Music and the Magnetosphere

I arrived in Iowa with the intention of studying the physics of music in early 1954. I had a degree in music, and at the time, the only thing required to be a graduate student in anything at the State University of Iowa was to have a bachelor's degree. Well, I had a bachelor's degree, so I signed up as a physics graduate student, and enrolled in undergraduate physics, mathematics, and chemistry courses. For three years at the North Texas State College School of Music, I had helped my flute teacher and mentor, Dr. George Morey, by teaching the secondary flutists. Upon my graduation, he arranged a student teaching position for me at SUI, his Alma Mater. Shortly after arrival, I played my flute for Himie Voxman, the Chairman of the SUI Music Department, and obtained a chair in the SUI orchestra (as second flutist, first chair positions being reserved for music majors). Later Chairman Voxman informed me that the flute teaching position I was supposed to take over had disappeared. The previous graduate student had decided to join the Iowa faculty, and would be continuing her role of teaching the flute students. Professor Betty Bang later became the President of the National Flute Association, and continues her illustrious career at the University of Iowa to this day.

At that time, Frank McDonald had recently arrived from Minnesota with a position as Research Associate. There were other Van Allen graduate students: Les Meredith was just finishing; Joe Kasper, Bill Webber, and Ernie Ray were well underway; Larry Cahill would arrive shortly; and George Ludwig was about to finish his undergraduate studies. Obviously, I was just beginning.

Dr. Van Allen was on sabbatical leave at Project Matterhorn at the time I arrived. Upon return, he found that he had some new graduate students, so he called them into his office, and put a list of potential projects and topics needing attention on the blackboard. Van Allen had developed a technique of getting extra altitude out of rockets by launching them from balloons, and he called the combination rockoons. He had the idea of launching rockoons, not only with just six-inch rockets, but also with easier-to-use and cheaper rockets that were only three-inches in diameter. One of the potential projects was to miniaturize the proton precession-magnetometer, which had just been invented. Larry Cahill thought that was interesting, so he took over that project of fitting the magnetometer into the three-inch rockets.

Another project was to use these small rockets to make a latitude survey of cosmic rays, i.e., to obtain the spectrum of cosmic rays by measuring the magnetic latitude dependence and using the Earth's magnetic field as a spectrometer. Not knowing any better, the flute player said, “OK, I'll do that.” (Dr. Van Allen probably did not know that this student was fresh out of music school.) The idea was to launch the three-inch Loki rockets from Skyhook balloons from a Navy ship during a voyage to Thule, Greenland in 1955. Even in those days, money was needed to do things. Therefore, Dr. Van Allen wrote a proposal to NSF. Figure 1 shows a copy of that proposal. Only one and a half pages, asking for only $2000. The estimated costs were later slightly reduced to make room for another item: fifteen percent university overhead. Dr. William H. Pickering, Director of JPL, helped obtain surplus Loki Phase I rockets, and that reduced the launch costs.

After designing an adapter for the three-quarter inch tip of the Loki rocket permitting a three-inch-diameter instrumentation, the next task was to try to package a Geiger tube, voltage sources, and telemetry system in a light-weight stack capable of withstanding the 270 g acceleration of the Loki rocket. The individual components were tested to 1000 g in a centrifuge and to temperatures below minus fifty degrees Centigrade (it is very cold at balloon altitudes).

Figure 1. Proposal for the first Loki rockoon program.

Hoping to detect the 'soft radiation' discovered on a previous rockoon expedition [Meredith et al, 1955; Van Allen, 1957], the nose cones were made of aluminum only 32 thousandths of an inch thick. Transistors were unknown, so the signal circuits, the transmitters, and high-voltage supplies all required vacuum tubes. The transmitter coils
were wound by hand and then adjusted for proper frequency (74 Mhz) and maximum power (one to two watts). The adapter included electrical insulation so the rocket motor served as one half of a dipole antenna. The broad lawn in front of the Old Capitol was used to simulate “space” for testing the transmitters.

