This breathless recommendation might stand as Joseph Ames's greatest error of judgment in a quarter century of service to the NACA, had he not suggested Bothezat be hired only conditionally by the Committee and given an opportunity to prove the assertions he had made in interviews with Ames. Stratton had interviewed Bothezat as well, shared Ames's enthusiasm, and also wanted to see some proof. Between them, Ames and Stratton had more than enough influence to get the NACA to adopt their suggestion. Bothezat was retained by the NACA, in a joint arrangement with the Army Air Service. He would advise the Committee on wind tunnels and a research program, and he would design for the Air Service a propeller suitable for the Liberty engine. If he succeeded in both, the NACA would consider hiring him full time.48

The man in the center is probably George de Bothezat. Far left is George Lewis; far right is Joseph Ames. The other two men seem to be army air service officers. This photograph was probably taken in Washington, D.C., or at Wright Field, Ohio, around 1920. (National Archives)

Bothezat began by examining the programs suggested by the NACA's staff engineers. He found them wanting. "They consist merely in a enumeration of different problems that can be investigated," he said, "but without any systematization of those problems." What was wanted was an understanding of "the general spirit that must animate all research in general but special [sic] all aerodynamical research." He explained:

Before a general conception of a problem to investigate is stated, one must take account of all the works made before and submit them to a critical investigation. Afterwards in the problem to investigate there must be reached as far as possible a certain general theoretical standpoint and clear understanding of the connections of the studied

problem to other problems and its relation to the general principals [sic] of dynamics and hydro-dynamics. The last constitutes only the fundamental demand of the continuity of scientifical evolution.49

In his awkward English, Bothezat was suggesting that the best researches are those enlightened by a knowledge of the previous work in the field and of the theoretical issues underlying the various problems retarding aeronautical progress. A laundry list of possible investigations is no substitute for a thorough understanding of the state of the art and an appreciation of what knowledge is required to advance that state. As he said earlier in the same report, with obvious reference to the kind of research being advocated by the NACA engineers:

Experimental researches or investigations can be of two kinds: Either they simply consist in measurements of some mechanical or physical quantities; Such measurements can be considered as scientific only when they are of a high grade of exactitude; In the other cases they simply constitute routine work. Or the experiments constitute a verification of a general conception of the studied phenomenon. It is the last investigations that generally have the most importance. This kind of conceptional investigations can be undertaken only when they are guided by a deep knowledge of all the studied phenomenon in its whole and its understanding from a unique philosophical standpoint.

Discounting the muddled syntax of that pronouncement, it is possible to see through his inadequate English to the essential idea he was trying to convey. Instead of the "measurements" and "routine work" he saw in progress around him, he advocated "conceptional investigations" based on "deep knowledge" and "understanding." That idea rang true to the charter of the NACA to engage in "the scientific study of the problems of flight." Presumably it was this kind of talk that had so impressed Ames and Stratton.

But Bothezat's influence with the Committee was shortlived, for he soon revealed himself as a better talker than performer. Ames had been warned, even before Bothezat was taken on by the NACA, that while he was "a brilliant mathematical physicist, and very suggestive,' he was nonetheless "wholly untrustworthy." Not only did it turn out that Bothezat promised more than he could deliver; he was soon discovered to have a temperament entirely unsuited to the Committee's needs. When an article he published was criticized by Jerome Hunsaker for claiming overmuch, Bothezat called the objections "ignorant criticism." At McCook Field to do propeller work for the army, Bothezat repeatedly modified the research program and never came to grips with the problems he had boasted of being able to solve. At the same

time, he gave overblown public lectures on the possibilities of using jet propulsion for interplanetary travel. His entire record with the Committee confirmed the hunch that he was brilliant but erratic. In the end, the NACA simply let him go as being too temperamental and poorly suited to working in American organizations. Though he seemed to have all the right ideas, he lacked the capacity to reduce them to practice, 50

Bothezat was soon forgotten at Langley and within the NACA. His successor, however, was unforgettable. Max Munk spent six years with the Committee and did more to shape the NACA's history than any other man in a comparable period of time. A protégé of Ludwig Prandtl, Munk came to the Committee in 1921 from the Zeppelin company in his native Germany, highly recommended by Jerome Hunsaker, who felt that his employment by the NACA would be the cheapest way of obtaining a great deal of unpublished aeronautical information generated in Germany during the war. Hunsaker also felt that Munk's abilities as a theoretician and generalist would allow him to draw conclusions from the work of others at the NACA—that is, from the engineers. Munk, in short, was to be the scientist providing the conceptual framework on which the NACA engineers would hang their researches. 51

