During the 1940s, investigators in the United States and Hungary bounced radar waves off the Moon for the first time, while others made the first systematic radar studies of meteors. These experiments constituted the initial exploration of the solar system with radar. In order to understand the beginnings of radar astronomy, we first must examine the origins of radar in radio, the decisive role of ionospheric research, and the rapid development of radar technology triggered by World War II.
As early as 20 June 1922, in an address to a joint meeting of the Institute of Electrical Engineers and the Institute of Radio Engineers in New York, the radio pioneer Guglielmo Marconi suggested using radio waves to detect ships:1
By the time Germany invaded Poland in September 1939 and World War II was underway, radio detection, location, and ranging technologies and techniques were available in Japan, France, Italy, Germany, England, Hungary, Russia, Holland, Canada, and the United States. Radar was not so much an invention, springing from the laboratory bench to the factory floor, but an ongoing adaptation and refinement of radio technology. The apparent emergence of radar in Japan, Europe, and North America more or less at the same time was less a case of simultaneous invention than a consequence of the global nature of radio research.2
Although radar is identified overwhelmingly with World War II, historian Sean S. Swords has argued that the rise of high-performance and long-range aircraft in the late 1930s would have promoted the design of advanced radio navigational aids, including radar, even without a war.3 More decisively, however, ionospheric research propelled radar development in the 1920s and 1930s. As historian Henry Guerlac has pointed out, "Radar was developed by men who were familiar with the ionospheric work. It was a relatively straightforward adaptation for military purposes of a widely-known scientific technique,  which explains why this adaptation--the development of radar--took place simultaneously in several different countries."4
The prominence of ionospheric research in the history of radar and later of radar astronomy cannot be ignored. Out of ionospheric research came the essential technology for the beginnings of military radar in Britain, as well as its first radar researchers and research institutions. After the war, as we shall see, ionospheric research also drove the emergence of radar astronomy.
Despite its scientific origins, radar made its mark and was baptized during World War II as an integral and necessary instrument of offensive and defensive warfare. Located on land, at sea, and in the air, radars detected enemy targets and determined their position and range for artillery and aircraft in direct enemy encounters on the battlefield. Other radars identified aircraft to ground bases as friend or foe, while others provided navigational assistance and coastal defense. World War II was the first electronic war, and radar was its prime agent.5
In 1940, nowhere did radar research achieve the same advanced state as in Britain. The British lead initially resulted from a decision to design and build a radar system for coastal defense, while subsequent research led to the invention of the cavity magnetron, which placed Britain in the forefront of microwave radar. The impetus to achieve that lead in radar came from a realization that the island nation was no longer safe from enemy invasion.
For centuries, Britain's insularity and navy protected it from invasion. The advent of long-range airplanes that routinely outperformed their wooden predecessors spelled the end of that protection. Existing aircraft warning methods were ineffectual. That Britain was virtually defenseless against an air assault became clear during the summer air exercises of 1934. In simulated night attacks on London and Coventry, both the Air Ministry and the Houses of Parliament were successfully "destroyed," while few "enemy" bombers were intercepted.6
International politics also had reached a critical point. The Geneva Disarmament Conference had collapsed, and Germany was rearming in defiance of the Treaty of Versailles. Under attack from Winston Churchill and the Tory opposition, the British government abandoned its disarmament policy and initiated a five-year expansion of the Royal Air Force. Simultaneously, the Air Ministry Director of Scientific Research, Henry Egerton Wimperis, created a committee to study air defense methods.
Just before the Committee for the Scientific Survey of Air Defence first met on 28 January 1935, Wimperis contacted fellow Radio Research Board member Robert (later Sir) Watson-Watt. Watson-Watt, who oversaw the Radio Research Station at Slough, was a scientist with twenty years of experience as a government researcher. Ionospheric research had been a principal component of Radio Research Station studies, and Watson-Watt fostered the development there of a pulse-height technique.7
 The pulse-height technique was to send short pulses of radio energy toward the ionosphere and to measure the time taken for them to return to Earth. The elapsed travel time of the radio waves gave the apparent height of the ionosphere. Merle A. Tuve, then of Johns Hopkins University, and Gregory Breit of the Carnegie Institution's Department of Terrestrial Magnetism in Washington, first developed the technique in the 1920s and undertook ionospheric research in collaboration with the Naval Research Laboratory and the Radio Corporation of America.8
In response to the wartime situation, Wimperis asked Watson-Watt to determine the practicality of using radio waves as a "death ray." Rather than address the proposed "death ray," Watson-Watt's memorandum reply drew upon his experience in ionospheric research. Years later, Watson-Watt contended, "I regard this Memorandum on the 'Detection and Location of Aircraft by Radio Methods' as marking the birth of radar and as being in fact the invention of radar." Biographer Ronald William Clark has termed the memorandum "the political birth of radar."9 Nonetheless, Watson-Watt's memorandum was really less an invention than a proposal for a new radar application.
