You are on page 1of 35

TOPOLOGY

FROM THE
DIFFERENTIABLE VIEWPOINT
By John W . Milnor
Princeton University

Based on notes by David W . Weaver

The University Press of Virginia Charlottesville

PREFACE

THESE lectures were delivered at the University of Virginia in December 1963 under the sponsorship of the Page-Barbour Lecture Foundation. They present some topics from the beginnings of topology, centering about L. E. J. Brouwer’s definition, in 1912, of the degree of a mapping. The methods used, however, are those of differential topology, rather than the combinatorial methods of Brouwer. The concept of regular value and the theorem of Sard and Brown, which asserts that every smooth mapping has regular values, play a central role. To simplify the presentation, all manifolds are taken to be infinitely differentiable and to be explicitly embedded in euclidean space. A small amount of point-set topology and of real variable theory is taken for granted. I would like here to express my gratitude to David Weaver, whose untimely death has saddened us all. His excellent set of notes made this manuscript possible.

J. W. M.
Princeton, New Jersey March 1965

vii

CONTENTS

Preface
1. Smooth manifolds and smooth maps Tangent spaces and derivatives Regular values The fundamental theorem of algebra

v11
1 2 7 8

2. The theorem of Sard and Brown Manifolds with boundary The Brouwer fixed point theorem
3. Proof of Sard’s theorem
4. The degree modulo 2 of a mapping Smooth homotopy and smooth isotopy

10 12 13
16

I

20 20 26 27 32
42 50

J

5 . Oriented manifolds The Brouwer degree
6. Vector fields and the Euler number 7. Framed cobordism; the Pontryagin construction The Hopf theorem
8. Exercises

52
55
59

Appendix: Classifying l-manifolds Bibliography Index

63

ix

ax. A map f : X Y is called smooth if for each x E X there exist an open set U C Rk containing x and a smooth mapping F : U --f R 1 that coincides with f throughout U A X . . thus a point x E Rk is an k-tuple x = (xi. ... however. ) . exist and are continuous. (The inverse diffeomorphism W A AI + U is called a system of coordinates on W A M . The following definition singles out a particularly attractive and useful class. We can now indicate roughly what diferential topology is about by saying that it studies those properties of a set X C Rhwhich are invariant under diffeomorphism. More generally let X C Rk and Y C R' be arbitrary subsets of euclidean spaces. note that the composition g o f : X + 2 is also smooth. . The identity map of any set X is automatically smooth. SMOOTH MANIFOLDS AND SMOOTH MAPS FIRST let us explain some of our terms. We do not. Let U C Rk and V C R' be open sets. want to look a t completely arbitrary sets X . A mapping f from U to V (written f : U . Any particular diffeomorphism g : U --f W A Af is called a parametrixation of the region W A M . A map f : X Y is called a di$eomorphisnz if f carries X homeomorphically onto Y and if both f and f-' are smooth. . Rk denotes the k-dimensional euclidean space.+ V ) is called smooth if all of the partial derivatives ay/ax. --f DEFINITION.Q1. xk) of real numbers. If f : X + Y and g : Y -+ 2 are smooth. A subset ilf C Rkis called a smooth inanifold of dimension m if each x E Af has a neighborhood W A M that is diffeomorphic to an open subset U of the euclidean space R". -- ---f DEFINITION.

will be a linear mapping from T M .consisting of all ( 2 . If I i s the identity map of U . By definition. if U C U' are open sets and i:U--+U' I): . evaluated a t 2. Rnkthere corresponds R'.. .f(x))/t EXAMPLES. . 1 for x2 y2 < 1.it follows that S2 is a smooth manifold. Translating both hyperplanes to the origin. is just that linear mapping which corresponds to the 1 X k matrix (af J a x i > . i s the identity map of R k . y ) E R2 with x # 0 and y = sin(l/x). then T M . x. Rkof dimension m called the tangent space of dd a t x. y . I f f : U + V and g : V + W are smooth maps. y > 0. for a smooth map f : M + N of smooth manifolds.(h) = Sometimes we will need to look a t manifolds of dimension zero. we first associate with each x E M C Rk a linear subspace T M . is the hyperplane through the of smooth maps between open subsets of Rk. Smooth manifolds Tangent spaces 3 1 Rm t Figure 1 .By iritcrchanging the roles of x. I n other words. z. then d(g 0 f& = dg. then d I . x < 0.(h) is a linear function of h. z ) E R3 with x2 y2 z2 = 1 is a smooth manifold of dimension 2. (Compare Figures 1 and 2. is defined to be the entire vector space Rk. where y = f ( x ) . we must study the special case of mappings between open sets. Intuitively one thinks of the m-dimensional hyperplane in Rk which 1 best approximates 14 near x. Elements of the vector space T M . Since these cover S2. with (x. M is a manifold of dimension zero if each x E M has a neighborhood W n M consisting of x alone. More generally. Y ) + ( x . The unit sphere S2.1. h E Rk. Then d f . Y .) Here are two fundamental properties of the derivative operation: 1 (Chain rule). and z < 0. More generally the sphere Sn-' C R" consisting of all (xl. For any smooth map f : U + V the derivatiue dfz . For example SoC R' is a manifold consisting of just two points. I n fact the diff eomorphism + + + for x E U .x2 . of first partial derivatives. For any open set U C Rkthe tangent space T U . y < 0. y . Parametrization of a region in IIf origin that is parallel to this.) Similarly one thinks of the nonhomogeneous linear mapping from the tangent hyperplane a t x to the tangent hyperplane a t y which best approximates f . are called tangent vectors to llil a t x. (In fact d f . 0 dfz. f(x) = y. we obtain similar parametrizations of the regions x > 0.. d . lim ( f ( x t-0 + th) . to every commutative triangle U w 9 Of r/\ r\ v TANGENT SPACES AND DERIVATIVES To define the notion of derivative df. Before giving the actual definition. a commutative triangle of linear maps R f Rd 2 49 R Y c f)= 2. Rk+ R' is defined by the formula df.) with x: = 1 is a smooth manifold of dimension n . to TN. and changing the signs of the variables.Q 1. one obtains dfz. parametrizes the region z > 0 of S2. A somewhat wilder example of a smooth manifold is given by the set of all ( x .Y".Clearly df.