The instrumentation assembly line for my ten little three-inch rockets occupied one of the lab benches in the basement of the Physics Building (Figure 2). The entire instrumentation assembly weighed 6.8 pounds.

Frank McDonald, Joe Kasper, and George Ludwig were also on the 1955 expedition. Joe Kasper served many roles, and also volunteered to help me with the contents of the classes being missed while on the ship. George Ludwig was helping Frank McDonald with instrumentation for the large six-inch Deacon rockets. George's presence on the expedition was very helpful for everyone. We four proceeded to go to the U.S.S. Ashland (LSD‑1) at Norfolk, Virginia, and found, to our dismay, that some of the three-inch rockets had arrived without their high-altitude fins, and only had their one-inch stubby fins for low-altitude launches. Frank McDonald and I ran into a Norfolk hardware store for some sheet aluminum, came back, cut some fins, and bolted them on. We took the precaution to use a hand drill, rather than an electric drill, to make the screw holes in order to avoid any possibility of accidental ignition.

Thus in August 1955, I found myself on a ship west of Greenland launching balloons carrying rockets. There were many experiences during this expedition, including coping with the ship’s rolling up to 45 degrees in the large waves left behind by hurricane Ione. Some of the launches were made during periods with substantial rolling (Figure 4), but it was calmer for the Loki rockoon launches (Figure 5). Figure 3 is a page from my notebook at the time of the first Loki launch. It is easy to imagine my elation when Ludwig reported that he had seen a rocket trail. So it did ignite ... that had been the first question: would it ignite? (with good cause, the first rockoon in 1952 did not ignite). But then there was great consternation: we could not find the signal. What happened? After an agonizing 20 seconds, I found the signal at a different frequency, and began recording the data from my first rocket. As an ex‑music student, I was obviously elated to be receiving real cosmic ray counts.

One particularly exciting experience came from Van Allen's idea of launching two-stage rockets from balloons. We had the large six-inch Deacon rockets, so it only called for sticking three-inch Loki rockets on top of them. As a music student, I, of course, knew how to design a coupler with a lanyard to ignite the second stage when they separated by differential air drag! We used a pressure sensitive switch as usual, and for extra safety, I devised a g-switch

apparently what happened, the expected velocity being over 8000 feet per second. Something we made may have achieved an altitude record at the time, but nothing survived to prove it. Also, “Retrospectively, it appears likely that this inexpensive technique, given a heat-resistant nose cone, would have resulted in discovery of the geomagnetically trapped radiation.” [Van Allen, 1983, p26].

Frank McDonald has documented another exciting but unwelcome event: the on-board accidental firing of a Loki rocket [McDonald, 1996]. In summary, this event consisted of the sequence:

1. Commander Ebel knelt down to adjust the timers in the control box and to check the igniter wires emerging from the Loki rocket on which was mounted the instrumentation containing the highest-power transmitter I had been able to make.

2. RF current in excess of 0.2 Ampere from the transmitter found its way up the igniter wires to the Dupont #201 Electrical Match setting it off. This set off the bag of igniter powder which then set off the main rocket propellant. The normal burning time of a Loki is less than one second.

3. Commander Ebel was badly burned (he made a complete recovery). The rocket’s blast centered on his shoulder where it burned through his thick arctic clothing and embedded bits of igniter wire into his flesh. Joe Kasper, standing nearby, had his eardrums ruptured, and his coat blown off. I was six feet to one side, and suffered only mild noise trauma.

4. The rocket accelerated sternward at such a rate that the tail fins I had sharpened to knife edges sliced through the saw-horses, went a few feet more and sliced through the cable of a phone held by a sailor who had been telling the bridge how to steer the ship to keep the balloon vertical. He instinctively leapt backward, but was lucky, and fell into a gun turret rather than into the icy arctic water. He was very lucky not to have been standing a few inches closer to the rocket’s path.

5. The rocket hit a stack of empty helium bottles and exploded sending parts of the rocket and burning propellant various places including to the balloon which caught on fire. The balloon was quickly cut loose.

6. The rocket to instrumentation adapter bounced off a helium bottle and landed on the bow of the ship where I found it the next day.

