In This Issue vice chancellor at UC Santa found ESL, Inc., and worked Cruz. He finally returned to with Sylvania/GTE. Caltech in 1969 as chairman of An article adapted from his the Division of Engineering and address begins on page 14. Infor­ Applied Science. Clauser mation about the 1987 Research became professor emeritus in Directors Conference can be 1980. found on page 32. The computer graphics of Clauser's ideas on the cover and Jet Start in the article were created by On October 31, 1936, near the Bob Bolender, a first-year grad future site of the Jet Propulsion student in mechanical engineer­ Laboratory, the jet age was born. ing. He programmed the designs On that date, just 50 years ago, in Caltech's Engineering Design Caltech graduate student Frank Research Laboratory, which is J. Malina and his cohorts con­ under the direction of Erik ducted the first test-firing of a Antonsson, assistant professor of liquid-fueled rocket motor. mechanical engineering, who Malina went took the pictures. 011 to co-found Hot Yachts JPL with On the cover - Computer Capital Ideas Theodore von graphics compare a conventional At last year's Research Directors Karman. In 12-meter yacht (top left), such as Conference, sponsored by 1946 he left those currently competing for Caltech's Industrial Associates, Caltech for a the America's Cup in Australia, William J. Perry delivered an job with with three innovative alterna­ enthusiastically received keynote UNESCO in tives to the yacht's heavy keel, address on "Entrepreneurship Europe. In his later years he set­ which normally provides upright and Advanced tled in Paris and devoted most stability. In his article, "The Technology." of his time to art - he was a Boat That Almost Was," which Perry, who pioneer in the kinetic art move­ begins on page 2, Francis earned his MS ment and the founder of Clauser explains why solutions from Stanford Leonardo, an influential art jour­ such as a broad-beamed hull and PhD from nal. Malina died in 1981. (top right), pontoons on out­ Penn State "The Rocket Pioneers," riggers (lower left), or under­ (1957), is a which begins on page 8, is water ailerons (lower right) could former Under­ adapted from an article Malina make an America's Cup yacht secretary of Defense for Research wrote for E&S in 1968, with the go faster. and Engineering and is currently addition of some material from Clauser, the Clark Blanchard president of H&Q Technology the oral history he gave to the Millikan Professor of Engineer­ Partners, Inc., in Menlo Park, Caltech Archives. (A separate ing, Emeritus, readily admits California. Before entering article on the Archives can be that he's not a naval architect. government service he helped found on page 19.) But as a participant in the aeronautics revolution in the 1930s, he has retained an Paul Winter - President of the Alumni Association interest in applying scientific Theodore Hurwitz - Vice President for Institute Relations solutions to problems previously Robert O'Rourke - Director of Public Relations left to intuition. His reputation reached the sailing world, and he STAFF: Editor - Jane Dietrich was recruited by the Eagle syndi­ Writer - Robert Finn cate to be chief scientist in the Production Artist - Barbara Wirick Newport Harbor Yacht Club's Business Manager - Lena L6nnberg-Hickling effort to win the America's Cup. Photographer - Robert Paz Clauser earned all his degrees from Caltech, PICTURE CREDITS: Cover, 6, 7, 27 - Bob Bolender/Erik Antonsson; 2 his PhD in - Bob Thomas & Associates; 3 - Jane's All the World's Aircraft; 26- 1937 under Dan Nerney/Dot; 27,28 - Francis Clauser; Inside Front Cover - Bob Theodore von Adler; 9, 13.,- Caltech Archives; II, 12 - JPL; 16 -.United Feature Syn­ Karman. dicate, Inc.; 23 - Giuseppe Attardi; 19,22, back cover - Bob Paz After spending the war years Enl'ineering & Science (ISSN 0013-7812) is published five times a year, September, at Douglas November, January, March, and May, at the California Institute of Technology, 120 I Aircraft Com­ East California Boulevard, Pasadena, California 91125. Annual Subscription $7.50 domestic, $20.00 foreign air mail, single copies, $ 1.50. Third class postagepaid at pany, he established the aeronau­ Pasadena, California. All rights reserved. Reproduction of material contained herein tics department at the Johns forbidden without authorization. © 1986 Alumni Association California Institute of Hopkins University in 1946. He Technology. Published by the California Institute of Technology and the Alumni remained there until 1965, when Association. Telephone: 818-356-3630. ' he took the post of academic Postmaster: Send change of address to Caltech, 1-71, Pasadena, CA ? 1125. ENGNE~~~~

CALIFORNIA INSTITUTE OF TECHNOLOGY NOVEMBER 1986 VOLUME L, NUMBER 2

The Boat That Almost Was - by Francis H. Clauser Page 2 An aeronautics expert brings science to sailing in the America's Cup race.

The Rocket Pioneers - by Frank J. Malina Page 8 An eyewitness account of the birth of the rocket age in the Arroyo Seco 50 years ago.

Entrepreneurship in Advanced Technology - by William J. Perry Page 14 An industrialist expresses optimism about the sources of leadership and financing in the continuing computer revolution.

The Past Recaptured Page 19 Caltech's Archives, established 18 years ago, has become almost too successful in preserving the Institute's history.

Departments

Research in Progress Page 23 Smog - in Bags and Balances One Genome, Fully Deciphered The Boat That Almost Was

by Francis H. Clauser

NTH E LATE SUMMER OF 1984 Chuck Newton yacht design had reached a pinnacle of perfec­ ! approached me at lunch in the Athenaeum tion in this country. When the Australians and asked if I would be willing to meet a won with their winged keel, the question man who wanted to talk to me about the naturally arose in my mind: Is the design of America's Cup. That man turned out to be these yachts as advanced as I had been led to George Tooby, BS '35 - almost a classmate believe? The Australian keel wasn't that radi­ of mine - who proposed that I sign on with cal. If this was really supposed to be the state the Newport Harbor Yacht Club to bring of high technology in yacht design, I thought aeronautical science into the design of its it might be interesting to see how it looked America's Cup challenger. from an insider's point of view. I had never designed boats before, but my So I began to study the various America's interest had already been piqued by the 1983 Cup contenders over the years, particularly America's Cup race of 12-meter yachts. the 12-meter yachts that had been racing Above: Eagle, cllrrellffy com· Before that race a number of my sailor since the competition was resumed in 1958 peling among Ihe America 's friends had assured me that the United States following the hiatus of World War II. I was CliP challengers ill {he wafers was sure to win the cup again, as it had con­ immediately struck by a parallel with my off Perth. was buill by rhe Newport Harbor YacJu Club sistently for the past 132 years. These friends early experience in aeronautics. In the 1920s as their conventional12-meter said that we had the finest crews in the world airplanes were designed mostly by intuition. yachl. and that furthermore, the art of 12-meter When scientific knowledge was applied to air-

2 ENGINEERI NG & SC IENCE I NOVEMBER 1986 plane design in the 1930s, earlier intuition The Fokker Trimotor 10- frequently turned out to have been grossly passenger transport (top) was wrong. For example, pilots often complained typical of the intuitive designs of the 1920s. The Douglas that the early airplanes had problems with Globemaster (bottom) was stability and control. Science in the 1930s designed during World War showed that the tail surfaces (elevators and II. It typifies the changes rudders) had been much too small. brought about by the revolu­ As I looked at the America's Cup yachts, tion in aeronautical science during the 30s and 40s. The my immediate reaction was that they must author designed the wing and have had stability and control problems with tail sections on this airplane, their very small rudders located in the wake and he holds the patents on of both the hull and the keel. And it turned the ailerons and flaps. out that indeed a previous America's Cup skipper, Bill Ficker of Intrepid in 1970, claimed that "the present breed of 12 meters is very difficult to steer and keep 'in the groove.' Intrepid's biggest difficulties were experienced when tacking in light weather. It was not easy at all to get her moving again on the wind and to regain the speed of the previ­ ous tack." Other helmsmen said the yachts needed constant control to keep from yawing New York Yacht Club had grown almost Figure 1: The 12-meter yacht in moderate weather. Comments from the insolent about its superior position in the Intrepid was typical of the designers of these yachts make it clear that race. Australia's winning of the cup made it America's Cup contenders of the 1960s and 70s. These they didn't know how to solve this serious possible for other American yacht clubs to yachts experienced the same problem. It was at this point that I con­ dream of winning the cup. At present a total type of stability and control cluded thatthe design of 12-meter yachts of 13 yachts are vying for the opportunity to problems that airplanes of the probably had not reached the peak of challenge Australia in the waters off Perth - 1920s did. scientific perfection that everyone seemed to one each from England, Canada, and New think. Zealand, two each from France and Italy, and Ever since a group of New York mil­ six from the United States. lionaires first challenged the British yacht When George Tooby decided that it clubs in 1851, the America's Cup has been a would be appropriate for the Newport Harbor race not between nations but between yacht Yacht Club to enter a challenger in the next clubs. It was set up as a series of match races America's Cup race, he formed the money­ every three years between a single defender raising Eagle syndicate and hired Johan and a single challenger. The defenders and Valentijn as Newport Harbor's chief designer. the challengers would hold their own series of Valentijn, a young, Dutch naval architect, eliminations to choose one boat to represent had designed Liberty, the unsuccessful Ameri­ them. But there are different rules for each can defender that had lost to the Australians side. The winner of the challengers' elimina­ in the 1983 cup race, as well as an earlier tion races would represent its home yacht contender, Magic, for the New York Yacht club, while the defender must sail under the Club. In the fall of 1984 I (as chief scientist) colors of the yacht club that holds the cup - joined forces with Valentijn to build one-third until 1983 the New York Yacht Club. This size models of Liberty and Magic (the latter had been a galling experience for all the other having been purchased by the Eagle syndi­ American yacht clubs, particularly since the cate) - to be tested in the large towing tank

3 in Escondido. This was to establish a data­ Unfortunately, his conventional boat ended base of known characteristics of these two up still sailing the wrong way around in my yachts so that we could compare future opinion. models with these two well-known quantities. What is a 12-meter yacht? For an The Eagle syndicate had laid out a pro­ America's Cup designer, this always looms as gram to build two yachts, for which their a key question, since it defines the limits to fund raising goal was $8-10 million. The his innovation and creativity. As originally first of these was to be a conventional yacht, laid down, the rules were intended to })e incorporating whatever improvements could sharply and clearly defined. As the rules are be made on Magic and Liberty. The second written, two quite different lengths are was to be as radical and innovative as we involved. One of these lengths, LWL, is the were capable of making it. Valentijn and I length on the waterline from the bow to the agreed that he would spend almost all of his stern (or to the rudder post if this is farther time on the conventional yacht, and 1 would aft), when the yacht is afloat without a crew. devote most of my effort to the more radical This length has nothing to do with the 12- second yacht. meter rating but serves to limit the yacht's Early in my work with Valentijn I was displacement and draft. The displacement struck by the great difference in our percep­ must not be less than the volume given by tions of what constitutes a good streamlined shape. Both theory and experiment long ago [LWL; 0.75 r Figure 2: The airfoil or airship led aerodynamicists to conclude that the typi­ (top) is the aerodynamicist's cal airfoil or airship shape shown in Figure 2 concept of a well-streamlined all in meters. Otherwise the yacht is penal­ shape. In contrast, the typical has the least resistance. Also shown in the ized in its rating by double the deficiency. boat shape (bottom) is figure is the typical shape associated with The draft must not exceed 16 percent of the designed to travel the other boats. Any tendency to see a similarity LWL plus 0.5 meters. Here the penalty is way around. between them is dispelled when one realizes three times the excess. The second length, L, that in one case the pointed end is forward is directly involved in the 12-meter rating. It and in the other it is backward. When we enters the all-important formula tested the model of Magic, I persuaded L + 2d - F +.JS Valentijn to tow it backward as well as for­ Rated length = 12 meters = 2.37 ward. The results, shown in Figure 3, were a shocking revelation to Valentijn. 1 kidded Here d is the girth difference, F is the free­ him by saying the reason we lost the cup was board, and S is the sail area, all in meters. (I that we were sailing the wrong way around. do not know the origin of the factor 2.37; without it, these would be 28.44-meter yachts.) Figure 3: Towing tank tests on 2000 Magic's hull (without keel and The definition of L is a study in complex­ rudder) showed that it had 1800 ity. Imagine a plane 0.18 meters (7 in.) less resistance when towed above the waterplane. This defines new bow backwards rather than 1600 and stern points, and the first part of L is the forwards. distance between these points. The second is (/) 1400 0 a girth measurement from a point on one Z => side of the bow 0.6 meters above the original 0 0.. 1200 waterline, around and under the bow, and up ~ to the corresponding point on the other side. W u 1000 Z From this is subtracted 1.2 meters, and the « MAGIC f- TOWED result is multiplied by 1.5 to give the second (/) FORWARDS UJ 800 part of L. The third is a corresponding mea­ w 0:: surement at the stern, except that here the f- 600 I multiplying factor is 0.33. The forward girth ~ 0:: measurement (when multiplied by 1.5) has a 0.. => 400 minimum value of 0.54 meters, and if the actual value is less than this, the minimum 200 value must be used. The aft girth measure­ ment similarly has a minimum value of 0.4 0 4 II meters. In practice most yachts have mea­ surements close to these minimums.