The memorandum outlined how a radar system could be put together and made to detect and locate enemy aircraft. The model for that radar system was the same pulse-height technique Watson-Watt had used at Slough. Prior to the memorandum in its final form going before the Committee, Wimperis had arranged for a test of Watson-Watt's idea that airplanes could reflect significant amounts of radio energy, using a BBC transmitter at Daventry. "Thus was the constricting 'red tape' of official niceties slashed by Harry Wimperis, before the Committee for the Scientific Survey of Air Defence had so much as met," Watson-Watt later recounted. The success of the Daventry test shortly led to the authorization of funding (£12,300 for the first year) and the creation of a small research and development project at Orford Ness and Bawdsey Manor that drew upon the expertise of the Slough Radio Research Station.
From then onwards, guided largely by Robert Watson-Watt, the foundation of the British radar effort, the early warning Chain Home, materialized. The Chain Home began in December 1935, with Treasury approval for a set of five stations to patrol the air approaches to the Thames estuary. Before the end of 1936, and long before the first test of the Thames stations in the autumn of 1937, plans were made to expand it into a network of nineteen stations along the entire east coast; later, an additional six stations were built to cover the south coast.
 The Chain Home played a crucial role in the Battle of Britain, which began in July 1940. The final turning point was on 15 September, when the Luftwaffe suffered a record number of planes lost in a single day. Never again did Germany attempt a massive daylight raid over Britain. However, if radar won the day, it lost the night. Nighttime air raids showed a desperate need for radar improvements.
In order to wage combat at night, fighters needed the equivalent of night vision--their own on-board radar, but the prevailing technology was inadequate. Radars operating at low wavelengths, around 1.5 meters (200 MHz), cast a beam that radiated both straight ahead and downwards. The radio energy reflected from the Earth was so much greater than that of the enemy aircraft echoes that the echoes were lost at distances greater than the altitude of the aircraft. At low altitudes, such as those used in bombing raids or in air-to-air combat, the lack of radar vision was grave. Microwave radars, operating at wavelengths of a few centimeters, could cast a narrower beam and provide enough resolution to locate enemy aircraft.10
Although several countries had been ahead of Britain in microwave radar technology before the war began, Britain leaped ahead in February 1940, with the invention of the cavity magnetron by Henry A. H. Boot and John T. Randall at the University of Birmingham.11 Klystrons were large vacuum tubes used to generate microwave power, but they did not operate adequately at microwave frequencies. The time required for electrons to flow through a klystron was too long to keep up with the frequency of the external oscillating circuit. The cavity magnetron resolved that problem and made possible the microwave radars of World War II. As Sean Swords asserted, "The emergence of the resonant-cavity magnetron was a turning point in radar history." 12 The cavity magnetron launched a line of microwave research and development that has persisted to this day.
The cavity magnetron had no technological equivalent in the United States, when the Tizard Mission arrived in late 1940 with one of the first ten magnetrons constructed. The Tizard Mission, known formally as the British Technical and Scientific Mission, had been arranged at the highest levels of government to exchange technical information between Britain and the United States. Its head and organizer, Henry Tizard, was a prominent physics professor and a former member of the committee that had approved Watson-Watt's radar project. As James P. Baxter wrote just after the war's end with a heavy handful of hyperbole, though not without some truth: "When the members of the Tizard Mission brought one [magnetron] to America in 1940, they carried the most valuable cargo ever brought to our shores. It sparked the whole development of microwave radar and constituted the most important item in reverse Lease-Lend." 13
 In late September 1940, Dr. Edward G. Bowen, the radar scientist on the Tizard Mission, showed a magnetron to members of the National Defense Research Committee (NDRC), which President Roosevelt had just created on 27 June 1940. One of the first acts of the NDRC, which later became the Office of Scientific Research and Development, was to establish a Microwave Committee, whose stated purpose was "to organize and consolidate research, invention, and development as to obtain the most effective military application of microwaves in the minimum time." 14
A few weeks after the magnetron demonstration, the NDRC decided to create the Radiation Laboratory at MIT. While the MIT Radiation Laboratory accounted for nearly 80 percent of the NDRC Microwave Division's contracts, an additional 136 contracts for radar research, development, and prototype work were let out to 16 colleges and universities, two private research institutions, and the major radio industrial concerns, with Western Electric taking the largest share. The MIT Radiation Laboratory personnel skyrocketed from thirty physicists, three guards, two stock clerks, and a secretary for the first year to a peak employment level of 3,897 (1,189 of whom were staff) on 1 August 1945. The most far-reaching early achievement, accomplished in the spring of 1941, was the creation of a new generation of radar equipment based on a magnetron operating at 3 cm. Experimental work in the one cm range led to numerous improvements in radars at 10 and 3 cm.15
Meanwhile, research and development of radars of longer wavelengths were carried out by the Navy and the Army Signal Corps, both of which had had active ongoing radar programs since the 1930s. The Navy started its research program at the Naval Research Laboratory (NRL) before that of the Signal Corps, but radar experimenters after the war used Signal Corps equipment, especially the SCR-270, mainly because of its wide availability. A mobile SCR-270, placed on Oahu as part of the Army's Aircraft Warning System, spotted incoming Japanese airplanes nearly 50 minutes before they bombed United States installations at Pearl Harbor on 7 December 1941. The warning was ignored, because an officer mistook the radar echoes for an expected flight of B-17s.16
Historians view the large-scale collection of technical and financial resources and manpower at the MIT Radiation Laboratory engaged in a concerted effort to research and develop new radar components and systems, along with the Manhattan Project, as  signalling the emergence of Big Science. Ultimately, from out of the concentration of personnel, expertise, materiel, and financial resources at the successor of the Radiation Laboratory, Lincoln Laboratory, arose the first attempts to detect the planet Venus with radar. The Radiation Laboratory Big Science venture, however, did not contribute immediately to the rise of radar astronomy.
