MATHEMATICAL MONOGRAPHS. MANSFIELD MERRIMAN axp ROBERT S, WOODWARD, No. 4. ma . iy HYPERBOLIC FUNCTIONS. JAMES McMAHON, Late Provisy ue Mast MATIC 18 Conseil UNIVERSELY FOURTIL EDITION, ENLARGED. NEW YORK: JOHN WILEY & SONS. Loxpon: CHAPMAN & HALL, Limtep,Comvarcnr, #896, MANSFIRI.D M¥RRIMAN aun RORERT § WOOMWARD HIGHER MATHEMATICS. Firat Haition, September, 1695. Second Kéition, January, 1898. Third Edison, August, 1900. Fourth Edition, January, 1906, Printed in U.S.A.EDITORS’ PREFACE. of which was published in 1896, contained eleven chapters by eleven authors, cach chapter being independent of the others, but all supposing the reader to have at least a mathematical training equivalent to that given in classival and engincering colleges. The publication of that volume is now discontinued and the chapters are issued in separate form. * Tn these reissues it will generally be found that the monographs” are enlarged by additional articles or appendices which either amplify the former presentation or record recent advances. This plin of publication has been arranged in order to meet the demand of teachers and the convenience of classes, but it is also thought th mathematical Hiewaiue, It is the intention of the publishers and editors to add other monographs to the series from time to time, if the call for the same seems to warrant it, Among the topics which are under consiceration are those of elliptic functions, the theory of num- bers, the group theary, the calculus of variations, and non- Euclidean geometry; possibly also monographs on branches of astronomy, mechanics, and mathematical physics may be included. It is the hope of the editors that this form of publication ‘may tend to promote mathematical study and research qyer a wider field than that which the former volume has occupied, December, 905, ieAUTHOR'S PREFACE, ‘This compendium of hyperbolic trigonometry was first published as a chapter in Merman and Woodward's Higher Mathematics, ‘There 1s reason to believe that it suppiies a need, being adapted to two or three «illerent types of readers. College students who have had elementary courses in trigonometry, analytic geometry, and differ- ential and integral calculus, and who wish to know something of the hyperbolic trigonometry on account of its important and historic rela- tions to each of those branches, will, it 1s hoped, find these relations presented in a simple and comprehensive way in’ the first half of the work. Readers who have some interest in imaginaries are then intro- duced to the more general trigonometry of the complex plane, where the circular and hyperbolic functions merge into one class of transcend- ents, the singly periodic functions, having either a real or a pure imag- inary period. For those who alo wish to view the subject in some of ns, numerous appheations have hern selected so as pil iy shows its mse tthe mode of approachmg the subject, and to the presentation of funda mental notions, and it is hoped that some improvements are diserni ble, For instance, it has been customary to define the hyperbole functions in relation to a sector of the rectangular hyperbola, and to lake the mitial radius of the sector coincident with the prneypal radius, of the curve, am the present work, these and similar restrictions are discarded mn the mterest of analogy and generality, with a gain in syme metry and simplicity, and the functions are defined as certain chitrace teristic ratios belonging to any sector of any hyperbola, Such defini- tions, m connection with the fruitful notion of correspondence of points ‘on conics, lead to simple and general proofs of the addition-theorems, from which easily follow the conversion-formulas, the derivatives, the Maclaurin expansions, and the exponential expressions, The proofs, are so arranged as to appiy equally to the circular tunctons, regarded as the characteristic ratios belonging to any elliptic sector, For those, however, who may wish to start with the exponential expressions as the definitions of the hyperbohe functions, the appropriate order of procedure is indicated on page 25, and a direct mode of bringing such exponential definitions into geometrical relation with the hyperbolic sector 1s shown in the Appendix. December. 1905Ant. |. CHARACTERISHIE K. . FuNctions OF PURE IN >. FUNCTIONS OF +1y IN THE FORM CORRESPONDENCE OF POINTS ON CONICS, AREAS OF CORRESPONDING ‘THIAS AREAS OF CORRESPONDING SECTORS Ratios As TRIANGLE-MFASURES. FUNCLIONAL REIATONS LOR ELLIPSE, FUNCTIONAL RELATIONS. FOR HIVPFRHOLA S. RRLALIONS BFIWEEN HIYPERHOLIC FUNCTIONS VARIATIONS oF THE HYPE RHOLIC FUNCLIONS, \NTEHYPERNOLIC FUSCHONS 6g FUNCTIONS OF SUMS AND DIFFERENCES Cosy esto: Fouts: Lawrie Ratios Denivarives oF Hyprnonie F DERIVATIVES OF ANUTIYPHRHOLIC FUNCTIONS SXPANSION OF TIYPHRHOLIC FUNCTIONS EXPONENTIAL EXPRESSIONS XPANSION OF ANTI-FENCTIONS . Locartritic EXPRESSION OF ANIEEUNCHIONS: ‘Tue. GuprRMaxtan FuNcrion CiRcULAR FUNCTIONS OF GUDERMANIAN Gupeeastan ANGLE oe DERIVATIVES OF GUDERMANIAN AND INVPRSE SERIES FOR GUDERMANIAN AND ITS INVERSE . GRAPHS OF HyPERHOLIC FUNCTIONS MENTARY INTEGRAL ae Fusctioss or Compiex NUMBRRS =. ADDITION THEOREMS FOR COMPLEXES. CINARIES. vay Tue CaTENARY an SARY OF UNIFORM STRENGTH. Tue Can ‘Tue Euastic CATENARY 5 es ire Teacromy | CU enna THE LOXODROME . se ee 08 OF SHCTORIAL MEASURE Page 4 6 6 8 ay 24 as 7 + 8 +8 + 9 + 3 nat na 35 ° a7 0 5° 26 conrenrs. Anz, 36 CompiNep FLExune AND TENSION 6 6 4s eee es 53 37. AtTERNATING CURRENTS. 25 ee ee ee 85 38. MISCELLANEOUS APPLICATIONS © 2 6 7 1 ee ee ee 39 Expuanarion or Tasies 6 ee ee ee Tante 1 Hyeronarie Frverions TL Varnes a enn (x4 jy) TIT Varnes ov glu a WV, Van Appenpix, Historicat, ax BiBir0cRaP Se EXPONENTIAL EXPRESSIONS AS DEFINITIONS © 5. TOK eee dl eioi elie siete iel sisi eis eisiicl aa's ON CONICS. ART. 1. CORRESPONDENCE OF POI To prepare the way for a general treatment of the hype bolic functions a preliminary discussion is given on the relations between hyperbolic sectors. The method adopted is such as to apply at the xame time to sectors of the ellipse, including the circle; and the analogy of the hyperbolic and circular functions will be obvious at every step, since the same set of he read in connectic th with atians thy hyperbola with the theury of e species, sc cllipset Ht fs convenient to beg correspondence of points on two central conics of lil i.e. either both ellipses or both hyperbol: To obtain a definition of corresponding points, let O,4,, 0,8, be conjugate radii of @ central conic, and O,d,. OB, conjugate radii of any other central conic of the same species} let #,, 2, be two points on the curves; and let their coordi- nates referred to the respective pairs of conjugate directions be (2,,.9,): (41,34); then, by analytic geometry, aie oy fear Sateen G at 6 al by had # The hyperdotie functions are nox so named on account of any anaiogy with wnat are termed Eihptic Functions,“ Ine elitie integrals, and thence the elliptic functions, derive their name from the early attempts of mathemati cians at the rectification of the ellipse... . To a certain extent this is a disadvantage: . . . because we employ the name hyperbolic function to de- note cosh w sinh 1, etc., by analogy with which the elliptic functions would be merely the circular functions cos @, sin , etc...” (Greenhill, Elliptic Functions, p. 175.)8 nyPRRn LAC KUNCTIONS, Now if the points P,, P, be so situated that 2 oe es a ee mes 0 the equalities referring to sign as well as magnitude, then P,, P, are called corresponding points in the two systems. If Q, angle P.0,Q, are said to correspond respectively with the sector and tangle 2,0,Qy These definitions will apply also coincide, the points 7, , 7, being then referred irs of conjugate diameters of the when the con to any two p ame conic, In discussing the relations between corresponding areas it is convenient to adopt the following use of the word “ measure” the ratio which it bears to the constant area of the triangie formed by two conjugate diameters of the same conic. Vor example, the measure of the sector 4,0,2, is the ratio sector A,0,P,_ ngle A,OB,AREAS OF CORRESPONDING SECTORS. 9 and is to be regarded as positive or negative according as A,0,P, and ,0,B, ave at the same or opposite sides of their common initial line. ART. 2. ARFAS OF CORRESPONDING TRIANGIES. ‘The areas of corresponding triangles have equal measures, For, let the coordinates of P., Q, he (x,,9,), (2,!. 9," and lot spomdents 2. Oy be (ys jh Gayl lh triangles #,0,Q,, /0,U, be 7,, 7, and let the measur angles ,0,8,, 4,0,B, be K,, K,, and their angles m,, a; then, by analytic geometry, taking account of both magnitude and direction of angles, areas, T, _ Mer! —x/p) sin oy K, 32,0, sin w, Ty Vey an) sin, K 42,0, sin @, Therefore. by (2). Arr. 3. Angas or Conresronpine Se The areas of corresponding sectors have equal measures. For conceive the sectors S,, S, divided up into infinitesimal corresponding sectors; then the respective infinitesimal corre- sponding triangles have equal measures (Art. 2); but the given sectors are the limits of the sums of these infinitesimal triangles, hence SS KrkK @) In particular, the sectors 4,0,?,, 4,0,P, have equal meas- ures; for the initial points 4,, 4, are corresponding poi Ti may be proved conversely by an obvious seductiv ad absurdum that if the initial points of two equal-measured sectors correspond, then their terminal points correspond. Thus if any radii O,A,, O,A, be the initial lines of two equal.measnred sectors whose terminal radii are O,P., O,P,,10 HYPERBOLIC FUNCTIONS. then P,, P, are corresponding points referred respectively to the pairs of conjugate directions O,4,, OB, and O,Ay Oy that is, Prob. 1. Prove that the sector #,0,@, is bisected by the line joining O, to the mid-point ot #”,Q,. (Keter the points /,, Q,, re spectively, to the median as common axis of x, and to the two opposite conjugate directions as axis of y, and show that P,, Q, are then corresponding points.) Prob, 2, Prove that the measure of a ci ar sector is equal to the radian measure of its angle. Mob. 3. Find the measure of an elliptic quadrant, and of the sectur included by conjugate radii, Ant, 4 CHARACTERISTIC RATIOS OF SECTORIAL Measures. jel io the tangent at vi a,,and the conjugate radius ne Sylh, 20 called the charace sof the given sectorial measure S//K,. These teristic rat ratios are constant both in magnitude and sign for all sectors of the same measure and species wherever these may be ated (Art, 3). Hence there exists a functional relation be- tween the sectorial measure and each of its characteristic ratios, Art. 5, Ratios EXPRESSED AS TRIANGLE-MEASURES. The triangle of a sector and its complementary triangle are measured by the two characteristic ratios. For, let the triangle A.0.P, and its complementary triangle P.0,B, be denoted by (5) 46,2, Bi K,~ Jad sin 0, ~ a,FUNCTIONAL RELATIONS FOR ELLIPSE, n ArT. 6. FUNCTIONAL RELATIONS FoR ELLIPSE. ‘The functional relations that exist between the sectorial ratios are the same measure and each of its