¥ y-


niversity Library


introduction to the theory of automor

3 1924 001 521 636





Cornell University Library


original of this


is in

the Cornell University Library.

There are no known copyright

restrictions in

the United States on the use of the



(EorttfU Hmueraiftj IGtbrarg BOUGHT WITH THE INCOME OF THE SAGE ENDOWMENT FUND THE GIFT OF Hetirg *m.. Sage 1891 -«..„ MATHEisAriCS .


OS .



and to describe in less detail certain of its important developments. of many important and attractive to present in as thorough subjects.PREFACE Owing domain largely to of the researches of Poincare and Klein the Automorphic Functions has expanded enormously last thirty-five years. or in some cases the exclusion. . The present tract had its origin in a series of lectures on to the Mathematical Research Class Automorphic Functions given of the University of Edinbui-gh during the Spring I Term of 1915. wish to express a grateful acknowledgment of my indebtedvaluable ness to Professor Whitta'ker. R. during the and the ramifications fields. F. L. of the subject have extended into many and diverse This has caused embarrassment in the selection of materials for a book of modest dimensions. and made many suggestions . The aim throughout has been a manner as possible the concepts and theorems on which the theory is founded. and to Mr Herbert Bell. who has assisted in the preparation of the Bibliography. and has necessitated a brief treatment. who has read the manuscript during has the course of its preparation. 28th July 1915. The Univebsity of Edinburgh.


.. Automorphic Functions Properties of Simple Automorphic Functions . 6 9 The Transformed Elements Types of Length and . 18. The Linear Transformation The Fixed Points of the Transformation . .. 4 . of Area . . 10. . 10 CHAPTER II Groups of Linear Transformations 7.. ..... 20 21 The Cycles The Transforms of the Fundamental Region Groups without Principal Circles . . . 5. 17 17 18 18 .44 . Notation . 22 25 26 29 32 35 CHAPTER III Automorphic Functions 20. Convergence of the Series .. Conformal Transformations Inversion in a Circle .. 3. 6.. 22. 17. 4. . 16 . Group Examples of Groups Transforming a Group Continuous and Discontinuous Groups The Modular Group The Fundamental Region Fundamental Region for the Modular Group Fuchsian Groups Fundamental Region for a Fuchsian Group Definition of a . .46 . 16. . 9. 2.37 38 . . . .. 1 3 . . . 12. . . . 19.. 8. of Linear Transformations . 11. 21. The Thetafuchsian Series 23. CONTENTS CHAPTER I Linear Transformations PAGE 1. 13. 14. 15.

46.. Plane. 42. . 50 51 Conformal Map of Half-Plane . . .. . The Three Types 35.. 28. . 56 57 59 • CHAPTER V Non-Euclidean Geometry 33. . . Geometry of .. . . .. Uniformisation . Riemann's Surfaces Genus 40. . .. .. 43. . Case III.76 77 78 81 85 87 Bibliography 88 .. . 44. . Cayleyan. . Genus of the Fundamental Region of a Group The Problem of Uniformisation Uniformisation by means of Automorphic Functions Whittaker's Group Weber's Group The Work of Schottky and Burnside .. . 63 64 36. 45.... .. .viii AUTOMORPHIC FUNCTIONS CHAPTER IV The Riemann-Schwarz Triangle Functions PAGE 24. . . or 34. . 53 54 54 55 . Case I. The Elliptic Case The Hyperbolic Case Application to Groups of Linear Transformations . The Differential Equation . Absolute. 27. 72 73 74 . The Triangle Function The Three Kinds of Groups . 39. 37. x + m + ">1 32. Representation of Geometry Non-Euclidean Straight Lines by Circles in the .65 67 69 CHAPTER VI 38. \ + ^ + *<l 30. Stereographic Projection 31. 25. Connectivity of Surfaces 41. . \+h+p= 1 Case II. 26. . The Regular Solids 29. .

If x. — The transformation cz +d where a. and areas in the z'-plane. where the infinite region is considered to be a line. d are constants (real or complex) and ad — be #= is called a linear transformation. c. of infinity differs from that employed in projective geometry.— CHAPTER I LINEAR TRANSFORMATIONS The Linear Transformation. + Also called a linear substitution. To points. b. it is customary to represent z by a point in a plane whose abscissa is x and whose ordinate is y. the point recedes to an infinite distance from the origin of coordinates. 1. the equation reduces to z'= constant graphic substitution. the coordinates being referred to perpendicular axes.* Consider z' =f(z). . or y. where f(z) is a function of z. and let the variable z' be represented on a second plane. defined as follows : Definition. The whole theory of automorphic functions depends upon a particular type of transformation. a homographic transformation.^ * The reason for this is that the kind of transformations most frequently conyariable transform the infinite region sidered in the theory of functions of a complex whereas ordinary projection in geometry into a point in the finite part of the plane : transforms the infinite region into a line. curves. or both become infinite. or a homobut this If ad. This is expressed by z = oo and infinity is considered to be a In this respect the conception single point in the complex plane. .be =0. If z is a complex quantity whose real part is x and whose imaginary part is iy. curves. case is without interest. — The configurations in the one plane are said to be trans- formed into configurations in the other plane. and areas in the z-plane there correspond by virtue of the equation points. . To each point z for which the function f(z) is defined there correspond one or more values of z given by the above equation.

(3) a (1). . Let equation (1) be solved for ± z. and only is following one-to-one Theorem 2. The inverse of a linear transformation is a linear transformation. To each value of z there corresponds one. the determinant of and of determinants (4) (1) . The z-plane is transformed into manner by a linear transformation. This is called the inverse of the transformation It is (1). (ya + /3c)z + ab + /3d + hc)z + yb + 8d ' ^ ' Making the transformation and then making to the single transformation Now (4) is equivalent (5) is also a linear trans- formation. its determinant in the form in which the fraction It is worth noting that if the is written is (ad . *=°^A yz + Expressing „ Z • • • • (4) v ' z" as a function of z. that follows but one plane will be used unless the contrary is The results just found can be stated in the explicitly stated. It will is called the determinant of the transbe convenient to have always ad-bc=l . I formation. value of z by (3). the z-plane one. ' +b z=.dz — cz . It will be most serviceable to represent the values of z' not on a different plane. Let the points z' itself in a be subjected to a linear transformation. and only one. . .— — — — 2 AUTOMORPHIC FUNCTIONS The quantity ad — bo [CH. but on the same* plane and with the same In all system of coordinates as are used for. a[(az ~y[(az + + b)l(cz + d)] + j3 = b)/(ez + d)] + S ~ (1) (5).representing z. In oraier words. one.1. a linear transformation with the state then same determinant as We Theorem.bc)(aS — /8y). (act. value of z' by (1) and conversely to each value of z' there corresponds . are each unity. transformed into the z'-plane in a one-to-one manner. ( 2) the determinant in the general case taking this value if the numerator and denominator of the fraction in the second member be divided by J (ad — be).

a linear transformation. '. 3. or it leaves . The Fixed Points of the Transformation. . when (10) ad . we get again a linear transformation of Thus we have. the fixed points are ^ = 00. (> If c = 0. in general z.— The points which remain unchanged under the transformation (1) are found by solving the equation z = (az + b )/(rz + d). .„. If c 4= 0. linear transformapoints z v z 2 z 3 into three tion which transforms three distinct distinct points z x z 2 z 3 We shall prove first that there is not more than one such transis one. z 2 z3 z± are the points into which they are transformed. have more than two roots only in case it is In this case is. . we have.«&/(<* if -a) . For if . and z(. '. (9) but if one fixed point (a — df + 4<bc = 0: or. '. '. (Zi'-^'K^'-Z/) = (Zl-3 2 )(Z 3 - Z )' 4 (6) 2. $. in position. and so on. There — and only '. following proposition Theorem 5. means of the last theorem we are able to establish the one. if c = d-a = b = 0. '. Theorem formation. or cz 2 + (d-a)z-b = (7) the fixed points are t _ ^ t ^ _a-d + J[(a-df + 4bc] 2c ' a-d- J[(a 2c rf) 2 ~~ + 46c] ' . It is the point $=(a-d)/2c Equation every point (7) will identically satisfied. — 3 2] LINEAR TRANSFORMATIONS If z" be subjected to (5) is unity also without further change. There is . transformation has the property invariant the cross-ratio of four points.be = 1 (a + df = 4. on substituting for z(.— — 1. linear transformations The successive 'performance of any number of is equivalent to a single linear trans- The that . z x z 2 z s z i are four points. z 2 z 3 \ zl their values from (1). Theorem — The only linear transformation with more than is the identical unchanged two fixed points By transformation z' = z. linear. homographic. that is 4.

and let . That there is always one such transformation we prove by actually setting it up. Section 1]. AUTOMORPHIC FUNCTIONS Let z' [CH. It obviously transforms z v z v z. <5) be two transformations carrying ing r. '. '.2 'z 3 ') and (zz v z 2 z s ). established. and . z" = (az + /3)/(yz + . When Tl ie z = z v z 2 or z 3 we have by hypothesis z" =z' = z 1 z 2 or z 3 points. z 2 z'. that z" depends linearly upon z from z and z" from s. Eliminat- we get z" as a linear function of =(Ajs' + B)/(C0' + D). for '. I = (az + b)j(cz + d). x z. Equation (11) is a convenient form for use in actually setting up the transformation carrying three given points into three given points. . zt or zs In any case there one transformation with the desired properties. If z1 = oo . in passing from a point on C t to a point on C 2 through the region 3. . We know '. If none of the six values is infinite. This is of the form (1) when solved for z in terms of z. . without actually performing for z is derived the elimination. z v z 2 z 3 into z(. transformation from z to z' has then three fixed = are transformations z3 Theorem 4 z" The and by z two thus the same. the angle between their tangents For convenience we shall consider the angle positive if at p). . . Conformal Transformations. it is necessary to replace the second member of (11) by z - z. 1 or z -z„ respectively member is is and a similar change is necessary in the first an infinite value of z{. C x and C 2 be two curves issuing from p. by a linear transformation.e. consider the transformation defined by an equation which expresses the equality of the cross-ratios (z'.4 formation. .A into z/. z/. Let 6 be the angle between the curves (i. '. or z3 = oo . and the theorem . z„'. '. —Let p be a point in a plane. . z 2 z s The equation (11) depends upon both variables if each set of three points is distinct the determinant of the transformation then does not vanish [see footnote. from Theorems 1 and 3. z'' z s '. . '. which proves that there is not more than one transformation with the desired properties. z2 = oo .

It is directly or inversely conformal according as the sign of the angle is C2 ' . z ./ The transformed angle & will be that anglebetween which includes the transforms of the points within the original angle It is positive if in passing from C/ to 0. — preserved or changed.c. except at the points where f(z) has singularities* Laying aside the two points z = <x> and z— — d. sec. It is a theorem in the elements of the Theory of Functions that a transformation of the complex plane z' =f(z). where f(z) is any analytic function of s.' therefrom (V and C. 253. v are transformed into a point p' and two curves C/ and issuing C. be written in the form (z ->/)'' (z-q) m ' + sec. where In general the correspondf(z) is an analytic function of z. Let z = q be the point for which z = x The function can f(z) then has a pole at q and nowhere else. preserves both the magnitudes and signs of angles.f ence of the points of the z-plane. t Ibid. For example z — Jz gives two values of z for each ' value of z . A transformation preserving the magnitude of angles is called a conformal transformation. that the most general directly conformal transformation of the 3-plane is of the form :'=f(z). For let z =f(z) be the most general function yielding a oneto-one transformation. 9. = sin z is gives infinitely many arises values of z for each value of z The question naturally transformation whether the linear the most general one possessing the quality of uniqueness mentioned in Theorem 2. Definition. Z-q —L -t- Au . formal. the latter of lines which gives z'=x since we have not defined the angle between meeting at oo we can state for reference the following Theorem 6. Theory of Functions. This is in fact the case. — LINEAR TRANSFORMATIONS 5 3] of the included angle the motion is counter-clockwise around p. It is a theorem of the Theory of Functions. . whose proof is beyond the scope of this tract.. through the included angle the motion is counter-clockwise around p' The cases with which we shall be chiefly concerned are those in which & = ±0. C C. . and these of the s'-plane is not one-to-one.— 2. . Now let a transformation of the plane be made such that p. The linear transformation is directly con. See Forsyth.

whose distance from O satisfies the equation OP' OP = r2 P and P' are said to be inverse with respect to the circle. 4. (x\ y') respectively. If 0. presently. Therefore m m m ^=-^.— — 6 — AUTOMORPHIC FUNCTIONS [CH. the circle becomes a straight approaches a position such that the line PP' is perpendicular to and is bisected by the given line. the straight line being included as a special case of the circle. . I where is the order of the pole. (y) Two points inverse with respect to a circle C are transformed into two points inverse with respect to the transformed circle C. To each value of z there are values of 2. ' We q have then established the following Theorem 7. y). P' have the coordinates (f.+ A = A 2 + Al ~ A z— z— ° ° g . g). the equations of the transformation are • r \ x -f) _ f_ '-(z-ffi + iy-g)*' 1 V ~9 _ = {x-fY + (y-gi r r2 (y - 9) to infinity. If the centre O moves The point P' inversions (a) : Angles are preserved in magnitude but changed in sign the transformation is inversely conformal. .' ." in the given circle and of radius r. Inversion in a Circle. The inversion is a one-to-one transformation of the plane into itself which preserves the magnitude that is. we shall see There is an intimate relation. (x. Clearing of fractions we get an equation of the m-th order in 2. which is of the form q ° (1). Inversion Consider a circle with centre at is the transformation resulting from replacing each point P of the plane by a point P' lying on the line OP and . (/3) Circles are transformed into circles. inversion in a straight line is a mere reflection in the line. Hence must equal 1. The most general one-to-one directly conformal transformation of the plane into itself is a linear transformation. The connection between inversions and linear transformations arises from (a) above. P. We state here for reference some well-known properties of line. That is. as between the linear transformation of — the complex variable and the geometrical transformation of the plane known as " inversion in a circle.

Let a translation. yl = y + a". the transformation. LINEAR TRANSFORMATIONS If of the angle but changes its sign. translated parallel to the line Let AB. A redistance a at and lines perpendicular to Oa Oa . is equivalent The successive performance of an even number to a linear transformation of the of this theorem will z plane into itself. be made is a one-toone transformation of the plane which preserves both the magnitudes and the signs of the original angles. z 1 = x 1 + iy v a = a' + ia".—— 3.4] — . . |Oa apart. This Oct is (fig. The converse transformation now be proved. or any even number. 1.. is the origin to Each point of the plane FlG. number of which proves that the given general linear transformation is equivalent to an even number of inversions. in succession. verified by eliminating We shall transformation is show that each of the five transformations is equivalent to an even inversions. be) = Ae £fl . (a) Putting z = x + iy. joining 1) be the line the point a. A> . such a transformation is a linear transformation. CD be two a distance equal to the length Oa. now two inversions. The = (az + b)/(cz + d) can be broken up into the succession of the following transformations : (a) (b) z z =z+a = e'% 2 l where o = d/c where c 2 /(ad - . on breaking and imaginary real into parts. the result Hence Theorem of inversions 8. According to Theorem 7. where fi = a/c. (<*) (•) z = zi + /3 are . given That these equivalent to the s v z 2 z3 zv . is Xl = x + a' .

yl sin 6 . . of inversions and by the linear In the present case we note that the points on AB are unchanged by a reflection in AB and are carried parallel to Oa a distance equal to the length Oa by a reflection in CD. P" are the successive OP" = Ar2 /OP' = Ar2/(r2/OP) = A OP. x.* Thus (a) is equivalent to two inversions.. to the rotation in This case is thus reducible to a pair of inversions. .AUTOMORPHIC FUNCTIONS flection in [CH. equating real and imaginary parts of the two members of the equation. (d) followed by (2) an inversion in a circle centre at rJA. ie in the form cos 0+i sin 6. y3 = Ay 2 . 2 = u-j cos . This follows number from Theorem 5. I AB (as in figure) followed by a reflection in CD is equivalent to the translation under consideration. y±3 2 + 2 2/ 3 This equivalent to an inversion in the )' y& unit circle with centre at the origin * 3 +ys * Vo-- z/+ys it is sufficient to We note that in this and succeeding cases verify that three points are transformed alike by the even transformation. For if P'. *4 + % = -z+Vs ys 2 2/3 ' 2 or V+ is . 2 . we get in this (6) Expressing e case. Vi = x sm + V\ \ cos > a rotation about the origin through an angle 6- Let A'A and B'B be two lines through the origin such that the angle AOB is \Q (fig. We verify at once that a reflection in A'A followed by a reflection in B'B is equivalent question. is equivalent to the two (1) inversions: an inversion in a circle centre at the origin and radius r . (c) This is the case of ex- pansion from the origin Ax It Fig. the origin and radius positions of P. 2. 2).

is a translation. Thus ((3) and (y) preceding yield the following important theorem Theorem 10. | dz2 has the area \dzxdz«\. 5] —— LINEAR TRANSFORMATIONS — — 9 followed by a reflection in the imaginary axis (e) This. increment dz' = (dz'\dz)dz. can be chosen in infinitely many ways. the transformed points are inverse with respect to the transformed circle. Since angles are preserved. carrying three distinct points of the first circle into three These points distinct points of the second circle (Theorem 5). the transformed point z is given an. Combining these theorem results. Since three points determine a circle. There exist infinitely many linear transformations which transform a given circle into a second given circle. of dz is the absolute value or modulus of dz and the length A small rectangle at the point z of sides dz l and dz'/dz \\dz\. There exist infinitely many linear transformations which transform a given circle into itself. (a) Circles are transformed into circles (6) If two points are inverse with respect to a circle. 5. let z be given " . Any linear transformation is equivalent to performance of an even number of inversions. like (a). a given circle can be transformed into a given circle by setting up a transformation — . we get the converse of the preceding Theorem the successive 9.— 4. Corollary. It is evident that the representation of a linear transformation as a succession of inversions tion can. The properties of inversions furnish us with valuable in- formation concerning linear transformations.— In the transformation = (az + b)l(cz + d) an increment dz. When the complex plane is subjected to a linear transformation. is not unique that the transformabroken up into an even number of inversions . Hence Theorem 11. The length of dz is \dz\. in fact. the . — The Transformed Elements z of Length and of Area. be in infinitely many ways.

The number of fixed points and the behaviour of the transformation 6. we note that when = oo . a transformation may have either two fixed points or one fixed point. Let chosen. particular point z 3 ' .c{2 & and |2 having the values in Equations (9).* The character of K is used to classify the transformations with two fixed points and we shall see presently that it determines the behaviour of the transformation with reference to the fixed points. (12) plier of the transformation... 1. .— — 10 AUTOMORPHIC FUNCTIONS is [CH. Laying aside the identical transformation z = Z. for !lz£i z z £2 where K = (z . If the determinant of the transformation is 1. is unity. I transformed area I a rectangle at dz z'. we put and z 2 the transformation £ and becomes = K?^ij z £. Its area is | dz(dz^ \. elements of length in the neighbourhood of a point z are multi-2 and elements of area are plied on transformation by |cz + d| 4 multiplied by |cz + d|~ . or dz'ldz 2 | 1 dz x dz. 1 with reference to classification. but such . (l3) * Considering the transformation in the 23 form Section ). 12. 23' = a/c. we see by Equation (6) that K is unchanged. its fixed points first furnish a useful basis of larger class with We turn to the two fixed points. if for z 3 z z we put any other pair of corresponding points. z z= -\^t' H(1).c£i)j(a . .£\)l(z ' 3 z U ~ fi)« ~ &) K is called the multi~ It . we can state Theorem Types of Linear Transformations. Using these values. the fixed points. It will simplify the problem to transform the plane in which z and z' are represented by a linear transformation. If in zx Equation (11) . depends apparently upon the is not the case. ad -be...Q(z . K = (a .— We have already noted a separation of linear transformations into two classes. |. Now Tz _ ad -be ~ (cz + df ' and if the determinant of the transformation.

and oo are inverse points with respect to any circle with centre at the origin. . F. . 71 K — A. since circle orthogonal to the fixed circles. K in terms of its modulus A (>0) amplitude so K = Ae* we distinguish the following cases (a) : The hyperbolic* transformation. applying Theorem 10 (b). since angles are preserved. Let the ZZ'-plane be now transformed back into the zz'-plane by means of Equations (13). a circle through the fixed points 0. and its (14) oo in which the fixed points and gv £2 are now shifted to Having studied this latter transformation. = AZ is the expansion from the origin observe at once the following facts through the origin (that is. parabolic. 3 way in shows the two families of circles just mentioned. The which a region is transformed is indicated. z'.. The transformation studied in Section 4 (c)..— 5 6] > . each shaded region being transformed into the next in the direction of the arrow. fixed circles. Klein in 1878. * The names elliptic. oo) is transformed into itself (2) any circle with centre at the origin (and hence orthogonal to the family of fixed lines) is transformed into some other circle We concerning it : (1) a straight line with its centre at the origin. LINEAR TRANSFORMATIONS 11 The transformation becomes Z' = KZ . were first used in this connection by of substitutions of these elliptic. be Writing 6. circles orthogonal to the family through £x and £2 In the hyperbolic transformation then (1) any circle through the fixed points is transformed into itself (2) any circle orthogonal to the family of fixed circles is transformed into some other We note also that. inverse points with respect to any circle orthogonal to the . hyperbolic. the points £ and f2 are. . the results can carried back by the inverse of (13) to the plane of z. The points and oo become £ and £2 respectively. and . . The straight lines through the origin become circles through the points £ and £2 and the circles with centre at the origin become. Fig. hyperbolic. They were suggested by the connection types with displacements in absolute geometry of the parabolic type.

(2) the circles orthogonal to the circles through the fixed points are the fixed circles (3) the fixed points are inverse points with respect to each of the fixed circles. Z' = Ae^Z. . Here 6 = ^tt. the fixed points is transformed into a circle through the fixed . but their r61es are changed.12 (b) AUTOMOBPHIC FUNCTIONS The elliptic [CH. 3. The circles of the preceding case play a part here. the points points with respect to the fixed circles. making an and co are inverse angle d with the given line. points intersecting the given circle at an angle 6 The behaviour of the transformation is shown in fig. . The circle with centre at the origin is transformed into itself a line through the origin - K . 4. e The transformation Z' = e' Z is the rotation about the origin discussed in Section 4 (6). K * The terra loxodromic was introduced by F. Each shaded region is transformed into the next in the direction of the arrow. = Aeie (c) The loxodromic * transformation. Transferring these results to the zz'-plane (1) a circle through : Fig. is transformed into another line through the origin. and only one of the fixed points is shown. Klein in 1882. is a combination of the preced. I = eie transformation. This transformation.

both a stretching from the origin and a In general there are no fixed circles but in the particular case 6 = tt. each line through the origin is transformed into itself. rotation about the origin. the two . then. the half of the line on the one side of the origin being transformed into the half on the other side. a stretching from the origin followed by a rotation through two right angles. In the particular case 8 = ir each circle through the fixed points is transformed into itself. the loxodromic transformation has general no fixed circles.. 6] LINEAR TRANSFORMATIONS It involves 13 ing ones. in In the zs'-plane.