The radar and digital technology used in those attempts on Venus was not available at the end of World War II, when the first lunar and meteor radar experiments were conducted. Moreover, the microwave radars that issued from Radiation Laboratory research were far too weak for planetary or lunar work and operated at frequencies too high to be useful in meteor studies. Outside the Radiation Laboratory, though, U.S. Army Signal Corps and Navy researchers had created radars, like the SCR-270, that were more powerful and operated at lower frequencies, in research and development programs that were less concentrated and conducted on a smaller scale than the Radiation Laboratory effort.
Wartime production created an incredible excess of such radar equipment. The end of fighting turned it into war surplus to be auctioned off, given away, or buried as waste. World War II also begot a large pool of scientists and engineers with radar expertise who sought peacetime scientific and technical careers at war's end. That pool of expertise, when combined with the cornucopia of high-power, low-frequency radar equipment and a pinch of curiosity, gave rise to radar astronomy.
A catalyst crucial to that rise was ionospheric research. In the decade and a half following World War II, ionospheric research underwent the kind of swift growth that is typical of Big Science. The ionospheric journal literature doubled every 2.9 years from 1926 to 1938, before stagnating during the war; but between 1947 and 1960, the literature doubled every 5.8 years, a rate several times faster than the growth rate of scientific literature as a whole.17 Interest in ionospheric phenomena, as expressed in the rapidly growing research literature, motivated many of the first radar astronomy experiments undertaken on targets beyond the Earth's atmosphere.
Typical was the first successful radar experiment aimed at the Moon. That experiment was performed with Signal Corps equipment at the Corps' Evans Signal Laboratory, near Belmar, New Jersey, under the direction of John H. DeWitt, Jr., Laboratory Director. DeWitt was born in Nashville and attended Vanderbilt University Engineering School for two years. Vanderbilt did not offer a program in electrical engineering, so DeWitt dropped out in order to satisfy his interest in broadcasting and amateur radio. After building Nashville's first broadcasting station, in 1929 DeWitt joined the Bell Telephone Laboratories technical staff in New York City, where he designed radio broadcasting transmitters. He returned to Nashville in 1932 to become Chief Engineer of radio station WSM. Intrigued by Karl Jansky's discovery of "cosmic noise," DeWitt built a radio telescope and searched for radio signals from the Milky Way.
In 1940, DeWitt attempted to bounce radio signals off the Moon in order to study the Earth's atmosphere. He wrote in his notebook: "It occurred to me that it might be possible to reflect ultrashort waves from the moon. If this could be done it would open up wide possibilities for the study of the upper atmosphere. So far as I know no one has ever  sent waves off the earth and measured their return through the entire atmosphere of the earth."18
On the night of 20 May 1940, using the receiver and 80-watt transmitter configured for radio station WSM, DeWitt tried to reflect 138-MHz (2-meter) radio waves off the Moon, but he failed because of insufficient receiver sensitivity. After joining the staff of Bell Telephone Laboratories in Whippany, New Jersey, in 1942, where he worked exclusively on the design of a radar antenna for the Navy, DeWitt was commissioned in the Signal Corps and was assigned to serve as Executive Officer, later as Director, of Evans Signal Laboratory.
On 10 August 1945, the day after the United States unleashed a second atomic bomb on Japan, military hostilities between the two countries ceased. DeWitt was not demobilized immediately, and he began to plan his pet project, the reflection of radio waves off the Moon. He dubbed the scheme Project Diana after the Roman mythological goddess of the Moon, partly because "the Greek [sic] mythology books said that she had never been cracked."
In September 1945, DeWitt assembled his team: Dr. Harold D. Webb, Herbert P. Kauffman, E. King Stodola, and Jack Mofenson. Dr. Walter S. McAfee, in the Laboratory's Theoretical Studies Group, calculated the reflectivity coefficient of the Moon. Members of the Antenna and Mechanical Design Group, Research Section, and other Laboratory groups contributed, too.
No attempt was made to design major components specifically for the experiment. The selection of the receiver, transmitter, and antenna was made from equipment already on hand, including a special crystal-controlled receiver and transmitter designed for the Signal Corps by radio pioneer Edwin H. Armstrong. Crystal control provided frequency stability, and the apparatus provided the power and bandwidth needed. The relative velocities of the Earth and the Moon caused the return signal to differ from the transmitted signal by as much as 300 Hz, a phenomenon known as Doppler shift. The narrow-band receiver permitted tuning to the exact radio frequency of the returning echo. As DeWitt later recalled: "We realized that the moon echoes would be very weak so we had to use a very narrow receiver bandwidth to reduce thermal noise to tolerable levels....We had to tune the receiver each time for a slightly different frequency from that sent out because of the Doppler shift due to the earth's rotation and the radial velocity of the moon at the time."19
The echoes were received both visually, on a nine-inch cathode-ray tube, and acoustically, as a 180 Hz beep. The aerial was a pair of "bedspring" antennas from an SCR-271 stationary radar positioned side by side to form a 32-dipole array antenna and mounted on a 30-meter (100-ft) tower. The antenna had only azimuth control; it had not been practical to secure a better mechanism. Hence, experiments were limited to the rising and setting of the Moon.
