eclipse
Introduction
in astronomy, complete or partial obscuring of a celestial body by another. An eclipse occurs when three celestial objects become aligned.
The Sun is eclipsed when the Moon comes between it and the Earth; the Moon is eclipsed when it moves into the shadow of the Earth cast by the Sun. Eclipses of natural or artificial satellites of a planet occur as the satellites move into the planet's shadow. The two component stars of an eclipsing binary star move around each other in such a way that their orbital plane passes through or very near the Earth, and each star periodically eclipses the other as seen from the Earth.
When the apparent size of the eclipsed body is much smaller than that of the eclipsing body, the phenomenon is known as an occultation. Examples are the disappearance of a star, nebula, or planet behind the Moon, or the vanishing of a natural satellite or space probe behind some body of the solar system.
A transit occurs when, as viewed from the Earth, a relatively small body passes across the disk of a larger body, usually the Sun or a planet, eclipsing only a very small area: Mercury and Venus periodically transit the Sun, and a satellite may transit its planet.

Phenomena observed during eclipses
Lunar eclipse phenomena
The Moon may, when full, enter the shadow of the Earth. The motion of the Moon around the Earth is from west to east relative to the Sun, so that, for an observer facing south, the shadowing of the Moon begins at its left edge (if the Moon were north of the observer, as, for example, in parts of the Southern Hemisphere, the opposite would be true). If the eclipse is a total one and circumstances are favourable, the Moon will pass through the umbra, or darkest part of the shadow, in about two hours (see Figure 1). During this time, the Moon is usually not quite dark. A part of the sunlight, especially the redder light, penetrates the Earth's atmosphere, is refracted into the shadow cone, and reaches the Moon. Meteorologic conditions on Earth strongly affect the amount and colour of light that can penetrate the atmosphere. Generally, the totally eclipsed Moon is clearly visible and has a reddish-brown, coppery colour, but the brightness varies strongly from one eclipse to another.
After the Moon leaves the umbra, it must still pass through the penumbra of the shadow. When the border between umbra and penumbra is visible on the Moon, the border is seen to be part of a circle, the projection of the circumference of the Earth. This is a direct proof of the spherical shape of the Earth. Because of the Earth's atmosphere, the edge of the umbra is rather diffuse, and the times of contact between the Moon and the umbra cannot be observed accurately.
During the eclipse, the surface of the Moon cools at a rate dependent on the constitution of the lunar soil, which is not everywhere the same. Many spots on the Moon sometimes remain brighter than their surroundings during totality—particularly in their output of infrared radiation—possibly because their heat conductivity is less, but the cause is not fully understood.
An eclipse of the Moon can be seen under similar conditions at all places on the Earth where the Moon is above the horizon.

Solar eclipse phenomena
Totality at any particular solar eclipse can only be seen from a relatively narrow belt on Earth. The various phases observable at a total solar eclipse are illustrated in Figure 2A. “First contact” designates the moment when the disk of the Moon, invisible against the bright sky background, just touches the disk of the Sun. The partial phase of the eclipse then begins, as a small indentation in the western rim of the Sun becomes noticeable. The dark disk of the Moon now gradually moves across the Sun's disk, and the bright area of the Sun is reduced to a crescent. The sunlight, shining through gaps in foliage and other small openings, is then seen to form little crescents of light that are images of the light source, the Sun. Toward the beginning of totality, the direct light from the Sun diminishes very quickly and the colour changes. The sky becomes dark, but, along the horizon, the Earth's atmosphere still appears bright because the umbra of the Moon's shadow on the Earth extends over a rather narrow region. The scattered light coming in from a distance beyond this region produces weird effects. Men, birds, and other animals react with fear; birds may go to roost as they do at sunset.
As the tiny, narrow crescent of sunlight disappears, little bright specks remain where depressions in the Moon's edge, the limb, are last to obscure the Sun's limb. These specks are known as Baily's beads, after the 18th-century English astronomer Francis Baily, who first drew attention to them. The beads vanish at the moment of second contact, when totality sets in. This is the climax of the eclipse. The reddish prominences and chromosphere of the Sun, around the Moon's limb, can now be seen. The brighter planets and stars become visible in the sky. The white corona extends out from the Sun to a distance greater than the Sun's diameter, at which point it fades completely. The temperature in the path of totality falls by some degrees. The light of totality is much brighter than that of the Full Moon but is quite different.
The moment of third contact approaches, at which time many of the phenomena of second contact appear again in reverse order. Suddenly the first Baily's bead appears, now on the other side of the Moon. More beads of light follow, the Sun's crescent grows again, the corona disappears, daylight brightens, and the stars and planets fade from view. The thin crescent of the Sun gradually widens, and about one and a quarter hours later the eclipse ends with fourth contact, when the last encroachment made by the Moon on the Sun's rim disappears.
During the partial phase, both before and after totality, it is absolutely essential to protect the eyes against injury by the intense brilliance of the Sun. It should never be viewed directly except through strong filters, a dark smoked glass, or a heavily fogged photographic plate or film.
When totality is imminent and only a small crescent of the Sun remains, the so-called shadow bands can often be seen on plain light-coloured surfaces, such as open floors and walls. These are striations of light and shade, moving and undulating, several centimetres (or inches) wide. Their velocity and direction depend on air currents at various heights, as they are caused by refraction of sunlight by small inhomogeneities in the Earth's atmosphere. A similar phenomenon is the projection of water waves on the bottom of a sunlit swimming pool or bath.

The geometry of eclipses, occultations, and transits
Eclipses of the Sun
An eclipse of the Sun takes place when the Moon comes between the Earth and the Sun so that the Moon's shadow sweeps over the face of the Earth (see Figure 3). This shadow consists of two parts: the umbra, or total shadow, a cone into which no direct sunlight penetrates; and the penumbra, or half shadow, which is reached by light from only a part of the Sun's disk.

To an observer within the umbra, the Sun's disk appears completely covered by the disk of the Moon; such an eclipse is called total (see video). To an observer within the penumbra, the Moon's disk appears projected against the Sun's disk so as to overlap it partly; the eclipse is then called partial for that observer. The umbra cone is narrow at the distance of the Earth, and a total eclipse is observable only within the narrow strip of land or sea over which the umbra passes. A partial eclipse may be seen from places within the large area covered by the penumbra. Sometimes the Earth intercepts the penumbra of the Moon but is missed by its umbra; only a partial eclipse of the Sun is then observed anywhere on the Earth.
By a remarkable coincidence, the sizes and distances of the Sun and Moon are such that they appear as very nearly the same angular size (about 0.5°) at the Earth, but their apparent sizes depend on their distances from the Earth. The Earth revolves around the Sun in an elliptical orbit, so that the distance of the Sun changes slightly during a year, with a correspondingly small change in the apparent size, the angular diameter, of the solar disk. In a similar way, the apparent size of the Moon's disk changes somewhat during the month because the Moon's orbit is also elliptical. When the Sun is nearest to the Earth and the Moon is at its greatest distance, the apparent disk of the Moon is smaller than that of the Sun. If an eclipse of the Sun occurs at this time, the Moon's disk passing over the Sun's disk cannot cover it completely but will leave the rim of the Sun visible all around it. Such an eclipse is said to be annular. Total eclipses and annular eclipses are called central.

In a partial eclipse (Figure 2B), the centre of the Moon's disk does not pass across the centre of the Sun. After the first contact, the visible crescent of the Sun decreases in width until the centres of the two disks reach their closest approach. This is the moment of maximum phase, and the extent is measured by the ratio between the smallest width of the crescent and the diameter of the Sun. After maximum phase, the crescent of the Sun widens again until the Moon passes out of the Sun's disk at the last contact.

Eclipses of the Moon
When the Moon moves through the shadow of the Earth (see Figure 1) it loses its bright direct illumination by the Sun, although its disk still remains faintly visible. As the shadow of the Earth is directed away from the Sun, a lunar eclipse can occur only at the time of Full Moon—that is, when the Moon is on the side of the Earth opposite to that of the Sun. A lunar eclipse appears much the same at all points of the Earth from which it can be seen. When the Moon enters the penumbra, a penumbral eclipse occurs. The dimming of the Moon's illumination by the penumbra is so slight as to be scarcely noticeable, and penumbral eclipses are rarely watched. After a part of the Moon's surface is in the umbra and thus darkened, the Moon is said to be in partial eclipse. After about an hour, when the whole disk of the Moon is within the umbra, the eclipse becomes total (see video). If the Moon's path leads through the centre of the umbra, the total eclipse can be expected to last about an hour and three-quarters.

Eclipses, occultations, and transits of satellites
These phenomena are conveniently illustrated by the four largest satellites of Jupiter, whose eclipses provide a frequently occurring and fascinating spectacle to the telescopic observer. The three innermost moons (Io, Europa, and Ganymede) disappear into the shadow of Jupiter at each revolution, though the fourth (Callisto) is not eclipsed every time. Because of the sizable dimensions of these bodies, some minutes elapse between first contact with the shadow and totality. The orbits of these satellites lie nearly in the same plane as Jupiter's orbit around the Sun, and, at practically every revolution of each satellite, the following four eclipse phenomena take place: (1) eclipse of the satellite when it passes through Jupiter's shadow; (2) occultation of the satellite when it disappears behind the planet, as seen from the Earth; (3) transit of the satellite across the disk of Jupiter; and (4) transit of the shadow of the satellite across the planet's disk.