7. Later, a report was filed which may have helped lead to strengthening the precautions against the accidental ignition of explosives by radio transmitters taken by both military and civilian organizations.

We tossed the tenth and last Loki rocket overboard, took Commander Ebel to the nearest port, and sailed home. The voyage had been eventful, and despite the unfortunate mishap, successful. It had yielded data for a master’s thesis [McIlwain, 1956] and helped develop the Loki rockoon technique. The most interesting non-scientific event had been the side trip to Pond Inlet to rescue a missionary (I was told that he had “lost his mind” over the

local custom of slipping the unwanted girl babies under the ice). The ship drove a herd of narwhals ahead of it on the way in. The Eskimos jumped into their kayaks, speared the narwhals, and later presented a narwhal tusk to the ship’s captain.

That winter, back in Iowa, on the morning of February 26, 1956, a breathless call was received from the University of Chicago, saying that a gigantic solar storm was bombarding the Earth with cosmic rays, and that we should launch something if at all possible to measure the solar particles which cannot penetrate the dense atmosphere. Knowing that such outbursts usually do not last very many hours, we searched for something that was all ready to go. The only thing that could be found was the unlaunched tenth Loki payload brought back from the expedition. We hurriedly took it to the Iowa football field and launched it with a cluster of rubber balloons. Fortunately, the weather was not bad. It was windy, however, and the Loki instrumentation with its dummy rocket body was carried across the field where it hit some trees, but it broke loose from the trees, and proceeded to go to high altitude where the balloons burst 99 minutes after launch. The solar bombardment had subsided by that time, but solar cosmic rays were still adding about 40 percent to the expected galactic cosmic ray counting rates and this furnished the data for my first published paper [Van Allen and McIlwain, 1956]. The Chicago group had also succeeded in getting instruments to high altitude that day, launching their balloon from Stagg Field [Meyer, Parker, and Simpson, 1956].

In 1950, Sidney Chapman, Van Allen, Lloyd Berkner and others had decided there should be an IGY (International Geophysical Year). In 1956, Dr. Van Allen presented his graduate students with the SUI IGY Program. The Scientific Purposes of the program included six cosmic ray studies using various vehicles (including satellites), two magnetic-field studies to measure currents in and above the ionosphere, and two soft-radiation studies. The Approved Operations included ground-launched Nike‑Cajun rockets, and two shipboard rockoon expeditions covering a wide range of magnetic latitudes. He described the capabilities and costs for six different rocket-vehicle combinations. The cost of a Loki‑rockoon, about $300 for the Loki and $200 for a 39 ft balloon, was a fraction of the cost of any of the others, such as the Nike‑Cajun, and thus was preferred for latitude surveys.

In January 1956, it was both an educational event and a real privilege to accompany Dr. Van Allen to Ann Arbor, Michigan for the Upper Atmosphere Rocket Research Panel’s historic symposium on the scientific uses of Earth satellites. Dr. Van Allen further enhanced the learning-by-doing process by having students attend meetings of the Panel he was unable to attend (Figure 10). Thus during 1956, Dr. Van Allen apparently developed enough confidence in his former music student to suggest that I take advantage of the opportunity of flying some of the Nike‑Cajun rockets from Fort Churchill, Canada. I thought that was a good idea, and began thinking about what would be interesting to do.

Sydney Chapman had spent some months in Iowa in 1954-5, and had told us all that was known about the aurora. What about studying the aurora? Nobody had actually measured auroral particles. I thought it would be very interesting to delve into that, particularly to measure more directly what was causing the aurora. This was not known at that time. A popular idea was that auroral light was produced by energetic 100 keV protons, but the rockoon discovery of soft radiation indicated the presence of electrons.

There was the practical problem of almost no way to measure low energy particles directly. When the characteristics of particles which can get down to the 100 kilometer region and no farther are estimated, it can be seen that such particles cannot penetrate very much at all, thus requiring essentially windowless detectors. Such detectors were rare in the laboratory, and were certainly not available for flight. So it was back to the laboratory where I started dreaming up detectors to measure the spectra of electrons and protons. With the help of a second-year electrical engineering student named Don Enemark and an undergraduate physics student, Don Stilwell, two copies of instruments capable of detecting auroral particles were designed, built, and calibrated in time to be ready for the first two flights scheduled for the fall of 1957.

