wasn't getting a very good grip on the salt, making the bike hard to steer. That winter, Wheeler whittled a wider wheel with a more aggressively grooved "tread" zone. No sooner was the new wheel finished, than the elements turned sour. Lousy weather and wet salt wiped out the entire '93 season. It was n't so bad if the crew knew in advance that the mee t was off- as in July, when the salt was st ill under 18 inches of water from the winter's snowmelt-but some­ times, to find out, they had to caravan up there Right: The crew tried and spend a day or tWO waiting in li ne not to run. everything during those Moore and Andrews were regulars on the prep frustrating post-crash team, and nearly always on the running team as well during those years. (The crew's other months. Here, Moore and regulars- Frank Sherri ll , Bill Ochelrree, Mike Andrews use a piece of Holliday and Chuck Bullwinkle- were racing string to do a quick-and­ cronies of W heeler's .) Bur as '94 rolled around, dirty rear-wheel alignment Moore's thesis research and Andrews ' involvement in the construction of the Gordon and Betty check between runs. Moore Laboratory of Eng ineering left no time to spare for the streamliner. Ronald Howe, of ]PL's atmospheric research group, picked up the slack. Howe, a mechanical engineer and machinist who designs and builds componencs for instrument packages flown on high-altitude balloons, had ac tually come to .bat for the project once before. "I made a maximum of maybe five percent of the [bike's] parts in my machine shop at home when Sam got rushed cowards the end ." Howe used to drag-race motOr­ cycles back in the '60s and '70s and had known W heeler by reputation then, but the twO didn't meet unci l much later. "1 met Sam about 15 years ago, out in the desert with a bunch of JPL guys

40 ENGINEERING & ICIENCE NO. 1997 "He was still accelerating because the course is tOO short. That makes him the

third-fastest bike ever-the fastest single-engine, fastest on gasoline."

riding motorcycles. I'd been crying to go up to makes him the third -fastest bike ever-the fastest Bonneville for years, but I never goc chere before single-engine, fastest on gasoline." The record Moore checks the tail fin's I went up with Sam." However, their dads, both for any kind of motorcycle is 323 mph, held by alignment. The fin, which test engineers/machinists and lifelong JPL a 2,600 pound behemorh sponsored by Easy Rider was damaged in the crash, employees, had worked together in the '50s, magazine and powered by twin 1.5-liter Harley­ eventually had to be building a hypersonic wind runnel (since decomis­ Davidson engines (a total of 3,000 cc) burning 80 sioned). The two also helped construct the Space percent nirromethane. Wheeler thinks he can take rebui lt completelYi sadly, Simulator, a high-vacuum chamber 25 feet in the Easy Rider; he'd just barely gotten into sixth only one set of GALen diameter and 85 feet tall in which full-sized gear at 301 mph. But he's going to need those decals had been made. spacecraft can be tested. missing mi les of salt-Moore has esrimated the The salt was wet during most of '94, too, but in drag based on the vehicle's performance in coast­ rhe one dry meet--Seprember- Wheeler did 206 down runs, in which they timed how the bike mph with the new wheel and the bike handled as slowed down when Wheeler shifted into neutral at well as it had before it got dropped. At the same a predetermined speed; knowing the bike's weight time, the SeTA started talking about banning and rhe deceleration rate allows you to calculate metal wheels, on the grounds chat they rutted drag. Moore's estimate is slightly below the values the salt and made it unsafe for other racers. This predicted in the Ae 104 class. "If there were two sparked a quest for yet another wheel, this one with more miles of salt, or 25 more horsepower in rhe a rubber tire. Goodyear's racing division eventu­ bike, we'd be there," he says. Absent the salt's ally donated a set of undersized tires that had been miraculous healing, a switch to alcohol fuel would custom-built for a streamlined motorcycle that get them the extra horsepower. But that's a proved to be too heavy to use them. The smallest project for the com ing year-the October meet tires Wheeler could find, they were still four inch­ gor canceled, so rhe books are closed on 1996. 0 es in diameter larger than the aluminum wheel, so an entirely new suspension system had to be built. In September 1995, with the rubber tire in hand, or rather on bike, Wheeler beat the class record, but it didn't count-the salt at the far end of the course was still soft from a rainstorm a few days earlier, and he couldn't make a rerurn run. That was the bad news. The good news was that the bike handled even better wich the rubber tire, and he was still only in fift h gear. The October meet got canceled for lack of entries. In 1996, everyrhing finally clicked. July gor rained out again, bur during Speed Week in August, Wheeler set the class record at 256 mph. On September 25, the first day of that month's meet, he raised it to 285 mph. Then, says Andrews, "Two days later he went 295 mph one­ way, with an exit speed of 301 mph. He was still accelerating because the course is too short. That