<8 zillzl -I. Transforming back to the cj'-plane.d/c into ajc. intersecting only at oo become circles intersecting only at £. straight line is transformed into itself any other transformed into a parallel straight line. s-t Substituting for £ its value. AUTOMORPHIC FUNCTIONS [ CH - 1 £ . this z-£ z-i _L = -L + r fi. . in the parabolic transformation there is a one parameter family of tangent circles through the fixed point each of which is trans- Any line parallel to 0/3 is . = -^-. Hence.— 14 tion carrying oo . Fig. we can write the trans- formation in the form ?1 lW~ . 5. .ajc -d/e-J z-£ {-ajc ' z-£becomes where z. a+d • ( 15 ) Transforming the plane in which z are represented as follows z-S the transformation becomes Z' ' z -t (17) = Z + /3 the fixed point now being at infinity. This is the plane parallel to the line 0/3 joining the origin to the point a translation of ft. £ oo . the point oo becomes the point f Parallel straight lines. . and hence tangent at £.

5 shows a parabolic transformation in which the fixed circles are those with a horizontal tangent. One °° . Fig. f in the zs'-plane that is. 15 formed into any other circle through the fixed point is itself * transformed into a second circle tangent to it. in the ZZ'-plane transformed into a circle through | + 1/j8. of the fixed circles is a straight line through is The fixed line through oo . Thus the fixed circles are those circles through whose tangent is parallel to the line joining and l/. .8. a straight line through £ f and | + l//3. The circles with a vertical tangent at £ furnish an example of a family each circle of which is transformed into some other * 0. |. B.6] LINEAR TRANSFORMATIONS . circle of the family. The shaded regions connected by the arrows show how the parts of the plane are transformed.

Linear transformations are thus T2 T3 . we have. when not likely to lead to confusion.e. and _1 T-" = (T )".. 1. No confusion then can arise from writing STU. Notation. as S. In combining transformations it is easily seen that the associative law holds. etc. If a transformation S be made. —The z \ linear transformation z' = (az + b)/(cz. i. and then to the transformed plane a second transformation T be applied. are the transformations equivalent to performing T _1 will represent the inverse of T. T by twice. . leaves the points of the plane T~ 1 T=1. the single linear transformation equivalent to the succession of the two will be represented by T(S). thrice. that S(TU) = ST(U). ST will be different in is general from TS. {a a (.— CHAPTER II GROUPS OF LINEAR TRANSFORMATIONS 7. Also it will often be convenient. U. . of Equation Section 1. The The identical transformation z' z is represented fact that a transformation followed by its inverse = unchanged gives the equation Combinations of T and its inverse T" 1 obey the ordinary laws of the addition of exponents in multiplication.rd) 1 is frequently denoted by v) + dj or \ > ( cz i \z .:a \' + by)z + a/3 + b8 + dy)z + c/i + dB which are different in general. etc. or simply TS._. to represent the transformation merely by a capital letter. Thus if S is the transformation above and T the transformation z = (az + jS)l{yz + <5). making use (5). not commutative. Tg = / Z ' aa + ( (ya \ f3c)z + (3d \ + 8c)z + yb + 84j' + ab t g^ "' /. T.f 2 • K z — - f2 / t > the transformation being the process of substituting the second term in the parentheses for the first.

after the manner of the preceding case. constant and m is a positive or negative integer or zero. but we shall be concerned only with groups of linear transformations. (1) Examples Of Groups. a. . is said to form a group if (1) the succession of any two transformations whatever of the set is a transformation of the set . . forms a group. we make on the plane transformed by ST first the transformation S -1 then T -1 T -1 -1 S (ST) = Tri(S-iS)T = T-iT=l. The definition applies to all kinds of transformations. . - applied V-ixj-i after . and U 2 . . T _1 also belongs to the group. Definition of a Group. The set z' = z+ma>. ST belongs to the group and (2) if T is a : . They are the successive rotations about the origin through * The order of procedure in a combination should be carefully noted. angle 2w/m. m-1. Group of Rotations about the Origin. the combination STU is z'=f3 {f^Lfii z )Vlis T is Thus z'=Mz). T _1 T( = 1) does also hence. 7-9] GROUPS OF LINEAR TRANSFORMATIONS 17 To find the inverse of ST. S is — — — . (5) Cyclic Group. Since T and T" 1 belong to the group. The transformations z' = eiiz. constitute a group. If k be restricted to positive values. In general the inverse of ST that is. The group properties. For if T is z' = z+k lt T~ l is z' = z-k u which belongs to the set.. using the notation of the last section. If S is z'=f3 (z). we do not have a group for T does . where k has any real value. z'=f1 (z). where any real quantity. to satisfy both group properties.— A set of transformations. forms a group. an form a group. — give here several examples of groups.1 S _1 . a are constants and m and m! are positive or negative integers or zero. . every group contains the identical transformation z' = z. — We Group of Real Translations. where a is (3) Group of the Simply Periodic Functions. where (4) Group of the Doubly Periodic Functions. — The m transformations z' = e2:n'n/mf^ n = 0. not then belong to the (2) set. ThusT-iS-i is the transformation which. If S is z' = z + k2 ST is z' = z + k 1 + k2 which belongs to _1 the set. The set z' = z + mm + m'm. finite or infinite in number. STU is U. followed by S. All transformations of the form z' = z + k. (2) the inverse of every transformation the set is a trans- of formation of the set. leaves the poiuts unchanged. transformation of the group. (ST). . T _1 S _1 * ST.1 = T. 1. 9. followed by T. UV is 8. It is assumed that the ratio a/u' is not real although this is not necessary to establish the group properties. are these (1) If S is a transformation of the group (not necessarily different from T). are easily shown.

and the first group property is satisfied. and let T carry becoming z and Let a transformation G be applied to the plane. All configurations in the Zj-plane can then be carried back to the s-plane by the transformation 11. Therefore. We shall find that it will frequently facilitate the study of a group to transform it in this manner. Thus.. z z to z/ zx and z{ respectively. Transforming a Group. two cases will arise : either all the transformed points will lie outside a sufficiently small circle with z as its centre (excluding. We shall show that the new set of this formed in transformations form a group.and this belongs to the set since ST is a transformation of the given group. the transformations of the given group be transmanner. S^T^GSG^GTG-^GSTG. the transformed set forms a group. —Let z be an arbitrary point in the plane. m z into z' a transformation of the original group. and the second group 1 . The transformation T x carrying x zx carrying that is a linear transformation equivalent to: (1) to z followed by (2) that carrying z to z' followed by (3) that carrying z' to z x \ Hence T = GTG x -1 . and let us apply to this point all Considering the the transformations of some given group. II six transformations z i_. property is satisfied. X JZl _i- form a group. And if S 1 = GSG" where S belongs to 1 the given group.— From any given group of groups can be linear transformations infinitely many other the plane derived by applying a linear transformation to Let T be which z and its transformed values are represented. We verify perties are satisfied. of course. 18 (6) AUTOMOKPHIC FUNCTIONS Group of the Anharmonic Ratios. let all . since T" 1 belongs to the original group. Tf^CGTG-^-^GT-iG" 1 which belongs to the set Now. the six anharmonic 10. positions of the points derived from z by these transformations. . group proby combining the transformations that both that if z is any one of reason the for named is so This group ratios are given by line. the anharmonic ratios of four points on a the transformations of the group. G _1 Continuous and Discontinuous Groups.-Tte 1 [CH. or there will be infinitely many of the trans- . the identical transformation of z which leaves the position unchanged) .

particularly by Lie. where n is a positive or negative integer or zero.. In the former case the group in the latter. and have yielded important However. by taking k sufficiently small.b are arbitrarily small quantities (but not all zero. and where 6 is incommensurable with ir. .Sz. whatever be the point whence. This group is continuous therefore. there is no transformed point within a circle of radius w drawn about z. reason that will appear later (Section 20. the group is continuous. If the group be continuous. there must exist transformations in which c. such groups . there will exist transformations. Hence. obtain a transformed point as near as we like to the given point. Thus consider the group z' = ein9z. such that Sz is arbitrarily small. If the transformed point be represented by z -f. (1) of Section 9). By rotating through the angle 6 a sufficient number of times any point can be brought back as near as desired to its original position hence. on the other hand. the transformation can be written z the identical + Sz = (az + b)l(cz + d) . continuous. (1) and (2). as Nos. Continuous groups have been made the subject of many investigations.— 9-11] GROUPS OF LINEAR TRANSFORMATIONS 19 z. which change the position of an arbitrary point of the z-plane by an arbitrarily small amount. A group with only a finite number of transformations. we can. (5) and (6) of the same section. The group is then discontinuous. cz2 + (d — a)z — b + (cz + d)Sz = 0. We lay down the following Definition. z. Thus. the group is continuous. footnote). This is the group of rotations through positive and negative multiples of the angle 0. (3) of Section' formed points clustering about said to be discontinuous .d — a. Similarly the group of the doubly periodic functions (No. is necessarily discontinuous. for a results in the theory of differential equations. Otherwise it is discontinuous. In order that the group be continuous it is not necessary that the coefficients of the transformation should vary continuously as is the case in Nos. for that is but the identical transformation). (4) of Section 9) is discontinuous. 9). // there exist transformations of the group. In the group of real translations (No. is given any finite point z. in the group of the simply periodic functions (No. — transformation excluded.

/3. If z' = (az + (3)l(yz + S) is a second transformation whose coefficients a. The inverse of such a transformation has the coefficients — d.bc)(aS . by y - and clearing of fractions. mailer. yb + Sd (Eqn. We have UT. denominator UT S = (s. therefore satisfied. s-2) or T1 = T. These coefficients are integers or zero. — 1/z) and T:(z. which are also positive or negative integers or zero and with determinant ad — bc = l. It will be shown in a moment that any transformation of the modular group can be formed by combinations of S and T with their inverses. The properties of this group exemplify many theorems to be subsequently established and the study of the group at this point furnishes a concrete background for future developments of the theory. y. : U= The method (z. ab + (3d. by z. Section 1). = fo Replacing x Put s-2 (4s 1/s 1/z + l)/(lls + 3)). z + n). . (3). Consider the transformations z' = (az + b)j(cz + d) in which . (4x25) -(9x11)=]. The Modular Group. T~" is (z. c. c. b.n so that 25 shall be replaced being chosen for by as small a number as possible. d are and ad — bc=l constitute the modular group. (5). . b. study of a well-known group. the first coefficient in the th< becomes the smaller. —a (Eqn. All transformations integers (z.20 AUTOMORPHIC FUNCTIONS [CH. ya + Sc. (4z + 9)/(lLs + 25)). We verify that S" 1 = S or S 2 = 1 T" is (z. II play no part in the theory to which this tract is devoted we shall not have occasion to consider them further. o are of the desired form then the first transformation followed by the second is a transformation with the coefficients aa + /3c. where T 1 = (z. taking first a particular example . and the determinant is (ad . b. I (-i/z + l)/(-ll/z + 3)) = (z. c. condition that . (z-4)/(3z-ll))./3y) = 1 Both group properties are a. Since 25 > 11 we replace z by z±n. to be followed is to reduce the coefficients in the denominator use of combinations of S and T. and In this section we make a detailed 12. — d are positive or negative integers or zero subject to the ad — be = 1. z-n). (az + b)/(cz + d)) in which a group called : a.z + 1).2 . Section 1). We shall now indicate a method of decomposing a transformation of the modular group into combinations of S and T. which belong to the modular group. Let us consider the two transformations S (z. .

11-13] Continuing. Definition. Hence. Definition. will then be expressible in terms of the new pair. z±n). with the study of the modular group it is desirable to introduce the definition of the fundamental region.T 1 -1 = ST. We eventually reduce the | | =t= denominator to unity. 1 1 21 =/(3. 3 . any transformation of the modular group can be expressed in combinations of S and T. or two areas are said — — be congruent with respect to a given group if one can be transformed into the other by some transformation of the group. we interchange c and d and can reduce again. The method followed in this particular case furnishes material for a complete proof for the most general case. Putting — 1/s for z. z±n) will not reduce d to zero.. where T2 = T4 =(~-.:. U = S. z^n) becomes equal to the identity.T3 U is thus -1 S. T = UV. The Fundamental Region. — — = = since and V (z.T 2 1 -1 S.4ST 2 expressed in terms of S and T. a conception of the foremost importance in the theory of discontinuous groups. The determinant at each step is unity. two curves. So long as c 1 the substitution (z. for the condition ad — be = 1 requires that c and d be prime to one another. All transformations. + 1)) is . and ad — be = a = 1 and the transformation of the form (z.ST. + 3) UT ST2ST S = 1. + l)). since the transformations which are combined have unit determinants. Any pair in terms of which S and T can be expressed will serve. We then express the given transformation in terms of the transformations used in reducing it as in the example.. -l/z). 1 1 Hence 3 . z±n) reduces the transformation to one in which \d\<\c\. 3 1 3 From this last equation U is 1 the inverse of T ST2 ST3 S. 1/z) might have used U (z. being expressible as combinations of S and Thus we T. which by means of (z. Two 'points. GROUPS OF LINEAR TRANSFORMATIONS UTjSTj^. It is clear that S and T are not the only transformations that can be regarded as generating transformations of the group. Before proceeding further 13. z + 4) UT ST2S=(„~. A region such that no two of its points are congruent with respect to a given group and such that every to — . l/(z S = U. where T3 = T = (. When \d\ >\c\ a transformation of the form (z.-l/( J -3)) UT ST 2ST = (.

for example. e two points in the shaded region can be carried into one another by rotations through multiples cyclic but points in any adjacent of the angle 2tt/6 region (and in this case any other point in the . 7 is a fundamental " parallelogram of periods. s + 1) represented geometrically by a . 6 shows a fundamental region imw/e for the No z). add to the given region a small adjacent region which has no two of its points con- provided we remove from the original region the part congruent to the added region. being congruent to the same would be congruent. We can. fundamental. which is contrary to hypothesis. or area in fig. note that the rays bounding the region 6 composed of congruent points. The 0. 14. Fig. Fundamental Region is for the Modular Group. 6 and 7 are congruent to the shaded regions and any one could be employed as a fundamental region. 6. Fir. "for the group (z. 7. No region . We in fig. It is clear that the mental region is fundanot uniquely determined. plane) can be carried into some point in the region by such a rotation. only one of these rays should be considered as belonging to the region. A similar remark applies to fig. 7.— The transformation T={s. II region adjacent to it contains points congruent to points in the given region is called a fundamental region for the group. for these two points.22 AUTOMORPHIC FUNCTIONS [CH. u. z of the doubly periodic functions + mu + m'w'). group of six rotations (s. For example. to ¥ CO are . the unshaded regions in figs. also serve region. Strictly speaking. We note also that any region congruent to the fundamental region will as a fundamental Fig. vertices of the parallelogram are the points qj . The shaded region. point can have two congruent points in the fundamental point. The new region is then gruent.

2 2 Consider the region lying outside the unit circle x +y = l and between the lines x = ±l. 8. The translation T carries it into a congruent region abutting along the boundary x=-\. The transformation S = (z.= 13. Hence the given region satisfies the second part of the definition of a -1 fundamental region. The transformation S carries it into a congruent region abutting along the circular portion of the boundary. 2/0 y . 14] GROUPS OF LINEAR TRANSFORMATIONS in 23 translation the positive direction parallel to the real axis through a unit distance. up into the real and imaginary parts — 1/z) breaks & + y* This is u? + ti- equivalent to ll = . a reflection in the imaginary axis followed by an Fig. . This is the shaded region in The translation T carries this region into a congruent fig.z . " v+ ^L y<f xo = . 8. one abutting along the boundary x=\. •v + y<f that is. inversion in the circle whose centre is at the origin and whose radius is unity.

transformation carrying the fundamental region into an adjacent R carries the three regions surrounding it into three regions surrounding R. found that the transformations. (3) |e|^2. The point z in the given region lies outside the unit circles drawn about the integral points on the real axis. The determinant ad= is d= + 1. T" 1 S. II It remains to show that there are no two points of the fundamental region congruent. since c 2 4. z + n) where n is a This translation carries a point of the region into a position outside the region. we can write the transformation z -a/c= \z' l/a(cz + d). a and d being integers. It is thus outside the given region. then \z + d/c\>±. The points of above the line y = \. The points and d/c lie the given region lie on the real axis. and | z' lies inside the unit circle about ±a. (az 1 + b)/(cz + d)) gives a = : = 0. It is necessary to consider three cases. To do this it will be proved that the transform of a point of the region lies outside the region by every transformation of the group except the identical transformation. | + a\--=l/\z±d\. The ^ We . according to the value of (1) c in the transformation c (z.— 24 AUTOMORPHIC FUNCTIONS [CH. The point z' is below y = $. If c 4= 0. and z'^a is the distance from z' to ±a. (2) c=±l. a/c \z'~a/c\=l/c*\z + djc\. \z'-a/c\< 2/c 2 ^|. The given region then contains no two congruent points and is therefore a fundamental region. Since we have proved that the modular group has a fundamental region. There are transformations of the group carrying the fundamental region into these regions adjoining region R. carry the fundamental region into congruent regions adjacent to it. T. | I Now z±d is the distance from z to the point ^d. we have established the fact that it is a discontinuous group. Hence when z lies in the given region no transformation of the group carries it into some other point of the region. . \z±d\>\. Hence | | z'T« < 1. and is therefore outside the given region. The transformation positive or negative integer. then a translation (z. Then.

Continuing this process. 3. and so on. 333. The fundamental region and its reflected region are inverses in the real axis . The real axis is then a. The points on this axis are for this reason called the points of discontinuity of the group. we cover the whole upper halfplane with regions congruent to the fundamental region. . Fuchsian Groups. Infinitely many regions cluster along the real axis. — 25 14. This group is a member of an important class of groups having the property designated and called Fuchsian groups.— modular group being real. shows that the transformations of a Fuchsian group are 15. but will be retained here. (1881). 8. Comptes rendus. or Groups with Principal Circle. p. any real point is transformed into a real point by a transformation of the group. We and by Theorem 10(6) the various transfoums of the region in the lower half- plane will be the reflection of the corresponding regions in the upper half-plane.* Definitions. and 5. The fixed circle is called the principal circle of the group. * This term was introduced by Poincare. A reference to the article on Types of Linear Transformations and to figs. 4. If all the transformations of a discontinuous group leave one and the same circle invariant. where the fixed circles are drawn. . and if points in the interior of the fixed circle are transformed into points in the interior by each of the transformations. It is not very appropriate. This partition of the plane is shown in fig. 15] GROUPS OF LINEAR TRANSFORMATIONS There are transformations of the group carrying the fundamental region into regions adjacent to the new regions. get a similar division of the lower half-plane by a reflection in the real axis. fixed circle for each transformation of the group. The coefficients of the — : (1) hyperbolic transformations with the fixed points on the principal circle (2) elliptic transformations with the fixed points inverse to one another in the principal circle (3) parabolic transformations with the fixed point on the principal circle.. xcii. the group is called a Fuchsian group.

transformation into its real and imaginary parts: . II loxodromic transformation can belong to the group for the only loxodromic transformations possessing fixed circles transform the interior of each of the circles into its exterior. which carry the interior of each fixed circle into its exterior. None of the points congruent to a lies in its neighbourhood.26 AUTOMORPHIC FUNCTIONS . ad-bc<0. the interior and exterior of the fixed circle becoming the two halves of the plane separated by the real axis. Since the group is assumed to be discontinuous. c. If z any real point is trans- formed into a real point. Let a be a point of the plane at which there are no infinitesimal transformations.hc)y ~ (ex + d) 2 + cty ' 8 we see that the necessary and sufficient condition that y' have the same sign as y. the determinant becomes +1. _ ac(x2 + y 2 ) 4. by a transformation No which carries the fixed circle into the real axis. and hence z' and z be on the same side Dividing numerator and denominator by the real quantity J (ad — be). The requirement that the interior of the fixed circle should be transformed into itself (and. = (az + b)/(cz + d) is a transforma- d are real quantities. The real axis being the principal circle. It will simplify matters to transform a given Fuchsian group. book that the determinant can the determinant becomes +1 an imaginary quantity.* of the real axis. [CH. "' V _ (ad . he). in the manner described in Section 10. and the coefficients no longer however.(ad + bc)x + Id (cx + d) 2 + ehj2 ~ . a. or at most contain a common complex factor which can be removed by division. -it . Any Fuchsian group can be thus transformed into one with the real axis as principal circle. b. Consider a Fuchsian group with the real axis as principal circle. We shall suppose a to be in the upper halfplane. tion of the group. such a point exists. the coefficients remaining real. of course. 16. is that ad — bc>0. It is easily seen that the real transformations with negative determinants are the loxodromic transformations of angle mentioned in Section 6 (c). only on division by \J(adremain real. There will exist a small region R about a all of whose points are transformed into points outside R by the transform a* — It was pointed out in the early pages of the If. Fundamental Region for a Fuchsian Group. the exterior into itself Breaking the also) gives a further condition on the constants. always be considered +1.

we could add to the given region a small adjacent region at b without introducing points congruent to any in the fundamental region which is contrary to hypothesis.15. if That R is sufficiently small. 9). by S into CD containing b'. . Call Let AB be the part of the boundary this transformation S. we get a portion EF congruent of the boundary containing c as interior or end point common boundary. no two of its points are congruent. 16] GROUPS OF LINEAR TRANSFORMATIONS 27 tions of the group. transformed is which containing b — is carried by S into an adjacent boundary. Let b be a point on the boundary of this region (see fig. is. 1 Now let R be enlarged by the addition of adjacent regions so long as this can be done without adding points congruent to each other or to points already in the region. CD the the fundamental region into an adjacent region. 9. of the boundary on one or both sides of b into some boundary point b' and a portion of the boundary near b'. AB being The fundamental region Considering a point c not on AB or CD. There is a transformation of the group carrying b and a portion as as possible in this way. and no adjacent region can be added without introducing points congruent to points already ia the region. If this were not the case. much for the group for \ / Fig. Similarly S" 1 carries common the being region. . the identical transformation excluded. When R is enlarged we have a fundamental region no two points of the region are congruent.

