Moon
Introduction
Earth's sole natural satellite and nearest large celestial body. Known since prehistoric times, it is the brightest object in the sky after the Sun. It is designated by the symbol ☽. Its name in English, like that of Earth, is of Germanic and Old English derivation.

The Moon's desolate beauty has been a source of fascination and curiosity throughout history and has inspired a rich cultural and symbolic tradition. In past civilizations the Moon was regarded as a deity, its dominion dramatically manifested in its rhythmic control over the tides and the cycle of female fertility. Ancient lore and legend tell of the power of the Moon to instill spells with magic, to transform humans into beasts, and to send people's behaviour swaying perilously between sanity and lunacy (from the Latin luna, “Moon”). Poets and composers were invoking the Moon's romantic charms and its darker side, and writers of fiction were conducting their readers on speculative lunar journeys long before Apollo astronauts, in orbit above the Moon, sent back photographs of the reality that human eyes were witnessing for the first time.
Centuries of observation and scientific investigation have been centred on the nature and origin of the Moon. Early studies of the Moon's motion and position allowed the prediction of tides and led to the development of calendars. The Moon was the first new world on which humans set foot; the information brought back from those expeditions, together with that collected by automated spacecraft and remote-sensing observations, has led to a knowledge of the Moon that surpasses that of any other cosmic body except Earth itself. Although many questions remain about its composition, structure, and history, it has become clear that the Moon holds keys to understanding the origin of Earth and the solar system. Moreover, given its nearness to Earth, its rich potential as a source of materials and energy, and its qualifications as a laboratory for planetary science and a place to learn how to live and work in space for extended times, the Moon remains a prime location for humankind's first settlements beyond Earth orbit.

Distinctive features
The Moon is a spherical rocky body, probably with a small metallic core, revolving around Earth in a slightly eccentric orbit at a mean distance of about 384,000 km (238,600 miles). Its equatorial radius is 1,738 km (1,080 miles), and its shape is slightly flattened in a such a way that it bulges a little in the direction of Earth. Its mass distribution is not uniform—the centre of mass is displaced about 2 km (1.2 miles) toward Earth relative to the centre of the lunar sphere, and it also has surface mass concentrations, called mascons for short, that cause the Moon's gravitational field to increase over local areas. The Moon has no global magnetic field like that of Earth, but some of its surface rocks have remanent magnetism, which indicates one or more periods of magnetic activity in the past. The Moon presently has very slight seismic activity and little heat flow from the interior, indications that most internal activity ceased long ago.
Scientists now believe that more than four billion years ago the Moon was subject to violent heating—probably from its formation—which resulted in its differentiation, or chemical separation, into a less dense crust and a more dense underlying mantle. This was followed hundreds of millions of years later by a second episode of heating—this time from internal radioactivity—which resulted in volcanic outpourings of lava. The Moon's mean density is 3.34 grams per cubic cm, close to that of Earth's mantle. Because of the Moon's small size and mass, its surface gravity is only about one-sixth of the planet's; it retains so little atmosphere that the molecules of any gases present on the surface move without collision. In the absence of an atmospheric shield to protect the surface from bombardment, countless bodies ranging in size from asteroids to tiny particles have struck and cratered the Moon. This has formed a debris layer, or regolith, consisting of rock fragments of all sizes down to the finest dust. In the ancient past the largest impacts made great basins, some of which were later partly filled by the enormous lava floods. These great dark plains, called maria (singular mare [Latin: “sea”]), are clearly visible to the naked eye from Earth. The dark maria and the lighter highlands, whose unchanging patterns many people recognize as the “man in the moon,” constitute the two main kinds of lunar territory. The mascons are regions where particularly dense lavas rose up from the mantle and flooded into basins. Lunar mountains, located mostly along the rims of ancient basins, are tall but not steep or sharp-peaked, because all lunar landforms have been eroded by the unending rain of impacts. For additional orbital and physical data, see the table.

Principal characteristics of the Earth-Moon system
In addition to its nearness to Earth, the Moon is relatively massive compared with the planet—the ratio of their masses is much larger than those of other natural satellites to the planets that they orbit. The Moon and Earth consequently exert a strong gravitational influence on each other, forming a system having distinct properties and behaviour of its own. The table compares some salient characteristics of the two bodies.

Although the Moon is commonly described as orbiting Earth, it is more accurate to say that the two bodies orbit each other about a common centre of mass. Called the barycentre, this point lies inside Earth about 4,700 km (2,900 miles) from its centre. Also more accurately, it is the barycentre, rather than the centre of Earth, that follows an elliptical path around the Sun in accord with Kepler's laws of planetary motion. The orbital geometry of the Moon, Earth, and the Sun gives rise to the Moon's phases and to the phenomena of lunar and solar eclipses. The geometry and motions of the Earth-Moon system are illustrated in the figure.
The distance between the Moon and Earth varies rather widely because of the combined gravity of Earth, the Sun, and the planets. For example, in the last three decades of the 20th century, the Moon's apogee—the farthest distance that it travels from Earth in a revolution—ranged between about 404,000 and 406,700 km (251,000 and 252,700 miles), while its perigee—the closest that it comes to Earth—ranged between about 356,500 and 370,400 km (221,500 and 230,200 miles). Tidal interactions, the cyclic deformations in each body caused by the gravitational attraction of the other, have braked the Moon's spin such that it now rotates at the same rate as it revolves around Earth and thus always keeps the same side facing the planet. As discovered by the Italian-born French astronomer Gian Domenico Cassini in 1692, the Moon's spin axis precesses with respect to its orbital plane; i.e., its orientation changes slowly over time, tracing out a circular path. (For the empirical rules that Cassini formulated about the Moon's motion, see Cassini's laws.)
In accord with Kepler's second law, the eccentricity of the Moon's orbit results in its traveling faster in that part of its orbit nearer Earth and slower in the part farther away. Combined with the Moon's constant spin rate, these changes in speed give rise to an apparent oscillation, or libration, which over time allows an observer on Earth to see more than half of the lunar surface. In addition to this apparent turning motion, the Moon actually does rock slightly to and fro in both longitude and latitude, and the observer's vantage point moves with Earth's rotation. As a result of all these motions, more than 59 percent of the lunar surface can be seen at one time or another from Earth.
The orbital eccentricity also affects solar eclipses, in which the Moon passes between the Sun and Earth, casting a moving shadow across Earth's sunlit surface. If a solar eclipse occurs when the Moon is near perigee, observers along the path of the Moon's dark inner shadow (umbra) see a total eclipse. If the Moon is near apogee, it does not quite cover the Sun; the resulting eclipse is annular, and observers can see a thin ring of the solar disk around the Moon's silhouette.
The Moon and Earth presently orbit the barycentre in 27.322 days, the sidereal month, or sidereal revolution period of the Moon. Because the whole system is moving around the Sun once per year, the angle of illumination changes about one degree per day, so that the time from one full moon to the next is 29.531 days, the synodic month, or synodic revolution period of the Moon. As a result, the Moon's terminator—the dividing line between dayside and nightside—moves once around the Moon in this synodic period, exposing most locations to alternating periods of sunlight and darkness each nearly 15 Earth days long. The sidereal and synodic periods are slowly changing with time because of tidal interactions. Although tidal friction is slowing Earth's rotation, conservation of momentum dictates that the angular momentum of the Earth-Moon system remain constant. Consequently, the Moon is slowly receding from Earth, with the result that both the day and the month are getting longer. Extending this relationship back into the past, both periods must have been significantly shorter hundreds of millions of years ago—a hypothesis confirmed from measurements of the daily and tide-related growth rings of fossil corals.

