•1* Of ]ty WZr^tjU A4A* /fa** c &$: / £ VXp &-£stn** SOLAR AND LUNAR ECLIPSES FAMILIARLY ILLUSTRATED AND EXPLAINED, WITH THE METHOD OF CALCULATING THEM ACC0RDI5G TO THE THEORY OF ASTRONOMY, AS TAUGHT IH NEW ENGLAND COLLEGES BY JAMES H. COFFIN, A.M. NEW YORK: COLLINS, BROTHER & CO 1845. Entered according to Act of Congress, in the year 1845, BY JAMES H. COFFIN, In the Clerk's Office of the District Court of Connecticut. 541 CC4 PREFACE The design of this treatise, is to explain the " rationale" of some of the most interesting astronomical calculations, in such a way that the student may clearly see the reason of every step, and its connection with the theory. In this respect it differs from many others, which give the rules for calculating merely, without any explanation of the reason of them. Being partly designed as a text hook for colleges, the author has endeavoured to adapt it to the design of college education, which is not so much to make adept practitioners in any particular science, as to give broad and com- prehensive views of the whole field. Hence, the principles of the several sciences should be thoroughly understood by the student ; but the application of them to practice by mere rules is foreign to the design of a collegiate course of study. If, therefore, the calcu- lations of astronomy are attended to at all in college, it should be in such a way, that the connection with the theory may be appa- rent, and that the two may mutually illustrate each other. In many of treatises for colleges, this point seems to be overlooked. Some of them contain tables for astronomical calculations which are very minute and accurate, and at the same time, so constructed and arranged as to reduce the labour of calculation as much as possible ; but the student can see no connection between them and the motions and perturbations which occupy his attention in the study of the theory. In fact, one who has studied the theory with ever so much thoroughness, has here very little advantage over one who is entirely ignorant of it ; each being guided wholly by rules that must appear entirely arbitrary. Such tables not being adapted to the design of this treatise, the author found it necessary to prepare a set differing somewhat in 111505 IV PREFACE. their construction from those in ordinary use. They are based, for the most part, on those of Delambre and Burg, with the more recent improvements of Airy and Bessel ; but varied in the plan of construction so as to adapt them to this work. Calculations may be made from them sufficiently accurate for the ordinary purposes of an almanac ; yet, as they are not designed especially for that purpose, the aim has been not so much to secure extreme accuracy in the results, as to render them easily understood. The quantities in the tables are given, for the most part, in degrees and decimals, instead of signs, degrees, minutes, and sec- onds, with a view to facilitate the labour of calculation, and to secure the same degree of accuracy with a less number of figures. The work is intended to be complete in itself, on the subject on which it treats, for those who have a general knowledge of the motions of the heavenly bodies ; yet, it would, doubtless, contribute to a better understanding of it, to read some such work as Olm- sted's or Herschell's Astronomy previously. Williams College, October, 1843. CONTENTS. PART I. CHAPTER I. GENERAL CONDITIONS OF AN ECLIPSE. 1, Causes of eclipses.— 2. Must occur near the Moon's nodes — 3. Moon's latitude in solar eclipses. — 4. Do. in lunar eclipses. — 5. Ecliptic limits.— 6. Solar. — 7. Lunar. — 8. Eclipses must occur six months apart. CHAPTER II. MEAN TIME OF AN ECLIPSE. 9. Time when the sun passes the moon's nodes, how found. — 10. Motions not uniform. — 11. Mean time of an eclipse, how found. — 12. Annual equations of moon's perigee and node explained. — 13. The same applied. CHAPTER ILL EQUATION OF THE CENTRE. 14. Velocity must be variable in eliptical orbits.— 15. The same shown by the principle of equal areas in equal times. — 15. Equation of the centre explained. — 17. The same applied. CHAPTER IV, TIONS OF LONGITUDE. 18. Moon's motion disturbed by sun's attraction. — 19. Annual equation of the moon's longi" tude explained and applied. — 20. Acceleration of the moon's mean motion. — 21. Its discov- ery. — 22. Theories to account for it. — 23. Earth's orbit becoming less eliptical. — 24. This shown to be the cause of the acceleration. — 25. Secular equation applied. CHAPTER V. PERTURBATIONS IN THE MOON's MOTION, CONTINUED. VARIATION. 26. Introduction to the subject of the chapter. — 27. Variation. — 28. Sun's attraction sometimes increases and sometimes diminishes moon's gravity. — 29. Moon's motion alternately ac- celerated and retarded. — 30. Effect upon the shape of the orbit. — 31. Combined effects of perturbation in velocity and shape. — 32. The Table for Variation explained.— 33. The equation applied. — 34. Annual equation of variation explained and applied. CHAPTER VI. PERTURBATION IN THE MOOn's MOTION, CONTINUED. EVECTION. 35. Difference between Variation and Evection. — 36. Evection caused, in part, by the une- quable motion of the apsides. — 37. Caused, in part, by a variation in the eccentricity of the orbit. — 38. Eccentricity depends on relative attraction. — 39. The principle applied to the moon's orbit. — 40. Apsides must progress on the whole. — 41. Their place, how deter- mined, physically. — 42. Their motion, how affected by an increase of gravity. — 43. How Vi CONTENTS. by a diminution of the same. — 44. How by a perturbation in velocity. — 45. Neutral points. — 46. Addititious force. — 47. Ablatitious force. — 48. The two combined. — 47. Neutral points determined. — 50. Increase of gravity in quadrature only equal to half the diminution in syzygy. — 51. Tangential force. — 52. Apsides progress when within 45° of syzygy. — 53. Remain nearly stationary when between 45° and 54° 43' 56". — 54. Regress when more distant. — 55. Progress exceeds the regress. — 56. Variation in eccentricity and motion of apsides represented by a figure. — 57. Evection represented by the same figure. — 58. Argument of evection explained. — 59. The same computed. — 60. Evectiun applied. — 61. Annual equation of evection. CHAPTER VII. PERTURBATIONS IN THE MOON'S MOTION, CONCLUDED. NODAL EQUATION OF THE MOON'S LONGITUDE. CORRECTION OF EQUATION OF THE CEN- TRE. REDUCTION TO THE ECLIPTIC. 62. Disturbing force acts out of the plane of the moon's orbit.— 63. Theory of nodal equation explained.— 64. Equation applied. — 65. Equation of the centre corrected.— 66. Reduction explained and applied. CHAPTER VIII. LUNAR OR MENSTRUAL EQUATION OF THE SUN*S LONGITUDE. 67. Theory of lunar equation.— 68. The same applied.— 69. Effect of inequalities inappre- ciable. CHAPTER IX. NUTATION IN LONGITUDE. 70. Longitude affected by precession of the equinoxes. — 71. Precession explained. — 72. Caused, in part, by the moon's attraction.— 73. When most rapid.— 74. Lunar nutation, how caused.— 75. Equation for lunar nutation explained.— 76. The same applied.— 77. Solar nutation. CHAPTER X. TRUE TIME LONGITUDES AND ANOMALIES. 78. Correction in the time of new moon. — 79. Greenwich time. — 80. Correction of the lon- gitudes.— 81. Synopsis.— 82. Preliminary equations explained.— 83. Application of the same.— 84. Synopsis. CHAPTER XL ELEMENTS OF AN ECLIPSE. 85. List of elements.— 86. Time and longitudes.— 87. Obliquity of ecliptic to equator.— 88. Moon's latitude.— 89. Hourly motions of sun and moon.— 90. Sun's apparent semidiame- ter.— 91. Moon's semidiameter and parallax.— 92. Semidiameter of earth's shadow.— 93. Angle of moon's path. — 94. Sun's declination. — 95. Elements collected. CHAPTER XII. DELINEATION OF A SOLAR ECLIPSE. 96. Appearance of the earth as viewed from the sun, while the latter advances from the vernal equinox to the summer solstice. — 97. The same, while it advances to the autumnal equinox. — 98. The same, while it advances through the winter solstice to the vernal equi- nox. — 99. Subject illustrated by a terrestrial globe. — 100. Relative size of objects, how measured in astronomy. — 101. To draw the northern half of the earth's disc, as seen from CONTENTS. VII the sun. 102. To find the position of the north pole on the disc. — 103. To find the position of any given place at noon on the disc. — 104. To find the same at midnight. — 105. To find the same at six o'clock, morning and evening. — 106. To find the same at any other hour, and to draw the apparent diurnal path of the place. — 107. The construction shows the time of sunrise or sunset. — 108. To find the position of the moon when new, as projected upon the earth's disc. — 109. To draw its path across the disc. — 110. To find its position at any time. — 111. To find how many digits will be eclipsed. — 112. To find when the eclipse will begin and end. — 113. To find the above by calculation. CHAPTER XIII. CENTRAL TRACK OF A SOLAR ECLIPSE. Hi. Track may be found by the drawing in the last chapter. — 115. A better method by a terrestrial globe. — 116. How to find where the centre of the moon's shadow first strikes the earth. — 117. How to find where it leaves it. — 118. How to find where the eclipse is central at any given time. — 119. Track may be found more accurately still by calcula- tion. — 120. How to find, by calculation, where the centre of the moon's shadow strikes or leaves the earth. — 121. How to find, by calculation, where the eclipse will be central at any given time. — 132. The same, more fully explained. — 123. Results of the calculation in the solar eclipse of May, 1854.— 124. General description of the track of that eclipse. CHAPTER XIV. DELINEATION OF A LUNAR ECLIPSE. 125. To draw the earth's shadow.— 126. To find the position of the moon when full, in res- pect to the shadow.— 127. To draw its path.— 128. To find the position of the moon at any lime during the eclipse.— 129. The same obtained by calculation.— 130. Results of the calculation in the lunar eclipse of November, 1844. 10 earth. The angle MES, is not difficult of cpmputation, being equal to SEA+MEC+AEC, the latter of which is equal (Euc. 1, 32) to ECB— EAB. Hence, MES=SEA+MEC+ECB— EAB, all of which are easily found. The two former are the apparent semidiameters of the sun and moon, as viewed from the earth, and the two latter the semidiameter of the earth, as viewed from the moon and sun, or, respectively, the moon's and sun's horizontal parallax. On account of the distance of the sun and moon from the earth not being constant, the angle MES is subject to a varia- tion in size, being sometimes 1° 38', and at other times not more than 1° 14'. The reader will perceive that, when the sun and moon are in conjunction, the angle MES is the moon's latitude ; and the conclusion to which we have just arrived, may be express- ed thus : If the latitude of the moon, when new, is less than 1° 38' there may be an eclipse of the sun, and if it is less than 1° 14' there must be one. Fig. 2. 4. Again, let S (Fig. 2) represent the centre of the sun, E that of the earth, and M that of the moon, just impinging upon the earth's shadow. The angle MET, as represented in the figure, is plainly the least possible, without producing a partial or total eclipse of the moon. The angle MET=MEF+FET, both of which can be easily computed. The former is the moon's appa- rent simidiameter, as seen from the earth, and FET is that of the section of the earth's shadow that eclipses the moon. Now, (Euc. 1, 32,) FET=BFE— FHE and FHE=AES— BAE ; therefore, FET=BFE+BAE — AES, the two former of which are the lunar and solar parallaxes, and the latter is the sun's apparent semidiam- eter, as seen from the earth. That is, the apparent semidiameter of the section of the earth's shadow thai eclipses the moon, is equal to the sum of the parallaxes of the sun and moon, diminished by the sun's apparent semidiameter. And if this angle be increased by the moon's apparent semidiameter, MEF, we shall have the whole angle MET, which is the least latitude the moon can have in oppo- 11 sition, without being eclipsed. This angle varies in size, like the analagous one in solar eclipses, just described, and for the same reason. Its maximum value is 1° 4', and its minimum 50'. 5. The centre of ihe section of the earth's shadow that eclipses the moon, must be situated in the plane of the ecliptic, directly op- posite to the sun, and must, therefore, be at the same distance from one of the moon's nodes that the sun is from the other. Now, we wish to know how near the sun, in its annual course, may approach to one of the moon's nodes, without occasioning eclipses ; or. in other words, at what distance from the node the moon's track and the ecliptic will have diverged, so as to be from 1° 14' to 1° 38' apart, if our inquiry relates to solar eclipses — or, from 50' to 1° 4'» if it relates to lunar. 6. Let AN (Fig. 3) represent a portion of the ecliptic, BN a por- tion of the moon's orbit, SM a por- tion of a secondary to the ecliptic, and N one of the moon's nodes. Then, in the right angled spheri- cal triangle, SMN, we have the angle SNM=5° 7' 47".9,* and for solar eclipses, the arc SM=1° 14' to 1° 38'. With these data, we find NS to be from 13° 14' to 19° 42', according to the value we give to SM. Hence, if the sun is within 19° 42' of the moon's node, on either side, at the time of new moon, it may be eclipsed ; and if it is within 13° 24', it must be. These distances are called the solar ecliptic limits. Since it takes the sun more than a lunar month, usually, to pass over one of these arcs, it follows, that it must be eclipsed at every passage, and, consequently, twice a year, at least. It may be eclipsed twice during one passage ; once, just after it enters the ecliptic limits, and again, just before it leaves them : but, if so, both of the eclipses will be small, and not central upon any part of the earth. 7. For lunar eclipses, the arc SM is 50' to 1° 4', and, by the same process as above, we find the lunar ecliptic limits to be, irom 7° 47" to 13° 21' on each side of the node.f They embrace an * This angle is subject to a slight variation, amounting, at it3 maximum, to 6' 47".lo. t Baiiy. 12 arc considerably less than is passed over by the sun in one luna- tion, so that it often happens, that the sun passes the node without there being any lunar eclipse. The lunar ecliptic limits being so much less than the solar, eclipses of the moon must be proportionably less frequent; yet, since a lunar eclipse is always visible over half the earth's surface, while one of the sun can be seen only over a very much smaller section, there will, on an average, be a greater number of visible eclipses of the moon, at any given place, than of the sun, .8. As the moon's nodes are 180° apart, or, in opposite points of the ecliptic, the interval between eclipses occurring at one node, and those occurring at the other, must be about six months. Also, since the nodes move backward about 19° in each year, eclipses must happen, on an average, nearly three weeks earlier every year than they did on the year preceding. The reader will see that these conclusions are verified by past experience, if he will take the trouble to examine the almanacs of former years. CHAPTER II. MEAN TIME OF AN ECLIPSE, AND THE MEAN LONGITUDES AND ANOMA- LIES OF THE SUN AND MOON. 9. The sun, in its apparent annual course, leaves the vernal equi- nox where its longitude is 0°, about the 21st of March, and moves eastward, towards the moon's nodes, about 1° each day. Conse- quently, it must arrive at either node, in about as many days after the 21st of March, in any given year, as the longitude of the node, in that year, contains degrees. At the next new moon before or after the date thus found, (more frequently the former.) there will be an eclipse of the sun. It is probable, though not certain, that there will also be a lunar eclipse at the nearest full moon. It will occur then if at all at that passage of the node. 10. The velocity of the motions of sun and moon in their respect- ive orbits, is variable ; but it is more convenient in astronomical calculations to regard it as uniform, and to make the necessary cor- rections for the inequalities afterward. 13 11. To explain the method of calculating the time of an eclipse, we will take, as an example, the solar eclipse that will occur when the sun passes the moon's ascending node, in the year 1854, Ta- ble 2d, at the end of this volume, contains the time of new moon in March of that year, as well as of every other during the present century, and the longitude of the sun, moon, and moon's ascending node, all calculated on the supposition of a uniform rate of motion. It also gives the .mean anomolies of the sun and moon, i. e., the dis- tance of each from its perigee. The longitude of the descending node may be found by adding, or subtracting, 180° to, or from, that of the ascending node. We might compute all these quanti- ties from the data given in table 1st, but table 2d supercedes the necessity. Although some of them will not be used in this chapter, it is most convenient to take them all out together, and write them as below. The longitude of the node on that year (see right hand column of the table) is 64°.2158 ; consequently, the sun will arrive at it about 64 days after the 21st of March, which carries the time to May 24th. The eclipse in question will occur at the new moon nearest that time. Entering table 3d,*' we next take out such a number of lunations (in this ease, two) as, when added to the time of new moon in March, will bring it near to the time when the sun reaches the moon's node, and write it down, with the longitudes and anomalies, under the corresponding quantities already taken from table 2d. These must be added together, (with the exception of the right hand column, where the lower number must be sub- tracted from the upper, because the motion of the node is retro- grade,) and we thus obtain the time of mean new moon in May. The following shows the operation : — Mean new moon in March,... Add two lunations, Mean new moon in May, Time. h. ?n. s, 28 10 42 50 59 1 28 6 26 12 10 5t Sun's Anom- aly. 85.586 58.21 1 143.797 Sun's Longi- tude. 6°0194 58.2135 Moon's Anom- aly. 93.665 51.634 64.2329|145.299 Moon's I Longi- Longi- tude of tude. Node. 6.0194 64°2158 58.2135l-3.1275 64.2329 i 61.0883 Table 5th shows the month and day to which any number of days found by the foregoing addition corresponds. At this stage of the calculation, it is well to compare the longi- tude of the sun and moon with that of the node, and if they do not * Table 3d shows the length of any number of mean lunations, from one to thirteen, with the mean motions of the sun and moon, both in longitude and anomaly, during the same ; also, the mean motion of the moon's nodes. 14 differ more than 20°, there may be an eclipse, though there will not, probably, be one, if the difference is over I62 . If the difference is too great, it shows that too many lunations were added, or too few, and a correction must be made accordingly. A difference of over 11° shows that another eclipse, at that passage of the node, is pos- sible, but not probable, unless it is as much as 14°. 