Figure 4 illustrates these phenomena; it shows Jupiter and the orbit of one of its satellites, the direction of the sunlight illuminating the system, and the direction toward the Earth, from where the observation is made. When the satellite arrives at the point S1 of its orbit, it enters Jupiter's shadow (eclipse) and vanishes. At S2 it comes out of the shadow, but, to the terrestrial observer, it is now hidden behind the planet (occultation) until at S3 it reappears at the limb. When the satellite reaches the position S4, its shadow falls on Jupiter, causing a small dark spot on its surface. Seen from the Earth, the satellite is to the left of Jupiter approaching Jupiter's limb, at the time that its shadow spot passes across the planet's disk (transit of shadow). At S5 the satellite starts to pass in front of the planet (transit of satellite), following its shadow spot. Both Jupiter and the satellite must have their illuminated sides facing the Earth. They differ little in total surface brightness; near the limb the satellite is somewhat brighter than the planet's surface on which it appears projected, but near the middle of the disk it is hardly distinguishable. At S6 the shadow leaves the planet, and at S7 the satellite emerges at the limb.
Historically, the eclipses of Jupiter's satellites are important, for they provided one of the earliest proofs of the finite speed of light. It is possible to calculate with considerable precision the times of disappearance and reappearance of a satellite undergoing eclipse. The Danish astronomer Ole Rømer in 1675 noticed discrepancies between the observed and calculated times of such eclipses, which he correctly explained as being due to the difference in the travel time of light when the Earth is nearest to Jupiter or farther away from it.
A related phenomenon is the occultation of a space probe by a planet. During the beginning and the end of such an occultation, signals sent out by the spacecraft and received on Earth have penetrated the planet's atmosphere and can yield information about atmospheric density and composition.
On March 10, 1977, the planet Uranus passed between the Earth and a bright star. The event was observed by several teams of astronomers, who hoped to derive an accurate estimate of the diameter of the planet from their data. To their surprise, however, the light from the star was briefly obscured several times before and after the disk of Uranus occulted it. It was concluded that Uranus has a system of rings somewhat like those of Saturn.

The frequency of solar and lunar eclipses
A solar eclipse, especially a total one, can be seen from only a limited part of the Earth, while the eclipsed Moon can be seen at the time of the eclipse wherever the Moon is above the horizon.
In most calendar years there are two lunar eclipses; in some years one or three or none occur. Solar eclipses occur two to five times a year, five being exceptional; there were five in 1935 and will be again in 2206. The average number of total solar eclipses in a century is 66 for the Earth as a whole.
Numbers of solar eclipses predicted to take place during the 20th to the 25th century are:
1901–2000: 228 eclipses, of which 145 are central
2001–2100: 224 eclipses, of which 144 are central
2101–2200: 235 eclipses, of which 151 are central
2201–2300: 248 eclipses, of which 156 are central
2301–2400: 248 eclipses, of which 160 are central
2401–2500: 237 eclipses, of which 153 are central
Any point on Earth may, on the average, experience no more than one total solar eclipse in three to four centuries. The situation is quite different for lunar eclipses. An observer remaining at the same place (and granted cloudless skies) can see 19 or 20 lunar eclipses in 18 years: three or four total eclipses and six or seven partial eclipses may be visible from beginning to end, and five total eclipses and four or five partial eclipses at least partially visible. All these numbers can be worked out from the geometry of the eclipses. A total lunar eclipse can last as long as an hour and three-quarters, but for a solar total eclipse maximum duration of totality is only 7 1/2 minutes. This difference results from the fact that the Moon is much smaller in cross section than the extension of the Earth's shadow but can be only a little greater in apparent size than the Sun.

Cycles of eclipses
The eclipses of the Sun and Moon occur at New Moon and Full Moon, respectively, so that one basic time period involved in the occurrence of eclipses is the synodic month; i.e., the time of one revolution of the Moon around the Earth with respect to the Sun.
A solar eclipse does not occur at every New Moon, because the Moon's orbit plane is inclined to the ecliptic, the plane of the orbit of the Earth around the Sun. The angle between the planes is about 5°; thus the Moon can pass well above or below the Sun. The line of intersection of the planes is called the line of the nodes, being the two points where the Moon's orbit intersects the ecliptic plane. The ascending node is the point where the Moon crosses the ecliptic from south to north, and the descending node that where it goes from north to south. The nodes move along the orbit from west to east, going completely around the ecliptic in about 19 years. The Moon's revolution from one node to the same node again (called the draconic month, 27.212220 days) takes somewhat less time than a revolution from Full Moon to Full Moon (the synodic month, 29.530589 days). For an eclipse to occur, the Moon has to be near one of the nodes of its orbit. The draconic month, therefore, is the other basic period of eclipses.
Resonance between these two periods produces what is called the saros period, after which time Moon and Sun return very nearly to the same relative positions. The saros was known to the ancient Babylonians (see below Use of eclipses for astronomical purposes: In ancient astronomy). It comprises 223 lunations (the time period from New Moon to New Moon)—that is, 6,585.321124 days, or 241.9986 draconic months. This is nearly a whole number, so that the Full Moon is in almost the same position (e.g., very near a node) at the beginning and end of a saros, a lapse of time equalling 18 years and 11 1/3 days or 10 1/3 days if five leap years fall within the period. Thus, there is usually a close resemblance between any eclipse and the one taking place 18 years and 11 days earlier or later. Since the date differs by only about 11 days in the calendar year, the latitudes on Earth of the two eclipses will be about the same, and so will the relative apparent sizes of Sun and Moon. The saros period also comprises 238.992 anomalistic months, again nearly a whole number. In one anomalistic month, the Moon describes its orbit from perigee to perigee, in which point it is nearest to the Earth. So the Moon's distance from the Earth is the same after a whole number of anomalistic months and very nearly the same after one saros. The saros period is, therefore, extremely useful for the prediction of both solar and lunar eclipses.
Because of the extra one-third day in the saros, the eclipse recurs each time approximately 120° farther west on the surface of the Earth. After three saroses, or 54 years and about a month, the longitude is repeated.
There is a regular shift on Earth to the north or to the south of successive eclipse tracks from one saros to the next. The eclipses occurring when the Moon is near its ascending node shift to the south, those happening when it is near its descending node shift to the north. A saros series of eclipses begins its life at one pole of the Earth and ends it at the other. Every saros series lasts about 1,300 years and comprises 73 eclipses. About 42 of these series overlap at any time.
Two consecutive saros series are separated by the inex period, 29 years minus 20 days—that is, 358 synodic months—after which time the New Moon has come from one node to the opposite node. The lifetime of an inex group is about 23,000 years, 70 groups coexisting, each comprising 780 eclipses. All other cycles in eclipses are combinations of the saros and the inex.

Prediction and calculation of solar and lunar eclipses
The problem may be divided into two parts. The first is to find out when an eclipse will occur, the other to determine when and where it will be visible.

For this purpose it is convenient first to consider the Earth as fixed and to suppose the observer looking out from its centre. To this observer, O in Figure 5, the Sun and Moon appear projected on the celestial sphere. While this sphere appears to him to rotate daily, as measured by the positions of the stars, around the line PP′ (the Earth's axis of rotation), the Sun's disk, S, appears to travel slowly along the great circle EE′ (the ecliptic), making a complete revolution in one year. At the same time the Moon's disk, M, revolves along the circle LL′ once during a lunar month. The angular diameters of the two disks S and M are each about 0.5° but vary slightly.
Every month the Moon's disk revolving along LL′ will overtake the more slowly moving Sun once, at the moment of New Moon. Usually the Moon's disk will pass above or below the Sun's disk. Overlapping of the two results in an eclipse of the Sun, which can happen only when the New Moon occurs at a moment when the Sun is near the points ☊ or ☋; these signs denote the ascending and descending nodes of the Moon's orbit.

The crosscut of the umbra, the shadow cone of the Earth, at the distance of the Moon (as shown in Figure 5), may be projected like a disk U onto the celestial sphere. It subtends an angle of about 1.4°; its centre will always be opposite to the Sun's disk and travel along EE′. A lunar eclipse occurs whenever the Moon's disk overlaps the shadow disk; this happens only when the shadow disk is near one of the nodes or the Sun is near the opposite node. The Sun's passage through the lunar nodes is thus the critical time for both solar and lunar eclipses. The Moon's orbit plane, represented by the circle LL′, is not fixed, and its nodes move slowly along the ecliptic in the direction indicated by the arrow, making a complete revolution in about 19 years. The interval between two successive passages of the Sun through one of the nodes is termed an “eclipse year,” and, since the Moon's node moves so as to meet the advancing Sun, this interval is about 18.6 days less than a tropical (or ordinary) year.
In Figure 6 the region of the ascending node as seen from the centre of the sphere is much enlarged. Here the node is kept fixed and the apparent motions of the Sun and the Moon are shown relative to the node. To the imaginary observer at the centre of the Earth, the Sun's disk will travel along the circle EE′, the Moon's disk along LL′. The Sun is so distant compared with the size of the Earth that, from all places on the Earth's surface, the Sun is seen nearly in the same position as from the centre. But the Moon is relatively near and its projected position on the celestial sphere is different for various observing stations on the Earth; it may be displaced as much as 1° from the position in which it is seen from the centre of the Earth. If the radius of the Moon's disk is enlarged by 1°, a circle, C, is obtained that encloses all possible positions of the Moon's disk seen from anywhere on the Earth. Conversely, if any circle of the Moon's size is drawn inside this “Moon circle,” C, there is a place on the Earth from which the Moon is seen in that position.