 The Signal Corps tried several times, but without success. "The equipment was very haywire," recalled DeWitt. Finally, at moonrise, 11:48 A.M., on 10 January 1946, they aimed the antenna at the horizon and began transmitting. Ironically, DeWitt was not present: "I was over in Belmar having lunch and picking up some items like cigarettes at the drug store (stopped smoking 1952 thank God)." 20 The first signals were detected at 11:58 A.M., and the experiment was concluded at 12:09 P.M., when the Moon moved out of the radar's range. The radio waves had taken about 2.5 seconds to travel from New Jersey to the Moon and back, a distance of over 800,000 km. The experiment was repeated daily over the next three days and on eight more days later that month.
The War Department withheld announcement of the success until the night of 24 January 1946. By then, a press release explained, "the Signal Corps was certain beyond doubt that the experiment was successful and that the results achieved were pain-stakingly [sic] verified."21
As DeWitt recounted years later: "We had trouble with General Van Deusen our head of R&D in Washington. When my C.O. Col. Victor Conrad told him about it over the telephone the General did not want the story released until it was confirmed by outsiders for fear it would embarrass the Sig[nal]. C[orps]." Two outsiders from the Radiation Laboratory, George E. Valley, Jr. and Donald G. Fink, arrived and, with Gen. Van Deusen, observed a moonrise test of the system carried out under the direction of King Stodola. Nothing happened. DeWitt explained: "You can imagine that at this point I was dying. Shortly a big truck passed by on the road next to the equipment and immediately the echoes popped up. I will always believe that one of the crystals was not oscillating until it was shaken up or there was a loose connection which fixed itself. Everyone cheered except the General who tried to look pleased." 22
Although he had had other motives for undertaking Project Diana, DeWitt had received a directive from the Chief Signal Officer, the head of the Signal Corps, to develop radars capable of detecting missiles coming from the Soviet Union. No missiles were available for tests, so the Moon experiment stood in their place. Several years later, the Signal Corps erected a new 50-ft (15 meters) Diana antenna and 108-MHz transmitter for ionospheric research. It carried out further lunar echo studies and participated in the tracking of Apollo launches. 23
The news also hit the popular press. The implications of the Signal Corps experiment were grasped by the War Department, although Newsweek cynically cast doubt on the War Department's predictions by calling them worthy of Jules Verne. Among those War Department predictions were the accurate topographical mapping of the Moon and planets, measurement and analysis of the ionosphere, and radio control from Earth of "space ships" and "jet or rocket-controlled missiles, circling the Earth above the stratosphere." Time reported that Diana might provide a test of Albert Einstein's Theory of Relativity. In contrast to the typically up-beat mood of Life, both news magazines were skeptical, and  rightly so; yet all of the predictions made by the War Department, including the relativity test, have come true in the manner of a Jules Verne novel. 24
Less than a month after DeWitt's initial experiment, a radar in Hungary replicated his results. The Hungarian apparatus differed from that of DeWitt in one key respect; it utilized a procedure, called integration, that was essential to the first attempt to bounce radar waves off Venus and that later became a standard planetary radar technique. The procedure's inventor was Hungarian physicist Zoltán Bay.
Bay graduated with highest honors from Budapest University with a Ph.D. in physics in 1926. Like many Hungarian physicists before him, Bay spent several years in Berlin on scholarships, doing research at both the prestigious Physikalisch-Technische-Reichanstalt and the Physikalisch-Chemisches-Institut of the University of Berlin. The results of his research tour of Berlin earned Bay the Chair of Theoretical Physics at the University of Szeged (Hungary), where he taught and conducted research on high intensity gas discharges.
Bay left the University of Szeged when the United Incandescent Lamps and Electric Company (Tungsram) invited him to head its industrial research laboratory in Budapest. Tungsram was the third largest manufacturer of incandescent lamps, radio tubes, and radio receivers in Europe and supplied a fifth of all radio tubes. As laboratory head, Zoltán Bay oversaw the improvement of high-intensity gas discharge lamps, fluorescent lamps, radio tubes, radio receiver circuitry, and decimeter radio wave techniques.25
Although Hungary sought to stay out of the war through diplomatic maneuvering, the threat of a German invasion remained real. In the fall of 1942, the Hungarian Minister of Defense asked Bay to organize an early-warning system. He achieved that goal, though the Germans occupied Hungary anyway. In March 1944, Bay recommended using the radar for scientific experimentation, including the detection of radar waves bounced off the Moon. The scientific interest in the experiment arose from the opportunity to test the theoretical notion that short wavelength radio waves could pass through the ionosphere without considerable absorption or reflection. Bay's calculations, however, showed that the equipment would be incapable of detecting the signals, since they would be significantly below the receiver's noise level.
The critical difference between the American and Hungarian apparatus was frequency stability, which DeWitt achieved through crystal control in both the transmitter and receiver. Without frequency stability, Bay had to find a means of accommodating the frequency drifts of the transmitter and receiver and the resulting inferior signal-to-noise ratio. He chose to boost the signal-to-noise ratio. His solution was both ingenious and far-reaching in its impact.