12. At the time of new moon, the longitudes of the sun and moon must be equal, and, according to our calculations, they are so at the time of the*mean new moon in May just found. But this is on the supposition, that their motions were uniform. To find whether or not their longitudes are truly equal, we shall proceed, in the fol- lowing chapters, to compute them, taking into account all the chief inequalities in their motions ; and, if they come out alike, the time of new moon is correctly found ; otherwise, we shall have to add or subtract such an amount of time, as, with the relative velocities of the sun and moon, at the time, will render them equal. As a preparatory step, it is necessary to know, more accurately, the the moon's anomaly, and the longitude of its node. The progressive motion of the moon's perigee, and the retro- grade motion of its nodes, being both caused by the sun's attrac- tion, are most rapid when the sun is in its perigee, and constantly grow slower and slower, till the sun reaches its apogee, where the motion becomes the slowest. Consequently, as the sun leaves its perigee, the moon's perigee immediately gets before, and its nodes behind their mean place, and continue so till the sun reaches its apogee, when, owing to the diminished rate of motion, their mean and true places again coincide. The contrary takes place when the sun is in the other half of its orbit. Hence it is apparent, that the moon's anomaly, being reckoned from its perigee, must be less than the mean when the sun's anomoly is less than 180°, and great- er when greater ; showing that something must be subtracted from the moon's anomaly in the former case, and added in the latter. The same must also be true of the longitude of the moon's nodes. These facts are indicated in Tables 6th and 7th, by the signs — and + placed at the head of the column containing the argument. By the term argument, is meant that quantity on which others depend, and which determines their value. Thus, in this case, the sun's anomaly determines what correction must be applied to the moon's anomaly, or to the longitude of its node, and is, therefore, the argument. 15 13. Entering tables 6th and 7th, with the sun's anomaly as an argument, we will take out the corrections which the foregoing considerations show to be necessary, and which are denominated Annual Equations of the moon's perigee and node, (taking care to make a proper allowance for the odd degree and decimals of the anomaly, as the tables give the equation only for every two de- grees,) and apply the former to the moon's anomaly, and the latter to the longitude of the node, according to the sign -f- or — at the head of the column of the argument. Observe in these, and most of the other tables, that the unit figure of the argument is placed at the top or bottom of the table, and the other figures at the right or left. When the latter is found at the left, we must look for the former at the top, but when the latter is at the right, the former must be sought for at the bottom. Opposite the latter, and in the same column with the former, the equation is found. Thus, in our eclipse, the sun's anomaly being 143°. 797, we look for the number 14 in the left hand column, and for 3 at the top. But, since the latter number is not found, the equation being given in the ta- ble only for 142° and 144°, we must note the difference between these equations, and take a proper proportion of it for the excess of the argument over 142°, viz. 1°.797. By this process, we find the annual equation of the perigee to be — 0°.221 ; and of the node — 0°.875. These, applied to the moon's anomaly, and the longitude of the node, make the former 145°.078, and the latter 61°.0008. CHAPTER III. EQUATION OP THE CENTRE. 14. The first inequality in the apparent motions of the sun and moon that claims attention, results from the eliptical form of their orbits, in consequence of which their motion is accelerated while passing from apogee to perigee, and retarded in the other half of their orbits, moving quickest in perigee, and slowest in apogee. The reason of this it is not difficult to discover. 16 Let AGH, &c. (Fig. 4) represent the moon's eliptical orbit, A and B the apsides, and E the earth. Let the moon start from perigee, at A, with its swiftest motion, and consequently with its greatest centrifugal force. The attraction of the earth, at E, not being able to retain it at that distance, G ' H it immediately begins to recede along the curve AGH. Being constantly pulled back by the earth's attrac- tion, since the angle EGH is obtuse, its motion is retarded, and when it approaches the apogee, at B, its velocity has become so much diminished, that the attractive force of the earth prevents it from receding further. It thus arrives at apogee with its slowest motion. Leaving B with a weak centrifugal force, the superior attractive power of the earth at E, immediately begins to draw the moon toward itself along the curve BOD, constantly hurrying it onward at every point, as O, the angle EOD, contained between the direction toward which it is drawn, and that toward which it moves being now acute. By the time it reaches A, its velocity becomes so much increased, that it is prepared again to leave peri- gee in the same condition as at first, to pursue another similar round. 15. We shall arrive at the same conclusion, if we apply the prin- ciple, that when a body is retained in its orbit by a force directed toward a fixed point, as the moon is toward the earth in this case, the radius vector must describe equal areas in equal times. Hence the moon must move slower when it is near B, than when it is near A, in about the same ratio that its distance from the earth is greater. Not only is the absolute velocity greatest in perigee, but the angular velocity, with which only we are now concerned, is rendered still greater, by reason of the diminished distance. The same reasoning will apply, in every respect, to the earth revolving in an eliptical orbit round the sun, and hence to the ap- parent motion of the sun round the earth. 16. If we consider the mean place of the sun and moon to be their true one, when in perigee, it will be, also, when in apogee ; because each half of the orbit is described in the same time ; but, as they start from perigee with their swiftest motion, they thus get ahead of their mean place, nor do they, though constantly retarded, 17 lose what they had gained, till the moment they arrive at apogee. Also, as they pass the apogee with their slowest motion, they di- rectly get behind their mean place, and it is not till the moment they reach the perigee, that the continued acceleration of their motion, in this half of their orbits, enables them to gain up what they had thus lost. Consequently, they must always be ahead of their mean place, when in the former part of their orbits, and behind . it in the latter. This difference between the mean and true place of the sun or moon, is termed the equation of its centre. 17. It is necessary to know in which half of their respective orbits the sun and moon are found at the time of our predicted eclipse : for, if either is moving from perigee to apogee, i. e., if its anomaly (4) is less than 180°, we must add something to its longi- tude already found, (11,) but subtract if in the other half of its orbit, i.e., if its anomaly is over 180°. The manner of computing the precise amount, will occupy our attention in another chapter. It is sufficient, here, to remark, that if, at any time, a line, OC, be drawn from the mean place of the moon to the centre of the elipse, the angle EOC, which this line makes with the radius vector, is very nearly equal to one-half the angular distance by which the moon is before or behind its mean place : so nearly that some au- thors have given this as a method of computing the equation of the centre. We shall use it hereafter as a convenient approxima- tion. For the present, we will dispense with the labour of compu- tation, and take the equation directly from tables already prepared. By the calculations in the last chapter, (11 and 13,) we found that the sun's anomaly was 143°.797, and the moon's, as corrected, 145°.078. It appears, then, that each is moving from perigee to apogee, and is, therefore, ahead of its mean place, so that we must add something to their respective longitudes ; with these anoma- lies, respectively, as arguments, we may now enter tables 8th and 9th, in the same manner as we did tables 6th and 7th, and take out the equation of the centre of the sun, and of the moon, applying the former to the the sun's longitude, and the latter to the moon's. The equations we find to be + 1M165 and +3°.4154, and the resulting longitudes 65°.3494 and 67°.6483. 2 18 CHAPTER IV. PERTURBATIONS IN THE MOON's MOTION. ANNUAL AND SECULAR EQUA- TIONS OF LONGITUDE. 18. In the preceding calculations, we first regarded the sun and .moon as revolving in circular orbits, with uniform angular velocity ; then, in elipses, describing equal areas by the radius vector, in equal times : but neither of these suppositions is strictly true. The sun's attraction disturbs the motion of the moon round the earth, producing numerous inequalities. The method of calculating these, will be treated of in another place. It will be sufficient for our purpose, here, simply to give the theory of them, and then take the corresponding corrections from the tables. 19. The most obvious effect of the sun's attraction, is to draw the moon away from the earth, and thus enlarge its orbit. If this influence were always the same, it would occasion no inequality : but when the sun is in perigee, it is nearer to the earth, and conse- quently to the moon, than at other times ; and the moon, therefore, will be more attracted by it. The moon being thus drawn farther away from the earth, when the sun is in this situation, and its peri- odic time consequently increased, it must fall behind its mean place. And although the attractive force of the sun diminishes as it leaves perigee, allowing the moon to contract its orbit and lessen its peri- odic time, it will not gain up what it had lost till the moment the sun reaches apogee. By similar reasoning, we may see that the moon must always be in advance of its mean place, so far as this cause is concerned, when the sun is in the other half of its orbit. There must then be applied to the moon's longitude a correction depending on the sun's anomaly ; .being additive when the anomaly is less than 180°, and subtractive when it is more. Such a correc- tion is supplied in table 10th, entering which, with the degrees of the sun's mean anomaly, in the manner described in article 13th, we find the equation to be -.1118, which is to be applied both to the moon's longitude and anomaly,* for the cause we have been considering obviously affects both alike. The same is true of most * The anomaly contains but three decimal places ; hence, in applying the corrections to it, the 3d figure is given according to its nearest value. d9 of the other corrections that remain to be applied. The longitude, as already obtained, (17,) is 67°.6483, and the anomaly (13) 145°.078. Applying trie above equation, they become 67°.5365, and 144°.9G6 respectively. 20. Connected with the foregoing inequality in the moon's mo- tion, there is another of great historical interest, from the theories to which it formerly gave rise, viz., the acceleration of the moon's mean motion. It is too small to be discovered by direct observa- tion, but becomes quite sensible in the lapse of ages. 21. Dr. Haliey, wishing to know the precise length of a lunation, went back to the ancient Chaldean observations, intending to ascer- tain how many new moons had occurred between that time and his own, and then to divide the time by this number, which would give the average length of each. But he was surprised to find that a lunation in those days was considerably longer than now. By comparing the Chaldean, Alexandrian, Arabian, and the pre- sent observations, he found that the lunar period grew successively shorter. 22. Astronomers doubted the fact when it was first announced ; but when they became satisfied of its truth, they set themselves to work to account for it. The most probable theory was, that the moon revolved in a resisting medium, which would cause it gradu- ally to fall toward the earth, and thus, by reducing the size of the orbit, make the periodic time less. It must seem paradoxical to those who have not thought upon the subject, that such a cause could produce the effect in question ; and that the retarding of the motion could make it revolve in less time. But it should be con- sidered, that by diminishing the moon's velocity, its centrifugal force is diminished in a more rapid ratio, which would allow the earth to draw it nearer to itself, and reduce the size of the orbit. And it is demonstrable, that the gain in time from the latter cir- cumstance, would more than counterbalance the loss from the for- mer ; so that on the whole, the moon's period would be shortened. The objection to this theory is, that comets, which are proved to be extremely light bodies, pass through this medium w^ith little or no resistance. Hence it was inferred that the cause, if it exist- 20 ed at all, was not sufficient to produce the effect of which we arc speaking. Other theories were advanced, but none were satisfactory ; and it was reserved for La Place to explain the true reason of this ac- celeration of the moon's period, about sixty years ago. 23. Owing to the attraction of the other planets, the earth's orbit is gradually becoming less and less eliptical, or, nearer and nearer to a circle ; so that the sun is every year about 39| miles nearer the centre of the elipse than it was on the year before. At this rate, the earth's orbit would become a circle in 40,315 years ; an event, however, that can never take place, for long before such a period shall elapse, the change of which we are speaking, and which is only an inequality of long period, will have reached its limit, when the eccentricity of the orbit would again increase. 24. If it can be shown that the sun's attraction diminishes the moon's gravity toward the earth, and thus increases the periodic time, more than it would do if the earth revolved in a circle at the same mean distance, it is manifest that so long as the change in the shape of the earth's orbit, of which we have just spoken, goes on, the moon's periodic time must grow less and less. Let ADBE (Fig. 5) represent an elipti- cal orbit, S the attracting body, placed in one of the foci, C the centre of the elipse, and F the other focus. It can be demon- strated, that the mean distance of S from all points in the orbit, is equal to AC or CB. Take any two points in the orbit G and H, equidistant from B and A. We propose to prove that the average attraction of S upon the moving body, when at these points, is greater than it is when the body is at its mean distance. And since these are any points in the arcs AD and DB, if we prove it for them, we prove it for the whole orbit. Join GS, GF and HS, and let SG bear any ratio, other than that of equality, to GF; say 6 : 4. Then, since by the properties of the elipse SG-f GF=AB=2CB, it follows that the ratio of SG to CB is 6 : 5, and of GF, or its equal HS, to CB, 4 : 5. Therefore, since the force of gravity is inversely proportioned to the square of the distance, the attraction at the former point will be §£, and at the 21 latter f- f of what it is at the mean distance. The average between them is fff of the attraction at the mean distance, — exceeding it Now the earth's orbit is much nearer circular than we have supposed this to be, and the excess of attraction must be propor- tionably less : but still there must be an excess, so long as it is eliptical at all. Hence, as the earth's orbit becomes nearer circular, the sun's attraction upon it, and consequently upon the moon, must continually grow less, allowing the orbit of the latter to contract. This would diminish the periodic time, and produce the very effect that excited so much wonder in the mind of Dr. Halley, and the astronomers of his time. 25. The tables for this work are based on the moon's motion, as it existed in the year 1800, and we must, therefore, add to its lon- gitude and anomaly the amount gained since that time, from the cause just explained. Table 11th contains the required correction, calculated at intervals of five years, during the present century. Look for the year in the left hand column of the table, except the unit figure, which is placed at the top, and opposite to the former, and under the latter will be found the correction required, express- ed in decimals of a degree, the first two places, which are ciphers, being omitted. The correction for the year 1854 is .0009, which, added to the longitude and anomaly already found, (19,) makes the former 67°.5374, and the latter 144°.967. CHAPTER V. PERTURBATIONS IN THE MOOn's MOTION, CONTINUED. VARIATION. 26. The moon's motions grow more complicated the farther we proceed. To investigate them thoroughly, is nothing less than a solution of the famed Problem of the Three Bodies. The moon's orbit, which we first regarded as a circle, and then an elipse, we shall now find to be neither a circle nor an elipse,* but an irregular * This statement seems to conflict with former ones, where the eliptical form of the moon's orbit was asserted; but its mean shape was then intended, without taking into account the irregularities. 22 oval shaped figure, which is constantly changing its form. The prospect before us, in trying to reduce such irregularities to order, so as to see their precise influence on the moon's longitude, is suffi- ciently appalling, but nevertheless, let us not be deterred from the attempt. To avoid misapprehension, it ought perhaps here to be remark- ed, that these irregularities are not of such a nature as to set aside our previous work, but only show that, under some circumstances, they may occasion necessary corrections. •27. And first let us, in this chapter, see what the shape of the orbit would be, and how the moon would revolve in it, on the sup- position that it was originally a circle round the earth, but drawn out of shape by the sun's attraction. We shall in this way disco- ver the cause of an observed inequality in the moon's motion, de- nominated variation, and discovered by Tycho Brahe, A. D. 1590. The modifications that the orbit would undergo, by supposing the original figure an elipse instead of a circle, will occupy our atten- tion in the next chapter. 28. Let S (Fig. 6) represent the sun, E the earth, and ADCB the moon's orbit ; and let us suppose, for a moment, that the moon, retains a circular orbit. Let D represent the place of the moon at conjunction, C at apposition, and A and B when it is at the same distance from the sun that the earth is, or very nearly in quadra- ture. First, let the moon be at A or B, in which case the moon and earth being equally distant from the sun, must be equally attracted by it, and consequently there would be no tendency to change their direction from each other, but only to draw them nearer to- gether, which would be precisely equivalent to increasing the earth's power of gravity. Next let it be in conjunction at D. Now the earth and moon are both in the same direction from the sun ; but the moon being nearest is more attracted, in the inverse ratio of the square of the distance, i. e., SE 2 : SD 2 - The only effect, therefore, is to draw the moon directly away from the earth, by virtue of the difference in the attractive forces, which would be equivalent to diminishing the earth's attraction. 