characteri for all elliptic, ine m cluding circular, sec- ~~ tore(Art 4) Let PB, PA ! ly [7 2, be conespond Lay points on an ellipse “ L and a circle, referred 0, ji rections O,d,, 0,B,, and O, A,.0,R,, the latter pair being at right angles; let the angle 1,0,?, — 4 in radian measure; then ae (©) (7) Prob. 4. Given x = hau; find the measure of the elliptic sector A,0,P1, Also tind its area when a, = 4, 6, = 3, @ = 60°. Prob. 5. Find the characteristic ratios of an elliptic sector whose measure is 17. Prob. 6. Write down the relation between an elliptic sector and its triangle. (See Art. 5.) Art, 7, FUNCTIONAL RELATIONS FOR HYPERBOLA, The functional relations between a scctorial measure and its characteristic ratios in the case of the hyperbola may be written in the form and these express that the ratio of the two lines on the left is acertain definite function of the ratio of the two areas on the right. These functions are called by analogy the hyperbolic2 HYPERBOLIC FUS TIONS. cosine and the hyperbolic sine. Thus, writing u for S,/K, the two equations cosh 1, sink w 8) 4, serve to define the hyperbolic cosine and sine of a rial measure 1; and the hyperboli iven secto- aud cosccant aie then defined as follows: inh w cosh tanh = coth i= Oh cosh a sinh w’ (9) The names of these functions may be read “h-cosine,” or ‘“hyper-cosine,” ete. (See “angloid” or ‘hyperbolic angle,” p. 73.) Apr & RELATIONS aMaNG HyprRroric Friverions Among the six functions there are five independent reia- tions, so that when the numerical value of one of the functions fs given, the » Four of these relations consist of the four defin fifth is derived from the equation of the hyperbola giving cosh" — sinh" # (10) By a combination of some of these equations other subsidi. ary relations may be obtained; thus, dividing (10) successively hx, and applying (9), give 1 —tanh’ w= sech* x, by cosh* 1, value of any fui when tanh # is given, coth w sech w= 4/1 —tanh'w, tanhRELATIONS BETWEEN HYPFRBOLIC FUNCTIONS. 13 l tanh a cosh w= » sinha = Via tanita Vr tana tanh a escha = ¥ © 3q) ually 5 ay wing considerations: The functions be removed by the f cosh, sech# are always positive, because the primary char- nd ic ratio x,/2, is positive, since the initial line Qj, 3 actel the abscissa O,J/, are similarly directed from Q,, on which: ever branch of the hyperbola ?, may be situated; but the func- tions sinh #, tanh w, coth a, esch w, involve the oth charac teristic Jr and 6, have the same or opposite signs, ic e or negative: hence these four fune all positive or all negative. ‘Thus when any one of the fune- tions sinh, tanh x, esch #, coth 7, is given in magnitude and . there is no ambiguity in the value of any of the six atio y,/6,, which is positive or negative according as ay the measure ther jons are wis pos given, there is ambiguity as to whether the other four functions shall be all positive or all negative. ‘The hyperbolic tangent may be expressed as the ratio of two lines. For draw the tangent : line AC =; then jp a nhnst tat t WA tanh weZ rah 5 a (12) & tt The hyperbolic tangent is the measure of the triangle 0.4C. For OAC _a_t iB 3 (23) OAB™ ab 6 Thus the sector AO/, and the triangies A0P, 708, AOC, are proportional to #, sinh 1, cosh x, tanh x (eqs. §, 13); hence sinhu > «> tanhy. (14)u HYPERBOLIC FUNCTIONS. Prob. 7. Express all the hyperbolic functions in terms ot sinh w. Given cosh # = 2, find the values of the other functions. Prob. 8, Prove from eqs. 10, 11, that coshw> sinh #, coshu>1, tanh a <1, ech <1. b. 9. in the figure of AU. 1, tet O42, OF —1, AOS = 60", and area ot sector 40/' = 3; find the sectoriai measure, and the two characteristic ratios, 1n the elliptic sector, and also in the hyper bolic sector; and find the area of the triangle AOP. (Use tables of cos, sin, cosh, sinh.) Prob, 10. Show that coth m sech w, esch w may each be ex- pressed as the ratio of two lines, as follows: Let the tangent at make on the conjugate axes OA, OB, intercepts OS = m, OT =n; let the tangent at &, to the conjugate hyperbola, meet OP in &, making BR = /; then cotha = Yu, sechu = m/a, eschu = n/b, Prob. 11, ‘The measure of segment AA/P is sinh w cosh u — w. Modify this for the ellipse. Modify also eqs. 10-14, and probs. Since the values of the hyperbulic fu tions depend only on the sectorial measure, it is convenient, in tracing their varie ations, to con half of aq conjugate radii prineip: line of all the sectors, The sectoral ider only sectors of one whose tangular byperbo are equal, and to take the axis OA as the common initial measure assumes every value from — 0, through 0, to -+ 0, as the terminal point P comes in from infinity on the lower branch, and passes to infinity on the upper swings irom the iower asymptotic posi- tion y= — 2. to the upper one, y= x, It is here assumed, but is proved in Art. 17, that the sector AOP becomes infinite as P passes to infinity. Since the functions cosh m, sinh #, tanh m, for any positionVARIATIONS OF THE HYPERBOLIC FUNCTIONS, 16 of OP, are equal to the ratios of , y, #, to the principal radius 4, it is evident from the figure that sinho=o0, tauho (15) positive i are positive and become infinite, but tanh approaches unity iy, conte, 9 cosh =, sinhoo =o, tanh o (06) Again, as ~ changes from zero towards the negative side, cosh positive and increases from unity to infinity, but sinh « is negative and increases numerically from zevo to a negative infinite, and tanh w is also negative and incteases numerically from zero to negal e unity; hence nh (— ¢) cosh (— 00 =e, sinh(— 0) =— 0, t —1.(17) For intermediate values of « the numerical values of these are tabulated at the end of tit their manner of vai n can be obt Art. 25, in which the sce abscissa, and the values of the functions cosh 1, sinh 1, ete., are represented by the ordinate. chapter, A generat ned from the curves in ial measure # is represented by the ‘The relations between the functions of — # and of u are evident from the defini Thus jons, as indicated above, and in Art. 8. cosh (— 0) = cosh u, sinh (— ) = — sinh », sech (— x) =+sech u, esch(— x)= —esch m } (18) tanh (— 2) = —tanh x, coth (— 1”) = — coth n. Prob. +9. ‘Trace the changes in sech 1, cath », sch w, a6 1 pacsoe appear 4, cosh w are infinites of an order infinitely higher than the order of u.) Prob. 13. Applying eq. (12) to figure, page 14, prove tanh m, = tan 40P.16 HYPERBOLIC FUNCTIONS Art. 10. ANTI-HYPERBOLIC FUNCTIONS. The equations = = cosh uv, % = sinh », 5 = tanh 4, ¢ten z may also be expressed by the inverse notation ~ = cosh t ye tanh“'5, etc, which may be read: “x is 4 = sinh” is the ratio x to the seciorial measure wituse hyperboiie cosis a," etc.; or “nis the anti-h-cosine of x/a,” etc. Since there are two values of u, with opposite signs, that correspond to a given value of cosh 1, it follows that if « be determined from the equation cosh w = m, where m is a number greater than unity, 7 is a two-valued function of m. The symbol cosh! m will be used to denote the po: of 1 that satisfies the equation cosh «=m. Similarly the symbol sech-!y will stand for the positive value of w that satisfies the equation sech # =m, The signs of the other functions sinh-'m, tanh“, coth~! m, sch! m, are the same ve value of reai numbers are one-vaiued. Prob. 14. Prove the following relations : cosh-'m = sinh“! 7m? — 1, sinh“! = 4 cosh-' Wm" | a, 1¢ upper or lower sign being used according as m is positive or negative. Modify these relations for sin“t, cos". Prob. 15, In figure, Art. 1,let OA = 2, OB = 1, AOB = 60°: find the area of the hyperbolic sector AOP, and of the segment AMP, if the abscissa of P is 3. (Find cosh“! from the tables for cosh.) Arr, 11, Functions or Sums anp DIrreRENcEs. (a) To prove the difference-formulas sinh (v — 2) = sinh «cosh vy — cosh sinh) = ¢ Oy cook (# Let OA be any radius of a hyperbola, and let the sectors AOP, A0Q sector QOP. Let OB, O9' be the radii conjugate to OA, 0; and let the codrdinates of P, Q, Q! be (#,,y,). (#9) (2.9) with reference to the axes O4, OB; thenFUNCTIONS OF SUMS AND DIFFERENCES. " sinh (u —v) = sinh S80t0# GOP _ triangle QOP [Art 5, RE x Ary,—xy)sino _ yx ya, ir ha, By alin ar sinh w cosh 7 — cosh # sinh \lecf Ee Pm > y cosh (1 ~ 2) = cosh 08 OOP triansle P00’ race, 5 We, aya but (20) since Q, Q! are extremities of conjugate radii; hence cosh (w — 7) = cosh w cosh 7 — sinh w sinh v, In the figures w is positive and v is positive or negative. Other figures may be drawn with # negative, and the language the text will apply to all. In the case of elliptic sectors, similar figures may be drawn, and the same language will apply, except that the second equation of (20) will be x’/a, = —¥/b; therefore sin (w — v) = sin # cos v — cos w sin % cos(u — 2) = cos w cos v-+sin usin sinh (w+ 0) = sinh w cosh v + cosh u sinh v,) cosh (w+ 0%) = cosh # cosh v + sinh u sinh ny These equations follow from (19) by changing 7 into — (21)1B HYPERBOLIC FUNCTIONS, and then for sinh(—2), cosh(— 2), writing — sinh », cosh o (Art. 9, eqs. (18). (©) To prove that tanh (w +t v) = tanh w+ tanh » Tdtonhwtenne 2) v Writing tanh (wv) = tained, Prob. 16. Given cosh # = 2, cosh # = 3, find cosh (u-| v)- Prob. 17. Prove the tollowing identities: 1. sinh av == 2 sinh w cosh 2. cosh 24 = cosh" + sinh" = 1-4 2 sinh w= 2 cosh? a — x 3. tb eosh w= 2 cosh? du, cosh u— 1 = 9 sinh? dy sinh «cosh w—1_ (cosh w —1\¢ 4 tanh fe = Tae ne = SS} . 2 tanh # ae — LE tanh! a mi i tanh w nie aa oe ea 6, sinh qu = 3 sinh # + 4 sinh’ «, cosh au = 4 cosh" —3 cosh u. rttanh de 7. cosh wf sinh w= MS Se 8. (cosh w+ sinh «)(cosh v + sinh 7) cosh (w+ v) + sinh (w+ 1). g. Gener: (8); and show also what it becomes when 10, sinh'x cos'y + cosh'x sin'y = sinh'x + sin’ 11. cosh"! + cosh“! = cosh![ma + V Ga? — 1) Ga@—7) |. [nVito tnd i pm} Prob. 18. What modifications of signs are required in (21), (22), in order to pass to circular functions? Prob. 19. Modify the identities of Prob. 17 for the same purpose. ae me sinh tn = Arr. 12. Conversion Formutas. To prove that cosh #.+ cosh u, = 2 cosh iu, ,) cosh 4 cosit m,— cosh m, = 2 sinh gu, + 1,) sinh 4(u,— 1), b es sinh ,--sinh 1, = 2 sinh 41, + 1) cosh 4x, — 1), sinh #, — sinh 4, = 2 cosh #(1, +1, sink 4, — 1%) |LIMITING RATIOS. 