C Thrz with irrclusion m u p di.* We say that x e n/r is a regular point of j if the derivative where U = g-'(x). If I i s the identity m a p of M . (Compare Figure 2. y. Since J is smooth there exist an open set W containing x and a smooth map F:W4R1 that coincides with f on W (7 M. If f : M + N and g : N --+ P are smooth. then df. Consider the commutative diagram 9 f P h-1 Figure 2. df.6 51. More generally. 7 of smooth mappings between open sets. i s an isomorphism of vector spaces.(v) to be equal to dF. Choose parametri~t' .) is defined as follows. R m d(h-' T 0 f 0 g > u >R" REGULAR VALUES Let f : ill . then d(g 0 f>r = dgu 0 dfz. As before. : TAT. (Chain rule). carries TAT. then T M .. The derivative dfz : T M . -+ T N .(v) belongs to T N . It follows that h-' o f o g : U -+ V is a well-defined smooth mapping. does not depend on the particular choice of F . 3 TN. -+ T N .+ N C R ' for neighborhoods g ( U ) of x and h ( V ) of y. we may assume that g ( U ) C W and that f maps g ( U ) into h(V). Smooth maniJolds Regular values 7 smooth map transformation by going around the bottom of the diagram. we obtain a commutative diagram of linear mappings RE dgu dFZ >R1 dha. v = h-'(y). with f(s)= y. the derivative operation has two fundamental properties: 1. = Image (dg. De&e df. Replacing U by a smaller set if necessary.) into TN. *4s before. = dh.a ions g:U+MCRE and h : V .(v) for all v e TM. Taking derivatiws. That is: f:M+N with f(x) = df. for we can obtain the same linear * This restriction will be removed in $2. 7'he tangent space of a submanifold fo U gJh V The proofs are straightforward. In payticular the dimension of ill must be equal to the dimension of N . = Image (dh. and that it does not depend on the particular choice of F . 2. . To justify this definition we must prove that dF. these two properties lead to the following: ASSERTION. is a well-defined linear mapping. 0 d(h-' This completes the proof that 0 f 0 g)u 0 (dgJ*. : T M . f f : M I N i s a difleonzoyphism.. if M C N with inclusion m a p i.). then d I z i s the identity m a p of TM.+ N be a smooth map between manifolds of the same dimension. Furthermore the resulting map df.. It follows immediately that d F ..

in N . A A V.. 1) of S2. . 0 . Hence the locally constant function #f-'(y) must actually be constant on this set. 1). for P fails to be a local diffeomorphism only a t the zeros of the derivative polynomial P'(z) = a. 1)) -+ R2 X 0 C R3 from the "north pole" (0. For the proof it is first necessary to pass from the plane of complex numbers to a compact manifold. 0 . 0. 1. [Let x l . and the image f ( x ) is called a critical value. 1) = = Thus Q is smooth in a neighborhood of 0. Since #f-'(y) can't be zero everywhere. and the polynomial P must have a zero. 0. and choose pairwise disjoint neighborhoods U. is singular.. and a regular value y E N . and there are only finitely many zeros since P' is not identically zero. Consider the unit sphere X2 C R3 and the stereographic projection h+ : Sz. where f(x) f ( 0 . and it follows that f = hI'Qhis smooth in a neighborhood of (0. -1) and set &(z) = h-fhI'(z). 0. . Smooth manifolds Fundamental theorem of algebra 9 df.8 01. - .f(M . For a smooth f : M --$ N . The point y E N is called a regular value if f-'(y) contains only regular points.A V. there i s a neighborhood V C N of y such that # . 0. I. For f-'(y) is in any case compact. In this case it follows from the inverse function theorem that f maps a neighborhood of x in M diffeomorphically onto an open set in N . space M . h. The first observation to be made about #f-'(y) is that it is locally constant as a function of y (where y ranges only through regular values!). with a. If df. Next observe that f has only a finite number of critical points. - Kotc. Observe that i M i s compact and ?J E N i s a regular value. we conclude that it is zero nowhere.e. We may then take Figure 3.( ' f ' y ) = #f-'(y) for any y' E V .((0. Ur of these which are mapped diffeomorphically onto neighborhoods V. + + . To see this we introduce the stereographic projection h.-i jzl-'. 0.from the south pole (0.UiJ .U. .hI'(z) S o w if P(z) = a.. x k be the points of f . * . The set of regular values of f . . . . then a short V = V. # 0. . with M compact. Thus f is an onto mapping. even in a neighbor- .' ( y ) . V .(See Figure 3. since f is one-one in a neighborhood of each x P f-'(y). . + a. ) for x # (0. 1) It is well known that this resulting map f is smooth. that h.z + * * * + an?).. Thus each y E N is either a critical value or a regular value according as f-'(y) does or does not contain a critical point.$ alzn-l computation shows that &(z) = = z / 1 ~ 1= ~ -- 1/Z. THE FUNDAMENTAL THEOREM OF ALGEBRA As an application of these notions. and f-'(y) is discrete. Stereographic projection hood of the north pole. is therefore connected. by elementary geometry.'Ph+(x) (0. being a closed subset of the compact. then f-'(y) f is afinite set (possibly empty). The polynomial map P from R2 x 0 itself corresponds to a map f from S2to itself.I z"/(a0 + d.) We will identify R2 X 0 with the plane of complex numbers. then x is called a critical point of f . is nonsingular.. being a sphere with finitely many points removed. we define #f-'(y) to be the number of points in f-'(y). we prove the fundamental theorem of algebra: every nonconstant complex polynomial P(z) must have a zero.

