ANALYTIC GEOMETRY OF SPACE BY VIRGIL SNYDER, Ps.D. (Gérrmazn) Proressor or Maruematics at CORNELL, UNIversiTyY AND C. H.SISAM, Pu.D. (Cornetz) Assistant Prorrsson of MATHEMATICS AT THE Unrverarry oF Int1No1s NEW YORK , HENRY HOLT AND COMPANY 1914Meth 8509.1 Gai. COLE “JAN 30 1915 CorrzicHr, 1914, BY HENRY HOLT AND COMPANY Norio Press ‘J.8. Cushing Co. — Berwick & Smith Co. ‘Norwood, Mass., U.8.A.PREFACE In this book, which is planned for an introductory course, the first eight chapters include the subjects usually treated in rectangular codrdinates. They presuppose as much knowledge of algebra, geometry, and trigonometry as is contained in the major requirement of the College Entrance Examination Board, and as much plane analytic geometry as is contained in the better elementary textbooks. In this portion, proofs of theorems from more advanced subjects in algebra are supplied as needed. Among the features of this part are the development of linear systems of planes, plane codrdinates, the-concept of infinity, the treatment of imaginaries, and the distinction between centers and vertices of quadric surfaces. The study of this portion can be regarded as a first course, not demanding more than thirty or forty lessons. In Chapter IX tetrahedral coérdinates are introduced by means of linear transformations, under which various invariant proper- ties are established. These codrdinates are used throughout the next three chapters. The notation is so chosen that no ambigu- ity can arise between tetrahedral and rectangular systems. The selection of subject matter is such as to be of greatest service for further study of algebraic geometry. In Chapter XIII a more advanced knowledge of plane analytic geometry is presupposed, but the part involving Plicker’s num- bers may be omitted without disturbing the continuity of the subject. In the last chapter extensive use is made of the cal- culus, including the use of partial differentiation and of the element of arc. The second part will require about fifty lessons.,GoogleCONTENTS CHAPTER I COORDINATES anmioe 1, Cobrdinates es 8. Orthogonal projection ee 8. Direction cosines ofaline . . . 6 +e 4. Distance between two points. 6. wee 5. Angle between two directed lines . =. . . @. Point dividing a segment inagivenratio . . . 7%. Polarcotrdinates . . . ee wee 8 Cylindrical cobrdinates . . . . wwe 9 Spherical colrdinates . . . 1. 1 we CHAPTER II PLANES AND LINES Equation of a plane . . . A Ba 2 - Plane through three points : Intercept form of the equation of a plane Normal form of the equation of a plane . | Reduction of a linear equation to the normal form . Angle between two planes Distance to a point from a plane Equations of a line . Direction cosines of the line of intersection of two planes Forms of the equations ofaline . . . Parametri¢ equations ofaline =. . Angle which a line makes with a plane . Distance from a point to a line free Distance between two non-intersecting lines | |. System of planes throughaline . . . . Application to descriptive geometry . . v Page Soaraacr 1 12 18 14 16 16 7 19 19 21 22 25vi CONTENTS armous M Biniiectpaies 27. Plane codrdinates . . . . . . . . . 88, Equation of a point : ee 29, Homogeneous cotrdinate of the point and of the plane . 30, Equation of the plane and of the point in homogeneous cotrdinates $1, Equation of the origin. Codrdinates of lanes through the onetn $2, Plane at infinity . . : oO ine wiiiity 34. Codrdinate tetrahedron . . . . . . . : 85. System offourplans . . . . . ww CHAPTER II ‘TRANSFORMATION OF COORDINATES Translation . . . . . . . . . . |. Rotation . fa . . . + . + Rotation and reflection of axes... . Euler's formulas for rotation of axes. : Degree unchanged by transformation of codrdinates SSS2s8 CHAPTER IV TYPES OF SURFACES . Imaginary points, lines, and Planes soe Loci of equations. . . . . . . . Cylindrical surfaces =. ww wee Projecting cylinders Plane sections of surfaces . . . . . . . Cones . . . . * . . . He Surfaces of revolution . . . . . . . . SBSRERE CHAPTER V THE SPHERE ‘The equation of the sphere oe The absolute . oe Tangent plane. =... we Angle between two spheres 2. : Spheres satisfying given conditims . . . . . Linear systemsof spheres... ww ee Stereographic projection. . . . . . . . SSSSSS5 si SRRRRREB BEESB AT al 49 SISAassCONTENTS vii CHAPTER VI FORMS OF QUADRIC SURFACES axriotz race Definition ofaquadrio ©... ww ew The ellipsoid. . . . . . ie . . . 68 ‘The hyperboloid of one ee es }. The hyperboloid of two sheets . . . . . . ae emee . Theimaginary ellipsoid . . . . 1 wwe. 8 . The elliptic paraboloid . . www kw. OD |. The hyperbolic paraboloid ee - 70 . The quadric cons... : _ 7 . The quadric cylinders. . . . . . eee Sage 72 CHAPTER VII CLASSIFICATION OF QUADRIC SURFACES @5. Intersection ofaquadricandaline =. . . . . . 7% 66. Diametral planes, center . is . . . . % 61. Equation of a quadric referred to itscenter - | |. 68. Principal planes. soe 8 @9. Reality of the roots of the discriminating cubio | |. | 79 70. Simplification of the equation of a quadric oo 80 71. Classification of quadric surfaces... we 81 ‘72. Invariants under motion . . . . . . 82 78. Proof that J, J, and D are invariant . . . ie . . 88 14 Proofthat Aisinvariant. =... 0. www, 7%. Discussion of numerical equations . =... : 86 CHAPTER VIII SOME PROPERTIES OF QUADRIC SURFACES 76. Tangent lines and planes. | o ‘77. Normal forms of the equation of the tangent pian ee or 78. Normaltoaquadric . . oe 79. Rectiinear generators. ws wwe ee 80. Asymptotic cone . ve eee eee On 81. Plane sections of quadrics oa 82. Circular sections =. ee 88. Real circles on types of quadriss . | |} ses 100 84. Confocal quadrics + + 10t 85. Confocal quadrics through a point. "Bite cobrdinates |. 105 88. Confocal quadries tangent toaline. 5 ior 87. Confocal quadrics in plane cobrdinates . 9. 0... + 108,viii CONTENTS CHAPTER Ix TETRAHEDRAL COORDINATES armoe 88. Definition of tetrahedral cordinates 89. Unit point toe ee 90. Equation of a plane. Plane coérdinates 81. Equation ofapoint . 2. 1. 98. Equations of a line 98. Duality . 94. Parametric equations of a plane and of a point 85. Parametric equations of aline, Range of podnta Pencils of planes 98. Transformation of point codrdinates 97. Transformation of plane codrdinates 98. Projective transformations 8 99, Invariant points... wwe 00. Cromratio . 1 ww wk lt ” CHAPTER X Pace 110 im 112 12 113 14 16 17 ne 120 121 121 QUADRIC SURFACES IN TETRAHEDRAL COORDINATES 101: Form of equation: 108. ‘Tangent lines and planes 108. Condition that the tangent plane is indeterminate - 104. The invariance of the discriminant 105. Lines on the quadric surface: . ao 108, Equation of a quadric in plane codrdinates . 107. Polar planes . 108. Harmonic property of conjugate points : 109, Locus of points which lie on their own polar planes 110, Tangent cone. es LLL, Conjugate lines as toaquadric |) | 118, Self-polar tetrahedron : 118, Equation referred to a self-polar tetrahedron 114, Lawof inertia =. os 116. Rectilinear generators. Reguli : 116. Hyperbolic codrdinates. Parametric equations 117. Projection of a quadric upon a plane 118. Equations of the projection 119. Quadries determined by three non-; intersecting lines 190. Transversals of fourakewlines . . . 181, The quadric cone. 189. Projection of a quadric cone upon a plane 124 124 126 126 129 130 182 132 133 133 184 185 135 136 187 138, 189 140 141 us 143 145CONTENTS CHAPTER XI LINEAR SYSTEMS OF QUADRICS anrioye 188. Pencilofquadris. . 2 - . wl 194. The A-discriminant =... 125, Invariant factors . 126. The characteristic . 127. 128. 129. 180. 181, 182. 183. 184. 185. 186, 187. 188, 189. 140. ui. 142. 143. 144, 145, 146. 147. 148, 149. 150. 161. 153. 158. 154, 155. 186. 187. 158. 159. 160. Peneil of quadrics having a common vertex. . Classification of pencils of quadrics : Quadrics having a double plane in common . Quadrics having a line of vertices in common Quadrics having vertex in common Quadrics having no vertex in common Forms of pencils of quadrics . Line conjugate to a point ‘ Equation of the peneil in plane codrdinates Bundle of quadries Representation of the quadrics of a bundle % the pelts ofa plane Singular quadrics of the bundle Intersection of the bundle ys a Plane The vertex locus J Polar theory in the bundle Some special bundles Webs of quadrics . ee The Jacobian surface ofaweb =... Correspondence with the planes of space Web with six basis points : Linear systems of rank r Linear systems of rank rin plane codrdinates | Apolarity Linear aystems of apolar quadrices. |, CHAPTER XII TRANSFORMATIONS OF SPACE Projective metric - Pole and polar as to the absolute . Equations of motion : Classification of projective transformations . Standard forms of equations of Projoctive transformations Birational transformations. ae Quadratic transformations Saeed area Quadratic inversion. oe ‘Transformation by relproca! radii on Pace 147 aT 148 150 161 161 161 161 162 166 163 165 166 167 168 168 169 170 am 173 175 175 v7 Ww 180 181 181 186 188 188 190 191 195 196 198 201 201x CONTENTS CHAPTER XII CURVES AND SURFACES IN TETRAHEDRAL COORDINATES I, Axoxsraic Surraces arcu raex 161. iret am tanar hela os. 206 168. Notation oe Pe, 1907, 168. Intersection of aline and asurtaco =|. |. |. 