4 ENGINEERING & SCIENCE / NOVEMBER 1986 The girth difference measurement, d in water as it moves along the wetted surface the formula, is. taken at the keel. It's the and from the eddies off the stern that are difference in length between a chain following created when the streamlines fail to close in along the hull from a point "a" 'above the behind the ship. These resistances increase waterline down along the keel toa point 1.5 roughly as the square of the ship's speed. meters below the waterline, and the length of At higher speeds wave resistance also a chain stretched tight between these same comes strongly into play. At first many small two points (Figure 4). The freeboard, F, is an waves form; as the speed increases, the waves average distance from the deck to the water increase in both length and height, and the plane. Since it is subtracted in the formula, it pattern becomes less complex. Finally there encourages the designer to build high decks. comes a speed for which there is simply a There is a maximum value of 1.21 meters wave crest at the bow, a trough along the permitted for F, however, and most designers midship and a final crest at the stern. use this maximum value.· When L, d, and F Spreading out from this wave pattern at the have been determined, the sail area must then ship itself is a great train of waves extending 1.5 METERS be chosen so as to give the yacht a 12-meter out in a chevron to the rear. At higher rating, that is, to fit the formula. If we add speeds this wave train carries off large the requirements that the beam must be at amounts of energy, resulting in a large 1 least 3.6 meters, the height of the mast must increase in resistance of the ship. This wave not exceed 25 meters, and the height of the resistance can become so great that it over­ jib must not exceed 18.75 meters, we have shadows the frictional resistance (Figure 5). the basic answer to the question: What is a It was an Englishman, William Froude, 12-meter yacht? who in the last century showed that the wave There are other subtle but important re­ pattern of a ship is governed by the dimen­ Figure 4: The chain midgirth strictions, however. One is that there be no sionless ratio V/.Jg[, that is, velocity divided measurement is the difference in length between a chain run­ hollows in the hull between the waterline and by the square root of the acceleration of grav­ ning from point "an along the the deckline except in the region near the ity times length. This important result hull to a point 1.5 meters rudder - a rule that excludes catamarans enables us to predict the wave resistance of below the waterline and a and other multi-hull craft. A second restric­ large ships from tests on smaller scale models. chain stretched tight between tion states that below 1.7 meters under the Another Englishman, Osborne Reynolds, also those points. water plane, no width can exceed 3.6 meters. working in the last century, showed that fluid This was adopted to limit the span of the frictional resistance is governed by the dimen­ wings on Australian-type keels. sionless ratio VL/v, that is, velocity times A third restriction requires construction in length divided by kinematic viscosity. Using accordance with the "scantlings" established by Lloyd's Register of Shipping. These are an 2000 Figure 5 shows results from archaic set of construction specifications that towing tank tests on a one­ almost completely rule out significant struc­ 1800 third-scale model of the 12- tural innovations. In effect they require that, meter yacht Liberty. Note regardless of a 12-meter yacht's total displace­ 1600 how wave formation causes the total resistance to rise ment, the hull minus its keel must weigh (j) 0 rapidly at higher speeds. approximately 17,000 Ibs. LL ~ 1400 00 And finally, for 12 meters competing in wD- 1200 TOTAL the America's Cup races, there is an addi­ ~~ RESISTANCE INCLUDING (j)f--~>- tional rule that requires the length of the _tl:: WAVE (j)W 1000 RESISTANCE yacht at a plane 50 mm (2 in.) above the wen water plane to be at least 44 ft. long. Q::::J f--f-- II 800 Given all the variations that are possible, <9u why is it that almost all 12-meter yachts g:>! ~w 600 weigh close to 57,000 Ibs. and have an overall I length of 66 ft. and a waterline length of 46 f-- ft.? To figure out why, we must first explore 400 some of the fundamental facts of hydro­ dynamics so that we can understand why a 200 good yacht goes fast. When a ship travels at 0 low speeds, the principal resistance it 4 II encounters is from the skin friction of the

5 2000 will be pushed right up into this high speed range. So it would seem that designers of 1800 1986-87 America's Cup yachts should strive

1600 for the greatest length they can get. And "'0z indeed, the first models that Valentijn and I => 14 00 built after our benchmark tests were larger, a.0 ;;; heavier ve rsions of Liberty. w 1200 But if we look back at the 12-meter rules, L> «Z we see that as the size of the hull (length, >- 1000 girth, and freeboard) goes up, the area of the "'iii w DISPLACEMEN sail must come down. For a typically propor­ 800 OF "'>- 96,000 tioned hull, this leads to the relationship of I POUND S 600 sail area to length shown in Figure 7. a. '""'=> Clearly, under the 12-meter rule the 96,000- 400 lb. yacht with its 54.8 ft. of length would be required to have a much smaller sail. This Figllfl' 6: Towing tank data 24,000 POUNDS 200 result seems to indicate that, instead of a hare been scaled liP fO show Ihe effeci oJsi=e on the resis· longer, larger hull to gain the advantage of taIlC(~ oflypical12-meler 4 5 6 7 8 9 10 high speed, the designer should be aiming for )'a£.'llIs. SPEED I N KNOTS " the smallest boat possible to have the greatest amount of sail area. this result we can scale up the fluid friction of The hull must, however, be constructed in a model to that of the full-sized ship. accordance with Lloyd's scantlings, which The first question that faces the 12-meter specify that the hull structure (minus keel) designer is: How heavy should the boat be? If shall weigh no less than approximately 17 ,000 Liberty were to be built in three sizes, 24,000, lbs. regardless of the displacement of the 57,000, and 96,000 Ibs. , with waterline yacht. The 57 ,000-lb. yacht will thus be able lengths of 33.6, 45.6 and 54.8 ft. respectively, to have 40,000 Ibs. of lead in her keel, while how would their resistances compare? Using the 24,000-lb. boat can have only 7,000 Ibs. our towing tank data and the scaling laws In a 12 meter the principal factor that enables described above, we can calculate the resis­ the yacht to stand up in heavy winds is the tance curves for these three yachts (Figure 6). enormous amount of lead ballast down in the At low speeds the lightest boat, being smaller, naturally has the least resistance. But at higher speeds the picture changes. For the longer boats the rapid rise in wave resistance FigW"e 7: Under Il·meler does nOl begin until proportionally higher mles. as the hllll of a com'en· speeds are reached. Consequently their total liollal yacht iI/creases in si=e, resistance is lower at high speeds, where they tlie sail area mllst be enjoy a superiority that progressively gets decreased. Th e lead bal/ast permil1cd /01" IIpright stability better with length. goes lip wit/t lel/gth, leading ru In Perth the winds are expected to be (/ desigll dilemma. strong, and in most of the races the yachts

3000 60,000 f- U) W w 0z "- 50,000 => 0 w LEAD BALLAST n. ~ 2000 40,000 f- => U) 0

0 20 40 60 80 WATERLINE LENGTH IN FEET

6 ~. N(;INt . ElU N C I\: SCIENCE / NOVEMBER IH!!f; keel. In fact, all 12 meters have keels much spite of the success of this idea, it doesn't larger than hydrodynamics alone would dic­ seem to have occurred to 12-meter designers tate, simply to house the lead. Our tank tests to explore this avenue. showed that the great advantage of the Aus­ What would a 12 meter look like as an tralian winged keel lay not in the hydro­ ultra-light? The conventional 12 meter with dynamic properties of the wings themselves, a gross weight of 57 ,000 Ibs. has a righting but in the fact that the shape of the keel and moment of about 160,000 ft.-lbs. when it is the wings permitted the center of gravity of heeled 30°. The keel with its 40,000 Ibs. of the lead to be significantly lower. lead provides about 145,000 ft.-lbs. of this. Figure 7 also shows the amount of lead The remainder is made up of an unstable ballast permitted in 12-meter yachts of vari­ moment of about 38,000 ft.-lbs. from the ous sizes. This now puts the designer's task in hull, mast, rigging, and sails, and a stable sharp focus. If he designs a small hull, it can hydrostatic moment of about 53,000 have large sails, but even light winds will ft.-Ibs. provided by a waterline beam of blow the boat over because of its small right­ approximately 11ft. ing moment. In contrast, a large hull will An ultra-light with a gross weight of have a very large upright stability, but it will 24,000 Ibs. (less than half that of the conven­ be permitted to use only small sails. These tional 12 meter) would have a length of 33.6 opposing constraints dictate that most 12- ft. compared to the 45.6 ft. of the conven­ meter yachts end up with waterline lengths of tional hull. To comply with the America's 43 to 48 ft. and with gross weights of 50,000 Cup rule, it would have to have a length of to 65 ,000 lbs. 44 ft. at a height 2 inches above the waterline. It was in the fall of 1985 that the design of So it would have long, nearly horizontal Valentijn's conventional yacht Eagle, was overhangs both fore and aft. completed, and my ideas about a radically One purpose of designing an ultra-light is new boat began to crystallize. In interna­ to be able to use much greater sail area. My tional offshore racing a new class of fast calculations indicated that to be able to stand ultra-light boats has been sweeping the field. up in the wind, it would need 40 percent How do they do it? There are several ways of more righting moment, that is 160,000 x gaining upright stability other than by placing lAO, or 224,000 ft.-lbs. when it heels 30°. lead in the keel , and ultra-lights do it by using Here the righting moment is to come from a broad beams with shallow-draft hulls. In broader beam, rather than from a heavy keel. This will require a waterline beam of 20 ft. rather than the conventional II ft. Being much lighter and broader of beam, the draft of the lightweight hull at 1.5 ft. will be much Figure 8 (fa r !eJi) shows a con­ shall ower than the usual 4.5 ft. Entering the ventional 12 meier in com ­ dimensions of such an ultra-light hull into the piller graphics creared by Bob 12-meter formula gives a sail area of approxi­ Bolender. Figllre 9 (to its mately 2,450 sq. ft. , which is much greater right) depicts a broad-beam ed. than the typical 1,750 sq. fl. for 12 meters. lightweight hI/II, which is per­ milled to carry a m ilch larger Referring back to Figure 6, we see that the sail. If can pla ne like a li ghtweight hull has the advantage of sur/board. significantly less resistance at low speeds, and the larger sail area would much more than make up for its greater resistance at higher speeds. It has been found that lightweight, broad­ beamed boats can readily get up and plane like a surfboard on the forward face of a wave. They frequently reach speeds of 14 to 18 knots. The reason for this is simple. For a planing hull the resistance, instead of rising sharply at high speeds as shown in earlier

conlinucd on page 26

7 The Rocket Pioneers_

by Frank J. Malina

The late Frank J. Malina inaugurated the jet age 50 years ago with the construction and test-firing of the first liquid-fueled rocket motor in the Arroyo Seco. In this article, adapted from one he wrote for E&S in 1968, he describes those exciting early days of rocketry at Caltech.

y INTEREST IN SPACE EXPLORATION was first mind turned more and more to the possibili­ Maroused when I read Jules Verne's De fa ties of rocket propulsion while we analyzed Terre a fa Lune in the Czech language as a the characteristics of propellers. boy of 12 in Czechoslovakia, where my fam­ In March 1935 at one of the weekly GAL­ ily lived from 1920 to 1925. On our return CIT seminars, William Bollay, then a gradu­ to Texas I followed reports on rocket work ate assistant under von Karman, reviewed the which appeared from time to time in popular possibilities of a rocket-powered aircraft based magazmes. upon a paper published in December 1934 by In 1934 I received a scholarship to study Eugen Sanger, who was then working in mechanical engineering at Caltech. Before Vienna. The following October Bollay gave a the end of my first year there I began part­ lecture on the subject before the Institute of time work as a member of the crew of the the Aeronautical Sciences in Los Angeles. GALCIT (Guggenheim Aeronautical Labora­ Local newspapers reported on Bollay's lec­ tory, California Institute of Technology) ture, which resulted in attracting to GALCIT 10-foot wind tunnel. This led to my appoint­ two rocket enthusiasts - John W. Parsons ment in 1935 as a graduate assistant in and Edward S. Forman. Parsons was a self­ GALCIT. trained chemist who, although he lacked the The Guggenheim laboratory at this time, a discipline of a formal higher education, had few years after its founding, was recognized as an uninhibited and fruitful imagination. He one of the world centers of aeronautical loved poetry and the exotic aspects of life. instruction and research. Under the leader­ Forman, a skilled mechanic, had been w()rk­ ship of Theodore von Karman, GALCIT spe­ ing for some time with Parsons on powder cialized in aerodynamics, fluid mechanics, rockets. They wanted to build a liquid­ and structures. Von Karman's senior staff propellant rocket motor but found that they included Clark B. Millikan, Ernest E. Sechler, lacked adequate technical and financial and Arthur L. Klein. The laboratory was resources for the task. They hoped to find already carrying out studies on the problems help at Caltech. They were sent to me, and of high-speed flight, and the limits of the that was the beginning of the story which led propeller-engine propulsion system for air­ to the establishment of the Jet Propulsion craft were beginning to be clearly recognized. Laboratory. In 1935-36 William W. Jenney and I con­ We reviewed the literature published by ducted experiments with model propellers in the first generation of space flight pioneers - the wind tunnel for our master's theses. My Ziolkowsky, Goddard, Esnault-Pelterie, and

8 . ENGINEERING & SCIENCE / NOVEMBER 1986 Oberth. In scientific circles this literature was We therefore set as our initial program the generally regarded as science fiction, primarily following: (a) theoretical studies of the ther­ because the gap 'between the experimental modynamical problems of the reaction princi­ demonstration of rocket-engine capabilities ple and of the flight performance require­ and the actual requirements of rocket propul­ ments of a sounding rocket; and (b) elemen­ sion for space flight was so fantastically great. tary experiments of liquid- and solid­ This negative attitude extended to rocket pro­ propellant rocket engines to determine the pulsion itself, in spite of the fact that God­ problems to. be met in making accurate sta­ dard realistically faced the situation by decid­ tistical tests. This approach was in the spirit ing to apply this type of propulsion to a vehi­ of von Karman's teaching. He always cle for carrying instruments to altitudes in stressed the importance of getting as clear an excess of those that can be reached by bal­ understanding as possible of the fundamental loons - an application for an engine of physical principles of a problem before initiat­ much more modest performance. ing experiments in a purely empirical man­ We concluded from our review of the ner, which can be very expensive in both time existing information on rocket-engine design and money. that it was not possible to design an engine to Parsons and Forman were not too pleased meet specified performance requirements for with an austere program that did not include a sounding rocket to surpass the altitudes at least the launching of model rockets. They attainable by balloons. After much argu­ could not resist the temptation of firing some ment, we decided that until someone could models with black powder motors during the design a workable engine with a reasonable next three years. Their attitude is symp­ specific impulse there was no point in devot­ tomatic of the anxiety of pioneers of new ing effort to the design of the rocket shell, technological developments. In order to propellant supply, stabilizer, launching obtain support for their dreams, they are method, and payload parachute. under pressure to demonstrate them before

This photo, taken in the Arroyo Seco in 1936, shows the first liquid-fueled rocket motor and the men who built it. The man at the lefi is unidentified. The others are, lefi to right, Apollo M.O. Smith, Frank J. Malina, Edward S. Forman, and John W. Parsons.