Thus d F . [S]. hence the theorem says simply that f ( U ) has measure zero. T H E THEOREM has rank less than n (i. under F . : T M . [HI. and i f y E N i s a regular value. In particular let M' = f ' y for a regular value y of f : M -+ N .f(C) must be everywhere denset in R". it follows that the complement R" . IN GENERAL.. [361. the derivative djl must map TAI. + * * * + xi. P. is not onto). and let C = { X E U I rankdf. for x E M'.'l the sense indicated in by the next theorem.L ( t ) ) . More generally consider a smooth map f : M -+ N . Any y # 0 is a regular value. C Rk. Brown). from a manifold of dimension m to a manifold of dimension n.m' called the space of normal vectors to M' in M at x. I n order to exploit this corollary we will need the following: Lemma 1. : Rk -+ R'''-n that is nonsingular on this subspace (31 C T M . Hence F maps some neighborhood U of x diffeoinorphically onto a neighborhood I' of (y.Regular values 11 82. (This agrees with our previous definitions in the case n~ = n.' ( y ) C M i s a smooth manifold of dimension m . Let x E f . it has already been f noted that T M : is a subspace of T M . Since y is a regular value. The null space 91 C T M . df. Then C will be called the set of critical points. then clearly C = U . The proof will be given in $3. of dfL. Now define c F :M by F(E) = -+ N X R"-" (f(t).Thc derivative d F . to the hyperplane y X R"'-". and the smooth manifold f-'(l) is the unit sphere. (References [51. Let C be the set of all x E M such that P ROOF.n. It is essential for the proof that f should have many derivatives.' ( y ) .) We will be mainly interested in the case m 2 n. * In other words. Let f : U + R" be a smooth m a p . T h e n the image f(C) C R" has Lebesgue measure zero. This proves that f'y is a smooth manifold o dimension m . therefore will be an (nz.n)-dimensional vector space. f(C) the set of critical values. In fact F maps f-'(y) r\ U diffeomorphically onto (y X R"-") A V.) I M' is a manifold which is contained in &I. If m < n.) Theorem. This result was rediscovered by Dubovickii in 1953 and by Thom in 1954.f (C) the set of repular values of f . But this set will be LLsmall. is then a vector space of dimension na . If M Rkl choose a linear map I. which was proved by A. Consider the function f : R" --+ R defined by f(. (Compare Whitney [38]. Illorse. -() . it is too much to hope that the set of critical values of a smooth map be finite. L ( x ) ) . < n ) . B.n.(. Brown in 1935. U V ) ) . onto 2'". deJined on a n open set U C Rm. (References [30]. T h e set of regular values of a smooth m a p --f N i s everywhere dense in N . is clearly given by the formula dPz(v) = (dfdv). is nonsingular.* Since a set of measure zero cannot contain any nonvacuous open set.e. + TN.it is possible to cover f(C) by a sequence of cubes in R" having total n-dimensional volume less than E . -() f As an example we can give an easy proof that the unit sphere Sm-' is a smooth manifold.. The orthogonal cornplement of TM: in T M . given any B > 0.) Since M can be covered by a countable collection of neighborhoods each diffeomorphic to a n open subset of R"'. and the complement N . t Proved by Arthur B. Sard in 19-12 following earlier work by A. If f : AT --+ N i s a smooth m a p between manifolds of dimension m 2 n.Kote that f-'(y) corresponds. then the set f .) = x: + x. we have: f :M O F SARD AND BROWN Corollary (A.