Because the Moon's spin axis is almost perpendicular to the plane of the ecliptic (the plane of Earth's orbit around the Sun)—inclined only 11/2° from the vertical—the Moon has no seasons. Sunlight is always nearly horizontal at the lunar poles, which results in permanently cold and dark environments at the bottoms of deep craters.

Motions of the Moon
The study of the Moon's motions has been central to the growth of knowledge not only about the Moon itself but also about fundamentals of celestial mechanics and physics. As the stars appear to move westward because of Earth's daily rotation and its annual motion about the Sun, so the Moon slowly moves eastward, rising later each day and passing through its phases: new, first quarter, full, last quarter, and new again each month. The long-running Chinese, Chaldean, and Mayan calendars were attempts to reconcile these repetitive but incommensurate movements. From the time of the Babylonian astrologers and the Greek astronomers up to the present, investigators looked for small departures from the motions predicted. The English physicist Isaac Newton used lunar observations in developing his theory of gravitation in the late 17th century, and he was able to show some effects of solar gravity in perturbing the Moon's motion. By the 18th and 19th centuries the mathematical study of lunar movements, both orbital and rotational, was advancing, driven in part by the need for precise tables of the predicted positions of celestial bodies (ephemerides) for navigation. While theory developed with improved observations, many small and puzzling discrepancies continued to appear. It gradually became evident that some arise from irregularities in Earth's rotation rate, others from minor tidal effects on Earth and the Moon.
Space exploration brought a need for greatly increased accuracy, and, at the same time, the availability of fast computers and new observational tools provided the means for attaining it. Analytic treatments—mathematical modeling of the Moon's motions with a series of terms representing the gravitational influence of Earth, the Sun, and the planets—gave way to methods based on direct numerical integration of equations of motion for the Moon. Both methods required significant input based on observation, but use of the latter led to great increases in the accuracy of predictions. At the same time, optical and radio observations vastly improved—retroreflectors placed on the lunar surface by Apollo astronauts allowed laser ranging of the Moon from Earth, and new techniques of radio astronomy, including very long baseline interferometry (see telescope: Very long baseline interferometry), permitted observations of celestial radio sources as the Moon occulted them. These observations, having precisions on the order of centimetres, have enabled scientists to measure changes in the Moon's speed caused by terrestrial tidal momentum exchange, have advanced understanding of the theories of relativity, and are leading to improved geophysical knowledge of both the Moon and Earth.

The atmosphere
Though the Moon is surrounded by a vacuum higher than is usually created in laboratories on Earth, its atmosphere is extensive and of high scientific interest. During the two-week daytime period, atoms and molecules are ejected by a variety of processes from the lunar surface, ionized by the solar wind, and then driven by electromagnetic effects as a collisionless plasma. The position of the Moon in its orbit determines the behaviour of the atmosphere. For part of each month, when the Moon is on the sunward side of Earth, atmospheric gases collide with the undisturbed solar wind; in other parts of the orbit, they move into and out of the elongated tail of Earth's magnetosphere, an enormous region of space where the planet's magnetic field dominates the behaviour of electrically charged particles. In addition, the low temperatures on the Moon's nightside and in permanently shaded polar craters provide cold traps for condensable gases.
Instruments placed on the lunar surface by Apollo astronauts measured various properties of the Moon's atmosphere, but analysis of the data was difficult because the atmosphere's extreme thinness made contamination from Apollo-originated gases a significant factor. The main gases naturally present are neon, hydrogen, helium, and argon. The argon is mostly radiogenic; i.e., it is released from lunar rocks by the decay of radioactive potassium. Lunar night temperatures are low enough for the argon to condense but not the neon, hydrogen, or helium, which originate in the solar wind and remain in the atmosphere as gases unless implanted in soil particles.
In addition to the near-surface gases and the extensive sodium-potassium cloud detected around the Moon (see the section Effects of impacts and volcanism below), a small amount of dust circulates within a few metres of the lunar surface. This is believed to be suspended electrostatically.

The lunar surface
Large-scale features
With binoculars or a small telescope, an observer can see details of the Moon's near side in addition to the pattern of maria and highlands. As the Moon passes through its phases, the terminator moves slowly across the Moon's disk, its long shadows revealing the relief of mountains and craters. At full moon the relief disappears, replaced by the contrast between lighter and darker surfaces. Though the full moon is brilliant at night, the Moon is actually a dark object, reflecting only a few percent (albedo 0.07) of the sunlight that strikes it. Beginning with the Italian scientist Galileo's sketches in the early 17th century and continuing into the 19th century, astronomers mapped and named the visible features down to a resolution of a few kilometres, the best that can be accomplished when viewing the Moon telescopically through Earth's turbulent atmosphere. The work culminated in a great hand-drawn lunar atlas made by observers in Berlin and Athens. This was followed by a lengthy hiatus as astronomers turned their attention beyond the Moon until the mid-20th century, when it became apparent that human travel to the Moon might eventually be possible. In the 1950s another great atlas was compiled, this time a photographic one published in 1960 under the sponsorship of the U.S. Air Force.
Astronomers long debated whether the Moon's topographic features had been caused by volcanism. Only in the 20th century did the dominance of impacts in the shaping of the lunar surface become clear. Every highland region is heavily cratered—evidence for repeated collisions with large bodies. (The survival of similar large impact structures on Earth is relatively rare because of Earth's geologic activity and weathering.) The maria, on the other hand, show much less cratering and thus must be significantly younger. Mountains are mostly parts of the upthrust rims of ancient impact basins. Volcanic activity has occurred within the Moon, but the results are mostly quite different from those on Earth. The lavas that upwelled in floods to form the maria were extremely fluid. Evidence of volcanic mountain building as has occurred on Earth is limited to a few fields of small, low domes.