207 164. Polar surfaces. ee 208 ies Chace es end plane 200 168, Inflexional tangents... 0. 0. eee 210 167. Double points. ee ai0) 168. The first polar surface and tangent cone. ai 169. Class of a surface. guation in plane cotrinaes aie 170, The Hessian. a 2 ais 171, The paraboliccurve . . . . . ww Od WB. TheSteimrian =. 5. ww wee II. Axozpraic Space Curves 178. Systems of equations defining a space curve. . . . . 216 174. Order of an algebraiceurve . . . 1 www. 18 175. Projectingoonss 2... eee 176. Monoidal representation. 2. 219 171. Number of intersections of algebraic curves and surfaces | ©. 221 178. Parametric equations of rational curves. rrer:,) 179, Tangent lines and developable surface of a curve . 4 2m 180. Osculating planes. Equation in plane codrdinates . 204 181. Singular points, lines, and planes. =. eT) 188. The Cayley-Salmon formulas. : ee 226 183. Curves on non-singular audit nracs : ae 228 184. Space cubic curves. sto 185. Metric clasefication of space cubic curves. + |. 984 186, Classification of space quarticcurves . . 0. ws 885 187. Non-eingular quartic eurves ofthe fst kind. eS Sea 188. Rational quartics . . oe CHAPTER XIV DIFFERENTIAL GEOMETRY I. Avaurmic Curves 189, Lengthofarcofaspacecurve . . . . « . 245 190, The movingtrihedral . . . we 246 ipl Cortare er usCONTENTS anno : 198. Torsion. ee 198. The Frenet-Serret formulas. |. 194. The osculating sphere . . 2. 0. we 195, Minimalcurvss 2 6. ee we Tl. Awazrric Surraces 196, Parametric equations of asurface. . . . 197. Systems of curvesonasurface . . . 198. Tangent plane. Normal line 199. Differential of are . : 200. Minimal curves 201. Angle between curves. Differential of surface 908. Radius of normal curvature. Meusnier’s theorem 203. Asymptotic tangents. Asymptotic curves . 904. Conjugate tangents. oe 205. Principal redii of normal curvature | | 906. Lines of curvature. ee 97. Theindicatrix =... ww ee AxeweRs : oo + 287 264 265 267 259 261 261 269Digitized by GoogleANALYTIC GEOMETRY OF SPACE CHAPTER I COORDINATES 1, Rectangular coérdinates. ‘The idea of rectangular codrdinates as developed in plane analytic geometry may be extended to space in the following manner. Let there be given three mutually perpendicular planes (Fig. 1) XOY, YOZ, ZOX, intersecting at O, the origin. These planes will be called coérdinate planes. The planes ZOX, XOY intersect in X’OX, the X-axis; the planes XOY, YOZ intersect in Y’OY, the Y-axis; the planes YOZ, ZOX intersect in Z'OZ, the Zaxis. Dis- tances measured in the directions X’OX, Y'OY, Z'OZ, respectively, will be considered positive; those measured in the opposite directions will be regarded as negative. The coérdi- Fra. 1. nates of any point P are its distances from the three codrdinate planes. The distance from the plane YOZ is denoted by 2, the distance from the plane ZOX is denoted by y, and the distance from the plane XOY is denoted by 2 These three numbers 2, y, 2 are spoken of as the 2, y-, zcodrdinates of P, respect- ively. Any point P in space has three real codrdinates. Con- versely, any three real numbers z, y, z, taken as x, y, and z codrdinates, respectively, determine a point P; for if we lay off a distance OA=z on the X-axis, OB=y on the Y-axis, OC =z on 12 COORDINATES (Caap. I. the Zaxis, and draw planes through A, B,C parallel to the co- ordinate planes, these planes will intersect in a point P whose codrdinates are x, y, and 2. It will frequently be more convenient to determine the point P whose codrdinates are'z, y, and z, as follows: Lay off the distance OA =2 on the X-axis (Fig. 2). From A lay off the distance AD =y on a parallel to the Y-axis. From D lay off the . distance DP =z on a parallel to the Zaxis. The eight portions of space separated by the codrdinate planes are called octants. If the codrdinates of a point P are a, 2, c, the points in the remaining octants at the same absolute distances from the codrdinate . planes are (—a, b, c), (a, —6, ¢), (a, b, — ¢), (— a,—b, €), (— a, b, —e), (a, — b, — €), (— a, — B,— 0). ‘Two points are symmetric with regard to a plane if the line joining the#i.is perpendicular to the plane and the segment between them'is bisected by the plane. They are symmetric with regard to a line if the line joining them is perpendicular to the given line and the segment between them is bisected by the line. They are symmetric with regard to a point if the segment be- tween them is bisected by the point. The problem of representing a figure in space on a plane is considered in descriptive geometry, where it-is solved in several ways by means of projections. In the figures appearing in this book a particular kind of parallel projection is used in which the X-axis and the Zaxis are represented by lines perpendicular to each other in the plane of the paper; the Y-axis is represented by a line making equal angles with the other two. Distances parallel to the X-axis or to the Z-axis are represented correctly to scale, but distances parallel to the Y-axis will be foreshortened, the amount of which may be chosen to suit the particular drawing considered. It will usually be convenient for the student, in drawing figures on cross section paper, to take a unit on the Y-axis 1/-V2 times as long as the unit on the other axes,Arr. 2] ORTHOGONAL PROJECTIONS 3 EXERCISES 1. Plot the following points to acale, using cross section paper: (1, 1, 1), 2, 0, 8), (—4, —1, —4), (— 8, —4, 1), (4, 4, — 1), (—7, 2, 3), (—1, 6, —5), (—4, 2,8), (3, -4, —1), (@, 1, —8), (—1, 0, 0), , —2, 2), (, 0, 2), @, —1, 9), (—8, 0, 0), (0, 0, 0). 2. What is the locus of a point for which = = 0? 3. What is the locus of a point for which z =0, y= 0? 4. What is the locus of a point for which z= a, y=? 5. Given a point (&, 1, m), write the codrdinates of the point symmetric with it as to the plane XOY; the plane ZOX; the X-axis; the Y-axis; the origin. 2. Orthogonal projections. The orthogonal projection of a point on a plane is the foot of the perpendicular from the point to the plane. The orthogonal projection ot a plane of a segment PQ of a line* is the segment P’Q! joining the projections P’ and Q! of Pand Q on the plane. The orthogonal projection of a point on a line is the point in which the line is intersected by a plane which passes through the given point and is perpendicular to the given line. The or- thogonal projection of a segment PQ of a line J on a second line Vig the segment P'Q! joining the projections P’ and Q! of P and Qonl. For the purpose of measuring distances and angles, one direc- tion along a line will be regarded as positive and the opposite direction as negative. A segment PQ on a directed line is positive or negative according as Q is in the positive or nega- tive direction from P. From this definition it follows that PQ =~ QP. The angle between two intersecting directed lines J and J’ will be defined as the smallest angle which has its sides extending in the positive directions along J and J’. We shall, in general, make no convention as to whether this angle is to be considered positive or negative. The angle between two non-intersecting directed lines 7 and /' will be defined as equal to the angle be- tween two intersecting lines m and m! having the same directions as land U', respectively, © We shall uso the word fine throughout to mean a straight line.4 COORDINATES [Cuar. I. Turorem I, The length of the projection of a segment of a directed line on a second directed line is equal to the length of the - given segment multiplied by the cosine of the angle between the lines. Let PQ (Figs. 3a, 38) be the given segment on J and let: P’Q! be its projection on /’. Denote the angle between 1 and 1’ by 6. It is required to prove that PQ! = PQ0os 6. Through P’ draw a line J" having the same direction as 7, The angle between U!' and 1” is equal to 6. Let Q” be the point in Fra. 3a. Fra, 3b. which 2 meets the plane through @ perpendicular to J’. Then the angle P'Q'Q” is a right angle. Hence, by trigonometry, we have PQ! = PQ" cos 6. But PQ" =PQ. It follows that P'Q'= PQcos 6. It should be observed that it makes no difference in this theorem whether the segment PQ is positive or negative. The segment PQ = +r will always be regarded as positive in defining cosines, Tuzorem II. The projection on a directed line 1 of a broken line made up of segments P,Py P,Ps ---, P,P, of different lines is the sum of the projections on 1 of its parts, and is equal to the pro- jection on 1 of the straight line P,P,.Arts. 2,3) DIRECTION COSINES OF A LINE 5 For, let Py Py P'y +++; Ply» P'y be the projections of P,, Py Py, +, P,., Py respectively. The sum of the projections is equal to P’,P’,; that is, PAPA PSP yt os + PyaP a= PPro But PP’, is the projection of P,P,. ‘The theorem therefore follows. Corottary. If Py, Py +++, P,, are the vertices of a polygon, the sum of the projections on any directed line 1 of the segments P,Py P;Py +++, Py1P, formed by the sides of the polygon is zero. Since in this case P, and P, coincide, it follows that P’, and P, also coincide. The sum of the projections is consequently zero. EXERCISES: 1. If O's the origin and P any point in space, show that the projections of the segment OP upon the cobrdinate axes are equal to the codrdinates of P. 2. If the cobrdinates of P; are x, y1, 21 and of P; are xa, yz, 22, show that the projections of the segment P,P; upon the codrdinate axes are equal to %a— By Ya— Yi, #2 — 21, Tespectively. 