Bay devised a process he called cumulation, which is known today as integration. His integrating device consisted of ten coulometers, in which electric currents broke down a watery solution and released hydrogen gas. The amount of gas released was directly proportional to the quantity of electric current. The coulometers were connected to the output of the radar receiver through a rotating switch. The radar echoes were expected  to return from the Moon in less than three seconds, so the rotating switch made a sweep of the ten coulometers every three seconds. The release of hydrogen gas left a record of both the echo signal and the receiver noise. As the number of signal echoes and sweeps of the coulometers added up, the signal-to-noise ratio improved. By increasing the total number of signal echoes, Bay believed that any signal could be raised above noise level and made observable, regardless of its amplitude and the value of the signal-to-noise ratio.26 Because the signal echoes have a more-or-less fixed structure, and the noise varies from pulse to pulse, echoes add up faster than noise.
Despite the conceptual breakthrough of the coulometer integrator, the construction and testing of the apparatus remained to be carried out. The menace of air raids drove the Tungsram research laboratory into the countryside in the fall of 1944. The subsequent siege of Budapest twice interrupted the work of Bay and his team until March 1945. The Ministry of Defense furnished Bay with war surplus parts for a 2.5-meter (120-MHz) radar manufactured by the Standard Electrical Co., a Hungarian subsidiary of IT&T. Work was again interrupted when the laboratory was dismantled and all equipment, including that for the lunar radar experiment, was carried off to the Soviet Union. For a third time, construction of entirely new equipment started in the workshops of the Tungsram Research Laboratory, beginning August 1945 and ending January 1946.
Electrical disturbances in the Tungsram plant were so great that measurements and tuning had to be done in the late afternoon or at night. The experiments were carried out on 6 February and 8 May 1946 at night by a pair of researchers. Without the handicap of operating in a war zone, Bay probably would have beaten the Signal Corps to the Moon, although he could not have been aware of DeWitt's experiment. More importantly, though, he invented the technique of  long-time integration generally used in radar astronomy. As the American radio astronomers Alex G. Smith and Thomas D. Carr wrote some years later: "The additional tremendous increase in sensitivity necessary to obtain radar echoes from Venus has been attained largely through the use of long-time integration techniques for detecting periodic signals that are far below the background noise level. The unique method devised by Bay in his pioneer lunar radar investigations is an example of such a technique."27
Both Zoltán Bay and John DeWitt had fired shots heard round the world, but there was no revolution, although others either proposed or attempted lunar radar experiments in the years immediately following World War II. Each man engaged in other projects shortly after completing his experiment. Bay left Hungary for the United States, where he taught at George Washington University and worked for the National Bureau of Standards, while DeWitt re-entered radio broadcasting and pursued his interest in astronomy.28
As an ongoing scientific activity, radar astronomy did not begin with the spectacular and singular experiments of DeWitt and Bay, but with an interest in meteors shared by researchers in Britain, Canada, and the United States. Big Science, that is, ionospheric physics and secure military communications, largely motivated that research. Moreover, just as the availability of captured V-2 parts made possible rocket-based ionospheric research after the war,29 so war-surplus radars facilitated the emergence of radar astronomy. Like the exploration of the ionosphere with rockets, radar astronomy was driven by the availability of technology.
Radar meteor studies, like much of radar history, grew out of ionospheric research. In the 1930s, ionospheric researchers became interested in meteors when it was hypothesized that the trail of electrons and ions left behind by falling meteors caused fluctuations in the density of the ionosphere.30 Edward Appleton and others with the Radio Research Board of the British Department of Scientific and Industrial Research, the same organization with which Watson-Watt had been associated, used war-surplus radar furnished by  the Air Ministry to study meteors immediately after World War II. They concluded that meteors caused abnormal bursts of ionization as they passed through the ionosphere.31
During the war, the military had investigated meteor trails with radar. When the Germans started bombarding London with V2 rockets, the Army's gun-laying radars were hastily pressed into service to detect the radar reflections from the rockets during their flight in order to give some warning of their arrival. In many cases alarms were sounded, but no rockets were aloft. James S. Hey, a physicist with the Operational Research Group, was charged with investigating these mistaken sightings. He believed that the false echoes probably originated in the ionosphere and might be associated with meteors.
Hey began studying the impact of meteors on the ionosphere in October 1944, using Army radar equipment at several locations until the end of the war. The Operational Research Group, Hey, G. S. Stewart (electrical engineer), S. J. Parsons (electrical and mechanical engineer), and J. W. Phillips (mathematician), found a correlation between visual sightings and radar echoes during the Giacobinid meteor shower of October 1946. Moreover, by using an improved photographic technique that better captured the echoes on the radar screen, they were able to determine the velocity of the meteors.