9 they can be technically accomplished. Thus motor. there were during this period attempts to The group heard with excitement that make rocket flights which were doomed to be Robert H. Goddard would come to Caltech disappointing and which made support even in August to visit Robert Millikan. Millikan more difficult to obtain. arranged for me to have a short discussion The undertaking we had set for ourselves with Goddard on August 28, during which I required, at a minimum, informal permission told him of our hopes and research plans. I from Caltech and from the Guggenheim also arranged to visit him at Roswell, New laboratory before we could begin. In March I Mexico, the next month, when I was going proposed to Clark Millikan that my thesis be for a holiday to my parents' home in Bren­ devoted to studies of the problems of rocket ham, Texas. propulsion and of sounding rocket flight per­ Both Dr. and Mrs. Goddard received me formance. He was, however, dubious about cordially. My day with him consisted of a the future of rocket propulsion and suggested tour of his shop (where I was not shown any I should, instead, take one of the many components of his sounding rocket), a drive engineering positions available in the aircraft to his launching range to see his launching industry at that time. His advice was no tower and 2,OOO-lb.-thrust static test stand, doubt also influenced by the fact that GAL­ and a general discussion during and after CIT was not then carrying out any research lunch. He did not wish to give any technical on aircraft power plants. Later he supported details of his current work beyond that which our work. he had published in his 1936 Smithsonian I knew that my hopes rested finally with Institution report, with which I was already von Karman. He was at this time studying familiar. This report was of a very general the aerodynamics of aircraft at high speeds nature and of limited usefulness to serious and was well aware of the need for a propul­ students. sion system which would surmount the limi­ The impression I obtained was that God­ tations of the engine-propeller combination. dard felt that rockets were his private After considering my proposals for a few preserve, so that any others working on them days, von Karman agreed to them and gave took on the aspect of intruders. He did not permission for Parsons and Forman to work appear to realize that in other countries there with me, even though they were neither stu­ were men who had arrived, independently of dents nor on the staff at Caltech. This deci­ him, at the same basic ideas for rocket pro­ sion was typical of his unorthodox attitude pulsion, as so frequently happens in within the academic world. He pointed out, technology. however, that he could not find funds. Von Karman in his autobiography, The During the next three years we received Wind and Beyond, writes: no pay for our work, and during the first year I believe Goddard became bitter in we bought equipment - some secondhand his later years because he had no real - with whatever money we could pool success with rockets, while Aerojet­ together. Most of our work was done on General Corporation and other organ­ weekends or at night. izations were making an industry out We began our experiments with the con­ of them. There is no direct line from struction of an uncooled rocket motor similar Goddard to present-day rocketry. He in design to one that had been previously is on a branch that died. He was an inventive man and had a good tried by the American Rocket Society. For scientific foundation, but he was not a propellants we chose gaseous oxygen and creator of science, and he took himself methyl alcohol. too seriously. If he had taken others Our work in the spring of 1936 attracted into his confidence, I think he would two GALCIT graduate students, A.M.O. have developed workable high-altitude Smith and Hsue Shen Tsien. Smith was rockets and his achievements would working on his master's degree in aeronautics; have been greater than they were. But Tsien, who became one of the outstanding not listening to, or communicating pupils of von Karman, was working on his with, other qualified people hindered doctorate. Smith and I began a theoretical his accomplishments. analysis of flight performance of a sounding On October 29, 1936, the first try of the rocket, while Tsien and I began studies of the portable test equipment was made for the thermodynamic problems of the rocket gaseous oxygen-methyl alcohol rocket motor

10 ENGINEERING & SCIENCE I NOVEMBER 1986 PRCT£CTIVE. PR~~!iUR£' /BA.JUtIK.R. A:>R RttGoVl.A."TOR Oe5&.RvE.RS

0, ZOOO/rt

5c~f.M"TIC DIAGRAM O~ PRCP~~f'O ROCK£"T" MOTOR Te.S"T Se'"ft;P "0"'_ ROCKET' Pltopul.S'Vi!! I.lNIT.5 Rf~loII"-.Lh ~L~I.NIII.'M ~V'TlC"1. L.AeoItA~V CALIFORNIA IN.S"T11V'£ OF" ~~J\IO ... OC,Y' DR,I,I','N e.o, RvNI< ..J. MALltolt.>. OAt.£. ' AUCiU~i 2BIJ93~ in the area of the Arroyo Seco back of Devil's Gate Dam on the western edge of Pasadena - a stone's throw from the present-day Jet Propulsion Laboratory. I learned several years later from Clarence N. Hickman that he and Goddard had conducted smokeless­ powder armament rocket experiments at this same location during Wodd War I. On October 31 we tested the rocket motor itself. The next day I wrote home as follows: This has been a very busy week. We made our first test on the rocket motor yesterday. It is almost incon­ ceivable how much there is to be done and thought of to make as simple a test as we made. We have been think­ ing about it for about six months now, although we had to get all the equipment together in two days, not by choice, but because there are classes, and hours in the wind tunnel to be spent. Friday we drove back and forth to Los Angeles picking up pressure tanks, fittings, and instru­ movie cameras for recording the The rocket motor fires. ments. Saturday morning at 3:30 a.m. action ofthe motor. Bill Bollay and we felt the setup was along far enough his wife also came to watch from to go home and snatch three hours of behind the dump. sleep. At 9 a.m. an Institute truck Very many things happened that took our heaviest parts to the Arroyo, will teach us what to do next time. about three miles above the Rose The most excitement took place on Bowl, where we found an idealloca­ the last "shot" when the oxygen hose tion. Besides Parsons and me, there for some reason ignited and swung were two students working in the around on the ground, 40 feet from N.Y.A. working for us. It was 1 p.m. us. We all tore out across the country before all our holes were dug, sand­ wondering if our check valves would bags filled, and equipment checked. work. Unfortunately, Carlos and By then Carlos Wood and Rockefeller Rocky had to leave just before this had arrived with two of the box type "shot" so that we have no record on

11 film of what happened. As a whole with the question, "How do we open a fund the test was successful. at Caltech for our project?" he was We made a number of tests with this flabbergasted. transportable setup, the last one on January When von Karman gave the group per­ 16, 1937, when the motor ran for 44 seconds mission to make small-scale experiments of at a chamber pressure of 75 lbs. per square rocket motors at GALCIT, we decided to inch. mount a motor and propellant supply on a In March 1937 Smith and I completed bob of a 50-foot ballistic pendulum,-using the our analysis of the flight performance of a deflection of the pendulum to measure thrust. constant-thrust sounding rocket. The results The pendulum was suspended from the third were so encouraging that our project obtained floor of the laboratory with the bob in the from von Karman the continued moral sup­ basement. We planned to make tests with port of GALCIT. We were authorized to various oxidizer-fuel combinations. conduct small-scale rocket motor tests in the We selected the combination of methyl laboratory. This permitted us to reduce the alcohol and nitrogen dioxide for our initial time we wasted putting up and down the try. Our first mishap occurred when Smith transportable equipment we had used in the and I were trying to get a quantity of the Arroyo Seco. Von Karman also asked me to nitrogen dioxide from a cylinder we had give a report on the results of our first year's placed on the lawn in front of Gates Chemis­ work at the GALCIT seminar at the end of try Laboratory. The valve on the cylinder April. jammed, causing a fountain of the corrosive The unexpected result of the seminar was liquid to erupt all over the laWn. This left a the offer of the first financial support for our brown patch there for several weeks, to the project. Weld Arnold, then an assistant in irritation of the gardener. the astrophysical laboratory at Caltech, came When we finally tried an experiment with to me and said that in return for being per­ the motor on the pendulum, there was a mis­ mitted to work with our group as a photogra­ fire. The result was that a cloud of nitrogen pher he would make a contribution of dioxide-alcohol mixture permeated most of $1,000. His offer was accepted with alacrity, GALCIT, leaving behind a thin layer of rust for our project was destitute. on much of the permanent equipment of the In 1942 the Jet Propulsion Arnold, who commuted the five miles laboratory. We were thrown out of the build­ laboratory consisted ofjust a betweeen Glendale and Caltech by bicycle, ing the next day, of course. few buildings in the Arroyo brought the first $100 for our project in one­ Then we built what we called the gas Seeo, close to the site of the dollar and five-dollar bills in a bundle apparatus on the outside of Guggenheim. It original rocket-motor firing. must have been about 1939 when that blew This location is at the eastern wrapped in newspaper. We never learned edge of the grounds of how he had accumulated them. When I up. It's quite possible that I might have been today's JPL. placed the bundle on Clark Millikan's desk done in by that explosion, but von Karman had called his secretary and asked if I would bring him a typewriter at home. I hopped in my Model A Ford, put the typewriter in it and drove to his house. When I came back I saw many people standing around. As I came closer and closer to the end of the building, I began to see pieces of the apparatus on the ground, and I realized some­ thing terrible had happened. Fortunately, neither Parsons nor Forman was hurt; they were shaken up a bit. But where I had been sitting before I left, a piece of a pressure gauge had blown right across where my head would have been and buried itself in a piece of wood. It was at this time that we began to be known on the campus as the "suicide squad." From the beginning the work of the group on rocket research at GALCIT attracted the attention of newspapers and popular scientific

12 ENGINEERING & SCIENCE I NOVEMBER 1986 journals. Since our work was not then sion, and (5) Research program for develop­ classified as "secret," we were not averse. to ing jet propulsion. discussing with journalists our plans and The word "rocket" was still in such bad results. There were times that we were repute in "serious" scientific circles at this abashed by the sensational interpretations time that it was felt advisable by von Karman given of our work, for we tended to be, if and myself to follow the precedent of the anything, too conservative in our estimates of Army Air Corps of dropping the use of the its implications. word. It die!. not return to our vocabulary The fact that our work was having a real until several years later, by which time the impact in America came from two sources. word "jet" had become part of the name of In May 1938 von Karman had received an our Laboratory (JPL) and of the Aerojet­ inkling that the U.S. Army Air Corps was get­ General Corporation. ting interested in rocket propulsion. I presented my report to the committee on Then, in August 1938, Ruben Fleet, the December 28, 1938, and shortly thereafter the president of the Consolidated Aircraft Co. of Academy accepted von Karman's offer to San Diego, approached GALCIT for informa­ study with our GALCIT rocket research tion on the possibility of using rockets for group the problem of the assisted takeoff of assisting the takeoff of large aircraft, especially aircraft on the basis of available information, flying boats. I went to San Diego to discuss and to prepare a proposal for a research pro­ the matter and prepared a report entitled gram. A sum of $1,000 was provided for this "The Rocket Motor and its Application as an work. Auxiliary to the Power Plants of Conven­ Parsons and Forman were delighted when tional Aircraft." I concluded that the rocket I returned from Washington with the news engine was particularly adaptable for assisting that the work we had done during the past the takeoff of aircraft, ascending to operating three years was to be rewarded by govern­ altitude, and reaching high speeds. The Con­ ment financial support and that von Karman solidated Aircraft Co. appears to have been would join us as director of the program. We the first American commercial organization could even expect to be paid for doing our to recognize the potential importance of rocket research! rocket-assisted aircraft takeoff. But in October Thus in 1939 the GALCIT Rocket 1938 a senior officer of the U.S. Army Ord­ Research Project became the Air Corps Jet nance Division paid a visit to Caltech and Propulsion Research Project. In 1944 I The first practical application informed our group that on the basis of the prepared a proposal for the creation of a sec­ of Malina's rocket work came Army's experience with rockets he thought tion of jet propulsion within the division of sometime between August 6 and 23, .1941 with the first there was little possibility of using them for engineering at Caltech. It was decided that it jet-assisted takeoff of an air­ military purposes! It was not until 1943 that would be premature to do so. Instead, von plane. Malina is the man liquid-propellant rocket engines, constructed Karman and I founded JPL. 0 furthest to the right. at Aerojet-General Corporation, were tested in a Consolidated Aircraft flying boat on San Diego Bay. In December 1938 I was informed by von Karman, Robert Millikan, and Max Mason that I was to go to Washington, D.C., to give expert information to the National Academy of Sciences Committee on Army Air Corps Research. One of the subjects on which General H.A. Arnold, then Commanding General of the Army Air Corps, asked the Academy to give advice was the possible use of rockets for the assisted takeoff of heavily loaded aircraft. I prepared a report which contained the fol­ lowing parts: (1) Fundamental concepts, (2) Classification of types of jet propulsors, (3) Possible applications of jet propulsion in connection with heavier-than-air craft, (4) Present state of development of jet propul-