Accordingly, an eclipse of the Sun occurs somewhere on Earth whenever the Moon overtakes the Sun in such a position that the Moon circle, C, passes over the Sun's disk; when the latter is entirely covered by the Moon circle, the eclipse will be central (i.e., total or annular). From Figure 6A, it is evident that a solar eclipse will take place if a New Moon occurs while the Sun moves from S1 to S4. This period is the eclipse season; it starts 19 days before the Sun passes a node and ends 19 days thereafter. Since there is a New Moon every month, at least one solar eclipse, and occasionally two, occurs during every eclipse season—of which there are two in each calendar year. A fifth solar eclipse during a calendar year is possible because part of a third eclipse season may occur at the beginning of January or at the end of December.

Figure 6B illustrates the condition necessary for a lunar eclipse. If a Full Moon occurs within 13 days of a node passage of the Sun (when the shadow disk, U, passes the ascending node), the Moon will be eclipsed. Most eclipse seasons, but not all, will thus also contain a lunar eclipse. When three eclipse seasons fall in a calendar year, there may be three lunar eclipses in that year. Eclipses of the Sun are evidently more frequent than those of the Moon. Solar eclipses, however, can only be seen from a very limited region of the Earth, whereas lunar eclipses are visible from an entire hemisphere.

During a solar eclipse the shadow cones (umbra and penumbra) of the Moon sweep across the face of the Earth (Figure 3), while at the same time the Earth is rotating on its axis. Within the narrow area covered by the umbra, the eclipse is total. Within the wider surrounding region covered by the penumbra, the eclipse is partial.
The astronomical ephemerides, or tables, published for each year provide maps tracing the paths of the more important eclipses in considerable detail, as well as data for accurate calculation of the times of contact at any given observing station. Calculations are made some years ahead in Ephemeris Time (ET), which is defined by the orbital motion of the Earth and the other planets. At the time of the eclipse, the correction is made to Universal Time (UT), which is defined by the rotation of the Earth and is not rigorously uniform.
It is possible with the aid of modern tables to accurately predict solar eclipses several years ahead. For predictions of longer range, the main uncertainty is that of the Moon's motion. Eclipses can of course be “predicted backward” as well as forward, and the calculation of ancient eclipses has been of value in historical research.

Eclipse research activities
Solar research
During a solar eclipse, the Moon serves as a screen outside the Earth's atmosphere. As seen from the Earth, the Moon's dark projection on the Sun crosses the Sun at approximately 350 kilometres (220 miles) per second. For a few seconds, the bright, ordinarily visible disk of the Sun, called the photosphere, is eclipsed, while the chromosphere, the lower solar atmosphere, remains visible at the Sun's edge. During this brief time, light emitted from the chromosphere can be studied and its decrease in intensity with height above the photosphere can be measured. Discrimination between layers within the chromosphere is possible to some extent. An instrument called the coronagraph has been developed to obscure the brilliant photosphere artificially at times outside of eclipses, but the study of the thin layers of the solar chromosphere and corona is still best done during an eclipse.
The first photograph of a solar eclipse was taken in 1851, and the first of scientific importance by the British scientist and inventor Warren de la Rue and the Italian astronomer Angelo Secchi in 1860.

Spectroscopic observations
The ordinary spectrum of the Sun contains a brilliant multicoloured background—the continuum. This radiation emanates primarily from the lower layers of the solar atmosphere. Cooler atoms higher up, however, selectively absorb the radiations, so that the solar spectrum consists of the bright rainbow background with many narrow gaps—dark lines—where the light has been absorbed.
Spectroscopy was first applied to the eclipsed Sun in 1868, when the path of totality passed over India and Malaya and the spectrum of bright lines originating in the solar prominences was observed. The British astronomer Joseph Norman Lockyer thought that one of these spectral lines was emitted by an unknown chemical element, which he called helium (from Greek hēlios, “sun”); helium was not identified on Earth until 1895.
At the moment of totality, when the Moon obliterates the last trace of the bright photosphere, with its dark-line spectrum, the light from the upper tenuous layers of the solar atmosphere (which is usually lost in the much brighter photospheric light) flashes into view with the characteristic bright-line spectrum of a luminous gas. This spectrum, first observed in 1870, disappears within four to five seconds, as the Moon moves across the disk of the Sun. Because of its evanescent character, it is called the flash spectrum. A second flash occurs at the end of totality. Analysis of flash spectra has led to some surprising results. The spectrum matches the dark-line spectrum only roughly. The lines of the neutral metals are of comparable strength in the two spectra, but those of the ionized metals (i.e., those made up of atoms that have lost one or more electrons) are markedly enhanced in the flash spectrum. The difference is attributed, in part, to lower pressures in the upper layers of the solar atmosphere, but high temperature appears to contribute to the increased excitation. This condition is especially true for the flash lines of ionized helium, which do not appear at all in the ordinary dark-line spectrum. Such flash lines require excitation temperatures of at least 25,000 K (about 45,000° F) for their production, whereas the temperature required to produce the observed quality and quantity of bright emission from the surface is only 5,800 K.

Other observations
Extending upward from the chromosphere, and closely related to it, are the so-called prominences, one of the striking features of a total eclipse, which project outward into space. They appear as rose-coloured patches of flame, projecting well beyond the limb of the Moon, and consist of long interlacing filaments of incandescent gas.
The corona of the Sun can be seen during totality. One of the most beautiful of natural phenomena, the solar corona shines like finely etched white frost against the deep blue of the eclipse-darkened sky. The form of the corona presented at different eclipses is almost infinitely variable. On occasion, usually when sunspots are near a minimum, long streamers extend four or five solar diameters away from the Sun. At other times, especially close to sunspot maximum, the corona is more nearly circular but with jagged, petallike extensions. The corona is faint, about 500,000 times less brilliant than the Sun itself. Consequently, the sky glare surrounding the Sun ordinarily hides the coronal details, and, before the development of the coronagraph, it was believed to be impossible to record the corona except during an eclipse.
Also interesting are observations of the radio emission of the Sun during a solar eclipse. When the Moon crosses different parts of the Sun, inhomogeneities in the generation of solar radio waves can be localized.
For geodetic purposes (measuring the Earth), the exact timing of the moments of the contacts of Moon and Sun are important, because these depend on the site of observation and so on the shape of the Earth.
The Sun's ultraviolet and X radiation help create and influence the ionosphere, a region in the Earth's upper atmosphere where many atoms are ionized. The ionosphere changes during an eclipse as the ionizing radiation is cut off. So that observations of the ionosphere made during an eclipse can be interpreted correctly, it is necessary to predict the times and phases of the eclipse as it reaches the high layers of the atmosphere, where they may be quite different from times and phases on the surface.

Tests of relativity during solar eclipse
One of the predictions of the general theory of relativity, as presented by Einstein in 1915, is the deflection of light rays by a gravitational field—that of the Sun, for example. According to this theory the deflection, which causes the image of a star to appear slightly too far from the Sun's image, amounts to 1.75 seconds of arc at the limb of the Sun and decreases in proportion to the apparent distance from the centre of the solar disk of the star whose light is deflected. This is twice the amount given by the older Newtonian dynamics if light is assumed to have inertial properties. If light does not have such properties, as is generally accepted now, the Newtonian deflection is zero. To test the theory, it is necessary to have extremely precise measurements of as many stars as possible around the Sun, and preferably close to it. The brightness of the corona prevents observation of stars closer to the Sun than one solar diameter. During totality the sky around the Sun is photographed with long-focus cameras to give the largest possible scale. The equipment is left in place, and about half a year later the same region of the sky with the same stars can be photographed again, this time during the night and without the Sun's disturbing gravity field. Comparison of the two sets of photographs shows the amount of gravitational displacement of starlight by the Sun and serves as a test of theory. Results of a number of eclipse observations made since 1918 have verified Einstein's prediction, though the deduced values are somewhat uncertain as a consequence of the small displacement.