1997 ENGIN£ERING & SCIENCE NO. 41 B a oks Is HEATHCLIFF A MURDERER? PUZZLES IN 19TH-CENTURY FICTION by John Sutherland Oxford University Press, 1996; 258 pages

So, does Heathcliff murder to the fact that he has edi ted Cathy's brother in Wuthering a number of these works for Heights? Reasonable doubt. the World's Classics series. And how about Becky Sharp Most of the readers of this in Vanity Fair? Does she kill paperback, which made The Jos Sedley in the end' Of Times bestseller List in course not, says John Suther­ London, probably never lost land, but Thackeray wants his any sleep over these puzzles readers ro suspect her anyway. in the original texts. But Victorian authors, not stupid even if you didn't notice that by any means or even simply Jane Austen lets apple trees careless, had various reasons blossom in June in Emma and for slipping such red herrings that Dickens gets sloppy with and other enigmas and anom­ his seasons in Martin Chuzzle­ alies into their novels, and wit, this remarkably unstuffy Sutherland plays detective in book, with evocations of such teasing out these reasons and nineties phenomena (19905, suggesting imaginative that is) as date rape and the interpretations of 37 literary movie spoof Frankenhooke1j "puzzles." Sutherland, the not to mention reasonable Lord Northcliffe Professor of doubt, is fun to read and Modern English Literature at might even lure you into the VISIONS OF THE PAST: University College London, as novels themselves. So why THE CHALLENGE OF FILM TO OUR IDEA OF HISTORY well as a visiting (annually) did Henry James rewrite the by Robert A. Rosenstone professor of literature at Cal­ ending of The Port,"ait of a Harvard University Press, 1995; 271 pages tech, where he taught from Lady' And why doesn't H. 1983 to 1992, is a closer G. Wells's invisible man reader than most people, make himself an invisible suit When Professor of History that film is an even more perhaps not wholly unrelated and some invisible food? Raben Rosenstone first appropriate medium for introduced movies in his showing us the past than are Caltech classes in the 1970s, words on a page. There are, and then in 1977 taught a however, different and more course entitled "History on complex rules for history on THE CHEERFULNESS OF DUTCH ART: Film," class enrollments film than for history on the A RESCUE OPERATION soared. But ultimately this page, and in his book by Oscar Mandel innovarion made an even Rosenstone discusses how Davaco Publishers (Netherlands), 1996; 128 pages more profound impact on his these "rules" are observed in own studies, luring the self­ the various forms of historical acknowledged "Dragnet" film: for example, documen­ English novelists in the beneath the surface of the historian (just the facts, taries, films that mix fictional 19th century may have most riotous peasant feasts, ma'am) into the theoret.ical and historical characters, planted puzzles in their work, merry companies, and"even issues of how film works to films from other cultures, but 17th-century Dutch innocent stilllifes and land­ create or "re-create" history. and experimental films with painters mOSt assuredly did scapes. Mandel chalks this up "History does not exist until deliberate anachronisms and not, according to Professor of to our own century's "assault it is created," writes Rosen­ inventions that "re-vision" Literature Oscar Mandel. In on euphoria" and sets out to stone. Film, he found, offers history. He discusses five this short book Mandel rakes liberate the "self-evidently a new relationship to the past films in depth, including on the current intel1ectual happy works" of the 17th­ and a new concept of what we Reds and The Good Fight. fashion of imposing 20th­ century Dutch painters "from mean by "history." His latest Rosenstone, who served as century interpretations of the excesses of academic book, comprising a collection historical consultant on the "semi-veiled meanings" on earnestness." The Dutch of essays exploring what hap­ former and narration writer these paintings, interpreta­ painted their hedonistic pens when words are trans­ on the latter, also practices tions that invariably see displays of food and flowers lated into images, suggests what he preaches. gloomy, moralistic lessons and depictions of the human

42 ENGINEERING & SCIENCE NO. 199 7 PREHISTORIES OF THE FUTURE: THE PRIMITIVIST PROJECT AND THE CULTURE OF MODERNISM Edited by Elazar Barkan and Ronald Bush, Stanford University Press, 1995; 449 pages