EF. and it is with this case only that we shall be concerned. It will be proved later that any axis. Since angles are preserved and the x-axis is transformed into itself. . this circle is transformed into a circle orthogonal to the x-axis which we know passes through C. HG by a transformation T of the HG contain no parts in common common group. D. 9). We propose now to simplify the region by replacing the boundary by another whose nature we know. finite number of pairs of congruent sides. . the fundamental region can be chosen as a circular polygon. the bounding circles having their centres on the x-axis and being arranged in congruent pairs. for a point lying just outside the region near a portion of the boundary would have two congruent points inside the Continuing the above process. It is with its centre on the x-axis. as in the case of the modular group or it may lie on both sides In the latter of the real axis and contain a portion of the axis.28 to another portion AUTOMORPHIC FUNCTIONS [CH. II Further. We shall speak of the transformations of the group by which the congruent pairs of sides are connected as the generating transformations of the group. the fundamental region. But it should be borne in mind that both portions belong to. case we need consider only that portion lying above the axis.] We can replace the former boundaries A6B and Cb'D by these two congruent circles. for any portion of the original region removed near one boundary is replaced by a congruent region near the other boundary. AB. conclude that the fundamental region is bounded by a finite or infinite number of pairs of congruent curves. The modified region is a fundamental region. we get the entire boundary composed of arcs of circles. This circle is perpendicular to the flection. The fundamental region may lie entirely above the real axis. we region. Hence. B is orthogonal to the x-axis. Let a circle with its centre on the x-axis be drawn through A and B (fig. which is impossible. CD. transformed by S into the circle through C and D [The circle through A. for the remaining portion can be derived from this by a re. Almost without exception the Fuchsian groups that have been investigated are those in which the fundamental region is bounded by a. Proceeding in a similar manner for the other parts of the boundary. Concerning the nature of the boundary of the above fundamental region we know nothing.

11. C being the transform of E. . there may be several. Each set of congruent vertices of the funda- — mental polygon is called a cycle. each of which lies on two sides. A In cycle may is consist of one vertex or of several. The two circles there are tangent. F. A vertex lying on the axis as A is called a parabolic point. The Cycles. Hence B. 17] GROUPS OF LINEAR TRANSFORMATIONS 29 transformation of the group can be expressed as a combination of these generating transformations. Corresponding to a vertex there may be one congruent vertex. 1 transform of A. 10. AB is transformed into ED. A being the transform of C.16. E being the Fig. The vertices A. D. C are congruent vertices. however. B. . T. Since the points congruent to a point on the . S. vertices of the polygon. the generating transformations. to 17. being each point on the boundary there corresponds in general one other point congruent to it and lying on the boundary. 11 there are three cycles: AC. B. Performing U. This is not generally true. fig. Per- U forming the transformation S. Performing T" EF goes into CB. CD goes into AF. D form a F. Starting _1 S" 1 we get the vertices with B and making in succession T. and constitute a cycle. . 10 shown a fundamental region in which the congruent sides being are connected by arrows. and D. U second cycle. In fig. or there may be none. of the pairs. —The sides of the fundamental region arranged in congruent Fig. Definition. E.

A„ be the vertices of the cycle. . . . . R x has a vertex of angle A„f Theorem 13. . .. Let T 1 be the Let A v A 2 transformation by which A x is transformed into A 2 a side adjacent to A 1 becoming a side adjacent to A 2 T 2 the transformation carrying A. * The treatment here departs from that of Poiucare in the papers in the early volumes of the Acta Mathematica. One of the vertices of the fundamental region for the modular group is a parabolic point —the vertex at infinity in the shaded region of fig. It constitutes a cycle. . .. .. 8.. II 30 axis also lie on the axis the vertices of a cycle containing a all parabolic point are parabolic points. . is one of the generating transformations or its inverse. The cycles of this kind are " open cycles " in Poincare's t No confusion will arise by representing by A. Consider the n regions congruent to the fundamental region by applying to the fundamental region the n transformations of T„ T. . . T„T. and an adjacent side into A 3 and an adjacent side and so on to T n which carries A„ and an adjacent side into A 1 and an adjacent side. 12 is an example of a fundamental region lying on both sides of the principal circle. . . T„ carries A„ and an adjacent side into A x and an adjacent side. Each of the transformations T. where the idea of the cycle first appeared. The region then has eight vertices and four cycles. . fundamental region for Weber's group mentioned vertices. . 12. Hence R t has a vertex at A x and adjoins R along one of the sides terminating at Aj. The region considered is that lying within the large circle and without the three small circles. Fig. . . .T t T„ the group: T„. the angle at the vertex A.. .- terminology. . . Since angles are preserved.• — AUTOMORPHIC FUNCTIONS [CH. It is the later.-.-u T 2T r Call the fundamental region R and the congruent regions just mentioned R 1( R 2 R„ respectively. .* This particular region has no Two useful theorems connected with the cycle will established — now be The sum of the angles in each cycle not containing parabolic points is a sub-multiple of"2tt. Fig. the four segments of the axis forming part of the boundary. There the region considered would be only that part above the axis.

A„ zero. 13. and the transformation — . A..— 17] GROUPS OF LINEAR TRANSFORMATIONS 31 at Ar The transformation T„Tn-1 carries A„_. If the anofle at the point A is not yet filled. . of sum the and Theorem 14. . . . one of these angles would Fig. the vertices at A having the angles A n _». . A 1. To each cycle not containing parabolic points there corresponds a relation between the generating transforma. none of the angles bounding sides are circles with and no two such circles can be i. T„ carries R and R' into R adjacent to Continuing... by the transformations into x 3 1. T„ is . all points are unchanged. .m coincides If R„. A 1 of R. when the transformation (T„ region R. Ar Further. we continue with the regions derived from R R«+:. . . But if three or more points are unchanged. angle A x for the points in these angles are congruent to points A„ of .„ R 2„. and the transformation it. where p is some integer. to . R 2 adjoins Rj along a side terminating at A. .T. is angles the 2ir/p. Now. each of the angles appears once. we get the regions R R„_ R„ as shown in the figure. above process the angle about the point is eventually filled.. T. can overlap the original . since otherwise the angle A t would tions of the group. have just found that We /' . for T„_ 1 carries R into a region R' adjacent to R at A„_ 2 t . contain pairs of congruent points which is impossible since A x belongs to the fundamental region.. R x a • T„ . and the sum is if R is sum 2-nthe and 2n appears each general in p times. none . with the angle A 1 of R each appears twice. by the on the axis. T/rx. for the their centres on the cc-axis etc. . in the angles A 2 region R and any original the . A. _. . A„. of the angles A 2. points common to the angle A l and have another congruent point in the fundamental region. Now. or R. . has the angle transformed the n T2 T 1 ) A in its original position. . 2x/2 which was to be proved.. The points in this angle must coincide with their original positions. tangent at a point not lying Hence. . which Hence the angle is impossible. is applied to R .

. As has been pointed out. For let AB be the side of the polygon R which is carried by the transformation lr . the generating transformations and transform it by means of their inverses into regions along its 2n sides. This can be verified directly from the . ." 1 Let T' be some transformation of the group. . Let T k be the generating transformation carrying adjacent to R along the side AB. . There is a transformation of the group transforming R into a region R" adjoining R' along the arc A'B'.. T 2 Tj) p = 1. of that part lying above the axis) cover the entire upper this. A returns to itself by the transformation T followed by S whence (ST) 3 = STSTST = 1. To prove be shown that any given point Q in the . . into Rfc R R ° into R".„ T^ 1 T. and let R' be the region into which it transforms R Let A'B' be any side of the polygon R'.— Let . . which is a transformation of the group carries R T' into A'B'. Tv T 2 the n . The point C (fig. if the region lies on both sides of the s. II the identity... 18. A and B make a cycle of angle 2tt/3. Unless the group consists of only a finite number of transformations we get in this way infinitely many transforms of the fundamental region. R . .-axis. . Starting with the region adjoining it R we .. equations of the transformations. . a relation between the generating transformations of the group.32 is AUTOMORPHIC FUNCTIONS Hence (T„Tn -i .— The transforms of the fundamental region in the above manner (or.. and so on. regions adjoining these new We can then transform R into regions along each of their sides. 8) constitutes a cycle of angle 2tt/2. is ? of the plane The question now arises How much covered by these repetitions of the fundamental : region Theorem 15. . . Then T' carries t into a region R" adjacent to R' along the side A'B'. Therefore T'Tj. The sides at C are connected by the transformation S whence the second theorem gives S 2 =l. R can be transformed into a region adjacent to it along any one of its sides by means of one of the transformations T T. These two theorems can be illustrated by the modular group. it will half-plane. . The Transforms of the Fundamental Region. T„ be the generating transformations connecting pairs of congruent sides of the fundamental region. [CH.

PQ lies entirely above the real axis let . —2 CiZ di\ Cfi +d i \ = (c x + d 2 + c y 2 . 14). 4 * In case 1'Q crosses a vertex of one of the polygons a slight deformation of the line will make it cross a side. if 2 t a^— t) D ci 2 t this length is ° Jo —— + \ \dz\ -4.* x Consider CJDi.Di . save for a finite number. R R (. be the transformation by which is congruent to an arc CD of joining two sides of R i The length =l. Their sum - . Let its be the side of the fundamental region from which the line (fig. etc. Now this latter sum is infinite.& 5 18 1 GROUPS OF LINEAR TRANSFORMATIONS 33 upper half-plane can be reached by repetitions of the fundamental region according to the method indicated in a finite number of steps. 1 Theorem | 12. Then C. Let M be its maximum distance from the z-axis. K be minimum distance from the axis. Let is z' = (a z + b )l(c z + d ) i i l i s derived from R . 3 . Let P be a point in the fundamental region. For the only case in which it would not be infinite is that in which all the arcs CD. The assumption that the number of segments C Di is infinite leads then to the conclusion that the CD. Construct the region R x adjacent to R along the side AB. Suppose there are infinitely many segments CiD4 is greater than K/M times the sum of the lengths of infinitely many arcs CD crossing R from side to side. By Now. dDi f / is I \dz'\. We shall suppose for the moment that the fundamental region does not extend to infinity. Let PQ leave R by the side x V Construct R 2 the transform of R adjoining R along the side A^. the portion of the line PQ lying in the region AB PQ issues \B . which is impossible. and from the equations of Section l^W'^iarc Therefore CA> s/ D| ^l = M x len s th of . Q by a straight line. and join P. cross the region near the vertices of the polygon but this would mean that PQ winds round a point at which vertices meet.

this new region . This is the case in which the fundamental region is that between two vertical lines. R must coincide with one of the regions R* already found for if there were overlapping. where Tt is the transformation carrying R into R4 and is a combination of the generating transformations. the " period strip " of the simply periodic functions. Whence T = T*. This proof does not hold if the fundamental region extends to infinity. . . In that One or case consider the congruent regions adjacent to R more of these adjacent regions will not extend to infinity save in the one case in which the fundamental region consists of only . in this case the half -plane is obviously covered. We note that all the transformations involved in these repetitions of the fundamental region have been combinations of T 1. [CH. any finite region lying . The group has no other transformations. . for in that case we cannot conclude that y. // Tv T n are the transformations connecting the n pairs of congruent sides of the fundamental region of a Fuchsian group. any transformation of the group can be expressed as a combination of Tv Tn with their inverses. two sides with infinity as a parabolic point. y is Since any point in the . . . the imaginary coordinate of a point on CD. and the transformations are real translations. take P in bounded. the half -plane will be divided into an infinite number of regions. it follows that the whole upper half -plane is covered by these repetitions of the fundamental region. Since any finite point Q above the axis can be reached in a finite number of steps. Now. upper half -plane can be reached in a Q finite number of steps. It carries R into a region R in the upper half-plane. Hence Q can be reached by repetitions of the fundamental region in a finite number of steps. and the preceding proof applies. the half-plane will be covered by a finite number of repetitions of the fundamental region. We lay this case aside. II sum sum But this is impossible since the of their lengths is not greater than PQ.. .. a finite quantity. Hence Theorem. If the group has an infinite number of If the transformations. . For let T be any transformation of the group. group has only a finite number of transformations. the region common to R and R* would contain points having two congruent points in R which is impossible.— 34 AUTOMORPHIC FUNCTIONS of their lengths is infinite. — . T„ with their inverses. is bounded. . .


18, 19]



entirely above the axis will contain in its interior only a finite


of regions or parts of regions.

Hence the points about
the real axis.


cluster points

many regions cluster lie on may comprise all the points


of the axis, as in the

case of the modular group
in the case of groups in

or only certain points of the axis, as

which the fundamental region

the axis.


division of the lower half -plane into regions can be


upper half -plane in the real axis. If R lies entirely above the axis, its reflection will be a fundamental region R lying entirely below the axis. The repetitions of R will, by
reflecting the
' '

virtue of

be the reflections of the repetitions of R If the fundamental region crosses the axis, the lower half of the region can be chosen as the reflection of the upper half and the reflection of the whole half-plane supplies the lower

Theorem 10

parts of all the transformed regions.
19. Groups without Principal group which possesses no principal




circle (called

Kleinian by

Poincare) exhibits several of the properties already found for Fuchsian groups. For example, the fundamental region can be simplified by the use of circular arc boundaries, although now

the bounding circles will not be orthogonal to a fixed circle. The sum of the angles of a cycle not composed of parabolic points will be a sub(i.e. vertices where the adjacent sides are tangent) relation between yields a multiple of 2tt and each such cycle And we can group. the the generating transformations of

partition the area surrounding the fundamental region by means It is here that we of transforms of the region, as in Section 18. copies of the fundathe however— come to an essential difference,

mental region do not fill the interior of a circle. In the simpler finite number of cases the whole plane may be covered by a clustering number infinite an regions (as shown in fig. 23), or by division is much the general in about a point (as in fig. 7), but which in that is phenomenon more complicated. A common
curve, is only a portion of the plane, bounded by an irregular necessary are regions fundamental covered and several different (just as in the Fuchsian to obtain the division of the whole plane principal circle, are the of case two regions, one on each side

usually required).






Poincar6 has introduced a method of treating these groups by means of three dimensional transformations. Each transformation is equivalent to a certain even number of inversions in circles in the plane. The transformations in the plane will
these inversions b« made in spheres with the same centres and radii and furthermore by the use of these spheres the transformations are denned for points above and below the complex plane. It can then be shown that there exists a fundamental region for these generalised transformations bounded by The parts of the plane contained spherical surfaces orthogonal to the complex plane. in this three-dimensional fundamental region furnish the several fundamental regions in the plane whose repetitions cover the whole plane.*

be the same


as the circles

* Poincare,

Mimoire sur Us groups


Acta Math.,



(1883), pp.



Forsyth, Theory of Functions, pp. 610-618.




20. Automorphic Functions.— Automorphic functions are the natural generalisation of the circular, hyperbolic, elliptic, and certain other functions of elementary analysis. A circular function, as sin 2, has the property that it is unchanged in value if z is replaced by z + 2mir, where is any integer that is, the function is unchanged in value if z be subjected to any transformation of the group (z, z + 2rmr). A hyperbolic function, as sinh z, is unchanged in value if z be subjected to a transformation of the group (z, z+2m-n-i); an elliptic function, as the Weierstrassian function p(z), retains its value under transformations of a group of the form (z, z+mw + m'w).



Definition.—J. function F(z)
with respect



to be


a discontinuous group if F

\ cz

(az — -4-b\

+ d/ =F(z),




any transformation of


It will be proved later that there are automorphic functions connected with any discontinuous group. For the present we assume this fact and proceed to derive some of the properties of the functions. We shall confine our attention to groups (Fuchsian or Kleinian) for which the fundamental region has

a finite number of sides, and to those automorphic functions having no other singularities than poles in the fundamental region. To avoid long circumlocutions in the statement of the
There are no functions, other than constants, which are unchanged in value all the transformations of a continuous group. For, let z be a point at which such a function has no singularity. Since the group has infinitesimal substitutions, It there are infinitely many points in the neighbourhood of z at which F(z) = F(z ). is a well-known theorem that a function which has the same value at infinitely many points in the neighbourhood of a point at which it has no singularity is identically equal to that value. Hence, F(z)=F(z„), * constant. It is for this reason that we confined our attention in the study of groups solely to those which are discontinuous.



where of angle 2ir/p. $(A) 0. * If the group is Fuchsian. Hence e2nUIP = l. s F(z) is said to take on its value times at a point z if it can be written F(z) = F(z ) + {z . since A' is a distant point. and we are to prove that s is a multiple of p.z).A)*0f». where (/>(z ) =t= O. the simple automorphic function is usually called a Fuchsian function. or s is a multiple of p which establishes the theorem. and. the function in square brackets can be made to differ arbitrarily little from <p(z). and let A be a vertex belonging to a cycle We can write F(z) = F( A) + (s . z is also near A. Ill theorems we shall call such functions simple automorphic functions* Owing to its automorphic character we shall have a complete knowledge of the function when we know its behaviour in the fundamental region. or some multiple thereof. If there are n vertices in the cycle. A simple automorphic function takes on its value p times.— 38 — AUTOMOKPHIC FUNCTIONS [CH. being automorphic.z ) s <j>(. is unchanged in value. at a vertex belonging to 21. Properties of Simple Automorphic Functions. A similar remark applies to each of the other congruent . A' being a point not in the neighbourhood function. a cycle the sum of whose angles is 2x/p. s/np will be said to belong to the fundamental region. It was found in Section 17 that in the Fuchsian case there is a transformation T of the group with A as fixed point which The other fixed rotates points near A through an angle 2-n-jp. F(z') Applying this transformation to the we get k) s <j>(z) = F(A) + (z - = F(A) + (z - A)»e*"*" '«<&} The function. Theorem A. — . Let the function be simply automorphic. =1= the transformation - — 0-A A. T then is point is A'. np regions have a vertex at A. =e 2. Dividing the s-fold value there equally among them. is absent and s is negative in the formula. By taking z near A. • A similar fact can be shown for the Kleinian of group.f In considering the number of points at which the function has the value F(%) the point s is counted s times.r * s-A j-. t In the case of a pole at s 0l F(z ) . the reflection of A in the real axis.

— At a parabolic point a simple automorphic function is a function of t having at most poles. In the neighbourhood of a parabolic point P the behaviour is less simple. 2nd ed. a group of simply periodic transformaFrom this we can show that F is a function of t having tions. then the integral q—-H log/0) taken around the contour is where in a counter-clockwise direction is equal to the in poles of number the the number of zeros and shall apply this theorem to the automorphic region* N-M. has as many zeros as it has poles in the fundamental region. If f(z) has no other singularities than poles in a region enclosed by a contour and if the function and its derivative are continuous on the contour and the function does not vanish m there. is unchanged by any repetition transformation: — rj= p + %/3. p. or Osgood. The number of times that F(s) takes on its value at the vertices of the cycle in the to which P belongs is determined by the index expression for F in the form F(z) = F(P) + *">(*). 21] vertices. Theorem C. We suppose first the of parts the Consider zeros nor poles on the boundary. —p = —p + . where P to infinity tions become t =e 2«7(z-. This kind of singularity is called logarithmic. not identically zero. — AUTOMORPHIC FUNXTIOXS 39 belonging to the yielding s/p (an integer) points fundamental region at which the value F(A) is taken on. 333. integrating around the boundary of the no has function the that mental region. A simple automorphic function. It —. a singularity which is not a mere pole..e. * Whittaker and Watson's Modern Analysis. /3 by which the function of this unchanged. %vVL^. N M We fundafunction F(z). Hence at most poles where t — e Theorem B.— 20. By transforming the point by the transformation Z=l/(s — P) the transformaZ' = Z + n/3. Lekrbuch der Funktioncntheorie. 119 . We can show that there is a parabolic transformation of the form is —. Let F(P) be the value approached by F(c) when z approaches P in the fundamental region. p.P)e. Since infinitely many regions cluster about P the function has there an essential singularity (i. for near a pole F(s) can take on no value more than a finite number of times). .

(s/2x)2 L A. / /A : Fig. and the integrals cancel as before. which F(s) has zeros or poles on the boundary require further consideration. fig.. 16.» . Since F(z) has the same values along the congruent sides. this sum is / d log F(z) + / d log F(z) = 0. Change the path of integration by means of two congruent arcs as shown in fig. the integral taken in the clockwise direction along the small arc shown in the figure is -s. We shall suppose it is a zero of order Let 2ir/p be belongs. the sum of the angles in the cycle to which A The integral —-. 16. Ill from two congruent sides AB. Similarly the parts arising from other pairs of congruent sides or N = M. But the function cannot have infinitely many . and N—M= The cases in A — Fig. 17.M = . A and these or N + s/p = M. since the points are congruent only one should be considered as belonging to the region. 40 integral re / AUTOMOEPHIC FUNCTIONS arising rB / [CH. Only one of the points is included in the new region and it is clear that cancel . there is one also on the side CD. 15. 15) . d log F(z) + d log F(z). A is ing to Hence N . also. Id log F(z) taken in a counter-clockwise direction entirely around is s hence. . We can show likewise that the case of a pole or a zero at a parabolic point produces no exception. 17. There will be a similar integral at each of the congruent vertices. Fig. I A/2ir. . and we see that the theorem holds here pole at The case of a treated similarly.= — sip But s/p is precisely the number of zeros belongthe region which have not been included in the contour will not cancel. CD (fig. which was to be proved. The methods above dispose of every case except those in which there are infinitely many zeros or poles on the boundary of the region. If F(z) has a zero or pole at a vertex A. we integrate as in s. If there is a zero or a pole on the side AB.