Lunar research
Lunar eclipses can yield information about the cooling of the Moon's soil when the Sun's radiation is suddenly removed and, therefore, about the soil's conductivity of heat and its structure. Infrared and radio-wavelength radiations from the Moon decline in intensity more slowly than does visible light emission during an eclipse because they are emitted from below the surface, and measurements indicate how far the different kinds of radiation penetrate into the lunar soil. Infrared observations show that at many “bright spots” the soil retains its heat much longer than in surrounding areas. Because of the absence of a lunar atmosphere, the solid surface is exposed to the full intensity of ultraviolet and particulate radiation from the Sun, which may give rise to fluorescence in some rock materials. Observations during lunar eclipses have given positive results for this phenomenon, with the appearance of abnormal bright regions in the obscured parts of the Moon.

Transits of Mercury and Venus
At the time of inferior conjunction (i.e., when moving between the Earth and the Sun) Mercury, seen from the Earth, usually passes north or south of the Sun because of the inclination of its orbit. But if the conjunction occurs when Mercury is near one of the nodes of its orbit, the planet crosses the disk of the Sun as a small black circular spot, visible only with a telescope. Since the Earth passes Mercury's nodes on May 7 and November 9, transits can occur only near those dates. The shortest interval between two successive transits is seven years.
Transits of Venus can be seen without a telescope if the eyes are properly protected. When the transit is central, it takes about eight hours. The phenomenon is rare and can happen only within a day or two of the dates when the Earth passes the nodes of Venus' orbit—that is, on June 7 and December 8. The transits occur in pairs, with an interval of eight years between members of a pair; between the pairs, more than 100 years elapse.
Transits of Venus are helpful in finding the parallax and from it the distance of the Sun, as first pointed out by the British astronomer Edmond Halley in 1679. Parallax is the apparent difference in direction of an object when observed from different positions. The transits of June 1761 and 1769 and those of December in 1874 and 1882 were, thus, extensively observed. The next pair of transits of Venus are expected on June 8, 2004, and June 6, 2012.

One kind of observation for determining parallax consists in fixing the times of the contacts of the disks of the planet and the Sun from different points on the Earth. The observers at past transits became aware of a few remarkable phenomena. When Venus was partially overlapping the disk of the Sun, the part of the limb of the planet that extended beyond the Sun was seen to be surrounded with a radiant aureole, which observers of the transit in 1761 ascribed to the presence of an atmosphere on Venus. A second phenomenon was seen just after second and before third contact (see Figure 1), when Venus just touched the Sun's limb on the inside; this consisted in the development of a little dark connection—the so-called black drop—between Venus and the limb. Because of the black drop, the times of contact could not be sharply defined. Presumed causes of the black drop are diffraction, atmospheric agitation, and instrumental factors.
The amount of sunlight intercepted during a transit depends on the diameter of the planet, and measuring this amount of sunlight may be one of the most accurate ways of determining the planet's diameter.

Occultations by the Sun and Moon
The occultation of the Crab Nebula by the solar corona, the extensive outer atmosphere of the Sun, takes place each year in June. The Crab Nebula is a radio source (Taurus A), and, when the occultation occurs, its pattern of radio emission is observed to broaden significantly. The broadening is attributed to scattering of radio-frequency radiation by the density irregularities in the extended corona. The nearest approach of Sun and Crab Nebula is at about five solar radii; the scattering is brought into evidence up to 60 solar radii.
The Moon sometimes occults a planet but very often occults a star. As a consequence of the inclination of the Moon's orbit with respect to the ecliptic (the Sun's apparent annual path) and the movement of the nodes of the Moon's orbit, all the stars in a belt of 10° around the ecliptic are occulted at some time during a period of about nine years. Among these are the bright stars Aldebaran, Regulus, Spica, and Antares, the star clusters Pleiades, Hyades, and Praesepe, and the Crab Nebula.
Since the Moon always moves eastward, an occulted star disappears at the Moon's eastern limb and reappears at the western. These phenomena can be best observed at the dark limb of the Moon.
Efforts have been made to determine the apparent diameter of stars by accurately estimating the time they take to disappear behind the Moon; with modern optics and electronic devices this would not be impossible in some cases. But there are complications of two kinds: first, many stars are close binaries (doubles); and, second, the decline of the light of the occulted star shows an optical phenomenon called a diffraction pattern. However, for the purpose of yielding basic data on the perturbations (disturbances) in the Moon's orbit and on irregularities in the rotation of the Earth, the exact times of star occultations are still of value.
The disappearance or reappearance of Jupiter takes more than one minute, but the time for a minor planet is under one second, and that for even a giant star like Aldebaran or Antares is less still, by a factor of 10. Other stars disappear nearly instantaneously, indicating that the Moon has no sensible permanent atmosphere.

Eclipsing binary stars
When, in a binary system (a double star whose components orbit a common centre), one component comes between the other and the Earth, an eclipse occurs and the amount of light received from both stars together is reduced; a total eclipse gives the greatest reduction, and the light is reduced to a minimum. Two eclipses occur during each revolution. The first star shown to have light that varied through eclipses in this way was Algol (Beta Persei). Its variability was discovered in 1667 by the Italian astronomer Geminiano Montanari, but its periodicity was not discovered and the explanation of the phenomenon as an eclipsing variable was not proposed until a century later by the English astronomer John Goodricke. At present, about 3,000 eclipsing binaries are known.
The many possible sizes of the two stars, in relation both to each other and to the size and tilt of their orbits, combined with the eccentricity (deviation from the circular) of the orbits, lead to a large variety of ways in which the light of the system changes during a revolution: the so-called light curve.
If the orbits of the two stars are not circular, then the motion in the orbits is not uniform but is most rapid near periastron (stars closest together) and least rapid near apastron (stars farthest apart). Therefore, the eclipses may not be equally spaced in time, as they are in the case of circular orbits. Study of the spacing and durations of the light minima will yield both the eccentricity and the orientation of the orbits. A steady change in the spacing of the minima as well as in their relative duration means that the line of apsides (major axis of the relative orbit) is rotating; the most rapid such rotation known is that of the eclipsing binary GL Carinae.
If the two stars are unequal in size, they may wholly overlap for some time. When the star of higher surface brightness is behind, the resulting eclipse is darker than that resulting when the brighter star cuts off the light of the dimmer one. Limb darkening, the tendency for a star to appear less brilliant near the edge of its apparent disk, also influences the shape of the light curves, and the amount of limb darkening present may be determined from those curves.
Spectroscopic studies of the light of a double star allow determination of the ratio of the masses of the components to each other. The sum of the masses, however, can only be determined if the inclination of the orbit is known. In the case of an eclipsing binary, the plane of the orbit is known to be directed to the Earth. Therefore, eclipsing binaries can be made to reveal complete information about their masses.
A rapidly rotating star bulges at its equator. Also, many double stars are so close together that they distort each other tidally; in the case of very close pairs, this distortion has caused them to turn always the same sides toward each other. If such stars are eclipsing binaries, the amount of distortion can, in some cases, be determined from the light curve. For some stars the ratio of the long axis of the star to the short has been found to be as great as 5 : 3.
There are a few star systems like Algol, notably Zeta Aurigae, in which the difference in the sizes of the components is so great that the crossing of the smaller, brighter star before the disk of the supergiant companion should really be called a transit rather than an eclipse. This particular star has the remarkably long orbital period of 972 days and consists of an orange supergiant having a diameter of almost 300,000,000 kilometres and a substantially smaller, brighter star, only about three times larger than the Sun. The secondary minimum lasts for 38 days, and a very sensitive photometer must be used to detect the extremely slight fading of light when the smaller star is shining through the extended, tenuous atmosphere of the large one.

Eclipses in history
In ancient and medieval times, eclipses of both the Sun and the Moon were often regarded as portents; hence, it is not surprising that many of these events are mentioned in history and in literature, as well as in astronomical writings.
Well over 1,000 individual eclipse records are extant from various parts of the ancient and medieval world. Most known ancient observations of these phenomena originate from three countries: Babylonia, China, and Greece. No records appear to have survived from ancient Egypt or India. Whereas virtually all Babylonian accounts of eclipses are confined to astronomical treatises, those from China and Greece are found in historical and literary works as well. Eclipses are noted from time to time in surviving European writings from the early Middle Ages. At this time only the Chinese, however, continued to observe and record such events on a regular basis, and this tradition continued almost uninterrupted down to recent centuries. Many eclipses were carefully observed by the astronomers of Baghdad and Cairo between about AD 800 and 1000. About AD 800 both European and Arab annalists began to include in their chronicles accounts of eclipses and other remarkable celestial phenomena. Some of these chronicles continued until the 14th or 15th century. Toward the end of this period, European astronomers commenced making fairly accurate measurements of the time of day or night when eclipses occurred, and this pursuit spread rapidly following the invention of the telescope.
The value of ancient and medieval records may be classified as follows, although it should be emphasized that there is some overlap between these individual categories: (1) literary and historical, depending on the interest that these records aroused and their connection with historical events; (2) chronological, insofar as they make it possible to verify chronological systems resting on other evidence and to supply dates for events concerned with eclipses; and (3) astronomical, including the determination by ancient astronomers of the periods and motions of the Sun and the Moon and by modern astronomers of variations in the length of the mean solar day.
The Sun is normally so brilliant that the casual observer is liable to overlook those eclipses in which less than about 80 percent of the solar disk is obscured. Only when a substantial proportion of the Sun is covered does the loss of daylight become noticeable. Hence it is rare to find references to small partial eclipses in literary and historical works. At various times, astronomers in Babylonia, China, and the Arab lands systematically reported eclipses of small magnitude but their vigilance was assisted by their ability to make approximate predictions. They thus knew roughly when to scrutinize the Sun. Arab astronomers sometimes viewed the Sun by reflection in water to diminish its brightness when watching for eclipses. The Roman philosopher and writer Seneca (c. 4 BC–AD 65), on the other hand, recounts that in his time pitch was employed for this purpose. It is not known, however, whether such artificial aids were used regularly.
When the Moon covers a large proportion of the Sun, the sky becomes appreciably darker and stars may appear. On those rare occasions when the whole of the Sun is obscured, the sudden occurrence of intense darkness, accompanied by a noticeable fall in temperature, may leave a profound impression on eyewitnesses. Total or near-total eclipses of the Sun are of special chronological importance. On average, they occur so infrequently at any particular location that if the date of such an event can be established by historical means to within about a decade, it may well prove possible to fix an exact date by astronomical calculation.
The Full Moon is much dimmer than the Sun, and lunar eclipses of even quite small magnitude are thus fairly readily visible to the unaided eye. Both partial and total obscurations are recorded in history with comparable frequency. As total eclipses of the Moon occur rather often (every three or four years on average at a given place), they are of less chronological importance than their solar counterparts. There are, however, several notable exceptions to this rule, as will be discussed below.