13 panies but of entire new industries. In fact, this technological revolution has provided the underpinnings for an economic revolution. In the 1960s and 1970s, at least, the United States not only provided leadership of this economic revolution but was also the primary Entrepreneurship beneficiary. If you look at the economic consequences, you might say that in the 19th century the new wealth in the United States In Advanced came from the gold in our mountains; for the last few decades our wealth has come from the silicon in our valleys. Technology We would like, of course, to extrapolate this exciting past into the future, but some prophets have said that's not going to happen. Some postulate that the rapid pace of techno­ logical innovation, which has driven these by William J. Perry economic changes forward, has flattened out and that the innovative phase of this revolu­ tion is over. Others believe that the techno­ logical revolution will continue, but the leadership of it will pass to other countries - E LIVE IN AN AGE that I believe historians to Japan or even to the Soviet Union - and Wwill call the Age of the Computer. The that the U.S. will end up as a second-class most profound technological revolution technological power. mankind has ever experienced is occurring Will this technological and economic right now, but because we are immersed in revolution continue? I believe that, not only this age we take it for granted. The computer will this revolution continue, it will has transformed the way we work; it has accelerate. To characterize this by a number, transformed our companies, our industries, I would say that in the next decade we can our economy, and our defense posture. expect an improvement of about a hundred Indeed, it is in the process of transforming times in price-performance of computers. In society more dramatically and more rapidly the transportation field a hundred-fold that the Industrial Revolution did in the 19th improvement in performance (speed) century. This technological revolution is represented a change from the horse and causing the same sort of turmoil and confu­ buggy to jet aircraft. In the same field, a sion in our ability to project the future as was hundred-fold improvement in price would the case with social and political revolutions require reducing the price of a $10,000 auto­ of an earlier day. mobile to $100. Charles Babbage invented the "analytical I think that the increase in density in engine" in England more than 150 years ago, integrated circuits, which has characterized but the computer was never realized in this price/performance improvement in the Babbage's lifetime. The enabling technologies last few decades, will continue for at least had to be invented before the real power of another 10 years. That is, we'll be going from the computer could be unleashed - first, the geometries in integrated circuits of a few invention of electricity, then the invention of microns to a few tenths of a micron; this ten­ the transistor, and finally the invention of the fold compression in linear dimension will integrated circuit 130 years later. result in about a hundred-fold increase in If the computer is the engine that drives density. To the extent that the history of this this new technological revolution, the industry is a valid predictor of the future, this integrated circuits provide the fuel for that increase in density will allow for approxi­ engine. That fuel has led to price/perfor­ mately a hundred-fold decrease in price per mance improvements in the computer of bit or per transistor. more than 20 percent per year for the last few This will require a whole new set of ena­ decades. Such rapid and steady improvement bling technologies, not the least of which will is unprecedented and has led to the develop­ be a whole new class of lithographic and etch­ ment not only of new products and new com- ing equipment. Those technologies are well

14 ENGINEERING & SCIENCE / NOVEMBER 1986 in hand, and it will be a matter of a relatively z 300 few years before .they are commercially intro­ (!) en duced. This continuing compression of the w density of integrated circuits, however, will c o/J lead to what has been commonly called Z "Moore's dilemma." Moore's dilemma states o i= 200 that the more transistors you put on a chip, Z ii: the harder it becomes to design it. w • The solution to Moore's dilemma is the C IX development of very sophisticated design o u.. tools that can be used not only by profes­ J: sional integrated circuit designers but by sys­ I"-Z 100 tems engineers as well. This technology was o ~ pioneered at Caltech by Carver Mead, the I oZ Gordon and Betty Moore Professor of Com­ (/) IX puter Science. But it leads us to a new W dilemma: As the number and specialization 0.. of computers proliferates, the next choke 1970 1975 1980 point comes in software. YEAR The solution to that problem - the development of design tools for writing ers, and that number will probably double in Gordon Moore expressed the software - is also under way but is not as far a few years. General Motors has contributed dilemma posed by the increas­ advanced as design tools for integrated cir­ five computers to my garage - two in one ing complexity oj integrated circuits: as it became possible cuits. A whole new industry is forming for car and three in the other. I never thought I to put more transistors on a companies that are building software design was buying a computer when I bought those chip, the design time increased tools. In certain classes of problems three- to cars, but there they are. General Electric has at an exponential rate. five-fold improvement in productivity has contributed three computers to my kitchen. already been demonstrated, and I think that There again I didn't know I was buying them. by the end of this decade we will see software My sprinkler system has two computers, and design tools that will allow an order of magni­ my hi-fi system (TVs, VCRs, and compact tude improvement in productivity in writing disk) has five. In my office I have two "real" code. computers, one sitting on a desk and another New architectures are also being designed that I carry around with me in my suitcase. for computers. After several decades of com­ The other half of this revolution consists puters based on von Neumann architecture, of products that ride up the performance we are now seeing a veritable explosion in curve. Among the applications that will be concurrent computers, or parallel processors, possible with IOO-times improvement in per­ an area in which Caltech has been a pioneer. formance are image processing, expert sys­ The economic revolution - the applica­ tems, and - perhaps most dramatic - simu­ tion of these new advantages to the develop­ lation. Weare already at the state in the ment of products - follows the technological design of integrated circuits, for example, revolution. The products include not only where simulation plays a crucial role in the new computers themselves but also a wide design process. No one would think today of range of other goods that can be made more designing a very large scale integrated circuit efficient and effective by embedding comput­ without the benefit of a computer to do the ers in them. We will be riding this price/­ simulation. That same process is going to be performance curve in two directions. Many applied to the design of missiles, automobiles, applications will ride the price curve down­ tanks, and airplanes. Where are we going to ward. Falling costs will lead to a proliferation find the leadership for all of this technological of small, embedded computers in the home, innovation? the office, the factory, and the automobile. To answer that question I want to go back Some pundits have criticized earlier fore­ in history and ask where we found the leader­ casts of the increase of computers in the ship for the last phase of this revolution - home and office as overblown. They have where the leadership for the development of already been proven wrong. In my own integrated circuits came from and why. My house recently I went around room by room authority on the subject is a British engineer, and counted computers. I have 17 comput- G. W. Dummer. Dummer is the man who

15 almost invented the integrated circuit. At a emulated, as a part share in the company's technical conference in 1952, several years prosperity gives an increased sense of respon­ before the invention of the integrated circuit, sibility. Successful businesses are almost Dummer said: "With the advent of the always dependent on a few people who are transistor and the work in semiconductors innovative and enthusiastic." generally, it seems now possible to envisage Dummer is pointing out the critical electronic equipment in a solid block with no importance of the entrepreneurial spirit and layers of insulating, conducting, rectifying, the availability of risk capital in the -U .S. As and amplifying materials, the electrical func­ a case in point: If a chief engineer in a major tions being connected directly by cutting out company in Europe or Japan left his job to areas of the various layers." You don't have start up a new company to follow up an; to be an IC designer to understand that Dum­ innovative idea, his friends and co-workers mer was describing the integrated circuit. He would think that there was something wrong not only described it, but he set out vigorous­ with him. In the U.S. if a chief engineer did ly to try to develop it. He did this with the not leave his job to follow up an innovative idea, his friends and co-workers would think .---_.JJOLLAR$ AND NON$EN$.a..:...'_" __ something was wrong with him. A difference in culture leads people to take action in one country that they would not take in another. These cultural traits do not change quickly. As far as I can tell from sitting near the fringes of the venture capital industry, that innovative spirit is as vigorous today as it was five years ago. I think that there are even more bright, enthusiastic people bringing for­ ward interesting ideas than was true five or ten years ago. As for the availability of risk capital, any­ one who tries to form a company soon finds out that banks and public stock are not useful or available sources for such capital. Neither "Yep, she's gonna cost ya-your of these institutions is created for the purpose microprocessor's shot." of providing risk capital for inventors. There are, however, other sources. A funding tech­ full support and cooperation of the British nique that was popular five years ago, and I government. think is still an appropriate form, is the R&D But it was not Dummer who invented the limited partnership. But it's currently out of integrated circuit, nor was England its pri­ favor and may not return to popularity for a mary beneficiary. The IC was invented by number of years, because a few very large Jack Kilby of Texas Instruments and Bob companies (Trilogy and STC Computer in Noyce of Fairchild, and the consequence of particular) bombed out and took with them that development occurring in the United about $50-100 million of R&D partnership States has been profound. investment. Not surprisingly, that had a chil­ Dummer, years later, looking back wist­ ling effect on investors, and it's not likely that fully at this missed opportunity, said: "It is we'll see many more R&D partnerships start worth remembering that American electronic technology companies. companies were formed since the war by a "Bootstrapping" - that is, doing it with relatively few enterprising electronics your own funds - is a time-honored way of engineers, setting up with either their own financing new companies in this country. I capital or risk capital from the bank. Often a would like to suggest that it is a greatly government contract would start them off. underrated technique. It's the way that I used Hard work was necessary and the large home in 1964 for my own company, ESL. It never market was a great asset, but the climate of had outside investment, never had venture innovation was such that any advanced capital, and is now a company with over technical product could be sold. The Ameri­ $200 million annual revenue. It was also the can system of encouraging employees to hold technique used by two engineers named Bill shares in the company is one which should be Hewlett and Dave Packard when they started

16 ENGINEERING & SCIENCE / NOVEMBER 1986 their company, which is now a multibillion­ italdid not flow into start-up companies in dollar enterprise. So it can be done, and it those countries. Both of those countries are has been done successfully. Th~ company trying to change that now. founder has to accept a much slower growth In the last year or so there has been much than he could achieve if someone else were gloomy talk about how venture funds have pouring money into the enterprise, but that's fallen off and how people with bright ideas not all bad. It forces the entrepreneur to can't get venture money to start new com­ avoid the pitfalls that he would inevitably panies anymore. This talk represents a gross face with rapid growth of a new company. misunderstanding of what the real situation The most popular form of risk capital in is. First, today there is well over $10 billion companies today is venture capital. This has in venture funds available for investment. undergone a remarkable transformation in That is plenty of money to invest relative to the last seven to eight years. During the any standard, and particularly relative to the 1970s there was about $100-200 million a standards of the 1970s. Two factors have year of venture capital flowing into the ven­ caused this atmosphere of reluctance on the ture pool, most· of what was available for new part of venture capitalists to invest. First, the company starts. This underwent a dramatic over-the-counter market was depressed for change in 1979. From then on there has high technology stocks, and therefore the ini­ been $2-4 billion a year of new money com­ tial public offerings almost completely dried ing into the field - in other words, an order up for a year and a half. With the absence of of magnitude increase in the investment in initial public offerings, venture companies this area. didn't have any way to graduate from a It would require a separate article to private to a public company, so their inves­ explain what really precipitated that rapid tors didn't have any way to achieve liquidity. change, but certainly the change in the capital When their companies needed a second or gains tax- an effective rate of 20 percent, third round of financing, and there was no down from 50 percent - was a primary con­ public market out there to provide it, they tributor to that growth. More important, and had to turn back to the original investors for less easy to analyze, is the fact that those ven­ it. So what has happened is that while the ture funds that were established in the 1970s amount of money flowing into venture com­ and were investing in companies during this panies is actually greater in the last few years period turned out to be fantastically success­ than previously, most of it is going into ful. Many of them were showing 40 percent second- and third-round financing rather than per year compounded growth rate return on into start-ups. So it indeed has been true that those investments, and that attracted the there has been less money for start-ups but attention of institutional investors. So it was not because of diminishing money in the ven­ a combination of several factors (primarily ture capital pool. these two) that pulled out the throttle and The situation obviously will change caused this surge of new money to come into rapidly as soon as the initial public offerings the venture capital business. start up again. And there's every indication Another related and very important factor that this new wave of public offerings is that is often neglected in discussions of this already under way. To the extent that hap­ situation is the existence of the public over­ pens, there will be a dramatic transformation the-counter market. Although I have already again in the use of funds that are now in this said that the public market was not a useful venture capital pool, and a much higher pro­ vehicle for a start-up company, its existence is portion of them will be eager to fund start-up critical, because it provides the mechanism by companies. which venture investors and entrepreneurs There's another, relatively new, source of eventually realize liquidity in their invest­ risk capital funds for start-up companies - ments. The vigorous over-the-counter market funds from large corporations. So far I've that exists in this country has often provided been talking about innovation through start­ the liquidity for these investors in five to up or emerging companies. But the amount seven years at 10 times their original invest­ of money that IBM invests in its R&D pro­ ment, thereby making these investments very gram in a given year is about the same order attractive. In Germany and Japan, for exam­ of magnitude as all of the venture money ple, no such healthy over-the-counter market going into start-up companies in a year. So has existed, which is one reason that risk cap- why don't these innovative developments all