Neither Hey nor Appleton pursued their radar investigations of meteors. During the war, Hey had detected radio emissions from the Sun and the first discrete source of radio emission outside the solar system in the direction of Cygnus. He left the Operational Research Group for the Royal Radar Establishment at Malvern, where he and his colleagues carried on research in radio astronomy. Appleton, by 1946 a Nobel Laureate and Secretary of the Department of Scientific and Industrial Research, also became thoroughly involved in the development of radio astronomy and became a member of the Radio Astronomy Committee of the Royal Astronomical Society in 1949.32
Instead, radar astronomy gained a foothold in Britain at the University of Manchester under A. C. (later Sir) Bernard Lovell, director of the University's Jodrell Bank Experimental Station. During the war, Lovell had been one of many scientists working on microwave radar.33 His superior, the head of the Physics Department, was Patrick M. S. Blackett, a member of the Committee for the Scientific Survey of Air Defence that approved Watson-Watt's radar memorandum. With the help of Hey and Parsons, Lovell borrowed some Army radar equipment. Finding too much interference in Manchester, he moved to the University's botanical research gardens, which became the Jodrell Bank Experimental Station. Lovell equipped the station with complete war-surplus radar systems, such as a 4.2-meter gun-laying radar and a mobile Park Royal radar. He purchased at rock-bottom prices or borrowed the radars from the Air Ministry, Army, and Navy, which were discarding the equipment down mine shafts.
Originally, Lovell wanted to undertake research on cosmic rays, which had been Blackett's interest, too. One of the primary research objectives of the Jodrell Bank facility, as well as one of the fundamental reasons for its founding, was cosmic ray research. Indeed, the interest in cosmic ray research also lay behind the design and construction of the 76-meter (250-ft) Jodrell Bank telescope. The search for cosmic rays never succeeded, however; Blackett and Lovell had introduced a significant error into their initial calculations.
Fortuitously, though, in the course of looking for cosmic rays, Lovell came to realize that they were receiving echoes from meteor ionization trails, and his small group of Jodrell Bank investigators began to concentrate on this more fertile line of research. Nicolai Herlofson, a Norwegian meteorologist who had recently joined the Department of Physics, put Lovell in contact with the director of the Meteor Section of the British Astronomical Association, J. P. Manning Prentice, a lawyer and amateur astronomer with a passion for meteors. Also joining the Jodrell Bank team was John A. Clegg, a physics teacher whom Lovell had known during the war. Clegg was a doctoral candidate at the University of Manchester and an expert in antenna design. He remained at Jodrell Bank until 1951 and eventually landed a position teaching physics in Nigeria. Clegg converted an Army searchlight into a radar antenna for studying meteors.34
 The small group of professional and amateur scientists began radar observations of the Perseid meteor showers in late July and August 1946. When Prentice spotted a meteor, he shouted. His sightings usually, though not always, correlated with an echo on the radar screen. Lovell thought that the radar echoes that did not correlate with Prentice's sightings might have been ionization trails created by cosmic ray showers. He did not believe, initially, that the radar might be detecting meteors too small to be seen by the human eye.
The next opportunity for a radar study of meteors came on the night of 9 October 1946, when the Earth crossed the orbit of the Giacobini-Zinner comet. Astronomers anticipated a spectacular meteor shower. A motion picture camera captured the radar echoes on film. The shower peaked around 3 A.M.; a radar echo rate of nearly a thousand meteors per hour was recorded. Lovell recalled that "the spectacle was memorable. It was like a great array of rockets coming towards one."35
The dramatic correlation of the echo rate with the meteors visible in the sky finally convinced Lovell and everyone else that the radar echoes came from meteor ionization trails, although it was equally obvious that many peculiarities needed to be investigated. The Jodrell Bank researchers learned that the best results were obtained when the aerial was positioned at a right angle to the radiant, the point in the sky from which meteor showers appear to emanate. When the aerial was pointed at the radiant, the echoes on the cathode-ray tube disappeared almost completely.36
Next joining the Jodrell Bank meteor group, in December 1946, was a doctoral student from New Zealand, Clifton D. Ellyett, followed in January 1947 by a Cambridge graduate, John G. Davies. Nicolai Herlofson developed a model of meteor trail ionization that Davies and Ellyett used to calculate meteor velocities based on the diffraction pattern produced during the formation of meteor trails. Clegg devised a radar technique for determining their radiant.37
At this point, the Jodrell Bank investigators had powerful radar techniques for studying meteors that were unavailable elsewhere, particularly the ability to detect and study previously unknown and unobservable daytime meteor showers. Lovell and his colleagues now became aware of the dispute over the nature of meteors and decided to attempt its resolution with these techniques.38
Astronomers specializing in meteors were concerned with the nature of sporadic meteors. One type of meteor enters the atmosphere from what appears to be a single point, the radiant. Most meteors, however, are not part of a shower, but appear to arrive irregularly from all directions and are called sporadic meteors. Most astronomers believed that sporadic meteors came from interstellar space; others argued that they were part of the solar system.
The debate could be resolved by determining the paths of sporadic meteors. If they followed parabolic or elliptical paths, they orbited the Sun; if their orbit were hyperbolic, they had an interstellar origin. The paths of sporadic meteors could be determined by an accurate measurement of both their velocities and radiants, but optical means were insufficiently precise to give unambiguous results. Fred L. Whipple, future director of the  Harvard College Observatory, a leading center of United States meteor research, attempted state-of-the-art optical studies of meteors with the Super Schmidt camera, but the first one was not operational until May 1951, at Las Cruces, New Mexico.39
Radar astronomers, then, attempted to accomplish what optical methods had failed to achieve. Such has been the pattern of radar astronomy to the present. Between 1948 and 1950, Lovell, Davies, and Mary Almond, a doctoral student, undertook a long series of sporadic meteor velocity measurements. They found no evidence for a significant hyperbolic velocity component; that is, there was no evidence for sporadic meteors coming from interstellar space. They then extended their work to fainter and smaller meteors with similar results.