29. Again, — suppose the moon at any point M in the quadrant AD. Being nearer to the sun than the earth is, it is more attract- ed by it, and the effect is nearly* the same as though the earth was not attracted at all, but the moon drawn along the line MS by a force equal to the difference of the attractions. The direction of this force, making an acute angle with that in which the moon moves, must accellerate the motion in its orbit; and the same would be true of every point in the quadrant AD. If the moon were at M' any point in the quadrant DB, the difference of attrac- tions acting along the line M'S, would tend to retard its motion. Once more : let the moon be at any point M" in the quadrant BC. Being further from the sun than the earth is, it is less attract- ed by it, which is nearly as though it were drawn in the opposite direction, along the line M"L. The effect would be to accelerate the motion, in nearly the same manner as in the quadrant AD. If the moon was at any point M'" in the quadrant CA, the difference of attractions acting, as it were, along the line M'"N, would retard its motion. Thus the moon is alternately accelerated and retarded in the differant quadrants ; moving swiftest in syzygy and slowest in quadrature. Hence, from this cause alone, the moon would be in advance of its mean place while passing from syzygy to quadrature, and behind it while passing from quadrature to syzygy. 80. But the above is not the only reason. We have thus far, in the present investigation, supposed the moon's orbit to retain its circular form, notwithstanding the disturbing influence of the sun : but this is not possible. To retain a body in a circular orbit, the centripetal and centrifugal forces must be equal. But we have just seen that the velocity in syzygy, as at C and D, (Fig. 7,) is * Sufficiently near for our present purpose. The subject will be investigated more critically hereafter. 24 greater than at A and B : and as the centrifugal force is propor- tioned to the square of the velocity, it must be greater. On the other hand, it was shown (28) that the sun's attraction diminished the moon's gravity toward the earth in syzygy, as at C and D, and in- creased it in quadrature, as at A and B. Taking both these facts into consideration, it is man. Fig. 7. ifest that at A and B, the centripetal force must con- siderably exceed the centri- fugal, while at C and D, the centrifugal will be the great- est, which would cause the moon's track to fall within the circle at the former points, as to L and N, and without it in the latter, as to F and G. The effect would be to Fig. 8. throw the orbit into some- thing such a shape as is represented in Fig. 8, viz: a kind of oval, with its longest diameter, AB, at right angles to line ES, drawn from the earth to the sun. Will this alteration in the the shape of the moon's orbit affect its longitude ? To aid us in this investigation, we will circumscribe the oval by a circle ; and to make the illustration more striking, we will suppose the oval very much flattened, so as nearly to coin- cide with AB, as in Fig. 9. Now, if the arc AF be divided in any given ratio at the point L, and LE be Fi £- 9 - drawn, it will cut the arc AD by no means in the same ratio. AM will bear a much greater ratio to MD than AL to LF. Hence, if two bodies, whose periodic times were equal, should start from A atf| the same time, and move with uniform velocity, one in the circle ALF, and the other in the oval AMD, the former would arrive at L before the latter would at M, leaving it behind perhaps at N. The same reasoning may be ap- plied to the other quadrants, though with the opposite effect in DB 25 and CA, when it will show that the moon must be in advance of its mean place. 31. There are two reasons, then, why the moon will be behind its mean place when passing from quadrature to syzygy, but in advance of it while passing from syzygy to quadrature ; 1st, from its unequal motion, (29,) and 2d, from the shape of its orbit. The maximum effect of the former to change the moon's place, is from 9' 17" to 10' 15", and the latter from 23' 56" to 20' 52", according to the distance of the earth from the sun. When at its mean dis- tance, the maximum effects are 9' 46" for the former, and 25' 24" for the latter, amounting to 35' 10" for both united. * 32. If the four quadrants were perfectly symmetrical, a table showing the correction required for each degree in one quadrant, would answer for all the rest ; only the equation would be additive when the moon is passing from syzygy to quadrature, i. e., in the arcs DB or CA, and subtractive when it is passing from quad- rature to syzygy, i. e., in the arcs AD and BC. But there is a slight difference ; for, 1st, the disturbing influence is a little less in the half of the orbit nearest the sun than in the other half, the dif- ference of the squares of the distance of the earth and moon from the sun, being a trifle less ; and 2d, the quadrants (so termed for the sake of conciseness) nearest the sun contain a little less than 90° each, and the other two quadrants, each a little more than 90°, for AB is not strictly a straight line, but an arc of the earth's or- bit. A table for two quadrants would, however, be sufficient — one in the half of the orbit next to the sun, and the other in the half most remote from it, as, for example, DB and BC. Table 12th is constructed in this way, where it will be seen that the equations are additive for a little less than 90° after the moon leaves D, and then subtractive to the end of the next quadrant. If the moon's angular distance from the sun exceeds 180°, which would carry it into the quadrants CA or AD, the degrees are found at the right hand and bottom of the table, and direction is given to " reverse the signs," so that the equations which were additive in DB become subtractive in AD, and those which were subtractive in BC become additive in CA. • 26 33. The angular distance of the moon from the sun, (found by- subtracting the longitude of the latter from that of the former, as thus far corrected, borrowing 360° if necessary,) shows in which quadrant the moon is. When the difference is from 0° to 90°, or from 180° to 270°, the moon is passing from syzygy to quadrature, but when it is from 90° to 180°, or from 270° to 360°, the moon is passing from quadrature to syzygy. In the present case, the lon- gitude of the sun (17) is 64°.3494, that of the moon (25) 67°.5374, and the excess of the latter 2°. 1880. Entering table 12th with this argument, in the same manner as directed in article 13th, the equation is found to be +.0445. This is to be applied to the moon's longitude and anomaly according to its sign. If the argument had been over 180°, the sign of the equation would have to be changed to — . After this equation is applied, the moon's longitude becomes 67°.5819, and the anomaly 145°.012. 34. The inequality to which this chapter is devoted, being occa- sioned by the disturbing influence of the sun, must be more or less according as the distance of that luminary varies, as we have al- ready observed, (31.) In table 12th, and, consequently, in the equation that was just applied, the sun is supposed to be at its mean distance. Hence another correction becomes necessary, which must evidently depend on the same circumstances as the last, to- gether with another, viz., the distance of the sun from the earth, which is determined by its anomaly. Accordingly in table 13th two arguments are employed ; viz., 1st, the argument just used for variation, which is to be sought for at the top or bottom of the table, and 2d, the sun's anomaly at the right or left. If the former is found at the top, we look for the latter at the left; but if at the bottom, at the right. The equation is found opposite the latter, and in the same column with the former. Since one argument is given only for every 5° and the other for 10°, it is necessary to institute a kind of double proportion for the units and decimals. It is further to be noticed, that if both arguments are to be found in the same gnomon, enclosed by the heavy lines about the table, the equation is to be applied with its proper sign, as found in the table ; but if one is found in the inner and the other in the outer gnomon, the sign before the equation is to be changed from + to — , or 27 from — to +. In the present case, the former argument (33) is 2°.1880, which being between 0° and 5°, is to be considered as found in the inner gnomon at the top; and the latter (11) is 143°.797, which is found in the outer gnomon, at the left. Making a proper allowance for the units and decimals, the equation is +.0053 ; but the arguments being found, one in the inner and the other in the outer gnomon, the sign must be changed, and the equa- tion becomes — .0053. This applied to the moon's longitude and anomaly, found in the last article, makes the former C7°.57G6, and the latter 145°.007. CHAPTER VI. PERTURBATIONS IN THE MOON's MOTION, CONTINUED. EVECTION. 35. The inequality which is to occupy our attention in this chap- ter was discovered by Ptolemy, A. D. 110, and is denominated Evection. For distinctness of conception, it is necessary to bear in mind the precise difference between this correction and that treated of in the last chapter, for there is danger of confounding them, since both are caused by the disturbing force of the sun in the plane of the ecliptic. That supposed the original form of the moon's orbit a circle, this an elipse, and wholly dependent on its eccentricity ; so that if the elipse had no eccentricity, there would be no correc- tion for evection. That always elongated the orbit in the direction of the quadratures ; this, we shall see, elongates it in the direction of the syzygyes. That regarded the shape of the orbit as constant ; this, as ever changing. An important element in this correction is the irregular motion of the line of apsides ; that had no such line to take into account. 36. It will be shown, that the progressive linjjpf the moon's ap- sides is quite irregular ; that it sometimes progreSes more and some- * times less rapidly ; sometimes remains stationary, and sometimes even goes backward. Now in determining the moon's mean ano- maly, (11,) all the motions were supposed uniform, and no correc- tion has been made for any irregularity in the motion of the moon's perigee, except that which resulted from the unequal distance of the sun, (13.) But since the anomaly is reckoned from the peri- gee, it must be subject to all the irregularities that the perigee itself is. Hence, in applying the equation of the centre, (17,) we used data that were erroneous, and the error that was introduced needs to be corrected. 27. But this is not all. The greater the eccentricity of an orbit is, the greater is the equation of the centre. Thus the equation of the moon's centre is much greater than that of the sun with the same anomaly, (compare tables 8th and 9th.) because the orbit of the former is much the most eccentric. Now it will be shown presently, that the eccentricity of the moon's orbit is ever varying, and the equation of the centre, which depends upon it, must vary likewise ; whereas table 9th is computed on the supposition of a constant mean eccentricity. So that we not only made use of a wrong anomaly in applying the equation of the centre, but also a wrong eccentricity in the moon's orbit. The correction for the combined effect of these two errors constitutes evection. 38. It will be demonstrated in its proper place, that if a revolving body be retained in an eliptical orbit, by a force directed toward one of the foci, the square of the distance of the body from that focus, at any point in its orbit, must always be inversely proportioned to the intensity of the attractive force at that point. Hence, if an in- increase or diminution of attraction were to take place throughout the orbit, proportional to the existing attractions at each point, the size, but not the form of the orbit would be changed. The ec- centricity would remain the same as before. But if the alteration in the attractive force were in any other ratio, it obviously would affect the shape of the orbit. If the perigeal gravity was made to bear too great a ratio to the apogeal, it would be drawn in too much at the former point, or too little at the latter, and the orbit would become rrqgfc eccentric. Or if the apogeal gravity became 29 too great in proportion to the perigeal, the orbit would be rendered less eccentric, or more nearly circular. 39. To apply this to our subject, let E (Fig. 10) represent the earth, ADBC the moon's orbit, A being the perigee and B the apo- gee, and FGHI the sun's apparent orbit. First, let the sun be at S, so that the line of apsides, AB, of the moon's orbit is directed to- Fg. 10. wards it, or lies in syzygy. The moon is more attracted by the earth at A than it is at B, in the inverse ratio of AE 2 to EB* ; and in order that the disturbing influence of the sun, which tends (28) to diminish the earth's attraction at these points, should effect no change in the shape of the moon's orbit, it must also (38) be more at A than at B, in the same ratio. But instead of that, the sun's disturbing influence is greater at B than tit A, for the difference between SE 2 and SB 2 is greater than between SA 2 and SE 2 ; con- sequently the relative difference in the attractive forces at A and B toward E is increased, and the orbit must become more eccen- 30 trie. In the same manner it may be shown that the eccentricity of the moon's orbit must be increased when the sun is at S". But if the sun were at S', and the moon at A or B, the latter would be drawn toward the earth by the sun's disturbing influence, and its gravity increased, (28 ;) but more at B than A in the ratio of EB to E A, as will appear if we resolve the force in the direction S'A into two others in the directions AE and ES', and that in the direction S'B into two in the directions BE and ES'. In this case the greatest addition is made to the least force, whereas to preserve the shape of the orbit unchanged, the additional gravities should be in proportion to the previously existing ones. The apogeai gravity thus becomes too great in proportion to the perigeal, and the eccentricity of the orbit is diminished, (38.) We shall arrive at the same conclusion if the sun be supposed to be at S'". Thus the eccentricity of the moon's orbit is greatest when the line of its ipsides lies in syzygy, and least when it lies in quadrature. It is plain that these changes in eccentricity occur, not instantane- ously, but gradually, as the sun progresses in its orbit. The ec- centricity must diminish while the sun is passing from S to S', or from S" to S'", and increase while it is passing from S' to S", or from S'" to S, being at its mean state when the sun is about half way between these points, as at L, M, N and K. The eccentricity must exceed the mean when it is in the quadrants KL, or MN, and be less than the mean when it is in the quadrants LM and NK. 40. The investigation of the irregular motion of the line of the apsides of the moon's orbit, on which the evection in part depends, is considerably more difficult than any of the preceding, and will require the reader's close attention. That it must, on the whole, progress, will appear, when we consider that the average effect of the sun's attraction is to draw the moon away from the earth, and thus to render its orbit less curved than it would otherwise be. Consequently, after the moon leaves its perigee, or apogee, where its motion is at right angles with the radius vector, its angular mo- tion round the earth must amount to more than 180° before its path will have been deflected enough to intersect the radius vector at right angles again. That is, it must be more than 180° from 31 • perigee to apogee, or from apogee to perigee.* And, further ; since the attraction of the sun sometimes increases, and sometimes diminishes the moon's gravity toward the earth, we should con- clude the line of its apsides must sometimes regress and sometimes progress. We must, however, go into a more minute investiga- tion of this motion, to account for all the phenomena to which it gives rise.f 41. If the moon, or any other body revolving in an eliptical orbit, should be deflected from its natural course at any point by some dis- turbing influence, so as to move at right angles to the radius vector, the point where such deflection occurred would thenceforward become one of the ' apsides of the orbit, provided it were not further dis- turbed. This is evident, from the fact that the apsides are the only points in an eliptical orbit, where the curve is at right angles to the radi- us vector. Also the body must still revolve in an elipse, or some other conic section, for we shall demonstrate hereafter that simple gravitation toward a fixed point can retain it in no other curve. Whether the point of deflection will be the perigee or the apogee, will depend on the velocity of the time ; if it be greater than the mean, the point will be the perigee, but if less, the apogee. 42. To apply this principle, let us inquire what alteration must be made in the attractive force of the earth, E, (Fig. 11.) to bring the motion of the moon at right angles to the radius vector at the points of its orbit C, D, F, and G, the two former being near the perigee, A, where the velocity exceeds the mean, and the two lat- ter near apogee, B, where it is less. While the moon is moving from A, through S to B, the direction of its motion constantly makes an obtuse angle with the radius vector, (as EFM ;) it would be * Playfair. t The articles between this and the 56th may be omitted, if the instructer should deem it expedient. 32 Fig. 11. necessary, therefore, at the point F, that the attractive force of E should be in- creased, to curve its mo- tion to K, and thus bring it at right angles with EF. And, if it were so increas- ed, the point F would (44) henceforth become the ap- ogee, instead of B. In other words, the apogee would have moved back- ward from B to F ; and consequently the perigee from A to T, for they must always be opposite each other. The same reasoning will apply to the point D ; yet, if the deflect- ing force should occur there, D would become the perigee instead of the apogee, on account of the moon's greater velocity at that point, (44,) and the apogee would be found in the direction of the line DO ; so that the apsides would have moved forward from A to D, and from B to O. In the other half of the moon's orbit, where the direction of the motion continually makes an acute angle (as EGR) with the radius vector EG or EC, it is manifest, that the attraction of E must be diminished, in order to bring the mo- tion at right angles, as GL and CH ; or, rather, a repulsive force must be given to it. Such a deflection occurring at G, would change the place of the apogee from B to G, or, would make it move forward. If occurring at C, it would change the place of perigee from A to C, or, would make it move backward. 