8. Ifthe lengths of the projections of P,P: upon the axes are respectively 8, — 2,7 and the codrdinates of P; are (— 4, 8, 2), find the codrdinates of P2. 4. Find the distance from the origin to the point (4, 8, 12). 5. Find the distance from the origin to the point (a, b,c). 6. Find the cosines of the angles made with the axes by the line joining the origin to each of the following points, 1, 2, 0) 1, 1) (7, 6, 2) (0, 2, 4) (1, - 4, 2) (tm 2) &. Direction cosines of a line. Let 1 be any directed line in space, and let 1! be a line through the origin which has the same direction. If @, By (Fig. 4) are the angles which 7’ makes with the codrdinate axes, these are also, by definition (Art. 2), the angles which 1 makes with the axes. They are called the direction angles of J and their cosines are called direction cosines. The latter will be denoted by A, y, v, respectively.6 COORDINATES [Cuar. I. Let P=(a, 6, c) be any point on?’ in the positive direction from the origin and let OP=r. Then, from trigonometry, we have a b e A=cosa=% =cosp=- =cosy=S. ee Cae ea But r is the diagonal of a rectangular parallelopiped whose edges are O0A=a, OB=d, O00=0 Henoe, we obtain «= r= Va EO FA In this equation, as in the formulas throughout the book, except when the contrary is stated, indicated roots are to be taken with the positive sign. By substituting this value of r in the above equations, we obtain A= 008 a = ——4*__ Vetere” b = cos B=" _ Be BS Ee e v= cosy=——°_.. Va+h+e By squaring each member of these equations and adding the results, we obtain Mag tah a hence we have the following theorem. Turorem. The sum of the squares of the direction cosines of a line i equal to unity. If Ay My v1 ANd Ay, pay ve are the direction cosines of two like directed lines, we have A= Dy = By 1 Ye If the lines are oppositely directed, we have A= Ay 4. Distance between two points. Let P,=(2, %, %1), r= (Cn Yo %) be any two points in space. Denote the direction cosines of the = Pay = Ve} ! Arrs, 4,5] ANGLE BETWEEN TWO DIRECTED LINES 7 line P,P, (Fig. 5) by A, », v and the length of the segment P,P, by d. The projection of the segment P,P, on each of the axes is equal to the sum of the projections of P,O and OP,, that is Ad=%— a, pd=y— fy, vd = —%. By squaring both members of these equations, adding, and extracting the N, ; ¥ 1 square root, we obtain Fre. 5. a= V1 — 21)? +(e — V3)? + Ge — 41) 8, (2) EXERCISES 1. Find tho distance between (3, 4, — 2) and (— 5, 1, — 6). 2. Show that the points (— 8, 2, — 7), (2, 2, — 8), and (— 8, 6, — 2) are vertices of an isosceles triangle. 3. Show that the points (4,3, —4), (—2, 9, —4), and (—2, 8, 2) are vertices of an equilateral triangle. 4. Express by an equation that the point (x, y, 2) is equidistant from 1, 1, 1) and (2, 8, 4). 5. Show that 2? + y? + 2% = 4 is the equation of a sphere whose center is the origin and whose radius is 2. 6. Find the direction cosines of the line P, Ps, given : = (0, 0, 0), P2= (2, 8, 5). Q,1,1), Pa= (2, 2, 2). 1=(1, —2, 8), Pr= (4,2, - 1). 7. What is known about the direction of a line if (a) cosa=0? (®) cose =0 and cosp=0? (c) cosa=1? 8. Show that the points (8, — 2, 7), (6,4, —2), and (5, 2, 1) are ona line, 9. Find the direction cosines of a line which makes equal angles with the codrdinate axes. 5. Angle between two directed lines. Let 4 and 4 be two directed lines having the direction cosines Ay, j4) ¥1 aNd Ay pay Yo) respectively. It is required to find an expression for the cosine of the angle between J, andl. Through O (Fig. 6) draw two8 COORDINATES [Cuar. I. lines OP; and OP, having the same di- rections as J, and 1, respectively. Let OP, = 7,and let the codrdinates of P, be ™%=0M, y,=MN, 4=NP. X The projection of OP, on OP, is equal to the sum of the projections of the Fra. 6. broken line OMNP, on OP, (Art. 2). Hence OP, 008 0 = OM + MN, + NP, y. But OP,=1y OM=a,=1y MN=y,=%yy NP=4 =" Hence, we obtain 7,008 6 or cos = rAide + Paptape + Tay = Mahe + Hata + v1¥2. ®) The condition that the two given lines are perpendicular is that cos@=0. Hence we have the following theorem: Taeorem. Two lines 1, and l; with direction cosines dy, py, v, and Day pay Vy Pespectively, are perpendicular if Dida + Pas + 112 = 0. “) The square of the sine of @ may be found from (1) and (3). Since sin* 6 = 1— cost 6, it follows that sin? = (AP + mat AP + pa? + 128) — Orda + Matte trim)? == dia — apt)? + Caive — oars)? + (rida — vedi) 6) 6. Point dividing a segment in a given ratio. Let P, = (a, 1%) and P, = (zy Y %) be two given points (Fig. 7). It is required to find the point P=(z, y, 2) onthe line P,P, such that P,P: PP, = m,:m, Let S 2, mv be the direction cosines of Fa 7. the line P,P, Then (Art. 2, Th. I) we have P,PA=2— 2, and PP,A=m—2. Hence P,P: PP,A= 2-2: %—L= mM: mM,Arr. 6] POINT DIVIDING A SEGMENT 9 On solving for # we obtain _ My + MD, any mg * ® Sa avi + mvs Similarly, ¥=— bm? Mat + mes | my + my It should be noticed that if m, and m, have the same sign, P,P and PP, are measured in the same direction so that Plies between P,and P, If m,and m, have opposite signs, P lies outside the segment P,P, By giving m, and m, suitable values, the coér- dinates of any point on the line P,P, can be represented in this way. In particular, if P is the mid-point of the segment P,Pz 11, = my 80 that the codrdinates of the mid-point are 2=8t5, yahth, rnhth. EXERCISES 1. Find the cosine of the angle between the two lines whose direction 1 2 3 2 1 5 are ——, ——, —— and —, ——, ——- vid’ vid Vid v30 .v80-V80 2. Find the direction cosines of each of the courdinate axes. 8. The direction cosines of a line are proportional to 4,—8, 1. Find their values, 4. The direction cosines of two lines are proportional to 6, 2, —1 and —8, 1, — 5, respectively, Find the cosine of the angle between the lines, 5. Show that the lines whose direction cosines are proportional to 3, 6, 2; —2,8, —6; —6, 2, 8 are mutually perpendicular. 6. Show that the points (7, 8, 4), (1, 0, 6), (4, 5, —2) are the vertices of a right triangle. 7. Show that the points (8, 7,2), (4, 3,1), (1, 6, 8), (2, 2,2) are the vertices of a parallelogram. 8. Find the cotrdinates of the intersection of the diagonals in the paral- lelogram of Ex. 7. 9. Show by two different methods that the three points (4, 13, 8), (8, 6, 4), (2, — 1, 8) are collinear.10 COORDINATES [Cuap. I. 10. A line makes an angle of 75° with the X-axis and 30° with the Y-axis. How many positions may it have? Find, for each position, the cosine of the angle it makes with the Z-axis, : 11. Determine the cotrdinates of the intersection of the medians of the triangle with vertices at (1, 2, 8), (2, 8, 1), (8, 1, 2). 12. Prove that the medians of any triangle meet in a point twice as far from each vertex as from the mid-point of the opposite side, ‘This point is called the center of gravity of the triangle. 13. Prove that the three straight lines joining the mid-points of oppo- site edges of any tetrahedron meet in a point, and are bisected by it. This point is called the center of gravity of the tetrahedron. 14. Show that the lines joining each vertex of a tetrahedron to the point of intersection of the medians of the opposite face pass through the center of gravity. 15. Show that the lines joining the middle points of the sides of any quadrilateral form a parallelogram. 16. Show how the ratio m,:m: (Art, 6) varies as P describes the line PiPy 7. Polar Coérdinates. Let OX, OY, OZ be a set of rectangular axes and P be any point in space. Let OP = p have the direc- AZ tion angles a, 8, y. The position of the line OP is determined by «, B, y and the position of Pon the line is given by p, so that the position of the point P in space is fixed when p, «, f, y are X known. These quantities p, a, 8, y are called the polar cotrdinates Y Fi. & of P. As a, B,y are direction angles, they are not independent, since by equation (1) cos? « + cos? B+ cos? y= 1. If the rectangular coérdinates of P are 2, y, z, then (Art. 3) z=poosa, y=pcosp, z=pcosy. 8. Cylindrical codrdinates. A point is determined when its directed distance from a fixed plane and the polar codrdinates of its orthogonal projection on that plane are khown. These co- ordinates are called the cylindrical cobrdinates of a point. If theAnts. 