The number of its poles is . . Since C may be any constant. This a simple automorphic function with the same poles as F(z). Hence the theorem is established. + A (m+1 „„ +]) = 0. A simple automorphic function which has no singularities in the fundamental region is a constant. etc.. Av A 2. Between two simple automorphic functions belonging to the same group there exists an algebraic relation.. not all zero.C=0.. . we assume does not have and it cannot have infinitely many zeros without being identically zero. Theorem E. .1. is a simple automorphic function with a zero at z and having Q poles. . whatever values be given independent of z. A l F 1 m (ci )Fi %c i) + .— — 21] AUTOMORPHIC FUNCTIONS 41 poles without having an essential singularity which it . We are to show that there exists a relation »+ 9(F lt F 2 ) = A 1 F 1 m F 2" + A 2 F 1m F2 1 . etc. if m and n be large enough (m + l)(n + l)-l>mk1 + nk 2 and the Fuchsian function $ has . which establishes the theorem. all zero. is a the poles of F x (z) and occur where points poles other than at the where A A 1( 2 . Theorem D. the relation being function $. (m+ l)(n + 1 . not greater than mA^ + rc/c 2 A (m+1 „ B+1) so that <£ will have etc. are constants not We can choose the constants A p zeros at (m+ 1)0+ 1) .2. It has as many zeros as it has poles. (m+l)(«+l)-l. . Such a functiqn. . . Therefore. is satisfy these equations since there can always be chosen to one more constant than equations to be satisfied.1 equations ) . . Constants. is identically zero.. F(z) — C . where C is a constant. Theorem F. The function which has no automorphic simple the constants. Let F(z) be such a simple automorphic function and let its value at z a point in the fundamental region. be C. F 2 (z). . not coinciding with the . i=l.. according to Theorem C. or F(z)=C. A simple automorphic function which is not a constant takes on every value in the fundamental region the no same number of times. Now . F(z) . is Consider the function F(z) — C. +A (m+1)(n+1) = 0. points c v c 2 poles of F x (z) or F 2 0) by satisfying the . Let Fx (s) and F2 (z) be the two functions with k x and k 2 poles respectively. The zeros of F(z) — C are the points at which F(z) takes on the value C. we conclude that the number of times F(s) takes on any given value is equal to the number of its poles.

. Hence. say ^(F^ F 2 )=0. Each pair of values of F v F 2 satisfying the algebraic equation is given by a single point z in the fundamental region (save for a few exceptional points). that $(F 1. Theorem C.z) takes on that value. For a fixed value of F 2 there are k. [CH. F 2 ) is reducible: #(F X F 2 ) = # 1 (F1 Fa^Fp F 2 ) Some one at least of the <&„(Fi>F. ^ 1 (F 1 F 2 ) = is of degree For particular k 2 in Fj. (z)-c=0. F2 )=0. Let F s (z) be a third Fuchsian function connected with the group. . Such a function. the number of values of F x satisfying the equation when F 2 is given a fixed value. This value of poles respectively is of degree k 2 in . . Ill more zeros than According to $(F 1. . ible relation will contain both functions unless one of them is a constant.— — 42 AOTOMORPHIC FUNCTIONS possible poles. F3 (z). the poles having mutually cancelled. . Let us consider two functions F t (z) and F 2 (z) with k x and k 2 x > . functions to be a rational function of the variables in the algebraic equation. as in the case of F we know from the theory of algebraic 3 z gives one value only of 1 function of X. 2 . It will generally happen. It is easy to see in the general case what the degree of this The degree in F x is irreducible equation in F t and in F 2 will be. . F whose irreducible algebraic relation <&(FV F2 ) = F 1 and k^ in F 2 as mentioned in the preceding paragraph. Hence which was to be established.). Hence the theorem. . This irreducirreducible factors must vanish. k 2 as before being the number of poles of F 2 (s). in which case the degree in F l is less than k 2 Theorem G. in which case the irreducible relation is of the form . Similarly it is of degree /Cj in F 2 functions some of the values of F x arising from the k 2 values of z may always be equal (as in the functions F and F 2 ). Any simple automorphic function can be expressed rationally in terms of two simple automorphic functions connected with the group. F3 is then a single-valued F and F 2 so long as these satisfy the algebraic relation $(F F ) = 0. when it has only polar singularities. if the algebraic relation be found in this manner. For each of these values F x (z) has a value satisfying the equation.2 values of z at which F 2 (.

Then F(s') = F(z) but Let z of the ™-^-^ £-'«<«<>• and F'(z') + F(z). Theorem H. 9 v ror -j^j^rf or .. not. Concerning the inverse function zix) we I. x(z). Two simple automorphic functions connected with a group satisfy. -tft is a t uchsian function. eZF. equations dx'/dz' = (cz + dfdx/dz and 1/y^-dz/dx. This condition holds only in case the transformations are all translations. s' = z + k* We note that the ratio of the derivatives of two automorphic functions is automorphic . in general. F{(z')lF^(z') =F 1 (s)/F 2 (z).— — — 21] — AUTOMORPHIC FUNCTIONS 43 is The derivative automorphic. where 4?(w. Consider the function y 1 (z) = J(dx/dz). an algebraic differential equation of the first order.(F1 . the pre- ceding theorem this is expressible rationally in general in terms Fx and F2 Hence. the variable z not explicitly appearing. . = (as + b)/(cz + el) be a transformation ad-bc = l. in general. shall now prove of z. x)y. where of an automorphic function . We shall show 1 first 72 that transformation of the group. . class in 2 y 1 dx t^tt is an automorphic function. For convenience let us represent an automorphic function by of .F 2 ). except in the case (cz + df = l for all transformations of the group. Let us subject z to a •which the derivative . and represent by y-[ and x'( = x) Making use of the the transformed values of the functions.e. dFjdF . we obtain the following: * The singly and doubly periodic functions constitute this exceptional is also automorphic. group. x) = 0. dx 2 R being rational in w and x and 3? a polynominal. „ By . the following remarkable theorem Theorem If x(z) is an automorphic function the quotient of z can two solutions order of the form of a linear differential equation of the second be expressed as -r-% a function of x by = i?(w. dF1/cZF2 = B. i. _ .

. ce~ . dx Whence. integrating.. —\ = U(w. where <£(?». d d!K = (« + d) '^1 +C + c. and w and x are joined by an algebraic relation $(w. In this case P(x) = 0. The theorem is thus established. 1_ cly 1 _yx 2 //j <fyi y x dx dx = _l2 = )/j c ^£. or where y a is a solution of the equation. 1 //j' dy_l <. Series. first y 2 be a second solution of this equation independent of y 1 and i/ s satisfy an equation of the form y1 d]h_. Now the let . We 22.*] Dividing by y^. dy1 = ^ j dx ' dx ' where c is pression equals a constant different from 0. satisfies an equation Whence i/. x)y. [In general the ex/Pdx where ¥(x) is the coefficient of dy/dx.Il-' K 2 dx' 'dx> dxdx 2 y^ dx = (cz + dfp ' dx Hence. The Thetafuchsian * Forsyth.44 AUTOMORPHIC FUNCTIONS [CH. x) = 0. 4th ed. note that two independent solutions of the equation are y = J(dxjdz) and y=z J{dxjdz). p. — We have heretofore assumed the existence of functions automorphic with respect to a given A Treatise on Differential Equatioiis.e' eft/! i/j dx 2 This function being automorphic. 114. is expressible rationally in terms of x and some other automorphic function w. x) = 0./. Ill dx dx v dx J \lx dx d Il V yl .

p. The first of these factors. be subjected to a If z z. We shall suppose the identical transformation to be z = z. being independent of is the same for all terms. - The or {CjZ + d r(c . with centre at the origin be the principal let group «£ and i +h = Q] T 2) 3.* We treat here the case of Fuchsian groups.2 + <*. principal circle. . (1881). 333. We shall now demonstrate their existence. vol. aA-6 A =l be the transformations belonging to the group. p. i. then is But the expression under the sign of summation is 9(s) all . Let H(z) be a rational function of z none of whose poles lie on the principal circle.21. provided Let us assume the series is called the Thetafuchsian series. Acta Mathematica. and in fact construct such functions. the series becomes the transform of the point z when the by the ^-transformation is made. the followed ^-transformation If we represent by % second factor of each term of the series firstfactor is i is H( ^±^t±^±Ml^ The series ^+ /j . for the {-transformations of the original series included * Oomptes rendus. 22] AUTOMORPHIC FUNCTIONS 45 group. by means of a series discovered by Poincare. i. We consider now the following series ®( z) = 2(c i2 + ^)—h(M±*A.)-. Mtmoire sur 193. (1883). . transformation of the group — iajZ + b^KcjZ + dj).. xcii. shall prove m convergence of the series and investigate the properties of the function 6(s) which it defines. . NCjZ + 'fj/ i We for certain presently that this series is convergent (save exceptional points) both within and without the This is an integer greater than 1. les the trans- /auctions fuchsiennes. . Let the unit circle of the circle .

Hence. at points congruent to the point infinity. A (a z + b i )/(c z l i term becomes infinite also when z = — djc that is.. Similarly let a large circle K be drawn with no point congruent to infinity on its exterior. a term becomes infinite whenever z approaches any point congruent to a pole of H(c). the integer 2 2 Consider the function F(z) = Qi(z)/Q 2 (z). we found that its derivative is multiplied by (cjz + dj) 2 on performing a transformation of the group. That is.. the function will be automorphic if F(s) is It is easy to m m . By means of Thetafuchsian series we can set up infinitely many automorphic functions. m + ?n + Convergence of the Series. F(r) (^+d r® J 1 (z) then an automorphic function. . and the denominator be also 2 v a sum of such terms. and we can readily deduce other combinations which are automorphic. Since each transformation carries { . +in k is the same for all the terms. ®(zJ ) = (c/ + dj Y'"®(z). . Ill formations of the group. —There are certain points of the plane at which individual terms of the series become infinite. H. form other combinations of G-series which are automorphic. {z) ^e ^ wo ser ies 1 formed from two rational functions m being the same. If F(z) is 1 2 an automorphic function. the order of the terms being changed. and let the region outside this circle and the congruent circular regions about the points — di/ct be excluded. Hence Q(z)/[F'(z)] m is an automorphic function. . there is an infinite term when + di) = a. If a is a pole of H(^). The portion of the plane that remains that inside K and outside the circles about the other excepted . 23.46 AUTOMORPHIC FUNCTIONS [CH. t (z) and H (z). the exterior of the circle into exterior to the circle. Thus if the numerator be a sum of terms of the form 0J02 Ok where the integers used in forming the series are m* respectively. these latter points are all Let a small circle be drawn about each pole of H(s) and let the regions within these circles and all the circular regions congruent to them be excluded from consideration. . and the ^/-transformations of the last series include also all the transformations of the group. itself. Let 9 (2) and Q.

./2 i { Ci z + dy-m K{z. 2. . ) = S™' H.(z ).nd \z + djc is the C distance from the point z to the excepted point ~-d lci This excepted point lies within the circle K and without the region If D be the greatest distance from a point of C to a point C on K. If z is in C each transformed point z lies outside the excepted regions around the poles of H(s). area than the less is A areas A the v /D A whatever the value of i. M < dm /D. C and let the Applying the formula 5. . nit > 1/Cfd whence Let C 1. . and vii be the maximum and minimum values of c. and the series UJ + U2+ converges.H#/DiA )(A +A + 1 . . 2 . M ^ + M 1*+ Hence m/<A.s + d )"2m 'H.1 = \Ci\. .1 z + di/dT 1 a. or it may contain a portion of the circle on its interior in case the fundamental region of the group contains congruent. . and d be the least distance from a point of C to any of < l/c D and the circles surrounding the excepted points. • • be the regions congruent to . We C may a portion of the We shall now prove that within C the circle. .M. ./D*A 4 4 since the sum of = quantity. <S^. of K. areas of the regions be A found for areas in Section A A 1( 2 . A )(A + ) + 1 <(d /D A Hence M«<S*. To do this we shall prove that whatever value z has within C the absolute value of the term (c. Let M. This inequality is true on the assumption that Consider now B. l \ i . . 4 4 2|/ i WA 4 i . and containing none suppose also that no two points within C are lie entirely within the principal circle. 2. . and M "" = M/"'.) C less in absolute of z within .*"H. . . . Hence greater than the H(>i) is finite and we can choose a constant excepted regions. . | t | . or entirely outside it. finite a S. i i ] .(z ) is less than a constant XL. we get the following A ( =| | | c tz +d t \-*dxdij > II mfdxdy = mM . ( .z + d -1 in Now \c z + d \. 23] AUTOMOEPHIC FUNCTIONS is finite 47 points — in area. Thetaf uchsian series converges uniformly. . C ./A and M/<^A. t 4 2 . we have t i . Let C be any contour lying within of the excepted regions on its this region interior. M . But of the series. . - Now.— 22. than the constant 2M 2m H<(S |m . whatever the value M. . the outside point every for function value of the the convergence establishing for material the now We have t t m> H is Each term value.

If H(a) has no poles within the principal circle. cited above. 48 AUTOMORPHIC FUNCTIONS series [CH. formed with a given value of m. 0(a) then has a pole at the excepted point. In a fundamental region outside the principal circle the number of poles of 0(s) is equal to the number of poles of H(z) outside the region increased by 2m. We conclude from this that for all functions Q{z) formed with a given its zeros arise from the zeros of the numerator the denominator. Hence.e. There is a point in the fundamental region congruent to each pole of H(a) within the principal circle. . and the order of the pole at their common congruent point in the fundamental region is not greater than the highest order of the infinite. z* of the excepted points a term of the series becomes but if this term be suppressed the series converges. Let us consider the function in a fundamental region within the principal circle. the latter arising from a term (c z + d )~'lm at a point congruent to infinity. the number of poles of 0(z) within the fundamental region is in general equal to the number of poles of H{z) within the principal circle. The poles of the automorphic function F(s) = 1 (z)/02 (0) arise from the poles of the numerator and the zeros of the denominator and the poles of F(z) has the same number of poles as of zeros in the fundamental region. Ill The 2M<H thus converges . In particular cases two poles of H(a) may be congruent points. verges absolutely and uniformly in analytic function of hence the original series conC and the sum Q(z) is an . the function m circle of a similarly formed series for a group without a principal can be established by slight and obvious changes of the proof just given provided infinity is not a point of discontinuity of the group. for a group whose fundamental region does not cross the principal circle]. In a group having a fundamental region crossing the circle the number of i l poles is the sum of the two preceding numbers. The series is called * The convergence Thetakleinian. The function has essential singularities at the points of the principal circle where the excepted points cluster.. Now. or if the poles are at congruent points and are of such a nature that the poles of the infinite terms arising from them cancel. that of all such functions. in the memoir larities within the circle. At one two poles. only a finite number are linearly independent [i. 0(z) has no singuPoincare has shown.

points at which infinitely many fundamental is regions cluster every point of the functions defined inside. by the same * Poincare proved that in the case of a fundamental region lying entirely within the principal circle. the two functions.1 . . the number of zeros exceeds the number of poles by m(n . are distinct functions. where 2» is the number of sides of the fundamental region and S is the sum of the angles of its vertices. circle an essential singularity for the by the series outside the circle and the function Each of the functions then has the circle as a natural boundary. defined Since analytically into the function inside cannot be continued the one outside. although series.— 23] AUTOMORPHIC FUNCTIONS 49 is the difference between the number of zeros and of poles constant.S/27r).* a It should be noted that in case all points of the principal circle are points of discontinuity of the given group —that is.

y{ = aj/ 1 + 6i/ ? y^ = cy x -\-dy 2 z = yjy. solutions and let z = y(\y 2 be the quotient of two other solutions. p. . this linear relation will vanish only if be other branches of y x and 2/ 2 in The determinant of y( and y 2 differ by a constant factor. Journal fiir Math. which we have already noted Let in Section 21. By the ordinary method of the substitution of a power series- in the differential equation and determining the 0. 2 . Theorem I. be the quotient of the two. oo . . The Differential Equation. We shall consider the singularities and with the all equation can in hypergeometric equation with three exponent differences X. — Given a linear differential equation of the second order 1 eh- dx . we find that if a is a point not coinciding with solutions of the form ubcr die or oo . 1. The cases be reduced to the following real an equation whose singularities are at the points 0. Then ' 2 = a 'Ji + b lh = a" + b +d cy l + dy 2 <:z _ In particular y x and y 2 ' ' may case these solutions are not single-valued. 75 (1873). Reihe in his Werke. fx.. arises in the following manner. coefficients. there are two independent * Riemann's Vorlesungen hyperg. Nachtrage - r Schwarz. v. 1. and two independent particular solutions y = yA {x) and y = y z (x) The conthen the general solution is of the form y = Ay 1 + By 2 nection with linear transformations. 50 .— — CHAPTER IV THE RIEMANN-SCHWARZ TRIANGLE FUNCTIONS* 24. 292. .

and the coefiicients in P x and P 2 are real if is real. 25. between The function za = y . the real axis near a is transformed into the real axis in the z„-plane near the origin. By virtue of a remark in the preceding section. the coefiicients in the series P3 a . Such a region is the upper half-plane.i=l+P. . and we now propose to find into what region in the z-plane the upper halfplane is mapped by the function z = yjy r and 1. we get the line 0-1 becoming. a linear transformation by which the Hence.0 we find the =1+P 4 (*) 2/1. Consider a point a on the real axis. say. solutions series in x — a having no Similarly at the singular points y Ii0 -a*{l + P s (a:)} 2/2. Conformal Map of Half-Plane. the real axis near a is mapped on an arc of a circle in the z-plane. If the series for z a does not converge throughout all the region consider z b the ratio of the principal solutions at a point such that the region of convergence of the series for z„ overlaps that of the preceding series. . we b. hence.Jy2ia = (x — a){l + 'P(x — a)} maps the neigh1 bourhood of the point x = aon the neighbourhood of z a = 0. real axis in the z a -plane becomes a circle in the z-plane.. a = (x-a){l + P (x~a)} = 1 + P 2 (x .. a circular arc in the z-plane.(as-l) q\Ki-\-n-0 \x) Further. P8 are real. 25] THE EIEMANN-SCHWAEZ TRIANGLE FUNCTIONS 2/t. by of convergence. — 1. We get the axis at b mapped on an arc in the z-plane. z = (a 1 z a + b 1 )/(c 1 z a + d 1 ). . z is in general a many-valued function of x.a 51 y..24. The coefficients of the series Y(x — a) are real. since it coincides with it in the map of the common region Continuing.— Let z = y l (x)/y 2 (x) be the ratio of two linearly independent solutions.1 =^{1 + P 5 (a. the transformation z = yjy 2 .-1)} y. but in a region not surrounding the singular points any branch of the function will be singlevalued. . It is a continuation of the arc previously found.a) l where P x and P 2 are ascending power constant terms.

x does not. two solutions of the preceding Taking the real branch positive part of the real of the function. pur. the region in the upper half -plane at x = being transformed into the region inside this angle. when x>0. /jl-w. vtt.$<1 at an angle p. conjugate imaginary points in the as-plane become conjugate imaginary points in the s„-plane.. the map of the lower half-plane just found is derived from the triangle Zoz iz a. Let us denote the vertices of this triangle by z z v z m starting at a. using the branch of the . = x^{l + F it} (x)}. we had traversed the real axis passing below the singular points we should have had a map of the lower half of the plane on a triangle z$z x z ^ with angles Xtt. x>l and x<0 „. a consideration of the 2/ L y2 . If. The axis x < becomes a line in the 0-plane making factor e an angle \tt with the positive real axis. „ shows that the two arcs meet at an angle result that Thus we have the The ratio of any two solutions z = y1 (x)/y2(x) maps the upper half-plane of the variable x on a circular arc triangle in the plane of z. But when x passes above the origin and then traces the negative portion Although P (x) of the axis. By means of z = (a z + b )/(c Q z + d ) this line becomes an arc in the z-plane making with the arc previously found an = . Whence x K = p e ".— 52 AUTOMORPHIC FUNCTIONS [CH. loo . 3 is a many-valued function of x. s is in general no longer real. Its angles are \tt. the sides of the triangle being the transforms of the lines 01. . z >0. Similarly we can show that the part of the real axis for x> 1 is transformed into an arc in the 2-plane meeting the transform If we continue the mapping of the axis 0<. of the parts solutions j/7t. vtt. .ir.. are series with real coefficients. the side z z l being common to the two triangles. Since each of the auxiliary functions introduced. etc. These go by the linear transformation into pairs of inverse points with respect to the circle into which the real axis is transformed in the z-plane. Hence. If we start with a point on the negative part of the axis. and oc 0. containing no fractional powers. and z has the iK ". to infinity. . IV At the section is origin the ratio of the z = y hl)/y. °V an inversion in the side z^ v ' (j . angle \tt. It is easy to find the relation between these triangles. the axis near x becoming the positive part at z = 0. z a z etc. Writing x = peie the circuit of the origin gives x = pe™ on the K ix negative part of the axis. remains real. Now.

of the other half -plane are in general infinite in 26. adjacent unshaded (shaded) region by inversion. . A similar remark applies to the other vertices. 26] THE RIEMANN-SCHWARZ TRIANGLE FUNCTIONS . Hence. is f(z) a single-valued function ? a ready answer. v are the reciprocals of vir are submultiples of it. the angles of the triangle \tt. passing below the singularities. or A.. in finding 2/1.25. be a submultiple of 2tt. other of the plane. If A = 0. and in order that there shall be no overlapping by continued reflections it is necessary and sufficient that in the sides meeting at z stances Under what circumThe triangles furnish .0 the undetermined coefficients yields but one solution.. the that A second solution with a logarithmic term can be found.0 and 2/2. It is that there must be no overlapping. These triangles The Triangle —The relation z =y l (x)/y. any shaded (unshaded) region is derived from any other shaded (unshaded) region by an even number of inversions that is. By some a transformation of the group this region is carried into satisfied) . for a point z belonging to two maps of the £-plane gives two different values of the function. by a It is easy to see that these transformalinear transformation. if z and z are any two of the same point in the x-plane.(x) defines x as a function of say x =f(z).„. by inversion in a side. are and »„*„ sides triangle _ the Vl possible also to have zero values. and we can show angle of the triangle is The at tangent . series of method of substituting a W m thus Air as before. we get a triangle which is an inversion of the original triangle in the side z z a mapping In general. /ait. . tions form a group (both group properties of Section 8 being 2 Air /j. At z the two triangles mapping the upper and lower half-plane form an angle 2\tt. there is a branch of the function either half -plane on a triangle derived from the map . congruent points being the transforms Hence. and that a shaded and an adjacent unshaded triangle together form a fundamental region for the group. number and overlap. Hence. f(z) is automorphic with respect or =/(s) points map congruent * It is f(z) . Function. z. The fundamental region is a map of the whole a-plane. 53 function which mapped the line Ooo on z z m and map the lower half-plane with this branch. The shaded and unshaded triangles Each shaded (unshaded) region is derived from an alternate. integers* For convenience let each triangle which is a map of the upper half -plane be shaded.