19 From the addition formulas it follows that cosh (w+ v)-+ cosh (w — 2) cosh (w+ 2) — cosh (w — 2) = 2 si = 2 cosh w cosh v, hw sinh 2, ty w= Kn, tH, H(u,— 1), these equations take the form required. Prob. 20. In passing to circular functions, show that the only modification to be made in the conversion formulas is in the alge- braic sign of the right-hand member of the second formula, cosh 24 + cosh cosh 24 + cosh 49, sinh auf sinh 42 cosh 24 — cosh 40 Prob. ar. Simplify Prob, 22. Prove sinh'x — sinh* y = sinh (x +y) sinh (x —y). Prob. 23. Simplify cosh*x cosh’ y + sinh*e sinh’y, Art, 13, LIMITING RaTios. To find the limit, as # approaches zero, of sinh tanh a u u which are then indeterminate in form, By eq. (14), sinh w > «> tanh w; and if sinh w and tanh w be successively divided by each term of these inequalities, it follows that sinh « 1 < S22" < cosh u, ” tanh « sech x < but when #0. cosli w = 1. sech w= 1. hence lim, tanh _ a20 HYPRRROLIC FUNCTIONS, Art, 14, Derivatives oF HYPERBOLIC FUNCTIONS. To prove that (@) ] @ | © Aeaoh) = sech* m, asech w) . (ay S80 oe — sech w tanh © ei x escn'y, (7) SEEN) exch w coth (a) Let y = sinh x, inh (1 4 An) — si = 2 cosh (24+ 4u) sinh 44%, 4 sinh 4.40 Fi = cosh (ut bay Take the limit of both sides, as 4u + 0, and put 4y _ dy _ d(sinh ») im. Au” du du Tim. cosh (u + $41) = cosh #, sinh 44a Jim Ge St Gee Art. 13) ne, HAsinh 2) i then SSE — cost a. (6) Similar to (2). di’ cosh & cosht « —DERIVATIVES OF HYPERBOLIC FUNCTIONS. at (@) Similar to (6). Asechu) _ d 1 __ sinha Oia du" coshu ~~ cosh’ w (7) Similar to (¢). tus appears ihai ihe funciions sinh x, cosh x reproduce themselves in two differentiations; and, similarly, that the —sech u tanh». ifferentiations. In this connection it may be noted that the frequent appearance of the hyperbolic (and circular) functions in the solution of physical problems is chiefly due to the fact that they answer the question: What function has its second derivative equal to a positive (or negative) constant multiple of the function itself? (See Probs. 28-30.) An answer such as cosh ms is not, however, to be understood as asserting that ‘mz is an actual sectorial measure and y its characteris same as the known reiation between the measure of a hyper~ bolic sector and its characteristic ratio: and that the numerical value of y could be found from a table of hyperbolic cost Prob. 25 Show that for circular functions the only modifica- tions required are in the algebraic signs of (0), (). Prob. 26. Show from their derivatives which of the hyperbolic and circular functions dimi as w increases, Prob. 27. Find the derivative of tanh w independently of the derivatives of sinh u, cosh «. Prob. 28, equation y= 4 cosh mx + B ate the constants by differentiation from the hh mx, and prove that a'y/dx* = m'y. Prob. a9. Eliminate the constants from the equation y= A cos mx + B sin mx, and prove that d *y/dx* = — my,22 HYPERROLIC FUNCTIONS, Ant. 15, DERIVATIVES OF ANTI-HYPERDOLIC FUNCTIONS. asin @ de vy cosh" x) | “i dz I vo ditanh—" 2) ! Cl areca (26) udu=Vi-a du, duxdxj¥i4+x. (6) Similar to (@). (9 Let w= tanh“, then 2 = tanh a, dx = sech’ udu = (1 — tanh" edu = (1 — 2d, du = ae /1EXPANSION OF HYPERBOLIC FUNCTIONS, 23 Prob. 32, Prove @ sinh”? Prob. 34. When tanh“ is real, prove that coth™? « is imagt nary, and conversely; except when x = 1. sinh-'x cosh~* x ob. 35. Eval seer Prob. 35, Evaluate “Tor a > when x = ArT. 16, EXPANSION OF IlyPERBOLIC FUNCTIONS. For this purpose take Maclaurin’s Theorem, Su Ko) + Hf'(0) + 18°70) + PO) Peer and put f(x) =sinh x, (2) = cosh xz, f"(u) = sinh #645 then flo) =sinho=0, (0) =cosho=1,..5 hence sinh w Sub ae tS tes (27) ilarly, or by differentiation, cosh wtb pat Dt : (28) By means of these series the numerical values of sinh x, cosh u. can be computed and tabulated for successive values of ot , because the ratio of the uth term to the preceding is in the fret case w*/(9" — 1\(ou — 9), and in the eecond #*/(an — 2'2n — 3), both of which ratios can be made less than unity by taking # large enough, no matter what value « has, Lagrange’s remainder shows equivalence of function and series.24 HYPERROLIC FUNCTIONS, From these series the following can be obtained by divisio tanh w= — gt gyn — aN fo, sech w= 1 — du} gut — yy 4. (20) u coth w= 1 — gy + a in observable law in the coefficients, and as the functions tanh x, sech 1, coth 1, csch u, can be found directly from the previously computed values of cosh 1, sinh Prob. 36. Show that these six developments can be adapted to the circular functions by changing the alternate signs. Art. 17, ExronentiaL EXPRESSIONS, Adding and subtracting (27), (28) give the identities cosh # — sinh « hence cosh w = Ae e*), oe tanh w= The analogous exponential expressions for sin 4, cos « are v¥—1) cos n= ete), sin w= Het ot), € «the result uf substituting wi fur + in the exponential development This will be more fully explained in treating of complex numbers, Arts. 28, 29EXPANSION OF ANTI-FUNCTIONS, 25 Prob. 37. Show that the properties of the hyperbolic fun could be placed on a purely algebraic basis by starting with cqua- tions (30) as their definitions ; for example, verify the identities : sinh (—u) = — sinh , cosh (— x) = cosh #, cosh’ w—sinh* w=1, sinh (4-+t) = sinh w cosh » + cosh # sinh 2 "(cosh mu at (sinh mu) ae a Prob. 38. Prove (cosh w+ sinh #)" = cosh mu + sinh nu, Prob. 39. Assuming from Art. 14 that cosh w, sinh w satisfy the differential equation d'y/du" =x, whose general solution may be written y — det-+ Be™*, where 4, B are arbitrary constants ; show how to determine 4, B in order to derive the expressions for cosh #, sinh w, respectively. [Use eq. (15)-] Prob. 40. Show how to construct a table of exponential func- tions from a table of hyperbolic sines and cosines, and rice versa, Prob. 41. Prove = log, (cosh w+ sinh 1), Tob. qa. Show that the area of any hyperboti wien its terminai iine is one of the asympivtes. Prob, 43. From the relation 2 cosh # = e* +e prove 2°-(cosh u)"=cosh au-+ cosh (n—2)u-+4n\n—1) cosh (4—4)u+.. and examine the last term when # is odd or even, Find also the corresponding expression for 2°-* (sinh u)". ART. 18. EXPANSION OF ANTI-FUNCTIONS, ; ian Since S80") Tae (hey hence, by integration, Q1) sinh 4 =x — the integration-constant being zero, since sinh~' x vanishes with x. ‘This series is convergent, and can be used in compu-6 HYPERBOLIC FUNCTIONS, sation, only when # <1. Another series, convergent when © > 1, is obtained by writing the above derivative in the form Asin) erp yet Arp 4) de where C is the integration-constant, which will be shown in Art. 19 to be equal to log, 2 A development of similar form is obtained for cosh x; for acosh™ x) _ yal Tea y rence sosh~" x = Clog x—+,—+ n which C is again equal to log, 2 [Art. 19, Prob. 46). In wrder that the function cosh~'z may be real, x must nat he ess than unity; but when # exceeds unity, this series is con- rergent, hence it is always available for computation, Sit tattatten Again, Fram (32), (32). (34) are derived : sech~! x = cosh + ¥ = Clog x —LOGARITHMIC EXPRESSION OF ANTICFUNCTIONS, a7 P3t 13s 1 esch”’ # = sinh™' — a (37) Frob. 44. Show that the series for tanh ‘x, cov +x, sech ‘x, are always available for computation. Prob. 45. Show that one or other of the two developments of the inverse hyperbolic cosecant is available. Art, 19, Loganiriic Expression or Anti Functions. Let #=cosh#, then sx therefore — cosh wfsinh w= 4, and = log (e+ ¥2'— 1). (38) Similarly, = log (4 + ¥° +1). (30) og LEVT 40) a log LEE (40) i Sn ee esch"ty = sinh" = log HEEL ES (41) Again, let therefore 14s, tanh“te=$ | 3 fanh“te=$ log 23 (4) > (43) rob. 40. Show trom (38), (39), that, when sinh“'x — log +log 2, cosh" log and hence show that the integration-constants in (32), (33) are each equal to log 2.28 HYPERROLIC FUNCTIONS, Prob. 47. Derive from (42) the series for tanh ‘x given in (34). Prob. 48. Prove the identities: =sinh" 4-2 ")=cosh Het); log sec x = 2 tanh“ tan" 4x; log ese. x = 2 tanh’ ' tan(da + 42); The correspandence of sectors af the same species was dis- cussed in Arts, 1-4. It is now convenient to treat of the correspondence that may exist between sectors of different specie: Two points P, P,, on any hyperbola and cllipse, are said to correspond with reference to two pairs of conjugates 0,4, O,B,, and 0,A,, 0,B,, respectively, when 4/4, = O/T y (44) and when y,.¥, have the same sign. The sectors 4,0,P. said te P Thas comesponding sectors of central conics ot different species are of the same sign and have the primary characteristic ratios reciprocal. Hence there is a fixed functional relation between their re~ spective measures, The elliptic sectorial measure is called the gudermanian of the corresponding hyperholic sectorial measure, and the latter the antigudermanian of the former. This relation is expressed by SSK, = gd S/K, or v=gdn, and (45) Art. 21. CrRcuran FUNCTIONS OF GUDERMANIAN. The six hyperbolic functions of u are expressible in terms he si gud for oi 2 in which m, 0 are the measures of corresponding hyperbolic and elliptic sectors,GUDERMANTAN ANGLE, 29 hence cosh =secv, (eq. (44)] sinh w= Vsecu—i = tan», tanh w= tan v/sec v = sin v, / 9) sech u = cosv. esch u = cot 9, The gudermanian is sometimes useful in computation ; for instance, if sinh be given, can be found from a table of natural tangents, and the other circular functions of v will give the remaining hyperbolic functions of x, Other uses of this function are given in Arts, 22-26, 32-36. Prob. 49. Prove that gd” — sec~'(cosh #) — tan-sinh u) cos '(sech u) = sin“ tanh 1), cosh” "(sec 0) = sinh“! (tan 2) Prob. 50. Prove gd~ = sech"(cos 2) = tanh “sin 0). —4s, si =)—— 4s, gd o = 0, gd“ (4) =e, gd —d2)= Prob 52. Show that gd # and gd-"v are odd functions of 4, &. Prob. 53. From the first identity in 4, Prob. 17, derive the rela- tion tanh du = tan 40. Prob. 54. Prove tanh” (tan #) = 4 gd 2w, and tan“ (tanh x) = $ gd-"2x, ArT. 22. GUDERMANIAN ANGLE If a circle be used instead of the ellipse of Art. 20, the gudermanian of the hyperbolic sectorial measure will be equal to the radian measure of the angle of the corresponding circular 24. (6) and Art. 3, Prob. 2). Th sector ( calied the gudermanian angle; but the gudenmanian function v, as above defined, is merely a number, or ratio; and this number is equal to the radian measure of the gudermanian angle 6, which is itself usually tabulated in degree measure ; thus O=10%r 2 ee. GD)30 HVPRRROLIC FUNCTIONS, Prob. 55. Show that the gudermanian angle of # may be construct- 2d as follows: ‘Take the principal radius OA of an equilateral hyperbola, as the & initial line, and OP as the terminal line, of the sector whose measure is 4; 7 | from M, the foot of the ordinate of / | P. draw MT tangent to the circle Ar / | Prob. 56. Show that the angle @ never exceeds go”. Prob. 57. ‘The bisector of angle 407’ ° * "A" pisects the sector AVP (see Prob. 13, Art. 9, and Prob. 3, Art. 21), and theline AP, (See Prob. 1, Art. Prob. 58. ‘This hisector is parallel to 7P, and the points 7; P are in line with the point diametrically opposite to A. Prob. sy. The tangent at ? passes through the foot of the ordinate of 7, and intersects 7Af on the tangent at A. Prob. 60. ‘The angle APA is half the gudermanian angle. e N AND INVERSE, vsgda, u=gd-2, sec v tan vd = sinh udu, sec tdu — du, therefore digd-* v) = sec vdv. 48) Again, dv = cosvdu = sech u du, therefore digd u) = sech u du, (49) Prob. 61. Differentiate: y = sinh w — gd wy, y =sinu + gdy, tanh w sech w+ gd, y= tanvseco + gdv, y This angle was called by Gudermann the longitude of w, and denoted by Jae “His inverse symboi was %i; thus w = i(‘u). (Creiie’s journai, voi. 6, 1830.) Lambert, who introduced the angle 8, named it the transcendent angle. (Hist. de Yacad. roy de Berlin, 1761). Hotel (Nouvelles Annales, vol. 3, 1863) called it the hyperbolic amplitude of