8, 9] SPHERICAL COORDINATES ll point P is referred to the rectangular axes x, y, 2, and the fixed plane is taken as z=0 and the waxis for polar axis, we may write (Fig. 9) z= pcos6, y=psin 6, =% in which p, 6, z are the cylindrical coordi- iP nates of P. Fa. 9. 9. Spherical codrdinates. Let OX, OY, OZ, and P be chosen as in Art. 7, and let P’ be the orthogonal projection of Pon the plane XOY. Draw OP. The position of P is defined by the distance p, the angle ¢= ZOP which the line OP makes with the z-axis, and the angle @ (measured by the angle XOP*) which the plane through P and the z-axis makes with the plane XOZ The num- bers p, ¢, @ are called the spherical coirdinates of P. The length pis called the radius vector, the angle ¢ is called the co-latitude, and 6 is called the longitude. If P=(#, y, 2), then, from the figure Gig. 10), OP' = pcos (90 — ¢)= pain ¢. Hence z=psin $cos 6, x y=psin sin 6, z=pcep * Y Wa. 10. On solving these equations for p, $, 6, we find p=vety+A, ¢=arecosx——*__ 6 =arctan Y. vetyte EXERCISES 1. What locus is defined by p = 1? 2. What locus is defined by a = 60°? 3. What locus is defined by 4. 5. . What locus is defined by . Transform 2% + y?+24= 4 to: (a) polar cotrdinates, (b) spherical codrdinates, (c) cylindrical coérdinates. 6. Transform 2! + y?= 22 into spherical cotrdinates; into cylindrical codrdinates. 7. Express the distance between two points in terms of their polar codrdinates.CHAPTER II PLANES AND LINES 10. Equation of a plane. A plane is characterized by the properties : : (a) It contains three points not on a line. (8) It contains every point on any line joining two points on it. - (©) It does not contain all the points of space. Tarorem. The locus of the points whose codrdinates satisfy a linear equation Az + By+ Cz+D=0 qd) with real coefficients is a plane. We shall prove this theorem on the supposition that C+ 0. Since A, B, C are not all zero, a proof for the case in which C=0 can be obtained in a similar way. It is seen by inspection that the codrdinates (0, 0, -3) (0 1, ~2eP), ( 0,- 242) satisfy the equation, ‘These threo points are not collinear, since no values of m,, m, other than zero satisfy the simultaneous equations (Art. 6) m=0, mn0, mA+m,B m0, Let P, = (2, Yy %) and P; = (%y Yy %) be any two points whose codrdinates satisfy (1). The codrdinates of any point P on the line P,P, are of the form ea Me tm y mit my, _ mime + ma, my + my m+ ‘my + my The equation (1) is satisfied by the codrdinates of P if (Ax, + By, + Cz, +D)+ m (Ax, + By + Cz + D)=0, but since the codrdinates of P, and P, satisfy (1), we have An + By+Cy+D=0, Am + By, +0z,+D=0, hence the coordinates of P satisfy (1) for all values of m, and m,. 12Arts. 11,12} INTERCEPT FORM OF THE EQUATION 13 Finally, not all the points of space lie on the locus defined by (1), since the cobrdinates @ 0, -2tO do not satisfy (1). This completes the proof of the theorem. 11. Plane through three points. Let (m, %, %); (t» Yn 4%)» (@y Ya %) be the codrdinates of three non-collinear points. The condition that these points all lie in the plane An + By +0z+D=0 is that their codrdinates satisfy this equation, thus Ax, + By, + Cz, +D=0, Az, + By, + 04+ D Ax, + By, + Ca +D The condition that four numbers A, B, C, D (not all zero) exist which satisfy the above four simultaneous equations is 2 y 2 im Hm 1 et. me 1 fe Ys % 1 This is the required equation, for it is the equation of a plane, since it is of first degree in 2, y, z (Art. 10). The plane passes through the given points, since the codrdinates of each of the given points satisfy the equation. @) 12. Intercept form of the equation of a plane. If a plane inter- sects the X-, Y-, Zaxes in three points A, B, C, respectively, the segments OA, OB, and OC are called the intercepts of the plane. Let A, B, C all be distinct from the origin and let the lengths of the intercepts be a, b, ¢, so that A =(a, 0,0), B=(0, , 0), C= (0, 0, ¢). The equation (2) of the plane determined by these three points (Art. 11) may be reduced to . @) This equation is called the intercept form of the equation of a plane. Ziuye ee ur14 PLANES AND LINES [Cuar. II. EXERCISES : 1. Find the equation of the plane through the points (1, 2, 8), (8 1, 2), (5, ~ 1,8). 2. Find the equation of the plane through the points (0, 0, 0), (1, 1, 1), (2,2, —2). What are ite intercepts ? 3. Prove that the four points (1, 2,8), (2,4, 1), (—1, 0, 1), (0, 0, 5) lie in a plane. Find the equation of the plane. 4. Determine & so that the points (1,2, —1), (8, —1, 2), (2, —2,°8), 1, —1, &) shalll ie in a plane. 5. Find the point of intersection of the three planes, x+y+2=6, Qe—y42e=0,2—-2y+82=4, 18. The normal form of the equation of a plane. Let ABC (Big. 11) be any plane. Let 0@ be drawn through the origin per- pendicular to the given plane and intersecting it at P. Let the direction cosines of OQ be 2, m, v and denote the length of the segment OF by p. Let P= (a, y, 2) be any point in the given plane. The projec- tion of P on OQ is P' (Art. 2). Draw OP and the broken line OMNP, made up of segments Fra, 11. M oy=«, MN=y, and NP=2, parallel to the X., ¥:,and Zaxes, respectively. The projections of OP and OMNP on OQ are equal (Art. 2, Th. II), The projection of the broken line is Ax + py + vz, the projection of OPis OP" or p, so that dat py tve=p. (4) This equation is satisfied by the coordinates of every point P in the given plane. It is not satisfied by the codrdinates of any other point. For, if P, is a point not lying in the given plane, it is similarly sben, since the projection of OP, on OQ is not equal to p, that the codrdinates of P, do not satisfy (4). Hence, (4) is the equation of the plane. It is called the normal form of the equation of the plane. The number p in this equa- tion is positive or negative, according as P’ is in the positive or negative direction from O on 0Q.Arr. 14] REDUCTION OF THE EQUATION 15 14, Reduction of the equation of a plane to the normal form. Let An + By +Cz+D=0 (6) be any equation of first degree with real coefficients. It isrequired to reduce this equation to the normal form. Let Q=(4, B, C) be the point whose codrdinates are the coefficients of 2, y,z in this equation. The direction cosines of the directed line OQ are (Art. 3) A=——_4___, p= Sa ea V+ PFO VA+EB+O? Vat RTO? If we transpose the constant term of (8) to the other member of the equation, and divide both numbers by VA?+B?+C%, we obtain (6) y ——4 ny z VEFREO | VAP BO oe eae @) VET EEO + B+ The plane determined by (7) is identical with that determined by (5) since the coordinates of a point will satisfy (7) if, and only if, they satisfy (6). By subtituting from (6) in (7) and comparing with (4), we see that the locus of the equation is a plane perpen- dicular to OQ and intersecting OQ at a point P' whose distance from O is p=. VA + B+ Ct In these equations, the radical is to be taken with the positive sign. The coefficients of z, y, z are proportional to A, », v in such a way that the direction cosines of the normal to the plane are fixed when the signs of A, B, C are known. But the plane is not changed if its equation is multiplied by — 1, hence the position of the plane alone is not sufficient to determine the direction of the normal. In order to define a positive and a negative side of a plane we shall first prove the following theorem: TuEoreM. Two points P,, P; are on the same side or on opposite sides of the plane Ax + By +Cz+D=0, according as their codrdi- nates make the first member of the equation of the plane have like or unlike signs, 8)16 PLANES AND LINES [Cuap. II. . For, let Py = (ay ¥y %), P= (2 n@ be two points not lying onthe plane. The point P =(z, y, z) in which the line P,P, inter- sects the plane is determined (Art. 6) by the values of m, m, which satisfy the equation sm Airy + By, + 0% + D) + (Ax, + By + On +D)=0. If Az, + By +C2+D and Az,+ By, +Cz+D have unlike signs, then m, and m, have the same sign, and the point P lies be- tween P, and P, If Am+ By +Cz+D and Ax, + By: + O% +D have the same sign, then the numbers m, m, have opposite signs, hence the point P is not between P, and P, When all the terms in the equation An+ By+@+D=0 are transposed to the first member, a point (2 yy, %) will be said to be on the positive side of the plane if Az, + By, +Cx+ Disa positive number; the point will be said to be on the negative side if this expression is a negative number. Finally, the point is on the plane if the expression vanishes. It should be observed that the equation must not be multiplied by —1 after the positive and negative sides have been chosen. 