On May 8, 1957 Dr. Van Allen had sent his description of the first of six Nike-Cajun operations planned for the launches beginning that fall to various officials, stating that “The chief SUI scientist for these operations will be Mr. C. E. McIlwain”. This is the first time anyone had referred to me as a scientist, much less Chief Scientist. As can be seen, Dr. Van Allen put a great deal of trust in his graduate students. Perhaps he had no choice (he was off on two long shipboard expeditions during the time period launching Loki rockoons). So, I took my rocket instrumentation up to Fort Churchill. Les Meredith, who as an SUI graduate student had helped discover the soft radiation [Meredith et al, 1955] was then at NRL. There, with Leo Davis, he had also developed low energy detectors, and they were already at Fort Churchill launching them on Aerobee rockets. When I arrived, they had just had a successful flight, and Les elatedly said "We have already found what causes the aurora. It is low energy electrons. You can just pack up and go back home". The Chief Scientist, however, sensed there were still some undiscovered things to learn about the aurora. I proceeded to check out my own instrumentation (Figures 11 and 12).

Figure 9. Deacon-Loki and balloon on the way to 70,000 feet and ignition of the rockets.

However, there were various minor problems. During the first Nike-Cajun flight, the Nike rocket decided to burn a bit more after separation, went up and hit the second stage breaking off the instrumentation. Searching in the muskeg the next day, I found the instrumentation and many biting flies. Some of the electronics still worked! The second Nike-Cajun took the payload into an aurora, but the Cajun rotated and pointed the detectors downward during part of the flight.

Fortunately, things went beautifully on the second expedition in February 1958. We got some nice quiescent aurora data, but I decided that we really wanted to get a bright aurora, an active aurora. So just visualize the scientists who were waiting around for me to get my last rocket off so they could fire theirs, and the impatience of the range safety people. Even though a graduate student, I still had control of when to launch. I told them, "Things are still not quite right”. We waited at T minus 5 minutes night after night, and they said "Come on, there is some aurora up there. Fire the thing", but I insisted on waiting, and was very lucky. Upon seeing an auroral breakup just to the south of Churchill, I finally decided it was the time to finish the countdown. Figure 13 is a picture of the launch. The burning Nike is at the bottom of the picture, the burning Cajun is at the top, and the trail of the sputtering Nike is in between. The bright aurora is approaching overhead. The Cajun got up to altitude right as the aurora came overhead. We received the very first measurements of particles producing a bright auroral display [Sullivan, 1961 p121; Hanle and Chamberlain 1981 p68; Van Allen 1995 p14486].

Thus, in the end I was very much luckier than Meredith and Davis had been. When holding at T minus five minutes, one can only guess exactly when to restart the countdown for the rocket to rendezvous with the auroral particles. Figure 14 is an all-sky camera picture of the encounter.

Luckily, the rocket remained pointing upward rather than downward. The detectors worked, and detected enormous fluxes of low energy electrons, with a different spectrum than both Meredith and I had found in diffuse aurora. Rather than having a distributed spectrum, this auroral spectrum had all the earmarks of being quasi‑monoenergetic [McIlwain, 1960b, McIlwain, 1960c]. This led to the conclusion that the electrons must have just fallen through a potential. I knew about electric fields parallel to the magnetic field at the time, but I also knew that you could not put anything about that subject in print. Theoreticians at the time knew for sure, that it was impossible to have parallel electric fields in a plasma.