w, and wrote it amh 1, in analogy with the amplitude of an elliptic function, as shown in Prob, 62, Cayley (Elliptic Functions, +876) made the usage uniform by attaching to the angle the name of the mathematician who had used it extensively in tabulation and in the theory of elliptic functions of modulus unity.SERIFS FOR GUDERMANIAN AND ITs INVERSE, at Prob. 62, Writing the “elliptic integral of the first kind” in the form being called the modulus, and @ the amplitude; that ¢ — am a, (mud. 9), show that, in the special case when « cos. am w = sech w, tan am w= sinh u; and that thus the elliptic functions sin am », ete., degenerate into the hyperbolic functions, when the modulus is unity." Art, 24, SERIES FOR GUDERMANIAN AND ITS INVERSE. Substitute for sech , sec » in (49), (48) their expansions, Art. 16, and integrate, then gd uu — dat shat — aha’ be (60) Bota taet teil te. Gt No constants of integration appear, since gd # vanishes with wand gd'e co! sand gate n co! tation, as gd u is best found and tabulated by means of tabies of natural tangents and hyperbolic sines, from the equation gd w= tan“(sinh «), and a table of the direct function can be used to furnish the numerical valucs of the inverse function ; or the latter can be obtained from the equation, gd~'v = sinh~“(tan v) = cosh~"(sec 2), To obtain a logarithmic expression for gd“"v, let gdv=x, v=gdu, 10, (mod. 1), led Hotel to name the function gd ty The relation gd the hyperbolic ampiiwude of ‘connection Cayley expressed the functions tan #, sechw, sinh w in tne torm sin gd w, cos gd u, tan gd u, and wrote them sg u, cg #, 1g m, to correspond with the abbreviations sn, cn, daw for sin am , cos amu, tan amu, Thus tanh # = sg 4 = sn u, (mod. 1); etc. It is well to note that neither the elliptic nor the hyperbol’c functions .d their names on account of the relation existing between them in a ease. (See foot-note, p. 7) wisee mote, Ant. 22}. i ti32. HYPERBOLIC FUNCTIONS, therefore sec v= cosh x, tan v= sinh #, secv-+tanv = cosh w+ sinh u =e, tabsin v1 — 005 +9) tan Gt do), cos D sin ($7 + ¥) ym gd ‘0, em logytan (ba | de) (32) ea Pie Gy Bvatuste HH #] , =P] Prob. 64, Prove that gd « — sin # is an infinitesimal of the fifth order, when «= 0. Prob. 65. Prove the relations dep des ane, de — bos tate, Art, 25, GRAPHS OF HYPERBOLIC FUNCTIONS, Drawing two rectangular axes, and laying down a series of points whose abscissas represent, on any convenient scale, suc- represent, preterably on the same seal , the corre. sponding values of ihe function to be plotted, the locus traced out by this series of points will be a graphical representation of the variation of the fune- tion as the sectorial meas-GRAPHS OF THE HYPERROLIC FUNCTIONS. 33 ure varies. The equations of the curves in the ordinary carte sian notation are: Fig. Full Lines. Dotted Lines. A Jy = cosh x, y = sech x; Boo y=sinhz, y= sche; c y= cot z; D Here x is written for the scctorial measure #, and y for the numerical value of cosh #, ete. It is thus to be noted that the variables 2, y are numbers, or ratios, and that the equation y = coshx merely expresses that the relation between the numbers # and y is taken to be the same as the relation be- tween a sectorial measure and its characteristic ratio. The numerical values of cosh x, sinh x, tanh x are given in the tables at the end of this chapter for values of « between o and 4. For greater values they may be computed trom the devel- onments of Art. 16. ‘The curves exhibit graphically the relations: 1 sech « , eschtu=—-—, cothn =; coshra sinh tanh w cosh uit, sech ast, tanhast, gdu 0%; a trent” (2-2) ] =coth~ '2—coth-"3 5)—tanh~"(.3333): —tanh-"(x >: anh: de 43 $494—.3466=.2028.* tanh~'o—tanh~(.5) = — 5494. (By interpreting these two integrals as areas, show graph- ically that the first is positive, and the second negative.) a ee ee & S Gx) Vab van Va-e For tanh=1(.5) interpolate between tanh (.54) — .4930, tanh (.56) = .5080 (see tables, pp. 64, 63); and similarly for tanh-1 (.3333).RLEMENTARY INTEGRALS, a7 the real form to be taken, (Put # — or coth’ Vv¥b—a the real form to be taken, (a — a’) — yy fe — ales = By means of a reduction-formula th integral is easily made to depend on 8. It may also be obtained by transforming the expression into hyperbolic functions by the assumption x =a cosh x, when the integral takes the form of sink! ndu= 5 J(cosh 20 — 1\du = “o'(sinh 24 — 2) dat which gives 17 on replacing @ cosh a by x, and a sink w by @ The geometrical interpretation of the result is evident, as it expresses that the area of a rectangular-hyper- bolic segment AMP is the difference between a triangle OP and a sector OAP. 8. fe — hie = se coc 3 sin 19. fet pat de = Sale + ath + Let sin, 20. Jvc odes =fo +b tan’ pd tan =f tan Ali t tan’ A) 4 4 sinh (tan A) inh w cosh w—x), 2 Prob. 71. What is the geometrical interpretation of 18, 19? Prob. 72. Show that,f"(ax* + aby + e)ldx reduces to 17, 18, 19,aR HYPERBOLIC FUNCTIONS. respectively: when a is positive, with ac <8 and when a is positive, with a¢ > 5". Prob. 73. Prove S sinh w tanh «de 3 when ais negative; sinh u— gd, Jf ‘cosh w coth w du = cosh « + log tanh 4S Prob. 74. Integrate (xt +axt Vix. (xt 4+axt Nd. (x* + ax + Vex. © parabula x? — apy i measured from the vertex, and @ the angle which the tangent line makes with the vertical tangent, prove that the intrinsic equation of the curve is ds/dp = 2f sec’ p, s = psec ptan b+ pgd"'d, Prob. 76. The polar equation of a parabola being r = @ sec’ 46, referred to its focus as pole, express s in terms of @. Prob. 77. Kind the intrinsic equation of the curve y/a = cosh x/a, and of the curve y/a = log sec */a, Prob. 78. Investigate a formula of reduction for f‘cosh*« ds; also integrate by parts cosh'x dx, tanh'x dx, (sinh“!x)'de5 and xd show that the ordinary methods of reduction for /f cos"=: ah" x de, can be applied taf cosh + plied taf ART. 27. FUNCTIONS OF COMPLEX NUMBERS. As vector quantities are of frequent occurence in Mathe- matical Physics; and as the numerical measure of a vector in terms of a standard vector is a complex number of the sorm x+y, in which x, y are real, and # stands for Y¥—1; it becomes necessary in treating of any class of functional oper- ations to consider the meaning of these operations when per- formed on such generalized numbers." The geometrical defini- tions of cosh, sinh x, given in Art. 7, being then no longer of the symbols of the symbols ‘The u ¢ of vectors in electrical theory js shown in Bedell and Crehore's Alternating Currents, Chaps. x1v-xx (Grst published in 1892). The advantage of introducing the complex measures of such vectors into the differential equa- tions is shown by Steinmetz. Proc. Elec. Congress, 1893: while the additional convenience of exprescing the xalution ia hyperbolic functions of these complex numbers is exemplified by Kennelly, Proc. American Institute Electrical Engincers, April 895. (See below, Art. 37.)FUNCTIONS OF COMPLEX NUMBERS, 39 cosh (x + #y), sinh (x + #y), a suitable algebraic meaning, which should be consistent with the known algebraic values of cosh x, sinh x, and include these values as a particular case when y =0. The meanings assigned should also, if possible, be such as to permit the addition-formuias of Art. 11 to be made general, with all the consequences that flow from them, Such definitions are iurnisied by the algebraic deveiop- ments in Art. 16, which are convergent for all values of #, real orcomplex. ‘Thus the definitions of cosh (x-+ #y),sinh (x -} é) are to be cosh (FO) =I ETA TET) bee (52) sinh (x ++ iy) e+e He tart From these series the numerical values of cosh (x + iy), inh (v4.43) could he computed ta any degree of tion, when 4 aud y ae given, In general out in the complex form* cosh (x + iy) = a+ 16, sinh (#4) = e+} td. The other functions are defined as in Art. 7, eq. (0). Prob. 79. Prove from these definitions that, whatever w may be, cosh (4) = cosh x, sinh (~ #) = — sinh », a ee ee Joeosh mu =m? cosh mu, 7 sinh meu = mt sinh mat IL 1s to be borne in mind that the symbols cosh, sian, nere stand for alge- braic operators which convert one number into another; or which, in the lane Buage of vector-analysis, change one vector into another, by stretching and turning. + The gener matical Physics as the solution of the differential equation d'¢/du where sp, m, w are complex numbers, the measures of vector quant Ant. 97.) fed hyperbolic functions usually present themselves in Mathe- mp, (See40 HVPRRROLIC FUNCTIONS, Art. 28. AppiTion-THEOREMS FoR COMPLEXES. The addition-theorems for cosh (w+ 0), etc., where % v are complex numbers, may be derived as follows, First take #9 as real numbers, then, by Art. 11, This equation is true when #, 7 are any real numbers. It must, then, be an algebraic identity. For, compare the terms of the 7th degree in the letters 2, v on cach side. Those on the left are 7(#-+ 0); and those on the right, when collected, ri form an rth-degree function which is numerically equal to the former for more than 7 values of # when v is constant, and for i t r val ant, of the iit degree on each side are algebraically identical fune- tions of v and v.* Similarly for the terms of any other degree. Thus the equation above written is an algebraic identity, and is true for all values of , v, whether real or complex. Then writing for each side its symbol, it follows that cosh (w+ 7) = cosh cosh #4-sinh w sinh v; (53) and by changing ¢ into — 2 cosh (v — 2) == cosh w cosh »— sinh w sinh 2%, (54) Ina similar manner is found sinh (vt 2) = sinh w cosh vt cosh w sinh. (58) In particular, for a complex argument, cust (2 $y) @zo)- +f two rth.degree functions of a single variahle he equal for more than values of the variable, then they are equal for all values of the variable, and are algebraically identical.”FUNCTIONS OF PURE IMAGINARIFS. 4t Prob. 79. Show, by a similar process of generalization,* that if sin u, cos «, exp «t be defined by their developments in powers of w, then, whatever w may be, (w+ 2) = sin w cos + cos # sin 2, 0 (u + 2) = cos w cos # — sin u sin 9 cap (uf &) — cap «capes Prob. 80. Prove that the following are identities: cosh” w — sinh* w= 1, cosh u4 sinh v= exp», cosh w — sinh # = exp (—1), cosh # = fexp # | exp (— )), sinh w = jfexp «—exp(—4)]. Art. 29, FUNCTIONS OF PURE IMAGINAKIES, sinh w= + 3 + get i put for # the pure imaginary zy, then cosh #y 1) at ge EOI: (57) sinh iy = iy + or + Oe ea . s =ily - Sut me | =ssiny, (58) and, by division, tanh dy = é tan y. (50) * This method of generalization is sometimes called the principle of the “permanence of equivalence of forms.” It is not, however, strictly speaking, a principle.” but a method: for, the validity of the generalization has to be demonstrated, for any particular form. hy means of the princi of the alge. braie identity of polynomials enunciated in the preceding foot-note. (See Annals of Mathematics, Vol. 6, p. 81.) } The «ymbol exp « stands for “exponential function of w,"" which is identi coal with e when 1 is real.4 HVPRRROLIC FUNCTIONS. These formulas serve to interchange hyperbolic and circular functions, The hyperbolic cosine of a pure imaginary is real, and the hyperbolic sine and tangent are pure imaginaries. The following table exhibits the variation of sinh , cosh x, tani, exp %, as 1 iakes a succession of pure imaginary vaiues, * In this table .7 is written for } 2, 707 Prob. 81. Prove the following identities : cosy = cosh y= 3fexp yt exp (— 5), siny = 5 sinh iy = [exp iy — exp (— 5)], cos y +i sin y — cosh ty + sinh iy — exp iy, xp (=o), cos iy = cosh y, sin y = sinh y. cos y — i sin y = cosh dy — sinh iy Prob, 83. Show that, in the identities (57) and (58), y may be replaced by a general complex, and hence that sinh (w & i) = 2 f sin (y T Hx),FUNCTIONS oF © + y IN THE FORM X44¥. 