15. Angle between two planes. The angle between two planes is equal to the angle between two directed normals to the planes; hence, by Arts. 5 and 14, we have at once the following theorem: Txeorem. The cosine of the angle 0 be- tween two planes Ax + By+ Cz+D=0, A'z+ By + O'2+D'=0 is defined by the equation oz AA' + BB'+CC' s (9) V49+ B+ V4? + BI+ OF In particular, the condition that the planes are perpendicular is AA! + BB +CC'=0. (10)Ants. 15,16] DISTANCE TO A POINT FROM A PLANE 17 ‘The conditions that the planes are parallel are (Art. 3) B 4=2-§. (11) The equations (11) are satisfied whether the normals have the same direction or opposite directions. From the definition of the angle between two planes it follows that in the first case the two planes are parallel and in the second case they make an angle of 180 degrees with each other. We shall say, however, that the planes are parallel in each case. 16. Distance toa point from a plane. Let P=(x, y,%) be a given point and Az+ By +Cz+D=0 be the equation of a given plane. The distance to P from the plane is equal to the distance from the given plane to a plane through P parallel to it. The equation As + By + Cz — (Az, + By, + Cy)=0 represents a plane, since it is of first degree with real coefficients (Art.10). It is parallel to the given plane by Eqs. (11). It passes through P since the codrdinates of P satisfy the equation. When the equations of the planes are reduced to the normal form, they become, respectively, A aie + VATB+O VA B+ oe VEER LO VETO ——4 «+ __8__, VATE VAT B+ 4 A+ B+ Cn, VA+B4+OQ VA BO ‘The second members of these two equations represent the dis- tances of the two planes from the origin, hence the distance from the first plane to the second, which is equal to the distance d to P from the given plane, is found by subtracting the former from the latter.18 PLANES AND LINES [Cuar. II. Red imed 4m+BuitOn+D, a2 Vat+ B+ CF The direction to P from the plane, along the normal, is positive or negative according as the expression in the numerator of the second member is positive or negative (Art. 14), that is, according as P is on the positive or negative side of the plane. EXERCISES 1. Reduce the equation 8 z — 12 y — 4 z — 26 = 0 to the normal form. 2. Write the equation of a plane through the origin parallel to the plane z+2y=6, 3, What is the distance from the plane 8z + 4y—2=65 to the point (2, 2, 2)? 4. Find the distance between the parallel planes Qe—yt+8e=4, Qz—y +8245 = 5. Which of the points (4, 8,1), (1, — 4, 8), (8, 6, 2), (—1, 2, —2), (5, 4, 6) are on the same side of the plane 62—2y—82=0 as the point (1, 6, — 8)? 6. Find the codrdinates of a point in each of the dihedral angles formed by the planes B24+2y+52—4=0, —2y—24+6=0. 7. Show that each of the planes 252 +39y+82—48=0 and 252 — 80 + 1122 + 113 =0 bisect a pair of vertical dihedral angles formed by the planes 52+ 122+7=0 and 8y—42—6 Which plane bisects the angle in which the origin lies ? 8. Find the equation of the plane which bisects that angle formed by the planes 82—2y+2—4=0, 22+ y—82—2=0, in which the point , 8, — 2) lies. 9. Find the equations of the planes which bisect the dihedral angles formed by the planes Aiz + By + Cie + Di =0, Age + Bayt Or2 + Da =0. 10. Find the equation of the locus of a point whose distance from the origin-is equal to its distance from the plane 8% + y—22=11. 11. Write the equation of a plane whose distance from the point (0, 2, 1) is 3, and which is perpendicular to the radius vector of the point (2, — 1, — 1). 12. Show that the planes 22—y+24+3=0, 2-y+4z2=0, 82+y —2248=0, 42—2y+22—6=0, 92+ 8y —62—7=0, and 7z—Ty +282 — 6 =0 bound a parallelopiped. 13. Write the equation of a plane through (1,2, —1), parallel to the plane z — 2y— 2 = 0, and find its intercepts.Arts. 17,18] DIRECTION COSINES OF THE LINE 19 14. Find the equation of the plane passing through the points (1, 2, 8), (2, — 8, 6) and perpendicular to the plane 42 + 2y +82 =1. 15. Find the equation of the plane through the point (1, 8, 2) perpen- dicular to the planes Qu48y—4e=2, 42—8y—22=5, 16. Show that the planes x +2y—2=0, y+72¢-2=0, z—2y—# —4=0, 248y+2=4, and 824+8y—z2=8 bound a quadrilateral pyramid, 17. Find the equation of the locus of s point which is 3 times as far from the plane 8z—6y—2z=0 as from the plane 2z—y+22=9. 18." Determine the value of m such that the plane mz +2y—382=14 shall be 2 units from the origin. 19. Determine & from the condition that x — ky + 3 2 = 2 shall be perpen- dicular to 82 +4y—22 =6. 17. Equations of aline. Let 42+ By+Cz+D,=Oand Ax + By + Cz + D,=0 be the equations of two non-parallel planes. The locus of the two equations considered as simultaneous is a line, namely, the line of intersection of the two planes (Art. 10). The simultaneous equations Ag+ By+Cz+D,=0, Apt By + Cg+D,=0 are called the equations of the line. The locus represented by the equations of two parallel planes, considered as simultaneous, will be considered later (Art. 33). 18. Direction cosines of the line of intersection of two planes. Let A, v be the direction cosines of the line of intersection of the two planes Aet+By + Cz+D,=0, 1, = Ag+ By + Cz+D,=0. Since the line lies in the plane L, =0, it is perpendicular to the normal to the plane. Hence, (Arts. 5, 14) AA, + mB, + vO, =0. Similarly, AA; + pB, + vO, =0. By solving these two equations for the ratios of A, p, v, we obtain 4 (13) —4+— =__* _.. BO,— BC, G,4,— 0,4, A,B, — A,B,20 PLANES AND LINES {Cuap. II. The denominators in these expressions are, therefore, proportional to the direction cosines. In many problems, they may be used instead of the direction cosines themselves, but, in any case, the actual cosines may be determined by dividing these denominators by the square root of the sum of their squares. It should be observed that the equations of a line are not sufficient to deter- mine a positive direction on it. 19. Forme of the equations of a line. If A, u,v are the direction cosines of a line, and if P,=(m, 1, %) is any point on it, the distance d from P, to another point P=(q, y, 2) on the line satis- fies the relations (Art. 4) Md=2—4, pd=y—y, vd=z—y, By eliminating d, we obtain the equations Ba % YW _ tw 14 : 7m a (14) which are called the symmetric form of the equations of the line. Instead of the direction cosines themselves, it is frequently convenient to use, in these equations, three numbers a, b, ¢, pro- portional, respectively, to A, 4, v. The equations then become Poh UWF, (15) a 6 c They may be reduced to the preceding form by dividing the de- nominator of each member by Va?+0?+ c (Art. 3). If the line (15) passes through the point P;=(m, Y» %), the codrdinates of P, satisfy the equations, so that On eliminating a, b, ¢ between these equations and (15), we obtain 24s =i’, Bh ww 2-H These equations are called the two-point form of the equations of a line, (16)| Arr. 20} PARAMETRIC EQUATIONS OF A LINE 21 20. Parametric equations of a line. Any point on a line may be defined in terms of a fixed point on it, the direction cosines of the line, and the distance d of the variable point from the fixed one. Thus, by Art. 4 z=mtdd, y=ytnd, z=m+u0. (17) If A,» v are given and (a,, y,, %) represents a fixed point, any point («, y, z) on the line may be defined in terms of d. To every real value of d corresponds a point on the line, and conversely. ‘These equations are called parametric equations of the line, the parameter being the distance. It is sometimes convenient to express the codrdinates of a point in terms of a parameter « which is defined in terms of d by a linear fractional equation of the form aatthr, y+ 8 in which a, 8, y, 8 are constants satisfying the inequality a8 — By #0. By substituting these values of « in (17) and simplifying, we obtain equations of the form atthe | the 7 ath BS tbe? ay bg ab be” (8) in which a, 0,, ete., are constants. Equations (18) are called the parametric equations of the line in terms of the parameter x. It should be observed that the denominators in the second members of equations (18) are all alike. Each value of «x for which a,+ #0 determines a definite point on the line. As a, + b,x approaches zero, the distance of the corresponding point from the origin increases without limit. To the value deter- mined by’ a,+x=0 we shall say that there corresponds a unique point which we shall call the point at infinity on the line. EXERCISES 1. Find the points in which the following lines plorce the codrdinate planes : (a) 2+2y—-82=1, 82-2y4+5e= (0) 24+ 8y+52=0, b2—B8y4+2= (c) z+2y—5=0, 22-8y4+22=7. 2.22 PLANES AND LINES [Cuap. II. 2. Write the equations of the line z+y—S2=6, 22—y422=7 in the symmetric form, the two-point form, the parametric form. 3. Show that the lines 42+ y—82=0, 22—-y+2z2+6=0, and8z ~yte=1, 102+ y—42+41=0 are parallel, 4. Write the equations of the line through (8, 7, 8) and (—1, 5, 6). Determine its direction cosines. 5. Find the equation of the plane passing through the point (2, — 2, 0) and perpendicular to the line # =8, y= 2x — 4. 248 6. Find the value of & for which the lines 7—8 — ¥+1_2+8 2k k+l 3 = ute. te are perpendicular. 7. Do the points (2, 4, 6), (4, 6, 2), (1, 8, 8) ie on a line? 8. For what value of & are the points (k, — 8,2), (2, — 2, 8), (6, —1, 4) collinear ? 