Literary and historical references
Old Babylonian
The earliest known references to eclipses for which dates can be established with reasonable confidence go back to the 21st century BC. These are recorded on the series of astrological tablets from Ur known as Enūma Anu Enlil. Several of these texts contain lunar eclipse omina—warnings of disasters that might follow an eclipse based on past coincidences between celestial and terrestrial occurrences. Some of the omina are so detailed that they are clearly based on observation of a specific eclipse. The example cited below is found on tablet 20 of the series:
If in Simanu [lunar month III] an eclipse occurs on day 14, the [Moon-] god in his eclipse is obscured on the east side above and clears on the west side below, the north wind blows, [the eclipse] commences in the first watch of the night and it touches the middle watch. . . . by this the [Moon-] god gives a decision for Ur and the king of Ur. The king of Ur will see a famine, there will be many deaths, the king of Ur will be wronged by his son; the son who has wronged his father, the Sun-god will catch him, and he will die at the burial of his father. A son of the king who was not named for kingship will then occupy the throne.
From a careful investigation of the historical and astronomical circumstances, it has been shown that the eclipse referred to here is very likely to have been associated with the murder of Shulgi by his son and the accession of Amar-Sin. The most probable date for the eclipse is April 4, 2094 BC. A further lunar eclipse 42 years later regarded as signaling the destruction of Ur has been dated to April 13, 2053 BC.

Chinese
According to long-established tradition, the history of astronomy in ancient China could be traced back to before 2000 BC. The earliest surviving relicts that are of astronomical significance date from nearly a millennium later, however. The An-yang oracle bones (inscribed turtle shells, ox bones, and so forth) of the Shang dynasty (c. 1550–1050 BC), which have been uncovered near An-yang in northeastern China, record several eclipses of both the Sun and the Moon. The following report is an example:
On day kuei-yu [the 10th day of a 60-day cycle], it was inquired [by divination]: “The Sun was eclipsed in the evening; is it good?” On day kuei-yu it was inquired: “The Sun was eclipsed in the evening; is it bad?”
The above text provides clear evidence that eclipses were regarded as omens at this early period (as is true of other celestial phenomena). Such a belief was extremely prevalent in China during later centuries. The term translated here as “eclipse” (chih) is the same as the word “eat.” The Shang people thought that some monster was actually devouring the Sun or Moon during an eclipse. Not until many centuries later was the true explanation known; but by then the use of the term chih was firmly established to describe eclipses, and so it continued throughout Chinese history. As the year in which an eclipse occurred is never mentioned on the preserved oracle bones (many of which are mere fragments), dating of these observations by astronomical calculation has proved extremely difficult. A recent investigation of five lunar eclipses yielded likely dates between 1200 and 1180 BC. Shang chronology, however, is still very uncertain.
The Shih ching (“Classic of Poetry”) contains a lamentation occasioned by an eclipse of the Moon followed by an eclipse of the Sun. The text, dating from the 8th century BC, may be translated:
The Sun was eclipsed, a thing of very evil omen. Then the Moon became small, and now the Sun became small. . . . For the Moon to be eclipsed is but an ordinary matter. Now that the Sun has been eclipsed—how bad it is!
The solar eclipse is said to have occurred on the day hsin-mao (the 28th day of the sexagenary cycle), which was the first day of the 10th lunar month. A date of 776 BC was formerly adopted for such an event, but modern computations show that no solar eclipse in that year was visible in China. A revised date of 735 BC has been proposed. The different attitudes toward solar and lunar eclipses at this time is interesting. Throughout the subsequent thousand years or so, lunar eclipses were hardly ever reported in China—in marked contrast to solar obscurations, which were systematically observed.
After about 200 BC, a wide variety of celestial phenomena began to be noted in China on a regular basis. Summaries of these records are found in astronomical treatises contained in the official dynastic histories. In many instances, a report is accompanied by a detailed astrological prognostication. For example, the Hou-Han shu (“History of the Later Han Dynasty”) contains the following account under a year corresponding to AD 119–120.
On the day wu-wu, the 1st day of the 12th lunar month, the Sun was eclipsed; it was almost complete. On the Earth it became like evening. It was 11 degrees in the constellation of the Maid. The woman ruler [i.e., the Empress Dowager] showed aversion to it. Two years and three months later, Teng, the Empress Dowager, died.
The date of this eclipse on the Chinese calendar is equivalent to January 18, AD 120. On this exact day there had occurred an eclipse of the Sun that was very large in China. Such chronological precision is typical of almost all Chinese records of celestial phenomena. The above-cited text is particularly interesting because it clearly describes an obscuration of the Sun, which, though causing dusk conditions, was not quite total where it was seen. The place of observation was probably Lo-yang, the Chinese capital of the time. With regard to the accompanying prognostication, it should be pointed out that a delay of two or three years between the occurrence of a celestial omen and its presumed fulfillment is quite typical of Chinese astrology.

Assyrian
The Assyrian eponym canon, which preserves the names of the annual magistrates who gave their names to the years (similar to the Athenian archons or Roman consuls), records under the year that corresponds to 763–762 BC: “Insurrection in the city of Ashur. In the month Sivan [equivalent to May–June], the Sun was eclipsed.” The reference must be to the eclipse of June 15, 763 BC, the only large solar eclipse visible in Assyria over a period of many years. A possible allusion to the same eclipse is found in the Old Testament: “ ‘And on that day,' says the Lord God, ‘I will make the sun go down at noon, and darken the earth in broad daylight' ” (Amos 8:9). Amos was prophesying during the reign of King Jeroboam II (786–746 BC), and the eclipse would be very large throughout Israel.

Jewish
Apart from Amos, the only Old Testament writer to allude to eclipses is Joel (2:31). The most direct account of an eclipse in ancient Jewish history occurs not in the Bible but in the writings of Flavius Josephus, the 1st-century-AD historian. Not long before the death of Herod the Great, Josephus recounts the occurrence of a lunar obscuration:
As for the other Matthias who had stirred up the sedition, he [Herod] had him burned alive along with some of his companions. And on that same night there was an eclipse of the Moon. But Herod's illness became more and more severe. . . .
This eclipse occurred shortly before the Passover festival. Calculation shows that the only springtime lunar eclipses visible in Israel between 17 BC and AD 3 took place on March 23, 5 BC and March 13, 4 BC. The former was total, while on the latter occasion about one-third of the Moon was covered. These two dates are conveniently close to one another, although the latter date is usually preferred by chronologists, implying that Herod died in the spring of 4 BC.