Simultaneously, George Ludwig was helping Van Allen prepare Explorer I, the very first US spacecraft, or, at least, the first one that worked and went into orbit. When I got back from my Churchill expedition, they were busy looking at the data, and scratching their heads. "Here is the normal cosmic ray counting rate, but here it is zero. Are there periods when something is not working properly?" I pointed out that another possibility was that the flux might sometimes be very high, driving the Geiger tube into such hard saturation that it did not count at all. Whether it was failure or high fluxes could be answered by seeing the transition from normal to zero rates. Did it just drop suddenly as in a failure, or did it smoothly rise to higher rates, go into saturation, and finally give only zeros? Unfortunately, only tiny fragments of data were available as only the scattered Minitrack Stations were being used at the

Figure 14. All-sky camera photograph taken from Fort Churchill 3 minutes after the launch. II6.27F’s position is indicated by the small circle.

time. We had to wait, not only until Explorer II ‑‑‑ it went into the ocean ‑‑‑ but until Explorer III was launched with the tape recorder that George Ludwig had developed. Two film strip copies of its first readout, recorded in San Diego on March 28, 1958, were sent. One was sent to Van Allen, who had gone to Washington after the launch. The other was sent to Iowa, where Assistant Professor Ernie Ray, Joe Kasper, and I promptly grabbed the reel of film, put it on a microfilm reader, and anxiously began looking for a transition. And there it was. So we knew at once that there was something of very high intensity out there. I immediately took the spare payload, and put it in front of an x‑ray machine (a 250 kV DC machine I had installed for calibrating my rocket instrumentation) where I generated what became known as a Van Allen r vs R plot [Figure 8 in Van Allen, 1958]. The results showed that fluxes that would ideally produce more than 35,000 counts per second, instead drove the rate to zero. We knew we had measurements of an exciting new phenomenon.

Simultaneously, having no tools other than his slide rule in his Washington hotel room, Van Allen [1983] bought graph paper and a ruler at a local drug store, and carefully plotted the counting rates for the entire 102 minutes of data. At 3:00 AM, he had turned in for the night "with the conviction that our instruments on both Explorers I and III were working reliably and giving reproducible results but that we were encountering a mysterious physical effect of a real nature" [Van Allen, 1983 p66]. Returning to Iowa, Dr. Van Allen proudly showed Ernie Ray and me his graph. I then showed him my x‑ray machine results. He instantly agreed that the satellites were encountering very high fluxes.

Van Allen announced the discovery to the world at a National Academy Meeting in Washington on May 1, 1958 [Van Allen et al, 1958; Berland, 1962; Hanle and Chamberlain 1981 p58]. It was clear that a spacecraft was needed to go up and study this new phenomenon. The preceding fall, Nicholas Christofilos had asked "What would happen if we set a high-altitude atomic bomb off; would it inject many trapped particles? Of course it would. So let's try it and measure what happens". Project Argus was conceived to do just this, and was now put in motion. Van Allen proposed to launch a satellite with better detectors to measure, without saturation, the trapped radiation that was already up there, and to detect the electrons injected by the atomic bomb blasts. An explicit requirement was to launch in time to beat the moratorium on high-altitude nuclear explosions. This was because the United States wanted to somehow set off high-altitude nuclear explosions in the time period before the moratorium, but after the spacecraft was up. Van Allen’s proposal was accepted in part because most people felt the required schedule was impossible and refused to propose.

We had less than three months to design, build, test, and launch instrumentation that could measure both the newly discovered radiation and any electrons injected by nuclear explosions. George Ludwig and I were quite busy for a while. We soon learned, however, that trying to continue working longer than 16 hours a day tended to produce more negative than positive results. The work directly related to the bombs was, of course, done in great secrecy. Ever since the success of Explorer I, Iowa had a continuous stream of media people coming to the basement of the Physics Building where the hall had been converted into a laboratory. The media included Time Magazine, and Walter Cronkite (Van Allen’s TV interview was held only a few feet from the bench in Figure 2). It was fortunate that our furious efforts designing and making the Explorer IV spacecraft itself were not required to be kept secret.