43. If instead of increasing the gravity at F, it were diminished, it is pretty clear that the apogee would move forward instead of backward. To illustrate it, let us suppose the gravity to be great- ly diminished, so much so as to be nearly destroyed. The moon, being scarcely attracted at all toward E, will fly off nearly in a tangent to the elipse at F, and the apogee will be found in that direction, but infinitely distant. If then, we draw NP parallel to the tangent FM, it will point to the place of the apogee, which has, therefore, moved forward, equivalent to the arc from B to V. If the attraction were less diminished, it would not move forward so far, but the reasoning would hold good. 33 Similar reasoning applied to the points D, C, and G, will show that, by reversing the supposed alteration in the force of gravity at those points, we shall reverse also the motion of the line of the apsides. Summing up our conclusions, we find that an increase of the moon's natural gravity toward the earth, near apogee, on either side, would cause the line of apsides to regress ; while a diminution would cause it to progress ; and that the reverse takes place by an alteration in the natural force of gravity when the moon is near its perigee. I employ the terms natural gravity, and natural velocity, to signify the gravity and velocity that the moon would have if it revolved regularly in its eliptic orbit, undisturbed by the attraction of any foreign body. 44. It is evident that an increase in the moon's velocity, and consequently of its centrifugal force, must have nearly the same effect as a diminution of the earth's attraction ; and vice versa. Hence, if its velocity near apogee is, from any cause, rendered greater than its natural velocity in that part of its orbit, the line of apsides must move forward, or progress ; and the reverse, if such an increase of velocity occurs near perigee. If both these causes conspire, (and we will proceed to show that they do,) the progress or regress of the apsides must be still more rapid. 45. It has been shown, (28,) that the sun's disturbing influence increased the moon's gravity toward the earth in quadrature, but diminished it in syzygy ; and we should suppose that there must be intervening points, where it exerted no influence either way. These points it is important for us to find, for when the moon is at these, the apsides of its orbit must be at rest, so far as their mo- tion is caused by a variation in gravity. 46. Let the moon be at any point M (Fig. 12) of its orbit, and let the sun's attraction on it at that point be represented by m. Resolving this force into two others in the directions ME and ES, the proportion for the former (called the addititious force, because it increases the moon's gravity toward the earth) will read SM : ME ::m: the addititious force =gM m= slfcfME m - 3 47. The proportion for the latter force, in the direction ES, will read SM : ES : : ?n : the force required =g^-?w. But the earth is at- tracted by the sun in the same direction, ES, and it is the differ- ence of the attractions only that exerts any disturbing influence on the moon in this direction. We will therefore find how much the attraction of the sun on the earth is, and subtract it from that just found, viz., — m. By the laws of gravity ES 2 : SM 2 : : m : the sun's attraction at the distance ES. Hence the earth is attracted with a force equal to ^m, which is to be sub- tracted from ^77i. Re- ducing the fractions to a common denominator, and subtracting, we have Now SM= very nearly; ES 3 — SM 3 -m. ES3+SM ES— EF, therefore, by involving both sides, and rejecting the 3d and 4th terms in the right hand member on account of their small- ness, we have SM 3 = ES 3 — 3ES 2 xEF. Sub- stituting this value in the place of SM 3 ' the above fraction, which expresses the disturbing influence of the sun in the direction ES, becomes *™l$m m = fgm. Resolving this force into two others, one in the direction EM, and the other at right angtes to it ; i. e., in the directions EG and SG, the proportion for the former (called the ablatitious force, because it diminishes the moon's gravity toward the earth) will read SE : EG, or (since the triangles EGS and EFM are similar) ME : EF : : ^m : the ablatitious force=^~^m. 48. The addititious and ablatitious forces acting in direct oppo- sition, must neutralize each other at the points in the orbit where they are equal, showing that, at such points, the sun's attraction produces no effect on the gravity of the moon toward the e«th. In that part of the orbit that lies between these points and quadra- ture,.there wilL.be an increase of gravity, and between those points and syzygy, at Cor D, (for the demonstration will apply to. either half of the orbit ACB or ADB,) a diminution. 49. But since the fractions representing these forces have a com- mon denominator, they will evidently be equal when their numera- tors are equal, i. e., when 3EP=ME 2 ; or (extracting the square root) when v/3xEF=ME; or (converting the equation into a proportion) when ME : EF : : V3 : 1. But ME : EF : : 1 : cos. MES ; therefore, by equality of ratios, v/3 : 1 : : 1 : cos. MES= .5773672, which is the cosine of 54° 43' 56". Hence the gravity of the moon toward the earth is diminished when it is within 54° 43' 56" of syzygy, and increased when it is within 35° 16' 4" -of quadrature. 50. It is worthy of notice here, that the diminution of gravity in syzygy is about double the increase in quadrature. The above reasoning shows, that the ablatitious force is to the addititious as 3EF 2 : ME 2 - But in syzygy EF=ME, and 3EF=3ME 2 ; so that the difference between them is 2ME 2 ; while, in quadrature, EF becomes 0, and the ablatitious force disappears, leaving the additi- tious force proportional to ME 2, which is half of 2ME 2 51. There remains, not yet investigated, the tangential force in the direction GS, or MH, one of the parts into which we resolved (47) that in the direction ES. Its precise amount it is not now ma- terial for us to know : but it is to be observed, that its only influ- ence is to retard the moon's motion from D to B ; since, being at right angles to EM, it neither increases nor diminishes the moon's gravity toward the earth. If the moon was supposed to be in any of the other quadrants, and figures constructed on the same princi- ple as this, we should see that the moon must be retarded in pass- 36 ing from D to B, and from C to A ; but accelerated from A to D, and from B to C. Hence its motion must be swiftest in syzygy at C and D, slowest in quadrature at A and B, and a mean half way between syzygy and quadrature. This is the same conclusion to which we arrived by a less rigid process, in article 29th. 5% To show how the motion of the line of apsides is affected by this perturbation in the moon's gravity and velocity, let us recur again to Fig. 10th. Let the sun be at S, or S", so that the line of apsides, AB, lies in syzygy. In this case, both the moon's velocity in its orbit will be increased, and its gravity towards the earth di- minished (48) at A and B. Consequently, (43 and 44,) the line of apsides must move forward when the moon is near apogee, at B ; but backward, when it is near perigee, at A ; and if the regress near perigee is equal to the progress near apogee, they will balance each other, so as, on the whole, to produce no change in the posi- tion of the line of apsides. But they are not equal, for several rea- sons. 1st. The diminution of the moon's gravity at these points, and the increase of velocity, are both caused by the force in the direc- tion ES, (Fig. 12,) which was obtained in article 47, by taking the the difference of the attractions of the sun upon the earth and moon, in that direction. Consequently, they must depend on the difference between the distances of the moon and earth from the sun ; and this difference is greater when the moon is in apogee, than when it is in perigee. 2d. The forces causing the line of apsides to progress, act for a longer time than those causing it to regress, because the moon is longer in describing the apogeal than the perigeal half of its orbit. 3d. A given force would produce more effect on the moon when it is in apogee than when it is in perigee, on account of its natural velocity being less at the former point, and therefore more easily deflected from its orbit. If a cannon ball were moving but one foot in a second, it would not be very difficult to turn it out of its course, but not so if it were moving 1000 feet per second. From all these circumstances combined, the line of apsides pro- gresses quite rapidly when the sun is at S, or S". And, since the same causes operate in the same way, one (51) while the sun is passing through the arcs KL and MN, extending 45° each way from S and S", the direction of the line of apsides of the moon's 57 orbit, and the other (49) through FG and HI, extending 54° 43' 56" from the same points, it follows that the line of apsides must pro- gress, at least, while the sun is in the quadrants KL and MN. 53. When the sun is in either of the small arcs FK, LG, HM or NI, containing 9° 43' 56" each, the line of apsides is nearly station- ary : for the perturbations, both in gravity and velocity, being near their limits are very weak, and what small force they do ex- ert is in opposition to each other, the former tending to make the apsides progress, and the latter, regress. 54. Now let us suppose the sun at S' or S'", so that the line of apsides shall be at right angles to ES', or shall be in quadrature. The moon's gravity toward the earth, when at A and B, will now be increased, (48,) and its velocity diminished, (51.) Consequently, (43 and 44,) the line of apsides must progress when the moon is near perigee, but regress when it is near apogee. And, by nearly the same reasoning as employed above, (52,) it may be shown, that the regress exceeds the progress ; so that, on the whole, the line of apsides regresses. In like manner, it may be shown, that it re- gresses, though less rapidly, when the sun is any where in the arcs IF or GH. 55. The regress here will, however, be less rapid than the pro- gress when the sun is in the arcs KL and MN, for the perturba- tion in the moon's gravity is (50) but half as great. It will also be of shorter duration, for the arcs IF and GH contain but 70° 32' 8" each, and the sun moving forward in its orbit about 1° per day, while the line of apsides moves backward, on an average, about 1-6 of a degree per day, each arc will be passed over in about 61 days; when 1 he line will become nearly stationary for about 10 days, (the time occupied in passing one of the small arcs, as FK,) and then begin to advance.* * It may seem to the reader erronious, to ascribe any part of the progress of the moon's apsides to the perturbation in velocity ; for, since that is equal in quadrature and syzygy, and extends the same distance, 45°, from each, it would seem that, so far as this cause is concerned, the regress, when the apsides are near quadrature, should be just equal to the progress when they are near syzygy. Sir Isaac Newton took this view of the subject, and was greatly perplexed at finding that he could account for but about half the motion of the line of the apsides. The explanation here given, is, in substance, that of Clairaut, who showed that, when the apsides regressed, they approached to meet the sun, thus shortening the regressive arc, and, consequently, diminishing the perturbation in velocity; but, when they progressed, they receded from the sun, lengthening the progressive arc, and thereby increasing the perturbation in velocity. So that the perturbation would not only be greater in the latter case than in the former, but extend through a greater arc. 38 Since the line of apsides remains nearly stationary about 40 days in the year, moves backward about 122, and forward during the remainder, amounting to about 203 days ; and since its forward motion is more rapid than its backward, it is evident that, in the 'course of a year, it must, on the whole, advance. The rate of ad- vance is found by observation, to be such as to carry it entirely round the orbit in 3232 days, IS hours, |P minutes, and 29.4 se- conds, or about nine years. The foregoing explanations have, I trust, made the theory of the variation in the eccentricity of the moon's orbit, and the irregular motion of the line of its apsides tolerably clear to the reader. It remains to make a practical application of it. 56. We have seen (39, 52 and 54) that the eccentricity of the moon's orbit exceeds the mean, when the sun is in those quadrants where it causes the line of the moon's apsides to advance, and is less than the mean when the sun is in the other parts of its orbit : also, that these changes occurred alternately, and nearly in alter- nate quadrants, the lines of division being not far from half way between syzygy and quadrature. We will eneavour to represent these changes by a figure, the construction of which was devised by Sir Isaac Newton for the purpose, and which observation shows to be very nearly correct. Let AaBb (Fig. 13) represent the moon's eliptical orbit, in its mean state, M the moon, E the earth, placed in one of the foci, and EC the mean eccentricity. Now, as the eccentricity varies, the centre C will sometimes approach toward E, as far as K, and sometimes recede from it to I ; the distances CK and CI being the greatest variation of the eccentricity from the mean. Describe the circle IDKH, and join MC and ME. Let ES represent the direction of the sun, and B the moon's mean perigee, i.e., the place of the perigee if it progressed uniformly. Now, since it has been shown (39) that when ES is at right angles to AB, the eccentricity is least, and when it coincides with it, the greatest, it is plain that it must be represented by EK in the former case, and EI in the latter. And when ES is in any other position, the eccentricity must be represented by a line longer than EK, and shorter than EI. We shall effect this for every possible position of ES, by al- ways making the angular distance of H from I, in the direction IDKH, equal to twice the angle BES, The length of EH will be 39 equal to the eccentricity at the time, very nearly. For example, when BES=45°, H will coincide with D, making EH very little longer than EC, and thus showing that the eccentricity exceeds the mean in the same small ratio. Again, if BES=90°, H would coincide with K, showing that the eccentricity was now a mini- mum, which agrees with what we have already seen to be true. In the same manner it may be shown, that at any other point, this construction brings out very nearly the result it should do accord- ing to our previous reasoning. Fig. 13. Not only will the length of EH represent the eccentricity of the moon's orbit at all times, but its position will represent that of the line of apsides very nearly, never varying from it more than 3'. Hence, the point H may always be considered as the centre of the eclipse ; thus changing the whole eclipse from the mean position AaBb, to that represented by the dotted curve WXYZ. 57. It was remarked, (17.) that if two lines were drawn from the mean place of the moon, one to the centre of the elipse, and the other to the focus, round which it revolved, the angle at the moon, 40 , contained by these lines, would be equal to half the equation of the centre, very nearly. If the eccentricity and position of the elipse remained unchanged, the angle in question would be EMC ; but when, in consequence of the change, H becomes the centre of the elipse, the angle becomes EMH. Therefore HMC, which is the difference between these two angles, must be half the effect of the disturbing force of the sun ; or, in other words, it is half the evec- tion we have been so long in quest of. Hence, if we can find out a method of determining the size of this angle, or the conditions on which its size depends, our task is over, for by doubling it we shall have the correction required. 58. Draw Hs at right angles to MC. Then, since small angles are nearly proportional to their sines, the line H? must always be nearly proportional to the angle HMS, and consequently to the evection. But Hs is also the sine of HCs, or its supplement HCR ; therefore the evection is always proportional to the sine of HCR. This angle we will proceed to find. The angle SEB=MEB— MES ; therefore the angular distance of H from I, in the direction IDK, (being, by construction, double of SEB,) =2MEB— 2MES. The angles MEB and MCB are nearly equal,* the eccentricity of the orbit being small ; therefore, subtracting MCB, or its equal ICR, from the first member of our equation, and MEB from the last, we have HCR=MEB— 2MES. Now MEB is the moon's mean anomaly, and MES is the angular distance between the sun and moon, or the excess of the mean longitude of the moon over the true longitude of the sun. Hence the evection, which has been shown to be proportional to the sine of HCR, is proportional to the sine of the moon's mean anomaly diminished by twice the excess of its mean longitude over the true longitude of the sun.f * There is danger that the proportions of the different lines, as they appear in the figure, may mislead the reader, and it is well to remember, that EC is but about 1-20, and KC about 1-100 of CB. t In this demonstration, the moon's longitude is supposed to exceed that of the sun. But we shall arrive at the same conclusion if we suppose the sun's longitude the greatest ; as, for example, if it be in the direction ES'. For, now, S'EB=MEB-f MES' ; and, consequently, by the same construction and reasoning as in the other case, HCR=MEB -f-2MES\ But the result is obviously the same, whether we add the angle MES', or subtract its supplement ; that is, the angular distance of M from S', reckoned in the other direction, S'AaBM. Now, this supplementary distance is the excess of the moon's lon- gitude over that of the sun, borrowing 360°, or one revolution ; therefore, the angle HCR is still equal to the moon's mean anomaly, diminished by twice the excess of its mean longitude over the true longitude of the sun ; and the principle becomes general in its application. 41 If the reader will now turn to table 14th, he will notice that this is the argument by which the evection is taken out in that table. 59. At the time of our predicted eclipse, the quantities are as follows : — The moon's mean anomaly is, (13,) - - ■ 145°.0780* The moon's mean longitude is, (11,) 64°.2329 Subtract sun's corrected longitude, (17,) 65 .3494 358 .8835x2=357 .7070 Argument of evection, 147.3110 60. Entering table 14th with this number as an argument, in the same manner as heretofore, the required correction is found to be .7156, which the sign — , placed at the head of the left hand col- umn, in which the argument is in this case found, shows to be subtractive. It is plain also, from the figure, that it should be sub- tractive; for, in adding the equation of the centre, (17,) we added twice the whole angle EMC, which was too much by twice the angle HMC. We now correct the error, by subtracting the equa- tion just found from the longitude and anomaly previously obtained, (34,) which leaves for the former 66°.8610, and for the latter, 144°.291. 