Greek
In a fragment of a lost poem by Archilochus occur the words:
Nothing there is beyond hope, nothing that can be sworn impossible, nothing wonderful, since Zeus, father of the Olympians, made night from mid-day, hiding the light of the shining Sun, and sore fear came upon men.
This seems a clear reference to a total eclipse. The phenomenon has been identified as the eclipse on April 6, 648 BC, which was total in the Aegean and occurred during Archilochus' lifetime.
Small fragments survive of other early Greek poetic descriptions of eclipses, and the ninth paean of Pindar, addressed to the Thebans, takes an eclipse of the Sun as its theme, as follows:
Beam of the Sun! O thou that seest from afar, what wilt thou be devising? O mother of mine eyes! O star supreme, reft from us in the daytime! Why hast thou perplexed the power of man and the way of wisdom, by rushing forth on a darksome track?
Pindar then proceeds to speculate on the meaning of this omen. Although he prays, “Change this worldwide portent into some painless blessing for Thebes,” he adds, “I in no wise lament whate'er I shall suffer with the rest.” This strongly suggests that Pindar, who was a Theban, had himself recently witnessed a great eclipse at his hometown. The most probable date for the eclipse is April 30, 463 BC; modern calculations indicate that the eclipse was nearly total at Thebes.
The historian Thucydides comments on the frequency of eclipses during the Peloponnesian War, which began in 431 BC and lasted for 27 years. The most interesting of these was a solar eclipse that occurred in the summer of the first year of the war (calculated date August 3, 431 BC) and a lunar eclipse that took place in the summer of the 19th year (calculated date August 27, 413 BC). On the former occasion, “the Sun assumed the shape of a crescent and became full again, and during the eclipse some stars became visible” (a statement that agrees well with modern computations). The latter date had been selected by the Athenian commanders Nicias and Demosthenes for the departure of their armies from Syracuse. All preparations were ready, but the signal had not been given when the Moon was eclipsed. The Athenian soldiers and sailors clamoured against departure, and Nicias, in obedience to the soothsayers, resolved to remain thrice nine days. This delay enabled the Syracusans to capture or destroy the whole of the Athenian fleet and army.
August 15, 310 BC, is the date of a total eclipse of the Sun that is said to have been seen at sea by Agathocles and his men after they had escaped from Syracuse and were on their way to Africa. Diodorus, a historian of the 1st century BC, reports that, “On the next day [after the escape] there occurred such an eclipse of the Sun that utter darkness set in and the stars were seen everywhere.” Modern computations of the eclipse track render it probable that Agathocles' ships passed along the north of Sicily during the course of the journey; the Sun would have only been partially obscured on the south side of the island.
In Plutarch's dialogue concerning the features of the Moon's disk, one of the characters, named Lucius, deduces from the phases of the Moon and the phenomenon of eclipses a similarity between the Earth and the Moon and illustrates his argument by means of a recent eclipse of the Sun, “which, beginning just after noon, showed us plainly many stars in all parts of the heavens, and produced a chill in the temperature like that of twilight.” This eclipse has been identified with one that occurred on March 20, AD 71, which was total in Greece. Whether Plutarch is describing a real, and therefore datable, event or is merely basing his description on accounts written by earlier authors has been disputed, however. Later in the same dialogue, Lucius refers to a brightness that appears around the Moon's rim in total eclipses of the Sun. This is one of the earliest known allusions to the solar corona. Plutarch was unusually interested in eclipses, and his Parallel Lives, an account of the deeds and character of illustrious Greeks and Romans, contains many references to both lunar and solar eclipses of considerable historical importance. There also are frequent records of eclipses in other ancient Greek literature.

Roman
Roman history is less replete with references to eclipses than that of Greece, but there are several interesting references to these events in Roman writings. Some, like the total solar eclipse said by Dio Cassius, a Roman historian of the 3rd century AD, to have occurred at the time of the funeral of Agrippina, the mother of Nero, never took place. One that has attracted the attention of students of astronomy and of the Roman calendar alike is stated by Cicero to have occurred in what may have been the 350th year from the founding of Rome. He also says that it was described by the poet Quintus Ennius: “On the nones of June the Sun was covered by the Moon and night.” This happening would appear to have been the total solar eclipse of June 21, 400 BC, which reached a total or almost total phase at Rome a few minutes after sunset. Its recorded date seems to show that in that year the calendar month of June began 16 days later than it did after the Julian reform. The eclipse of the Moon on June 21–22, 168 BC, has attracted much attention. The Romans were at that time at war with Macedonia, and Polybius says that this eclipse was interpreted as an omen of the eclipse of a king and thus encouraged the Romans and discouraged the Macedonians.
What may well be an indirect allusion to a total eclipse of the Sun that caused darkness at Rome is recorded by Livy for a time corresponding to 188–187 BC (the consulship of Valerius Messalla and Livius Salinator):
Before the new magistrates departed for their provinces, a three-day period of prayer was proclaimed in the name of the College of Decemvirs at all the street-corner shrines because in the daytime at the third hour darkness had covered everything.
The darkness took place some time after the election of the consuls (Ides of March), and, allowing for the confusion of the Roman calendar at this time, the total eclipse of July 17, 188 BC, would be the most satisfactory explanation for the unusual darkness. Since the Sun is not mentioned in the text, the phenomenon possibly occurred on a cloudy day. Two years earlier (190 BC), Livy records an eclipse as happening at the beginning of July. The calculated date, however, is March 14 in that year. Consequently, the Roman calendar in that year must have been as much as 3 1/2 months out of adjustment.

Medieval European
Following the close of the Classical Age, eclipses were in general only rarely recorded by European writers for several centuries. Not until after about AD 800 did eclipses and other celestial phenomena begin to be frequently reported again, especially in monastic chronicles. Hydatius, bishop of Chaves (in Portugal), was one of the few known chroniclers of the early Middle Ages. He seems to have had an unusual interest in eclipses, and he recounted the occurrence of five such events (involving both the Sun and the Moon) between AD 447 and 464. In each case, only brief details are given, and Hydatius gives the years of occurrence in terms of the Olympiads (i.e., reckoning time from the first Olympic Games, in 776 BC). During the total lunar eclipse of March 2, AD 462 (this date is known to be accurate), the Moon is said to have been “turned into blood.” Statements of this kind are common throughout the Middle Ages, presumably inspired by the Old Testament allusion in Joel (2:31). Similar descriptions, however, are occasionally found in non-Christian sources, as, for example, a Chinese one of AD 498.
Given below is a selection from the vast number of extant medieval European reports of eclipses. In many cases, the date is accurately recorded, but there also are frequent instances of chronological error.
An occultation of a bright star by the eclipsed Moon in AD 756 (actually the previous year) is the subject of an entry in the chronicle of Simeon of Durham, compiled some four centuries after the event:
Moreover, the Moon was covered with a blood-red color on the 8th day before the Kalends of December [i.e., November 24] when 15 days old, that is, the Full Moon; and then the darkness gradually decreased and it returned to its original brightness. And remarkably indeed, a bright star following the Moon itself passed through it, and after the return to brightness it preceded the Moon by the same distance as it had followed the Moon before it was obscured.
The text gives no hint of the identity of the star. Modern computations show that the Moon was totally eclipsed on the evening of November 23, AD 755. During the closing stages of the eclipse, Jupiter would have been occulted by the Moon, as seen from England. This is an example of the care with which an observer who was not an astronomer could describe a compound astronomical event without having any real understanding of what was happening.
Several eclipses are recorded in Byzantine history, beginning in the 6th century AD. By far the most vivid account relates to the solar eclipse of December 22, AD 968. This was penned by the contemporary chronicler Leo the Deacon:
At the winter solstice there was an eclipse of the Sun such as has never happened before. . . . It occurred on the 22nd day of the month of December, at the 4th hour of the day, the air being calm. Darkness fell upon the Earth and all the brighter stars revealed themselves. Everyone could see the disk of the Sun without brightness, deprived of light, and a certain dull and feeble glow, like a narrow headband, shining round the extreme parts of the edge of the disk. However, the Sun gradually going past the Moon (for this appeared covering it directly) sent out its original rays, and light filled the Earth again.
This is the earliest account of the solar corona that can be definitely linked to a datable eclipse. Although the appearance of the corona during totality is rather impressive, early descriptions of it are extremely rare. Possibly many ancient and medieval eyewitnesses were so terrified by the onset of sudden darkness that they failed to notice that the darkened Sun was surrounded by a diffuse envelope of light.
In a chronicle of the Norman rule in Sicily and southern Italy during the 11th century, Goffredo Malaterra records an eclipse of the Sun that, even though it caused alarm to some people, was evidently regarded by others as no more than a practical inconvenience:
[AD 1084] On the sixth day of the month of February between the sixth and ninth hours the Sun was obscured for the space of three hours; it was so great that any people who were working indoors could only continue if in the meantime they lit lamps. Indeed some people went from house to house to get lanterns or torches. Many were terrified.
This eclipse actually occurred on February 16, AD 1086. It was the only large eclipse visible in southern Italy for several years around this time; hence, the chronicler had mistaken both the year and day.