At first, Munk was spectacularly successful. In five years with the Committee he authored or co-authored 57 reports, more than any other writer in NACA history except Edward P. Warner. As early as 1922, Dr. Ames-himself a scientist-reported to the NACA annual meeting that "Dr. Munk's work during the past year, in the theoretical side of aerodynamics, has placed the Committee in the forefront of the world." Two years later, Ames said that "the papers written by Dr. Munk during the past year have done more to make aerodynamics and hydrodynamics a living thing than anything that has happened during the past twenty years." Even George Lewis, the engineer, shared this original enthusiasm for Munk's work and its value to the Committee. At Lewis's suggestion, Dr. Ames summarized six of Munk's most important reports in a form more understandable and appealing to aeronautical engineers who lacked the "very extensive training in mathematics and physics" necessary to understand them. Lewis wanted to make Munk's work more "readily appreciated by the average aeronautical engineer or designer and to further stimulate his interest so that he will undertake the reading of Doctor Munk's papers in detail."52

Munk's greatest contribution to the NACA, however, was not the papers he wrote but the wind tunnel he built. When Munk joined the NACA, only a simple atmospheric wind tunnel based on European design was in operation at Langley. Through a 5-foot-diameter test section in the tunnel, a propeller pushed air at varying speeds across a

model of an aircraft or wing section to simulate conditions of flight. The forces on the model were measured by a set of balances. One of the main questions of aerodynamical research at the time was the scale effect: Did the air act on the model in a way determinably proportional to the way it acted on the full-scale body? Even as the NACA's first atmospheric wind tunnel went into operation, it was realized that the answer was no; a scale effect compromised wind-tunnel results, and required correction in a way that could then only be guessed at.53 Munk's contribution was to create a wind tunnel in which the conditions in the tunnel were directly comparable to those in flight. The key to the problem was air density. The forces acting on bodies immersed in a moving fluid (like air) depend on the Reynolds number-a dimensionless mathematical quantity that varies directly with the size of the body, the velocity of the stream, and the density of the air, and inversely with the viscosity of the air. Results from use of a small model could be made comparable to those from a full-size aircraft by increasing the speed or the density of the air, or by reducing its viscosity. The first method was impractical because a one-tenth-scale model would require air speeds in the tunnel ten times those encountered in flight. Even had such a supersonic tunnel been technologically feasible in the 1920s, which it was not, it would have produced compressibility effects even more distorting than the scale effect. The third method, reducing viscosity, was theoretically possible by reducing the temperature in the tunnel, but this too was beyond the technology of the 1920s. The only practical factor to vary was the air density. This could be done by increasing the pressure in the tunnel. If ten atmospheres of pressure could be generated in a wind tunnel, then tests of a one-tenth-scale model would produce usable results. 54

Though all this was known before Munk began his work for the NACA, no use had been made of it. It was Munk who translated the theory into a practical proposal and designed the variable-density wind tunnel, essentially a tunnel in a bottle. A conventional annular-return wind tunnel with a five-foot test section was entirely enclosed in a steel tank 10.5m long and 4.5m in diameter. The tank could withstand pressures of 21 atmospheres, so that almost any model of reasonable size could be tested under conditions comparable to those encountered by a full-scale aircraft in flight.

Though greeted with some skepticism at home and abroad, the variable-density tunnel worked, and it began a revolution in aeronautical research. By the end of the decade, other countries began building similar tunnels. The NACA became famous for innovative research techniques and tools, and used this fame to win more funds from Congress for equally innovative facilities and equipment in the years to come. 55 Munk's fame also increased, not only for the tunnel but also

for the research he conducted with it. Early in the tunnel's life, Munk used it to test his new theory of airfoils. From this work flowed the achievement for which the NACA is perhaps best known among aircraft designers: the NACA family of airfoil shapes. By designing, modeling, and testing whole series of airfoils in which such characteristics as camber, maximum thickness, and chordwise thickness distribution were varied slightly and systematically in each successive model, the NACA was able to provide designers with a wing section for every purpose: that is, a family of sections in which the characteristics of each were so well defined that a designer could simply select off the shelf the one best suited to the aircraft he was designing.56

All of this began with Max Munk, but he was not around to share in the harvest. However popular he may have been with Joseph Ames and other members of the NACA, he was not well liked at the Langley laboratory, where he headed the aerodynamic research section. Like Bothezat before him, Munk was arrogant and eccentric, and the English language would never quite do what he wanted. Like many geniuses, he had a mind filled with as many crackpot schemes as flashes of bril