9. Is there a value of & for which the points (k, 2, — 2), (2, —2, &), and (— 2, 1, 8) are collinear ? 10. Show that the line “= 2 — ¥+2 — —8 ties in the plane 22 +2y 3 =-1 —2+8=0. 11. In equations (18) show that, as « approaches infinity, the correspond- ing point approaches a definite point asa limit, Does this limiting point lle on the given line ? 21. Angle which a line makes with a plane. Given the plane Ae + By+ Cz2+D=0 and the line ‘The angle which the line makes with the plane is the complement of the angle which it makes with the normal to the plane. The direction cosines of the normal to the plane are proportional to A, B, Cand the direction cosines of the line are proportional to a, b, ¢, hence the angle @ between the plane and the line is de- termined (Art. 5) by the formula aA+dB+00 : as) VET BLOVa pore sind =Arrs. 21, 22} DISTANCE FROM A POINT TO A LINE 23 EXERCISES 1. Show that the planes 22 —8y+2+1=0, 62+2—1=0, 42+ 9y—z—6=0 havea line in common, and find its direction cosines. 2. Write the equations of a line which passes through (6, 2, 6) and is parallel to the line 22—82+y—2=0, 2+y+e+1=0. 3. Find the angle which the line x++y+2z=0, 22—y+22—-1=0 makes with the plane 8x + 6 2—5y +1=0. 4. Find the equation of the plane through the point (2, —2, 0) and perpendicular to the line z + 2y—82=4,2%—8y+42=0. Find the equation of the plane determined by the parallel lines e-8_yt4is—l 8 2 1 6. For what value of k will the two lines z+ 2y—2+3=0, 82—y+ 2241=0; 22-y+2—-2=0,2+y—2+k =0 intersect ? 7. Find the equation of the plane through the points (1, — 1,2) and (8, 0, 1), parallel to the line z+ y—2=0,22+y+2=0. . Show that the lines 22 =Htte 5 intersect, and find the equation of the plane determined by them. 9. Find the equation of the plane through the point (a, b, ¢), parallel to each of the lines, 2—%1 —¥—¥1 _2— #1; 2—% _Y— a _ 2 22, 4 m m 4 m, Ty 10. Find the equation of the plane through the origin and perpendicular tothe line 82—y+42+5=0,2+y—2=0, 11. Find the value of & for which the lines 2=$ = ¥+1 2k k+l 2 and a¥t5_ 242 3 1 “k-2 12. Find the values of & for which the planes kr — 6 y +(k + 6)z+8=0 and (k— 1)x + ky + 2 =0 are perpendicular. 13. Find the equations of the line through the point (2, 8, 4) which meets the Y-axis at right angles. are perpendicular, 22. Distance from a point toa line. Given the line eh YH _t*—-4 x » ¥ and the point P; =(%y ys) %) not lying on it, It is required to find the distance between the point and the line.Dy PLANES AND LINES [Cuar. I. Let P, = (ty vy %) (Fig. 13) be any point on the line; let P be the foot of the perpendicular from P, on the line; 6 the angle between the given lineand x the line P,P,; let d be the length of the segment P,P, We have (Fig. 13) Y Fig. 13. P,P? = P,P? sin* 6 = @ — & cos? 6. The direction cosines of the line P,P, are aH, a4 a4, from which (Art. 5) HaSr%4y, Bony ,2=%, cos 6 =A pote! tae Hence, PP} = @ — Boos 6 =(% — %)* +(%s — Hh)" + ( — 4 —QG—a)+ HG —")t%—%))*% (20) 28. Distance between two non-intersecting lines. . Given the two lines Bom YW 2% ang ZY _ 2 yu Pa % 4 Pa Ye which do not intersect. It is required to find the shortest dis- tance between them. Let A, », v be the direction cosines of the line on which the distance is measured. Since this line is per- pendicular to each of the given lines, we have, by Art. 5, Equations (4) and (5), oe Pave — Vita vg — vey a where @ is the angle between the given lines. The length d of the required perpendicular is equal to the projection on the common per- pendicular of the segment PP’, “Y Fra, 14. and is equal to the projection of the broken line PMNP” (Fig. 14).Arts, 23,24] SYSTEM OF PLANES THROUGH A LINE 25 d=A@ — %)+ oO — mH) +¥G% — 4) © tee Mal d=t|n-% mh | @1) aoe ain 6 EXERCISES 1. Find the distance from the origin to the line 2. Find the distance from (1, 1, 1) toz +y+2=0,82—2y+42=0. 3. Find the perpendicular distance from the point (— 2, 1, 8) to the line z+2y—z24+3=0,82—-y4+2241=0, 4. What are the direction cosines of the line through the origin and the point of intersection of the lines 2 +2y—2+8=0, 8z—y+22+1=0; Q2—2y +82—2=0,2-y—24+8=0. 5. Determine the distance of the point (1, 1, 1) to the line z= 0, y =0 and the direction cosines of the line on which it is measured. G. Find the distance between the lines gut? ey-8_e241 2 =1 7. Find the equations of the line along which the distance in Ex. 7 is measured. 8. Find the distance between the lines 22 +y—z2=0, z—y+22=8 and 2 4+2y—82=4,22—8y4+42=6. 9. Express the condition that the lines *—*! -¥—¥1—_2— #1, £—% 4 m m’ sto £3 intersect. 24. System of planes through a line. If L,= Ag+ By + Cz +D,=0, L,= Ag +By + Cz+D,=0 are the equations of two intersecting planes, the equation k,Z, + k, L, =0 is, for all real values of k, and ky the equation of a plane passing through the line L,=0, Zy=0. For, hl, + kyl, = 0 is always of the first degree with real coefficients, and is therefore the equation of a plane (Art. 10); this plane passes through the line Z,=0, L,=0, since the codrdinates of every point on the line satisfy Z,=0 and Z_=0 and consequently satisfy the equation26 PLANES AND LINES (Caap. II. kL, + kjLz=0. Conversely, the equation of any plane passing through the line can be expressed in the form kL, + kyl, =0, since k, and k, can be so chosen that the plane kJ, + il, =0 will contain any point in space. Since any plane through the given line is determined by the line and a point not lying on it, the theorem follows. To find the equations of the plane determined by the line Z, = 0, I, =0,and a point P, not lying on it, let the codrdinates of P, be (@y Yy %)- If P, lies in the plane k,L, + kL, = 0, its cobrdinates must satisfy the equation of the plane; thus (Am, + Ban + Crr+Di) + ky Am, + Bay, + Om + Ds) = 0. On eliminating &, and k, between this equation and kL, + kely = 0, we obtain 0 = (Ag, + By, + Cx, + D,)(Aze + By + Oz + D;) — (Az, + Bin + Cm + D) (Aw + By + Cz + D,), as the equation of the plane determined by the line Z,=0, I, = 0, and the point P,. It will be convenient to write the above equation in the abbre- viated form s(x) Ig (2) — Iy(24)La(2) = 0. The totality of planes passing through a line is called a pencil of planes. The number k,/, which determines a plane of the pencil is called the parameter of the pencil. If, in the equation KL, + kL, = 0, i, and iy are given such values that the coefficient of a is equal to zero, the corresponding plane is perpendicular to the plane z= 0. Since this plane contains the line, it intersects the plane # = 0 in the orthogonal projection of the line Z,=0, L,=0. Similarly, if k and , are given such values that the coefficient of y is equal to zero, the corresponding plane is perpendicular to the plane y=0 and will cut the plane y=0 in the projection of I, =0, L,=0 on that plane; if the coefficient of z is made to vanish, the plane will contain the projection of the given line upon the plane z =0. The three planes of the system KZ, + k,l» =0 obtained in this way are called the three projecting planes of the line Z, = 0, Ly =0 on the codrdinate planes.Arr. 24] SYSTEM OF PLANES THROUGH A LINE 27 Since two distinct planes passing through a line are sufficient to determine the line, two projecting planes of a line may always be em- ployed to define the line. If the line is not parallel to the plane z=0, its projecting planes on « =0 and y = 0 are distinct and the equations of the line may be reduced to the form (Fig. 15) Y Fra. 15. eameta, yoneth. (22) If the line is parallel to z = 0, the value of k for which the coeffi- cient of 2 is made to vanish will also reduce the coefficient of y to zero, 80 that the projecting planes on z=0 and on y=0 coincide. This projecting plane z =c and the projec- ting plane on z = 0 may now be chosen to define the line. If the line is not x parallel to the X-axis, the equations of the line may be reduced to (Fig. 16) ra. 16. . z=pyto z=e. (23) Finally, if the line is parallel to the X-axis, its equations may be reduced to (Fig. 17) Y 4 y=b 250. (24) If the planes J, = 0, Ly = 0 are par- (B allel but distinct, so that ‘0 x A_B_G A BG, ¥ A, B, OQ,” D,’ Fa. 17. then every equation of the form k,l, + %,L,=0, except when A BG Fm G defines a plane parallel to the given ones. Conversely, the equation of any plane parallel to the given ones can be written in the form kL, + kl; =0 by so choosing hy : ky28 PLANES AND LINES (Cuap. II, that the plane will pass through a given point. In this case the system of planes kyZ, + Kyl = 0 is called a pencil of parallel planes. ‘Two equations, L,= Az+ By + Oz + D, =0, 1,= Ag+ By+Cg+DP=0, will represent the same plane when, and only when, the coefficients A, By G, D, are respectively proportional to A, By, Cy D,; thus, when ‘These conditions may be expressed by saying that every deter- minant of order two formed by any square array in the system ja 2 om : A BG Dy shall vanish. In this case multipliers %,, k, can be found such that the equa- tion kL, + k,L, = 0 is identically satisfied. Conversely, if multipliers k,, K; can be found such that the pre- ceding identity is satisfied, then the equations L,=0, L,=0 define the same plane. EXERCISES 1. Write the equation of a plane through the line 7z + 2y 82—8y +22 —6 =0 perpendicular to the plane 22 + y — 22 = 2. What is the equation of the plane determined by the line 22 —8y— 24+2=0,2—y+42=8 and the point (3, 2, — 2)? 