43 cosh (x #9) = cos (y ¥ ix), sin (x £ 9) = & ésinh (y ¥ ax), cosh (y T éx). ucteceries for cine sinh «= a(x + £( ART. 30, FUNCTIONS OF 4+ 4y IN THE Form X+4/¥. By the addition-formulas, cosh (x + 4y) = cosh x cosh fy + sinh x sinh ty, sinh (x + #y) but cosh iy = cosy, sinh iy =i sin y, inh x cosh ¢y -+ cosh x sinh iy, hence cosh (x + iy) = cosh x cos y+ sinh x sin y, } (60) sinh (x + ty) = sinh & cos y+ cosh x sin y.) e+e a=cosh xcosy, b= sinh x sin y, ius if cosh (2-5) = ab, al (1) ¢=sinh x cosy, d= cosh x sin y. From these expressions the complex tables at the end of this chapter have been computed, Writing cosh s=Z, where s— 2+ iy, Z=X+i¥; let the complex numbers 2, Z be represented on Argand diagrams, in the usual way, by the points whose coordinates are (x, 9), (X, ¥); and let the point s move parallel to the y-axis, on a given line x =m, then the point Z will describe an ellipse 1g v between the equa- jis ee (Cosh mT (sink my and which, as the parameter m varies, represents a series of confocal ellipses, the distance between whose foci is unity.44 HYPERBOLIC FUNCTIONS, ilarly, if the point x move parallel to the x-axis, on a given line y —n, the point Z will describe an hyperbula whose equi tion, obtained by eliminating the variable x from the equations X= cosh x cos n, V h x sin x, is x yo (cos n)* (sin n)* and which, as the parameter n.varies, represents a series vi hyperbolas confocal with the former series of ellipses. These two systems of curves, when accurately drawn at close intervals on the Z pline, constitute a chart of the hyper- bolic cosine; and the numerical value of cosh (m -+ i) can be read off at the intersection of the ellipse whose parameter is 1 with the hyperbola whose parameter is n* A similar chart can be drawn for sinh (x+y), as indicated in Prob. 85. Periodicity of Hyperbolic Functions—The functions sinh « and cosh « have the pure imaginary period 2iz, For sinh (u-+2iz) =sinh u cos o-+i cash usin oz—sinh u, cosh (4+ The functions sinh « and cosh » each change sign when the argument 1 is increased by the half period #x. For sinh (u+in) =sinh u cos x+i cosh u sin x= —sinh 1, cosh (u+in)=cosh u cos x+i sinh « sin x= —cosh %#, The function tanh u has the period ix. For, it follows from the last two identities, by dividing member by member, that tanh (u-+ix)—tanh 1, By a similar use of the addition formulas it is shown that sinh (u-+}in) i cosh u, cosh (u-+$in) =i sinh w. By means of these periodic, half-periodic, and quarter-periodic relations, the hyperbolic functions of «-+iy are easily expressible * Such a chart is given by Kennelly, Proc. A. I. E. E., April 1898, and is used by him to obtain the numerical values of cosh (x-+ iy) sinh (x-biv). which present themselves as the measures of certain vector quantities in the theory of alternating currents, (See Art. 37.) The chart ie constructed for values of = and of y hetween o and 1.2; but it is available for all values of y, on account of the periodicity of the functions.FUNCTIONS OF x-tiy IN THE FoRM X+i¥. 45 ‘The hyperbolic functions are classed in the modern function- theory of a complex variable as functions that are singly periodic with a pure imaginary period, just as the circular functions are singly periodic with a real period, and the elliptic functions are functions sinh~! m and cosh~! m have each an indefinite number of values arranged in a series at intervals of 2ix. ‘That par ular valuc of sinh~'m which has the coefficient of 7 not greater than 4a nor less than —47 is called the principal value of sinh~ m; and that particular value of cosh~! m which has the coefficient of 7 not greater than x nor less than zero is called the principal value of cosh—4m, When it is necessary to distinguish between the general value and the principal value the symbol of the former will be capitalized; thus Sinh-! m =sinh tm-+2irx, Cosh-i m =cosh~! m-+2irn, Tanh™ m= tanh! m-+irx, 1 vis any integer, posi Complex Roots of Cubic Equations.—It is well known that when the roots of a cubic equation are all real they are expressible in terms of circular functions. Analogous hyperbolic expressions are casily found when two of the roots are complex. Let the cubic, with second term removed, be written w+ jhe 20, Consider first the case in which b has the positive sign. Let aw—r sinh u, substitute, and divide by r’, then ve oF negative, sinh? w+ 32 sinh « Comparison with the formula siuh" w+ sinh w= ¥ sink gw . 30 3 26 sinh ju gives pop poe? whence revabl, sinh gum, w= sinh Piet fz iacuaiae) therefore = 2b! sinh G sinh a)46 TYPERROLIC FUNCTIONS. in which the sign of b is to be taken the same as the sign of ¢ Now let the principal value of sinh*;5, found from the tables, be m; then two of the imaginary values are n+2ix, hence the ‘). The 3/ three values of « are 2¢ sinh © and 26+ sin’ (Gs . a ea inst two reduce wo — (sinh T tiv cosh ). 3 3 Next, let the coefficient of « be negative and equal to ~3b. Tt may then be shown similarly that the substitution x=r sin @ Teads to the three solutions Fen where m=sin-! 55, a a ‘These roots are all real when cB. If c>d', the substitution =r cosh u leads to the solution —2b* sin > (sin Fav 5 eos Cee Re) x= 264 cosh ( ~ cosh #7), 3 which gives the three roots abicosh , —0+(cosh *+ v7 sinh}, wherein n=cosh-"* 3 7 (com serv ssn); oe in which the sign of 6 is to be taken the same as the sign of ¢, Prob. 85. Show that the chart of cosh (x + 4) can be adapted to sinh (x + iy), by turning through a right angle; also to sin (x +4). Prob. 86, Prove the idemtity tanh (x + yy) = ST 27, ee ae z. Prob. 87. If cosh (x + i), = a+ i, be written in the “ modulus and amplitude” form as r(cos # + ‘sin #), = r exp 6, then P= at+0' = cosh? x —sin*y = cos*y — sinh* x, tan @ = b/a = tanh x tan y. Prob. 88, Kind the modulus and amplitude of sinh (x + ty). aaa Prob. 8g. Show that the period of exp is fa, When m is real and < 1, cosh“! m = feos! m,THE CATENARY. ay Art, 31, THE CATENARY. A flexible inextensible string is suspended from two fixed points, and takes up a position of equilibrium under the action of gravity. It is required to find the equation of the curve in which it hangs. ty and s the length of ae AP measured from the lowest point A; then ws is the weight of the portion AP. This is balanced by the terminal tensions, T acting in the tangent line at P, and /7 in the horizontal tangent. Resolving horizontally and vertically gives Teosp=H, Tsing = us, in which ¢ is the inclin n of the tangent at P; hence tno = Had wheie ¢ is written for H/z, the length whose weight is the constant horizontal tension ; therefore oy dz z Hs See sinh”, sinh © == = ¢ cunaie! which is the required equation of the catenary, referred to an axis of + drawn at a distance ¢ bclow A. The following trigonometric method illustrates the use of the gudermanian: The “intrinsic equation,” s = ¢ tan, gives ds =csect dd; hence dr, = ds cos, = c sec bd; dy, =ds sin gy =csec @ tan dp; thus a= gd", y=esec ; whence y/e = sec =secgd 2/c=coshx/c; and t/e—tan gd r/o — sink of) oe fengih is 5 suspended from two points 20 feet apart in the same hori- zontal; find the parameter ¢, and the depth of the lowest48 HYPRRROLIC FUNCTIONS, The equation s/c = sinh x/e gives 15/e = sinh 10/2, which, by putting 10/¢ = s, may be written 1.5¢= sinh, By exam. ining the intersection of the graphs of y = sinhs, y = 1.5%, it appears that the root of this equation is « = 1.6, nearly, ‘To find a closer approximation to the root, write the equation in the form fiz) inh 2 — t.g2 =a, then, hy the tables, = = 0024, At.64) = 2.4806 — 2.4600 = + .0206 whence, by interpolation, it is found that f{1.6221) — 0, and 2 = 1.6221, ¢ = 10/s = 6.1649. The ordinate of either of the fixed points is given by the equation ye = cosh x/e = cosh 10/e = cosh 1.6221 = 2.6306, from tables: hence y = 16.2174. and req ed depth of the votes —y eo = Prob. gt Tn the ahove num of the terminal tangent to the horizon. Prob, 92. If a perpendicular 4/4 be drawn from the foot of the ordinate to the tangent at P, prove that ACV is equal to the con- stant 4, and that WP is equal to the arc A?. Heace show that the locus of Vis the involute of the catenary, and has the prop- erty that the length of the tangent, from the point of contact to the axis of «is constant. (This is the characteristic property of the tractory). Prob. 93. The tension 7 at any point is equal to the weight of a portion of the string whose length is equal to the ordinate y of that point. Prob. 94 An arch in the form of an inverted catenaryt is 30 feet wide and 10 feet high: show that the length of the arch can be obtained from the equations cosh s — * See a similar problem in Chap. T, Art. 7. 4 For the theory of this form of arch, see “Arch” in the Encyclopsedia Britannica,CATENARY OF UNIFORM STRENGTH. 49 ArT. 82, CATENARY OF UNIFORM STRENGTH. If the area of the normal section at any point be made proportional to the tension at that point, there will then be a constant tension per unit of area, and the tendency to break will be the same at all points. ‘To find the equation of the ium of nder gravity, concider the an element 22 whose ss, aud whose wi is gps, where @ is the section at P, and p the uniform density. This weight is balanced by the difference of the vertical components ‘of the tensions at Pand 2”, hence d(T sin b) = garnds, MT cos) = 0; therefore 7 cos @ = H, the tension at the lowest point, and T=H sec @ Again, if @, be the section at the lowest point, then by hypothesis «/a, = 7/H = sec ¢, and the first equation Healsec o sin $) = gow, sec pds, or edtan @ = sec pas, where ¢ stands for the constant 7/gpw,, the length of string (of section @,) whose weight is equal to the tension at the lowest point ; hence, ds =csec pdp, s/c = gi-', the intrinsic equation of the catenary of uniform strength. Also. dx = dscos b= edd, dy =ds sin @ =c tan pdb; hence —-x = cd, y=e log sec ¢, and thus the Cartesian equation is log sec x/e, Prob. 93. Using the same data as in Art. 31, find the parameter € and the depth of the lowest point. (‘The equation x/c = gd s/e gives 10/c = gd15/c which, by putting 15/¢ = 4, becomes50 HYPERBOLIC FUNCTIONS. gds=4s. From the graph it is seen that « is nearly 1.8 If Als) = gd — 4s, then, from the tables of the gudermanian at the end of this chapter, S(1.80) = 1.2432 — 1.2000 = + .0432, Fit99) + 0072, fl 9g) = 12881 — egnne = — arty, 1.2739 — 1.2667 whence, by interpolation, + = 1.9189 and ¢= 7.8170. Again, s/e = loge sec x/e; but a/e = 10/e = 1.2793; and 1.2703 radians = 73° 17’ 55/7; hence y = 7.8170 X .