For millennia people wondered about the appearance of the Moon's unseen side. The mystery began to be dispelled with the flight of the Soviet space probe Luna 3 in 1959, which returned the first photographs of the far side. In contrast to the near side, the surface displayed in the Luna 3 images consisted mostly of highlands, with only small areas of dark mare material. Later missions showed that the ancient far-side highlands are scarred by huge basins but that these basins are not filled with lava.

Effects of impacts and volcanism
The dominant consequences of impacts are observed in every lunar scene. At the largest scale are the ancient basins, which extend hundreds of kilometres across. A beautiful example is Orientale Basin, or Mare Orientale, whose mountain walls can just be seen from Earth near the Moon's limb (the apparent edge of the lunar disk) when the lunar libration is favourable. Its multiring ramparts are characteristic of the largest basins; they are accented by the partial lava flooding of low regions between the rings. Orientale Basin appears to be the youngest large impact basin on the Moon.
Orientale's name arises from lunar-mapping conventions. During the great age of telescopic observation in the 17th–19th centuries, portrayals of the Moon usually showed south at the top because the telescopes inverted the image. East and west referred to those directions in the sky—i.e., the Moon moves eastward and so its leading limb was east, and the portion of the basin that could be seen from Earth was accordingly called Mare Orientale. For mapping purposes lunar coordinates were taken to originate near the centre of the near-side face, at the intersection of the equator and a meridian defined by the mean librations. A small crater, Mösting A, was agreed upon as the reference point. With the Moon considered as a world, rather than just a disk moving across the sky, east and west are interchanged. Thus, Orientale, despite its name, is located at west lunar longitudes.

Smaller impact features, ranging in diameter from tens of kilometres to microscopic size, are described by the term crater. The relative ages of lunar craters are indicated by their form and structural features. Young craters have rugged profiles and are surrounded by hummocky blankets of debris, called ejecta, and long light-coloured rays made by expelled material hitting the lunar surface. Older craters have rounded and subdued profiles, the result of continued bombardment.
A crater's form and structure also yield information about the impact process. When a body strikes a much larger one at speeds of many kilometres per second, the available kinetic energy is enough to completely melt, even partly vaporize, the impacting body along with a small portion of its target material. On impact, a melt sheet is thrown out, along with quantities of rubble, to form the ejecta blanket around the contact site. Meanwhile, a shock travels into the subsurface, shattering mineral structures and leaving a telltale signature in the rocks. The initial cup-shaped cavity is unstable and, depending on its size, evolves in different ways. A typical end result is the great crater Aristarchus, with slumping terraces in its walls and a central peak. Aristarchus is about 40 km (25 miles) in diameter and 4 km (2.5 miles) deep. (For additional details about the formation of impact craters and large impact basins, see the article impact crater.)

The region around Aristarchus shows a number of peculiar lunar features, some of which have origins not yet well explained. The Aristarchus impact occurred on an elevated, old-looking surface surrounded by lavas of the northern part of the mare known as Oceanus Procellarum. These lava flows inundated the older crater Prinz, whose rim is now only partly visible. At one point on the rim, an apparently volcanic event produced a crater; subsequently, a long, winding channel, called a sinuous rille, emerged to flow across the mare. Other sinuous rilles are found nearby, including the largest one on the Moon, discovered by the German astronomer Johann Schröter in 1787. Named in his honour, Schröter's Valley is a deep, winding channel, hundreds of kilometres long, with a smaller inner channel that meanders just as slow rivers do on Earth. The end of this “river” simply tapers away to nothing and disappears on the mare plains. In some way that remains to be accounted for, hundreds of cubic kilometres of fluid and excavated mare material vanished.
The results of seismic and heat-flow measurements suggest that any volcanic activity that persists on the Moon is slight by comparison with that of Earth. Over the years reliable observers have reported seeing transient events of a possibly volcanic nature, and some spectroscopic evidence for them exists. In the late 1980s a cloud of sodium and potassium atoms was observed around the Moon, but it was not necessarily the result of volcanic emissions. It is possible that interactions of the lunar surface and the solar wind produced the cloud. In any case, the question of whether the Moon is volcanically active remains open.
Telescopic observers beginning in the 19th century applied the term rille to several types of trenchlike lunar features. In addition to sinuous rilles, there are straight and branching rilles that appear to be tension cracks, and some of these—such as Rima Hyginus and the rilles around the floor of the large old crater Alphonsus—are peppered with rimless eruption craters. Though the Moon shows both tension and compression features (low wrinkle ridges, usually near mare margins, may result from compression), it gives no evidence of having experienced the large, lateral motions of plate tectonics marked by faults in Earth's crust.

Among the most enigmatic features of the lunar surface are several light, swirling patterns with no associated topography. A prime example is Reiner Gamma, located in the southeastern portion of Oceanus Procellarum. Whereas other relatively bright features exist—e.g., crater rays—they are explained as consequences of the impact process. Features such as Reiner Gamma have no clear explanation. Some scientists have suggested that they are the marks of comet impacts, in which the impacting body was large in size but had so little density as to produce no crater. Reiner Gamma is also unusual in that it coincides with a large magnetic anomaly (region of magnetic irregularity) in the crust.