Medieval Islāmic
Like their Christian counterparts, medieval Islāmic chroniclers record a number of detailed and often vivid descriptions of eclipses. Usually the exact date of occurrence is given (on the lunar calendar). A graphic narrative of the total solar eclipse of June 20, AD 1061, is recorded by the Baghdad annalist Ibn al-Jawzι, who wrote approximately a century after the event:
On Wednesday, when two nights remained to the completion of the month Jumādā I [in AH 453], two hours after daybreak, the Sun was eclipsed totally. There was darkness and the birds fell whilst flying. The astrologers claimed that one-sixth of the Sun should have remained [uneclipsed] but nothing of it did so. The Sun reappeared after four hours and a fraction. The eclipse was not in the whole of the Sun in places other than Baghdad and its provinces.
The date corresponds exactly to June 20, AD 1061, on the morning of which there was a total eclipse of the Sun visible in Baghdad. The duration of totality is much exaggerated, but this is common in medieval accounts of eclipses. The phenomenon of birds falling from the sky at the onset of the total phase was also noticed in Europe during several eclipses in the Middle Ages.
Two independent accounts of the total solar eclipse of AD 1176 are recorded in contemporary Arab history. Ibn al-Athιr, who was age 16 at the time, described the event as follows:
In this year [AH 571] the Sun was eclipsed totally and the Earth was in darkness so that it was like a dark night and the stars appeared. That was the forenoon of Friday the 29th of the month Ramaḍān at Jazιrat Ibn ʿUmar, when I was young and in the company of my arithmetic teacher. When I saw it I was very much afraid; I held on to him and my heart was strengthened. My teacher was learned about the stars and told me, “Now, you will see that all of this will go away,” and it went quickly.
The date of the eclipse is given correctly apart from the weekday (actually Sunday) and is equivalent to April 11, AD 1176. Calculation shows that the whole of the Sun would have been obscured over a wide region around Jazιrat Ibn ʿUmar (now Cizre in Turkey). Farther south, totality was also witnessed by Saladin and his army while crossing the Orontes River near Hamāh (in present-day Syria). The chronicler ʿImād al-Dīn, who was with Saladin at the time, noted that, “The Sun was eclipsed and it became dark in the daytime. People were frightened and stars appeared.” As it happens, ʿImād al-Dīn dates the event one year too early (AH 570), but the only large eclipse visible in this region for several years occurred in AD 1176.
Lunar and solar eclipses are fairly frequently visible on the Earth's surface 15 days apart, and from time to time such a pair of eclipses may be seen from one and the same location. Such was the case in the summer of AD 1433, but this occurrence caused some surprise to the contemporary Cairo chronicler al-Maqrīzī:
On Wednesday the 28th of Shawwāl [i.e., June 17], the Sun was eclipsed by about two-thirds in the sign of Cancer more than one hour after the afternoon prayer. The eclipse cleared at sunset. During the eclipse there was darkness and some stars appeared. . . . On Friday night the 14th of Dhu l-Quʿda [July 3], most of the Moon was eclipsed. It rose eclipsed from the eastern horizon. The eclipse cleared in the time of the nightfall prayer. This is a rarity—the occurrence of a lunar eclipse 15 days after a solar eclipse.
The loss of daylight produced by the solar eclipse is much exaggerated but otherwise the description is fairly careful.

Uses of eclipses for chronological purposes
Several examples of the value of eclipses in chronology have already been mentioned in passing. No one system of dating has been continuously in use since ancient times, although some, like the Olympiads, persisted for many centuries. Dates were frequently expressed in terms of a king's reign; years were also named after officials of whom lists have been preserved (the eponym canons mentioned above). In such cases, it is important to be able to equate certain specific years thus defined with years before the Christian Era (BC). This correspondence can be made whenever the date of an eclipse is given in an ancient record. In this regard, eclipses have distinct advantages over other celestial phenomena such as comets: in addition to being frequently recorded in history, their dates of occurrence can be calculated exactly.
Chinese chronology can be confirmed accurately by eclipses from the 8th century BC (during the Chou dynasty) onward. The Ch'un-ch'iu (“Spring and Autumn Annals”), a chronicle covering the period from 722 to 481 BC, notes the occurrence of 36 solar eclipses during this interval. This is the earliest surviving series of eclipse observations from any part of the world. The records give the date of each event in the following form: year of the ruler, lunar month, and day of the 60-day cycle. Three of the eclipses (occurring in 709, 601, and 549 BC) were described as total. As many as 32 of the eclipses cited in the Ch'un-ch'iu can be identified by modern calculations. Errors in the recorded lunar month (typically amounting to no more then a single month) are fairly common, but both the year and the recorded day of the sexagenary cycle are invariably correct.
The chronology of Ptolemy's canon list of kings, which gives the Babylonian series from 747 to 539 BC, the Persian series from 538 to 324 BC, the Alexandrian series from 323 to 30 BC, and the Roman series from 30 BC onward, is confirmed by eclipses. The eclipse of 763 BC, recorded in the Assyrian eponym canon, makes it possible to carry the chronology back with certainty through the period covered by that canon to 893 BC. Identifiable eclipses that were recorded under named Roman consuls extend back to 217 BC. The dated solar eclipse of Ennius (400 BC), the lunar eclipse seen at Pydna in Macedonia on June 21–22, 168 BC, and the solar eclipse recorded at Rome in 190 BC can be used to determine months in the Roman calendar in the natural year. Furthermore, eclipses occasionally help to fix the precise dates of a series of events, such as those associated with the Athenian disaster at Syracuse.
The late Babylonian astronomical texts occasionally mention major historical events, as, for example, the dates when Xerxes and Alexander the Great died. Most of these clay tablets, inscribed with a cuneiform script, are now found in the British Museum. The preserved texts are mainly in the form of day-to-day diaries of celestial observations and summary tables that abstract specific types of observation from the diaries. To illustrate the potential of this material for chronological purposes, the date of the death of Xerxes may be accurately fixed by reference to eclipses. On a tablet that lists lunar eclipses at 18-year intervals occurs the following brief announcement between two eclipse records: “Month V, day 14 [?], Xerxes was murdered by his son.” Unfortunately, the cuneiform sign for the day of the month is damaged, and a viable reading could be anything from 14 to 18. The year is missing, but it can be deduced from the 18-year sequence as 465 BC. This identification is confirmed by calculating the dates of the two eclipses stated to have occurred in the same year that Xerxes died. The first of these happened when the Moon was in the constellation of Sagittarius, while the second took place on the 14th day of the 8th lunar month. For many years both before and after 465 BC, no such combination of eclipses can be found; it occurs only in 465 BC itself. The dates deduced for the two eclipses are June 5 and November 30 of that year. Mention of an intercalary sixth month on the same tablet enables the date of the death of Xerxes to be fixed as some time between August 4 and 8 in 465 BC.

Uses of eclipses for astronomical purposes
In ancient astronomy
It is known from both Babylonian and Greek history that at least from the time of King Nabonassar (whose reign began in 747 BC), a dated canon of astronomical observations was preserved at Babylon. This included numerous eclipses of both the Sun and the Moon. Reference to the list of lunar eclipses would enable the Babylonian astronomers to determine accurately the intervals between such eclipses and must have facilitated the discovery of the 18-year cycle (more exactly the cycle of 6,585 1/3 days that the 10th-century Greek lexicographer Suidas named the saros). This cycle is attested in Babylonian astronomy at a fairly early period. In contrast to lunar obscurations, the visibility of solar eclipses at a given place on the Earth's surface is complicated by geographic considerations. Hence, even a lengthy list of solar eclipse dates would be of limited value in deducing the saros.
The surviving late Babylonian astronomical texts contain many examples of predictions of both lunar and solar eclipses using cycles. Comparison with modern computations shows that, although some predictions were successful, in other instances an eclipse was visible only elsewhere on the Earth's surface or did not occur at all. A theory based on a detailed knowledge of apparent lunar and solar motions is necessary to enable eclipses to be predicted accurately. Lunar eclipses indicate more accurately than any other phenomena the actual time when the Sun and the Moon are in opposition. From an early date the Babylonian astronomers must have deduced from lunar eclipses not only the mean interval between two conjunctions (closest apparent approaches) of the Sun and the Moon but also the principal inequality (change of speed) in the motion of the Moon and the similar inequality in the motion of the Earth (or, as according to their geocentric theory they conceived it, of the Sun). The Babylonians were able to define the periods of these inequalities (the cause of which lies in the ellipticities of the orbits), which astronomers refer to as the anomalistic month and year.
In the same way, since eclipses happen only when the Sun and the Moon are at the intersections of their orbital planes called nodes, and since the path of the shadow in a lunar eclipse depends on the position of the centre of the Sun in relation to the node, the Babylonians were also able to determine the position and motion of the nodes. By assuming, as is approximately true, that the saros of 6,585 1/3 days contained an exact number: (1) of synodic months, or revolutions of the Moon measured from the Sun, (2) of anomalistic months, or revolutions of the Moon measured from its apogee or perigee (i.e., from its farthest distance from and closest approach to the Earth), and (3) of draconic months, or revolutions of the Moon measured from its node, astronomers, perhaps as early as the 6th century BC, computed the relative motions of the Sun and the Moon, the lunar perigee and apogee, and the nodes. About 500 BC the Babylonian astronomer Nabu-rimmani, apparently from a more accurate study of eclipse observations, obtained improved values that: for the motion of the Moon relative to the Sun were 10″ of arc per annum too small; for the Moon's perigee motion 20″ of arc per annum too great; and for the motion of its node 5″ of arc too small. Still more accurate values were obtained by Kidinnu about 383 BC, from whom they passed to the Greek astronomer Hipparchus. In the system of Nabu-rimmani the distance of the Moon from its node was used for the prediction of the magnitude of lunar eclipses.