The Jodrell Bank radar meteor studies determined unambiguously that meteors form part of the solar system. As Whipple declared in 1955, "We may now accept as proven the fact that bodies moving in hyperbolic orbits about the sun play no important role in producing meteoric phenomena brighter than about the 8th effective magnitude."40 Astronomers describe the brightness of a body in terms of magnitude; the larger the magnitude, the fainter the body.
The highly convincing evidence of the Jodrell Bank scientists was corroborated by Canadian radar research carried out by researchers of the Radio and Electrical Engineering Division of the National Research Council under Donald W. R. McKinley. McKinley had joined the Council's Radio Section (later Branch) before World War II and, like Lovell, had participated actively in wartime radar work.
McKinley conducted his meteor research with radars built around Ottawa in 1947 and 1948 as part of various National Research Council laboratories, such as the Flight Research Center at Arnprior Airport. Earle L. R. Webb, Radio and Electrical Engineering Division of the National Research Council, supervised the design, construction, and operation of the radar equipment. From as early as the summer of 1947, the Canadian radar studies were undertaken jointly with Peter M. Millman of the Dominion Observatory. They coordinated spectrographic, photographic, radar, and visual observations. The National Research Council investigators employed the Jodrell Bank technique to determine meteor velocities, a benefit of following in the footsteps of the British.41
Their first radar observations took place during the Perseid shower of August 1947, as the first radar station reached completion. Later studies collected data from the Geminid shower of December 1947 and the Lyrid shower of April 1948, with more radar stations brought into play as they became available. Following the success of Jodrell Bank,  McKinley's group initiated their own study of sporadic meteors. By 1951, with data on 10,933 sporadic meteors, McKinley's group reached the same conclusion as their British colleagues: meteors were part of the solar system. Soon, radar techniques became an integral part of Canadian meteor research with the establishment in 1957 of the National Research Council Springhill Meteor Observatory outside Ottawa. The Observatory concentrated on scientific meteor research with radar, visual, photographic, and spectroscopic methods.42
These meteor studies at Jodrell Bank and the National Research Council, and only at those institutions, arose from the union of radar and astronomy; they were the beginnings of radar astronomy. Radar studies of meteors were not limited to Jodrell Bank and the National Research Council, however. With support from the National Bureau of Standards, in 1957 Harvard College Observatory initiated a radar meteor project under the direction of Fred Whipple. Furthermore, radar continues today as an integral and vital part of worldwide meteor research. Its forte is the ability to determine orbits better than any other technique. In the last five years, a number of recently built radars have studied meteors in Britain (MST Radar, Aberytswyth, Wales), New Zealand (AMOR, Meteor Orbit Radar, Christchurch), and Japan (MU Radar, Shigaraki), not to mention earlier work in Czechoslovakia and Sweden.43
Unlike the Jodrell Bank and National Research Council cases, the radar meteor studies started in the United States in the early 1950s were driven by civilian scientists doing ionospheric and communications research and by the military's desire for jam-proof, point-to-point secure communications. While various military laboratories undertook their own research programs, most of the civilian U.S. radar meteor research was carried out at Stanford University and the National Bureau of Standards, where investigators fruitfully cross-fertilized ionospheric and military communications research. The Stanford case is worth examining not only for its later connections to radar astronomy, but also for its pioneering radar study of the Sun that arose out of an interest in ionospheric and radio propagation research.
In contrast to the Stanford work, many radar meteor experiments carried out in the United States in the 1940s were unique events. As early as August and November 1944, for instance, workers in the Federal Communications Commission Engineering Department associated visual observations of meteors and radio bursts. In January 1946, Oliver Perry Ferrell of the Signal Corps reported using a Signal Corps SCR-270B radar to detect meteor ionization trails.44 The major radar meteor event in the United States and elsewhere,  however, was the spectacular meteor shower associated with the Giacobini-Zinner comet.
On the night of 9 October 1946, 21 Army radars were aimed toward the sky in order to observe any unusual phenomena. The Signal Corps organized the experiment, which fit nicely with their mission of developing missile detection and ranging capabilities. The equipment was operated by volunteer crews of the Army ground forces, the Army Air Forces, and the Signal Corps located across the country in Idaho, New Mexico, Texas, and New Jersey. For mainly meteorological reasons, only the Signal Corps SCR-270 radar successfully detected meteor ionization trails. No attempt was made to correlate visual observations and radar echoes. A Princeton University undergraduate, Francis B. Shaffer, who had received radar training in the Navy, analyzed photographs of the radar screen echoes at the Signal Corps laboratory in Belmar, New Jersey.