61. All the remarks that were made in article 34 on the subject of variation, will apply also to evection, since both are caused by the sun's disturbing influence. The method of taking the requisite correction from the table (table 15) is also the same, only that in this case, the argument for evection, viz., the moon's mean anomaly diminished by twice the excess of the moon's longitude over that of the sun, is to be sought for at the top or bottom of the table, in- stead of the argument for variation. The correction, as found in the table, is — .0100, but since one argument is found in the inner gnomon, at the bottom, and the other in the outer one, at the right, the sign is to be changed, (34,) and the correction becomes -f .0100, which, added to the longitude and anomaly last found, makes them respectively 67°.8710, and 144°.301. * This is the moon's mean anomaly, corrected by the annual equation of its perigee, which it is proper to do, because that inequality affects only the average progressive mo- tion of the perigee at different seasons of the year, and is in no way connected with that which we are now considering, or any other which has reference to the position of the moon in its orbit 42 CHAPTER VIL NODAL EQUATION OF THE MOON's LONGITUDE, AND REDUCTION TO THE ECLIPTIC. 62. In the numerous corrections that we have had occasion to apply to the moon's longitude and anomaly, growing out of the dis- turbing influence of the sun, the orbits of both have been supposed to lie in the same plane ; or the latter to lie in the plane of the or- bit of the former. The first of these suppositions is never true, and the latter only twice in a year ; viz., when the sun passes the moon's nodes. At all other times, it is either on one side of the plane of the moon's orbit or the other. Now it is evident that the sun's disturbing influence, in the various ways we have been speak- ing of, must be less than if it lay in the plane of the moon's orbit ; for* in order to make our reasoning good, its attraction must be resolved into two forces, one lying in the plane of the orbit, and the other at right angles to it, which necessarily creates a loss- of force. If it were always at a fixed mean distance from the plane, a proper allowance might be made in computing the inequalities, and the work would thus be accurate without further correction. In fact, the quantities in the tables which we have been using, were calculated on that supposition. But since the distance is variable, additional corrections are necessary for all that we have applied in the three preceding chapters. We will select, as an example, the annual equation of the moon's longitude, discussed in chapter 4th, remarking, as we pass, that if we were to attempt to apply all the corrections resulting from causes like that under consideration, and from the effect of one correction in altering the argument from which others had been obtained, our task would be endless. The business is, at best, only a series of approximations. 63. When the sun is passing one of the moon's nodes, being in the plane of the orbit, its attractive force exceeds the mean, so far as the circumstance now under consideration is concerned, dilating the moon's orbit (19) and increasing the periodic time more than usual. The moon must therefore fall behind its mean place, and continue to do so more and more, till the sun reaches its point of mean distance, about 45° from the node. As the sun continues to recede from the plane of the moon's orbit, its disturbing influence 43 must grow less* allowing the moon to contract its orbit and shorten its periodic time, till finally, when the former is 90°* from the node, the latter will have gained up what it had lost, so that its mean and true place will again coincide. The reverse of all this will take place when the sun is in the next quadrant. Its disturbing influence being a minimum at the outset, the moon must get ahead of its mean place ; and -it wtfl not lose what it thus gains till the sun reaches the next node. Hence, if the sun's longitude exceeds that of one of the moon's nodes by- less than 90°, something must be subtracted from the longitude and anomaly, as already obtained ; but added, if the excess is greater than 90°. Or, reckoning from the ascending node, a subtractive equation must be applied in the 1st and 3d quadrants, and an addi- tive one in the 2d and 4th. Such an equation is termed the Nodal Equation of the moon's longitude.* 64. To find how far the sun is from the node, the longitude of the latter (13) must be subtracted from that of the former, (17.) Entering table 16th with the argument thus found, viz., 4°.3486, in the same manner as we did table 6th and others, we find the equa- tion to be .0026, which the sign — at the head of the column con- taining the argument, as well as our previous reasoning, shows must be subtractive. The resulting longitude of the moon becomes 66°.8684, and the anomaly 144°.298. , 65. The moon's anomaly is now altered considerably, by reason of the various equations that have been applied to it, from what it was when we used it to take out the equation of the centre, in arti- cle 17th ; and since this equation is a very important one, our work will be more accurate if we now take it out again, and by what- ever amount it differs from what it was as first taken out, correct the moon's longitude. The anomaly, as used in article 17th, was 145°.078, which gave as an equation +3°.4154, while now it is but 144°.298, which gives for the equation -f-3°.4834, so that we did not add enough to the moon's longitade by 0°.0680. Adding this now, we have 66°.0364, which may be regarded as the true longitude of the moon, reckoned on its orbit, or, as it is usually termed, the true Orbit Longitude. * I find no name for this equation in any treatise on astronomy that I have met with, and have given it one that seems to be indicative of its character. 44 66. The plane of the lunar orbit being inclined to that of the ecliptic, causes longitudes reckoned on it to be different from what they would be if reckoned on the ecliptic. And since the longitudes of the heavenly bodies are referred to the latter, the orbit longitude just found needs one more correction to reduce it to the ecliptic. The argument, found by subtracting the longitude of the moon's node from that of the moon itself, is 5°. 9356, and the corresponding equation, obtained from table 17th, is — .0233. This applied, leaves 66°.9131 for the moon's true longitude from the mean vernal equinox. CHAPTER VIII. LUNAR, OR MENSTRUAL EQUATION OF THE SUn's LONGITUDE AND NUTA- TION. 67. It was observed in article 2d, that any motion or change of motion in the earth, produced apparently a precisely similar one in the sun. Now, the earth, like the moon, revolves round the common centre of gravity of the two, and is, therefore, subject to inequalities in this motion, the same in kind as those we have been considering in that of the moon, though far less in degree, owing to the earth's greater weight, and consequently close proximity to the centre of gravity. These inequalities, small in themselves, are rendered vastly smaller in their effect upon the sun's apparent mo- tion, by reason of the great distance of the latter. Fig. 14. Let S (Fig. 14) represent the sun, ABF the earth, E its centre, M the moon, and C the common centre of gravity between the earth and moon, about which both revolve. The distance from E 45 to C is not far from 2970 miles, or about three-fourths of the earth's radius. It is manifest that the longitude of the sun, as seen from E, will differ from its longitude as seen from C, by the angle CSE. When the angle MES is either 0° or 180°, the angle CSE will disappear, and when it is of any other size, the latter angle can be calculated ; for, in the triangle CES, the two sides, CS and CE, and the angle CES are known. We assume here, that E revolves in a circle round C, keeping CE of uniform length. It is plain from the dia- gram, that if the longitude of the moon exceeds that of the sun, the latter will be increased by the angle CSE ; but the contrary, if the longitude of the sun is greatest. In other words, if the longi- tude of the moon, diminished by that of the sun, is less than 180°, the equation will be additive, but if greater, subtractive. » 68. The longitude of the sun, as found in chapter 3d, is 65°.3494, and that of the moon, as finally corrected, (67,) 66°.9131. The excess of the latter above the former is 1°.5637. Entering table 18th with this number, in the usual way, we find that, in the pres- ent case, the correction is inappreciable, unless we extend our de- cimals further ; so that the longitude obtained in chapter 3d is to be considered correct. 69. The error occasioned by regarding the orbit of the earth's centre as a circle instead x>f an elipse, as it in fact is, might be cor- rected by introducing an equation corresponding to the equation of the centre of the sun or moon. But it would never be necessary, unless extreme accuracy were required ; for the whole correction which we just undertook to apply, when a maximum, is but the decimals of a degree, .0021 ; and since the eccentricity of the or- bit amounts to hardly more than 1-20 of the radius, the error can never be more than about .0001, which is less than half a second. Much less then must the inequalities in the motion be appreciable. 46 CHAPTER IX. ■ NUTATION IN LONGITUDE. .70. The equinoxes are not stationary, but move slowly, west- ward, which necessarily affects the longitudes of all the heavenly bodies, since they are reckoned from the vernal 'equinox. If the rate of motion were uniform, the longitudes of the sun, moon, and moon's nodes, which we have obtained in the preceding chapters, would nevertheless be correct ; for the tables from which we ob- tained the mean longitudes in article 11th, are based upon the sup- position of a uniform rate of precession of the equinoxes, and allow- ance is consequently made for it. But it is not uniform, and we are now to look into the causes of the inequality, and make the requisite correction in the longitudes on account of it. 71. If. a body were to revolve round the earth, in an orbit not coinciding with the ecliptic, so that it would be sometimes. north and sometimes south of the plane of the latter, we can see that whenever it were thus situated, the sun's attraction must tend to draw it back into the aforesaid plane. To illustrate by a diagram, let SD (Fig. 1 5) rep- resent the plane of the ecliptic, and MM' that of the revolving body, both seen edgewise,* S the sun, and E the earth. Fig. 15. When the body is at M the sun's attration on it, in the di- rection SM, may be resolved into t^o other forces, in the directions SC and CM, the latter of which tends to draw the body directly into the plane SD. In like manner, when the body is at M', it is drawn toward the plane SD, by a force represented by M'D ; and so of any other point out of the plane of the ecliptic. The consequence is, that the« body, as it revolves round its orbit, which we will suppose it to do in an easterly direction, is drawn into the plane of the ecliptic, and made to cross it sooner, that is, farther westward, every succeeding revolution.* The same would be true of any number of bodies similarly situated, so that our reasoning will be good, even if they were multiplied to such a * The reader will perceive, that the condition of our supposed body corresponds, in every respect, with that of the moon revolving round the earth, and hence will see the cause of the retrograde motion of the moon's nodes. 47 degree as to form a continuous ring entirely round the orbit. Nor will it alter the principle, if we suppose the bodies, or ring, very near to the earth, or even attached to it, only that, in < the latter case, they would communicate their * motion to the earth, and so could not move without dragging the earth with them, which would greatly deaden any motion, or .change of motion they would otherwise have. Now, from the spheroidal form of the earth, there is a protruber- ant mass of matter about eighteen miles thick girdling its equator, every particle of which is situated precisely as we supposed our imaginary bodies or ring to be. The effect we have described must therefore follow, and every place — as Quito, for example — must every day, by the diurnal motion of the earth, cross the eclip- tic at a point farther west than on the day previous. Or,, to illus- trate farther ; suppose the plane of the ecliptic to be a vast sheet of some material substance, with an orifice of sufficient size to ad- mit the earth, and to allow it to revolve freely on its axis. Now, if a man were stationed on Mount Chimborazo, (which we will suppose to be on the equator, though it is not precisely so,) and every time the earth rolled round, so as to carry him under the plane, which would be every twelve hours, should mark on the edge the place under which he passed it, these marks would be continually farther and farther west by about fifteen feet. 72. The attraction of the moon also conspires with that of the sun in causing a precession of the equinoxes ; for the plane of its orbit being much more nearly coincident with the ecliptic than that of the equator is, it may be regarded as another body lying in the plane of the ecliptic, and conspiring with the sun in its influence upon the earth. 73. It is evident from Fig. 15th, that the greater the inclination of the planes SD and MM', the greater must be the forces repre- sented by CM and DM', and consequently the more rapid must be the retrograde motion of the points of intersection. So far, there- fore, as the moon's influence is concerned, the greater the obliquity of its orbit to the equator, the more rapid must be the precession of the equinoxes. 48 Let CD (Fig. 16) represent a portion of the ecliptic, seen edgewise, V the vernal equi- nox, and EF and GH portions of the moon's orbit, making an ^_ angle of about 5° with the eclip- tic ; GH representing it when the ascending node is at V, and EF when the ascending node is there. Also, let AB represent a portion of the equator, making an angle of about 23*° with the ecliptic. Then HVB, the inclina- tion of the moon's orbit to the equator, when the ascending node is at V, equals about 28|° ; and FVB, the inclination when the as- cending node is there, equals about 18£°. 74. When, therefore, the ascending node, in its retrograde course, passes the vernal equinox, which it does once in about nineteen years, the rate of precession must considerably exceed the mean, and the equinoxes must immediately get too far west, which would increase the longitude af all the heavenly bodies. The same would be true all the while that the node was slowly working its way backward round to the autumnal equinox ; for though the rate of precession would continually diminish, and become a mean when the node was 90° back, or west of the vernal equinox, yet it would take the whole of the next quadrant for it to lose what it had gain- ed in the first. Thus, when the ascending node gets round to the autumnal equinox, which would bring the descending node to the vernal, all the longitudes would become right, or in their mean state again. But the rate of precession is now a minimum, and on principles similar to those we have been discussing, it is apparent, that, while the node is passing through the other half of its orbit, the longitude of all the heavenly bodies must be less than the mean. Thus, for a period of about nine and a half years, all longitudes are greater than the mean, and then, for the same period, less ; and so on, alternately. This is called Lunar Nutation in Longitude. 75. It is plain, that when the ascending node is passing from the vernal back to the autumnal equinox, its longitude must exceed 180°, and be less than 180° when it is in the other half of its orbit; so that we can know by the longitude of the node, whether to add to or subtract from our mean longitudes. 49 76. At the time for which we are calculating, the mean longi- tude of the ascending node (11) is 61°0883 ; and entering table 19th with this argument, we find the amount to be subtracted from the longitudes of all the heavenly bodies at that time, is .0042, which will leave us for the moon's longitude (G6) 6G°.9089 ; for the sun's (18) G5°.3452, and for the moon's node (13) 60°.9966. 77. There is another inequality in the rate of the Precession of the Equinoxes, called Solar Nutation, and occasioned by the vari- able distance of the sun from the plane of the equator in the course of a year. But it is so small (never amounting to much over one second) that it may be disregarded without material error. CHAPTER X. TRUE TIME LONGITUDES AXD ANOMALIES. 78. By comparing the true longitudes of the sun and moon, found in the last chapter, we find that the latter is greatest by 1°.5637, which shows that the moon has passed by the sun, and that the eclipse is over. It remains (12) for us to subtract such an amount from the mean time of new moon (11) as it must have taken the moon to gain this difference, and in order to do so, we must know the relative velocities of the sun and moon in their orbits at the time. Their motions are swiftest in perigee, and grow slower as they recede from it ; hence their anomalies are the proper arguments for determining their motions. We may therefore enter tables 20th and 21st, with the anomalies of the sun and moon respectively as arguments, and take out their hourly motions. The former we find to be .0401, and the latter .5018. All the other inequalities treated of in the preceding chapters, must likewise affect the moon's hourly motion, of which Variation and Evection are the most important. The effect of Variation, as is plain from the theory, is to increase the moon's velocity in syzy- gy and diminish it in quadrature. Now, in an eclipse, the moon is always in syzygy, and hence we must add to its hourly motion the quantity given in the margin of table 21st. To correct the 4 50 moon's hourly motion for Evection, we must enter table 22d with the same argument that was used for that inequality in article GOth, and in the middle column we find the equation, which is to be ap- plied to the hourly motion according to its sign. The following is the operation in the case before us : — Moon's hourly motion, by table 21st, - - .5018 Add for Variation, .0115 .5133 Subtract for Evection, by table 22d, - - .0092 .5041 Subtract sun's hourly motion, - .0401 Hourly gain of the moon upon the sun, - - .4640 Now, by simple proportion, we can find how long it must have taken the moon to gain the difference in the longitudes of the sun and moon, viz. 1°.5637. Thus, .4640 : 1 hour : : 1°.5637 : the time required, which is thus found to be 3 hours, 22 minutes, and 12 seconds. This subtracted from the time of mean new moon, found in article 11th, leaves for the true time of new moon in May, 26d. 8h. 48m. 44sec. 79. The time thus found is Greenwich time, and to reduce it to that of any other place, allowance must be made for the difference of longitude, viz., 4 minutes of time for each degree of longitude. It is also to be observed, that the astronomical day begins at noon, and counts the 24 hours round to the next noon. 80. The longitudes and anomalies of the sun and moon must now be corrected, by subtracting their motions during the correc- tion just applied to the time. If that correction had been additive, this would be so also. The amount can be easily found from their hourly motions, thus — One hour : the moon's hourly motion, viz., .5041 : : 3h. 22m. 12 sec. : the correction required in its longitude and anomaly, which is thus found to be 1°.6988. One hour : the sun's hourly motion, viz., .0401 : : 3h. 22m. 12 sec. : the correction required in its longitude and anomaly, which is thus found to be 0M351. 