0 . both for f and for the restriction is a smooth (m . is a smooth manifold. maps the space o j normal vectors to M' isomorphically onto TN. where m > n. Let IT : g-l(y) --f R denote the coordinate projection. we see that df. dg. If z is an interior point. The interior X . it suffices to consider the special case of a map f : H" + R". Here is one method for generating examples. so that T X . 2. C R". with boundary equal to g-'(O) . 2 O}. THE BROUWER FIXED POINT THEOREM We now apply this result to prove the key lemma leading to the classical Brouwer fixed point theorem.dX is a smooth manifold of dimension m. is a full m-dimensional vector space. Furthermore the boundary a ( f . . The unit disk Dm.' ( y ) with aX. Therefore the set g-'(y) f l H" = f-'(y) A U . we may assume that g has no critical points. MANIFOLDS WITH BOUNDARY The lemmas above can be sharpened so as to apply to a map defined on a smooth "manifold with boundary. . then as before f-'(y) is a smooth manifold in the neighborhood of 2. x. Let X be a compact manifold with boundary.) E R" I X. with regular value y E R". This completes the proof. e . Counting dimensions we see that df. Replacing U by a smaller neighborhood if necessary. The null space of df. to zero. with boundary equal to n-'(O).. Suppose that 3 is a boundary point. If y c N i s a regular value." Consider first the closed half-space PROOF. i s precisely equal to the tangent space T M : C T M . A subset X C Rk is called a smooth m-manifold with boundary if each x E X has a neighborhood U n X diffeomorphic to an open subset V n H" of H". ~ .' ( O ) is equal to the null space of DEFINITION.n.1. = d f . : R" + R". Lemma 3. f 1 ax. T h e set of x in M with g(x) 2 0 i s a smooth manifold.. For the tangent space of g-'(y) at a point x c r . Since we have to prove a local property. Hence g-'(y) is a smooth manifold of dimension m . Let Z E f-'(y). by Lemma 3. Sard-Brown theorem Brouwer fixed point theorem 13 E EXAMPLE.) = x. The boundary dH" is defined to be the hyperplane R"-' X 0 We claim that IT has 0 as a regular value. Choose a smooth map g : U + R" that is defined throughout a neighborhood of 3 in R" and coincides with f on U A H". then f-'(y) C X Lemma 4. of the submanifold M' = f-'(y). * . maps the subspace TA!!: C T M . It is not hard to show that dX is a well-defined smooth manifold of dimension nz .' ( y ) ) i s precisely equal to the intersection of f . The boundary aX is the set of all points in X which correspond to points of dH" under such a diffeomorphism. Let M be a manifold without boundary and let g : M ---f R have 0 as regular value. consisting of all x 1- R'" with cx: 0. smooth manifold.n)-manifold with boundary. T N .16 Lemma 2. n(x1. From the diagram - 52. 2 M' i is a. even if x is a boundary point. H" = { ( x 1 . : T M . Hence d f l maps the orthogonal complement of T M : isomorphically onto TN. PROOF. is defined just as in 51. consisting of all x E g-'(y) with n ( x ) 2 0. with boundary equal to A " ' S-. The tangent space T X . Now consider a smooth map f : X -+N from an m-manifold with boundary to an n-manifold. The proof is just like the proof of Lemma 1. but the hypothesis that f I aH" is regular at x guarantees that this null space cannot be completely contained in R"-' X 0..

) the expression under the square root sign being strictly positive.G ( x )I I < p for all x. Suppose g has no fixed point. However. I + U = IIX x . There i s no smooth map f : X wise jixed. with I IP(x) . We reduce this theorem to the lemma by approximating G by a smooth mapping. Problem 4. and the only compact 1-manifolds are finite disjoint unions of circles and segments. .g(4ll ' t = -x. which is impossible by Lemma 5 . Let y E ax be a regular value for f. Then the continuous function 1 IG(x) .14 $2. p. + D" has aJixed point (i. Hirsch).u + d . Suppose that G ( x ) # x for all x E D". according to the Weierstrass approximation theorem. 1 + + The procedure employed here can frequently be applied in more general situations: to prove a proposition about continuous mappings. This contradiction establishes the lemma. . PROOF. aX that leaves a X point- P ROOF (following M. This contradicts Lemma 6 . For x E D". z P(x> = P'(X)/(l + 4. .u)z.) Then f : D" --+ s"-' is a smooth map with f(x) = x for x E s"-'.g(x> . Choosing 1' : D"+ I)" as above. Suppose there were such a map f. we clearly have P ( x ) # x. with boundary consisting of the single point f-'(y) n ax = {v). Here x :) and subsequently llxll denotes the euclidean length d .$1 I must take on :L minimum p > 0 on W . and completes the proof. Any continuous function G : D" + D" has a Jixed point. Fiyrtre But f-'(y) is also compact. (To see that f is smooth we make the following explicit computation: f(x) = x tu. (Compare 58. let f(x) E 8'-' be the point nearer x on the line through x and g(x). may send points * A proof is given in the Appendix. (See Figure 4. a point P ROOF. To correct this we set is a compact manifold bounded by the unit sphere x"-'. .G(x)II < t for 2 E D".x ' x + (x. : R" -+ R" with IIP1(x) .G(s)1 I < 2e for x E D". P . we first establish the result for smooth mappings and then try to use an approximation theorem to pass to the continuous case. Given t > 0. Thus P is a smooth map from D" to itself without a fixed point. f See for example Dieudonn6 [7. Brouwer Fixed Point Theorem. P Lemma 6. 1331.e. x . Since y is certainly a regular value for the identity map f I a also. where Then clearly P maps D" into D" and I IP(x) . A n y smooth map g : D" D" with g(x) = x ) . it follows that f-'(y) is a smooth 1X manifold.* so that aff'(y) must consist of a n even number of points.t there is a polynomial function P . In particular the unit disk 4 D"= (xER"Ix:+ +z:< 1) of D" into points outside of D". Sard-Brown theorem + Brouwer jixed point theorem 15 Lemma 5. Hence as a special case we have proved that the identity map of A"-' cannot be extended to a smooth map D" --f AY-'.