§41§3 X 2.3026 = 9.7472, the required depth.) Prob. 96. Find the inclination of the terminal tangent. Prob. 97. Show that the curve has two vertical asymptotes. Prob. 98. Prove that the law of the tension Z; and of the section @, at a distance s, measured from the lowest point along the curve, is is 3.48 times, Prob. 99. Prove that the radius of curvature is cosh s/c. Also that the weight of the arc s is given by H sinh s/c, in which s is measured from the vertex. ArT. 33, THE ELASTIC CATENARY. An elastic string of uniform section and density in its natue ral state is suspended from two points, Find its equation of equilibrium. Let the element do stretch into ds; then, by Hooke’s law, ds = dot AT), where d is the elastic constant of the string; hence the weight of the stretched element ds, = goude, gowis/(a PAT). Accordingly, as before, ad hence ed{tan 9) = ds/(t + psec 9), in which js stands for AH, the extension at the lowest point ;He TRACTORY. BL therefore ds = dsec' p+ p sect p)dg, s/e = tan $+ 4u(sec Htan d + gd-1 4), [prob. 20, p. 37 which is the int of ihe common caicnary when yo sic equation of the curve, and reduces to that . The coordinates x, y may be expressed in terms of the single parameter @ by put- ct Ade, (sec @ + msec’ g) sin pag. a/c=gd' p+pmtan gy, y/e=sec p+ gy tan’ p. ‘These equations are more convenient than the result of eliminating ~p, which is somewhat complicated. ting de — Wience Art, 34. THE TRACTORY.* To find the equation of the curve which possesses the secutive tangents such that 7 PT =P'T' =<, and let OT ) = 6; draw 7S perpendiculae | | Se to P'1; then if PP’ = ds, it is evident that S7’ differs ? ™ bil from ds by an infinitesimal of a higher order. Let PT make an augle @ with OA, the axis of y; then (to the first order of infinitesimals) PT = TS = TT" cos @; that is, cd = cos pdt, t=c gd-'d, x =t—csing, =cgd"'g—sing), y=ccosg ‘This is a convenient singie-parameter torm, which gives all ‘* This curve is used in Schiele's anti-friction pivot (Minchin's Statics, Vol. , p. 242); and in the theory of the skew circular arch, the horizontal projection of the joints being a tractory. (See “Arch,” Encyclopedia Britannica.) The = gd /¢ furnishes a convenient method of plotting the curve. equationa HYPRRROLIC FUNCTIONS, values of x, y as @ increases from oto $x. The value of s, ex. pressed in the same form, is found from the relation ds = ST! = dtsing =ctangdd, At the point 4, @=0, 7=0, log. sec @. If w be put for ¢/c, and be taken as independent variable, pagdu, xfesu—tanhn, yfe=sechu, s/e =log cosh. Prob. 100. Given f= 2¢, show that @ = 74? 35’, 5 = 1.3249¢, = .2658¢, 2 = 1.0360, At what point is #=c? Prob. rot. Show that the evolute of the tractory is the eatenary. (See Prob. 92.) Prob. 102. Find the radius of curvature of the tractory in terms of @; and derive the intrinsic equation of the involute. Art 88 Tie Lovonrome. On the surface of a sphere a curve starts from the equator a given direction and cuts all th angle, ‘To find its equation in latitude.and longitude co. ordinates : Let the loxodrome cross two consecutive meridians AM, ANin the points P,Q let PR be a parallel of lati tude; let OM =x, MP=y, radian measure; and let the angle MN = dx, RQ = ay, all MOP = RPQ =a; then quired equation is y = gd (tana). * Jones, Trigonometry (Ithaca, 1890), p. 185.COMBINED FLEXURE AND TENSION, Ba To find the length of the arc OP: Integrate the equation ds = dy csca, whence s =y csc, To illustrate numerically, suppose a ship sails northeast, from a point on the equator, until her difference of longitude is 4°, find her lat nde and distance + = tandy = gd x = gd 47 = gd(.7854) = ¥2 — Loreg radii 1 152 Here tan @ 40.980. If the ship set out from latitude y,, the formula must be modified as follows: Integrating the above differential equa- tion between the limits (x,, y,) and (+4, ,) gives (#, — 4%) tan a = gd7'y, ~ do's hence gd~'y, = gd-'y, +(x, — x) tan a, from which the final latitude can be found when the initial latitude and the differ- ence of longitude are given, The distance sailed is equal to Gey Mercator's Chart—In this projection the meridians are line y’ = x tan a, hence the relations between the coordinates of corresponding points on the plane and sphere are 2’ 9’ = gd-'y, Thus the latitude y is magnified into gd is tabulated under the name of “ y"; the values of y and of y’ being given in minutes, A chart constructed accurately from the tables can be used to furnish graphical solutions of problems like the one proposed above, wctidional part for latitude Prob. 103. Find the distance on a rhumb line between the points (30° N, 20° E) and (30° 8, 40” E). Art, 36. COMBINED FLEXURE AND TENSION, other, and is also subjected to a horizontal tensile force Q ap= plied at the same point; to find the equation of the curve assumed by its neutral surface: Let x,y be any point of theBA HYPERBOLIC FUNCTIONS. elastic curve, referred to the free end as origin, then the bend- ing moment for this point is Qy — Pr. Hence, with the usual notation of the theory of flexure,* andd *y/dz* = d*u/d2*, becomes late mat whence u=Acoshnx4Bsinhnz, [probs 28, 30 that is, y= mx + A coshux + B sinh nx. The arbitrary constants A, B are to be deter terminal conditions. At the free end x=0,9 must be zero, and yamx4Bs & B= m+ nB coshnr; inh nz, xed end, a — 4 and dy/ide — 0, hence Ba —m/n cosh nt, and accordingly m sinh ne = mz — SS es cosh wl To obtain the deflection of the loaded end, find the ordinate of the fixed end by putting x = /, giving deflectio1 = m{I— “tanh nl), Prob. 104. Compute the deflection of a cast-iron beam, 2 X 2 ches section, and 6 feet span, hh at one end and carrying to a horizontal tension of Sooo pounds. [in this case = 4/3, E=15X 10", Q=8000, P= 100; hence #=1/s0, m= 1/80, deflection = gy(72 — go tanh 1.44) = gy(72 — 44.69) = .341 inches.] * Merriman, Mechanics of Materials (New York, 1895), pp. 70-77, 267-269ALTERNATING CURRENTS, BB Prob. 105. If the load be uniformly distributed over the beam, say w per linear unit, prove that the differential equation is TY _ oy s So say BID = Or dest, or Th = Wty mat, feos na +B sinh nx nx? 2t, eosin ns + B sins naj ma" and thai the soiuiion is y: ART. 37. ALTERNATING CURRENTS.* In the general problem treated the cable or wire is regarded as having resistance, distributed capacity, self-induction, and leakage; although some of these may be zero in special cases, The line will also be considered to feed into a receiver circuit of any description; and the general solution will in- clude the particular cases in which the receiving end is either grounded or insulated. ‘The electromotive force may, without loss of generality, be taken as a simple harmonic function of ime, by jodi y perio - dina ple hanmonien| The EMF, and the current, which may differ in phase by any angle, will be supposed to have given values at the terminals of the receiver circuit; and the problem then is to determine the E.M.F. and current that must be kept up at the generator terminals ; and also to express the values of these quantities at any inter- mediate point, distant # from the receiving end; the four line-constants being supposed known, r= resistance, in ohms per mile, = coefficient of self-induction, in henrys per mile, capacity, in farads per mile, B= welch of leakage, It is shown in standard works§ that if any simple harmonic + dee reierences in footnote, Ar. 27. Byerly, Harmonic Functions. This article follows the notation of Kennelly’s ‘Treatise on the Application of Hyperbolic Functions to Electrical Engineering Problems, p. 70. § Thomson and Tait, Natural Philosophy, Vol. I. p. 4o; Raleigh, Theory of Sound, Vol. I. p. 20; ‘Bedell and Crehore, Alternating Currents, p. 214.8 HVPRRROLIC FUNCTIONS, iunction @ sin (wt + 8) be represented by a vector of length a and angle 8, then two simple harmonics of the same period 2n/c, but having different values of the phase-angle 8, can be combined by adding their representative vectors. Now the E.M.F. and tie current at any point of the circuit, distant x from the receiving end, are of the form Hyg +, (64) in which the maximum values ¢y é,, and the phase-angles 0, 0’, are all functions of x. ‘These simple harmonics will be repre- sented by the vectors «,/4, 4,/8'; whose numerical measures are the complexes ¢, (cos # + sin OF, i, (Cos O’ -+7 sin 6”), which will be denoted by ¢,% The relations between é and 7 may be obtained from the ordinary equations + Es an ee AE r Fae det ai \ in (at -+ 4+ de), then he vector wei/? for, since de/dt = « 08 (ial +) = we, de/at will be represented by by the sum of the two vectors ge/é, Gots bes whose numerical measures are the complexes gé, jucé; and similarly for de/dx in the second equation; thus the relations between the complexes ¢, # are Be griwye, Lay tiene (66) * In electrical theory the symbol j ie used, instead of i, for VT, Bedell and Crehore, Alternating Currents, p. 181. The sign of de is changed, because x is measured from the recaiving end. The coeficient of Feaiage, 4, is usually taken zero, bui is inere reiained for generafity and sym- metry, J These relations have the advantage of not involving the time, Steinmetz derives them from first principles without using the variable ¢. For instance, he renards 1 +l as a generalized resistance-cocffcient, which, when applied to i, gives an E.MLF., part of which is in phase with i, and part in quadrature with i, Kennelly calls + ++ jal the conductor impedance; and g + jue the dielectric admittance; the reciprocal of which is the diclectric impedance.ALTERNATING CURRENTS oT Differentiating and substituting give (+ ful)(g + jucdé, (67) } (r + jul)(g + juc)t. and thus ¢, + are simiiar functions of x, to be distinguished only by their terminal values. It is now convenient to define two constants a, % by the equations * = (r+ jul) (g + jer), % —a/(g + jor)j (68) and the diffcrential cquations may then be written a o Seo, Teal, 6. ante Jane (69) the solutions of which are f €=Acoshax +Bsinhax, 7= A’ cosh ax + B sinhax, wherein oniy two of the four constants are arbitrary; for substituting in either of the equations (66), and equating coefficients, give (g + juc)A — 0B’, (g + juc)B= ad’, whenee Bl=A/sy A! =B/ay Next let the assigned terminal values of g, i, at the re« ceiver be denoted by Z, 7; then putting x =o gives E= A, T=A', whence B =I, B! = E/; and thus the general so- lution is (70) * Professor Kennelly calls a the attenuation-constant, and r» the surge- impedance of the line + Sco Art, 14, Proba, 28-30; and Art. 27, foot-note.ba iyrérnoiic ruNcriONs, If desired, these expressions could be thrown into the ordi- nary complex form X + JY, X’-+,¥’, by putting for the let- ters their complex values, and applying the addition-theorems for the hyperbolic sine and cosine. The quantities X, Y,X’, of x; and the repre /¥,whereet=X*+Y, y ¥" would then be expressed as funciio sentative vectors of ¢, i, would be e, /6, 4, #4" tan A= V/ a For purposes of numerical computation, however, the for- mulas (70) are the most convenient, when either a chart,# or a table,t of cosh x, sinh x, is available, for complex values of x. Prob. 106.f Given the four line-constants: r= 2 ohms per mile, 1= 20 millihenrys per mile, = 1/2 microfarad per mile, g£=0; and given w, the angular velocity of E.M.F. to be 2000 radians per second; then al = 4o ohms, conductor reactance per mile; tb jol = 2+ 40j ohms, conductor impedance per mile; tw = .v01 iy, dietecirie suscepianee per mile; oor mho, dielectric admittance per mile: aan .04 +.002), which is the measure of .ogo0s/177° 8’; therefore a= measure of .2001/88° 34/ — .00$0 + .2000f, an abe stract coefficient per mile,of dimensions [length], % = a/(g + jwc) = 200 — sj ohms. Next let the assigned terminal conditions at the receiver be: =o (line insulated); and E = rooo volts, whose phase may be taken as the standard (or zero) phase; then at any distance ~, by (70), Ecosh ax. musi be kepi up ai a generaivr 100 miles away; iden * Art. 30, foot-note. See Table TI t'The data for this example are taken from Kennelly’s article Qc. P38).ALTERNATING CURRENTS. 