Small-scale features
On a small-to-microscopic scale, the properties of the lunar surface are governed by a combination of phenomena—impact effects due to the arrival, at speeds up to tens of kilometres per second, of meteoritic material ranging in size down to fractions of a micrometre; bombardment by solar-wind, cosmic-ray, and solar-flare particles; ionizing radiation; and temperature extremes. Subject to no meteorological effects and unprotected by a substantial atmosphere, the uppermost surface reaches almost 400 kelvins (K; 260 °F, 127 °C) during the day and plunges to below 100 K (−279 °F, −173 °C) at night. The top layer of regolith, however, serves as an efficient insulator because of its high porosity (large number of voids, or pore spaces, per unit of volume). As a result, the daily temperature swings penetrate into the soil to less than one metre (about three feet).
Long before human beings could observe the regolith firsthand, Earth-based astronomers concluded from several kinds of measurements that the Moon's surface must be very peculiar. The evidence from photometry (brightness measurements) is particularly striking. From Earth the fully illuminated Moon is 11 times as bright as one only half illuminated, and it appears bright up to the edge of the disk. Measurements of the amount of sunlight reflected back in the direction of illumination indicate the reason: on a small scale the surface is extremely rough, and light reflected from within mineral grains and deep cavities remains shadowed until the illumination source is directly behind the observer—i.e., until the full moon—at which time light abruptly reflects out of the cavities. The polarization properties of the reflected light show that the surface is rough even at a microscopic scale.
Before spacecraft landed on the Moon, astronomers had no straightforward means by which to measure the depth of the regolith layer. Nevertheless, after the development of infrared detectors allowed them to make accurate thermal observations through the telescope, they could finally draw some reasonable conclusions about the outer surface characteristics. As Earth's shadow falls across the Moon during a lunar eclipse, the lunar surface cools rapidly, but the cooling is uneven, being slower near relatively young craters where exposed rock fields are to be expected. This behaviour could be interpreted to show that the highly insulating layer is fairly shallow, a few metres at most. Though not all astronomers accepted this conclusion at first, it was confirmed in the mid-1960s when the first robotic spacecraft soft-landed and sank only a few centimetres instead of disappearing completely into the regolith.

Lunar rocks and soil
General characteristics
As noted above, the lunar regolith comprises rock fragments in a continuous distribution of particle sizes. It includes a fine fraction— dirtlike in character—that, for convenience, is called soil. The term, however, does not imply a biological contribution to its origin as it does on Earth.

Almost all the rocks at the lunar surface are igneous—they formed from the cooling of lava. (By contrast, the most prevalent rocks exposed on Earth's surface are sedimentary, which required the action of water or wind for their formation.) The two most common kinds are basalts and anorthosites. The lunar basalts, relatively rich in iron and many also in titanium, are found in the maria. In the highlands the rocks are largely anorthosites, which are relatively rich in aluminum, calcium, and silicon. Some of the rocks in both the maria and the highlands are breccias; i.e., they are composed of fragments produced by an initial impact and then reagglomerated by later impacts. The physical compositions of lunar breccias range from broken and shock-altered fragments, called clasts, to a matrix of completely impact-melted material that has lost its original mineral character. The repeated impact history of a particular rock can result in a breccia welded either into a strong, coherent mass or into a weak, crumbly mixture in which the matrix consists of poorly aggregated or metamorphosed fragments. Massive bedrock—that is, bedrock not excavated by natural processes—is absent from the lunar samples so far collected.
Lunar soils are derived from lunar rocks, but they have a distinctive character. They represent the end result of micrometeoroid bombardment and of the Moon's thermal, particulate, and radiation environments. In the ancient past the stream of impacting bodies, some of which were quite large, turned over—or “gardened”—the lunar surface to a depth that is unknown but may have been as much as tens of kilometres. As the frequency of large impacts decreased, the gardening depth became shallower. It is estimated that the top centimetre of the surface at a particular site presently has a 50 percent chance of being turned over every million years, while during the same period the top millimetre is turned over a few dozen times and the outermost tenth of a millimetre is gardened hundreds of times. One result of this process is the presence in the soil of a large fraction of glassy particles forming agglutinates, aggregates of lunar soil fragments set in a glassy cement. The agglutinate fraction is a measure of soil maturity—i.e., of how long a particular sample has been exposed to the continuing rain of tiny impacts.

Although the chemical and mineralogical properties of soil particles show that they were derived from native lunar rocks, they also contain small amounts of meteoritic iron and other materials from impacting bodies. Volatile substances from comets, such as carbon compounds and water, would be expected to be mostly driven off by the heat generated by the impact, but the small amounts of carbon found in lunar soils may include atoms of cometary origin.
A fascinating and scientifically important property of lunar soils is the implantation of solar wind particles. Unimpeded by atmospheric or electromagnetic effects, protons, electrons, and atoms arrive at speeds of hundreds of kilometres per second and are driven into the outermost surfaces of soil grains. Lunar soils thus contain a collection of material from the Sun. Because of their gardening history, soils obtained from different depths have been exposed to the solar wind at the surface at different times and therefore can reveal some aspects of ancient solar behaviour. In addition to its scientific interest, this implantation phenomenon may have implications for long-term human habitation of the Moon in the future, as discussed in the section Lunar resources below.
The chemical and mineral properties of lunar rocks and soils hold clues to the Moon's history, and the study of lunar samples has become an extensive field of science. To date, scientists have obtained lunar material from three sources: six U.S. Apollo Moon-landing missions (1969–72), which collectively brought back almost 382 kg (842 pounds) of samples; three Soviet Luna automated sampling missions (1970–76), which returned about 300 grams (0.66 pound) of material; and scientific expeditions to Antarctica, which have collected meteorites on the ice fields since 1969. Some of these meteorites are rocks that were blasted out of the Moon by impacts, found their way to Earth, and have been confirmed as lunar in origin by comparison with the samples returned by spacecraft.
The mineral constituents of a rock reflect its chemical composition and thermal history. Rock textures—i.e., the shapes and sizes of mineral grains and the nature of their interfaces—provide clues as to the conditions under which the rock cooled and solidified from a melt. The most common minerals in lunar rocks are silicates (including pyroxene, olivine, and feldspar) and oxides (including ilmenite, spinel, and a mineral discovered in rocks collected by Apollo 11 astronauts and named armalcolite, a word made from the first letters of the astronauts' surnames—Armstrong, Aldrin, and Collins). The properties of lunar minerals reflect the many differences between the history of the Moon and that of Earth. Lunar rocks appear to have formed in the total absence of water. Many minor mineral constituents in lunar rocks reflect the history of formation of the lunar mantle and crust (see the section Origin and evolution below), and they confirm the hypothesis that most rocks now found at the lunar surface formed under reducing conditions—i.e., those in which oxygen was scarce.