We had no idea what was up there. What we knew about radiation belts then was that a Geiger tube would saturate upon entering them. The only upper limit we had was how much particle energy the magnetic field could hold. At these low altitudes and low magnetic latitudes, this was enormous. I decided to put on a detector that could look at low energy particles, but could not be easily saturated [McIlwain, 1960a]. This detector, consisting of a scintillator on a photomultiplier tube, looked into space through a nickel foil only one milligram per square centimeter thick (and fortunately did not rupture during launch even though there was no protective nose cone). A circuit of special diodes and multi-billion ohm resistors provided a wide dynamic range for the current to voltage conversion. Field effect transistors had not been invented, so a vacuum tube was required to take this voltage and drive one of the subcarrier oscillators feeding signals to the transmitter. Knowing vacuum tubes tend to drift, I included a miniature mechanical relay to periodically provide the zero signal level. This system performed well in orbit, and did not go near the upper limits of its dynamic range.

There had been a problem, however. During vibration tests, two parts in the photomultiplier tube failed. Knowing our urgent need (and perhaps our lofty official DX A2 priority rating), RCA quickly redesigned the tube and promptly delivered some to us. RCA later put the new design into production. It remained a standard item for rocket and satellite experiments for many years.

George Ludwig and I went to Cape Canaveral to help with the launch preparations. There we had many unique experiences, and witnessed some spectacular unintended fireworks generated by early ICBM test launch failures. Once, curious about a Redstone rocket on a neighboring launch pad, we climbed the gantry and found a dummy test capsule for manned flight. There, high above the ground inside the capsule, we tried to imagine what it would be like to have the rocket beneath us ignite and carry us into space.

During an Explorer IV press conference, we two students received little attention compared to that given to Wernher von Braun. Later, when von Braun and I were having a quick lunch at a roadside cafe, he told me "You are the important ones. I'm just the trucker".

Explorer IV was successfully launched 77 days after conception. This included time the engineers at the Redstone Arsenal needed for their tests when we shipped it to them after giving it a good luck blessing (Figure 15). The data arrived at SUI on analog tapes. The tapes were played back through various analog electronics which wrote their output on rolls of paper with eight pens. We hired a cadre of students to measure the switching times of the scalers off the long strip charts and pencil their measurements in standard MIT lab notebooks. We had to edit the notebooks and quickly erase data that looked as if it had anything to do with the bomb blasts. Explorer IV proceeded to measure a great deal of what was to be known about the radiation belts for some time. Perhaps the majority of what was known for the first two years about the radiation belts was from Explorer IV [Van Allen et al, 1959a]. It did most of the mapping, much of the composition estimates, and was up in time to measure the results of the high-altitude nuclear explosions.

In February 1959, there was a classified workshop at the Lawrence Livermore Laboratory. Earlier, Edward Teller had asked Ted Northrop to "see if you can find out how particles drift in longitude". Nobody knew. We knew that particles spiraling around the magnetic field lines would bounce and be trapped, but did not know how they would drift around the Earth. Ted Northrop found the key: the Rosenbluth longitudinal invariant. At the workshop, he gave an impromptu seminar on the invariant to Dr. Van Allen, me, and other interested people. Later, the invariant was described in the open literature (Northop and Teller, 1960).

This invariant formed the basis for devising a way of mapping trapped radiation, the B,L coordinate system [McIlwain, 1961]. There was the fortunate circumstance of Ted Northrop finding exactly what was needed, even though it was then known only in a few classified circles. The nuclear explosions gave markers on magnetic shells, which told us where the particles were drifting [Van Allen et al, 1959b], and provided one of the first confirmations that adiabatic invariants really worked. Explorer IV thus provided a firm observational basis for the B,L coordinate system [Van Allen, 1962].

In conclusion, it is now recognized that radiation belts are an important and common aspect in many parts of our universe. We at the State University of Iowa who were involved with the Explorer I, III, and IV spacecraft were exceedingly lucky be there to help produce mankind’s first view of this wondrous new phenomenon. The ex-musician would have liked more time to perform music, but he has never had regrets concerning his Iowa transformation.

Acknowledgments.