17 come out of a company like IBM? Why do companies. I believe that when the statistics we depend on start-ups at all for this on this are available they will demonstrate a innovation? high ratio of failure. The problem with the The large companies are effective in using acquisition of small companies is that the their R&D funds to develop products that large company snuffs out the very flame it is evolve from predecessor products, but they trying to capture. Realizing this, some com­ are not very effective in using their funds to panies have switched over to the venture cap­ take an innovative technology that leads to a ital business~ Exxon and General Electric are new product - something that really breaks cases in point. My opinion is that these ven­ ground with predecessors. The size and the tures will turn out to be unsuccessful as well. bureaucratic nature of a large company is The companies may be successful as vent4re simply incompatible with seizing a new idea capitalists, depending on whom they have and taking it rapidly to the marketplace. running the activity, but what they're really There is also another much more subtle prob­ trying to achieve - a transfer of technology lem - what I call the "liability of and a head start on new products - will not leadership. " come this easily. I know this from personal experience. At In the past few years a number of com­ the time that the transistor was being com­ panies have tried an innovative approach mercialized, I worked for Sylvania Electric known as corporate partnering, or strategic Products, one of the three world leaders in partnering, as an alternative to acquisition or the production of vacuum tubes. The venturing. In this arrangement the large research director clearly saw that the transis­ company forms a business relationship with tor was the successor product to the vacuum the small company that usually involves some tube, so he launched an energetic R&D pro­ technology transfer, and it also makes a gram. Then the president assigned responsi­ minority investment in the company. A bility for the commercialization of the transis­ number of companies have been experiment­ tor to the manager of the vacuum tube divi­ ing in one way or another with the technique sion. The rest is history. in the last few years - IBM, General Motors, The psychological problems, not to men­ Eastman Kodak, Lockheed, Rockwell, TRW. tion the very real financial problems, of a In each case what they are trying to do is to manager vigorously introducing a new get access to innovative applications of tech­ product that kills off the product that is his nology and rapid seizure of new product bread and butter are not to be underesti­ opportunities. The benefits to the small com­ mated. A similar situation happened when pany are an access to risk capital, to a broad IBM stood by and watched Digital Equip­ base of technology, and to markets and credi­ ment invent the minicomputer and capture bility. Basically this marriage combines the that market. This was not because IBM mass of a large company with the velocity of didn't know how to build a minicomputer. the small company to provide momentum for But had they successfully produced a mini­ both companies to move forward. computer, it would have cut the legs off the In summary, I believe that the technologi­ low end of their mainframe (a very profitable .cal revolution that has taken place these last part of their line), and they just couldn't quite few decades not only will continue through bring themselves to do that. Somewhat later this decade but will in all probability accel­ Digital Equipment did the same thing with erate in both its technical and economic man­ the personal computer market. Ironically, by ifestations. Second, tbe culture - the that time IBM saw the opportunity develop­ entrepreneurial spirit - that underlies this ing for the personal computer, and, since it revolution in the United States is alive and didn't compete with their mainframes, was well and will continue to support leadership able to go in and make a major impact on in technological innovation in this country. that market. Third, the large pool of risk capital that has More and more of the large companies been formed during the last seven to eight facing this problem are asking themselves the years will maintain itself and will be available question: What can we do to participate in in ever increasing quantities to fund these these venture or start-up activities? They are start-ups; the risk capital will come both from concluding that if they can't beat them they'll the conventional venture funds and from a join them. There have been a number of transfer of funds from large corporations to attempts to participate by acquiring small small corporations. 0

18 ENGINEERING & SCIENCE / NOVEMBER 1986 The Past Recaptured

Judith Goodstein (foreground), Paula Hurwitz (center) and Carol Buge inspect the recently delivered boxes of the HEN CHARLES RICHTER died last year, he left had already begun to organize the Millikan papers offormer Caltech Whis papers - correspondence, unpub­ and Hale collections, also getting the first President Harold Brown - lished writings, course notes, and technical crack at their contents. (This resulted in stacked on the floor in the reports - to Caltech. The collection from Kevles's prizewinning book, The Physicists, Archives' basement hallway. the inventor of the well-known earthquake scale was packed into dozens of cartons and delivered to the Caltech Archives, and then left in the hallway of the Millikan Library basement. There was no other place to put it. The next morning Archivist Judith Good­ stein and her assistant Carol Buge discovered to their horror that the janitor had tossed out six or seven boxes, thinking they were trash. Buge tracked down the garbage truck driver, who held up unloading his cargo until the Archives rescue team could reach the city dump. Fortunately the driver knew his route across campus and was able to determine approximately where in his truck the library trash might be located. "Then he dumped it inch by inch in front of us," remembers Goodstein. They managed to recover some but not all of the Richter cartons. "It broke my heart," says the archivist, even though she had figured that something like this was bound to happen sooner or later. The Archives had long since outgrown its space. Although Goodstein has managed to seize some qcdjacent space over the years, the center of the Archives is still the same room that she walked into in 1968 as Caltech's first archi­ vist. The room contained a map case, a filing cabinet full of medals, a bunch of boxes wrapped in brown paper - and the papers of Robert A. Millikan and George Ellery Hale. Theodore von Karman's papers arrived a week later. Daniel J. Kevles, now professor of history,

19 published in 1978. Scholars since have of the leading institutions in the world and as tended to avoid the Hale papers, presuming, a community of scholars, Caltech has an obli­ A 1926 letter from Einstein to falsely, that it has all been done, says Good­ gation ... to preserve the record of its life." Paul Epstein, then professor of stein.) Kevles played an important role in The record of Caltech's life had already theoretical physics at Caltech, urges him to try to arrange an getting Caltech's Archives established. In a been pretty well preserved, chiefly through the academic post in America for memo to then-provost Robert Christy, he efforts of Roger Stanton, director of the Insti­ a young Jewish colleague, who wrote: "The principal reason [for maintain­ tute libraries from 1948 to 1963. Stanton was denied a teaching post in ing an archive] is that the study of history was careful to save all the papers and files Berlin for "political" reasons. rests heavily upon the records of institutions given to him over the years. The creation of He concludes with the com­ ment that everyone there was and of the activities of its members .... the Archives provided a place to put these caught up in SchrMinger's Given the central importance of science in records, as well as a person to organize and new theory of quantum states. the 20th century, it would seem that, as one make them available to the scholarly world. The boxes that Goodstein found when she arrived had been brought from various store­ PreuSische krli. .'1t1. JZ 19.?& rooms around campus to this "dumping Akodemle der Wissenscnilften IW 7. Uwtft ... u....le ground" in the basement of the new Millikan Library when it was built in 1967. "What I inherited was everybody's little attic, plus Hale and Millikan," comments Goodstein. Opening up the boxes and sifting through d4~{J~~~ the papers was a heady experience. From Amos Throop's death mask to letters from eU.~. 'Jz.:~/ t/~ ~~ Einstein, from long-lost photographs to the ~~~~~-I~~ personal notes of Thomas Hunt Morgan - ~~'--'~~~. Goodstein found herself spellbound. The Caltech cache and its documentation of the ~·~~~~·~1~ rise of science and technology as a dominant ~~,:..-.~-..Ie.vr. ~ 4-~ influence in the 20th-century exerted a con­ ~p. ~~~~ siderable influence on her own studies. ~I ~.P'A-L ~ ~ ~~ Although her PhD work at the University of JI.~::tJ?~ _ ~ ~ ~"" Washington dealt with the history of science in the 1800s, after arriving at Caltech she 1'" ~.~~ ~..h/~­ "never again looked back to the 19th cen­ ./!...... L...... ~ "...... _~~:...... -L..- fI~J tury." Goodstein is currently writing a book ~~ ~....r--:h...-.. r L-4- ..I/2...t:..z-- ~~- ttl; on Caltech's history. ~k.../~. The immediate task at hand was catalog­ ~4c. --'.-..4 t!l:-.~ ~ ~ ing 96 manuscript boxes (containing 76,000 ~/ .L.~--'~ ~f ~ ~- documents and covering 144 linear feet) of r0-J~(h."",:-' ~ "'"""~#- f ------l ~~ Theodore von Karman's papers, left to Cal­ ./~~~~..:-~ tech in his will. But Goodstein lost no timein i£.....----.:7':~ :I...f~"<-~ ~b..tt- laying the groundwork for future donations. ~. d~:~_ 7k/T. 'i'...f../"-..u. t~, ';/..... "".~3-<~ She sent a memo to all faculty members and

.....,...,:....,...t.. ..4!t-.. ~";;t ,'h~:....c..t..--L.F~, ~!..L .... :-...4...-- administrators raising their historical con­ ,... #J .h, r-...... :...,..4...... :k.. ~~,.,...~--~I' sciousness and urging them not to be hasty in throwing things out. "Students of the history ""f;:...... -:e~ r- /l'L...... - __ r';.. of 20th-century science turn to unpublished M,,- 4..:...t.-. ~..<>(...... Lh '; r' C • .u-~ primary sources ... when they want to ~ ~r~tJ6..."'f~~~ unravel the genesis of a particular idea, or ___ ~. ::1.- .--<,/~ £A how a particular research field grew, or how r- u..- b.bt ,,"-- r- ~4-f ---- knowledge of it was communicated within ~~ -:.....r-~:r.;;«.t-" ~~- and outside of the scientific community." r"L.~ de., ~. Personal records that she considered worth c.t.a ,...... ,.--...r ~-L ~ preserving included "preliminary drafts of published papers, privately circulated mem­ ~ oranda, data books, records of experiments 4 (the unsuccessfuJ ones, too), original drawings or photographs of equipment ...."

20 ENGINEERING & SCIENCE / NOVEMBER 1986 Her plea was heeded only too well. There are more than .100 manuscript collections (70 of which have guides prepared py Goodstein L(]IU(~ctlOllS in and her staff) splitting the Archives' seams, and now Goodstein reluctantly has to turn material away. The one missing collection she coveted was Linus Pauling's; he gave it to Oregon State, his alma mater - including Walter Adams the Caltech chemistry division papers from the years Pauling was chairman. Such divi­ sion records provide a wealth of information. "You can get a whole new perspective on the history of biology," says Goodstein, "in the Division of Biology papers between 1928 and 1955, the years of Thomas Hunt Morgan and George Beadle. And it's not just budgets ..." Sometimes the art of the archivist has demanded more than just sitting in the library basement waiting for the stuff to roll in. Frank Malina was an avid collector of primary source material relating to Caltech's rocket research project and the founding of the Jet Propulsion Laboratory (see article on page 8). When he left his collection to the Library of Congress on his death in 1981, Prints, and Slides Goodstein flew to Paris to survey the papers Arthur Il"YTnm,,1 and subsequently shipped 302 kilos of paper Charles F. Howard P. Robertson back to Caltech for microfilming before relin­ quishing it to Washington. Not content with collections of papers, Bruce H. Rule which might not tap the details and nuances of important events, Goodstein also began to mine people's memories. She started the oral Sei.sIDl()lo!4~ [Vller... !"-',,, Publications*

history project in 1978, and it now consists of L. Sin:~heimer nearly 70 transcribed interviews with people William II:. Smythe who were on the scene during Caltech's for­ Royal W Sorensen mative years. (Portions of a number of the Chester Stock oral histories have been published in E&S.) H. Since many of these world-famous scientists Alfred H. Sturtevant led other lives before (and after) Caltech, their stories encompass a much broader scientific and political arena than the Pasadena campus. Richard C. This is what makes the Caltech Archives Albert Tyler such a great resource of intellectual history - Anlhonie Van Harreveld Viden Cassettes including its collection does not just concern Caltech. Ditch activities Most other places are far more parochial and The 1\'le"·h~,"i",,l hence far less interesting, but here "these guys sat at the center of the storm," says Good­ stein. Caltech people corresponded with just about ev_eryone who was anyone in 20th­ Famest C. 'Vatson century science. And science wasn't the only thing they wrote about. For example, the horrors of the rise of Nazi Germany and fas­ Charles C. Lauritsen Vost cist Italy can be read in some of the scientists' Thomas Lauritsen Laszlo Zechmeister letters. Paul Epstein's papers contain un­ opened letters from relatives in Europe plead-