51 At the same time the longitude of the node must be corrected, by taking from table 4th its motion during the same time, and ap- plying it, with the contrary sign from that of the other motions, because it moves in the opposite direction. 81. The reader will get a clearer idea of the process of calcula- ting the time of an eclipse, if we now give a synopsis of the work that we have been through in the foregoing chapters. EXAMPLE. Showing the method of calculating the time of a Solar Eclipse. Time. Sun's Anom-» aly. Sun's Longi- tude. Moon't Anom- aly. Moon's Longi- tude. Longi- tude of Node. Mean new moon,'March, 1854 d. h. m. s. 28 10 42 50 59 1 28 6 85.586 58.211 6.0194 58.2135 93.665 51.634 6.0194 58.2135 64.2158 -3.1275 Mean new moon in May, An. equa'n moon's per. & node 26 12 10 56 143.797 64.2329 145.299 —.221 64.2329 61.0883 —.0875 + 1.1165 +3.4154 Equation of the centre, 145.078 61.0008 —.0042 An'l equation of moon's long. 65.3494 145.078 —.112 67.6483 —.1118 Secular equa. of moon's long., 144.966 +.001 67.53615 +.0009 Variation, v 144.967 +.045 67.5374 + .0445 Annual equation of variation, . 145.012 —.005 67.5819 —.0053 Evection, 145.007 —.716 67.5766 —.7156 Annual equation of evection, . 144.291 +.010 66.8610 +.0100 Nodal equation of the ) 144.301 —.003 66.8710 —.0026 moon's longitude, $ Corrrection of the equa- ) 144.298 66.8684 +.0680 tion of the centre, £ * * Reduction to the ecliptic, 66.9364 —.0233 —.0042 —1.699 Lunar nutation, 66.9131 —.0042 3 22 12 —.135 Correct for difference in ) long, of sun & moon, $ 65.3452 —.1351 66.9089 —1.6988 60.9966 +.0074 True time, long's & anomalies, 26 8 48 44 143.662 65.2101jl42.599 65.2101 61.0040 52 82. It can hardly fail to suggest itself to the attentive reader, that so great an alteration in the time, as that which was made in article 78th, must in some degree vitiate the result of our work. The anomalies, and nearly all the arguments for the inequalities would vary in the interim. It would hence seem desirable to have obtained, if possible, some nearer approximation to the true time at the outset. The " preliminary equations"* in tables 27 and 28 are designed for this purpose, but the theory of them could not be well explained at that stage of our work, where it was necessary to in- troduce them, if at all. The construction of these tables is as follows. Since at the time of new or full moon, the argument for evection is the same as that for the equation of the moon's centre, viz., the moon's mean anom- aly, the separate effects of the two on the time are united in the first preliminary equation. Also, since both the equation of the sun's centre, and the annual equation of the moon's longitude, de- pend on the sun's anomaly, they are united in like manner in the second preliminary equation. After having found the time of mean new or full moon, as des- cribed in article 11th, we apply to it these equations, and then take from table 4th the mean motions in longitude and anomaly during the time so applied. These motions applied to the mean longitudes and anomalies at the mean time, give the mean longitudes and an- omalies at the corrected time ; and w r e then proceed to calculate the true longitudes at the corrected time, in the same manner as we have done for the mean time in the foregoing chapters. We will illustrate, by example, the method of using these tables, and at the same time show how to calculate a lunar eclipse. 83. It will not be necessary to give much more than a synopsis of the operation, as the time of a lunar eclipse is calculated precise- ly in the same manner as one of the sun, only that the half lunation in table 3d, is used in order to give the time of mean full moon, and the longitudes of the sun and moon, instead of being made to agree, are made to differ just 180°. Let us inquire whether there will be an eclipse of the moon when the sun passes its ascending node in the year 1844. Turning to table 2d, we find that the longitude of the ascending * These are the same as those usually termed 1st and 2d equations of the mean to ihe true syzygy. 53 node in March of that year is 258°, and we therefore (11) take from table 3d such a number of lunations as, added to the half lu- nation at the foot of the table, will contain 258 days, or thereabouts. In 8£ lunations there are 251 days, which is the nearest to 258 that we can get from the table. This will carry the time forward to November, and will give us for the longitude of the sun 244°, and of the node about 245°. The sun will, therefore, be but 1° from the node, and so far within the lunar ecliptic limit (7) that there cannot fail to be an eclipse of considerable size. We will proceed to calculate it. From tables 2d, 3d and 5th, we obtain the following : — Time. Sun's Anom- aly. Sun's Longi- tude. Moon's Anom- aly. Moon's Longi- tude. Longi- tude of Node. Mean new moon in March, 1844, Add eight lunations, d. h. m. s. 18 15 40 54 246 5 52 23 14 18 22 1 76.522 232.843 14.553 356.7831 232.8530 14 5534 132.366 206 535 192.908 356.7831 232.8530 194.5534 25°. 1245 —12.5102 — .7819 Mean full moon in November,... 24 15 55 18 323.9181 244.1895 171809 64.1895 244.8324 Entering tables 27 and 28, with the anomalies of the moon and sun respectively as arguments, and making the proper proportions, we obtian the " preliminary equations," the first of which is lh. 30m. lsec, subtractive, and the second 2h. 30m. J9sec, also sub- tractive. The sum of the two is 4h. 0m. 20sec, which must be subtracted from the mean time found above. Also the motions of the sun and moon during this time, both in longitude and anomaly, must be subtracted, and that of the node added, because it moves the other way. The quantities in table 4th, from which we obtain these motions, are given only for the units of the arguments, but will answer just as well for tens, by removing the decimal point one place to the right. Thus the sun's motion in anomaly, accord- ing to the table, is, for 2 days, 1°.9712, and for 20 days 19°.712. When used for units, the right hand figure may be omitted, and the next reckoned according to its nearest value. The following is the operation for finding the motions in the case before us : — Motion in 4 hours, Do. in 20 seconds, Total motion in 4h. 0m. 20sec, Sun's Anom- aly. 0°.164 0°.000 0°.164 Sun's Longi- tude. 0°.1643 0°.0002 0°.1645 MoonV Anom- aly. Moon's Longi- tude. 2°.177 2°.1961 0°.003i 0°.O030 2°.180,2°.1991 Longi- tude of Node. 0°.00P l 0°.000( ! ! 0°.008^ 54 84. After applying these corrections to the time, longitudes and anomalies, the process of calculation is, throughout, the same as for a solar eclipse, with the exception already mentioned* and it is unnecessary to go through it in detail. The following is a synop- sis of the calculation : — EXAMPLE. Showing the method of calculating the time of a Lunar Eclipse. Time. Sun's Anom- aly. Sun's Longi- tude. Moon's Anom- aly. Moen's Longi- tude. Longi- tude of Node. Mean new moon in March, 1844, d. h. m. s. 18 15 40 54 136 5 52 23 14 18 22 1 7°6.522 356.7831 132.366 356.7831 25°3.1245 14.553 14.5534 192 908 lOAjaan -12.5102 — .7819 244.8324 +.0088 64.1895 24 15 55 18 — 4 20 323.918 244.1895 — .164: —.1645 171.809 1st cor. in time, with cor'pond'g motions, —2.180—2.1991 Annual equation of moon's per. and node, 24 11 54 58 323.754 244.0250 169.629 + .217 61.9904 +1.0412 244.8412 +.0897 —1.1560 169.846 + .109 169.955 + .001 169.956 + .006 244.9309 +.0044 242.8690 63.0316 +.1087 63.1403 +.0006 63.1409 +.0058 169.902 + .003 169.965 — .193 63.1467 +.0034 63.1501 —.1930 169.772 + .007 62.9571 +.0068 169.779 + .001 169.780 62.9639 +.0013 62.9652 +.0063 62.9720 +.0077 + .0044 242.8734 — .0103 — .121 62.9797 +.0044 — 14 41 — ".010 Cor. for difference* in Ion. of sun & moon, 02.984 1 —.1210 244.9353 +.0004 7-010 time, longitudes and anomalies, 24 11 40 17 323.744 242.8631 169.659 62.8631 244.9357 * In lunar eclipses it is this difference, + or —180°. 55 CHAPTER XL ELEMENTS OF AN ECLIPSE. 85. The following elements or data, are all that are needed for making any calculation that we may desire in regard to an eclipse, either solar or lunar ; such as the place on the earth's surface, where the sun will be centrally eclipsed at any given time, while the eclipse lasts ; the portions of the earth where an eclipse will be visible, and the time when it will commence, become a maxi- mum, and terminate ; or the size of an eclipse, at any given place and time. The 10th is not needed in solar eclipses, nor the 2d, 3d, and 12th in lunar. 1st. The time of new or full moon. 2d. The longitudes of the sun and moon. 3d. The obliquity of the ecliptic to the equator. 4th. The moon's latitude. 5th. The sun's hourly motion. 6th. The moon's relative hourly motion, or the excess of its hourly motion over that of the sun. 7th. The sun's apparent semidiameter as seen from the earth. 8th. The moon's do. 9th. The apparent semidiameter of the earth as seen from the moon, which is the same as the moon's horizontal parallax. 10th. The apparent semidiameter of the earth's shadow where it eclipses the moon, as seen from the earth. 11th. The angle of the moon's visible path with the ecliptic. 12th. The sun's declination. 86. The method of obtaining the first two, was explained in the last chapter. 87. The mean obliquity of the ecliptic to the equator in the year 1840, was 24° 27' 62".52, but it decreases at the rate of about half a second a year, owing to the attraction of the planets. Il is also subject to an inequality, whose period is about 19 years, depending on the longitude of the moon's node : for it is evident from the theory of lunar nutation, that the moon's influence must affect the 56 obliquity of the equator to the ecliptic, as well as the place of their points of intersection. Table 23d gives the obliquity on the 1st of January in each year of the present century after 1840, taking both these causes into account, from which it can be readily found for any time in the year by inspection. In January, 1854, it is 23° 27' 33".6, and in January, 1855, it is 23° 27' 35".7. Hence, at the time of our solar eclipse in May, 1854, it is 23° 27' 34".5. 88. The moon's latitude depends on its distance from the node, and the inclination of the plane of its orbit ; but the inclination, and also the place of the node, varies according to the situation of the sun in respect to the node. Hence there are two principal equa- tions of the moon's latitude, one depending on the distance of the moon from the node, and the other on that of the sun. Now in eclipses, the distance of both luminaries from the node is the same, so that the two equations may be combined into one, which is done in table 24th. The argument is found by subtracting the longitude of the node from that of the sun and moon, which leaves, in our solar eclipse, 4°.2061, and in the lunar, 177°.9274. The moon's latitude, as determined by these arguments, is, in the former case, the decimals of a degree .3664, or a little over one-third of a de- gree, and in the latter, .1810. It is to be noticed, that in table 24th the figures of the argument at the right and left are whole degrees, and those at the top and bottom the first place of decimals. It must be readily seen, that the moon's latitude must be north for the first 180° after it leaves the ascending node ; and that it moves northerly, or ascends, through the first quadrant, and south- erly, or descends, through the second : also, that in the other half of the orbit, its latitude must be south, being descending in the first quadrant, and ascending in the second. These facts are indicated in the table by capital letters at the head of the columns containing the argument. 89. The method of finding the 5th and 6th elements was explain- ed in the last chapter, (78 :) but if much accuracy were required, they w r ould have to be now computed over again for our solar eclipse, because the anomalies have been changed. Calculated from the anomalies as finally corrected, in the same manner as was done in article 78, the hourly motion of the sun in our solar 57 eclipse we find to be .0401, and in the lunar .0421 ; and the rela- tive hourly motion of the moon is .4649 in the solar eclipse, and .4523 in the lunar. 90. The 7th element is obtained from the 1st column of table 26th, where it is sufficiently explained. 91. Table 21st, columns 1st and 3d, give the 8th and 9th ele- ments, so far as they depend on the elliptical form of the moon's orbit. But the effect of Variation is to throw the orbit into a kind of oval, with its shortest diameter lying in syzygy. From this cause, the distance of the moon from the earth is less when it is new or full than at other times, which must increase their apparent size as viewed from each other. Hence the corrections in the margin of table 21st. Also Evection, by altering the shape of the moon's orbit, affects its apparent size and parallax, so that a far- ther correction becomes necessary from table 22d. The following shows the method of obtaining these elements for the eclipse of May, 1854 :— Semidiarneter. Parallax. Values taken from table 21st, .2482 -f .0020 .9097 + .0073 "Variation, Evection, .2502 — .0023 .9170 — .0080 True semidiameter and parallax, .2479 .9090 The semidiameter and parallax at the time of the lunar eclipse in November, 1844, obtained in the same way, are .2453 and .8989 respectively. 92. It was shown in article 4th, that our 10th element, viz., the apparent semidiameter of the section of the earth's shadow that eclipses the moon, is equal to the sum of the parallaxes of the sun and moon, diminished by the sun's apparent semidiameter. The sun's parallax may always be put down at .0024, and the methods of obtaining the other data for finding this element have been al- ready explained. Thus in the lunar eclipse which we have taken as an example, 58 The sun's parallax is .0024 The moon's do., as just found, is - - - - .8989 .9013 The sun's semidiameter (see 7th element) is - - .2707 The semidiameter of earth's shadow is - - - .6306 93. The angle which the moon's path makes with the eclip- tic varies according to the moon's distance from the node. When at the node, it makes the same angle as the planes of the two or- bits; but when it is 90° from it, its motion becomes parallel to the ecliptic. But the inclination of the planes also varies, depending, as was remarked above, on the distance of the sun from the node. The two influences may, however, be combined into one at the time of an eclipse, in the same manner as in table 24th. And not only does the real angle vary from both these causes, but the ap- parent angle is increased by the earth's motion in the same direc- tion ; and since the rate of the latter, as compared with the moon's motion, is quite variable, the apparent angle must vary also from this cause. All these causes are taken into account in table 25th. This table has two arguments, viz., 1st, the difference between the hourly motions of the sun and moon, (for the motion of the earth is measured by the apparent motion of the sun,) the first two decimal places of which are placed at the top ; and 2d, the distance of the sun or moon from the node, which is placed at the right and left. In the solar eclipse we are calculating, the former (89) is .4649, and the latter (found by subtracting the longitude of the node from that of the moon) 4°.2061. These give the angle 5° 44' 33", ascending. In the same way the angle at the time of our lunar eclipse is found to be 5° 45' 41" descending. 94. The sun's declination, which is our 12th element, can easily be computed from its longitude by spherical trigonometry, since the obliquity of the ecliptic is known, (87 ;) for its longitude, right ascension, and declination form a right-angled spherical triangle, of which an angle and one side is known. Table 26th gives the declination calculated from the obliquity in the year 1840, which is sufficiently exact for our purpose, though, after a lapse of years, it must evidently need correction. Entering this table, with the sun's longitude at the time of our solar eclipse as an argument, we obtain the declination 21°.1871. 59 Since the sun starts northerly from the vernal equinox, its decli- nation must be north for the first 180°, and south through the rest of the orbit. This fact is indicated by the capital letters at the head of the columns of the argument. 95. The elements collected are as follows : — 1. True time of the eclipse, 2. Longitude of the sun and moon,.... 3. Obliquity of ecliptic to equator, 4. Moon's latitude, 5. Sun's hourly motion, 6. Moon's relative do 7. Sun's apparent semidiameter, 8. Moon's do 9. Moon's horizontal parallax,, 10. Semidiameter of earth's shadow, .... 1 1. Angle of moon's visible path with eclip. 12. Sun's declination, Solar Eclipse. d. k. m. 8. May 26 8 48 44 , 65°.2101 23° 27' 34".5 0°.3664 (north) 0°.0401 0°.4649 0°.2635 0°.2479 0°.9090 5° 44' 33" (ascend.) 2l°.187l (north) Lunar Eclipse. d. h. m. s. N"ov. 24 11 40 17 0°.1810 (north) ()°.0421 0°.4523 0°.2707 Q°.2453 3°.8989 (J°.6306 5° 45' 4i" (descend.) CHAPTER XII. DELINEATION OF A SOLAR. ECLIPSE. 