\ + + v = 1II. There are four possible cases in which \. X + m + "< 1 CnseIIL\ + + v>l. The Three Kinds of Groups.* On trans- * It will be observed that by combining several of the triangles in this case we can get a parallelogram whose opposite sides are congruent by transformations of the group.— 54 to the AUTOMORPHIC FUNCTIONS group. z to infinity. Thus in fig. less ir.—The groups arising in connection with the triangle functions exhibit strikingly different characteristics according as the sum of the angles of the triangle is equal than.— reflec- In this case the third side is also a straight line. IV This automorphic function is called a triangle function. 18. 19 the twelve triangles having either a side or a vertex . . then into the three cases : Fig. h. fx. (iii. /ul - iuL z meet again at z.) is h i> 4 > and (iv. (ii. 27. 18 is shown the character of the third side in each of the three cases. then become In fig. Let a linear transformation be made carrying the sides Two of of the triangle straight lines. 28.) (ii. [CH.) The division of the plane for shown in the accompanying figure. the triangle meeting at. Case Case I. say. for purposes of study. v have the values (i.) A. : Fig. We divide the groups. or greater than to. 19.) i. Case I. It will be a simplification the z-plane as transform to Let the sides of follows. \+p+ v = 1. It is obvi- ous that by continued tions in the sides the triangles are infinite in number and cover the whole plane.

\ + n + v<l— The third side BC (fig. 18) convex toward A. group. liecuexl Math. the many common 29. 424.26-29] THE RIEMANN-SCHWARZ TRIANGLE FUNCTIONS infinitely is 55 forming back to the original s-plane we get triangles with the point z. p. eroup. . which S which a + m + v (1911). Case II. the through the centre of the figure is such a the system of side into an opposite side transforming for a doubly periodic This is then the period parallelogram . which in this case intersection of the three sides. hence A lies outside the circle of which BC is is Fig. is a subgroup of the original . to elliptic integrals in the four cases in functions triangle the of reduction The = 1 has been effected by A. parallelogram. 20. Polakov. on the vertical line can draw a real tangent AT to this circle. as a cluster point. an arc. . point of as centre and AT as radius (T being the orthogonal to is constructed. de Moscou. be let a circle A we K By an inversion in any side translations which carry a triangles into itself. 27 ^ . From With A K tangency) each of the sides of the triangle.

The single triangle is the half of the shaded region formed by a vertical line through the point C. The same figure can be used to illustrate the case of the values \. It follows The new triangles and for all The linear trans- K from Theorem 15. arising will also have their sides orthogonal to K. There are infinitely many possible values for A. succeeding inversions K will be a fixed circle. The group is no longer the modular group. case we Stereograpbic Projection. \. The two outstanding properties of inversion in a sphere are (1) that spheres (including planes) are transformed into spheres. This called the stereographic projection of the sphere the plane or of the plane on the sphere. 30. The modular group is an example of the case of the values £. / / L O . Now let an inversion be made in a sphere with its centre the point P. Obviously the correspondence of the points of S and T is one-to-one. \. 0. In this case the shaded region is a single triangle and the fundamental region consists of the shaded region and an adjacent region. \. that the triangles will be -infinite in number and entirely fill the circle K. 20 shows the case in which the values are §. From these we can state the following facts concerning corre- .— Before discussing the third shall introduce a widely used method of representing r /^"^ the z-plane on the surface of a sphere. . 8 is composed of two triangles. formations arising from an even number of these inversions have as a principal circle.56 of the triangle. and (2) that the angles between lines are preserved in magnitude. K Z inversion is formed into a sphere S passing through P (fig. \ \ J / / \ \^ J ~FlG 21 ''' /at T / / and let P be a point not lying in the plane. IV K is transformed into itself. 21). Section 18. The same inversion transforms on the sphere S into T. Fig. being composed of only a part of the transformations of that group that is. The group is Fuchsian. AUTOMORPHIC FUNCTIONS [CH. Let T be the n. 0. The fundamental region shaded in fig. T is trans0-plane. §. /j. it is a subgroup of the modular group.. v such that their sum is less than unity.

since every also fixed in rigid linear transformation has a fixed point. This sphere is orthogonal to the plane. we can state at once that every one-to-one and directly conformed transformation of the sphere into itself corresponds to a linear transformation in the plane. The succession of two such reflections (or in general of an even number) is equivalent to a rotation of the sphere. having the values (i. but an even number of the inversions preserves the signs of the angles and corresponds to a linear transformation of the plane. v Case III. sphere in planes perpendicular to OA are the fixed circles. The circles is elliptic in type. the transform of R by means of the inversion of the stereographic projection. OA is a fixed axis.-\-v>1. we note that the transformation corresponding to a rotation of the sphere motion on the Projecting these on the . . reflection in a diametral plane does not change the magnitudes of figures on the sphere. Hence inversions of 8 hi orthogonal spheres correspond to ordinary inversions in the plane T. The centre and the motion is rotation. are four possible cases. is any integer. where n A is a. by rigid motion. there is one fixed point on the sphere. * This is the For. Such an inversion changes the sign of the angles on S. the circle corresponding to C. 1/n.) any —There \. \. ft. Thus the rotation. A particularly useful inversion in an orthogonal sphere is a reflection in a diametral plane. \ + ij. the line of intersection of the two diametral planes being the axis of rotation. and conversely. and transforms great circles into great circles. X. given most general transformation of the sphere into itself rigid motion. of the sphere about an axis corre- sponds to a transformation of the plane whose fixed points are the stereographic projections of the extremities of the axis. By virtue of Thereom 7. R' is orthogonal to S and intersects it in C. Further. hence the line plane.29-31] THE RIEMANN-SCHWARZ TRIANGLE FUNCTIONS 57 sponding figures on S and T: (1) angles are preserved in magnitude (2) a circle in T is transformed into a circle on S and conversely (3) points of T inverse with respect to some sphere R will correspond to points on S inverse with respect to the transformed sphere R'. Section 3. . 31. The effect on the plane is the same if the inversion be made in a sphere R with the same centre and radius as C.* It is easy to find the transformation of the sphere into itself corresponding to an inversion of the plane in a circle C. . By means of (3) above we see that the corresponding transformation of the sphere is an inversion in R'.

The group of linear transformations arising from combinations of even numbers of the inversions in the plane is a group of rotations of the sphere. The sides of the new triangles arising by these reflections are also arcs of great circles and the continued inversions in the sides of succeeding triangles in the plane are all reflections in the diametral planes of the corresponding sides of triangles on the sphere. The area of the triangle A'B'C is unchanged by reflection in a diametral plane. and no common is intersected triangle at at by each side of the two points which are diameter. which an arc . \. It will now be shown that ABC can be projected stereographically on a sphere in such a manner that the three sides become arcs of great circles. Projecting these triangles on the plane.) [CH. triangles. projected ends of a plane be stereographically on a with A as centre and AT opposite let Now the sphere S are as radius. we can state that in the case \ + + v>1 the triangles are finite in number and fill the whole plane. A. with A as centre and AT as radius. £.) AUTOMORPHIC FUNCTIONS i.58 (ii. \. and (iv. Through A draw the chord TT' of the circle of the third side which is bisected at A (TT' is perpendicular to the radius through A). \ .T) 2 (X+fx + v— 1).) fig. IV \. /ul We The area can determine the number of the triangles in any case. That is. The circle K. In these cases the vertex A BC of the triangle (see is 22) lies within the circle of there is no real tangent to the circle from orthogonal circle. -J. \. Hence by repeated reflections the whole surface of the sphere will be covered by a finite number of section . (iii. of the spherical triangle A'B'C is Tr(A. . the sides of the corresponding triangle A'B'C on the sphere are arcs of great circles. From the remarks of the last paragraph of the preceding we see that inversions in the sides of the triangle ABC correspond to reflections in the diametral planes of the sides of the triangle A'B'C. sides of the triangle The three become three which passes through opposite ends of a diacircular arcs each of meter. The points on the circle K unchanged in position. \.

In each case half will be shaded triangles is (i. 32] THE RIEMANN-SCHWARZ TRIANGLE FUNCTIONS of the whole spherical surface of is 59 . These are shown in fig. tt. 31. In the four possible cases mentioned above we have N = 4n. 3-71-2 about an axis connecting two Now Klein. is N/2 from which we conclude that the group of linear transformations has N/2 distinct transformations. N = 24. 9 (1875).). 4tt(AT) 2 Hence (ii. Each of these larger composed of six of . L. in fig.f let us consider the rotations of the sphere which correspond to the transformations in the plane by which a shaded triangle.. (iii. transformation. The transformations by which one of the small shaded triangles at the centre of fig.. The triangles are not equilateral. i>.) N = 120. 7r/2 an equilateral triangle an octant on the sphere. . The angles and each of these triangles becomes tt/2. 23. We ^.) and the area the number . 23 is carried into the other three shaded triangles at the centre correspond on the sphere to rotations through angles tt/2. Ann. the smaller ones. say. Now let the bounding arcs of these triangles be replaced by the chords joining their extremities.— Case in the plane can be projected III. — — octahedron inscribed in the sphere. Math. £. t I am — . * Schwarz. 32. 24. including the identical region. This octahedron is shown 24. N = 4/(X + M + „-l). The number of copies of the fundamental which consists of some two adjacent triangles. in which the triangles stereographically into ordinary spherical triangles all equal in size. is shown fig.. cit. If these be projected stereographically a sphere S of which K is a great circle.* (iii. R. and we have the edges of a regular of each are 7r/2.) and half unshaded. is carried into another shaded triangle. 183. loc. F. we get a system of forty- eight spherical triangles. v have the values on The in division of the plane into the forty-eight triangles 23. mode] of indebted to Professor Crum Brown for the construction of the which this figure is a photograph. treat here only case in which X. /jl. £. is connected with the groups of rotations about axes and reflections in planes of symmetry by which a regular solid is transformed into itself.) N = 48 and (iv. Now consider the division of the plane into eight triangles indicated triangles by heavy is lines in fig. The Regular Solids.

IV opposite vertices of the octahedron. The inversions by which a shaded triangle is carried into an adjacent unshaded triangle. carry each of the eight large triangle .60 AUTOMORPHIC FUNCTIONS [CH. 23.three shaded triangles corresponds to a rotation of the sphere by which the octahedron is transformed into itself. rotations through angles 2tt/3. there is a group of twenty-four rotations by which the octahedron is transformed into itself. and in fact into any shaded triangle. 4tt/3 about rotations the octahedron we transformed into itself. including the case of no rotation (the identical transformation). Similarly can show that the linear transformation by which a shaded is Fig. is transformed into any of the other twenty. But we can say more. The transformations by which one of these triangles is carried into the other two shaded triangles lying in the same large triangle correspond to an axis through the In each of these centres of opposite faces of the octahedron. Hence.

we now group them triangles. instead of grouping the forty-eight triangles into eight large equiangular triangles each containing six. By projection on the sphere the central figure becomes a spherical quadrilateral . Suppose that. On the sphere these inversions are reflections in diametral planes by which the octahedron is transformed into itself.32] THE RIEMAXN-SCHWARZ TRIANGLE FUNCTIONS 61 triangles either into itself or into some other large triangle. two reflections is not a reflection but a rotation. each of which passes through either four vertices or the mid-points of tour faces. 24. The four quadrilaterals arising by in its sides together with that 23 as inversions composed of the eight triangles meeting at infinity constitute the other five. Since a shaded triangle may be carried into any one of the unshaded regions. But the twentyfour reflections together with the twenty -four rotations mentioned above constitute a growp of transformations by which the octahedron is transformed into itself. They do not form a group. into six quadrilaterals each containing eight fig. since the succession of Fig. The planes in question are the planes of symmetry of the solid. there are twenty-four of these reflections. Take the eight triangles at the centre of one quadrilateral.

IV. The radii to the vertices of the cube pass through the mid-points of the faces of the octahedron and the radii to the . where N is the total number of triangles.* vertices of the octahedron pass cube. or 7i-gon. * In Klein's Lectures Special attention oil the Ikosahedron will be found a treatment of all the solids. Re- placing the arcs by their chords.). AUTOMORPHIC FUNCTIONS [CH. we have the tetrahedron and in (iv. . since they are reflections of the quadrilateral in diametral planes. ^. The remaining cases of III. furnish the groups of rotations and reflections for the other regular solids. the group of sixty rotations and sixty reflections being used in investigating the general equation of the fifth degree. 1/n. In each case the solid is transformed into itself by a group of N/2 rotations and N/2 reflections in planes of symmetry. The two faces coincide.) i.) J. A. 32 The same first is true of the others. we have the ikosahedron and dodekahedron. Each of the faces is a regular polygon of n sides. In (i. is paid to case (iv. we have the dihedron. we get the edges of an inscribed cube. ' through the mid-points of the have the same planes of The two faces of the symmetry. ^. -3. solids . a regular " solid " of two faces. and the figure has zero volume. •$•.62 of four equal sides. In (ii. Because of this fact (or we can show it directly as before) the twenty-four reflections and twenty-four rotations of the preceding group transform the cube into itself.) 4.

that V U. in this — to be described section called the absolute.CHAPTER V NOX -EUCLIDEAN GEOMETRY 33. Then the distance PQ is (fig. Dist. Quantics. 1859. P. These definitions are made to depend upon a chosen fixed conic Cayleyan. Trans. and let the line PQ meet the Fig. 25. T be the the lines * Sixth Memoir on ii. tangents from O to the . Phil. Geometry.. M be two lines intersecting at between angle The absolute.PQ = Hog|^|I. Absolute. 63 . and let S. Let P. of the cross ratio logarithm the to proportional be to defined is. of the V . absolute in the points U. or The system of geometry was introduced by Cayley. Q be two points. 25). 149. vol. reprinted in Cayley's Scientific Papers.* " It is a geometry in which the terms " distance " and " angle are shorn of their usual meanings and are given new definitions. Let L. points four Q.

Thus in the given figure where we consider points interior to the ellipse the two lines through U. . In parabolic geometry there is one parallel to a given line through a given point. absolute a conic whose equation has real coefficients. that all geometry hold with the exception of the parallel postulate. taii 7ilog M=k LO sin - LOS - 'sin sin MOT . angle LOM + angle MON = angle LON. AUTOMORPHIC FUNCTIONS [CH. In elliptic geometry there are none. T that is. The line. If that is. the angle coincidence and the angle LOM approaches between two lines meeting at infinity is zero. (2) The geometry is called hyperbolic. since the two intersections with the absolute coincide. (3) double fundamental facts of the two lines can be drawn through a given point parallel to a given line. 34. The Three Types of Geometry. . We note also from the definition that if P approaches U the distance PQ becomes infinite. — parabola. PQ + Dist. Dist. since a given real line has no real intersection with the absolute. * The sum of the angles about a point being called four right ancles. geometry is called parabolic. It can also be shown without difficulty that in hyperbolic geometry the sum of the angles of a triangle is less than two right angles * easily discover If the We many of the is various types. . This type is the limiting case separating the two preceding. S and T approach becomes infinite. and that if L approaches S the angle LOM approaches the absolute. QR = Dist. as follows: (1) The absolute has a real locus.— 64 L.„ —^r-=-rLOTsmMOS . We on the verify directly from the definitions that line if R is a point PQ. PR. ordinary Euclidean the axioms of can show. The absolute has no real locus. and that if N We is a line through 0. The absolute is a degenerate conic consisting of a real The plane is not divided into two parts. V M is ratio of the four lines L. — We choose for the There are then three types of geometry. said to be parallel. The geometry is called elliptic. S. in fact. This is the case of the ordinary real hyperbola. and divides the plane into two parts.V and the given point are parallel to PQ. L and M are then . Angle ° . geometry hyperbolic. defined to be proportional to the logarithm of the cross M. or ellipse.

It follows from this that each straight line is closed and of finite length. 1 y v z x ) and Q(a. Fuchsian concerning facts certain arrive at an interpretation of non-Euclidean of ideas the of and other groups by means plan it will now be shown how geometry.. which need be constant only for a fixed value of e.} 33-35] (the NON-EUCLIDEAN GEOMETRY is 65 absolute) . The plane of elliptic geometry is characterised by the fact that there are no infinite points.— Elliptic The Circles in the Plane. The homogeneous equation of the absolute can by means of a projection be taken in the form i£ .y2 . 2 s/ {[e(x1 xi + y. When T Lim . is made a function of e.z x z.z^\e{x^ + y 22 ) . ) ) \ > im 1 \ " - ^/g 2 1 ) «(Oi a: 2 + yi2/2)--i~2 _ s/jfaz.M^r + Vi) ~ *FM** + 2/2 .1 *Jtfe(x. to be positive. elliptic.{e(x? + y?) . 2 . lie on the geometry the sum is equal to two right angles in elliptic geometry the sum is greater than two right angles. ir e{x x x. the absolute Euclidean geometry is Straight Lines by 35 Representation of Non-Euclidean It is our aim to Case. tangential the from made be can angle similar treatment of that the case e = equation of the absolute.z*\[c(xf + y?) -z?\\ l0g 1+ <XiX* + VOti-*to t -2V« . i~J . line the being geometry. on forming the cross ratio of the definition. .{x? + y. 2 y v z. the formula just found approaches zero unless k.^-J2 . at infinity counted twice. the result being which is yields the ordinary definition of angle. In pursuance of this .. for the infinite points lie on the absolute which is imaginary. Taking k = — 1/2 Je we get for the distance when e approaches zero shall We now show that when e = e definition of distance.*2 ] V ) we have the usual Euclidean approaches zero. negative. or zero.z&f . -x9z + {y& .zfo . sum zero the vertices of the triangle in parabolic and the geometry as e is is hyperbolic. -1.x i + yiy i )-zl z^-[e(x l + y 1i )-zfl[<^ + yi*)-zf) _. .zf] 2 2 2 e^Xt + 2/jJAj) . 2 When e = the absolute thus a special case of absolute is s = 0.\e. or parabolic according The Cayleyan distance between two points P(a. f..2 ) we find.V{[«(^i»2 + ViVt) .Vri? } ^ Z^Z-2 A the ordinary Euclidean formula for distance.2 + \J{ [ejx^ + y x y 2 ) .2 °" + y^j 2 ) .

Y. x y 1 = X Y Z.} The straight lines of the plane project into great circles on the sphere. . 1).— 66 AUTOMOEPHIC FUNCTIONS [CH. due to . + s/{ (X^. If P(x. the centre of the sphere being the centre of projection.and Y-axes of space. We shall work out the formulae for distance only. the tangential equation of the absolute being used.r 1. The pro- we have just made the inverse of this transformation. the corresponding points on the sphere beingP'^. Y 2 + Z. the a-and j/-axes of the X plane being parallel to the X. since P. P'. Z) the corresponding point on the sphere.Z. merely stating the results in the case of angle. y) is a point in the plane and P'(X. using the formula of the preceding section : : : : . The absolute distance PQ is.Z 28 ) \ X.(XI + Y.* + Z*)(X<? + Y. Let P(. . Ann. [The projection from its centre of a sphere on a plane jection is called a is gnomonic projection. V the straight lines of non-Euclidean geometry can be represented by circles in the plane. X gX t 1 Z x Ztf . and the origin lie in a line.' 4. X. Let the plane of the absolute be placed tangent to the sphere 2 + Y 2 + Z 2 = l. replacing the co- ordinates of points in the plane by their values in terms of the coordinates of the corresponding points on the sphere. be the equation of the absolute. Math. The familiar example of a geometry in which the is sum of the angles of a triangle greater than two right angles on the surface of a sphere. 0. since the plane through the straight line of projection cuts the sphere in a great circle. 4 of elliptic geometry with ordinary spherical geometry is (1872) and 6 (1873). at the point (0. we have. Y. Now let the plane be projected on the sphere. we shall have a consistent system of elliptic geometry on the sphere in which straight lines are represented by great circles* Let us see what these formulae become. and the distance between the points is the length of the arc of the great circle joining them. The treatment of angle is very similar to that of distance. Z can then be used as the homogeneous coordinates of the point P. Y v Zx ) and Q'(X 2 Y 2 Z2 ).. Angle here has its ing. + X 2 + Y I Y 2 + Z 1 Z2 -V{(X l X 2 + Y1 Y2 + Z I Z2 )2-(X 1 ^+Y^ + V-)(X^ + Y 2 + Z^)}>i Y^ * The connection Klein. geometry ordinary meanis We shall see how this geometry arises from Let x z + y 2 +l = elliptic absolute geometry. + Y. y t ) and Q(x 2 y 2 ) be two points in the plane. If and the centre now we trans- form the formulae for distance and angle.

is unchanged. ice have the Cayleyan distance equal to the spherical arc joining P'Q'. If we let angle have its ordinary meaning and define the length of a curve to be that function of the curve which is equal to the length of its stereographic projection on the sphere— we do not need to derive the function. the great circle common to the sphere and plane. 0) and the x. 67 Since (X 1( Y Z ) and (X.35 36 ] ' NON-EUCLIDEAN GEOMETRY . circles X 2 -|-Y 2 -)-AX-|-BY = l. 0. The k=-ii.and y-axes being parallel respectively to the X. Then cos = X X X +Y _ formula then becomes t 2 + Y^+ZZ. project the sphere stereographicaliy on a plane.and Z-axes. If now we — It is easy to see what these circles are. Y Z 3 ) are the non-homogeneous x s coordinates of a point on the sphere. — the plane Y = meets the sphere.— If it 2 the absolute 2 2 is real. X 2 + Y 2 = l. We we can by shall now 2 as follows. the great circles become circles in the plane. The projections of a point in the plane are the two points in which a line through the given point perpendicular to (0. The straight lines are then represented by the two-parameter family of 36. The Hyperbolic Case. and the same is true of the projection of the great circle. Each great circle on the sphere intersects this circle at opposite ends of a diameter. 0. Place the plane of the absolute in coincidence with the plane = 0. the origin in the plane being at the point (0. 0. 0) infinity. The straight lines in the plane project into circles on the sphere which are orthogonal to the plane Y = 0. We project the plane on the sphere X +Y +Z = 1 Y take as centre of projection the point in homogeneous coordinates the point where the Y-axis meets the plane at 1. A similar treatment of angle will show that the Cayleyan angle between two lines is equal to the angle (in the ordinary sense) betiveen the corresponding great circles Putting on the sphere.. a projection take to be the circle x 2 + y = l. we have a system of elliptic geometry in which straight lines are represented by the circles just found. If the projection is made from the point (0. of' a ereat circle joining P' and Q'. 1) on the plane Z = 0. 2 we have X 2 +Z 2 = 2 2 2 Let 6 be the length of the arc 2 + Y 2 + Z 2 = l.. .

ive have a representation of hyperbolic Y= geometry in the plane in which straight lines are represented by circles orthogonal to a fixed line. £:»j:l=X:Y:l-Z.1 t x* + y*-\ F Ax J = o. V system of circles be projected stereographically from the point (0.VI [«Ax + yAyf . if the definitions of distance and angle be carried over. Hence. It will be r.[i? + y 2 . rj) of the plane. y). Ay 1 2 "I + x&y/(l . y. Let V(x. P'(£ >)). Q(x + Ax. 2 we 2 derive the relations 1. Y. 0. Z) on the sphere and the point (0. y = Z.1 ][( Axf + Ayf] }" 2 ( Dropping infinitesimals of higher order than the be written xA 4z y - first. and the system of circles is projected into the circles orthogonal to this line. The circle in which Now let this the plane cuts the sphere (i. 1) and (x+Ax. most convenient to get the formula for distance in the form of an integral. 1). 2 0.—— — 68 AUTOMORPHIC FUNCTIONS [CH.e. y2 e . coordinate axes of the line through a point (£. The equations connecting the points on the Cayleyan plane with those on the sphere are evidently x = X. we have. y + Ay) be neighbouring points in the ay-plane. which was previous projection) is projected into the line the unchanged by Y = 0. where get X +Y +Z = From these we the equations of the transformation £=x/(l-y). Let the coordinates in the plane on which the sphere is stereographically By considering the projections on the projected be £ r\. (x. = J(l-x*-y*)l(l-y). the absolute.i•-1 pn-H ^ ~ g s2 + f- 1 + sAg + yAy + ^{[xAx + yAyf-^ + y' '. z2) now being tv + U1St - 1 x2 + y. y + Ay. . its corresponding point (X.l][(Az) 8 + (At/) 2] } + xAx + yAy . J V . 1) on the plane Z = 0. this can I 1 ) " V \\z 2 + y 2 -lJ /J ptAx + yAy Lx2 + z/ 2 . Substituting j. Q'(£+A£ + Aij) being the corresponding points in the ^ij-plane. respectively in 1) for (x v y v z x ) and (x 2 the formula of Section 34. We shall now determine the formula for distance.y) ~| 2 -2/J L yAl-x*-**) J .