Since C .) has measure zero for k sufficiently large.1)-dimensional measure zero. p 2 1... and let C be the set of critical points. The details are somewhat easier since we assume that f is infinitely differentiable. The cases where n _< p are comparatively easy. R EMARK. . h maps some neighborhood V of 3 diffeomorphically onto a n open set V’.C. Let C. then the intersection of I Figure 5 . * * 1 2 . 51-52]. * * The proof will be divided into three steps as follows: STEP 1. Construction of the map g * Our proof is based on that given by Pontryagin [28].) To start the induction. xz. since C = C. STEP 2. Let f : U . for i 2 1. pp. The proof will be by induction on n. hence the set g(C’) of critical values of g is equal to f ( V n C). however. c R” has measure zero. . this will prove that f (C .Step 1 17 the C. 3 c. We will need the well known theorem of Fubini* which asserts that a measurable set A SARD’S THEOREM* C R” = R‘ X R”-’ iriust have measure xero i it intersects each hyperplane f in a set of ( p . 3 . 3 c. which is not zero a t 3. with U open in Rn. there is some partial derivative.) has measure FIRST let us recall the statement: Theorem of Sard.) has measure zero. Consider the map h : U --f R” defined by < p. give a unified proof which makes these cases look just as bad as the others.C.C. c 3 c. Thus we have a descending sequence of closed sets h(x) = ( f l ( X > . ) . (By definition Ro consists of a single point. More generally let C. the theorem is certainly true for n = 0. Note that the set C’ of critical points of g is precisely h ( V n C). (Compare de Rham [29. (constant) X Rp-’ For each f E C . denote the set of x such that all partial derivatives of f of order < i vanish a t x. is nonsingular. Note that the statement makes sense for n 2 0. Hcnce in this case it is sufficient to carry out Steps 1 and 2. Sincc f # C. 101.C. STEP 3. If f happens to be real analytic. We may assume that p 2 2. say dfl/dzl. The coniposition g = f o h-’ will then map V’ into R”. is covered by countably many of these neighborhoods.C. we will find an open neighborhood V C R” so that f ( V (7 C) has measure zero. P R O O F O F P ROOF O F STEP 1. The image f(C. but the general case follows easily from this special case. p. (REMARK. * For an easy proof (as well as an alternative proof of Sard’s theorem) see Sternberg [Xi. The image f(C.) Q3. C C denote the set of all x E U such that the first derivative d f z is zero. is vacuous unless f is constant on an entire component of U . ..) We will. that is the set of all x E U with rank d f z Then f(C) zero. Sternberg assumes that A is compact.+R” be a smooth map.) has measure zero. Sincc dh. This first step is perhaps the hardest. The image f(C . . when p = 1.

Since c.. the compactness of I".C. Again we consider g = f o h-' : V' -+ R"../ax8. . be a cube of the subdivision which contains a point x of c k . does not. this will prove that f(C. Hence. Suppose for definiteness that s. .4 18 $3. > carries some neighborhood V of diffeomorphically onto an open set V'.* 1 z .. 2. ax. I'rom 1) it follows that / ( I . Ck+l) has measure zero. . ) lies in a cube of edge a / ~ ~ ccntered +' about f(z). . Note that h carries Ck A V into the hyperplane 0 X R"-'. * . . Proof of Surd's theorem Step 3 19 For each ( t .) E V' note that g(t. . where 1) + h) ll&. C) PROOFF S TEP 2.( k + l ) p = If k 1 > n / p .1 to be precise) we will prove that f(Ck A I") has measure zero. x2. This completes the proof of Sard's theorem. But each point in h(C. by Fubini's theorem. akf. .*+. Now subdivide I" into Y* cubes of edge 6/r. x h E I". -. By induction.Ck+1 is covered by countably many such sets V . Then the map h : U -+ R" defined by 1. the set I for x E C. Let I" C U be a cube with edge 6. then evidently V tends to 0 as 1 -+ m . Therefore gh(Ck n V ) = f(Ck n V ) has measure zero. Let I . Then any point of I1 can be written as x h. we see that f(. it follows that f(C. If k is sufficiently OF large ( k > n / p . A V ) is certainly a critical point of 0 (since all derivatives of order _<k vanish). and Step 1 is complete.) has measure zero.+. n In) ' must have measure zero./ax. the set of critical values of 0 has measure zero in R". 22. h(z>= (W(Z>. the set of critical values of g t has measure zero in t X R"-'. Here c is a coiistant wliich depends oiily on f and I". From Taylor's theorem..az. there is some (Ic O I)-'' derivative ak*'f. Let gt : ( t X Rn-l) A V' -+ t X R"-' 1 denote the restriction of g. Thus the function W(X) = + llhll I di(6h). where a = 2c (di 6 ) k + 1 is constant. Therefore the set of critical values of g intersects each hyperplane t x R"-' in a set of measure zero.. xpl . and the definition of ck. Since c k can be covered by countably many such cubes. = + vanishes at but aw/dx.J belongs to the hyperplane t X R"-' C R": thus g carries hyperplanes into hyperplanes. . since it can be expressed as a countable union (' of compact subsets.. Let 0 : (0 X Rn-') n V' -+ R" denote the restriction of g. EIence f(ck I") is contained in a union of a t most 1 cubes having total volume ' 2) V 5 rn(a/l'k+l)n aD1'n .+. with + WlI I c + g(C'> = j(V has measure zero. . This set g C ) is measurable.which is not zero. x. = f(x) + R(z1 h) llhll"' According to the induction hypothesis. For each z E c k .. for the matrix of first derivatives of g has the form PROOF S TEP 3. Note that a point of t X R"-' is critical for gt if and only if it is critical for g. so f(C. A I".