In modern astronomy
Ancient and medieval observations of eclipses are of the highest value for investigating long-term variations in the length of the day. Early investigators such as Edmond Halley deduced from eclipse observations that the Moon's motion was subject to an acceleration. However, not until 1939 was it demonstrated (by Harold Spencer Jones) that only part of this acceleration was real. The remainder was apparent and was a consequence of the practice of measuring time relative to a non-uniform unit—namely, the rotation of the Earth. Time determined in this way is termed Universal Time. For astronomical purposes, it is preferable to utilize an invariant time-frame such as Ephemeris Time.
Lunar and solar tidal friction, occurring especially in the seas and oceans of the Earth, is now known to be responsible for a gradual decrease in the terrestrial rate of rotation. Apart from slowing down the Earth's rotation, lunar tides produce a reciprocal effect on the Moon's motion, causing a gradual increase in the mean distance of the Moon from the Earth (at about 3.5 centimetres [1.4 inches] per year) and a consequent real retardation of its motion. Hence, the length of the month is slowly increasing. These changes in the Moon's orbit can now be accurately fixed by lunar laser ranging, and it seems likely that they have proceeded at an essentially constant rate for many centuries. The history of the Earth's rotation, however, is complicated by effects of nontidal origin, and in order to obtain maximum information it is necessary to utilize both modern and ancient observations. Telescopic observations reveal fluctuations in the length of the day on time scales as long as decades, and these fluctuations are mainly attributed to interactions between the fluid core of the Earth and the surrounding solid mantle. Ancient and medieval observations also suggest the presence of longer term variations, which could be produced by alterations in the moment of inertia of the Earth resulting from both the ongoing rise of land that was glaciated during the Pleistocene ice age and sea-level changes associated with the freezing and melting of polar ice.
Records of large solar eclipses preserved in literary and historical works have made an important contribution to the study of past variations in the Earth's rate of rotation. Nevertheless, the major contribution has come from the analysis of timings of lunar and solar eclipses by ancient Babylonian and medieval Arab astronomers. (Unfortunately, there are very few measurements of intermediate date.) Although many Babylonian texts are fragmentary, nearly 100 usable timings of eclipse contacts are accessible (including measurements at different phases of the same eclipse). These observations date primarily from between about 550 and 50 BC. By comparison, only a handful of similar Greek measurements are preserved, and these are far less precise. About 50 eclipse timings by medieval Arab astronomers are preserved; these are mainly contained in the Hakemite Tables compiled by Ibn Yūnus about AD 1005.
Tidal computations indicate a steady increase in the length of the mean solar day by about 1/40 second every millennium, with other nontidal causes producing additional smaller effects. Although this seemingly represents a minute change, the long time scale covered by ancient observations is an important asset. Approximately one million days—each marginally shorter than at present—have elapsed since the earliest reliable eclipse observations were made, about 700 BC. As a result, present-day computations of ancient eclipses that make no allowance for any increase in the length of the day may be as much as five or six hours ahead of the observed time of occurrence. In the case of total solar eclipses, the path of the Moon's shadow across the Earth's surface may appear to be displaced by thousands of kilometres.
The technique of using ancient observations to investigate changes in the rate of the Earth's rotation is well illustrated by a total solar eclipse observed by Babylonian astronomers on a date corresponding to April 15 in 136 BC. This event is recorded on two damaged tablets, a composite translation of which follows:
At 24 degrees after sunrise, there was a solar eclipse beginning on the southwest side. After 18 degrees it became total such that there was complete night. Venus, Mercury, and the normal stars were visible. Jupiter and Mars, which were in their period of disappearance, were visible in that eclipse. [The shadow] moved from southwest to northeast. [Time interval of] 35 degrees for obscuration and clearing up.
This is an exceptionally fine account of a total solar eclipse and is by far the best preserved from the ancient world. As will be seen, the Babylonians were able to detect a number of stars, as well as four planets, during the few minutes of darkness. Modern calculations confirm that Jupiter and Mars were too near the Sun to be observed under normal circumstances; Jupiter was very close to the solar disk.
Time intervals were expressed by the Babylonians in degrees, each equivalent to 4 minutes of time. Hence the eclipse is recorded as beginning 96 minutes after sunrise (or about 7:10 AM), becoming total 72 minutes later and lasting from start to finish for 140 minutes. Computations that make no allowance for changes in the length of the day suggest that this eclipse was barely visible at Babylon, with as little as 15 percent of the Sun being covered. Furthermore, the computed time of onset is around noon rather than in the early morning. In order to best comply with the record, it is necessary to assume that the length of the day has increased by about 1/20 second in the intervening two millennia.
Numerous eclipses of both the Sun and the Moon were timed by the Babylonian astronomers with similar care, and analysis of the available records closely confirms the above result for the change in the length of the day. Presumably some kind of clepsydra, or water clock, was used to measure time intervals. Although such a device is likely to have been of low precision, many eclipse observations were made fairly close to the reference moments of sunrise or sunset. Hence, the measured intervals would be so short that clock errors may be presumed to be small.
The latest known Babylonian observations date from about 50 BC. After this date, eclipse measurements of comparable precision—by Arab astronomers—are not found until AD 800. The following observations of the lunar eclipse of September 17, AD 1019, made by al-Bīrūnī at Ghazna (now Ghaznī, Afghanistan) attest to the quality of these more recent data:
When I observed it, the altitude of Capella above the eastern horizon was slightly less than 60 degrees when the cut at the edge of the Full Moon had become visible; the altitude of Sirius was [then] 17 degrees, that of Procyon was 22 degrees and that of Aldebaran was 63 degrees, where all altitudes are measured from the eastern horizon.
All of these various measurements are in agreement that the eclipse began at around 2:15 AM, but calculations that make no allowance for any change in the length of the day indicate a time approximately 1/2 hour later.
Combining the various results obtained from analysis of ancient and medieval data, it is possible to show that over the last 1,000 years the average rate of increase in the length of the day was only about two-thirds of what it was in the previous millennium. This emphasizes the importance of nontidal effects in producing changes in the rate of the Earth's rotation period. In sum, the history of the Earth's rotation is extremely complex.

Jakob HoutgastF. Richard Stephenson
Additional Reading
Bryan Brewer, Eclipse (1978); and David Allen and Carol Allen, Eclipse (1987), explain and recount the history of eclipses for the general reader. Donald H. Menzel and Jay M. Pasachoff, “Solar Eclipse: Nature's Super Spectacular,” National Geographic, 138(2):222–233 (August 1970), chronicles the events of a solar eclipse expedition. Frank Dyson and R.v.d.r. Woolley, Eclipses of the Sun and Moon (1937); and J.B. Zirker, Total Eclipses of the Sun (1984), discuss in considerable detail the history, methods, and results of eclipse observations. W.M. Smart, Textbook on Spherical Astronomy, 6th ed., rev. by R.M. Green (1977), presents the basic mathematical tools for calculating occultations and eclipses. Great Britain Nautical Almanac Office, Explanatory Supplement to the Astronomical Ephemeris and the American Ephemeris and Nautical Almanac (1961, reissued 1977), presents in comprehensive fashion data for the calculation of astronomical phenomena. F. Link, Eclipse Phenomena in Astronomy (1969), treats modern developments in eclipse problems, excluding solar eclipses and eclipsing variables. Works on historical eclipses include F. Richard Stephenson, “Historical Eclipses,” Scientific American, 247(4):170–183 (October 1982); Said S. Said, F. Richard Stephenson, and Wafiq Rada, “Records of Solar Eclipses in Arabic Chronicles,” Bulletin of the School of Oriental and African Studies, 52:38–64 (1989); and F. Richard Stephenson and S.S. Said, “Non-tidal Changes in the Earth's Rate of Rotation as Deduced from Medieval Eclipse Observations,” Astronomy and Astrophysics, 215(1):181–189 (1989). Catalogs of eclipse data include Th. Ritter Von Oppolzer, Canon of Eclipses (1962; originally published in German, 1887), astronomical data of all solar eclipses between 1207 BC and AD 2161 and eclipses of the Moon from 1206 BC to AD 2163, with maps of the central lines of the solar eclipses over the Earth; George Van Den Bergh, Eclipses in the Second Millennium BC (-1600 to -1207) (1954), a demonstration of a method for computing each of these eclipses with simple arithmetic, and Periodicity and Variation of Solar (and Lunar) Eclipses (1955; originally published in Dutch, 1951), an arrangement of all the eclipses in Oppolzer's Canon into the saros and the inex periods in a newly developed panorama; Robert R. Newton, Ancient Astronomical Observations and the Accelerations of the Earth and Moon (1970), and Medieval Chronicles and the Rotation of the Earth (1972); and Hermann Mucke and Jean Meeus, Canon of Solar Eclipses −2003 to +2526 (1983), and Canon of Lunar Eclipses −2002 to +2526, 2nd ed. (1983), more accurate data and maps of central lines over the Earth of all solar and lunar eclipses over 4,500 years. Joseph Needham, Science and Civilisation in China, vol. 3, Mathematics and the Sciences of the Heavens and the Earth (1959), includes a discussion of eclipses placed in the context of Chinese astronomy with extensive references to original literature. Bernard Lovell (ed.), Astronomy, 2 vol. (1970), contains discourses in the physical sciences from 1851 to 1939, a large number of which are devoted to solar eclipses.