3. Determine the equation of the plane passing through the line 2422 =4,y—2= 8 and parallel to the line 2—8 -¥+4 _2—7, 4. Does the plane x+2y—z2+3=0 have more than one point in common with the line 82—y +22+1=0, 22—8y+82—-2=0? 5. Determine the equations of the line through (1, 2, 8) intersecting the two lines x + 2y—82=0, y—42=4 and 22—-y +82=8, 82 +y + 22 +1=0. 25. Application in descriptive geometry. A line may be repre- sented by the thrée orthogonal projections of a segment of the line, each drawn to scale. Consider the XZplane (elevation, or verti- cal plane) as the plane of the paper, and the XY-plane as turned about the X-axis until it coincides with the XZplane. The pro-Arts. 25, 26] BUNDLES ‘OF PLANES 29 jections in the XY-plane are thus drawn to scale on the same paper as projections on the XZ-plane, but points are distinguished by different symbols, as PY, P,. Q Zz The X¥-plane is called the plan or horizontal plane. Finally, let the YZ-plane be turned about the Zaxis until it coincides with the XZplane, and let figures in the new position be drawn to scale. This is called the end or profile plane. Thus, in the figure (Fig. 18), » segment PQ, wherein ¥ P=(7, 4, 8), Q=(13, 9, 12), Fro. 18. may be indicated by the three segments P'Q’, P,Q, P,Q,. Examere. Find the equations of the projecting planes of the line Qu+8y—4e=5, c—4y+52=6, Here, = Ly=224+8y—42—-5, Ip=z—4y +526, Kay + KaLa =(2 ky + ko) + (8 hy — 4 ha)y + (— 4h +6 ke) + (— 5k — 6x) = 0. It ky = —2k;, the coefiicient of x disappears; thus the equation of the plane projecting the given line on the plane z = 0 is Wy- 14247 I Bot the coeficient of y vanishes; the projecting plane on y =0 is 4 fonnd to be 1x — 2 = 88, Finally, Bf, tho projecting plane on z =0 is found. Its equation is dz —y = 49. EXERCISES Find the equations of the projecting planes of each of the following lines : Lox+2y—-82=4, 22—-Sy+42=5. 2 224+y+2=0, n—yt2e=8. Bo xtyte2=4, go-yt8z2=4. 4 A+ By+ Ce+D,=0, Age+ By + Cz +Di=0. 26, Bundles of planes. The plane L,= Ag+ By+Oz+D, =0will belong to the pencil determined by the planes Z,= 0, L,=0, assumed distinct, when three numbers ky ky ky not all zero, can be found such that the equation k,l, + kL, + kL, = 0 is identi-30 PLANES AND LINES {Cuar. II. cally satisfied for all values of 2, y, 2. This condition requires that the four equations k,A, + #4, -+ ky A,=0, 1B, + kB, + #,B,=0, kG +h0:+hC,=0, %,D,+h,D,+hD,=0 are satisfied by three numbers k, ky ky not all zero; hence, that the four equa- tions |A,BG,|=0, |B,C,D,|=0, |C.DA|=0, | D,A,B,| =0 are all satisfied, wherein we have written for brevity, A BG A, B, G, A BG These simultaneous conditions may be expressed by saying that every determinant of order three formed by the elements contained in any square array in the system [- BG D, shall vanish. 4, B, C, D,| A, BG Dy Conversely, if these conditions are satisfied, then three con- stants , ky, k, can be found such that the equation kL, + kL, + Iyly=0 is identically satisfied, and the three planes Z,=0, L,=0, L,=0 belong to the same pencil. Let L=Age+ By + Oz+D,=0, L,= Ag+ By + Cz+D,=0, L,= Aw+ By + Cz+D;=0 be the equations of three planes not belonging to a pencil. If we solve these three equations for (2, y, 2), we find for the codrdinates of the point of intersection of the three planes, in case | 4,B,C,| #0, |AB,G| = 1 oto. Moe aoe | to PAC | (oa4eG) aaiaee oo If | .4,B,C,| =, but not all the determinants in the numerators of (25) are zero, no set of values of 2, y, 2 will satisfy all three equations. -In this case, the line of intersection of any two of the planes is parallel to the third. For, if Z,= 0 and L, = 0 intersect,Aarts, 26, 27]. PLANE COORDINATES 31 the direction cosines of their line of intersection are proportional (Art. 18) to BOBO, CA,—GA, “4B, — A.B, The condition that this line is parallel to the plane L,=0 is (Art. 21) A,(B,C;— B,C,) + B,(C,A, — C,A,) + C,(A,B; — A,B,)= 0, which is exactly the condition |.4,B,C,;|=0. The proof for the other lines and planes is found in the same way. If at least one of the determinants | A,B,C, |, | D,B,C,|, |4,D:Cs|, and |.4,B,D,| is not zero, the system of planes kyLy + gly + byl, =0 is called a bundle. If |ABC|+#0, all the planes of the bundle pass through the point (25), since the codrdinates of this point satisfy the equation of every plane of the bundle. Conversely, the equation of every plane passing through the point (25) can be expressed in this form. This point is called the vertex of the bundle. If |ABC|=0, all the planes of the bundle are parallel to a fixed line (such as L,=0, Z,=0). In this case, the bundle is called a parallel bundle. 27, Plane cotrdinates. The equation of any plane not passing through the origin may be reduced to the form uz +vy+uz+1=0. (26) ‘When the equation is in this form, the position of the plane is fixed when the values of the coefficients w, v, w (not all zero) are known; and conversely, if the position of the plane (not passing through the origin) is known, the values of the coefficients are fixed. Since the numbers (w, v, w) determine a plane definitely, just as (x, y, 2) determine a point, we shall call the set of num- bers (wu, v, w) the coordinates of the plane represented by equation (26). ‘Thus, the plane (3, 8, 2) will be understood to mean the plane whose equation is 32+5y+2z+1=0. Similarly, the equation of the plane (2, 0, —1) is 22—z+1 I£ u, v, w are different from zero, they are the negative recipro- cals of the intercepts of the plane (u, v, w) on the axes (Art. 12).32 PLANES AND LINES [Cuap. IL. If u=0, the plane is parallel to the X-axis; if u=0, 0 =0, the plane is parallel to the X¥-plane. The vanishing of the other coefficients may be interpreted in a similar way. 28, Equation of a point. If the point («, y,, %) lies in the plane (26), the equation um +, + wy +1=0 (21) must be satisfied. If 2, x, % are considered fixed and u, v, w variable, (27) is the condition that the plane (u, , w) passes through the point (m, yz). For this reason, equation (27) is called the equation of the point (2, 1, 2) in plane codrdinates. Thus, u—5v+2w4+1=0 is the equation of the point (1, —5, 2); similarly, 8utw+1=0 is the equation of the point (3, 0, 1). If equation (27) is multiplied by any constant different from zero, the locus of the equation is unchanged. Hence, we have the following theorem : Turorem. The linear equation Au + Bu+Ow+D=0 (D#0) : - nie BON. ‘, is the equation of the point (4, 2, ©) in plane codrdinates. D’ D’ D, Thus, u-5v—3w—2=0 is the equation of the point foot 2222 The condition that the codrdinates (v, u, w) of a plane satisfy two linear equations way + ry, tum +1=0, um + my, +um+1=0 is that the plane passes through the two points (a, y, %) and (@ Yx %) and therefore through the line joining the two points. The two equations are called the equations of the line in plane~ codrdinates.Ants. 28, 29] HOMOGENEOUS COORDINATES 33 EXERCISES 1. Plot the following planes and write their equations: (1, 2, }), (8, — 4, —t)(-L4-b- 2. Find the volume of the tetrahedron bounded by the codrdinate planes and the plane (— 4, — }, —4)- 3. What are the codrdinates of the planes whose equations are Tz+5y—2+1=0, 2—-6y+1le+5=0, O2—4=0? 4. Find the angle which the plane (2, 6, 5) makes with the plano (-1, 42). 5. Write the equations of the points (1, 1, 1), (2, — 1, 4), (6, —2, 1). 6. What are the cotrdinates of the points whose equations are Qu-c—-Swt1=0, ut+2w—-8=0, w—2=0? 7. Find the direction cosines of the line Su-o+2wt1=0, ut+bo+2u 0. 8. What locus is determined by three simultaneous linear equations in (u, 0, w) ? 9. Write the equation satisfied by the codrdinates of the planes whose distance from the origin is 2. What is the locus of » plane which satisfies this condition ? 29. Homogeneous cotfdinates of the point and of the plane. It is sometimes convenient to express the courdinates 2, y, z of a point in terms of four numbers 2’, y', z', t’ by means of the equations , ' : A set of four numbers (2', y', 2’, t'), not all of which are zero, that satisfy these equations are said to be the homogeneous codrdinates of a point. If the codrdinates (2, y', 2’, ') are given, the point is uniquely determined (for the case t'=0, compare Art. 32), but if (2, y, 2) are given, only the ratios of the homogeneous coordinates are determined, since (2’, y', z', t') and (ka', ky’, kz', kt") define the same point, k being an arbitrary constant, different from zero. Similarly, if the codrdinates of a plane are (u, v, w), four num- bers (w', o', w', s'), not all of which are zero, may be found such that w.34 PLANES AND LINES [Cuap. II. The set of numbers (u', v', w', s') are called the homogeneous cotrdi- nates of the plane. Where no ambiguity arises, the accents will be omitted from the homogeneous codrdinates. 