Main groupings
The materials formed of these minerals are classified into four main groups: (1) basaltic volcanics, the rocks forming the maria, (2) pristine highland rocks uncontaminated by impact mixing, (3) breccias and impact melts, formed by impacts that disassembled and reassembled mixtures of rocks, and (4) soils, defined as unconsolidated aggregates of particles less than 1 cm (0.4 inch) in size, derived from all the rock types. All these materials are of igneous origin, but their melting and crystallization history is complex.
The mare basalts, when in liquid form, were much less viscous than typical lavas on Earth; they flowed like heavy oil. This was due to the low availability of oxygen and the absence of water in the regions where they formed. The melting temperature of the parent rock was higher than in Earth's volcanic source regions. As the lunar lavas rose to the surface and poured out in thin layers, they filled the basins of the Moon's near side and flowed out over plains, drowning older craters and embaying the basin margins. Some of the lavas contained dissolved gases, as shown by the presence of vesicles (bubbles) in certain rock samples and by the existence of pyroclastic glass (essentially volcanic ash) at some locations. There are also rimless craters, surrounded by dark halos, which do not have the characteristic shape of an impact scar but instead appear to have been formed by eruptions.
Most mare basalts differ from Earthly lavas not only in the lack of evidence of water but also in depletion of other volatile substances such as potassium, sodium, and carbon compounds. They also are depleted of elements classified geochemically as siderophiles—elements that tend to affiliate with iron when rocks cool from a melt. (This siderophile depletion is an important clue to the history of the Earth-Moon system, as discussed in the section Origin and evolution, below.)Some lavas were relatively rich in elements whose atoms do not readily fit into the crystal lattice sites of the common lunar minerals and are thus called incompatible elements. They tend to remain uncombined in a melt—of either mare or highland composition—and to become concentrated in the last portions to solidify upon cooling. Lunar scientists gave these lavas the name KREEP, an acronym for potassium (chemical symbol K), rare-earth elements, and phosphorus (P). These rocks give information as to the history of partial melting in the lunar mantle and the subsequent rise of lavas through the crust. Radiometric age dating (see below Mission results) reveals that the great eruptions that formed the maria occurred hundreds of millions of years later than the more extensive heating that produced the lunar highlands.
Ancient highland material that is considered pristine is relatively rare because most highland rocks have been subjected to repeated smashing and reagglomeration by impacts and are therefore in brecciated form. A few of the collected lunar samples, however, appear to have been essentially unaltered since they solidified in the primeval lunar crust. These rocks, some rich in aluminum and calcium or magnesium and others showing the KREEP chemical signature, suggest that late in its formation the Moon was covered by a deep magma ocean. The slow cooling of this enormous molten body, in which lighter minerals rose as they formed and heavier ones sank, appears to have resulted in the crust and mantle that exists today (see below Origin and evolution).

The lunar interior
Structure and composition
Most of the knowledge about the lunar interior has come from the Apollo missions and from robotic spacecraft, including Galileo, Clementine, and Lunar Prospector, which observed the Moon in the 1990s. Combining all available data, scientists have created a picture of the Moon (see figure) as a layered body comprising a low-density crust, which ranges from 60 to 100 km (40 to 60 miles) in thickness, overlying a denser mantle, which constitutes the great majority of the Moon's volume. At the centre there probably is a small iron-rich metallic core with a radius of about 400 km (250 miles) at most. The core may once have been an electromagnetic dynamo like that of Earth (see geomagnetic field), which could account for the remanent magnetism observed in some lunar rocks, but it appears that such internal activity has long ceased.
Despite these gains in knowledge, important uncertainties remain. For example, there seems to be no generally accepted explanation for the evidence that the crust is asymmetrical: thicker on the Moon's far side, with the maria predominantly on the near side. Examination of naturally excavated samples from large impact basins may help to resolve this and other questions in lunar history.

Internal activity of the past and present
The idea that the lunar crust is the product of differentiation in an ancient magma ocean is supported to some extent by compositional data, which show that lightweight rocks, containing such minerals as plagioclase, rose while denser materials, such as pyroxene and olivine, sank to become the source regions for the later radioactive heating episode that resulted in the outflows of mare basalts. Whether or not there ever was a uniform global ocean of molten rock, it is clear that the Moon's history is one of much heating and melting in a complex series of events that would have driven off volatiles (if any were present) and erased the record of earlier mineral compositions.
At present all evidence points to the Moon as a body in which, given its small size, all heat-driven internal processes have run down. Its heat flow near the surface, as measured at two sites by Apollo instruments, appears to be less than half that of Earth. Seismic activity is probably far less than that of Earth, though this conclusion needs to be verified by longer-running observations than Apollo provided. Many of the moonquakes detected seem to be only small “creaks” during the Moon's continual adjustment to gravity gradients in its eccentric orbit, while others are due to meteorite impacts or thermal effects. Quakes of truly tectonic origin seem to be uncommon. The small quakes that do occur demonstrate distinct differences from Earth in the way seismic waves are transmitted, both in the regolith and in deeper layers. The seismic data suggest that impacts have fragmented and mixed the upper part of the lunar crust in a manner that left a high proportion of void space. At depths beyond tens of kilometres, the crust behaves as consolidated dry rock.

Origin and evolution
With the rise of scientific inquiry in the Renaissance, investigators attempted to fit theories on the origin of the Moon to the available information, and the question of the Moon's formation became a part of the attempt to explain the observed properties of the solar system (see Solar system: Origin of the solar system). At first the approach was largely founded on a mathematical examination of the dynamics of the Earth-Moon system. Rigorous analysis of careful observations over a period of more than 200 years gradually revealed that, because of tidal effects (see tide), the rotations of both the Moon and Earth are slowing and the Moon is receding from Earth. Studies then turned back to consider the state of the system when the Moon was closer to Earth. Throughout the 17th, 18th, and 19th centuries, investigators examined different theories on lunar origin in an attempt to find one that would agree with the observations.
Lunar origin theories can be divided into three main categories: coaccretion, fission, and capture. Coaccretion suggests that the Moon and Earth were formed together from a primordial cloud of gas and dust. This scenario, however, cannot explain the large angular momentum of the present system. In fission theories a fluid proto-Earth began rotating so rapidly that it flung off a mass of material that formed the Moon. Although persuasive, the theory eventually failed when examined in detail; scientists could not find a combination of properties for a spinning proto-Earth that would eject the right kind of proto-Moon. According to capture theories, the Moon formed elsewhere in the solar system and was later trapped by the strong gravitational field of Earth. This scenario remained popular for a long time, even though the circumstances needed in celestial mechanics to brake a passing Moon into just the right orbit always seemed unlikely.