59 € = 1000 cosh (.5 + 209), 7 = 200(40 — /)~' sinh (.5 + 20/), but, by page 44, cosh (15 + 20/) = cosh (.5 + 20/~ 67) = cosh (.5 + 1.157) = .4600 + .47507 obtained from Table II, by interpolation between cosh (.5 + 1-1/) and cosh (.§ + 1.2/)} hence sho + 4767 = rene O47 sin A), = 45° 55, and where log tan = log 475 — log 460 = .0139, 2, = 460 sec 8 = 661.2 volts, the required E.M.F. Similarly sinh (-5 | 207) = sinh (.5 + 1.15/) = .2126 1 1.02807, nd hence 1 Sol t495 + 82667) (cos 6” + 7 sin 6”), where log tan 6 = 10.7427, # = 79° 45", #, = 1495 sec O/160 5-25 amperes, the phase and magnitude of required current. Woglte + (2126 + 1.028) = Next let it be required to find ¢ at x= 8; then 1000 cosh (.04 + 1.6/) = 1000/ sinh (.04 + .03/), Jr aud appiying page 44. tween sinh (0+ 07} and sinh (0 +.17) gives sinh (0 |-.03 /) = o0000 | .02995 j- Similarly sinh (1 + .03/) =.10004 + .03004 4 Interpolation between the last two gives: sinh (.04 + .03 /) = 04002 +.02999/. Henceé where 40.02 +29.99/) = —29.99-+-40.02/ =e,(cos 0+-/ sin 6), log tan 6 = .12530, 0 = 126° 5 volts, Again, let it be required to find ¢ at x = 163 here — 29.99 sec 126" 51” = 50.01 1000 cosh (.08 + 3.2/) = — 1000 cosh (08 + .06/), bs 2) bu <6) hence cosh (.08 +.06/)=1.0010 +,0048/, and €= — 1001.44.87 = ¢,(cos O47 sin 4), where # = 180° 17', ¢, = toor volts. ‘Thus at a distance of about 16 miles the E.M-F. is the same as at the receiver, but in opposite60 HYPERBOLIC FUNCTIONS, phase. Since ¢ is proportional to cosh (.oos + .2/)x, the value of for which the phase is exactly 180° is 7/2 = 15.7. Similarly the phase of the EMF. at x= 7.85 is 90°. ‘There is agreement in phase at any two points whose distance apart is 31.4 miles. In conclusion take the more general terminal conditions in which the line feeds into a receiver circuit, and suppose the current is to be kept at so amperes, in a phase 40° in advance of the elec- tromotive force: then /== so(cos 40° +f sin 40°) = 28.30 + 22-14 7 and substituting the constants in €= 1609 cosh (,005 + .2/)x + (7821 + 6236/) sinh (.005 + .2/)= 4604 4757 —4748-+-9366/= — 4288-+9841/=e,(cos +/sin 4}, where @= 113° 55", ¢, = 10730 volts, the E.M.I. at sending cnd. ‘This is 47 times what was reyuived when the other end was insulated, Prob. 107. If} =0, g=o0, 1=o; then a=(1+)n, m= (1+ fm, where m* = wre/2, n? = r/2uc; and the solution is & E Vcosh anx + cos 2nz, tan @ = tan nxtanh nx, v2 eee: ee i aa Cos aay tad! — ian me evil na, Prob 108, If self-inductioa and capacity Le aw, aud the se ceiving end be insulated, show that the graph of the ciectromotive force is a catenary if g # 0, a line if g=o. Prob. 100. Neglecting leakage and capacity. prove that. the solution of equations (66) is 7 — I, @=- E+ (7 + jul)Ix. Prob. 110. If x be measured from the scnding end, show how equations (65), (69) are to be modified; and prove that @=Eycoshax — afysinhax, i= I,coshax ~ 1 E,cinh ax, where E, I, refer to the sending end. ArT. 88. MISCELLANEOUS Appr “ATIONS. 1, The length of the arc of the logarithmic curve y = a* is $= M(cosh u--log tanh 4u).in which A= 1/log a. sinh w 2. The lengi = ta(sinh 2u-+ 2), where sinh 7 —7/e f ture is p=(a" sinh” «+6 cosh" u)i/ab; in which w is the measure of the sector AOP, i.e. cosh w= x/a, sinh x= y/b, 4. In an oblate spheroid, the superficial arca of the zone 2. In the hypMISCRILANROUS APPLICATIONS. 61 between the equator and a parallel plane at a distance y is S= m0\(sinh 2 + 2u)/2e, wherein 6 is the axial radius, ¢ eccen- tricity, sinh «= ey/p, and p parameter of generating ellipse. 5. The length of the arc of the parabola y* = 2px, measured from the vertex of the curve, 1s 2= 4p(sinh 204+ 2m), in which sinh u = y/p =tan 6, where @ is the inclination af the termi tangent tv the 6, The centre of gravity of this arc is given by 3lz = p(cosh"u — 1), 64/7 = p*(sinh 4 — au); and the surface of a paraboloid of revolution is $= 27 yt 7. The moment of inertia of the same arc about its ter. al ordinate is = p[xllx— 2x) + yy pV], where p is the mass of unit length, and N= u— sinh 20 — 4 sinh 4u-+ gy sinh 60. & The centre af gravity of the are of a catenary meacn from the iowest point is given by 4Uy= (sinh 2 + 2u), Ze = cu sinh “— cosh u + 1), in which w =.2/c; and the moment of inertia of this arc about its terminal abscissa is T= poly sinh gu + f sinh # — 1 cosh x). 9. Applications to the vibrations of bars are given in Ray- leigh, Theory of Sound, Vol. I, art. 170: to the torsion of prisms in Love, Elasticity, pp. 166-74; to the flow of heat and electricity in Byerly, Fourier Series, pp. 75-81; to wave . B. 477, and in the theory of potential in Byerly p. 135, and in Maxweii, Eiectricity, arts, 172-4: to Non-Fuclidian geometry and many other subjects in Ginther, Hyperbelfunktionen, Chaps. V and VI. Several numerical examples are worked out in Laisant, Essai sur les fonctions hyperboliques.ee HYPEREOLIC FUNCTIONS Art. 39. EXPLANATION OF TABLES. In Table I the numerical values of the hyperbolic functions sinh #, cosh #, tanh w are tabulated for values of w increasing from 0 to 4 at intervals of 02, When w exceeds 4, Table IV may he used Tahle IT gives hyperholic functions of complew argu cosh (wt iy) =a + 15, a id, h (xt fy) = and the values of a, 4, ¢, d are tabulated for values of x and of y ranging separately from 0 to 1.5 at intervals of .1. When interpolation is necessary it may be performed in three stages. For example, to find cosh (.82-+ 1.344): First find cosh (.82 + 1.32), by keeping y at 1.3 and interpolating between the entries under x =.8 and x by ke z= 8and x cosh (.82 + 1.3) and cosh (.82 + 1.4) find cosh( .82 + 1.34/), ing gat ig and y, as before; then by interpolation beiween kept of y, however great, by means of the formulas on page 44: sinh (2 2i2) =sinh2, cosh (x 2in) = cosh 2, ete. It does not apply wh dom occurs in practice. ‘This table can also be used as a com. plex table of circular functions, for cos (y tr) and, moreover, the exponential function is given by n a is greater than 1.5, but this case sel ax, sin(y £iz) dics exp (+24 ty) =atct+ibtd). the signs of x and of / on the ieft, (See Appendix, C.) Table III gives the values of v= gd », and of the guder. manian angle 6= 180° v/m, as u changes from 0 to 1 at interBxPLAWATION oF TARLES. 6h vals of .02, from 1 to 2 at intervals of .05, and from 2 to 4 at intervals of .1. In Table IV are given the values of gd x, log sinh x, log cosh x, as u increases from 4 to 6 at intervals of .1, from 6 to 7 ai int 7 wy at iniervais uf 45. In the rare cases in which more extensive tables are neces- OE Guderman- ian angle (written £) is taken as the independent variable, and increases from 0 to 100 grades at intervals af o1, the corres sponding value of w (written ZA) being tabulated. Inthe usual case, in which the table is entered with the value of 1, it gives by interpolation the value of the gudermanian angle, whose circular functions would then give the hyperbolic fi of «When # is large, this angle is so neatly right that polation is not reliable. ‘Lo remedy this inconvenience Gu- netions 2 to 5, and by .or irom 5 to 1 Glaicher has tahulated the values af o and o-#, te nine sig. ificant figures, a 4 varies by .oor from 0 to .1, by .or from o to 2, by .1 from 0 to 10, and by t from 0 to soo. From these the values of cosh x. sinh # are easily obtained. Geipel and Kilgour's handbook gives the values of cosh a, sinh r, to seven figures, as x varies by .o1 from 0 to 4. There are also extensive tables by Forti, Gronau, Vassal, Callet, and Hoiiel ; and there are four-place tables in Byerly’s Fourier Series, and in Wheeler's Trigonometry, (See Ap- pendix, C.) In the following tables a dash over a final digit indicates that the number has heen increased, *Gudermann in Crelle's Journal, vols. 6-9, 1831-2 (published separately under the title Theorie der hyperbolischen Functionen, Rerlin, 1833). Glaisher in Cambridge Phil, Lrans., vol. 13, 1881, Geipel and Kilgour's Electrical Hand-| tinh 00 | .0000 02 | “0200 rv 06 | | Pat | 10 | | 2 | | 16 18 8 70 7” Er cy 8 80 82 as “0 | 1 ossd oo | Ltd 96 | 1 1448 | HVPERROLIC FUNCTIONS, Tanux I.—Hyrennouic Fuxcrions. cohen | uahw [| vow | ohm 1.0000 | .0000 1.00 11 1.5481 120003 | “ox | 1.03 | 1 1560) 0008 ro | i L308 11,0018 1.06 1 16164 r 108 | 1. Lea 1 110 | 1 16685 : Lay | Lamy | Lows 1 116 1 17317 i wae | 1 188 1 ig | a 18108 i | Ltd vat | Lyah | 18725 1.26 1 1.9045 ras | Costs 1.30 1.6984 | 1,97 use | Last | 2.0008, tat | tres | Soe tas | oy | atin 13 | 186rF | 2.1182 ian | tan | 91m 1a | 1 1804 ret $25, 222 ge Be a ek Bo 38 zi 14 oa 1 1 1 1 4 " 4s 31 E ig BEE65 TABLES, 99924 9027 ‘99980 99918 90021 2932 e528 SERA SEER 2 Serss es 2 Reet BSE Z aRaRA Rae é SS REGSS SStas Beene Seon Sams Rens RaRAS BARES S 8 geeee i Seledsts E = = 9906 001 o018 9917 9920 9923 ‘Taste I. 6.0048 rod Tie 7.8988re . =k deeb HOee Rawk wic-o % x Bm eis HYPERROLIC FUNCTIONS. Taste I, VaLues oF cost (x + iy) AND SIN (x + i “tous 1997 0 8014 “aR18 5675 ui 7200 Tar 8457 8997 0 snot 0.0684 ov708 a'fng | 00709 1.0025 1.0000TABLES. er ‘Tas TI. Vatugs oF cosu (x + i) ann simu (x + i)68 HYPRRROLIC FUNCTIONS, Tanta II. Vatues op costt (x + iy) AND sini (x + iy). et : i 0000 | ‘ 285.0000 ; 1 | t 18 : ! i : i e “3538 7 m8 1. 884 i] ; : Al | : ; 3 n | at eee | eee 0 | 1.8107 0009 |1.5098 0000 | 1.9109 9090 | 1.8982 oon AA | Asie “Tor [5091808] 19617 “ton8 | T6880 “Tons Bera aw sive xe | une ace | 1. 