The displacements i. . pp. Since distances are to be unchanged. A distance and angle save possibly for sign. We shall call a transformation of the plane by which both distances and angles are unchanged a displacement. The length of a curve in the . of the absolute imposes five conditions on the collineation leaves invariant the cross ratio of four points on a line or of four lines through a point and we observe from the definitions of distance and of angle that the collineations which transform the absolute into itself preserve . 37.)«}/.)« + (^A?) 2 = 2AV{(A^)2 + (A. an point must remain at infinity. A consideration of angle shows that. as angle has its ordinary meaning. 8.. taken along the curve. and we shall have for the transformed points Dist. Acta Mathematica. in the elliptic case. that is. R lie on a line. transformed into itself. Dist. are the directly * Poincar£. 7. The displacement is then a infinitely distant is the absolute . PQ + Dist. It follows also that lines are transformed into lines for if P. vol. us consider the representation of the hyperbolic plane in which the straight lines are represented by circles orthogonal to the real axis. and this is the value to be uged for the infinitesimal distance P'Q'. We found that the absolute is let Now transformed into that axis.. | j use the customary complex variable z = £+i„ this becomes.— 36 37 ] > NON-EUCLIDEAN GEOMETRY 69. Q'R' = Dist. putting 2fc = 1 A z /. P'Q' + Dist. If now we fr-plane is the value of the integral ||dz|/..— Let us return to the Cayleyan plane. PR. =a x + b x y + gi x a 3 x+b s y + c s . Q'. Q.* and the shortest distance between two points is A and B the integral j' dz | | /. Application to Groups of Linear Transformations. R' are col linear. ^ ' = a 2 x + b$ + e„ a sx + b 3y + c 3 ' The invariance coefficients. QR = Dist. From the form in which we have written we verify at once from the equations of the Dist. the path of integration being the circle through the points orthogonal to the £-axis. } the last equation transformation that PQ = 2* > /{(Af/. . which is true only if P'. collineation z . P'R'.

the angles hyperbolic less . and any one can be brought into coincidence with any other without alteration of shape or magnitude Certain facts of previous proofs become obvious for example. V conformal transformations which leave the real axis invariant and transform circles orthogonal to the real axis into circles orthogonal to that axis that is. projected the plane on the sphere in such a manner that great circles play the role of straight lines. rotating the plane through 180° about one of In this hyperbolic geometry. In each case. The various transforms of the fundamental region which partition the plane are equal in all respects. They preserve lengths but change the signs of angles.70 AUTOMORPHIC FUNCTIONS [CH. When X+/x + = l the sum of the angles of the triangle is two right angles and the i/ and reflections mentioned. The transformations of the group are all displacements. the absolute is projected into the circle at infinity. the fundamental region of a Fuchsian group is a rectilinear polygon. the linear transformations. The displacements are the rigid motions of the sphere. the In the we transformations which preserve lengths but change the signs [It is easy to show that in the transformation we made. and that the regions will cluster .] The three kinds of triangles that arose geometry is Euclidean is (fig. and the geometry is (fig. which is unchanged by the of angles are the reflections in diametral planes. In each case a shaded triangle can be brought into coincidence with any other shaded triangle by a displacement or into coincidence with any unshaded triangle by a displacement together with a turning over. in infinite number along the elliptic case axis. The transformation amounts. When \ + /U + 1/> i the sum of the angles is greater than two right angles and the geometry is elliptic (fig. to use the Euclidean analogy. 23). 19). The inversely conformal transformations leaving the absolute . When X+ M + „<1 the sum of than two right angles. all the triangles of a network are equal to each other in magnitude when the appropriate type of geometry is used. rotations in connection with the triangle functions can be interpreted in the light of the three kinds of geometry. invariant are the inversions in circles orthogonal to the real axis. to its lines. 20). that only a finite number of regions will be contained within an area not extending to the real axis.

. 26. is shown The Fig. Math. It is possible also to discover independently discontinuous groups of collineations which leave a certain conic invariant. 0. 38 (1890). Fricke.37] NON-EUCLIDEAN GEOMETRY 71 The Fuchsian groups furnish us with discontinuous groups of collineations which leave a real conic invariant. 50. the division for the modular group carried over to the Cayleyan plane by means of the equations of Section 36. each transformation of the group can be expressed as a transformation of the points in the Cayleyan plane by means of the equations defining the transformation from one plane to the other. The circular-arc fundamental region in the complex plane is transformed into a rectilinear polygon in the Cayleyan plane. Studies of the subject from this point of view have been made by Poincare and by Fricke. Journal de Math. 717 3. tt/3. which is a fundamental region for the group of collineations. 26. 3 (1887). Ann. -461. and from these to derive new discontinuous groups of linear transformations. in fact. For. The angles of each triangle are figure is. .* * Poincare. 405. The division of the plane for such a group of collineations in fig..

Riemann's Surfaces. 72 ." the function is single-valued. —The functions of greatest simplicity . and w acquires the value «. Let x = p e tS ° be a point in the z-plane. describes any closed path not encircliDg the origin.CHAPTER VI UNIFORMISATION 38. . the function u? = x. being the axis. In the case of functions which are many- theorems of valued. If r. there are complications owing to the fact that it is necessary to deal with a particular one of the many branches of the function. that is. say along the negative half of the real axis. and w 2 if x is in the second plane. and returns to x with the original value wx If. The two values of w are w 1 = \Jp eie °l'2 and w 2 = . Each value of x. circuit not if it crosses . or the negative part of the real axis into the second plane it recrosses into the original plane. either x remains always in the original plane. 1 2 another circuit of the origin brings the value back to w v Now let us superpose on the x-plane a second a-plane infinitesimally near it. If x makes a surrounding the origin. the "Riemann's surface. . Now let us consider the variation of w. the angle is increased or diminished by 2ir. w returns to the value w x If. in analysis are the single-valued functions to such the simple the theory of functions can be applied without reservation.ri Let us fix our attention on to 1 while x is varied. x makes a circuit of the origin. We shall recall here a device introduced by Riemann in which the plane of the independent variable is multiplicity of planes in such a is replaced by a manner that on the resulting configuration the function Consider co. Join the part above the axis in each sheet to the part below the axis in the . each point of which represents the same value of a: as the point immediately below it. excepting the points and two values of w. The points and °° at which the line of juncture of the sheets terminates are called branch points. however. however. other sheet. both the distance from the origin and the angle return to their original values.v: 1 = \Zp e iB «l 2+. In the resulting surface a moving point x passes from one sheet into the other on crossing the negative half of the real axis. On this configuration. yields first single-valued. it returns to x„ in the second plane with the value m>2 In general whatever path x makes returning to x the function has the value w if x x is in the original plane. Let the two planes be cut along curves extending from the origin to infinity. x passes once around the origin. p distance from the origin to x and B the angle between 0x and the real . starting with the value io x at x„ in the original plane.

its and w = R/fW. that single-valued function of a. For a given value of x the w This simple example exhibits the essential features Riemann's surface for the general algebraic _ . .b^x + c s _ a2w + b2x + c2 a 3 iv + b 3 x + c s + there are certain numbers connected with the curve which are unchanged. on a Riemann's surface composed of sheets joined along certain lines terminating in and the solution of the equation gives m wisa m branch points. The fact is. I. either for . and no two points of the given curve are transformed into the same point. coefficients P. the number of sheets equalling the degree of the equation in x. w=f{x) X It or for the inverse function is x=f~\w). its class. It is known that the most general such transformation is formation from. a 3 m -f. are fixed values of w. function. x) = real locus.f 39. For example. x = ~R '(w'.3 8 39] » UNIFORMISATION of 73 the Let be defined as a function of x by the irreducible* equation <f>(wx) = ? (x)w>» + V x)iv^ + +?m (x) = 0. means of plane curves although in are complex and the curve has no : . and the number of the original curve. Genus. not breaking up into factors of lower degree. helpful to interpret our statements by general the coefficients in the equation *(jf.. If we make a projection — w _a > 1 'f + 6 x+g 1 1 . x'). but the nont There exists a Riemann's surface for any function w=f(x) algebraic cases are characterised by the fact that the number of sheets.x). x'). The study of the algebraic equation into which a given algebraic equation §(w. Owing to tion * is form this transforma- called birational. . x) = can be transformed has led to the important conception of genus.e. In other words. the points on the Riemann's surfaces of the two curves are to correspond in a one-to-one manner. is infinite. wliere P^') is l( a polynomial in x..the points on one curve to the points on the other are expressible rationally 2 : one in which the transx) iv' = ~R 1 (w. the degree its of the transformed curve. double points are the same as in Consider now the most general kind of transformation by which each point of the given curve is transformed into one and only one point of the transformed curve. x'=H2 (w. as the preceding example would lead us to expect. Similarly the same equation defines i as a function of w single-valued on a Riemann's surface covering the w-plane.

A surface is said to be connected if we can join any point of the surface to any other of its points by a curve lying entirely within the surface. an important invariant number.-'* = x'(x' . it [CH. formation transformation w = w. The French term is genre. be written rationally thus: w = w\ An exceptional case. Then two points on the given curve yield the same point on the transformed curve.occurs for this transformation if the given curve is symmetrical with respect to the tu-axis. however. x) = 0. n be the degrees of the original and the transformed curves. The transformation is not birational— we are unable by using the equation tion is w = x-. If the inverse called a rational without using the equation of the curve the transforma. Cremona transformation for example. by considering the properties of the Riemann's surface quite apart from the equation connected with it. The which an irreducible curve of the nth degree can have is (n—l)(n — 2)/2 hence the genus of a curve is the number of double points by which it fails to attain It is usually represented maximum number of double points . 1 {w. The general inverse which transforms it into «. a 40. But it must be possible to do so by The using the fact that w. nor the class. There is.(!P. commonly called deficiency. Let n. resulting rational equations are not the general inverse transformation. of the curve the inverse transformation can x = io' a l(x' as The transformation is thus birational.2)/2 . and the trans- = R. x' = w=w 1). — different the German. and one more useful for our purposes. It is necessary first to state some general results concerning what is termed the connectivity of a surface. + Multiple points of higher order than simple double points are to be counted as the equivalent * In curve theory it is number of double points. .2)/2 .d' = (n . x = l/x'.i~ . Connectivity of Surfaces. is called the genus* of the algebraic relation. Funktionen. x' = l/x. For example. Crelle's J.1 ) 2 x — -Jx' is not rational but if we make use of the equation . be possible to solve the equations w' = ~R. x are connected by the algebraic equation *(w. . Consider the curve w" = x?-x. nor the is number of double unchanged. VI x). This invariant number. Genus was introduced in Riemann's Theorie der Abel'schen.. w'=x/w. but an An inverse transformation which holds only for the points on the two curves.1 ){n' . x'. 54 (1851). d' the numbers of their double points. w=x 3 to get is an inverse function involving only rational functions. this maximum.t 74 It is x' AUTOMORPHIC FUNCTIONS not necessary that x) for w. (n — l)(n — 2)/2 — d. x rationally in term of if'. Gcschlecht. The term deficiency is due to Cayley. which has The Cremona transformation is thus a special case of the birational transformation. and d.d. the surface of a sphere. by the letter p. the inverse w = 1/w'x'. In all cases (»' . . We points readily find that in the case of the birational transforma- tions neither the degree.. We can take an entirely view of the genus of an algebraic equation. example will make this clearer.l)(n .

We can then represent its connectivity cuts rendering the It by 2p + l. the surface is divided into two pieces. two sides of the In general. The oftenquoted example of such a surface is that formed by a strip of paper whose ends are joined after one of them has been given a twist through 180°. The first and second of the surfaces just mentioned are simply connected. its connectivity is unchanged. . 27. or is of con- nectivity n + 1. where p class of is an It integer. such as a closed surface with a small hole in it. We shall assume that the surface has a boundary. or be transformed in a one-to-one and continuous manner into any other surface. can be proved without difficulty that a surface with a single boundary. a surface can be transformed into any other of the same connectivity if they have the same number of boundaries. The number p is called the can be shown that a surface can be transformed in a continuous and one-to-one manner into any other surface of the same class. since a cut extending from one boundary circle to the other does not cut the . however this cut be made. if it is a closed surface. the surface is said to be simply connected otherwise it is multiply connected. the surface .f the surface. is of odd connectivity. plane surface bounded by a a plane circular ring we shall make boundary* such as the a small hole in the surface to provide an let us cut the surface along a curve beginning and ending in the boundary. for the cuts which render it simply connected will be transformed into new surface simply connected.39 40 J ' UNIFORMISATION circle. the surface is (7i + l)-ply connected. It is clear that if a surface be continuously deformed in any way. * We shall not consider in this section the so-called one-sided surfaces. said to be doubly connected otherwise we can make a second cut beginning and ending in the boundary (which now first includes the cut). Now surface into two pieces. If. if n cuts are necessary to make the surface simply connected. t In fact. 75 sphere initial bounded by two concentric circles are connected. surface is renders the If one cut simply con- nected. the third is multiply connected. Fig.

point of the fundamental pegion. Now let us consider the Riemann's surface in the light of the preceding remarks. the congruent sides being connected by arrows. each point of the fundamental region corresponds to one. Genus of the Fundamental Region of a Group. Functions. VI The simplest example closed surface with of a surface of class it. 28 we give an example of this process. . of the same work. of the algebraic relation connected with it* In view of this fact it becomes evident intuitively why the genus is unchanged by the birational. . we found that two functions. this surface is the same as that define the genus of the fundamental region to be the genus of this closed surface formed by of the Riemann's its We bringing In fig. and only one. each point being brought into coincidence with its congruent point. Since w and x are automorphic.76 AUTOMOEPHIC FUNCTIONS [CH. Section 21. surface. or one-to-one. Theory of An excellent treatment of the connectivity of surfaces will be found in Chapter XIV. 355. x{z) satisfying the equation and in general no two points of the region yield the same pair of values. * For the proof of this fact in an important special case see Forsyth. say — automorphic with respect to a group are connected = 0. To avoid considering the points at infinity. In (I) is shown a fundamental Bringing the sides B and H. 27 shown such and the six a surface of class 3. In other words. The initial boundary is at cuts which render the surface simply connected are shown. transformations. x) . congruent values of z correspond to the same point on the surface. Each point z of the fundamental region yields a pair of values w(z). p. The surface a closed surface consisting of several spherical sheets and it can be shown that the class of this surface is equal to the genus let it . region. In Theorem F. surface. by an algebraic relation §(tu. is A be projected stereographically on the sphere. Let us now form the fundamental region into a closed surface by bringing congruent edges together. We thus get a closed surface which corresponds point for point with the Riemann's The genus (class) of surface. is that of a p holes through In fig. point of the Riemann's surface and conversely each point of the surface is derived from one. p is . Thus two congruent edges of the fundamental region correspond to the same curve on the w and x. congruent edges together. 41. and only one.

or at least partially dissected. ei S ht h • . of many. 12. and since neighbouring points on one surface correspond to neighbouring points on the other. it follows that this new surface can be used as one on which to represent the pairs of values (w.valued functions. it would serve as a Riemann's surface for the algebraic equation quite as well as the original Since T (i) Q 1 (in) (17) Fro. valued functions is both simpler and more developed than thar. The theory of single42. surface composed of several plane sheets. 77 and that it is D and F into coincidence. !nd and these are k brought into coincidence on the ' lZ* ° f genUS 2 We *««»» form a cycle the points of this closed surface formed from the fundamental region correspond in a one-to-one manner with the points of the Riemann's surface for <£(iv. where some further cuts are necessary to make the region simply connected. [The fundamental region is not always simply connected. In other words. The Problem of Uniformisation. Viewed in this light.] This use of the fundamental region as a dissected Riemann's surface is particularly useful because of the fact that both w and x are single-valued analytic functions of the variable z of the fundamental region. we get necessary to bnng the opposite ends of S - A and E are now closed curves G and C together. x) = 0. we see that the fundamental region before being joined into a closed surface is the Riemann's surface dissected into a simply connected region. as in (III) so n0te that the surface. See fig.40-42] and UNIFORMISATION (II). x) satisfying the equation. and if we can express each of the variables of the many-valued functional relation as singlevalued functions of a third variable a substantial contribution — . 28.

The curves of genus 1 can be uniformised by means of doubly periodic functions. it is the problem of expressing the curve in parametric form by means of single-valued functions.p'(z)]. however. gram the 43. the two-sheeted Riemann's surface. where p(z) is the Weierwell that those of genus . x and w as single. x). that there are no exceptions. w(z). It has been established. x(z). x = 4>(z).valued function. It is the fact. We can employ functions of the form x — (j>[p(z). x) = x = x(z).p'(z)].— 78 to — AUTOMORPHIC FUNCTIONS — [CH. VI is the study of the function is made. This the problem of uniformisation. <w = yfr(s). resulting surface is of class 1. that J. We observe that by joining congruent edges of the period parallelostrassian function of their arguments. and each pair (w. there exists an algebraic relation and this is obviously uniformised by the functions <J?(w. From it is the problem of finding two singlevalued functions which map a portion of the 2-plane in a one-to-one and generally conformal manner upon the Riemann's another point of view surface of the functions =f(x). 7. known and where <p and \{f are rational functions The functions map each of the period parallelograms on the Riemann's surface. two such functions. brought out in preceding sections lead us to hope that the general algebraic curve can be uniformised by means of For we found that between simple automorphic functions. w = w(z). w2 = l — x? is uniformised by the functions x = 2s/(l + z 2 ) w — 2 l These functions map the whole z-plane upon (1 — z' )/(l + z ). The facts . satisfying the equation is given by one point z at least. [See fig. w = \fs[p(z). by means of a general existence theorem.] Uniformisation by Means of Automorphic Functions.valued functions of a third variable. Looked at from the standpoint of curve theory. Theorem Any algebraic curve can be uniformised by . We wish to express Let w =f(x) be a many. But there was no indication as to whether all algebraic relations could arise in this way. w It is can be uniformised by means For example. such that for each value of z for which both functions are defined the corresponding values of x and w satisfy the equation w =f(x). We consider here particularly the algebraic curves. the two-valued function of rational functions.

The functions have no other singularities than poles within the principal circle. in establishing ^ functfo^^thaf Theorem K. i... Ann. and the fundamental region for this group is the whole plane.. 21 (1883). group to which fundamental region of the the the functions belong lying within principal L-et with a principal circle. chapter xiv. III. and in the Mathematische Annalen. it follows that the fundamental region of the group contains portions of the principal circle on its interior. it is clear that the uniformising functions cannot be the simple automorphic these functions are conSince the uniformising functions have only polar singularities within the region. Let us examine the statements in this theorem. 2nd ed. So far as I am aware.42 43 1 > UNIFORMISATION the >jq means of Fuchsian functions. The proofs of these theorems. Math. Further papers by Koebe have since appeared in vols. If the function to be uniformised is transcendental. general existence theorem. in fact. iv=f{x) be a multiple.valued function of any kind It has Can be Unif0rmised b means of automorphic ^ functions studied in Chapter nected by an algebraic relation. for the theorem to admit the single transformation z' = z as a possible group for this case. It is necessary.. It was Klein who first saw the possibility of arriving at a the latter publication. involve considerable detail in the execution .—The most general many -valued function can be uniformised by means of automorphic functions of a group Also a much more general theorem has been established. and Abschnitt iii. and the circle is a natural boundary for one at least of the functions. 1907-9. The treatment is based on o series of papers by Koebe in the Gottinger Nachrichten. p. 162. where one at least of the functions has sino-ularities other than poles. while simple enough ception.. or at least has vertices on the principal circle. 67 (1909) and 69 (1910). We shall presently give an example of the uniformisation of algebraic functions by means of functions whose group possesses a fundamental region not confined to the interior of * Vol. §10. Cf. . in con- and they cannot be given here. the only complete and simple presentation is to be found in Osgood's Funktionentheorie* The automorphic functions of the types just mentioned are not the only ones which can be used in the uniformisation of functions.

and R is a : To take a simple example from the calculus 2 the integral \dx> J{\ —x % ) —that is. the become or w x branch points another. not in the point of departure. where xv and x satisfy an algebraic equation $(iv. The advantages of uniformisation are many. x)dx. VI the principal circle. by uniformising the curve iv 2 = 1 — x 2 by means of auto- morphic functions. The integral is then a single. where §(w. where w =\—x . The uniformisation can also be accomplished by functions whose group preserves no circle invariant.valued on the surface when certain closed paths are described. In the study of functions on a Riemann's surface functions of the form f(w. 80 AUTOMORPHIC FUNCTIONS [CH.w = cos z in other words. The points at which many the considering involve one kind infinite of series expansion. but in a point congruent to it. But in the plane of the uniformising variable where the fundamental region is simply connected these integrals are single-valued.— . for the fundainfinite points mental region has no and no branch points. form JR(w. both differing from that at ordinary points. But functions of the kind mentioned in the theorem would seem to be usually the most simple. x). x) = 0. dx/w. There is one advantage in the fact that a fundamental region within the principal circle is always simply connected. the integral is changed by the addition of a constant. x) = the — — owing to the necessity of exceptional points. investigations are lacking in simplicity On variable the fundamental region in the plane of the uniformising these exceptional points do not occur.valued function which changes by an additive constant (which may be zero) in passing from any point . 2 — is most easily found by putting x = sin z. In case the genus of the surface is greater than the integrals are always infinitely many. The integrals become singlevalued (provided they have no logarithmic singularities) when the surface is dissected into a simply connected piece. A path on the Riemann's surface by which the integral returns to its original position with an added modulus corresponds to a path in the plane of the uniformising variable which terminates. Not least among isation is to be by uniformfound in the treatment of Abelian integrals the simplifications introduced integrals of the rational function.