21 seismograms on 276,000 individual photo­ graphic sheets) available to the scientific com­ munity as part of an international earthquake data bank. It took the Archives staff 3If2 years to film and catalog the records between 1923 and 1962, which provide a continuous long­ term chronicle of local and distant earth­ quakes over much of the period for which instrumental data exist. Earthquake data over long periods of time are important in under­ standing seismicity rates a.nd changes - in order to assess current risks, for one thing. Financial support for this project came from the U.S. Geological Survey and the National Oceanic and Atmospheric Administration. It's always nice to have grants, as every scientist knows, but they don't solve all budget problems. So Goodstein thought up a more enterprising form of financing to counter budget cuts in the early 1970s - Among the more peculiar ing for help in getting out; Epstein felt so marketing the now well-known photograph of items in the Archives collec­ helpless and depressed that he couldn't read Einstein on a bicycle. The poster, distributed tion is the death mask of them. "I opened some of the letters," admits by the Smithsonian Institution and several Amos Throop, Caltech's other outlets, has netted about $40,000, founder, here staring out of Goodstein. "I didn't feel so good about that." the yellowed newspaper in The intermingling of science and politics which has helped fund many of the oral his­ which it was wrapped. can also be read in the Archives. For exam­ tory projects. Is the original photograph itself ple, Hale's attempt to create the International in the Caltech Archives? "Actually, I think I Research Council after World War I, which first saw it in E&S," admits Goodstein. foundered on the enmity of the scientists of About 500 researchers, a steadily increas­ the Allied nations as well as on Woodrow ing number, now come to use the Archives Wilson's political commitment to the League every year and there are numerous other of Nations, is well documented in Hale's requests to the staff for information and pho­ letters. tographs. In the early 1970s most Archives Other scientists' personal papers reveal users were Caltech faculty; now almost all somewhat off-beat personality traits. (Trivia come from outside, most of them scholars question: Which famous Caltech professor (with an occasional "respectable" journalist) frequented nudist camps?) A donor can place and most of them in history of science. "By access restrictions on any material of a and large we're pretty open," says Goodstein, private or controversial nature, and many do. "but you have to have a legitimate reason." But "whether your interest is prurient, or Goodstein has lately noticed a significant more academic, in science or politics, it's all switch in users' interests. Last year for the here," says Goodstein. first time more historians consulted the Grants have enabled the Archives staff papers of biologists, biophysicists, and geneti­ (which includes, besides Buge, Assistant cists than those of physicists, mathematicians, Archivist Paula Hurwitz) to make the Caltech and chemists. Most popular were the collec­ collection more available to scholars. A grant tions of Max Delbriick, Thomas Hunt Mor­ from the National Endowment for the gan, and George Beadle, in that order. Humanities helped process the von Karman Researchers using the Archives are con­ and DuBridge papers, and the Smithsonian signed to two desks in the hallway - the Air and Space Museum financed publication same catch-all hallway from which the of von Karman's papers in microfiche. Hale's Richter cartons disappeared. Goodstein has documents had earlier been published in been trying to get the work ~rea improved for microfilm under a grant from the National the past 16 years. The day may come, how­ Historical Publications Commission. ever, when this space too is completely eaten Another large microfilming project made up by more cartons of documents awaiting the earthquake records of the Kresge Seismo­ either cataloging '- or an errant garbage logical Laboratory (including single-copy truck.o - JD

22 ENGINEERING & SCIENCE / NOVEMBER 1986 Research in Progress synthesizing factories that translate messenger RNA (mRNA). These mito­ chondrial ribosomes are scaled-down versions of the ribosomes used in the cytoplasm for the same purpose. And mtDNA codes for 22 different tRNAs. These are the molecules that bring the One Genome, Fully Deciphered 20 different amino acids to the lengthening polypeptide chain as a protein is actually being assembled on the mRNA template. INETEEN YEARS of concentrated the proto-mitochondrion became its At first, this number of tRNAs was N labor have had their final payoff: host's partner, providing quantities of something of a puzzle - the cyto­ this year Giuseppe Attardi, the Grace energy in the form of adenosine tri­ plasm requires 32 tRNAs because in C. Steele Professor of Molecular Biol­ phosphate (ATP) in return for a pro­ the universal genetic code a single ogy, and his research group have com­ tected environment. Eventually, after amino acid is often coded for by two pleted the functional identification of hundreds of millions of years of evolu­ or more nucleotide triplets, and each all the genes in human mitochondrial tion, most mitochondrial genes tRNA can recognize only one or two DNA (mtDNA). In the course of this became incorporated into the host's triplets. Although it initially appeared work they've uncovered quite a few genetic machinery. In mammalian possible that mtDNA-coded proteins surprises, including the highly efficient cells this continued until the mito­ might simply lack several amino acids mode in which information is packed chondrial genome was a ghost of its or that the missing tRNAs were into the genome, a mode that Attardi former self - it contains just 16,500 imported, Attardi and others were able calls "a lesson in economy." base pairs in a circle 5 micrometers in to discard these two proposed solutions In most genetic systems, including circumference. Because of this limited to the discrepancy. The actual expla­ the chromosomes in the nucleus of coding capacity, the mitochondrion nation is that mitochondria use a human cells, the great majority of must import at least 95 percent of the simplified version of the genetic code, DNA does not actually code for any proteins it needs to function. one in which (among other differences) gene product. There are great stretches Much of the content of mtDNA is each of several amino acids is coded of "nonsense" DNA in the spaces devoted to coding for the mitochon­ for by four triplets that can be recog­ between genes, and even within most drion's own protein synthesizing nized by a single tRNA. eukaryotic genes there are long non­ machinery. With some help from the Apart from the different RNA coding regions called introns. Not so nucleo-cytoplasmic compartment of species coded for by mtDNA, it in mtDNA. Except for one short the cell, mitochondria produce their became obvious, when the DNA region, which anchors the mtDNA to own ribosomes - the protein- sequence was determined in F. San- the inner mitochondrial membrane and is important in initiating transcrip­ tion, every part of the mitochondrial genome codes for some product, either a protein, or a ribosomal RNA (rRNA), or a transfer RNA (tRNA). Also, in most genetic systems only one strand of the double-stranded DNA molecule in a given segment codes for gene products. In some cases genes can be found on both strands, but normally when this happens one strand is transcribed in one segment and the other strand is transcribed in the next segment. Human mtDNA is unique in that both strands are com­ pletely transcribed. But even with this extreme degree of information compression, mito­ chondria (the cell's power plants) are far from autonomous. Attardi thinks that this may not always have been the case. Like many biologists, he believes that the mitochondrion was once an This high·resolution electron micrograph shows one molecule of human mitochondrial DNA. independent cell that long ago The molecule forms a circle 5 micrometers in circumference, and it contains about 16,500 mounted an invasion of another cell. base pairs. The dark knob at bottom center represents a residue of the inner mitochondrial In a process known as endosymbiosis, membrane, to which the DNA molecule is aI/ached.

23 ger's laboratory in England, that finally been identified, Attardi claims the mitochondrial genome codes for that he has not put himself out of 13 different proteins. It proved rela­ business. "Our effort now goes in the tively easy to identify six of these pro­ direction of studying how the expres­ teins because of their amino-acid sion of this genome is regulated. sequence homology to known proteins. These are very intriguing problems." Four turned out to be proteins impor­ And Attardi also hopes to study a tant in the respiratory chain - a series number of genetic diseases that seem of electron carriers that transfer to to result from damage to the mito­ oxygen the electrons released during chondrial genome. "There is a whole the final stages of the oxidation of food group of diseases called mitochondrial molecules. (These four proteins are myopathies, which mostly affect the cytochrome b and three subunits of muscular system but sometimes also cytochrome c oxidase.) The fifth and affect the nervous system and the sixth proteins were found to be two heart. All these myopathies are subunits of ATPase, the enzyme that characterized by alterations in mito­ uses the energy derived from the oxi­ chondria - both structural alterations dation of food molecules to synthesize and functional alterations. And the ATP. enzyme complex that is most fre­ This left seven "unidentified read­ quently altered is NADH ing frames" (URFs) in mtDNA - 60 dehydrogenase. " percent of its protein coding capacity. This brings Attardi back to his ori­ Although it was possible that the ginal motivation for working with URFs were nonsense sequences, human cells rather than with yeast or Attardi and his colleagues were able to Neurospora, both of which are more reject that hypothesis for two reasons. amenable to genetic analysis. "I'm an For one thing, these sequences were M.D. I'm interested in man not only highly conserved in different mam­ as an example of a mammalian species malian species, something that never or a higher eukaryotic cell, but also as happens with nonsense DNA. For an organism for which there are medi­ another, the researchers had clear evi­ cal problems, practical problems to be dence that the URFs were expressed solved." D - RF - that is, transcribed into mRNA, which in turn is translated into pro­ tein. But, says Attardi, determining the functions of these proteins "appeared to be a terrible job. I Smog- thought that a possible shortcut would be to show that these proteins were In Bags and Balances somehow related to each other, that they belonged to the same complex." This indeed turned out to be the URING AN AVERAGE Los Angeles chemical engineering, has developed a case. Two of Attardi's postdocs, Anne D morning rush hour, thousands of detailed computer model of Los Chomyn and Paolo Mariottini, collab­ tons of pollutants from hundreds of Angeles smog. orating with Russell Doolittle's group thousands of cars join thousandii of "You could, in principle, use this at UC San Diego, and using antibodies tons of pollutants from industrial model to predict tomorrow's smog," raised to the proteins encoded by indi­ smokestacks. As the prevailing winds says Seinfeld. "The only problem is vidual URFs, were able to show that carry these compounds in a generally the expense. You would need a super­ each of these antibodies under certain westerly direction, the sun causes them .computer more or less dedicated to conditions precipitated six of the pro­ to begin reacting with each other. this use. The chemical mechanisms of teins produced by the URFs, indicat­ These chemical reactions produce smog involve 100 to 150 reactions. ing that they were all part of the same ozone, nitric and sulfuric acid, and Then on top of that you've got the enzyme complex. Further investiga­ many other things as well. Some of particle formation process,. which is of tion showed that this enzyme complex these condense to form suspended comparable complexity. Our long­ is another important component of the aerosol particles, which cause the haze term goal is to have such a model on a respiratory chain - NADH dehydro­ that we call smog. computer that the Air Quality genase. Finally, this year the remain­ Understanding this process is a job Management DistIjct could use." ing URF was shown to encode another every bit as complex as predicting However, ~'a computer model like subunit of this same enzyme, a finding tomorrow's weather, but John Seinfeld this is only as good as the fundamental that will appear soon in the journal and his colleagues are very close to chemistry and physics in it," notes Science. achieving this goal. Seinfeld, the Louis Seinfeld. In particular, the formation Even though the last URF has E. Nohl Professor and professor of and composition of aerosol particles

24 ENGINEERING & SCIENCE / NOVEMBER 1986 and thus its composition. Determining the particle's size is fairly straightforward - they simply turn off the charge on the electrodes and let gravity take over briefly. By measuring how long the particle takes to fall a predetermined distance, a computer can automatically calculate its size. Sageev has recently developed an ingenious way to determine the infrared spectrum of an aerosol parti­ cle. Normally, to determine an infrared spectrum, light of various infrared wavelengths is shined through a sample of liquid or gas. When the wavelength of the light matches a par­ ticular molecular resonance, the mole­ cules absorb the light and this absorp­ tion is inferred by detectors that see a decrease in light transmission through the sample. But with such a small par­ ticle the amount of light absorbed is infinitesimal, so measuring it is impossible. Instead, the researchers take advan­ tage of the fact that when the droplet On the roof of the Keck Laboratory is a large Teflon bag that Caltech researchers fill with absorbs some infrared energy, it heats hydrocarbons and nitrogen oxides to simulate early morning air. Sunlight drives the reac­ up and a tiny bit of the liquid evap­ tions that produce smog, which is analyzed by instruments housed in the structure in the orates. Less than a single layer of center of this photograph. atoms on the surface of the particle evaporates, but the laser light scattered are still only poorly understood. In an existing particles. Nucleation occurs by the particle changes enough to sig­ effort to learn more about atmospheric when the partial pressure of a gas gets nal even this minuscule change in size. aerosol, Seinfeld and Richard Flagan, so high that aerosol particles precipi­ The device slowly steps through the professor of environmental engineering tate spontaneously from the saturated infrared wavelengths, measuring size science and mechanical engineering, air. It had been thought that nuclea­ changes with each step, a process that together with graduate students Mark tion played an insignificant role in takes about one hour. Cohen, Carol Jones, Gideon Sageev atmospheric aerosol formation, since The researchers plan to use the and Jennifer Stern, are conducting condensation will continually occur device to look at the composition of both large-scale and small-scale when the atmosphere is full of particles particles containing sulfates. Such par­ experiments. on which gases can condense. But ticles are a common by-product of To study large-scale smog Jennifer Stern has found that nuclea­ industries that burn coal. This is an phenomena, the researchers have tion can and does occur at a significant important question since Carol Jones installed a smog chamber - a room­ rate. This happens when the concen­ has shown that the composition of sized Teflon bag - on the roof of the tration of a gas increases at such a rate these particles changes with time, espe­ Keck Laboratory. They fill the bag at that it exceeds the capacity of the con­ cially if they contain small quantities night with precisely controlled concen­ densation process. of catalysts such as manganese. trations of hydrocarbons and nitrogen The small-scale studies of aerosols As the results of these experiments oxides to simulate early morning air, are conducted at a very small scale come in, they will add detail to and then they cover the bag with a tar­ indeed - that of individual aerosol Seinfeld's numerical model of smog. paulin. The next morning they particles. In a variant of the Millikan It's particularly appropriate that these remove the tarp, and the sunlight oil-drop experiment, the researchers developments have happened at Cal­ begins driving the chemical reactions. use a device called an electrodynamic tech, since the late Arie Haagen-Smit Periodically they withdraw air through balance to levitate a single charged virtually invented smog research here. ports in the bag and measure gas con­ droplet, just five micrometers or so in Notes Seinfeld, "Around 1950 Arie centrations and particle numbers and diameter, between hyperbolic elec­ Haagen-Smit really figured out the SIzes. trodes. Once the particle is in place, relationship between oxides of nitrogen One important question that they can determine its precise size and, and ozone - a critical step in under­ remains to be answered is the extent to in a most remarkable recent accom­ standing the system. In a sense my which aerosol particles are formed by plishment, Gideon Sageev has mea­ research is a natural progression from nucleation or by condensation on pre- sured the particle's infrared spectrum what Haagen-Smit started." 0 - RF

25 America's CLIP contenders in 1983. Liberty (top) and Aus­ tral ia 11. illustrate [he rIIl/ning room [lUll they require when bOlh hare mainsails and spin­ nakers 0111 .