Dr. James A. Van Allen cannot be thanked too much for providing the many opportunities for research and personal development, and for his continued support over the years. Mary, my wife of 44 years, also deserves great credit for her assistance and encouragement. I thank Richard Maheu and Stephen Kerr for their help in preparing this paper. The National Science Foundation and the Office of Naval Research provided partial financial support for the research projects and we had operational support from the U. S. Navy and the U. S. Army. The operational support included the use of facilities, transportation, food, clothing, and photographers.

REFERENCES

Berland, T., The Scientific Life, Coward-McCann, Inc., New York, 1962.

Davis, L. R., D. E. Berg, and L. H. Meredith, Direct measurements of particle fluxes in and near auroras, Proc. Cospar Space Sci. Symposium, North Holland Pub. Co., Amsterdam, 721-35, 1960.

Hanle, P. A., and V. D. Chamberlain, Space Science Comes of Age, National Air and Space Museum , Smithsonian Institution Press, Washington, D. C., 194pp, 1981

McDonald, F. B., IMPs, EGOs, and Skyhooks, J. Geophys. Res., 101: No. A5, 10,521-30, 1996.

McIlwain, C. E., Cosmic ray intensity above the atmosphere at northern latitudes, M.S. thesis, University of Iowa, 55pp, 1956.

McIlwain, C. E., Scintillation counters in rockets and satellites, Institute of Radio Engineers Transactions in Nuclear Science, N57, 7:159, 1960a.

McIlwain, C. E., Direct measurement of protons and electrons in visible aurorae, Proc. Cospar Space Sci. Symposium, North Holland Pub. Co., Amsterdam, 715-20, 1960b.

McIlwain, C. E., Direct measurement of particles producing visible aurorae, J. Geophys. Res. 65, 2727-47, 1960c.

McIlwain, C. E., Coordinates for mapping the distribution of magnetically trapped particles, J. Geophys. Res., 66, 3681, 1961.

Meredith, L. H., M. B. Gottlieb, and J. A. Van Allen, Direct Detection of soft radiation above 50 kilometers in the auroral zone, Phys. Rev., 97, 201-5, 1955.

Meyer, P., E. N. Parker, and J. A. Simpson, Solar Cosmic Rays of February, 1956 and their propagation through interplanetary space, Phys. Rev., 104, No. 3, 768-783, 1956.

Northop, T. G. and E. Teller, Stability of the adiabatic motion of charged particles in the Earth’s field, Phys. Rev., 117, 215, 1960.

Sullivan, Walter; Assault on the Unknown, McGraw Hill, 460pp, 1961.

Van Allen, J. A., and C. E. McIlwain, Cosmic ray intensity of high altitudes on February 23, 1956, J. Geophys. Res. 61, 569-570, 1956.

Van Allen, J. A., Direct detection of auroral radiation with rocket equipment, Proc. Nat. Acad. Sciences, 43, 57-62, 1957.

Van Allen, J. A., G. H. Ludwig, E. C. Ray, and C. E. McIlwain, Observations of high intensity radiation by satellite 1958 Alpha and Gamma, Jet Propulsion, 588-592, 1958.

Van Allen, J. A., C. E. McIlwain, and G. H. Ludwig, Radiation observations with satellite 1958 Epsilon, J. Geophys. Res. 64, 271-286, 1959a.

Van Allen , J. A., C. E. McIlwain, and G. H. Ludwig, Satellite observations of electrons artificially injected into the geomagnetic field, National Academy Sciences Proceedings 45, 1152-66, l959b; and J. Geophys. Res. 64, 877-891, 1959b.

Van Allen, J. A., Dynamics, composition and origin of the geomagnetically-trapped corpuscular radiation, Transactions of the International Astronomical Union Vol. XIB, pages 99-136, 1962.

Van Allen, J. A., Origins of Magnetospheric Physics, Smithsonian Institution Press, 144pp Washington, D.C. 1983.

Van Allen, J. A., Early rocket observations of auroral bremsstrahlung and its absorption in the mesosphere, J. Geophys. Res, 100, 14485-97, 1995.

C. E. McIlwain, Center for Astrophysics and Space Sciences, University of California San Diego

La Jolla, California, 92093-0424 USA.

(e-mail: cmcilwain@ucsd.edu)