9G. To find whether a solar eclipse will be visible at a particu- lar place, and if so, its size and general appearance there, it is more convenient to first reverse the order of viewing the phenomenon, and to suppose the spectator placed at the centre of the sun, to look down upon the earth, and see the moon passing across its disc. From so vast a distance, the earth wofla 1 appear to him, as the sun does to us, like a flat circular disc. The circle of illumination would be to our observer the circle of the disc, and all circles whose planes were perpendicular to this would be seen edgewise, and would appear to him like straight lines. Their arcs would seem to be only of the length of the straight lines that they would sub- tend, as viewed by him. Such circles as were seen obliquely, would appear elliptical in their shape. Let us suppose him to take 60 his station when the sun is in the vernal equinox, about the 21st of March, and to retain it for a year, accompanying the sun in its ap- parent annual round. Being always in the plane of the ecliptic, it would appear to him like a straight line dividing the earth into two equal parts, one half lying north and the other south of it. At first also, being in the plane of the equator, it too would be seen as a straight line, as well as all the parallels of latitude ; yet not paral- lel with the ecliptic. The west end of the equator would be north of the ecliptic and the east end south, crossing it in the centre at an angle of about 23^°, as in Fig. 17, where AB represents the eclip- tic, CD the equator, PP' the earth's axis, P and P' its poles, GH the axis of the ecliptic, and 1 1, 2 2, 3 3, &c, parallels of latitude. As the sun advances, it gets north of the plane of the equator and of the parallels of latitude, and they will no longer appear as straight lines, but will seem bent downward toward the south. The earth's axis will approach to parallelism with that of the eclip- tic ; and the poles revolving in circles whose planes are parallel to that of the ecliptic, will seem to move in straight lines toward n and s till on the 20th of June, or thereabouts, when the sun reaches the summer solstice, the two axes will coincide, and the north pole will be seen at n. The south pole will be invisible, being hid be- hind a segment of the earth ; but if the earth were transparent it would appear at s. 97. The sun still advancing, the earth's axis will appear again on the other side of GH, the poles will approach toward N and S, and the parallels of latitude will become less curved. And when the sun reaches the auturmtal equinox, in September, the latter will again become straight lines, but lying the opposite way from what they did in March, and the poles will appear at N and S. 08. As soon as the sun has passed the autumnal equinox, the parallels of latitude will appear curved again, but upward toward the north, instead of downward, because the sun will now be on the south side of their planes. The poles will recede along the 61, iines NP and SP', passing n and s when the sun is at the winter solstice, in December, and finally arriving at P and P' about the '21st of March, when everything assumes the same aspect»as when he started. 99. A common terrestrial globe will serve to present these vari- ous appearances to the reader's view, in a much clearer light than can be done by any verbal description. Let the north pole be elevated about 662 above the point marked " North," on the wood- en horizon, and then the latter will represent the ecliptic. If now the globe be placed upon a table in the centre of the room, so that the wooden horizon may be on a level with the eye, and the read- er, after having found the 21st of March on the horizon, should retire across the room in that direction, he will see the globe pre- cisely as represented in Fig. 17. Let him now pass slowly round the room in the order of the months on the horizon, and all the ap- pearances we have described will be presented to his view. 100. When he is in the direction marked May 26, he will have a true representation of the earth, as it would appear to our obser- ver at the sun,, at the time of the solar eclipse we have been calcu- lating. We will endeavor to represent the same by a figure, and also the appearance of the moon passing over the earth's disc. In order to give proper proportions to the several parts of our figure, we must be able to mark down their relative dimensions. In com- mon plans and drawings these are given in miles, feet, inches, or some other direct measure of length ; but in astronomy it is found more convenient to determine them by the angle which they would subtend when viewed from a given distance, as we did in the last chapter. And this answers the purpose just as well, provided the distance be sufficiently great, for the apparent would be very near- ly proportional to the real size of the object. The dimensions which respect the moon and earth, given in the last chapter, are the angles that the objects would subtend at a distance of about 237,000 miles, or the distance between the earth and moon. The reader must not here fall into the error of supposing that our obser- ver has changed his position. He is still at the sun, and these an- gles are given merely for the purpose of determining the relative sizes of the objects that we wish to draw. 62 *%; *; fj, t) 89 ?t> u so so 4-0 SO GO 70 80 so ic a 3 ^yrft T ri^fi^.l'j^iTufir'T tirrrl^TTl iTr j'i^^rr nn]i j Yrlrn- = T -:=Sin. Pi, f p X COS. P *S±Pl=P2, sin. mxcos. P2=sin. P3, tan. ???xsin. p2=tan. P4, fsin. (P±P4)xcos. P3=sin. a?=the latitude, cot. s' cos. m tan. P2 =tan. Ps, Pe, cos. m sin. P3 cos. x sin. #xtan. P7=Ps, jp 5 — P 6 +^±Ps=y=the longitude. The chief difficulty in applying these equations consists in know- ing which of the four possible values to give to P, Pi, P2, &c. The following statements will remove all doubt. P, Pi, P3 and P4 are each always less than 90°. Ps is always of the same affection as s' increased by 90°. P6 is always of the same affection as P2. P7 is less or greater than 90°, according as P is less or greater than the complement of P4; it never exceeds 180°. Ps is always of the same affection as P7. 122. It is impossible, by a mere description, to convey to the reader a clear idea of the reason of the several steps of this pro- cess ; but if he will take his globe, and adjust it in the same manner as he would do to find the position of the centre of the eclipse by the previous mechanical process, he may be able to discover suc- * If after the new moon -j- ; if before it — . t The sign before P4 is + if P2 is less than 180° ; but — if it is greater. And in the latter case if P4 is greater than P, the latitude of the place will be opposite in character to that of the moon ; i. e. if the moon's latitude is north that of the place will be south, and the contrary. X The sign before Ps is -f- if Ps is between 0° and 90°, or between 270° and 360° ; but — if P2 is between 90° and 270°. 78 cessively the arcs and angles expressed byP, Pi, &c. ; and hence to understand the method by which they are obtained. P is the distance of the centre of -the eclipse from the ecliptic, measured on a secondary to it drawn upon the earth's surface, as m 10, (Fig. 19.) Or, it is the latitude of that point in the heavens, on which an observer, placed at the centre of the earth, would see the centre of the shadow at the earth's surface projected, if the earth were transparent. Pi is an arc of the ecliptic, intercepted between the aforesaid secondary and the point where the sun is vertical. Or, it is the difference between the sun's longitude and that of the aforesaid point in the heavens. Pa is the same arc increased by the sun's longitude. Or, it is the longitude of the aforesaid point in the heavens. P3 is an arc of &. great circle, drawn from the north pole of the equator, perpendicular to the aforesaid secondary. P4 is the arc of the secondary, intercepted between this perpen- dicular and the north pole of the ecliptic. Ps is the right ascension of that point in the equator where it is cut by a secondary to the ecliptic passing through the centre of the shadow on the earth's surface, at the time that, it leaves the earth ; or, it is the right ascension of that point in the heavens on which the centre of the shadow would be seen projected at that time, by an observer at the centre of the earth. P6 is the right ascension of that point in the equator, where it is cut by the first mentioned secondary. P7 is the angle at the centre of the shadow, or at the first men- tioned point in the heavens, contained between secondaries to the ecliptic and equator passing through that point. Ps is the arc of the equator intercepted between these two secon- daries. 133. To apply the process to a particular case, let it be required to find the place where the solar eclipse which we have taken as an example, will be central at 20 minutes and 51 seconds past 10, 79 by Greenwich time. After the preparatory steps described in articles 120 and 121, we have the following data and results : — DATA. RESULTS. Time=10£. 20m. 51sec. P = 28° 58' 22" s'= 65° 16' 50" Pi = 66 29 27 5=65° 16' 34" P 2 =131 46 1 /= 0°.4394 P 3 = 15 22 40 d= 0°.7278 P 4 = 17 56 12 Z=65° 12' 45" a;= 44 45 28=the latitude. p= 0°.9072 Ps=153 21 5 ?rc=23° 27' 34".5 Pg=129 19 34 Pt= 21 55 45 P 8 == 15 49 38 y= 73 24 38— the longitude Thus we find that the centre of the shadow, at the time just mentioned, is in lat. 44° 45' 28", and Ion. 73° 24' 38", which is on the west shore of Lake Champlain, about four miles north of the village of Piattsburg. J24. In bringing to a close our examination of the solar eclipse of 1854, it may not be uninteresting to give a general description of it. The centre of the eclipse first strikes the earth at 54 minutes and 27 seconds past 6 P. M., (Greenwich time,) in the Pacific ocean, not far from the Caroline Islands, and travelling northeast- wardly, nearly over the Sandwich Islands, strikes the American coast a little north of Astoria, in Oregon Territory, about 24 min- utes past 9. Crossing the Oregon Territory, it enters the British possessions, and turning easterly, and then southeasterly, re-enters the United States territories west of Lake Superior. At 13 min- utes past 10 it crosses the outlet of Lake Superior, about 100 miles N. W. from Michilimackinack. After reaching the settled parts of Canada, it passes a little south of Bytown, travelling at a rate of 70 miles per minute, and reaches the St. Lawrence about 20 miles below St. Regis, at 20 minutes past 10. Again entering the United States, on the north line of New York, it arrives at the west shore of Lake Champlain, about four miles north of Pittsburgh, at 20 minutes and 51 seconds past 10. It strikes the opposite shore in 80 the south part of the town of Georgia, and from thence passes through the following towns in Vermont, viz., Fairfax, Fletcher, Cambridge, Stirling, Morriston, Elmore, Woodbury, Cabot, Dan- ville, Barnet, Waterford, and reaches the Connecticut river at 21 minutes and 45 seconds past 10. Travelling now about 100 miles per minute, it passes through the towns of Littleton and Bethlehem, in New Hampshire, and from thence directly over the Notch in the White Mountains, and through Adams and Chatham into Maine. After passing through Fryeburg, Denmark, Bridgetown, Sebago, and across the pond into Windham, Gray, Cumberland, and Yarmouth, it strikes the Atlantic in the latter town, about ten miles, in a direct line, from Portland. It leaves the earth at 28 minutes and 55 seconds past 10, in lat. 32° 36' 6", Ion. 49° 44' 12", which is in the Atlantic ocean, about 800 miles east of Bermuda.* Since the apparent size of the moon at the time of the eclipse is less than that of the sun, (95,) the eclipse cannot be total at any place ; but along the line we have described, the visible part of the sun will appear as a very slender bright ring, encircling the moon. This appearance will extend about fifty miles on each side, taking in Burlington, Middlebury, Dartmouth, Bowdoin, and Waterville colleges ; the ring will appear of uniform width only along the central line. Such eclipses as this are called annular. CHAPTER XIV. DELINEATION OF A LUNAR ECLIPSE. 125. The delineation of a lunar eclipse is extremely simple, since it consists merely in representing the passage of the moon across the earth's shadow. To show the method of effecting it, we will proceed to delineate the lunar eclipse of November, 1844, from the elements obtained in chapter 11th. The angular dhnen- * By taking into account several minute circumstances that we have disregarded, the central track of the eclipse may vary slightly from this description, probably passing ten or fifteen miles further north, and nearly over Bowdoin college. 81 sions given in the 4th, 6th, 8th and 10th elements being supposed to be all taken at the same distance, viz., the distance from the earth to the moon, will serve as a measure for their absolute dimen- sions, in the same manner as they did in the solar eclipse. The first step is to make the larger and smaller scales SS and ss (Fig. 21) just as was done in the solar eclipse, (101 and 109.) Take from the longer one the semidiameter of the earth's shadow, viz., 0°.6306, and with it describe the graduated circle BDAE, which will represent the shadow. It is evident that the plane of the ecliptic must bisect the shadow, apd we therefore draw the two diameters AB and DE at right angles to each other, the former to represent a section of the plane of the ecliptic, and the latter its axis. 120. The moon's latitude is 0°.1810 north. We therefore take this distance from the scale SS, and measure it upward from C toward D, which gives M as the place of the moon's centre at the time of full moon. If the latitude were south, the centre would be found in the line CE. 127. Its path makes an angle of 5° 45' 41" with the ecliptic, tending south. If, therefore, we draw CF, making an angle of that size with CB, and YMZ parallel to it, the latter line will represent the track of the moon's centre across the shadow. It passes M at 40 minutes and 14 seconds past 11 in the evening, by Greenwich time, anp! its position at any other hour and minute may be found by graduating the line YZ, as directed in article 109th. 128. By taking the moon's semidiameter, 0°.2453, from the scale SS, and with it describing a circle from any point in its path as a centre, the position of the entire disc will be shown, as it must exist at the time indicated at its centre on the path. It is drawn in the plate in five different positions : 1st, when it begins to impinge on the shadow at G, which must be the commencement of the eclipse : 2d, when it just falls wholly within the shadow at H, at which time the eclipse must begin to be total : 3d, when its centre is at N, found by drawing CN perpendicular to YZ, and thus (Euc. 3, 2) bisecting the chord, which must be the middle of the eclipse : 4th, when it begins to leave the shadow at L, at which time it 6 82 must cease to be total, and 5th, when it entirely leaves the shadow at P, which must be the end of the eclipse. The respective cen- tres are at R, S, N, T and V, and the time may be determined very nearly by the drawing. 129. More accurate results may, however, be obtained by cal- culating the length of MR, MS, MN, MT and MY, and then find- ing by the relative hourly motion of the moon, how long it must take it to pass over them. In the right angled triapgle CNM, the side CM and the angle MCN are known, being our 4th and lltfr elements, and we can find CN and MN. The sirfe CN is common to the two right angled triangles SNC and RNC, and the sides CS and CR are also known, the former being the difference, and the lattjr the sum of our 8th and 10th elements. Jlpnce NS anc( NR can be found, and likewise their equals NT and NV. Now, by adding and subtracting MN, which is known, we shall have the lines required. 130. The times obtained by this process are as follows :— ~. Commencement of the eclipse, - ? Begins to be total, - Middle, - Ceases to be total, End of the eclipse, Duration of total obscuration, - Duration of the eclipse, - The foregoing is Greenwich time, but can readily be converted into that of any other place, by allowing for the difference of lon- gitude. If strict accuracy were aimed at, the elements should be calcu- lated at several intervals during the eclipse, as they are liable to vary considerably. h. m. aec. 9 48 59 10 57 31 11 42 42 12 27 53 1 36 24 1 30 22 3 47 26 N 2 S O a 02 H H « M EXPLANATION OF TERMS, AS USED IN THIS WORK Ecliptic. The apparent annual path of the sun through the heavens. Nodes. The points where the orbit of a planet intersects the plane of the ecliptic. Ascending Node. That through which the planet passes from the south side of the ecliptic to the north side. Descending Node. That through which it returns 'to the south side. Line of the Nodes. A straight line connecting the two nodes. Equinoctial Points, or Equinoxes. The point where the ecliptic intersects the plane of the equator. Vernal Equinox. That through which the sun apparently passes from the south side of the equator to the north side. Autumnal Equinox. That through which it returns to the south side. Solstitial Points, or Solstices. Points in the ecliptic midway between the equi- noxes. Perigee. The point where the sun or moon approaches nearest the earth. Apogee. The point where they are most distant from the earth. Apsis. The common name for apogee or perigee. Apsides. The plural of apsis. Line of the Apsides. A straight line joining the two apsides. Conjunction. In the same direction as the sun. Opposition. In an opposite direction from the sun. Stzygy. The common name for conjunction and opposition. Quadrature. Points in the moon's orbit midway between the syzygies. Radius Vector. A straight line drawn from a revolving body to the centre about which it revolves. Latitude of a heavenly body. Its distance north or south of the ecliptic. Longitude of a heavenly body. Its distance eastwardly from the vernal equinox, measured on the ecliptic. Right Ascension. The same, measured on the equator. Declination. The distance of a heavenly body, north or south, from the equator. Lunation. The time from one new or full moon to another. Parallax. The apparent change in the place of a heavenly body, when viewed from different points. It is always equal to the angle which a line connecting the points of observation would subtend, when viewed from the body. Note. — It is thought best to omit, in the present edition of this work, a sequel, or second part, now in manuscript, and to which there have been several references in the foregoing pages, ex- plaining the method of calculating most of the lunar motions and inequalities directly from the laws of elliptical motion and the principles of gravitation, without the aid of tables. It may ap- pear hereafter. Errata. — Page 15th, &c. for elipse read ellipse; page 27, near the bottom, for syzygyes read syzygies. Page 32, Fig. 11, the lines NE and EP should be in the same straight line. I ASTRONOMICAL TABLES LIST OF TABLES 1. Elements of Orbits of Sun and Moon. 2. Mean New Moons in March, &c. 3. Mean Lunations. 4. Mean Motions in hours, minutes and seconds. 5 Days of the Year reckoned from March. 6. Annual Equation of the Moon's Perigee. 7. Annual Equation of the Moon's Node. 8. Equation of the Sun's Centre. 9. Equation of the Moon's Centre. 10. Annual Equation of the Moon's Longitude. 11. Secular Equation of the Moon's Longitude. 12 Variation. 13. Evection. 14. Annual Equation of Variation. Annual Equation of Evection. Nodal Equation of Moon's Longitude. Reduction to the Ecliptic. 18. Lunar or Menstrual Equation of the Sun's Longitude. 19. Lunar Nutation in Longitude. Sun's Semidiameter and Hourly Motion. Moon's Semidiameter, Hourly Motion and Equatorial Parallax. Do. as affected by Evection. Obliquity of the Ecliptic to the Equator. Moon's Latitude in Eclipses. Angle of the visible path of the Moon with the Ecliptic in Eclij Sun's Declination. 27. First Preliminary Equation. 28. Second Preliminary Equation. 29. Augmentation of the Moon's Semidiameter. 30. To convert minutes into decimals of a degree. 31. To convert seconds into decimals of a degree. 15. 16. 