The Boat That Almost Was continued./i'om page 7

FigllYl) 10 indicme.\' Ihal Ihe upright resis­ Imlce of a pklllillg hull (\\'ilh a gross weight 0/24.000Ibs.) levels off at high speeds. This lI'i/l permil il 10 surf 011 larger 11'Q\'es.

'''''' """

"""

."" '''''

~ 6 e 'iI 10 II ' 2 13 ,. 15 SPEED IN KNOTS figures. levels off as indicated in Figure lead ba ll ast. If the pontoons are out of (and far greater than provided by the 10. Although sails cannot provide the the water when the boat is in the conventional lead keel). Perhaps even several thousand pounds of thrust to upright position (all 12-meter measure­ more important. the pontoons would cause a hull 10 plane. ocean waves fre­ ments are taken in the upright posi­ prevent the yacht rrom heeling more quently have slopes of 20 percent or ti on). they would not be classed as than about 15°. This would be a more, and for a 24,000-lb. yacht, this extra "hulls." In fact , they can be significant advantage, since a great deal gives a forward thrust of 4,800 Ibs. or lying on th e deck when the yacht is of sa il power of a conventional yacht more - quite enough to cause it to measured. And if the outriggers are is lost when it's heeled as much as 30°. plane. deployed above the deck level. they Pontoons could also be changed just as Besid es a wide beam, there are would be classed as "booms" and thus sails are changed. When light winds other wa ys to provide upright stability not in violation of the "hollows in the are anticipated. small pontoons with for an ultra-light hull. When a sail­ hull" rul e. less resistance co uld be used, and boat is running before the wind. it has Ir the pontoons take the place of extra-large pontoons could be substi­ the mainsa il out on one side and the lead ballast, the wetted area of the keel tuted in heavy winds. spinnaker out on the other. On a 12 can be reduced by about 140 sq. ft. If The required upright stability for meter each of these sweeps out to a one of the pontoons has this much ultra-lights ca n also be provided by di stance of about 35 ft. on each side of wetted area, then using an airship "ailerons" ex tending out 35 ft. on each the boat. This establishes the "running shape. such a pontoon could have a side of the boat. Each of these long room" that must be permitted under buoyancy of 6.000 Ibs. When deployed thin wi ngs could be rotated about its the rules. on a boom 35 ft. long, such a pontoon ax is so as to provide hydrodynamic This suggests the possibility of would give a righting moment of hee ling moments when the yacht is deploying pontoons on outriggers 230,000 ft.-Ibs. , an even greater under way. acting in the same way as within this running room in order to moment than provided by the wide ailerons on an airplane. Since almost obtain upright stability instead of using beam of our earlier li ghtweight hull all of the 12·meter measurements are

26 f..NG 1N t:[R1NG& SC1ENCE/NOVEMBm 1986 Figure 11: Pontoons on oU/riggers (farlejr) can provide IIpriglll stabililY and also keep Ihe yacht from heeling more Ilion J5·. The IIllderlt'arer "ailerons" in Figllre 12 (to the immediate left) can also pro­ vide upright stability. Such SOlllliolls as broad beams, POI/tOOIlS. and ailerons render a heavy keellillnecessary and make an /litra-light hull feasible. Be/ow: Later this last summer the owhor had Ihis 40- percellt-scale model buill olld tested to explore/lirther the potentialities of the underwa­ ter proboscis. taken at or above the waterline, there is nothing to prevent this. In fact, an official interpretation of the rule says that the hull may be of any shape below 150 mm (6 in.) below the water­ line as long as it does not exceed the length between the fore and aft points 180 mm above the waterline. If these wings, made of carbon fiber composites, have chords of 12 in. and thicknesses of 2.5 in., they can support the loads that are required to keep the yacht from heeling under the force of the sail. Such wings would have about 140 sq. fl. of wetted area, about the same amount that would be saved if the lead were removed from the keel. So the fluid friction would be approximately the same as that of the larger keel, but with the very great give the designer great freedom in the waves from the sphere start off savings in weight that would make the shaping the portion of the hull that is with a large trough. If this trough ultra-light hull possible. 6 in. or more below the waterline sug­ could be positioned to counteract the Ailerons would have a significant gests still another avenue to explore: crest, they would cancel each other, advantage in heavy winds. As wind Is there a way of making a boat with a and the wave resistance would be strength increases, the forces go up as small waterline length have a much nearly eliminated. It turned out the square of the speed. With a con­ larger wavemaking length? indeed that by placing a bulb under­ ventional keel , the heeling of the yacht We often see supertankers and water out ahead of the ship's bow and increases rapidly with increasing winds, other commercial vessels with bulbous joining it smoothly into th e hull lines, and once it has reached 30·, the sail bows that are designed to reduce their the bow waves could be dramatically area must be decreased (reefed) to wave resistance at a specific speed. reduced. prevent the yacht from heeling farther. The reasoning that leads to the bul­ One problem was that the interac­ As a result of the decreased sail area, bous bow concept is relatively simple. tion of the two wave systems was the yacht's spced no longer increases The bow of a ship CUlling through the altered by a change in the ship's speed. with in creasing wind speed. In con­ water creates a wave crest, which ini­ A bulb that would produce a favorable trast, the power of the ailerons goes up tiates the train of waves that spread interaction at one speed could produce with the square of the boat's speed, out behind. It is known that if a unfavorable interactions at other enabling the yacht to stand up straight sphere is towed just under the surface speeds. For commercial vessels that even in the strongest of winds. So the of the water, there is a slight upwelling spend their time traveling at one yacht could take full advantage of ahead of the sphere, but a large trough speed. this was an acceptable restric­ higher winds without sacrificing sail is created immediately behind it. This tion. But for sailboats, which travel at area and thus speed, as shown in Fig­ also initiates a train of waves that a wide variety of speeds, bulbous bows ure 13 on the next page. spread out behind it. But where the have appeared in the past to have lillie The fact that the 12-meter rules ship's waves start off with a large crest, chance of success.

27 Earlier we have seen that it is the underwater proboscis can thus extend light weight (24,000 Ibs. vs. 57,000 Froude number, V/ $, that governs far out ahead of the bow without lbs.) and the long underwater proboscis the wave patterns of ships. Commer­ increasing the measured length. And it lead to a hull whose slenderness ratio cial vessels typically travel at Froude can provide a significant addition to is about half that of the conventional numbers between 0.25 and 0.30. In the displacement so that the hull itself hull. This implies that the wave resis­ contrast, 12-meter yachts in moderate can be greatly slenderized. tance will be reduced to a small frac­ to heavy winds travel at Froude Let us consider again an ultra-light tion of that of the conventional hull. numbers of 0.37 to 0.45. At these with a gross weight of 24,000 lbs. and So much for theory and computer cal­ higher values the wave pattern is a waterline length of 33.6 ft. Again, culations. How will such a hull really simpler. This should make the design using nearly horizontal bow and stern perform? of a 12-meter bulbous bow simpler overhangs, we can meet the America's When all of this was coming into and should permit it, if properly Cup rule for a minimum length of 44 focus in my mind in November 1985, designed, to be effective over a wider ft. at a plane 2 in. above the water­ construction was just beginning on range of speeds. plane. And with proper shaping of the Valentijn's conventional yacht Eagle. The concept of a bulb producing a bow and stern we can have a measured Unfortunately, the Eagle syndicate was beneficial interaction with the bow is a length which will again allow a sail a half-million dollars in debt. The useful one, but it misses an important area of 2,450 sq. ft. Further, the 12- construction costs for Eagle began to point. The wave resistance of a hull is meter rule will permit a total underwa­ mount rapidly, and ahead still lay the proportional to the fourth power of the ter length of 44 ft Hence the pro­ need for sails, masts, booms, and so slenderness ratio of the hull. The bul­ boscis can extend 44 minus 33.6 ft., or orr. bous bows on tankers are merely small 10.4 ft out ahead of the bow. Under these circumstances the add-ons. However, the 12-meter rule All of this accomplishes two board of directors of the Eagle syndi­ gives the designer great freedom to important things. First, the wavemak­ cate became deeply divided on the make the bulbous bow almost any ing length of 44 ft is now nearly equal question of planning for the second, shape he desires as long as it is at least to the 45.6 ft. of conventional 12 more radical boat. Tooby, chairman 6 in. under the waterline. Such an meters. But more importantly, the of the board, and Gary Thomson, syn-

2000.---,----r---.r---.----r--~--__,

YACHT WITH 1800 AILERONS

10 1600 ~ (f) o 0 z Z ~ :::J 0 1400 z n. o w z w W 1200 n. u (f) 5 z LIBERTY l- « I tn 1000 U (f) >! W 0:: 800 l- I <:2 0:: 600 5 10 15 20 n. WINDSPEED IN KNOTS :::J 400

LIGHTWEIGHT YACHT WITH 200 UNDERWATER PROBOSCIS

0 4 6 7 8 9 10 II SPEED IN KNOTS

Figure 13 (above left): Since the power of the ailerons goes up with the square of the boat's speed, a yacht with ailerons could take full advantage of high winds. A con­ ventional yacht would have to reef its sail, thus leveling off its speed, to keep from heeling too far.

Figure 14 (above): Towing tank tests showed the slender, lightweight model with an underwater proboscis with the same resistance at 11 knots as Liberty had at 9 knots. the Eagle syndicate insisted that these tests on the model at the left be condUCted with Eagle's keel rather than with pontoons or ailerons.

28 ENGINEERING & SCIENCE / NOVEMBER 1986 dicate president, were enthusiastic it was clear that there was no chance most of these yachts are outspoken about building and testing a model of of meeting this goal, and the syndicate proponents of the "2 percent school," my proposed hull with the underwater decided to focus its efforts on the con­ that is, that what it takes to win the proboscis. But Valentijn and otheF ventional boat. America's Cup race is a 2 percent members of the syndicate were At this point Tooby and I decided advantage. This is being borne out in opposed on the grounds that this to resign from the Eagle syndicate and Perth. In a four-hour race, the would take away funds needed for approach the New York Yacht Club difference between the winner and the Eagle. (of which Tooby was a member) to see loser is usually less than five minutes. It was not until March 1986 that a if at this late date they would have any Still, I am convinced by our tests decision was made to make funds interest in undertaking the design and that improvements of tens of percent available to build a 40-percent scale construction of such a radical hull. are possible. What lies behind this model of the lightweight boat with the The members of the New York Yacht difference in point of view? Almost underwater proboscis. Even though Club syndicate expressed enthusiastic certainly it can be traced to the secrecy this boat was designed to· be used with interest, but the boat-building firm that that surrounds the design of 12-meter either pontoons or ailerons, members had constructed all of the NYYC's yachts. There is almost no scientific of the syndicate insisted that it be 12-meter yachts (as well as Eagle) told literature on the subject. Each tested with the keel and rudder used them that it was too late to design and designer jealously guards his store of on Eagle, ostensibly to have a direct build a new hull, particularly such a information. The usual intellectual comparison. radical one. discipline that exists when scientific Early in April we were able to test By this time the New York Yacht ideas are published and subjected to this model, and the results were little Club had already built two versions of the examination of others is almost short of spectacular. They are shown its yacht America II and was in the entirely lacking. I have become aware in Figure 14 in comparison with process of launching a third, since the of a long list of intuitive ideas that Liberty. At low speeds the resistance first of these had proven to be a poor sailors and designers alike cling to that is lower because of the smaller size of contender. Although it was too late to I am convinced are nonsense. Perhaps the lightweight hull. At higher speeds, design and build a new boat, the boat with this article I can convince others instead of the usual disadvantage that builders said that it might be possible to publish their ideas and, in the pro­ a smaller boat has, the underwater pro­ to take the first boat, cut the entire cess, to see how well these ideas stand boscis, combined with the slenderness bottom out of it, and put in a new up under scientific scrutiny. that it makes possible, gives the ultra­ bottom. They thought they could do One risk I run in doing this is that light the truly remarkable advantage in that by September 1. publishing this article may cause the decreased resistance that theory had Tooby and I went back East for a rules to be changed. Although the rules predicted. The ultra-light has the same lengthy meeting with representatives of as originallY written were intended to resistance at II knots that the conven­ the New York Yacht Club and their be well defined, the keepers of the rule tional hull has at 9 knots. And on top designer. We laid all our cards on the book, being human, couldn't quite let of this, the ultra-light can carry 40 per­ table, describing in detail the design of go of the strings of control. Not in the cent more sail area. Advantages such the hull and the towing tank tests on rules themselves but in the instructions as this usually occur only in a it. The NYYC members were at first to the measurers, they inserted the fol­ yachtsman's wildest dreams. skeptical but then admitted that in all lowing paragraph: These results created quite a com­ their computer studies they had never "If from any peculiarity ... the motion within the Eagle syndicate. uncovered the great potential that our National Yacht Racing Union ... is in Tooby and Thomson (who had wit­ tests had demonstrated. They said doubt as to the application of the rules nessed the towing tank tests) were they would have to confer with their or instructions ... it shall report the highly enthusiastic. Others were less various committees. Tooby and I case to the International Yacht Racing so. Because of an ever-increasing came away buoyed up. Union, who ... shall award such dearth of funds, there was now the A week later we received a letter certificate of rating as it may deem danger that if a new hull were to be from the chief operating officer of the equitable; and the measurement shall built, work on the original hull might America II syndicate full of praise for be deemed incomplete until this has have to be stopped. But perhaps the what we had done but then saying that been done." most critical factor facing us all was they had decided not to go ahead with As a result, almost every innova­ the question of time. To compete in such a radical project. It thus became tion is challenged on the basis of this the America's Cup, all yachts had to clear that the current America's Cup "peculiarity" instruction. The IYRU be measured and certificated by Sep­ races would be run without any of the has maneuvered so that it can make tember I, 1986. In order to meet this ideas that I have described here. quite subjective judgments as to what date, construction on the new hull As I look back on my two years' it will permit. The New York Yacht would have to begin by May I, and it initiation into the design of America's Club people told me that a number of would have to be flown to Perth by air Cup yachts, I am amazed at the great decisions have been made solely on the cargo. difference between the hyperbolic basis of what the IYRU deemed "good During the last two weeks in April, claims of "high tech" innovation that for the sport." It remains to be seen Thomson and Tooby made a heroic appear in the press and the design of whether broad beams, pontoons, effort to raise an extra million dollars the actual yachts now contending in ailerons, or underwater proboscises will to build the new boat. But by May I the waters off Perth. The designers of be good for the sport. D