THE DEGREE MODULO 2 O F A MAPPING for all z E X . since otherwise the mod 2 degree would necessarily be zero.) Given a smooth homotopy F between f and g.-I) for 7 > 0. I] denote the subset* of Rk” consisting let of all (x. then. then M X [0. t ) = g(x) - 5 for Q 5 t 5 + 1. y is also a regular value for F . the formula G(x.0 Y. More generally this same definition works for any smooth (For example. We will prove that the residue class modulo 2 of #f-’(y) not depend on does the choice of the regular value y.t) with z E X and 0 5 t 5 1. To see that it is transitive we use the existence of a smooth function rp : [0. * If M is a smooth manifold without boundary. then #f-’(y> = #9-’(y> :x+ Y are called smoothly homotopic (abbreviated f - g) if there exists a (mod 2). This map F is called a smooth homotopy between f and g. g : M --+ N be smoothly homotopic m a p s between manifolds of the same dimension. This residue class is called the mod 2 degree of f . PROOF. I f y E N i s a regular value for both f and g .8 ) A($ . Fb. Let f. t ) = f(x) for 0 5 t I f:M--+N where M is compact without boundary. CONSIDER a smooth map f : S“ + S”. I] 3 [0. Given X C Rk. 1 for each t E [0. 13. t) maps X diffeomorphically ont. let p(t) = X(t . recall that #f-’(y) denotes the number of solutions x to the equation f(z) = y . 1 -+ Y from f to g so that. I] with p(t) = 0 for 0 5 t 5 for $ 5 p(t) = 1 5 t 5 1. This also will be an equivalence relation. 11 is a smooth manifold bounded by two “copies” of M . Note that the relation of smooth homotopy is an equivalence relation. The diffeomorphism f is smoothly isotopic to g if there exists a smooth homotopy F : X X [O.Homotopy and isotopy 21 smooth map F : X X [O. we can also define the concept of a “smooth isotopy” between f and g.t ) ) . Two mappings f. Boundary points of M will give rise to “corner” points of M X [O.Q ) / ( X ( t . N is connected. Then F-’(y) . with the aid of this construction. 1) = 9(z) 84. 0) I = fb). 11 + N be a smooth homotopy between f and g. g It will turn out that the mod 2 degree of a map depends only on its smooth homotopy class: Homotopy Lemma. Let F : M X [0. If y is a regular value. X X [0. where X ( T ) = 0 for T 5 0 and X(7) = exp( -. If f and g happen to be diffeomorphisms from X to Y . Now if f g and g h. - G(z. the correspondence z -+ SMOOTH HOMOTOPY AND SMOOTH ISOTOPY F(z. . (We may as well assume also that N is compact without boundary. I] + Y with F ( x . it is easy to prove that f h. First suppose that. 11. DEFINITION. p ( t ) ) defines a smooth homotopy G with + G(z. where M i s compact and without boundary.) For the proof we introduce two new concepts. and both manifolds have the same dimension. t ) = F ( x .

n. Dejorming the unit ball Mx 0 Mxl Figure 6 . The number of boundary points o n the left i s congruent to the number o n the right modulo 2 Let cp : R" --+ R be a smooth function which satisfies &) q(x) >0 = . Now suppose that y is not a regular value of F. - #-(' f'Y> for all y' E = #f-W .). But we recall from $2 that a compact 1-manifold always has an even number of boundary points. . -. Then clearly 1) F . Choose a regular value z of F within V .(Z). 3) F . C Figure 7 .ci&.he differential equations dxi d t -. 0 F. Thus #j-'(y) #g-'(y) is even. C N of y .22 $4. connected manifold N . We will use the notation x ( t ) = F t ( 2 ) for this solution. and therefore #f-'(y) = #g-'(Y>T(mod 2 ) . + .) Given any fixed unit vector c E S"". I ~ * Compare [22.41. and 2 ) slides the origin to any desired point of the open unit ball. We will also need the following: Homogeneity Lemma. consisting of regular values of f . i = 1. ( 2 ) is defined for all t and 2 and depends smoothly on t and 2.l l ~ 1 1 ~ ) where A ( t ) = 0 for t 5 0 and A(t) = exp(-t-') for t > 0. defined for all* real numbers which satisfies the initial condition x(0) = 2. so that (For example let p(z) = A(l . #g-'(u') = #g-'(Y) c N so that For any 2 E R" these equations have a unique solution x = x ( t ) .. consider t. for all y' E V. ( 3 ) = F . 2) F"(3) = 3. - * * .. z. Thus the total number of boundary points of F-'(y) is equal to + #s-'(y). which completes the proof. Then #f-'(Y> = #f-'(z) = #g-%) = #s-YY)). Thus there is a neighborhood V. 02. Let y and z be arbitrary interior points of the smooth. A V..for 11x1 I <1 0 for 11x11 2 1. and there is an analogous neighborhood V . Recall (from 51) that #f-'(y') and #g-'(y') are locally constant functions of y' (as long as we stay away from critical values).) The proof in general proceeds as follows: We will first construct a smooth isotopy from R" to itself which 1) leaves all points outside of the unit ball fixed. Degree modulo 2 Homotopy and isotopy 23 is a compact 1-manifold. + (For the special case N = x" the proof is easy: simply choose h to be the rotation which carries y into z and leaves fixed all vectors orthogonal to the plane through y and z. with boundary equal to ~ . V..y y n (M ) x o v ~I/Ix #f-'(y) 1) = f-'(y) x o v g-l(y) x 1. Then there exists a diffeomorphism h: N +N that i s smoothly isotopic to the identity and carries y into z.