By the mid-20th century, scientists had imposed additional requirements for a viable lunar-origin theory. Of great importance is the observation that the Moon is much less dense than Earth, and the only likely reason is that the Moon contains significantly less iron. Such a large chemical difference argued against a common origin for the two bodies. Independent-origin theories, however, had their own problems. The question remained unresolved even after the scientifically productive Apollo missions, and it was only in the early 1980s that a model emerged—the giant-impact hypothesis—that eventually gained the support of most lunar scientists.
In this scenario the proto-Earth, shortly after its formation from the solar nebula about 4.6 billion years ago, was struck a glancing blow by a body the size of Mars. Prior to the impact, both bodies already had undergone differentiation into core and mantle. The titanic collision ejected a cloud of fragments, which aggregated into a full or partial ring around Earth and then coalesced into a proto-Moon. The ejected matter consisted mainly of mantle material from the colliding body and the proto-Earth, and it experienced enormous heating from the collision. As a result, the proto-Moon that formed was highly depleted in volatiles and relatively depleted in iron (and thus also in siderophiles). Computer modeling of the collision shows that, given the right initial conditions, an orbiting cloud of debris as massive as the Moon could indeed have formed.
Once a proto-Moon was present in the debris cloud, it would have quickly swept up the remaining fragments in a tremendous bombardment. Then, over a period of 100 million years or so, the rate of impacting bodies diminished, although there still occurred occasional collisions with large objects. Perhaps this was the time of the putative magma ocean and the differentiation of the ancient plagioclase-rich crust. After the Moon had cooled and solidified enough to preserve impact scars, it began to retain the huge signatures of basin-forming collisions with asteroid-sized bodies left over from the formation of the solar system. About 3.9 billion years ago, one of these formed the great Imbrium Basin, or Mare Imbrium, and its mountain ramparts. During some period over the next several hundred million years there occurred the long sequence of volcanic events that filled the near-side basins with mare lavas.
In an effort to unravel the history of this period, scientists have applied modern analytic techniques to lunar rock samples. The mare basalts show a wide range of chemical and mineral compositions reflecting different conditions in the deep regions of the mantle where, presumably because of heating from radioactive elements in the rock, primordial lunar materials were partly remelted and fractionated so that the lavas carried unique trace-element signatures up to the surface. By studying the past events and processes reflected in the mineral, chemical, and isotopic properties of these rocks, lunar scientists have slowly built a picture of a variegated Moon. Their findings have provided valuable background information for Earth- and spacecraft-based efforts to map how the content of important materials varies over the lunar surface.
Once the huge mare lava outflows had diminished, apparently the Moon's heat source had run down. The last few billion years of its history have been calm and essentially geologically inactive except for the continuing rain of impacts, which is also declining over time, and the microscopic weathering due to bombardment by solar and cosmic radiation and particles.

Lunar exploration
Early studies
Investigations of the Moon and some understanding of lunar phenomena can be traced back to a few centuries BC. In ancient China the Moon's motion was carefully recorded as part of a grand structure of astrological thought. In both China and the Middle East, observations became accurate enough to enable the prediction of eclipses, and the recording of eclipses left data of great value for later scientists interested in tracing the history of the Earth-Moon system. (See eclipse: Uses of eclipses for astronomical purposes.) Several early Greek philosophers saw reason to believe that the Moon was inhabited, although they did not base their conclusion on scientific principles. The Greek astronomer and mathematician Hipparchus, on the other hand, took an experimental approach: observing Earth's round shadow creeping across the Moon during a lunar eclipse, he concluded that Earth must be spherical and that the Moon was an independent world, and he correctly explained the Moon's phases and accurately estimated the distance between the two bodies. Later, Mayan calendars were constructed that reflected the results of careful observation and long-range prediction.

For centuries, knowledge about the Moon accumulated slowly, driven by astrological and navigational needs, until an outburst of progress began in the Renaissance. In the early 1600s the German astronomer Johannes Kepler used observations made by Tycho Brahe of Denmark to find empirically the laws governing planetary motion. Kepler wrote a remarkable work of science fiction, Somnium (“The Dream”), that describes the life of imagined inhabitants of the Moon and correctly portrays such facts as the high temperature of the Moon's sunlit side. In 1609–10 Galileo began his telescopic observations that forever changed human understanding of the Moon. Most effort hitherto had been devoted to understanding the movements of the Moon through space, but now astronomers began to focus their attention on the character of the Moon itself. Some milestones in human exploration and understanding of the Moon are given in the table.

Exploration by spacecraft
First robotic missions
Following the launch in 1957 of the U.S.S.R.'s satellite Sputnik, the first spacecraft to orbit Earth, it became obvious that the next major goal of both the Soviet and the U.S. space programs would be the Moon (see space exploration). The United States quickly prepared and launched a few robotic lunar probes, most of which failed and none of which reached the Moon. The Soviet Union had more success, achieving in 1959 the first escape from Earth's gravity with Luna 1, the first impact on the lunar surface with Luna 2, and the first photographic survey of the Moon's far side with Luna 3. After the National Aeronautics and Space Administration (NASA) was founded in 1958, the U.S. program became more ambitious technically and more scientifically oriented. Initial spacecraft investigations were geared toward studying the Moon's fundamental character as a planetary body by means of seismic observation, gamma-ray spectrometry, and close-up imaging. Scientists believed that even limited seismic data would give clues toward resolving the question as to whether the Moon was a primitive, undifferentiated body or one that had been heated and modified by physical and chemical processes such as those on Earth. Gamma-ray measurements would complement the seismic results by showing whether the Moon's interior had sufficient radioactivity to serve as an active heat engine, and they would also give some information on the chemical composition of the lunar surface. Imaging would reveal features too small to be seen from Earth, perhaps providing information on lunar surface processes and also arousing public interest.
Among nine U.S. Ranger missions launched between 1961 and 1965, Ranger 4 (1962) became the first U.S. spacecraft to strike the Moon. Only the last three craft, however, avoided the plaguing malfunctions that limited or prematurely ended the missions of their predecessors. Ranger 7 (1964) returned thousands of excellent television images before impacting as designed, and Rangers 8 and 9 (both 1965) followed successfully. The impact locale of Ranger 7 was named Mare Cognitum for the new knowledge gained, a major example of which was the discovery that even small lunar features have been mostly subdued from incessant meteorite impacts.

After a number of failures in the mid-1960s, the Soviet Union scored several notable achievements: the first successful lunar soft landing by Luna 9 and the first lunar orbit by Luna 10, both in 1966. Pictures from Luna 9 revealed the soft, rubbly nature of the regolith and, because the landing capsule did not sink out of sight, confirmed its approximate bearing strength. Gamma-ray data from Luna 10 hinted at a basaltic composition for near-side regions. In 1965 the Soviet flyby mission designated Zond 3 returned good pictures of the Moon's far side.
In the mid-1960s the United States carried out its own soft-landing and orbital missions. In 1966 Surveyor 1 touched down on the Moon and returned panoramic television images. Six more Surveyors followed between 1966 and 1968, with two failures; they provided not only detailed television views of lunar scenery but also the first chemical data on lunar soil and the first soil-mechanics information showing mechanical properties of the top few centimetres of the regolith. Also, during 1966–67 five U.S. Lunar Orbiters made photographic surveys of most of the lunar surface, providing the mapping essential for planning the Apollo missions.