This was the first attempt to utilize microwave radars to detect astronomical objects. The equipment operated at 1,200 MHz (25 cm), 3,000 MHz (10 cm), and 10,000 MHz (3 cm), frequencies in the L, S, and X radar bands that radar astronomy later used. "On the basis of this night's experiments," the Signal Corps experimenters decided, "we cannot conclude that microwave radars do not detect meteor-formed ion clouds."45
In contrast to the Signal Corps experiment, radar meteor studies formed part of ongoing research at the National Bureau of Standards. Organized from the Bureau's Radio Section in May 1946 and located at Sterling, Virginia, the Central Radio Propagation Laboratory (CRPL) division had three laboratories, one of which concerned itself exclusively with ionospheric research and radio propagation and was especially interested in the impact of meteors on the ionosphere. In October 1946, Victor C. Pineo and others associated with the CRPL used a borrowed SCR-270-D Signal Corps radar to observe the Giacobinid meteor shower. Over the next five years, Pineo continued research on the effects of meteors on the ionosphere, using a standard ionospheric research instrument called an ionosonde and publishing his results in Science.
Pineo's interest was in ionospheric physics, not astronomy. Underwriting his research at the Ionospheric Research Section of the National Bureau of Standards was the Air Force Cambridge Research Center (known later as the Cambridge Research Laboratories and today as Phillips Laboratory). His meteor work did not contribute to knowledge about the origin of meteors, as such work had in Britain and Canada, but it supported efforts to create secure military communications using meteor ionization trails.46 Also, it related to similar research being carried out concurrently at Stanford University.
The 1946 CRPL experiment, in fact, had been suggested by Robert A. Helliwell of the Stanford Radio Propagation Laboratory (SRPL). Frederick E. Terman, who had headed the Harvard Radio Research Laboratory and its radar countermeasures research during the war, "virtually organized radio and electronic engineering on the West Coast" as  Stanford Dean of Engineering, according to historian C. Stewart Gillmor. Terman negotiated a contract with the three military services for the funding of a broad range of research, including the SRPL's long-standing ionospheric research program.47
Helliwell, whose career was built on ionospheric research, was joined at the SRPL by Oswald G. Villard, Jr. Villard had earned his engineering degree during the war for the design of an ionosphere sounder. As an amateur radio operator in Cambridge, Massachusetts, he had noted the interference caused by meteor ionizations at shortwave frequencies called Doppler whistles.48
In October 1946, during the Giacobinid meteor shower, Helliwell, Villard, Laurence A. Manning, and W. E. Evans, Jr., detected meteor ion trails by listening for Doppler whistles with radios operating at 15 MHz (20 meters) and 29 MHz (10 meters). Manning then developed a method of measuring meteor velocities using the Doppler frequency shift of a continuous-wave signal reflected from the ionization trail. Manning, Villard, and Allen M. Peterson then applied Manning's technique to a continuous-wave radio study of the Perseid meteor shower in August 1948. The initial Stanford technique was significantly different from that developed at Jodrell Bank; it relied on continuous-wave radio, rather than pulsed radar, echoes.49
One of those conducting meteor studies at Stanford was Von R. Eshleman, a graduate student in electrical engineering who worked under both Manning and Villard. While serving in the Navy during World War II, Eshleman had studied, then taught, radar at the Navy's radar electronics school in Washington, DC. In 1946, while returning from the war on the U.S.S. Missouri, Eshleman unsuccessfully attempted to bounce radar waves off the Moon using the ship's radar. Support for his graduate research at Stanford came through contracts between the University and both the Office of Naval Research and the Air Force.
Eshleman's dissertation considered the theory of detecting meteor ionization trails and its application in actual experiments. Unlike the British and Canadian meteor studies, the primary research interest of Eshleman, Manning, Villard, and the other Stanford investigators was information about the winds and turbulence in the upper atmosphere. Their investigations of meteor velocities, the length of ionized meteor trails, and the fading and polarization of meteor echoes were part of that larger research interest, while Eshleman's dissertation was an integral part of the meteor research program.
Eshleman also considered the use of meteor ionization trails for secure military communications. His dissertation did not explicitly state that application, which he took up after completing the thesis. The Air Force supported the Stanford meteor research mainly to use meteor ionization trails for secure, point-to-point communications. The Stanford meteor research thus served a variety of scientific and military purposes simultaneously.50
 The meteor research carried out at Stanford had nontrivial consequences. Eshleman's dissertation has continued to provide the theoretical foundation of modern meteor burst communications, a communication mode that promises to function even after a nuclear holocaust has rendered useless all normal wireless communications. The pioneering work at Stanford, the National Bureau of Standards, and the Air Force Cambridge Research Laboratories received new attention in the 1980s, when the Space Defense Initiative ("Star Wars") revitalized interest in using meteor ionization trails for classified communications. Non-military applications of meteor burst communications also have arisen in recent years. 51
Early meteor burst communications research was not limited to Stanford and the National Bureau of Standards. American military funding of early meteor burst communications research extended beyond its shores to Britain. Historians of Jodrell Bank radio astronomy and meteor radar research stated that radio astronomy had surpassed meteor studies at the observatory by 1955. However, that meteor work persisted until 1964 through a contract with the U.S. Air Force, though as a cover for classified military research.52
Auroras provided additional radar targets in the 1950s. A major initiator of radar auroral studies was Jodrell Bank. As early as August 1947, while conducting meteor research, the Jodrell Bank scientists Lovell, Clegg, and Ellyett received echoes from an aurora display. Arnold Aspinall and G. S. Hawkins then continued the radar auroral studies at Jodrell Bank in collaboration with W. B. Housman, Director of the Aurora Section of the British Astronomy Association, and the aurora observers of that Section. In Canada, McKinley and Millman also observed an aurora during their meteor research in April 1948.53