will carry the origin to any desired point in the open unit ball. A constant map c : M -+ M has even mod 2 degree. Call two points of N "isotopic" if there exists a smooth isotopy carrying one to the other. Theorem. This completes the proof. with suitable choice of c and t. ) For such a map f would give rise to a smooth homotopy I ! F ( z . Given regular values y and z.this result implies the assertion that no smooth map f : Dn+' S" leaves the sphere pointwise fixed.Now suppose that f is smoothly homotopic to g. and completes the proof. We can now prove the main result of this section. f is a smooth homotopy invariant. Assume that M is compact and boundaryless. Hence the identity m a p of a compact boundaryless manifold i s not homotopic to a constant. let h be a diffeomorphism from N to N which is isotopic to the identity and which carries y to z. Compare $2. which is a regular value for both f and g. the Homotopy Lemma asserts that #(h 0 f)-'(z) = #f-'(z) (mod 2). Therefore #f-'(y) = #f-'(z) (mod 21. The identity map I of M has odd degree. that N is connected. hence there can be only one such isotopy class. hence the above argument shows that every point sufficiently close to y is "isotopic" to y. and the interior of N is partitioned into disjoint open isotopy classes. If y and z are regular values of f then #-() f'y = #f-'(z) (modulo 2). But clearly.24 $4. I n the case M = s". (I. t) between a constant map and the identity. PROOF. EXAMPLES. -() so that #(h O f)-'(z) = #f-'(y>. is a diffeomorphism from R" onto itself. then it has a neighborhood diffeomorphic to R". the diffeomorphism F . there exists a n element y P N . Since h o f is homotopic to I . 1 = f(tz). Call this common residue class deg2(f). we see that each F . each "isotopy class" of points in the interior of N is an open set. F : S" X [0.. which i s called the mod 2 degree of f . The congruence (mod 2) deg. Letting t vary. If y is an interior point. depends only on the smooth homotopy class of f. Lemma 5 . 1 -+ Sn. g now shows that deg. Then z is a regular value of the composition h o f . But the interior of N is connected. By Sard's theorem. c But (h 0 f)-'(z) = f-'h-'(z) = f'y. and that f : A4 -+ N is smooth. the sphere -+ is not a smooth ((retract" of the disk. I n other words. T h i s common residue class. f = #f-'(y) = #g-'(y) = deg. is smoothly isotopic to the identity under an isotopy which leaves all points outside of the unit ball fixed. This is clearly a n equivalence relation. Now consider a connected manifold N . Degree modulo 2 Homotopy and isotopy 25 Therefore each F . as required.e.

) determines the same orientation as the basis (6.1)dimensional subspace T ( d M ) .1 or 1 according as a positively oriented vector a t x points inward or outward. preserves or reverses orientation. An orientation for a finite dimensional real vector space is an equivalence class of ordered bases as follows: the ordered basis (bll .. An oriented smooth manifold consists of a manifold M together with a choice of orientation for each tangent space TM. at a boundary point: --.. is an ‘(outward” vector.. If M is compact and N is connected. * . carries the specified orientation for T M .. . .- -- - + As an example the unit sphere Sm-’ R” can be oriented as the C boundary of the disk D”. -+ TNt(. If m 2 1. 0. . b. How to orient a boundary . DEFINITIONS. 1). I n the case of the zero dimensional vector space it is convenient to define an ‘(orientation” as the symbol 1 or . . 3) there are the (‘inward” vectors forming a complementary half space.1. we can distinguish three kinds of vectors in f the tangent space T M . these are required to fit together as follows: For each point of M there should exist a neighborhood U C M and a diffeomorphism h mapping U onto an open subset of R“ or H”’ which is orientation preserving. U . 2) there are the (‘outward” vectors. into the standard orientation for R”.) -- ) + IN ORDER to define the degree as an integer (rather than an integer modulo 2) we must introduce orientations. be a smooth map. THE BROUWER DEGREE Now let M and N be oriented n-dimensional manifolds without boundary and let f :M+N 1) there are the vectors tangent to the boundary. . (See Figure 8.. 0 ) )(0. If the dimension of M is 1. forming an open half space bounded by T(dM). (0. O R I E N T E D MANIFOLDS Each orientation for M determines an orientation for dM as follows: For x E dM choose a positively oriented basis (ul. U) for T M .) > 0. in the sense that for each x E U the isomorphism dh. this integer deg(f... C TM. . Thus each positive dimensional vector space has precisely two orientations.. Define the sign of df. I M has a boundary. 0).b. I M is connected and orientable. . : T M .T h e Rrouwer. are tangent to the boundary (assuming that m 2 2 ) and that v. . U. with det(a. It determines the opposite orientation if det(a.1 according as df. As in $1. then the degree of f is defined as follows: Let x E M be a regular point of f . then it has precisely two f orientations. 0. 1I Figure 8. v2) . in such a way that v2. . forming an (m . It is defined on a dense open subset of N . degree 27 Q5. 0. The vector space R” has a standard orientation corresponding to the basis (1. y) = zrf-’(u) sign df. is a linear isomorphism between oriented vector spaces. .. . . 6 9 if b: = a. y) is a locally constant function of y. to be 1 or . so that df. . Then (u2. 1..) determines the required orientation for dM a t x. For any regular value y E N define + deg(f.. then each boundary point x is assigned the orientation .) < 0.