. . Since the real the circle orthogonal Hence . 1-32..though rotates 1. It seems probable that automorphic functions are destined to play an important rdle in the theory of algebraic functions and their to a congruent point. S^ axis g. S 3 S 4 S 5 S 6 be real self -inverse transformations satis. made in succession. pp.(281*- fi . we get C 3 orthogonal to the axis.t The transformations e JJ. 44. Led points'lying in the upper b 5 are transformations S v forms of the point D 6 when the line short a . 44] UNIFORMISATION 81 There is a marked gain in simplicity.. 192 (1899). —We turn now to the question of actually effecting the uniformisation of algebraic curves by means of functions connected with specific groups.43. C D C 2 bemg C4 C5 .. 6 D ( se el6*29£ . and it is not type the fixed are elliptic formations. integrals.(M2.. . ™ . * Phil. angle an Dl through into xtself D through DlD is transformed m . opposite sid e of formed into a point C. Trans. transformations). . points being conjugate trans the Consider half-plane. . [. . Whittaker's Group..* The groups are constructed from transformations S< which have the property that S 2 = 1 (called self-inverse. S. Groups of the type discussed in this section were introduced by Whittaker. Fio. on the C2 the circle point into a Similarly S 2 carries X Continuing.-ie76)/(lU-188)]. This equality gives much = fying the equation S 6S 5 S 4S 3S 2 S 1 L difficult to determine such transfreedom of choice. [. De D„ Let imaginanes.. 29.. Let S v S. 4786)/(17 S -281)]. .

fl . and so on. 6 congruent sides together to form a closed surface we see that the genus of the fundamental region is 0. . Since x has the same values along the side CjDj as along DgDj. .82 AUTOMORPHIC FUNCTIONS . Section 21.. . [CH. It is known that when the genus of the fundamental region is there exist functions having only one pole in the region. VI S 5 S 4 S 3 S 2 S 1 = S 6 _1 = S 6 a transformation which leaves the point D 6 unchanged.. sum -n- of the angles at the vertices D C . x takes on its value twice there. . and only once. where \f^(D 1 ) * 0. . and (SSS 4 S 3S 2 S L) 2 = S 62 = 1. . . x acquires any given value once. As z moves from 6 to x x moves along some curve from e6 topoints . . congruent bringing the 1. &* = l. with this group.. Hence . Hence C 5 coincides with D6 and we have the closed circular polygon shown in the figure. T) 2 C 1 and D 2 C 2 The generating transformations satisfy the equations etc. as z moves along DjC^ x traverses the curve just found from e x back to e6 C x alsobelongs to a cycle of angle 27I-/2. in the fundamental region. and that in terms of any one of these functions all the automorphic functions of the group are expressible rationally. . C4 is (Section 17). Let e v . [We can place the pole arbitrarily by using a function of the form (ax + b)/(cx + d). . . D . e 5 be the values of x at the points D v D 5 and e6 its value at the congruent equation see that the . From the last . . Let x be one of these functions. x — e 1 = (z — Djz-yf^z).. In other words. .2 and back again. . D 1 D 6 and D^. Dc Cp We shall start with the point 6 and. . Thus we find that the function x(z) -maps the fundamental region on the .J According toTheorem E. . and we shall suppose for convenience that its pole does not lie on the boundary. which also has but one pole. we . C 4 tracing the boundary of the fundamental region in a counterclockwise direction. This polygon is in fact the fundamental region for the group There whose generating transformations are S v S 2 S 3 S 4 S 5 are five pairs of congruent sides. . . Let us examine this mapping. and the angle -k at D x is mapped on an angle 2tt in the cc-plane. . according to1 of Section 21.. S 52 = l. find the corresponding locus in the x-plane. By . Now let us consider the automorphic functions connected. D D ev Now D Theorem A constitutes a cycle of angle 2tt/2 and. the function x(z) maps the fundamental region on the whole £-plane. and the angle at C x is mapped on an angle twice as great at e The side C1 D 1 C2 is mapped on a curve from e 6 to e.

the opposite sides being congruent by the transformation S r The side Be C v and the new region .J{(x — e 1 )(x — e 2 )(x — e 3 )(x — ei )(x — e 5 )(x — e 6 )}. boundary of the region being mapped on five curves . v 6 ±iut this group. The functions w and x are connected by an algebraic relation $(w. .e3 )(x - e 4 )(x - e ){x b - e a ). . single cycle of angle 2tt. and is hence an automorphic function of z with respect to the second group. 30). say S Erase the . Consider a function w(z) automorphic with respect to the new group. Consider in particular the function . Adjoin to the region the region congruent toit by one of the transformations. Hence the automorphic functions of this group enable us to tions Hence . the UNIFORMISATION 83 extending from the points e e 5 to the point e (see fig. will uniformise only the curves of genus 0. which is of the second degree in w. S S S S 4 lf 5 r The eight vertices of the polygon form a is for example. x) = 0. x) = 0. since all its transforma- belong to the former group. w= uniformise the curve w i = (x. since to each value of x correspond two points z of the region and hence two values of w.44] K-plaTie. We get a group of greater genus in the following manner.. xiz) is an automorphic function of the new group it acquires a given value twice in the double polygon. Joining congruent edges of the new fundamental polygon to form a closed region amounts to joining the two x-planes along the remaining branch lines. 29). It maps the polygon on two planes joined together along the branch line e x e§. common r a polygon of eight sides. This group is a subgroup of the preceding. The two-sheeted surface just found is the Riemann's surface for Q(w. is side C 4 D 5 D6 The transformation S 5 followed by S transforms C 4 D 5 Dg into x C'D/C^ and transforms the new polygon into one adjacent to it along the side C'D 'C A similar treatment of the other 5 r sides shows that the double polygon the fundamental region for the group whose generating transformations are S^. S^. is transformed into C-fl^'C (see fig. This function is single-valued on the surface just found. being of genus 0. We find that this region is of genus 2.e x ){x - — e 2 )(x .

real constants. there are six points e v e6 . . the number of essential constants is the . . So there are altogether six essential real constants at our dis- posal in forming the group. Also this group is not essentially different from one obtained by transforming it by any real transformation. . that the general automorphic functions of the group are rational functions of w and x. . . Hence. which shows that the groups have the generality requisite for the uniformisation. S 6 contains three essential imposes one condition on leaving twelve constants. that a= —d . The the relation rest.. which is of genus p. of these is defined by two real coordinates. S6 S 5 St = 1 defines three of these in terms of which is which shows that three more of the constants are non-essential. . does there exist a group of the type we have been discussing whose automorphic functions effect its uniformisation ? As an indication that the uniformisation can always be made by these groups. By considering 2p + 2 self -in verse transformations and forming the double polygon as before. . _1 4 . we should be able to uniformise the relation M.84 AUTOMORPHIC FUNCTIONS [CH. VI We can show by means of Theorem G. it will be shown that the constants at our disposal are equal in number to the conditions to be fulfilled. . (a. each [namely. 2j)+2 ). so that the groups in question can be applied to uniformise any algebraic relation of genus 2. So we have six essential constants in the a>figure. .2 = (aj-ea ) . same in the s-figure as in the z-figure. — e6 ). . . . The values e v e e depend upon the function x(z) and upon the group... presents itself is : If the function w = (x — e 2 1) . .. twelve real constants. Section 21. giving . (x — ee ) be arbitrarily chosen. (x - e. (Section 1)]. Now and each considering the as-plane. These groups can be generalised without difficulty. The question that . . But if we make a linear transformation we can transform any three points into the points which shows that six of the constants can be disev e2 e 3 regarded as non-essential. It is known that any algebraic relation of genus 2 can be transformed birationally into a relation of the form w2 = (x — e 1) . of the «-plane. and are not all at our disposal. Each of the transformations Sv The condition S = S ..

returning to the initial mapped on the x-plane thus is region given the circuit. as z traces the Flo 32 circle x moves along the value on the completion of real axis. researches in the general theory of conformal representation! Weber shows that there is a function x(z) with the following properties (1) x is analytic within the given region except at infinity. 32. 31) . the region to be considered lies outside the circles. ~ P P Cl £2 Si ^ 2 gj real along a bounding circle. pp. 31. . For simplicity we shall take the case of three circles (C v C 2 C 3> . and is continuous on the boundary circles (2) x is real on the circumferences of the and only once region.] bounding the near points property (3) for arising by us add to the given region the region ^— . t Monatsberichten der . 1886. UNIFORMISATION 85 Weber's Group. Fig. of —In the group employed by Weber* for purposes uniformisation an initial region bounded by non-intersecting circles orthogonal to the real axis is considered. . Let us see what the boundary Since a. save for points on the bounded by It follows that the function x(z) maps the region the circles on the whole x-plane. By a well-known theorem on Ein Btitrag zur Poirwart's Theorie der Fuchs'chen Functioned Gbttinger Nachrichten.) [It is e e e e e e slits 5 6 3 4 bounded by three x 2 twice—once in each only slit each trace can easy to see that x without violating overlap cannot slits the direction— and that circles. fig. The (See fig. Berliner Ahademie. . 359-370. . 1870.— 44. From Schwarz's : where it has a simple pole . the in once value (3) x acquires each boundary. 45J 45. is becomes. Now let an inversion in the * circle C3 .

vol. are generating transformations of a group with the real axis whose fundamental region is the combined region is an automorphic function with respect to this group. . The group is formed from three circles with their centres on the real axis. exactly as in the preceding section. C2 ' along which the function x has its values unchanged. this imposes three conditions on the group. S r say. which acquires each value twice in the region once in each of the combined portions. and we is suffi- can show by the count of constants that this group ciently general to uniformise w = (x — e x) . And x — sponding to together. If we bring congruent edges is of the region it . Similarly inversions in to a linear transformation as fixed circle C 3 and C 2 are equivalent S 2 carrying C 2 into C 2 S x and S 2 ' '. carrying C x into C/. Now.e4 )(x - e )(x 6 - e 6) w is an automorphic function of functions of the group enable us curve of genus 2. an inversion in C 3 followed 31). 2nd ed. Since the quantities are real. We see.. preceding case.. save for certain inequalities necessary to ensure that the circles be exterior to one . e6 are real . that to* if = (x . we get a closed surface with two holes through and the genus of the region thus 2. 666 et seq. Each circle has two degrees of freedom its centre and its radius may be chosen arbitrarily. * Osgood. the imaginary part of x changes sign again. Hence the automorphic this algebraic to uniformise . (x — e6 ). Lehrbuch der Funktionentheorie. observe that the quantities ev . any algebraic relation of the form where ev e & are real. have the same real part but an imaginary part of opposite sign at points inverse with respect to the circle. If we invert in one of these by an inversion in Cx ' is a linear transformation. . . As in the . Closing the region corresponds in the as-plane to joining the two sheets along the remaining branch lines. pp. . ..e x ){x - e )(x 2 - e z ){x . The circles C 1. C2 invert into C/. circles. (fig. C3 . VI harmonic continuation * a function which is real along a circle will. We 1 z. and the result of the two inversions is that the value of x is unchanged. if continued analytically across. i.. The function x(z) maps the region on two planes joined along the branch line e s e t correjust found.86 AUTOMORPHIC FUNCTIONS [CH. . only three of these quantities are essential.

welche bei einer bestimmten linearen Transformation ihres Argumentes unverandert bleibt..45 46 J ' UNIFORMISATION 87 There are thus six real constants at our disposal.] If we remove the condition of orthogonality. for example. Schottky makes a study of the automorphic functions and the Abelian integrals connected with the groups.— In an Journal* Schottky deals with groups derived of Weber's. Weber.. . and obtains . [Certain other. 227-272. in Grelle's The Work of Schottky and Burnside. . It is * Vol.. and the automorphic functions of this group effect the uni2 formisation of the relation = (x-e 1 ) . . pp. and derives their properties in an exceedingly simple manner. including those of = 1. (1892). . another . : . w . There are thus oo 3 groups of this type which can be used for the uniformisation of the function w 2 = (x — e1 ) (x — ee ). . that x has the values ev e 6 at the points where the circles intersect the axis.. vol. Burnside sets up the various types of integrals in this way. (1837). t On a Class of Automorphic Functions. . which shows that we have three more constants at our disposal in groups of the Schottky type than in those of Weber (for the case of three circles). Similar groups are studied by Burnside. manner . oi. . . but article in the where the circles are not restricted to those orthogonal to a circle. the Poincare theta series (Section 22) converge for m then possible to get the Abelian integrals by integrating these series. But three of these can be regarded as non-essential since the group is not essentially different when subjected to a linear transformation. xxiii. which shows that the groups have the requisite generality to effect the uniformisation We have made no use of the fact that the circles are orthogonal to the real axis except to conclude that the group is Fuchsian. pp. 49-88. We are left with three essential real constants for the fulfilling of the three conditions imposed. 46. where e v e6 are real.f who makes the discovery that in a large class of groups. where ev e 6 are real. we get a Kleinian group of the Schottky type (to be mentioned in the next section). . results that we have not mentioned do follow from this fact. Proceedings of the London Mathematical Society. (x-e 6 ). Ueber eine specielle Function. infinite products for their expression.. Each circle has an added degree of freedom since its centre is not restricted to lie on the axis.

Crelle's J. Fuohs.. e. . Gbtt. 95 (1882). H. O. A. H. 150. (1882). Poincare. F.. H.. Acta Math. 1485. H. Ann. 193. Ann. 293. . . Ann. Acta Math. p. Rausenberger. — Many particular classes of functions. L. 20 (1882).. 163. however. the circular functions. H.. Crelle's J. Sur les fonctions uniformes qui se reproduisent par des substitutions Math. and the" present list will therefore begin with the papers in which Poincare created the general theory. Ann. welche durch Umkehrung der Integrate von Lbsungen der linearen Differentialgleichungen mit rationalen Coeffizienten entstehen. Klein. 81. A. 1274. 49. Schottky.. It seems undesirable to attempt here a bibliography of these earlier researches. welche durch lineare Substitutionen unverandert bleiben. Poincaee. 1484 lineaires. Uber eindeutige Functionen mit linearen Transformation in sich. 301... L. Math. Gruppen ganzzahliger Thiorie des grmipes fuchsiens. 1 (1882). be made of one or two memoirs of the earlier period in which the general theory was functions. 19 (1882). 89 (1880). 565 20 (1882). 1 (1883). C.. 626 .. Poincare. 297. 300. 1166 . (1883). Nach. welche die absoluten Invarianten linearer Transformationen bilden. Math. . p. 581.. Mention should. 20 (1882). Mimoire sur Us fonctions fuchsiennes. p. Sur une classe de groupes discontinus de substitutions lineaires et sur les fonctions de deux variables indipendantes restant invariables par ces substitutions.. 97 . 52. Ueber eindeutige Functionen mit mehreren nicht vertauschbaren Perioden. 553 20 (1882). Uber die conforme Abbildung mehrfach zusammenhdngender ebener Fldchen. E. Sur 1198. Poincare. 93 (1881). 83 (1877). Schwarz. 75 (1872). Schottky. Math. 20 (1882) 187 21 (1883). 1038. 393. p. . .R. 19 (1882). 92 (1881). elliptic and triangle-functions (including the elliptic-modular and polyhedral functions). had been studied extensively for many years before the general theory of automorphic functions was created. Poincare. 292.. les fonctions fuchsiennes. notably foreshadowed ..R. C. Ueber eindeutige Functionen mit linearen Transformationen in Math... H. Uber eine Klasse von Funktionen mehrerer Variabeln. —-? Picard. Crelle's J. Acta Math. Uebereine Eeihe neuer Functionen.g.... . 859. Fuchs... F. especially Uber diejenigen Falle.. 59. 125.. in welchen die Oaussische hyper geometrische Reihe cine algebraische Funktion ihres vierten Elernentes darstellt. F. sich. gewisser Huewitz. Ann. 1 (^882). Sur les fonctions fuchsiennes.— A BIBLIOGRAPHY OF AUTOMORPHIC FUNCTIONS Note.. 333. 94 (1882). Ueber Functionen. 1.

Ofv... inditerminies conjuguies binaires d quadratiques les formes Sur Picard. unendliche durch reproduciren. 1. Gbtt. (1885).. 227. l'Ec. I. Svenska Vet. Bull. J.. Sur les fonctions fuchsiennes et les formes quadratiques ternaires indifinies. de la Soc. Les fonctions fuchsiennes (1887). de Norm. Leipz.. Demonstration d'un thiorime general sur les fonctions uniformes Hies par une relation algebrique. fonctions fuchsiennes. Gbtt.. Acta Math. Crelle's J. 405. Weber. Poincare. . 3 (1883).. 41. Ber. 11 (1887). PoiNCARE.. Acta Math ..M. (1886)...- l'Ec. X. 49... . 46 (1889). 1137. welche bei eincr bestimmten linearen Transformation ihres Argumentes unverdndert bleibt. 33 Math.. F. Neue Beitrdge zur Riemann'schcn Functionentheorie..> Bemsfbr nagra Forh. H. Norm. 2. H. Sitzungsb. Ann 21 (1883). 9. Naeh. E. 3 (1887). Crelle 3 J. Poincare. H. F. Schlesingeb.. af K. Picard. Sur 201.. 92 VON Mangoldt. (1883). 61. Liouville'9 J. de (3). 5 (1884). Acta Math.. 239. E. Math. 359. Fuchs. teorem af Poincare. H. Biermann. d'Abel et quelques-unes de ses applications giomitriques. Liouville'9 J. 11 (1883) 112.. «-) Humbert. Application de la thiorie des fonctions fuchsiennes a Vitude des (4). 141. 5 (1884). (4). E. H. E.. der Fuchs'schen Functionen. 1. Varithmitique. 735. Acta Math. Memoire sur les fauctions zitafuchsiennes.. 7 (1885). Sur un thiorime de M. 0. 12 (1884). 219. Picard.. Sur une transformation des fmictions fuchsiennes et la reduction des intigrales aUliennes. 43. Lincei Rend. les les formes quadratiques binaires indifinies.. Sur le thiorime (4). Sur les formes quadratiques ternaires indifinies a indeterminees conjugieset sur les fonctions hyperfuchsiennes correspondantes.B. L. Dyck. 677. groupes des Equations liniaires. Schottkt. . H... et les 5 (1888). Bull. PlUARD. Acta Math.BIBLIOGRAPHY 89 Poincare. Substitutionen lineare (1888). 4 (1884) mime c6U 209.. Bianchi. L... PlOARD. 291.. Zur Theorie der Fuchs 'schen Functionen. Wien . Producte. 3 Stouff. Memoire sur (1884). Poincare. de France. Ueber ein Verfahren zur Darstellung elliptischer Modulfunctionen durch unendliche Producte nebst einer Ausdehnung dieses Verfahrens auf allgemeinere Functionen.. (4). Poincare. H.. et 1.-Ak. Ueber die durch Gruppen linearer Transformationen gegebenen regularen Gebietseinteilungen des Raumes. F. Math. Nach. 102(1886). 4 2 (1888). Klein. H. Ein Beitrag zu Poincare" s Theorie der Fuchs' schen Functionen. H. Ueber die Darstellung der eindeutigen Functionen. Am. Sulle superficie Fuchsiane. "W. 187.. die sich Ann.R. Poincare. . courbes algibriques. Sur un groupe des transformations des points de I'espace situes du d'un plan. Liouville's J. 161. Humbert. C. durch Stahl. 102 (1886). E.F... H. C. Ueber eine specielle Function. G. Acta Math. 327. Mimoiresur les groupes kleiniens. 121. Ann. 101 (1887). 105 deBrun. Zur Theorie 11 (1888).. (1886). H. S. Sur la transformation des fonctions fuchsiennes. Surun thiorime de la thtlorie ginirale des /mictions. Math. Poincare. G. (3). (1889) 181. Ann.

" Bihang till Sv. 9. (2).. Burnside. Ueber die Differentialgleichungen dritter Ordnung. 15(1891).... Stouff.. On a class of automorphic functions. 268. (1891). Ueber die bei den linearen Differentialgleichungen zweiter Ordnung aufbietenden Primformen. L. Sur la theorie des fonctions fuchsiennes. Math. Ann. 1 (1890). Sui gruppi Kleineani a due variabili.. Klein. . .. deren Substitutionscoefficienten ganze Zahlen eines biquadratischen Kbrpers sind. Math. ville's J. X. Entwickelungen zur Transformation fiinfter und siebenter Ordnung einiger specieller automorpher Functionen. 4 (1890). Nach. (3). 1893). Math.M. 130.. 185. 85. 283. Burnside. Nach. 143. I.. 41 (1892). A. 705. Acta Math. E. Napoli Rend. P. Biermann.. Fricke.. 16 (1891). 38 (1890).. Zur gruppentheoretischen Grundlegung der automorphen Functionen.. Painleve.. Math. Ueber ein allgemeines arithmetisch-gruppentheoretisches Princip in der Theorie der automorphen Functionen. 345. Ann. P.90 AUTOMOKPHIC FUNCTIONS Hurwitz. P. (1892). Sur un problime de representation conforme.. 564. Ueber die Darstellung der Fuchs'schen Functionen erster Familie durck unendliche Producte. del Pezzo. Fricke. Burckbardt. Gott. Ueber cine besondere Klasse diskontinuierlicher Gruppen reeller linearer Substitutionen. 38(1890). Ueber Normirung der linearen Differentialgleichungen zweiter Ordnung. Toulouse Ann. 1. 267. Math. 9 (1893). Quadratwurzeln Fricke. welchen die Formen mil linearen Transformaiionen in sich genugcn. 22 (1893). G.. E. . Cassel. R. 114 (1892). P. 110 (1892). (1892). Fricke. Ann. Mimoire sur les Equations diffirentialles du premier ordre. ' . 121. Note on linear substitutions.... Nach.. 33. bindmes. O. Substitutionen einer Variabeln. 3 (1892). Nach.. Gott. 273. Stackel. Rittek. 112 (1893) 287. 17 (1893). Acta Math. 2.. H. R. specielle Thetareihen. E. 1. 190. R. 62. Ann. Klein. Fricke. Zur Theorie der Lame'schen Functionen. p. Crelle's J. 345. . 144. Math. Crelle's J. (1890). 36 (1890). Ueber discontinuirliche Gruppen. Nach. W. Weitere Untersuchungen iiber automorphe Gruppen solcher linearen . u-elche lineare Substitutionen in sich zulassen. Annales . Proc. R. Schlesinger. 453... 7 (1893). Mess. Nach. G. R F. Ueber indefinite quadratische Formen mitdreiundvier Verdnderlichen. Weber med titel : Beitrag zu Poincard's Theorie der Fuchs'schen Functionen. F. Monatshefte f. 33 (1889). Ann. No. 130. 201. Afd.. Schlesinger. Gott.. Zur Theorie der Abel'schen Functional.. (4). Math. 23(1892). Gott. Math. Math. L. Ueber algebraische Gleichungen zuAsehen eindeutigen Functionen. F. 42 (1893).R. \V. 39 (1891). 49. 461. C. Klein... Picard. Die eindeutigen automorphen Formen vorn Geschlecht Null. R. ' Mn de 1'Ec. Norm. Ritter. 103. 1100. Klein. Ueber die Darstellung einiger Fdlle der automorphen Primformen durch 123. (1892).. Fricke. 40 (1892). (1893).... Fricke. Ann. VIII. Ann. Ann. M. 49.. L. Ofver en afhandling afH.S.. Ak. Die automorphen Formen von beliebigem Geschlechie.. 1409. Cassel. De V equation Au = ke u sur une surface de Riemann fermie. 50. Ueber den Begriff des functionentheoretischen Fundamentalbereichs. F. Sur certains groupes fuchsiens forme's avec Us racines d'iquations . (2). R. Liou. Gott. Gott.. deren Substitutionskoeffizienten ganzer Zahlen enthalten.. Ann 42 (1893). 281.