Most of the 16 essays in new fascination with the contributors shift backwatd this book exploring the primitive pervades much of and dig deeper into the influence of ethnography on the literature, art, and music political, social, and racial what has become popularly of the early 20th centuty. antecedents and complexities known as modernism were The book's editors, who also of encounters with primitive originally presented at a 1991 organized the original con­ societies. Some of the essays conference jointly sponsored ference, Elazar Barkan, deal with academic anthro­ by Caltech and the Claremont associate professor of history pology, but tOpics also drama of daily domestic life, Graduate School Humanities at Claremont Graduate encompass vampires and he writes, to create images for Center. In the late 19th School (as well as director of violence, Gauguin in Tahiti, pleasure and joy-and as a century various technologies its Humanities Center and Josephine Baker in Paris, the telief from the incessant (for example, railroads, teleg­ previously instructor in influence of African American moralizing of the past. He raphy, photography) btought history at Cal tech), and music on Irving Berlin, T. S. also argues that the heroic, Western culture into closer Ronald Bush, professor of Eliot's fascination with primi­ allegorical paintings of the encounter with primitive literacure at Caltech, don't tive peoples, and the effect of time, which appear to be cultures, ushering in a pro­ follow the easiet, more heavi­ ethnographic photography's loaded with obvious high­ found alteration in how ly traveled routes through the erotic images on Victorian minded meaning, are really Westerners perceived oth­ familiat modernism terrain. morality. using Biblical and classical ers-and themselves. This Rather, they and the other themes as a front to indulge in painting nudes- and some TECHNICALLY SOUND quite erotic ones at that. Caltee h-Occi d e nta I Mandel grants that a tradi­ Concert Band tion of vanitas paintings did compact disk exist, with unambiguous, and legitimate, symbols-skulls, skelerons-of the transitori­ The Cal tech-Occidental ness of life. But, he claims, Concert Band, directed by not every snail nibbling a Bill Bing, director of Cal­ tulip petal connotes mortal­ tech's instrumental music ity, not every bird is a lewd program, has recorded its first proposition, and sometimes CD. It's loaded with such an empty shoe is only an Caltechiana as the "Centen­ empty shoe. nial Suite," written by alum­ nus Les Deutsch (BS '76, PhD '80) for Caltech's lOOth birth­ day; "Throop March," written in 1900 and "unearthed" in To HEAR OURSELVES AS OTHERS HEAR US: 1987; and a medley of unfot­ TAPE RECORDING AS A TOOL IN MUSIC PRACTICING AND TEACHING gettable songs from the by James Boyk, MMB Music, Inc., 1996,78 pages 1920s including "Lead Us On, Our Fighting Beavers," "Fight, Men of California James Boyk, Cal tech as pianist and teachet and equipment in his Caltech lab. Tech," and the "Gnome lecturer in electrical engineer­ with reflections on making Yehudi Menuhin has called Sweetheart Song" (all sans ing and music, explains his music. Among other things, the book "valuable to both lyrics, unfortunately). There own coaching techniques "for it advises us to "squint our teacher and student," and are pieces by two other local music students, teachers, ears" when listening to tape Andre Watts contributes that composers with a Caltech performers, and those who playbacks, and to dance and it's "a treasure-trove of connection (but no beavers or en j oy a peek behind the sing along. But the technical information, advice and Gnomes) and, oh yes, some scenes." Not just a technical side is not overlooked: Boyk entertaining musical insights rves, Sousa, and Mozart roo. how-to manual, the book also includes a chapter on for both amateur and profes­ The CD cao be ordered from teaches how to listen to audio systems and compo­ sional musicians .... [which] the Cal tech Bookstote (818- oneself, and it is tichly nents, giving readers the should be required reading 395-6161) for $12.95 plus illustrated with anecdotes inside scoop from his many for all lovers of music." shipping and handling. from the author's own career years testing recording D- JD