Apollo to the present
After the Soviet cosmonaut Yury Gagarin pioneered human Earth-orbital flight in April 1961, U.S. President John F. Kennedy established the national objective of landing a man on the Moon and returning him safely by the end of the decade. Apollo was the result of that effort.
Within a few years the Soviet Union and the United States were heavily engaged in a political and technological race to launch manned flights to the Moon. At the time, the Soviets did not publicly acknowledge the full extent of their program, but they did launch a number of human-precursor circumlunar missions between 1968 and 1970 under the generic name Zond, using spacecraft derived from their piloted Soyuz design. Some of the Zond flights brought back colour photographs of the Moon's far side and safely carried live tortoises and other organisms around the Moon and back to Earth. In parallel with these developments, Soviet scientists began launching a series of robotic Luna spacecraft designed to go into lunar orbit and then land with heavy payloads. This series, continuing to 1976, eventually returned drill-core samples of regolith to Earth and also landed two wheeled rovers, Lunokhod 1 and 2 (1970 and 1973), that pioneered robotic mobile exploration of the Moon.

In December 1968, acting partly out of concern that the Soviet Union might be first in getting people to the Moon's vicinity, the United States employed the Apollo 8 mission to take three astronauts—Frank Borman, James Lovell, and William Anders—into lunar orbit. After circling the Moon three times, the crew returned home safely with hundreds of photographs. The Apollo 9 and 10 missions completed the remaining tests of the systems needed for landing on and ascending from the Moon. On July 20, 1969, Apollo 11 astronauts Neil Armstrong and Edwin (“Buzz”) Aldrin set foot on the Moon while Michael Collins orbited above them. Five more successful manned landing missions followed, ending with Apollo 17 in 1972; at the completion of the program, a total of 12 astronauts had set foot on the Moon.
Twenty years later the Soviet Union admitted that it had indeed been aiming at the same goal as Apollo, not only with a set of spacecraft modules for landing on and returning from the Moon but also with the development of a huge launch vehicle, called the N1, comparable to the Apollo program's Saturn V. After several launch failures of the N1, the program was canceled in 1974.

After the Apollo missions, lunar scientists continued to conduct multispectral remote-sensing observations from Earth and perfected instrumental and data-analysis techniques. During Galileo's flybys of Earth and the Moon in December 1990 and 1992 en route to Jupiter, the spacecraft demonstrated the potential for spaceborne multispectral observations—i.e., imaging the Moon in several discrete wavelength ranges—to gather geochemical data (see figure. As a next logical step, scientists generally agreed on a global survey of physical and geochemical properties by an automated spacecraft in polar orbit above the Moon and employing techniques evolved from those used during the Apollo missions. Finally, after a long hiatus, orbital mapping of the Moon resumed with the flights of the Clementine and Lunar Prospector spacecraft, launched in 1994 and 1998, respectively.
New human missions to the Moon, though they have been the subject of advocacy and some detailed studies, will require a rebirth of public interest, perhaps after a period of human activity in Earth-orbiting space stations. Meanwhile, additional robotic missions will be developed and launched. In addition to expanding scientific knowledge, these missions will have another role—namely, to prepare the way for human residence on the Moon.

Mission results
The Apollo program revolutionized human understanding of the Moon. The samples collected and the human and instrumental observations have continued to be studied into the 21st century. Analyses of samples from the Luna missions have continued as well and are valuable because they were collected from eastern equatorial areas far from the Apollo sites.
One new and fundamental result has come from radiometric age dating of the samples. When a rock cools from the molten to the solid state, its radioactive isotopes are immobilized in mineral crystal lattices and then decay in place. Knowing the rate of decay of one nuclear species (nuclide) into another, scientists can, in principle, use the ratios of decay products as a clock to measure the time elapsed since the rock cooled. Some nuclides, such as isotopes of rubidium and strontium, can be used to date rocks that are billions of years old (see rubidium-strontium dating). The required measurements are threatened by contamination and other problems, such as past events that might have reset the clock. Nevertheless, with great care in sample preparation and mass spectrometry techniques, the isotopic ratios can be found and converted into age estimates. By the time of the Apollo sample returns, scientists had refined this art, and, using meteorite samples, they were already investigating the early history of the solar system.
Analysis of the first lunar samples confirmed that the Moon is an evolved body with a long history of differentiation and volcanic activity. Unlike the crust of Earth, however, the lunar crust is not recycled by tectonic processes, so it has preserved the records of ancient events. Highland rock samples returned by the later Apollo missions are nearly four billion years old, revealing that the Moon's crust was already solid soon after the planets condensed out of the solar nebula. The mare basalts, though they cover a wide range of ages, generally show that the basin-filling volcanic outpourings occurred long after the formation of the highlands; this is the reason they are believed to have originated from later radioactive heating within the Moon rather than during the primordial heating event. Trace-element analyses indicate that the magmatic processes of partial melting gave rise to different lavas.

In addition to collecting samples, Apollo astronauts made geologic observations, took photographs, and placed long-lived instrument arrays and retroreflectors on the lunar surface. Not only the landing expeditions but also the Apollo orbital observations yielded important new knowledge. On each mission the Moon-orbiting Command and Service modules carried cameras and remote-sensing instruments for gathering compositional information.

The Clementine and Lunar Prospector spacecraft, operating in lunar polar orbits, used complementary suites of remote-sensing instruments to map the entire Moon, measuring its surface composition, geomorphology, topography, and gravitational and magnetic anomalies. The topographic data highlighted the huge South Pole–Aitken Basin, which, like the other basins on the far side, is devoid of lava filling. Measuring roughly 2,500 km (1,550 miles) in diameter and 13 km (8 miles) deep, it is the largest impact feature on the Moon and the largest known in the solar system; because of its location, its existence was not confirmed until the Lunar Orbiter missions in the 1960s. The gravity data collected by the spacecraft, combined with topography, confirmed the existence of a thick, rigid crust, giving yet more evidence that the Moon's heat source has expired. Both spacecraft missions hinted at the long-considered possibility that water ice exists in permanently shadowed polar craters. The most persuasive evidence came from the neutron spectrometer of Lunar Prospector (see below Lunar resources).