Proc... Toulouse . Gbtt.. 288. Ueber Fuchs'sche Functionen zweier Variabeln. Sulle funzioni Thetafuchsiane. On a certain automorphic function.. Fricke. Ann . 2 Bde. 157.. Ueber Eiemann'sche Formenschaaren auf einem beliebigen algebra ischen Gebilde... *. Fricke. Ann. G. 1900).. Cambr. Stouff. 44 (1894). Math. 557 Fricke. Ann. 46 (1895). E.. 31. Schoenflies. L. 97. Math. Ann. 606.. R.. Die Stctigkeit der automorphen Functionen bei stetiger Abanderung des Fundamentalbereichs. Forsyth. 261. Gbtt. L. Congres (1893). 565. Ann. Math. Cambridge (1893).. E. Ueber die Theorie der automorphen Modulgruppen. Sine Anwendung der Idealtheorie auf die Xubstitidionsgruppen der automorphen Functionen. 473. Ritter. Ann. 8 (1895) 322. de Cracovie (1895). 106. F. Kapteyn. 1894: Math. Ritter. Gottingen. Ueber die Discontinuitdtsbereiche der Gruppen reeller Hnearen Szib stitutionen einer complexen Variabeln. Note on the equation y* = x{x i -l). Baker. Math. d'Ovidio. A... 91.) Ueber unverzweigte lineare Differentialgleichungen der zweiten Ord nung auf ebenen Curven vierten Grades. 29 (1894). 46 (1895).. Cassel. Kempinski. Sul teorema dell' esistenza delle funzioni Fuchsiane. 17. Ann. {1896). Fricke. Math. Le equazioni differenziali lineari a coefficienti algebrici integrabili . 44(1894). 573 : S. 149... X. 43 (1893). R. F. 105. 47 (1895). Ueber lineare Differentialgleichungen der zweiten Ordnung.. Notiz iiber die Discontinuitdt gewisscr Collineationsgruppen. (1894). Sopra alcune classi di gruppi di sostituzioni lineari a coefficienti •complessi. v h 4de Nederl Nat. Gokdan. Fricke. Math. Leipzig (1896). Die multiplicativen Formen auf algebraischen Gebilde beliebigen Geschlechtes. Nach.. Ueber die Transformalionstheorie der automorphen Functional. 1'. Ann. Kriliska studier ofver teorin for de automorfa funktionerna. Vorlesung. {second edition. 489. Die Theorie der automorphen Functionen und die Arithmetik.. Nach. Fricke. 72... A.. viii.BIBLIOGRAPHY 91 Burnside. Ann. Math. 5(1896). Diss Upsala (1894). Gott.. 42 (1893) 377. Bianchi.. Nach. R. Schlesinger. Klein. R. 45 (1S94). A. Rivista di Mat.. 47 (1896). 46(1895).. Math. R. Die Kreisbogenvierseite und das Princip der Symmetric. 101. E.. 10 G "' UebeT daS FormensVste m eines Kreisbogenpolygons vom Geschlecht Null. . 200. F. Sou. Schoenflies. Viterbi. 42 (1893). (1895). R. Auto graph. Math. 360. (1896). 44(1894). Bagnera. Handel.. A.. 77. intern. <1896). . Fricke. Uur diffe'rents points de la thiorie des fonctions fuchsiennes. Handbuch der Theorie der Hnearen Differentialgleichungen. I. Theory of Functions of a complex variable. Ritter. ^' Math.... mit Anwendung auf die Theorie der automorphen Formen. Math. Ueber Kreisbogendreiecke und Kreisbogenvierecke 44 Math Ann (1894). Torino Atti. R.S. Ann. Ann. W. H. Papers published by the Amer. Een paar slellingen van Poincari. E. . Proc. Ueber Kreisbogenpolygone. W. L. 47 Bull. (1894) D.. R. 24 (1893). Ann. Math... 741. Ann.M. A.

Math. 1 (1901). On the connexion of algebraic functions with automorphic . L. Wien Wjrtinger. I?. Zur Theorie der automorphen Functionen von n Veranderlichen.M. Frioke. R. 10. 7 (1898).. Dixon. 1'Ec. I. Fricke. du premier . J. 78. Deutsch. R. Gott. 314. Trans. 253 . .. A. E. (3). 1. Sitzungsber. 14o. R.. . 1239. Cambridge (1897). Ver.S. Poincaue. 31(1902).. Vorlesungen iiber die Theorie der automorphen Functionen. Bagneka. 380. and Klkin.M. 19 (1902). 261. Teubner.. C. Nach. Ueber die in der Theorie der automorphen Funktionen auftretendenPolygoncontinua. Wellstein. 33 d. Kluyver. (1900). Zur Theorie der Poincarff schen Eeihen. 126 (1898). H. Poincare. 150 36 (1898).. Fricke. 1. Sur un vrobleme de Riemann. R. F.. Phil. Fricke. . 32 Dixon. degre. H. Theorie der automorphen Functionen. 35 (1897). E. O.. 11 (1897). R. u. Gott. T. Frioke. Die Discontinuitatsbereiche der Gncppen reeller linearer Substitutionen einer complexen Veranderlichen. Acta Giornale di Batt. Schlesinger. Frioke. H. Jahresb.. L. Un Kleinieno. 723. . Nach. A. D. Mathematische Probleme. Notes on theory of automorphic functions. .. Deutschen M. d. d. (1900). Leipzig. On a class of automorphic functions. 137 . Hilbert. A. L. 353. F. the theory of automorphic functions. Lindemaxn. 9 (1900).R. 1897-1912.. Ver. Palermo Rend..M. de .. 627. 85. T. Frioke. . Baker. 193. Les fonctions fuchsiennes et V Equation Ai( = e u . . Nach. Proc. Liouville's 4 (1898). surface... (1903). R. C. Whittaker. Festschr. 340. Proc. Ver. Prime functions on a Riemann (1900). Veber den arithmetischen Charakter gewisser Netze von unendlich vielen congruenten Vierecken. 29 (1899). Veber Primformen auf Riemann' schen Flachen. 151. 94. Ueber die Beziehungen zwischen der Zahlentheorie tind der Theorie der automorphen Functionen. C. (1900). (1900). 331.. Naeh. (3).. 265. Gott. (1897). 192 A (1898)... functions. R Die Bitter'sehe Frimform auf einer beliebigen Riemann 'schen Flache. L.. R. Math. M.. Jahresb. W. C. Deutsche Math.S. 423 : 30 (1900).. 297. Proc. Notes on (1900). A special case of Dirichlet's problem for two dimensions. d. teorema relativo agli invarianti delle sostituzioni di Lincei (5 a ). Braunsohw. 493. Sur une classe de fonctions hyperfuchsiennes et sur certaines sub- stitutions lineaires qui s'y rapportent. Frioke.. 3 Alezais. U.S.. 21(1897).. 92 AUTOMORPHIC FUNCTIONS algebricamevte. R. J. F. .S. studiate in base alia teoria delle "funzione fuchsiane" del Poincari. M. Ber. Whittaker. Hutchinson. Dedekind Festschrift (1901). the un gruppo 31 Dixon. Die automorphen Elementarformen. 108 (1899). Fricke. Math. Gott. 44. (5). Ann.. Trans. Amer. (1899). Note on a function analogous to Weierslrass' sigma-function.. 55. Miinch. 6 (1898). Ueber die Poincari' schen Reihen der {-l)-ten Dimension.. (1902)... 213. 4(1897). J. 303. Mess. C. Norm. 126 (1898). Sur I'inUgration algibrique des Equations differenticlles ordre et du premier J. Arch. Abel's theorem and the allied theory. Nach. Gott. P. (J. Zur . Rendic.

. 1. Inst. Atti d. J. Ueber die Uniformisierung beliebiger analytischer Kurven. 89. 129 (1905). 184. 329. .. O. Sulla teoria dei gruppi discontinui. 3 2. W. R. Beoden. 62 (1906). 65 (1905). R. 62 (1906).. Amer. 7 (1906). Heidelberg dritt Math. On automorphic groups whose coefficients are integers in a quadratic field.. Trans. I.. S.. Sur VinUgration algibrique des equatiuns des integrates abiliennes. Levi. T. 120. Beweis der Existenz linear-polymorpher Funktionen vom Grenzkreistypus auf Biemann' schtn Fldchen.. *5 (1904). E. Araer. Crelle's Journ. Sur Huewitz. (3). oo ) and the functions belonging to it.. Math. On the group of the sign (0. Mat.. 61 (1905). 139. Heeglotz. les invariants arithme'tiques. 497.S. M. Math. Gioenia (4). 6.. Blumenthal.. 0. 7 (1906). Ueber Modulfunktionen von mehreren Veranderlichen 93 liniaires et les periodes S6 M Ann (1903).. J. 530. Koebe P. Eend. On the automorphic functions of the group (0.. (5). 8 (1907). G. 191. 177. 13 (1904). Zur Theorie der mehrdeutigen automorphen Funktionen. On the aut'omorphic functions of the group (0. JTach. 12 (1906). tli Mat. 54. A.. 347. Math. Am. Una questione fondamentale per la teoria dei gruppi e delle funzioni automorfe. . (1904). Fubini. Trans. E. Ann. Ann. (3). Morris.. d. tfach. (1907)..S. 246. 13 (1904). . Sulla costruzione dei compi fundamentals di un gruppo discontinue. Ann. 40 (1904).. 15 (1906).. 447. Trans.. Johansson. 3 2. Annali di Mat. . Liouville's J.. Deutsch. Feicke. 6). Ver. 5 (1904). . 62 (1906). M. Math.. Nach. Ricerche sulla teoria delle funzioni automorfe. Math. 425. Ann... the fundamental region of a discontinuous group of linear transformations. Am. P. Atti R. Hutchinson. 449. G. Soc. Sulla teoria delle funzioni automorfe.. Ven. It. Bin Satz konforme Abbildung einfach zusammenhdng- ender Biemann'scher Fldchen auf den Einheltskreis. 9 (1903). H. 4. Lincei <5). Uniformisierung reeller algebraischer Kurven. A method for constructing {1907). Ueber Thctafunlet ionen und Modulfunktionen mehrerer Veranderlichen. ' Koebe. Lunds Poincake. 3 Z„ l2 l3 Tr. Zur Theorie der automorphen Funktionen von beliebig vielen Variabeln. (1904) .. iiber die Johansson. . Math. 590. 58 (1904). L. Beitrage zum Continuitdts beweise dcr Existenz linear -polymorpher Funktionen auf Biemannschen Fldchen. Kongress (1904). Gbtt. 'J. I. Sulla funzioni automorfe cl iperfuchsiane di piu variabili indipendenti. Soe. Fubini. 10 (1904). 633. )t Universitets Ars-skrift. G.. Ann.S. Young. Bemerkungen zur Theorie der automorphen Funktionen. Ann. 261. 17 (1903). 92. Blttmenthal.. 177. I. Ann. Lincei. Math. 81. Ueber die Gestaltder auf algebraischen Kurven nirgendssinguldren linearen Differential eichungen zweiter Ordnung... Sulla teoria delle forme qaadratiche Hermitiane e dei slate-mi di tali forme. 1297. G. Fkicke. H 159. 59 (1904). Fubini G. Fubixi... Math. Ueber die. G. Blttmenthal. H. 509 . M. Ace.BIBLIOGRAPHY Poincake.... Gbtt. Hutchinson. 0. Hutchinson. . Neue Entwicklungen iiber den Existenz-bewei3 der polymorphen Funktionen. Gbtt. Tr. S. --82. Fubini. Fubini. J.. (3). Ann. 325. Atti R.

G. Ueber die zu einer ebenen algebraischen Kurve gchbrigen transzendenten Formen und Koebe. 64(1907). Young. G. M. Nach.. Sur runiformisation (1907). 915. 19 (1908).E. 10(1906). Poincare. M. Stahl. 33. Am. H. Palermo Rend. .. Whittaker. les equations Aiffirenticlles est . C. Journ. 537. A fundamental invariant of the discontinuous (-groups defined by the normal curves of order n in a space of n dimensions. Koebe. Klein. . Ann. Introduzione alia teoria dei gruppi discontinui e delle funzioni automorfe.. H. 169. (3). Nach.. Zur Uniformisierung der algebraischen Kurven. W. M. Tubingen (1908). G. P. Ann. 336. Rothe. H. P. Sur . E. AUTOMORPHIC FUNCTIONS FuBiNr. 23 (1907). (1907).. Pisa (1908). Math. W. 14 (1907). e delle Fubini. Gbtt. Circolo d. Hilb. d. J. 67(1909). R. 1. relation to the general theory of 40 (1909). M. Nuove ricerche interna ad alcune classi di gruppi discontinui.. u. et les fonetions fuchsiennes. Richmond. E. ii. 94 .. J. M.. Phys... Ueber die Uniformisierung der algebraischen Kurven. 119. Differentialgleichungen. Math. P. Math. 66 (1908). Klbin. POINCAB. Diss. Berlin Sitzungsb. 230. Grossen.. 410. 16 (1909). 258.R. F.. 363. F.. 18 (1907). intern. Sur les solutions uniforms de certaines equations fonctionnelles. Young. (1907). Sulla teoria delle funzioni automorfe loro transformazioni. 145. (2). H.. (1908). Young. (1906). R. Act.. On automorphic functions in algebraic curves.. Bull. Math. On the discontinuous (-groups defined by rational normal curves in a space of n dimensions. On a class of discontinuous (-groups defined by normal curves of . 147 (1908).. the fourth order. u.. Ueber die Darstellbarkeit der homomorphen Formenscharen durch Poincare'sche Zetareihen. Gbtt. (1908). Caspar. G. 14 (1908). 919 and Ueber die Zusammenhang zwischen dem sogenannten Oszillations- theorem der linear Differentialgleichungen und dem Fundamentaltheorem der automorphea Funktionen. matem. vol. F. Fubini.. P. Sur la reduction des integrates abiliennes Rend. P.S.. 143 (3).. Berechtigung einer Arbeit von fferrn E. W. Ver.. die (1908).. Pick. Hilb. Arch. 231 Garnier. Nach. des fonetions analytiques. Monatshefte f. ginirale du troisieme ordre dont Vintigrale uniforme. (1909). Bemerkungen zur Theorie der lineaien Differentialgleichungen zweiter Ordnung. auf gegebeue Gebiete beschrdnkt sind. Quart. . Koebe. P. Ueber Kleinsche Theoreme in der Theorie der linearen Differential68 (1909). u. Gbtt. Die Uniformisierung der algebraischen Kurven. (1908). . 175. d. 546. C. Mat. Deutsch. 27 (1908). Monat. 21 (1906).S. Ueber das Grundtheorem und die Obertheoreme der auiomorphen Funktionen im Falle der Hermite-Lamisclien Gleichung mit vier singularen Punkten. W. gleichungen. congresso. Math. Neue Entwicklungen iiber lineare Differentialgleichungen. 219... . 215 . H. (2). 363. J. Ann.. Mitteilung eines Grenzubergangs durch iterierendes Verfahren.. 24. 258... Fubini. Bull. T. Palermo Rend. Ann. Atti iv. M. 281. 16 (1907). (Rome). di Math. 177. Fatou. 97. 31 Schottky. Ueber Beziehungen zwischen verdnderlichen . 112. Palermo. Am. G. Sulla discontinuity propria dei gruppi discontinui.

E. Nach. L.. Sechs Vortrdge iiber ausgewahlte Gegenstdncle aus der reinen Mathematik und mathematischen Pltysik. 138 (1910). Johansson. Amer. Brottwer. 15. E. 499. A. Leipzig (2nd edition) (1912). 53 (1910). Nach. P. C. Au=e u und die automorphen Funktionen. Zur Transformationstheorie der automorphen Funktionen.. J. L.. LlLJESTROM. 240. Zur theorie der Konvergenz der Poincare''schen Reihen bei den Hauptkreisgruppen. Ueber die Hilbertsche Uniformisierungsmethode. Fys. BlANCHl. W. Nach. 68 .. Hbhere Modulfunktionen und . Brotjwek. Ueber die Uniformisierung der algebraischen 1. 59. J. 192 139 (1911). E. Hecke. fallet... Bieberbach. Leipzig (1911). 37.. Koebe. (1912). (1911). Ueber die Konstruktion der Klassenkorper reeller quadratischer K'drper mit Hilfe von Prym. Ueber Kleinsche Theoreme in der Theorie der linearen Differentialgleichungen. 181. Gott...R. Leipzig (1910). Gott. 24. Math der algebraischen Koebe. (1910). E. 603. Sechste Mitteilung. Ueber die . 337 (1909). . Forhand.. 180. Gott (1910). R. Math. (1910).. HlLBEitT. 324. (1910). Poixcaue. 11 (1910). 305. Gott. Gott. Trans.. D. 803. Courant. 599... E. Lehrbuch der Funktionentheorie. 355. P. R. Zur Theorie der automorphen Funktionen. polymorpher Funktionen auf Osgood. 1.. Om konvergensen af de Poincare" ska d-serierna i hufvudcirkel2d Congr. Johansson. Nach. Kurven Math \nn Gott 69 (1910). L. Zur Transformationstheorie der automorphen Funktionen. S. . Heche. 6 (1910). class of Diss Gottineen s Soc Craig. L. Ann. 71 (1911). P. Ueber die Anwendung des Dirichletschen Prinzipes aufdie Problcme der konformen Abbildung. 68 (1910). Theorie der Prymschen Funktionen 619. die Uniformisierung automorphc Funktionen mit imagindrer Substitutionsgruppe. G. . erster Ordnung.. Nach. On a Ueber hyperfuchsian functions. Gott. (1912). Ueber die Singularitdtenfreiheit der Modulmannigfaltigkeit.. P. Ann. Ofversigt af Finska Vetenskaps-Soc. Arkiv Matem. J. Math. 518. Gott.. Fricke. Uniformisierung beliebigtr analytischer Kurven.. E. Koebe. . 71 (1911).. Nach.. Atti Accad.. F. H. L. Nach.... 25. (1912). Sul gruppo automorfo delle forme ternarie quadratiche suscettibili di rappresentare lo zero. Scand. Nach. Fricke. Gott. Ueber die Uniformisierung beliebiger analytischer Kurven. Nach.. Math. (1911). S. Bieberbach. Chazt. Astr. Sur les fonctions fudisiennes a multiplicateurs constants. and Rost. F.. 18.. Grundzilge einer allgemeinen Theorie der linearen Integralgleichungen. Crelle's J. C. F. automorphen Funktionen. P. (1912). Ann. (1908). Gott. Ueber die topologischen Schwierigkeiten des Kontinuitats beweises der Existenztheoreme eindeutig umkehrbarer RUmannschen Fldchen. math. (1910).. Koebe. de leurs 152 (1911).BIBLIOGRAPHY 95 Hilb. 145. Lincei (5) 21 (1912). p. 251. Sur Vinditermination des fonctions uniformes au voisinage coupures. ihre Anwendung aufdie Zahlentheorie. Kurven durch Nach (1909) Koebe.

. Math. Sur des equations differentielles du troisieme ordre donl Vintegrale generale est uniforme. 39. Princeton (2) 14 . Astro. 21 Zu {1912). F. Koebe.. P. ses points critiques fixe <. Begrundung der Kontinuitatsmethode im. Monats. Koebe. Fricke. (1912).. Deut.PRINTED IN GREAT BRITAIN Bv NEILI. Wiss. 1. L. Fys.. A. R.. 437.. p. J.. und Hilb... M. Liljestrom. Osgood. de l'eoole supeY (3). u. W. P. Ver. Koebe... EDINBURGH. Leip. Automorphe Funktionen mit Einschluss der elliptischen Modulfunktionen. E. 7 Plemelj.. Bieberbach.. F. Math. R. Band ii. 2. den Verhandlungen betreffend automorphen Funktionen. (1913).96 AUTOMORPHIC FUNCTIONS Klein... Ueber die Uniformisierung der algebraischen A'urven. Gbtt. 72 (1912).. Existenzbeweis betreffend Funktionen.. . J. Zur Begrundung der Kontinuitatsmethode. Math. Arkiv Matem. Osgood. 59. welche zu einer eigentlich discontinuierlichen automorphen Gruppe gehoren. 153. Gebiete der konformen Abbildung und Uniformisierung.. 143. 56. Leipzig (1913). Koebe. 103. Die Grenzkreis uniformisierung analytischer Gebilde. Ann. Etudes sur la thiorie du potential logarithmigue. W.. Annal. 64 (1912). F. E. 23 (1912). P. P. L.. Ber. On the uniformisation of algebraic functions. Ann. 29 (1912). 297. Palermo Rend. LTD. 879... Encyclopadie der math. 35 (1913). AND CO. f.. Garniee. P. et sur une classe liquations nouvelles d'ordre supirieur dont VinMgrale generale u. (1912). Nach. Brouwer.




^ifleilOmVersity libraries MAR 1 7 1988 nvrmmvcs librae .

Sign up to vote on this title
UsefulNot useful