30. Equation of a plane and of a point in homogeneous cotrdinates. If, in the equation Ax + By + Cz+ D=0 (D+0, and A, B, C are not all zero) the homogeneous codrdi- nates of a point are substituted, we obtain, after multiplying by t, the equation of the plane in homogeneous coérdinates Az + By + Cz+ Dt=0. The homogeneous codrdinates of this plane are (A, B, C, D). If, in the equation Au + Bu+Cw+D=0 (D+, and A, B, Care not all zero) the homogeneous codrdi- nates of a plane are substituted, we obtain, after multiplying by 8, the equation of the point in homogeneous codrdinates Au + Bu+ Cw+ Ds=0. The homogeneous codrdinates of this point are (4, B, C, D). 81, Equation of the origin. Coordinates of planes through the origin. The necessary and sufficient condition that the plane whose equation is uz+ vy+wz+st=0 shall pass through the origin is s=0. We see then that s=0 is the equation of the origin, and that (u, v, w, 0) are the homogeneous coérdinates of a plane through the origin. Since s = 0, it follows from Art. 29 that the non-homogeneous codrdinates of such a plane cease to exist. 32, The plane at infinity. Let (2, y, 2, t) be the homogeneous codrdinates of a point. If we assign fixed values (not all zero) to 2, y, 2 and allow ¢ to vary, the corresponding point will vary in such a way that, as t= 0, one or more of the non-homogeneous co- ordinates of the point increases without limit. If ¢=0, the non- homogeneous codrdinates cease to exist, but it is assumed that there still exists a corresponding point which is said to be at infinity. It is also assumed that two points at infinity coincide if, and only if, their homogeneous coérdinates are proportional. The equation of the locus of the points at infinity is t=0. Since this equation is homogeneous of the first degree in a, y, 2, t, it will be said that ¢= 0 is the equation of a plane. This plane ‘ called the plane at infinity.Arts. 33, 34) COORDINATE TETRAHEDRON 35 33, Lines at infinity. Any finite plane is said to intersect the plane at infinity in a line. This line is called the infinitely dis- tant line in the plane. The equations of the infinitely distant line in the plane Az + By + Cz + Dt=0 are Ax + By + Cz=0,t=0. Tuxorem. The condition that two finite planes are parallel is that they intersect the plane at infinity in the same line. , If the planes are parallel, their equations may be written in the form (Art. 15) Az + By+Oz+Dt=0, A+ By+Oz+Dit=0. (28) It follows that they both pass through the line Az+ By+(z=0, t=0. (29) Conversely, the equations of any two finite planes through the line (29) may be written in the form (28). The planes are there- fore parallel. 84, Coérdinate tetrahedron. The four planes whose equations in point coérdinates are 2=0, y=0, z=0, t=0 will be called the four codrdinate planes in homogeneous codrdi- nates. Since the planes do not all pass through a common point, they will be regarded as forming a tetrahedron, called the codrdi- nate tetrahedron. The coérdinates of the vertices of this tetra- hedron are (0,0, 0,1), (0,0, 1,0), (0, 1,0,0), (4, 0, 0, 0). The codrdinates of the four faces in plane coordinates are ©, 0, 0,1), (0,0, 1,0), ©,1,0,0), (1,0, 0, 0). ‘The equations of the vertices are u=0, v=0, w=0, 8=0. EXERCISES 1. Find the non-homogeneous codrdinates of the following pointe and planes : (a) 72+8y+4+32-4=0, @ 9u—0—8w4+2=0, (0) 102—8y+16=0, (e-) w+v—w-T=0, (c) 2-2=0, (f) 2w411=0. 2. Determine the cobrdinates of the infinitely distant point on the line Ba4+2y+5t=0, 22—10248t=0.36 PLANES AND LINES (Cuar. IT. 3. Show that if Z)(u)= Aju + Bio + Ciw + Dis =0, and Ls(u)= Agu + By + Cyw + D238 = 0 are the equations of two points, the equation of any point on the joining line may be written in the form k,L; + k2Z, =0. 4. Show that the planes x + 2y+72—8t=0,2+8y+62=0,2+4y +62 —2t=0 determine a parallel bundle, Find the equation of the plane of the bundle through the points (2, — 1, 1, 1), (2, 6, 0, 1). 35. Systemof four planes. The condition that four given planes L,=Agzt+ By + Cz + Dt=0, L,= Ag+ By + Cz + Dt Ag + By + Oe + Dt = Ag+ By t+ O¢+ De= 0 all pass through a nn is that four numbers (2, y, z, #), not all zero, exist which satisfy the four simultaneous equations. ‘The condition is, consequently, that the determinant. A BG D, A, B, Q, D, A BG D, ABO D is equal to zero. If this condition is not satisfied, the four planes are said to be independent. When the given planes are independ- ent, four numbers k,, k,, k,, k, can always be found such that the equation KL, + kyly + byly + iD, = 0 shall represent any given plane. For, let ax-+ by-+cz +d=0 be the equation of the given plane. The two equations will repre- sent the same plane if their coefficients are proportional, that is, if numbers hy, ky ky ky not all zero, can be found such that KA, + kA, + gy t+ ky Ay KB, + kB, + By + By CHhC, +O, +h, + kCy d=hD, + kD, + hDy + kDy Since the planes are independent, the determinant of the coeffi- cients in the second members of these equations is not zero, and the numbers k,, ky k;, k, can always be determined so as to satisfy these equations. a=Arr. 35] SYSTEM OF FOUR PLANES 37 ‘These results, together with those of Arts. 24, 26,.may be ex- pressed as follows: The necessary and sufficient condition that a system of planes have no point in common is that the matrix® formed by their coefficients is of rank four; the planes belong to a bundle when the matrix is of rank three; the planes belong to a pencil when the matrix is of rank two; finally, the planes all coincide when the matrix is of rank one. We shall use the ex- pression “rank of the system of planes” to mean the rank of the matrix of coefficients in the equations of the planes. EXERCISES 1. Determine the nature of the following systems of planes : (a) 22—6y+2—8t=0, c+ y+42—5t=0, 2+ 8y4+62—t=0. (0) 8244y+62—5t=0, 62+5y4+92—10t=0, 84+ 3y +52 —5t=0, c—-y+22=0. (0) 22-4+4y=0, Gx+Ty +22 =0, S2+4y—2243t (a) 224 by+82=0, Ty—624+4t=0, 2—y44e=86. 2. Show that the line z+ 3y—#+¢=0,22—y+22—8t=0 lies in the plane 7z +7y +2—8t=0. 3. Determine the conditions that the planes eacy+bz, y=ax+cz, 2=be tay shall have just one common point ; a common line ; are identical. 4. Prove that the planes 2z—8y—72=0, 32—My—182=0, 8x —81y— 882 =0 have a line in common, and find its direction cosines, 5. Show that the planes 82—2y—¢=0, 42-22-2¢=0, 4z+4y — 62 =0 belong to a parallel bundle. 0, 2=0. * Any rectangular array of numbers A, BO, Dy Ay By Cy Dy - As Bs Cy Dg ~~ Ms An Bu Cy Da Mn is called a matrix, Associated with every matrix are other matrices obtained by suppressing one or more of the rows or one or more of the columns of the given matrix, or both ; in particular, associated with every square matrix, that is, one in which the number of rows is equal to the number of columns, is a de- terminant whose elements are the elements of the matrix. Conversely, associated with every determinant is a square matrix, formed by its elements. We shall use the word rank to define the order of the non-vanishing determinant of high- est order contained in any given matrix. ‘The rank of the determinant is defined as the rank of the matrix formed by the elements of the determinant.CHAPTER III TRANSFORMATION OF COORDINATES The codrdinates of a point, referred to two different systems of axes, are connected by certain relations which will now be determined. The process of changing from one system of axes to another is called a transformation of cobrdinates. 36. Translation. Let the codrdinates of a point P with respect to a set of rectangular axes OX, OY, OZ be (a, y, z) and with respect to a set of axes 0X’, OY’, O'Z’, parallel respectively to the first set, be (a’,y,2’). If the codrdinates of 0’, referred to the axes OX, OY, OZ are (h, k, 2) we have (Fig. 19) a=ad+th, yay +h, z=74+L (1) For, the projection on OX of OP is equal to the sum of the pro- jeotions of OO’ and O'P (Art, 2), but the projection of OP is z, of OO’ is h, and of O'Pis x’; hence a=a+h. The other formulas are derived in a similar way. Since the new axes can be obtained from the old ones by moving the three codrdinate planes parallel to the Xaxis a distance h, then parallel to the Y-axis a distance k, and Fia. 19. parallel to the Zaxis a distance 1, without changing their directions, the transformation (1) is called a translation of axes. 87. Rotation. Let the codrdinates of a point P, referred to a set of rectangular axes OX, OY, OZ, be a, y,z, and referred to another rectangular system OX', OY’, O2' having the same origin, be 2, y', 2. Let 2! = OL', y'= L'M, z' = M'P(Hig. 20); and let the direction cosines of OX', referred to OX, OY, OZ, be Ay pny 13 hose of OF" be Ay Hay ve and of OZ! be Ay py Ve 38