Lunar resources
Scientists and space planners have long acknowledged that extended human residence on the Moon would be greatly aided by the use of local resources. This would avoid the high cost of lifting payloads against Earth's strong gravity. Certainly, lunar soil could be used for shielding habitats against the radiation environment. More advanced uses of lunar resources are clearly possible, but how advantageous they would be is presently unknown. For example, most lunar rocks are about 40 percent oxygen, and chemical and electrochemical methods for extracting it have been demonstrated in laboratories. Nevertheless, significant engineering advances would be needed before the cost and difficulty of operating an industrial-scale mining and oxygen-production facility on the Moon could be estimated and its advantages over transporting oxygen from Earth could be evaluated. In the long run, however, some form of extractive industry on the Moon is likely, in part because launching fleets of large rockets continuously from Earth would be too costly and too polluting of the atmosphere.
The solar wind has implanted hydrogen, helium, and other elements in the surfaces of fine grains of lunar soil. Though their amounts are small—they constitute about 100 parts per million in the soil—they may someday serve as a resource. They are easily released by moderate heating, but large volumes of soil would need to be processed to obtain useful amounts of the desired materials. Helium-3, a helium isotope that is rare on Earth and that has been deposited on the Moon by the solar wind, has been proposed as a fuel for nuclear fusion reactors in the future.

One natural resource uniquely available on the Moon is its polar environment. Because the Moon's axis is nearly perpendicular to the plane of the ecliptic, sunlight is always horizontal at the lunar poles, and certain areas, such as crater bottoms, exist in perpetual shadow. Under these conditions the surface may reach temperatures as low as 40 K (−388 °F, −233 °C). Some scientists have theorized that these cold traps may have collected volatile substances, including water ice, over geologic time, though others have expressed doubt that ice deposits could have survived there.
The Lunar Prospector spacecraft, which orbited the Moon for a year and a half, carried a neutron spectrometer to investigate the composition of the regolith within about a metre (three feet) of the surface. Neutrons originating underground owing to radioactivity and cosmic-ray bombardment interact with the nuclei of elements in the regolith en route to space, where they can be detected from orbit. A neutron loses more energy in an interaction with a light nucleus than with a heavy one, so the observed neutron spectrum can reveal whether light elements are present in the regolith. Lunar Prospector gave clear indications of light-element concentrations at both poles, interpreted as proof of excess hydrogen atoms. The observed hydrogen signature may represent the theoretically predicted deposits of water ice.
A high priority for future lunar exploration is to send an autonomous robotic rover into a dark polar region to confirm the putative ice deposits, find out the form of the ice if it exists, and begin assessing its possible utility. If lunar ice can be mined economically, it can serve as a source of rocket propellants when split into its hydrogen and oxygen components. From a longer-term perspective, however, the ice would better be regarded as a limited, recyclable resource for life support (in the form of drinking water and perhaps breathable oxygen). Should this resource exist, an international policy for its conservation and management would be needed. Discovery of volatile substances anywhere on the Moon would be important both scientifically and for potential human habitation because all known lunar rocks are totally dry.
Even if no icy bonanza is discovered, the lunar polar regions still represent an important resource. Only there can be found not only continuous darkness but also continuous sunlight. A solar collector tracking the Sun from a high peak near a lunar pole could provide essentially uninterrupted heat and electric power. Also, the radiators required for eliminating waste heat could be positioned in areas of continuous darkness, where the heat could be dissipated into space.
The lunar poles also could serve as good sites for certain astronomical observations. To observe objects in the cosmos that radiate in the infrared and millimetre-wavelength regions of the spectrum, astronomers need telescopes and detectors that are cold enough to limit the interference generated by the instruments' own heat (see infrared astronomy). To date, such telescopes launched into space have carried cryogenic coolants, which eventually run out. A telescope permanently sited in a lunar polar cold region and insulated from local heat sources might cool on its own to 40 K (−388 °F, −233 °C) or lower. Although such an instrument could observe less than half the sky—ideally, one would be placed at each lunar pole—it would enable uninterrupted viewing of any object above its horizon.

James D. Burke
Additional Reading
Revival of interest in the Moon and in lunar exploration after the long post-Apollo hiatus is reflected in Paul D. Spudis, The Once and Future Moon (1996), a readable and well-illustrated summary of recent scientific knowledge as well as arguments advocating continued exploration and eventual human return to the Moon. The rich and expanding scientific literature of the Moon is well represented in Proceedings of Lunar and Planetary Science, published collections of papers from the annual Lunar and Planetary Science Conference. The finest book about lunar exploration remains Davis Thomas (ed.), Moon: Man's Greatest Adventure, rev. ed. (1973), splendidly illustrated, with text by scholars describing the history and culture, engineering and projects, and early scientific results of the great human drive that began in ancient times and culminated in the Apollo missions. A more modest but still comprehensive and well-illustrated book is Patrick Moore, The Moon (1981, reprinted 1984). Don E. Wilhelms, The Geologic History of the Moon (1987), is illustrated with many beautiful explanatory pictures and drawings. Examples of relevant literature on theories of lunar origin include W.K. Hartmann, R.J. Phillips, and G.J. Taylor (eds.), Origin of the Moon (1986); and A.V. Vitiazev, G.V. Pechernikova, and V.S. Safronov, Planety zemnoy gruppy: proiskhozhdeniye i rannyaya evolyutsiya (“Terrestrial Planets: Their Origin and Early Evolution”; 1990). Stuart Ross Taylor, Planetary Science: A Lunar Perspective (1982); and W.W. Mendell (ed.), Lunar Bases and Space Activities of the 21st Century (1985), are collections of specialized papers, although most of the material is accessible to the general reader. The realization that the Moon is an enormous storehouse of resources that may be useful to humankind in the future has prompted several publications, including Gerard K. O'Neill, The High Frontier, 3rd ed. (2000); a series of conference papers, Space Manufacturing (biennial); and a book commissioned by NASA to serve as a primary reference, Grant Heiken, David Vaniman, and Bevan M. French (eds.), Lunar Sourcebook: A User's Guide to the Moon (1991), which with its compendious bibliography is an excellent summary of what is known about the Moon. Peter Eckart (ed.), The Lunar Base Handbook: An Introduction to Lunar Base Design, Development, and Operations (1999), gives up-to-date information on lunar resources and their uses, plus engineering and scientific aspects of the creation of human habitats on the Moon.