29 You'd be receiving full tuition, university; Full Study Fellows attend Fellowships books and fees, an educational stipend, classes full-time. full employee benefits, relocation ex­ Since Hughes is involved with Pioneer the future. penses, professional-level salary, sum­ more than 90 technologies, a wide All of the technological ad­ mer employment, technical experience range of technical assignments is vancememts that have been pioneered ...with a combined value of $25,000 available. An Engineering Rotation by Hughes are merely an introduction to $50,000 a year. Program is also available for those interested in diversifying their work to what will come. While you're completing your experience. degree, you'll also have the opportunity And what's coming will be as­ Since 1949, more than 5,000 men to gain valuable experience at Hughes tounding. and women have earned advanced facilities in Southern California, Arizona As a Hughes Fellow, you could be degrees in engineering and science or Colorado. pioneering your own future, studying with the help of Hughes fellowships. for your Master's or Doctorate in Engi­ Hughes Fellows work full-time We hope you'll join us in creating the neering, (Electrical, Mechanical, during the summer. During the aca­ next generation of technological Manufacturing), Computer Science demic year, Work Study Fellows Work wonders, by pioneering the future­ or Physics. part-time while studying at a nearby yours and ours.

Hughes Aircraft Company, Corporate Fellowship Office Dept. MSETC, Bldg. C1/B168, P.O. Box 45066, Los Angeles, CA 90045-0066 Please consider me a candidate for a Hughes Fellowship and send me the necessary information and application materials.

PLEASE PRINT: Name Date

Address City State Zip

I am interested in obtaining a Master's _____ Doctorate _____ in the field of: ______

Rotation Program Interest: Yes ______No ______-:-

DEGREES NOW HELD (OR EXPECTED):

Bachelor's: Date Field School GPA HUGHES AIRCRAFT COMPANY

Master's: Date Field School GPA

Minimum GPA - 3.0/4.0. Proof of U.S. Citizenship Required. Equal Opportunity Employer. The Case of the Unconfirmed Phenomenon

It waS cold, dark, alumni - thousands and cloudy day­ of them-got gravity wave together through the weather. Gravity Alumni Fund and waves. Hmmm. Were chipped in to sup­ they for real or was port this project and Einstein just pulling a whole lot of other my leg? I had to find things too. You out for myself. know, 'cutting edge; I stopped in at my 'state of the art; local hardware store 'tomorrow's discov­ and asked Harvey, eries today; that the high school kid kind of stuff. Alumni behind the counter, are our silent part­ where they kept the ners in a lot of what gravity wave detec­ goes on here:' tors. He gave me a "How can I get in look that told me I on this?" I said. should have gone "How can I find the straight to Ed, the Alumni Fund?" owner. "Maybe you "Don't worry,' he should talk to Ed, the owner,' the kid said. Ed said with a mischievous grin, "the Alumni Fund could be found somewhere in aisle three by the will find you. You just be ready to make out your sixteen-penny nails. check when they do. Now get out of here; you're "What are you looking for buddy?" Ed said blocking my laser!" And so I left. as he hoisted a case of claw hammers to the I don't worry about gravity waves anymore. I upper shelf. just wait for the Alumni Fund solicitation to "Gravity waves;' I replied. arrive. And when it does, I'm going to send them a "S'pose you'd be fixin' to find a gravity wave check, a big one. detector then?" Because if a gravity wave turns up, I want a part "That's right:' of the credit for detecting it. "Well I can't help you ... try Caltech:' He'd been right about the weather stripping I'd asked about last winter, so I trusted him about Shore in the ochievementl this, too. And since Caltech was just a hop, skip, ,and a three hour plane flight from home I Support the Coltech Alumni Fund dropped in at the Institute. "$500,000! That's a little more than I was hop­ California Institute of Technology ing to spend!" I protested when told the tab just to Mail Code 105-40 house the prototype. Actually I knew that the cost Pasadena, California 91125 was quite reasonable; it takes only chalk and a 818/356-6285 blackboard to conceive of the universe's funda­ NOTE: The kind folks in Gravity Physics want you to know that mental natural phenomena, but a few million dol­ none of them are wizened. They also point out that the prototype lars to confirm them. I just figured that if I acted a detector they are working on has two arms, each 130 feet long, little indignant I might be in a better position to and incorporates vacuum tanks, suspended masses, and several lasers. The proposed upscaling of this model will have arms haggle. over one mile long. Unrestricted funds provided the seed money "It's not for sale anyway,' the wizened professor for this project which, in turn, is helping Caltech obtain-addi­ told me, "and don't call me wizened!Caltech tional funding from the National Science Foundation. Whether you're a research manager, a professor, 01 just interested in the latest technologies, plan to attend these conferences presented by Caltech's DNice for Industrial Associates.

Research Directors Conference i'ebruary10-11,1987 Caltech faculty will describe their rese.arch on computa- tion, communication and infonuation detectors and polymers. Admiral Bobby R. Inman, Westmark Systems Inc., will present the address. Also are Caltech talks by corporate researeh executives and an afternoon to focus Research groups with Caltech faculty. 1987 Electronic Materials and Devices March 3-4, 1987 Chairman: Dr. David B. Rutledge get the edge Associate Professor of Electrical Engineering Topics to be addressed will cover current advances in electronic materials and devices that are revolutionizing electrical solid state ics and materials science,

Chemical Frontiers in Biotechnology March 31-April1, 1987 Chairman: Dr. Peter B. Dervan Professor of Chemistry Emerging biotechnology, its current status and future direction, is the focus of this conference. Topics will include automated ON A sequelllemg, the mapping of the human chromosome, at the gene tran&-nption, and messenger RNA "P"',",'U!,>' ------Please send me the program and registration fonu for the following conferences: Research Directors Conference February 1987 Conference on Electronic Materials and Devices March 1987

L;ojntelren~oo on Chemical Froutiers in Biotechnology March31-Aprill,1987 universities. The Associates Linda McManus California Institute of lechm)lo~~ Events Coordinator Ucv'Clo}:)melllt 10540 356-6599 Ya8ad€~na, LalnOITna 91125 The Travel Program Of Alumni Flights Abroad

This is a private travel program especially planned for the alumni of Harvard, Yale, Princeton and certain other distinguished universities. Designed for the educated and intelligent traveler, it is specifically planned for the person who might normally prefer to travel independently, visiting distant lands and regions where it is advantageous to travel as a group. The itineraries follow a carefully planned pace which offers a more comprehensive and rewarding manner of travel, and the programs include great civilizations, beautiful scenery and important sights in diverse and interesting portions of the world:

TREASURES OF ANTIQUITY: The treasures of classical antiquity in Greece and Asia Minor and the Aegean Isles, from the actual ruins of Troy and the capital of the Hittites at Hattusas to the great city-states such as Athens and Sparta and to cities conquered by Alexander the Great (16 to 38 days). VAllEY OF mE NILE: An unusually careful survey of ancient Egypt that unfolds the art, the history and the achievements of one of the most remarkable civilizations the world has ever known (19 days). MEDITERRANEAN ODYSSEY: The sites of antiquity in the western Mediterra­ nean, from Carthage and the Roman cities of North Africa to the surprising ancient Greek ruins on the island of Sicily, together with the island of Malta (23 days).

EXPEDITION TO NEW GUINEA: The primitive stone-age culture of Papua-New Guinea, from the spectacular Highlands to the tribes of the Sepik River and the Karawari, as well as the Baining tribes on the island of New Britain (22 days). The soum PACIFIC: a magnificent journey through the "down under" world of New Zealand and Australia, including the Southern Alps, the New Zealand Fiords, Tasmania, the Great Barrier Reef, the Australian Out­ back, and a host of other sights. 28 days, plus optional visits to South Seas islands such as Fiji and Tahiti.

INDIA, CENTRAL ASIA AND mE HIMALAYAS: The romantic world of the Moghul Empire and a far-reaching group of sights, ranging from the Khyber Pass and the Taj Mahal to lavish forts and palaces and the snow-capped Himalayas of Kashmir and Nepal (26 or 31 days). soum OF BOMBAY: The unique and different world of south India and Sri Lanka (Ceylon) that offers ancient civilizations and works of art, palaces and celebrated temples, historic cities, and magnificent beaches and lush tropical lagoons and canals (23 or 31 days). mE ORIENT' The serene beauty of ancient and modern Japan explored in depth, together with the classic sights and civilizations of southeast Asia (30 days). BEYOND mEjAVA SEA: A different perspective of Asia, from headhunter villages in the jungle of Borneo and Batak tribal villages in Sumatra to the ancient civilizations of Ceylon and the thousand-year-old temples of central Java (34 days).

EAST AFRICA AND mE SEYCHEllES: A superb program of safaris in the great wilderness areas of Kenya and Tan­ zania and with the beautiful scenery and unusual birds and vegetation of the islands of the Seychelles (14 to 32 days).

DISCOVERIES IN mE soum: An unusual program that offers cruising among the islands of the Galapagos, the jungle of the Amazon, and astonishing ancient civilizations of the Andes and the southern desert of Peru (12 to 36 days), and soum AMERICA, which covers the continent from the ancient sites and Spanish colonial cities of the Andes to Buenos Aires, the spectacular Iguassu Falls, Rio de Janeiro, and the futuristic city of Brasilia (23 days).

In addition to these far-reaching surveys, there is a special program entitled "EUROPE REVISITED, "which is design­ ed to offer a new perspective for those who have already visited Europe in the past and who are already familiar with the major cities such as London, Paris and Rome. Included are medieval and Roman sites and the civilizations, cuisine and vineyards of BURGUNDY AND PROVENCE; medieval towns and cities, ancient abbeys in the Pyrenees and the astonishing prehistoric cave art of SOUmWEST FRANCE; the heritage of NORmERN ITALY, with Milan, Lake Como, Verona, Mantua, Vicenza, the villas of Palladio, Padua, Bologna, Ravenna and Venice; a survey of the works of Rembrandt, Rubens, Van Dyck, Vermeer, Brueghel and other old masters, together with historic towns and cities in HOLLAND AND FLANDERS: and a series of unusual journeys to the heritage of WALES, SCOTLAND AND ENGLAND.

Prices range from $2,225 to $5,895. Fully descriptive brochures are available, giving the itineraries in complete detail. For further information, please contact: Alumni Flights Abroad Dept. CT 19 A.F.A. Plaza, 425 Cherry Street Bedford Hills, New York 10507 TOLL FREE 1-800-AFA-8700 N.Y. State (914) 241-0111 NON· PROFIT ORG. ENGINE~~~~ U.S. POSTAGE PAID California Institute of Technology PASADENA, CA Pasadena, California 91125 PERMIT NO. 583

ADDRESS CORRECTION REQUESTED

Callech Logo Sweatshirts AduH sizes: Small, Medium, Large, Ex·Large THE CALTECH BOOKSTORE Colors: Mail Code 1·51, Pasadena, CA 91125 Grey with Powder Blue & Pink Logo Navy Blue with Powder Blue & Whrte Logo Black with White Logo ~~'------White with Orange & Black Logo A~, ______Price: $15.00 City ______State ____Zip, ______o Check or Money Order Please make check payable 10 btal Cost 01 Merchandise CaHech Booi<~O!e. California Residents Add 61,-'2 % Sales Tax o MasterCard 0 Visa

Cord No. Exp ___ 'Ii .) TOTAL AMOUNT OF ORDER ~MW~I ______