199 7 ENGINEERING & SCIENCE NO. 4' HONORS AND AWAR DS

Tom Ahrens, MS '58, Engineers. Carreira has also Neumann Medal for 1966. professor of geophysics, has received an American Chem­ H e has also received the been selected by [he Meteor­ ical Society Award in Pure 1996 Phil Kaufman Award Faculty File itical Society to receive its Chemistry ftom the Alpha from the Electron ic Design 1997 Barringer Award, which Chi Sigma Fraternity, and Automation Companies recognizes outstanding con­ an Arthur C. Cope Scholar (EDAC). tributions to the study of Award from the American John Seinfeld, rhe Louis E. impact craters and associated Chemical Society. Nohl Professor and professor phenomena. Donald Coles, PhD '53, pro­ of chemical engineering, and J acqueline Barton, profes­ fessor of aeronautics, emeritus, chair of the Division of sor of chemistry, is the re­ has been awarded the 1996 Engineering and Applied cipient of the 1997 William Otro Laporte Award by the Science, has been elected a N. Nichols Medal, p resented American Physical Society. Fellow of the American Insti­ by the Nichols Medal Jury Siobodan Cuk, PhD '77 , tute of Chemical Engineers. and the New York Section associate professor of electri­ Robert Shacp '34, MS '35, of the American Chern ieal cal engineering, has been the Robert P. Sharp Professor Society. elected a Fellow of the Insti­ of Geology, Emeritus, has J ohn Bercaw, Centennial tute of Electrical and Elec­ been selected by the Geologi­ Professor of Chemistry, has tronics Engineers. cal Society of America as been awarded an American Barbara Imperiali, associate recipient of the Distinguished Chemical Society Award for professor of chemistry, has Career Award of the Quarter­ Distinguished Service in the received a 1996 Arthur C. nary Geology and Geomor­ Advancement of Inorganic Cope Scholar Award from the phology Division. Chemistry, sponsored by American Chemical Society. Edward Stolper, t he Wil­ Mallinckrodr Baker, Inc. Wolfgang Knauss '58, liam E. Leonhard Professor Norman Brooks, PhD '54, PhD '63, professor of aero­ of Geology, and chair of the the James Irvine Professor of nautics and applied mechan­ Division of Geolog ical and Environmental and Civil ics, has been elected to the Planetary Sciences, has been Engineering, Emeritlls, has International Academy of chosen by the European been elected an Honorary Engineering (form erly the Union of Geosciences as a M ember of the American Russian or Soviet Academy joint recipient of the Arthur Society of Civil Engineeering. of Engineering). Holmes medal. Erick Carreira, associate Carver Mead '5 6, PhD '60, Edward Stone, the D avid professor of chemistry, is one the Gordon and Betty Moore Morrisroe Professor of of 60 young researchers Professor of Engineering and Physics, vice president, and named by Presidem Clinton Applied Science, was selected director of JPL, has been to receive the first annual by the Institure of Elecrrical selected by the California Presidemial Early Career and Electronics Engineers to Museum of Science and Award for Sc.ientists and receive the IEEE J ohn Von Industry Foundation to receive the International von Karman Wings Award. Peter Wyllie, professor of geology, has been invited to become a Foreign Member of Anneila Sargent has been named executive director of Caltech's Owens the Academia Europaea in the Valley Radio Observatory (OYRO) in Big Pine, California, succeeding Nick Earth & Cosmic Sciences Scoville, the Francis L. Moseley Professor of Astronomy, who has been Section. director for II years. Sargent, a senior research associate in astronomy Ahmed Zewail, the Linus Pauling Professor of Chemical who is widely known for her research into the formation and evolution of Physics and professor of stars and protoplanetary systems, has been serving as associate director physics, has received [he for millimeter-wave operations at OYRO since 1992. She earned her BS 1996 Kirkwood Medal, presented by the New Haven from the University of Edinburgh (1963) and her MS (1967) and PhD (1977) Section of the American from Caltech, where she has remained ever since. OYRO will celebrate its Chemical Society and Yale's 40th anniversary in 1998. Department of Chemistry. 0

44 ENGINEERING & SCIENCE NO. 1997 Offlce of Gift and Estate Planning

ENGINEERING THE STOCK M ARKET

Caltech electrical engineer­ year of tetirement as he did If you've enjoyed success in ing students aren't expected the yeat before at Texaco. the stock market, why not to graduate with a knowledge He continues to be assiduous consider following Maloney's of the stock matket, but about his securities, doing example and establish a Fred Maloney attributes his his own research and seldom charitable trust at Caltech? profitable investment prac­ relying on the advice of a The ad vantages to you are tices to his Caltech experi­ broker. considerable: an income tax ence. Maloney received his Maloney has supported deduction, avoidance of bachelor's deg ree in electrical Cal tech through donations to capital gains taxes when the engineering in 1935 and his the Institute's Alumni Fund trustee sells your appreciated master's in electrical eng i­ and through several chari ­ stock, broader diversification, neering a year later. He says table remainder trusts. As hi s an income to you fot life, and, that he was impressed by a stock portfolio increased in ultimately, significant graduate course in economics value, the concept of a chari­ support for the Institute. taught by Horace Gilbert, table trUSt became increas­ Contact us (see below) for Caltech's well-respected ingly important for tax more information, or ask for business economics professor purposes. our brochure that explains who died in 1990. these trUStS in greater detail. Gilbert, an efficiency expett, taught Maloney how to evaluate a company, and by following his advice, Maloney says he has been able to amass significant wealth. "If it Thomas W. Smith, CFP hadn't been for Gilbert, I Susan A. Walker, CFP probably wouldn't have done Office of Gift and Est at e Planning so well," says Maloney. California Institute of Technology Although Maloney started investing in the stock market Mail Code 105-40 as soon as be gtaduated from Pasadena, California 91125 Caltech, he devoted mOst of (818) 395-2927 his energy to his in ves tments planned ... gifts@starbase l.ultech .edu after he retired as division superintendent at Texaco. He http://www.caltech.edu/-deve lopment/gep/gep.html says he made four times as much money during his first