17. TABLE I. Elements of Orbits of Sun and Moon. Mean longitude, Jan. 1, 1801, .... Sun. Moon. O / // 280 39 13.17 118 17' 8.3 Motion in 100 years or 36525 days, - - - - 36000 46 0.77 481267 52 41.6 Mean longitude of perigee, Jan. 1, 1801, 279 31 9.71 266 10 7.5 Motion of do. eastward in 100 vears, ... 1 42 56.0 4069 2 46.6 Longitude of Moon's JNode, Jan. 1, 1801, 13 53 17.7 Motion of do. westward, in 100 years, ... 1934 9 57.5 TABLE II. Mean New Moon, SfC. in March. Mean New Moon in Sun's Mean Moon's Mean Sun and Moon's Longitude of Node. Year. March. Anomaly. Anomaly. Mean Longitude. 1800 d. 25 h. m. 18 sec. 49 83°208 127.960 2.7° 39 28.8207 1810 5 6 36 43 63.167 63.441 342.8436 196.4758 1820 14 1 38 40 72.232 24.740 352.0799 2.5671 1830 23 20 40 37 81.295 346.038 1.3163 168.6535 1840 3 2 58 31 61.254 281.520 341.4458 336.3135 1841 22 31 8 79.624 257.140 359.8335 315.9844 1842 11 9 19 42 68.888 206.943 349.1144 297.2190 1843 18 8 17 53.152 156.746 333.3954 278.4536 1844 18 15 40 54 76.522 132.366 356.7831 253.1245 1845 8 29 28 65.786 82.169 346.0640 23^.3591 1846 26 22 2 6 84.156 57.789 4.4517 219.0300 1847 16 6 50 40 73.420 7.592 353.7326 200.2646 1848 4 15 39 15 62.684 317.395 343.0135 181.4991 1849 23 13 11 52 81.054 293.015 1.4012 161.1702 1850 12 22 27 70 318 242.819 350.6822 142.4048 1851 2 6 49 1 59.583 192.622 339.9631 123.6394 1852 20 4 21 38 77.952 168.242 353.3508 103.3103 1853 9 13 10 13 67.217 118-045 347.6317 84.5449 1854 28 10 42 50 85.586 93.665 6.0194 64.2158 1855 17 19 31 24 74.851 43.468 355.3003 45.4504 1856 7 4 19 59 64.115 353.271 344.5813 26.6851 1857 25 1 52 36 82.485 328.891 2.9689 6.3560 1858 14 10 41 11 71.749 278.694 352.2499 347.5906 1859 3 19 29 46 61.013 228.497 341.5308 328.8252 1860 21 17 2 24 79.383 204.117 359.9185 308.4961 1861 11 1 50 58 68.647 153920 349.1994 289.7307 1862 10 39 33 57.911 103 723 338.4803 270.9653 1863 19 8 12 10 76.281 79.343 356.8680 250.6362 1864 7 17 45 65.545 29.146 346.1490 231.8708 1865 26 14 33 22 83.915 4.766 4.5366 211.5417 1866 15 23 21 57 73.179 314.569 353.8176 192.7763 1867 5 8 10 31 62.443 264.372 343.0985 174.0110 1868 23 5 43 9 80.813 239.992 1.4862 153 6819 1869 12 14 31 43 70.077 189.795 350.7671 134.9165 1870 1 23 20 18 59.34i 139.599 340.0481 116.1511 1871 20 20 52 55 77.711 115.219 358.4357 95 8220 1872 9 5 41 29 66.976 65.022 347.7167 77 0567 1873 28 3 14 6 85.346 40.642 6.1043 56'7275 1874 17 12 2 41 74.610 350.445 355.3853 37*9622 1875 6 20 51 15 63.874 300.248 344.6662 19*1968 1876 24 18 23 53 82.244 275.868 3.0539 358*8677 1877 14 3 12 28 71.508 225.671 352.3348 340 1023 1878 3 12 1 3 60.772 175.474 341.6158 3213370 1879 22 9 33 40 79.142 151.094 0.0034 301 0079 1880 10 18 22 15 68.406 100.897 349.2844 2822425 1881 3 10 50 57.670 50.700 338.5653 2634771 1882 19 43 27 76.040 26.320 356.9530 243 1480 1883 8 9 32 2 65.304 336.123 346.2339 2243826 1884 26 7 4 39 83.674 311.743 4.6216 2040535 1885 15 15 53 14 72.938 261.546 353.9025 185'2881 1886 5 41 49 62.202 211.349 343.1835 1665228 1887 23 22 14 26 80.572 186.969 1.5712 1461937 1888 1889 12 7 3 69.836 136.772 350.8521 127*4283 1 15 51 35 59.100 86.575 340.1330 108'6629 1890 20 13 24 12 77.470 62.196 353.5207 88*3337 1891 9 22 12 47 66.734 11.999 347.8016 69*5683 1892 27 19 45 24 85.104 347.619 6 1893 492392 1893 17 4 33 59 74.368 297 422 3554703 30*473;) 1894 6 13 22 33 63.632 247 225 344.7512 11*7085 1895 1896 25 10 55 10 82.002 222 845 3 1389 351*3794 13 19 43 45 71.266 172 648 352*4 198 332.6140 1897 3 4 32 19 60.530 122 451 341.7007 3 13 8486 1898 22 2 4 56 78.900 98 071 0.0884 2^.5195 1899 11 10 53 31 68.164 47.874 349.3694 274.7542 1900 19 42 6 57.428 357.677 338.6503 255.9888 This Table shows the time of New Moon in March, of each year, with the longitudes, anom- alies, &c. of the Sun and Moon at that time, on the supposition that all the motions are per- torrned with uniform angular velocity. r TABLE III. Mean Lunations. No. Sun's Mean Moon's Mean Sun and Moon's Longitude of Lun. Mean Lunations. Anomaly. Anomaly. Mean Longitude. Node. d. A. m. sec. 1 29 12 44 3 29.105 2%.817 2§.1067 1.5%38 2 59 1 28 6 58.211 51634 58 2135 3.1275 3 88 14 12 9 87.316 77.451 87.3202 4.6914 4 118 2 56 12 116.421 103.268 116.4270 6.2551 5 147 15 40 14 145.527 129.085 145.5/337 174.6405 7.8189 5 177 4 24 17 174.632 154.906 9.3827 7 206 17 8 20 203.738 180.718 203.7472 109465 8 236 5 52 23 232.843 206.535 232.8530 12.5102 9 265 18 36 26 261.948 232.352 261.9607 14.0740 10 295 7 20 29 291.054 258.169 291.0674 15.6378 11 324 20 4 32 320.159 283.986 320.1742 17.2016 12 354 8 48 35 349.264 j 309.803 349.2809 18.7654 13 383 21 32 37 378.370 335.620 378.3877 20.3291 h 14 18 22 1 14.553 192.908 *14.5534 .7819 Note. — The true quantities for one lunation, from which these tables are calcu- lated, are as follows, viz. Length of a lunation, - 29d. Sun's mean motion in Anomaly in one lunation, Moon's do. ----- Sun and Moon's mean motion in Longitude, do. Mean motion of the Node, do. 12A. 44m. 2.88sec. 29°.10535764 25 .81692410 29 .10674457 1 .56377989 TABLE IV. Mean Motions of the Sun and Moon. Sun's Sun's Moon's Moon's Longitude of Days. Anomaly. Longitude. Anomaly. Longitude. Node. 1 8.9856 8.98565 1§.0650 1§.17640 .05295 2 1.9712 1.97129 26.1300 26 35279 .10591 3 2.9568 2.95694 39.1950 39.52919 .15886 4 3.9424 3.94259 52 2600 52 70559 .21182 5 4.9280 4.92824 65.3250 65.88198 .26477 6 5.9136 5.91388 78 3900 79.05838 .31773 7 6 8992 6.89953 91.4549 92.23477 .37068 8 7.8848 7.88518 104.5199 105.41117 .42364 9 8 8704 8.87083 117.5849 118.58757 .47659 Hours. bun's Anomaly. .0411 .0821 .1232 .1643 .2053 .2464 .2875 .3285 .3696 feun's Longitude. .84107 .08214 .12321 .16427 .205:34 .24641 .28748 .32855 .36962 Moon's Anomaly. 0.5444 1.0887 1.6331 2.1775 2.7219 3.2662 3.8106 4.3550 4.8994 Moon's Longitude. .5^902 1.09803 1.64705 2.19607 2.74508 3.29410 3.84312 4 39213 4.94115 Longitude of Node. .00221 .00441 .00662 .00883 .01104 .01324 .01545 .01765 .01986 For the moon increase this by 180°. TABLE NO. IV CONTINUED. Min- Sun's Lon. Moon's Moon's Lon- Lon. of Se- Sun's Lon. Moon's Lon. utes. and Anom. Anom. gitude. Node. conds. and Anom. and Anom. 1 .00 45 .2693 .5911 .9871 315 50 .2682 .5864 .9833 310 55 .2671 .5816 .9792 305 60 .2659 .5767 .9749 300 65 .2647 .5714 .9705 295 70 .2635 .5661 .9659 290 75 .2623 .5608 .9613 285 80 .2611 .5556 .9567 280 85 .2599 .5503 .9521 275 90 .2586 .5450 .9475 270 95 .2574 .5397 - .9429 265 100 •2562 .5350 .9384 260 105 .2550 .5303 .9342 255 110 .2539 .5256 .9302 250 115 .2528 .5214 .9263 245 320 .2518 .5173 .9227 240 125 .2509 .5136 .9194 235 130 .2500 .5100 .9162 230 135 .2492 .5069 .9133 225 140 .2485 .5039 .9108 220 145 .2479 .5014 .9086 215 150 .2474 .4992 .9066 210 155 .2469 .4972 .9048 205 160 .2465 .4958 .9034 200 165 .2462 .4944 .9023 195 370 .2460 .4936 .9016 190 175 .2459 .4932 .9011 185 180 .2458 .4930 .9009 180 The principle and construction of this Table is the same as that of Table 20th. At the time of new or full moon the quantities in this Table must be increased for the effect of Variation as follows, viz. 1st column, .0020; 2d do. .0115; 3d do. .0073. TABLE XXII. Moon's Semi-diameter, Hourly Motion, and Equatorial Parallax, as affected by Evection. Argument — The same as for Evection, Table 14th. Arg. Semi-diameter. Hourly motion. Equatorial Parallax. Arg. 0° + .0029 +.0112 +.0105 360° 10 -f.0028 +.0109 +.0103 350 20 +.0027 +.0103 +.0098 340 30 +.0025 +.0097 +.0091 330 40 +.0022 +.0086 +.0081 320 50 +.0019 +.0070 + .0068 310 60 +.0014 + .0055 +.0052 300 70 + .0009 + .0036 +.0035 • 290 80 +.0005 +.0019 +.0018 280 90 - .0000 - .0002 - .0001 270 100 - .0005 - .0020 - .0019 260 110 -.0010 - .0038 - .0037 250 120 - .0014 - .0055 - .0054 240 130 - .0019 - .0071 - .0068 230 140 -.0022 -.0083 - .0080 220 150 - .0024 -.0094 - .0089 210 160 - .0026 - .0103 - .0096 200 . 170 -.0027 -.0107 - .0100 190 180 -.0028 - .0109 - .0103 180 All the inequalities in the moon's longitude, for which the foregoing Tables give the corrections, must likewise affect its apparent size, hourly motion, and equatorial parallax. Variation and Evection are the only ones that it is important to take into account, the former of which may be considered constant at the time of new or full moon, and this Table gives the requisite correction for the latter. TABLE XXIII. Obliquity of the Ecliptic to the Equator. Argument — The date. Arg. 184 185 186 18? 188 189 23 27 15.1 23 27 24 3 23 27 43.2 23 27 26.0 23 27 33.6;23 27 30 7 23 27 18.4 23 27 20.1 23 27 20.8:23 2717.4 23 27 13.4i23 2716.0 3 23 27 40.423 27 37. 1 23 27 33.5 23 27 28.3 23 27 31.0 23 27 33.6 23 27 27.2 23 27 23.7 23 27 21.3 23 27 235 23 27 26 23 27 27.9 23 27 13.9 23 27 10.8 23 27 8.4 23 27 18.3 23 27 20.0 23 27 20.7 o / II 23 27 29.6 23 27 35.7 23 27 17.6 23 27 29.0 23 27 6.8 23 27 20 5 6 23 27 27.1 23 27 37.2 23 27 15.8 23 27 29. 23 27 6.3 23 27 19.4 23 2724.8 23 27 377 23 2714.9 27 28.5 23 27 7.0 23 2717.3 2 23 23 2723.6 23 27 37.3 23 27 15.1 23 2726 7 23 27 8.6 23 27 14.4 23 27 23.4 23 27 36.0 23 27 15.9 23 27 24.1 23 27 10.8 23 2711.0 The obliquity of the ecliptic to the equator is slowly diminishing, owing to the attraction of the planets, and is also subject to an inequality whose period is about nineteen years, caused by the attraction of the moon, and called Nutation. This Table gives the obliquity on the 1st of January in each year, taking both these causes into account. TABLE XXIV. MoorCs Latitude in Eclipses. Argument — Moon's Longitude diminished by that of its Node. Arg. S.D. lsoio.oboo 181J0.0873 182 ! 0.1746 1880.6950 0.2617 0.3486 0.4353 0.5219 0.6085 0.7811 0.8672 0.9529 1.0382 1.1233 1.2082 1.2925 1.3765 Arg. 1°.0 0087 0960 1834 2704 3572 4440 5306 6173 7036 7897 8758 9614 0467 1318 2167 3009 3849 0.0175 .9 0.1048 0.1921 0.2791 0.3659 0.4526 0.5392 0.6258 0.7122 0.7983 0.8844 0.9700 1.0552 1.14031 1.2251 1 1.3093 1 1.3933 1 JB 0262 1135 2008 287810 37460 46130 54780 63430 72080 80690 8930;0 97850 0638 1 1488 2335 3177 4017 .7 0350 1222 2095 ,2965 3832 4699 5565 6431 7295 8155 9015 9870 0723 1573 2420 3261 4101 .6 0.0437 0.1310 0.2182 0.3052 0.3919 0.4786 0.5652 0.6517 0.7381 0.8241 0.9101 0.9955 1.0808 1.1658 1.2504 1.3345 1.4185 .5 0524 1397 2269 3139 4005 4873 5739 6604 7467 8327 9187 0041 0893 1743 2588 3428 4269 0.0611 0.1484 0.2356 0.3226 0.4092 0.4960 0.5826 0.6690 0.7553 0.8413 0.9273 1.0126 1.0978 1.1828 1.2672 1.3512 1.4353 4 1 3. .8 0.0698 0.1572 0.2443 0.3313 0.4180 0.5046 0.5912 0.6778 0.7639 0.8499 0.9358 1.0211 1.1063 1.1913 1.2756 1.3596 1.4437 .2 .9 lo.O Arg. 0.07850 0.1659 0.2530 0.3399 0.4267 0.5133 0.5999 0.6864 0.7725 0.8585 0.9444 1.0296 1.1148 1 1998 I l!2841 1 1 3680 1 1.4521 N.D. S-A 73 1791359 1746 178 358 2617,177 357 3486176356 4353 175,355 5219! 174 1 354 60851173 353 6950'172 352 7811J171 351 8672170i 350 9529169 349 0382|168 348 1233167,347 2082166 346 .2925165 345 .3765 164 344 1-4605 163] 343 Arg. The moon has sometimes a north and sometimes a south latitude, owing to the obliquity of the plane of its orbit to that of the ecliptic. This Table gives the latitude for every tenth of a degree of longitude, reckoned 17° either way from each node. The capital letters at the hea'd of the columns of the argument show whether the latitude is north or south, and whether it is ascending or descending. TABLE XXV. Angle of the visible path of the Moon with the Ecliptic in Eclipses. Arguments — Horary motion of the Moon from the Sun at the top, and the Moon's distance from the Node at the right and left. N. A. ~6° S. D. 180° .44 .46 .48 .50 .52 .54 .56 .58 .60 N. D. S. A. 5°47' 5°46' 5°45' 5044/ 5°43' 5°42' 5°41' 5°40' 5°39' 180° 360° 3 183 5 46 5 45 5 44 5 43 5 42 5 41 5 40 5 40 5 39 177 357 6 186 5 45 5 44 5 43 5 42 5 41 5 40 5 39 5 39 5 38 174 354 9 189 5 42 5 41 5 40 5 39 5 38 5 38 5 37 5 36 5 35 171 351 12 192 5 39 5 38 5 37 5 36 5 35 5 34 5 34 5 33 5 32 168 348 15 195 5 35 5 34 5 33 5 32 5 31 5 30 5 30 5 29 5 28 165 345 The angle of the moon's path with the ecliptic, which depends upon its dis- tance from the node, is apparently increased by the earth's motion in the same direction. This Table gives the apparent angle, taking both these facts into con- sideration. TABLE XXVI. The Sun's Declination. Argument — Sun's Longitude. / g 6 2^/?>.? Arg. Oo 1° 2° j 3o j 40 50 60 7o 80 9° lOo Arg. N. s. IS 0.0000 0°3980 0°7961 1°1939 1.5911 1°9S83 2°3S50 2°7809 3°1761 3°5703 3°9639 s. 35 N. 17 I iy 3.9639 4.3564 4.7477 5.1380 5.5267 5.9142 6.3000 6.6839 7.0664 7.4472 7.8259 34 16 Si 20 7.8259 8.2025 8.5767 8.9489 9.3189 9.6858 10.0506 10.4128 10.7720 11.1284 11.4817 33 15 B 2 l 11.4817 11.8317 12.1789,12.5225 12.8628 13.1997 13.5328 13.8622 14.1877 14.5092 14.8270 32 14 4 22 14.8270 15.1402 I5.4494jl5.7539 16.0542 16.3500 16.6408 16.9273 17.2086 17.4850 17.7561 31 13 b 23 17.7561 18.0223 18.2833.18.5386 18.7886 19.0331 19.2717 19.5042 19.7314 19.9528 20.1681 30 12 6 24 20.1681 20.3770 20.579520.776120.9664 21.1500 21.3267 21.4975 21.6614 21.8186 21.9689 29 11 V 25 21.9689 221122 22.2486 22. 378 1I22.5003 22.6153 22.7234 22.8242 22.9178 23.0042 23.0831 28 10 8 26 23.0831 23.1544 23.2184 23.2750 23.3242 6° 23.3658 5° 23.3997 23.4261 23.4450 23.4564 23.4603 27 9 Arg. 10o 90 8° ) 7° 40 3o 2° 10 Oo Arg. The plane of the ecliptic not coinciding with that of the equator, the sun is sometimes north of the equator and sometimes south. This is called its declina- tion, and this Table shows its amount for every degree of longitude. The epoch of the Table is 1840. TABLE XXVII. 1st Preliminary Equation. Argument — Moon's Anomaly. 0° 10 2o 30 40 50 6° 70 8° 9=> 10O > + h. ra. s. h. m. s. h. 111. s. h. m. s. h. m. s. h. m. s. h. m. s. h. m. s. h. ra. s. h. m. s. h. m. s. 9 34 19 8 28 41 38 13 47 44 57 13 1 6 41 1 16 7 125 31 134 54 35 1 134 54 144 16 153 36 2 2 53 2 12 8 2 21 19 2 30 28 2 39 34 2 48 39 2 57 43 3 6 45 34 2 3 6 45 3 15 44 3 24 42 3 33 38 3 42 32 3 123 4 7 4 8 47 4 17 25 4 26 1 4 34 33 33 3 4 34 33 4 43 2 4 51 15 4 59 42 5 7 56 5 16 5 5 24 9 5 32 9 5 40 4 5 47 54 5 55 38 32 4 5 55 38 6 3 16 6 10 49 6 18 18 6 25 40 6 32 56 6 40 6 6 47 6 6 54 8 7 1 2 7 7 50 31 5 7 7 50 7 14 30 7 21 2 7 27 22 7 33 36 7 39 46 7 45 46 7 5133 7 57 23 8 3 12 8 8 59 30 6 8 8 59 8 14 33 8 20 18 8 25 44 8 31 8 36 6 8 41 2 8 45 48 8 50 24 8 54 50 8 58 6 29 7 8 58 6 9 3 13 9 7 9 9 10 54 9 14 28 9 17 51 9 21 3 9 24 4 9 26 54 9 29 33 9 32 1 28 A 9 32 1 9 34 18 9 36 24 9 38 19 9 40 3 9 4136 9 42 59 9 44 11 9 45 12 9 46 3 9 46 44 27 9 9 46 44 9 47 14 9 47 33 9 47 46 9 47 54 9 47 49 9 47 36 9 47 13 9 46 38 9 45 52 9 44 53 26 K) 9 44 53 9 43 42 9 42 21 9 40 51 9 39 8 9 37 14 9 35 12 9 32 58 9 30 32 9 27 58 9 25 12 25 11 9 25 12 9 22 14 9 19 5 9 15 43 9 12 9 9 8 25 9 4 31 9 25 8 56 10 8 5145 8 47 8 24 12 8 47 8 8 42 18 8 37 19 8 32 11 8 26 53 8 2124 8 15 46 8 9 57 8 3 56 7 57 45 7 5124 23 13 7 5124 7 44 51 7 38 9 7 31 18 7 24 10 7 17 9 7 9 52 7 2 24 6 54 46 6 47 6 39 4 22 14 6 39 4 6 30 57 6 22 41 6 14 19 6 5 51 5 57 17 5 48 37 5 39 51 5 30 57 5 2156 5 12 48 21 If) 5 12 48 5 3 33 4 54 11 4 44 42 4 35 6 4 25 20 4 15 26 4 5 26 3 55 21 3 45 11 3 34 58 20 16 3 34 58 3 24 42 3 14 24 3 4 3 2 53 38 2 43 9 2 32 34 2 21 54 2 11 10 2 23 149 33 19 17 < 149 33 138 40 127 44 1 16 46 1 5 48 54 50 43 52 32 54 2156 10 58 000 18 < 10° 90 8° 70 6« 50 40 3° 2o 1° 0o When the moon's anomaly is less than 180°, it is in advance of its mean place at time of new or full moon by reason of the Equation of the Centre, but behind it by Evection, (Tables 9 and 14,) yet on the whole it is in advance; conse- quently it will overtake the sun sooner than it would otherwise do, and something must be subtracted from the mean time. The contrary takes place when the anom- aly is more than 180°. This Table shows the amount of time to be added or sub- tracted from these causes. TABLE XXVIII. 2d Preliminary Equation. Argument — Sun's Anomaly. OP + 1 2 ■3 4 5 6 7 8 9 10 II 12 13 11 152 16 1? m. s. 44 28 27 31 7 45 43 57 14 49 39 30 57 27 7 59 10 53 6 10 54 4 35 3 9 36 38 44 3 12 24 10 42 39 10° 1Q m. s 4 29 48 52 3141 1135 47 18 17 35 4140 53 52 8 37 10 45 5 18 52 29 32 45 6 45 35 22 59 26 20 6 38 26 9 go m. s. 8 56 53 13 35 49 15 20 50 36 20 20 43 45 12 9 10 10 33 422 50 50 30 26 3 51 3157 55 37 16 34 11 HO :p 13 23 57 36 39 56 19 5 53 49 23 45 44 126 9 39 10 16 3 23 49 7 28 3 40 ii. in. 17 1 1 141 2 22 2 57 3 25 3 47 4 2 4 10 4 9 4 2 3 47 3 25 50 54 2 57 28 29 5146 1153 28 55 7« 2 25 147 1 7 25 h. m. 500 22 56 1 47 35 38 35 4 55 18 1? 36 53 9 54 45 39 60 60 2 26 3 3 28 3 49 4 3 4 10 4 9 4 1 3 45 3 23 2 54 2 21 144 1 3 21 h. m. 26 1 10 52 2 30 3 3 3 30 3 51 4 4 4 10 4 8 3 59 3 43 3 20 2 51 2 17 1 40 59 17 5° 4° 70 h. m 31 10 1 14 49 156 5 2 33 35 3 6 10 3 32 50 3 52 49 4 5 37 4 10 49 4 8 21 3 58 27 3 4123 3 17 51 2 48 30 2 14 14 1 36 10 55 1 12 51 30 9° HO m. s. ft. in. 35 36 40 19 5 123 12 3 6 2 40 6311 3 37 37 9 35 6 54 26 6 29 10 54 7 41 57 2 39 18 15 9 45 18 10 36 32 12 51 4 8 35 2 rj 3 55 4 7 4 10 4 6 3 55 3 37 3 12 2 42 2 6 128 46 h. in. s. 44 28 35 10O 10 27 31 2 7 45 2 43 57 3 14 49 3 39 30 3 57 27 4 7 59 4 10 53 4 6 10 3 54 4 3 35 3 9 36 2 38 44 2 3 12 124 10 42 39 0_4_0 Oo When the sun's anomaly is less than 180°, it is before and the moon behind the mean place, by reason of the Equation of the Centre (Table 8) of the former, and the Annual Equation of the Longitude (Table 10) of the latter. For both rea- sons, then, the moon will not overtake the sun so soon as it would otherwise do, and consequently something must be added to the mean time of New or Full Moon. The contrary takes place when the anomaly is more than 180° ; and this Table shows the amount of time to be added or subtracted from these causes. TABLE XXIX. Augmentation of the Moon's Semi- diameter. Argument — Distance of the place (as projected on the disc) from the earth's centre. Tables 21 and 22 show us the apparent semi-diameter of the moon as viewed from the centre of the earth; but the distance of the moon from any place on the earth's surface at which it is visible (save when it is in the horizon) is less than from the centre, which must cause it to subtend a greater angle. This Table shows the augmentation in the moon's apparent semi- diameter from this cause. Arg. + .0045 10 .0045 20 .0044 30 .0043 40 .0041 50 .0038 60 .0035- 70 .0031 80 .0024 90 .0015 100 .0000 TABLE XXX. To convert minutes into decimals of a degree. Argument — The number of minutes. Arg. 0' I' 2' 3' 4' 5' 6' 7' 8' .1333 9' .0000 .0167 .0333 .0500 .0667 .0833 .1000 .1167 .1500 1 .1667 .1833 .2000 .2167 .2333 .2500 .2667 .2833 .3000 .3167 2 .3333 .3500 .3667 .3833 .4000 .4167 .4333 .4500 .4667 .4833 3 .5000 .5167 .5333 .5500 .5667 .5833 .6000 .6167 .6333 .6500 4 .6667 .6833 .7000 .7167 .7333 .7500 .7667 .7833 .8000 .8167 5 .8333 .8500 .8667 .8833 .9000 .9167 .9333 .9500 .9667 .9833 TABLE XXXI. To convert seconds into decimals of a degree. Argument — The number of seconds. Arg. 0" 1" 2" 3" 4" 5" 6" .0017 7" 8" 9" .0000 .0003 .0006 .0008 .0011 .0014 .0019 .0022 .0025 1 .0028 .0031 .0033 .0036 .0039 .0042 .0044 .0047 .0050 .0053 2 .0056 .0058 .0061 .0064 .0067 .0069 .0072 .0075 .0078 .0081 o .0083 .0086 .0089 .0092 .0094 .0097 .0100 .0103 .0106 .0108 4 .0111 .0114 .0117 .0119 .0122 .0125 .0128 .0131 .0133 .0136 5 .0139 .0142 .0144 .0147 .0150 .0153 .0156 .0158 .0161 .0164 QB5HI an