3b | a0hta8 Mor [20 sao] 8800 bono | Lom 4 | 1.0077 arg 999 ost | agus) taut | aman 0 3b | 5800 Te57 [ea7 oest | 17308 “sia | 14905 949 6 | Uaiae B25 [Rass Lowe | L626? oo00 | 14007 1.1181 Sr | Aleedd gree iSto8 11605 | 10074 soma | 1.250 C367 alt r alt ty i 110 | 0:0783 0 ie Li jo v. po re |e & rearel rs aise 1a | 4 1 14 : 15 : i |TABLES, 69 Tavue I, Vatugs oF cost (x + iy) AND sixut(x + iy.) y 78 11,5431 0 PE 3 1.4313 4 1.8542, 5 1.2788 6 Lie 4 a 393 6046 a 4128 4463 1.3 2623 2836 14 * Ls. aun —i eye e, ee ee ine — 0000 | 2.1293 0 2128 | 2.1187 al 200 | 2080 ‘3 m2 | Bowe 3 0.8208 | 1.9613 4 1.0208 | 1.8686 5 Yeues | aot ‘6 iar | 1.0280 a 1.5375 | 1.4885 8 6679 | 1.8236 2 V7O17 | 11505, 10 Lav76 | u ws UI rood | rove 1g om? | “sAG8 1a 2.0983, 3619 14 Zia | 1506 15 2.1908} 0000 2.8804 | be10 HYPERROLIC FUNCTIONS, ‘Tani II. 2881 73.803 00 | 1.8017 74.584 1) | 1.8871 76.087 20 | 1.3501 77.854 iw | 1819 BeaHe 140 | 1.8800 79.688 50 | 1.4075 80.615 60 | 1.4927 81.513 70 | 114868 82.510 20.195 ‘80 | 1.4493. 85.040 ‘98 21.267 .00 | 1.4000 9.707 40 | aaa [TOO) 800 | 1.4713 84.801 ‘Bl Bee Lise | aise | dees Se'See aa | xara [11m | sao | Tart 8x75 48 | 26.503 | 1.20 | 8.40 | 1.5041 86.177 10 arse [125 | ory aso | 1.510 86.541 +53 28.535 | 1.80 | 1.0387 8.60 | 15163 86.870 oa 20537 |1.85 | 1.0885, B70 | 1.5214 87168 ‘96 so.o29 | 140 | 1.0873 8.80 | 1.0261 87.487 58 Bratt | 145 | 11100 8.90 | 115008 87.681 3 SEES) seal 335) BE BS. Z | SERGE ERS ios HG g googe senas seenel & !A. HISTORICAL AND BIBLIOGRAPHICAL, What is probably the earliest suggestion of the analogy between the sector of the circle and that of the hyperbola is found in Newton's Principia (Bk. 2, prop. 8 et seq.) in connection with the solution of a dynamical problem. On the analytical side, the first hint of the modi- fied sine and cosine is seen in Roger Cotes’ Harmonica Mensurarum (1722), where he suggests the possibility of modifying the expression for the area of the prolaie spheivid so as iv give ihai of ihe oblate one, by a certain use of the operator V—1. The actual inventor of the incanza Ricea Phys. Bononis, 1757). Te adopied the nviation Shp, Chop for the hyperbolic functions, and Sep, Ce.g for the cir- cular ones, He proved the addition theorem geometrically and derived a construction for the solution of a cubic equation. Soon after, Daviet de Foncenex showed how to interchange circular and hyperbolic fun tions by the use of V’—1, and gave the analogue of De Moivre’s theorem, the work resting more on analogy, however, than on clear definition (Reffex. sur lee quant. imag., Miscel. Turin Soc, Tom. x). Johann Heinrich Lambert systematized the subject, and gave the serial devel- opments and the exponential expressions. He adopted the notation sinh u, etc., and introduced the transcendent angle, now called the edema: Gc. page go). mportan history of the subject is indicated on page 30. iy played a considerable part in the controversy regarding the doctrine of imaginaries, which occupied so much attention in the eighteenth cen- tury, and which gave birth to the modem theory of functions of the2 FAYPERROLIC FUNCTIONS. complex variable. In the growth of the general complex theory, the importance of the “ singly periodic functions” became still clearer, and was gradually developed by such writers as Ferroni (Magnit. expon. log. et trig., Florence, 1782); Dirksen (Organon der tran. Anal., Ber- kt, Crelle, x: ; Schollbach (Die cinfa: 1, 2256); Schellbach (Die einfas » Crelle, 3854)5 ‘Ohm » #243) miberg, 1855); Hotel (Theor. des quant. complex, Paris, 1870). Many other writers have helped in systematizing and tabulating these functions, and in adapting them to a variety of applications. ‘The following works may be espe- cially mentioned: Gronau (Tafeln, 862, Theor. und Anwend., 1865); Forti (Tavoli e teoria, 1870); Laisant (Essai, 1874); Gunther (Die Lehre..., 1881). ‘The last-named work contains a very full history and bibliography with numerous applications. Professor A. G. Green- hill, in various places in his writings, has shown the importance of both the direct and inverse hyperbolic functions, and has done much to pop- ularize their use (see Diff. and Int. Calc., 1891). ‘The following articles ‘on fundamental conceptions should be noticed: Macfarlane, On the defini functions (Buii, NW. ¥. MM. Soc., 1895). Attention has been caiied in Arts, 30 and 37 to the work of Arthur K. Kennelly in applying the hyperbolic complex theory to the plane vectors which present them- selves in the theory of alternating currents; and his chart has been described on page 44 as a useful substitute for a numerical complex table (Proc. A. I. E. E., 1895). It may be worth mentioning in this connection that the present wnter’s complex table in Art. 39 is believed to he the earliest of its kind for any function of the general argument x+y. (See Appendix, C.) B. EXPONENTIAL EXPRESSIONS AS DEFINITIONS. For those who wish to start with the exponential expressions as the definitions of sinh u and cosh u, as indicated on page 25, it is here pro- posed to show how these definitions can be easily brought into direct : fea vio el eee : and the differential relations ¢ cosh w=sinh du, d sinh u=cosh u du, which are themselves immediate consequences of those exponential definitions. Tet OA, the initial radius of the hyperbolic sector, beEXPONENTIAL EXPRESSIONS AS DEFINITIONS. 73 taken as axis of x, and its conjugate radius OB as axis of y; let OA =a, OB=6, angle AOB=w, and area of triangle AOB=K, then K= dab sin w. Let the coordinates of a point P on the hyperbola be x and y, then *2/a?—32/82—1, Comparison of this equation with the ideniiiy cos u=1 permits the wo assumptions x/a=cosh « and y/=sinh w, wherein wis a single auxiliary variable; and it now remains to give a geometrical interpretation to 1, and to prove that “ AP. nates IK, of a second point Q be x+dx and y+dy, then the area of the triangle POQ is, by analytic geometry, #(x4y—ydx)sin w, Now the sector POQ bears to the triangle POQ a ratio whose limit is unity, hence the differential of the sector S may be written dS—4(x dy _y dx)sin w— Jab sin w(cosh® u—sink® udu=K du. By integration S=Ku, hence u=5/K, the sectorial measure (p. 10); this establishes the fundamental ‘geometrical relations /a=cosh S/K, »/b=sinh S/K. pared by G. F. Becker and C. E. Van Orstrand (1909). For complex arguments the most elaborate tables are those of Professor A. E. Kennelly: “Tables of Complex Hyperbolic and Circular Functions ” (Harvard University Press, 1944). ‘Yhree-digit tables of sinh and cosh of x+iy, up to x=1 and y=1 by steps of .o1, are given by W. E. Miller in a paper. “Formule, Constants, and Hyperbolic Functions for Transmission- line Problems” in the General Electric Review Supplement, Schen- ectady, N. Y., May, 1910. There are interesting applications and an extensive bibliography in Professor Kennelly’s treatise on “The Application of Hyperbolic Funciiony io Elecisital Engineering Pivbleus" (University of London Press, 1912). Tt should be noted that this author uses the term “ hyperbolic to the fact that the “sector measure" for the circle and ellipse is an actual angle (p. rr), ‘he convenient term “ angloid” has been suggested hy Professor S. Epsteen.Additionstheorems, pages 16, 40. ‘Admittance of dieleetrie, 56 ‘Algebraic hens, 8. ‘Alternating currents, 58,46, 53: | | | | | ty of value, 13, 16, 45. Amplitude, hyperbolic, 3x. of complex number, 46. Applications. 46 ef sea. Arch. 48. st. ‘Areas, 8, 0. 14, 36, 37, 60. Axgand diagram, 43, 58. Bassett’s Hydrodynamics, 61. Beams, flexure of, $4. Becker and Van Orstrand, 73. Bedell and Crehore, 38, 56. Byerly's Fourier Series, cley G1, 65. Callet’s Tables, 63. Capacity of conductor, $5. Catenary, 47. 1m strength, 49. ® je Functions, 30, 31. Center of gravity, 61, | Characteristic ratios, 10. | (Chart ot nyperboiic functions, 44, 58. | ____-Mereators, 53. i | Circular functions, 7, 44) 44) 18, 24) 28, | Cayley’s Ell 29, 351 441 45: ‘of complex numbers, 395 42,42 of gudermanian, 28 Complementary triangles, 10. Complex numbers, 38-46. Applications of, $5~6o. Tables, 63, 66. Conductor ‘resistance and impedance, 38 Construction for gudermanian, 30. af cherte, 43. fal graphs, 30 Convergence, 23. 25. Conversion-formulas. 18. Corresponding points on conics. 7. 28 sectors and triangles, 0, 28. Currents, alternating, 55. Curvature, 50, 52, 60. Cotes, reference to, 71. Deflection of beams, 54: Derived functions, 26, 22) 50. Difference formula, 16. Differential equation, 21, 25, 47, 495 Ste 52,87. Dirksen’s Organon, 71. Distributed load, 55. Electromotive force, §5, 58 Elimination of constants, 21. lipses, chart of confocal, 43. umpue tunctions, 7, 30, 31 integrais, 7, 3t- sectors, 7, 3% Equations, Didleveniiat (see). Numerical, 45) 48; se. Evolute of tractory, 52. Expansion in series, 23, 25) 3%76 Exponential expressions, 24, 25, 72+ Ferroni, reference to, 71. Flexure of beams, 53. Foncenex, reference to, 71. Forti’s Tavoli e teoria, 63, 71. Fourier series, 55, 61. antichyperbolic (sce). circular (see). elliptic (see. gudermanian (see). hyperbolic, defined, 11. of complex numbers, 38. of pure imaginaries, 41. of sum and difference, 16. periodic, 44. Geipel and Kilgour's Electrical Hand. Graphs, 32. Greeahill’s Calculus, 72. Elliptic Function Gronau’s Tafeln, 63, 72. ‘Theor. und Anwend., 72. Gudermann’s notation, 30. Gudermanian, angle, 29. function, 28, 31, 34 47, 53; 03, 79. Gunther's Die Lehre, etc, 61, 71. ‘Haskell on fundamental notions, 72. “Hoviel’s notation, etc., 30, 31, 71. Hyporbala, 7 lg, 30, 37, 44, 60. iyperhalie functions, defined, innathenreme far, 16 expansions of, 23, ‘exponential expressions for, 24. graphs of, 32 Integrals involving, 35. INDEX. Hyperbolic functions of complex nume bers, 3B et seg. relations among, 12. relations to gudermanian, 29. Tmaginary, eee eompler Trapedance, 34 Integrals, 35. Interchange of hyperbolic and circular functions, 42. Interpolation, 30, 48) 50, 59, 62+ Intrinsic equation, 38, 475 49, 5% Involute of catenary, 42 of tractory, 50. Jones' Trigonometiy, 2 Kennelly on alternating currents, 38, 58. Kennelly's chart, 46, §8; treatise, 73. Limiting ratios, 19, 23, 32+ Logarithmic curve, 60. expressions, 27, 32 Love's elasticity, 61. Loxodrome, 52. ‘Macfarlane on definitions, 72. ‘Maxwell's Electricity, 62. “Measure, defined, 8; of sector, 9 et sea. ‘Mercator's chart, 53. Miller, W. E., Tables, etc., 73, Modulus, 31, 46. Newton, reference to, 71» ‘Numbers, complex, 38 ef seq. ‘Ohm, reference to, 71. Operators, generalized, 39, 56. Parabola, 38, 6. Periodicity, 44, 62.INDEX, Permanence of equivalence, 41. Phase angle. 56. 50. Physical problems, 21, 38, 47 ef seg. Potential theory, 61. Product-series, 43. Pure imaginary. ax. Katios, characteristic, 10, Limiting, 19. Rayleigh’s Theory of Sound, Or. Keactance of conductor, 58. Reduction formula, 37, 38. Relations among functions, 12, 29, 42. Resistance of conductor. 56, Rhumb line, 53. Riccati’s place in the history, 71. Schelibach, roference to, 72- Sectors of conics, 9, 28 Self-induction of conductor, 55. Series, 23, 31. Spheroid, area of oblate, 58 Spiral of Archimedes, 60. Steinmetz on alternating currents, 38 Sueceptance af dielectric, g& Tables, 69,23, ‘Terminal conditions, Teactary, 48, gt sf, bo. Van Orstrand, C. F., Tables, 73. Variation of hyperbolic functions, 1a. Vaseal's Tables, 63, Vectars, 38, 56. Vibrations of hare, 61. Wheeler's Trigonometry, 6t-