we have proved that deg(f. If y and z are both regular values for f : M ---$ N . f . More generally. How to orient F . y ) is constant within some neigliborhood U of y. Figure 9. y ) deg(f. z).) The degenerate mapping f :M --+ constant e N . y ) = 0 for every regular value y . maps the unit circle onto itself with degree k . choose a diffeomorphism h : N -+ N that carries y to z and is isotopic to the identity. The compact. y ) does not depend o n the choice of regular value y . Hence. EXAMPLES. I] x Ai? -+ N g b ) = F(1. Y) . . determines the required orientation for T A . T h e degree deg(g. First consider the following situation: Suppose that M is the boundary of a compact oriented manifold X and that M is oriented as the boundary of with sum zero. Lemma 1.. I] X M" can be oriented as a product. Y ) = 0. as in $4. Theorem B. z) - by Lemma 2. deg(f.+') into a positively oriented f basis for TN. The complex function z -+ zk. suppose that yo is a regular value for f. with U.. : X --f N . Adding up over all such arcs A . between Lemma 2. If f : M -+ N extends to a smooth m a p F deg(f. 1 X M") a t a regular value y is equal to the difference deg(g. The remainder of the proof of Theorems A and B is completely analogous to the argument in 54. z ) . and + and hence (summing over all such arcs) that deg(/ . Let A C F-'(y) be one of these arcs. The function deg(f. carries (vz.' ( y ) deg(f.deg(f. with only the boundary points of the arcs lying on M = a X . Oriented manifolds T h e Brouwer degree 29 Theorem A. Clearly U. signdf.(x) points outward a t one boundary point (say b) and inward a t the other boundary point a.z # 0. Thus the degree 1 of F I a([O. negative. is a smooth function. PROOF. Yo) This proves Lemma 1. y) is equal to deg(f. v. let (vl. Y>* According to Lemma 1 this difference must be zero. hence deg(h 0 The orientations for X and N determine an orientation for A as follows: Given z e A . Y ) = deg(h O f. tangent to A. = x. Then h will preserve orientation. U. y) for a n y common regular value y . as well as for f = F I M . First suppose that y is a regular value for F . h ( d ) by inspection. . with aA = { a ) U ( 0 ) . = -1. y ) = 0. Let v1 ( x ) denote the positively oriented unit vector tangent to A a t x. i and only if d F . It will be called the degree of f (denoted deg f). . we can choose a regular value y for F within U and observe that d 4 f . (Here k may be positive. which completes the proof. X I . F : [0. T h e integer deg(f.. Therefore deg(f. and v. = +l. z) = = deg(f.. We will show that sign dfa sign dfb = 0. manifold [0. * * .28 $5. then PROOF. It follows immediately that signdf. y) = 0. But f is homotopic to h o f . The and will then have boundary consisting of 1 X M" (with the correct orientation) and 0 X M" (with the wrong orientation).+J be a positively oriented basis for T X . but not for F . Now consider a smooth homotopy two mappings fb) = F(O.' ( y ) is a finite union of arcs and circles.It is only necessary to keep careful control of orientations. then deg f = deg g. or zero. 1-manifold F . T h e n U. . If f i s smoothly homotopic to g .. The proof will be essentially the same as that in $4.

then we may as well suppose that 2) 31 has degree zero. (Compare Figures 10 and 11. A smooth tangent vector jield on M C Rk is a smooth map v : M -+ Rksuch that ~ ( x E) T M .2r(x) = 1 for all z E S".23. following Brouwer. : S" ---f x". that the antipodal map of S" i s homotopic to the identity for n odd. 2n+1) = ( ~ 1 . I n the case of R " this is clearly equivalent to the condition " the sphere S" c 1) V(X). . Attempts for n = B DEFINITION.0) i = z.F(z. r. e) and that Y 1 T h u s if n i s even..-. Computation shows that = -z = r1 o r 2 o 0 .cl. A diffeomorphism f : Jl + N has dtgree 1 or . On the other hand. -L. where + v(z). . .* . if n = 2k . 54. . Oriented manifolds T h e Brouwer degree using the euclidean inner product. This completes the proof. T h u s a n orientation reversing difeoinorphism of a compact boundaryless manifold i s not smoothly homotopic to the identity. F(z. a fact not detected by the dcgree modulo 2. e). the explicit formula U(X1.. 7 X2ki -22k-1) defines a nonzero tangent vector field on S". F ( X . incidentally. Figure 10 (above). as we can see by noting that the antipodal map is the composition of n 1 reflections: + F : S" X [0.. . 6) = z + S" cos 0 + v ( x ) sin 6. . * . For in any case 8(z) = v ( ~ ) / ~ ~ would)be ~ vector field which does v~~ ~ a satisfy this condition.1 according as f preserves or reverses orientation. * . Now define a smooth homotopy The antipodal map of S" has degree (. I n 57 f f we will prove a more general result which implies Hopf's theorem.(zl.1. Rut for n even we have seen that this is impossible. 7r] by the formula F(z. .30 $5. G+J. If v(z) is nonzero for all z. L2k) = (22. One example of an orientation reversing difi'eomorphism is provided by the reflection r . the antipodal m a p of x" i s not smoothly homotopic to the identity.. OT.l)"". A nonzero vector jield on the 1-sphere Figure 11 (below). I t follows. .) F(z. a - . A famous theorem due to Heinz Hopf asserts that two mappings from a connected n-manifold to the n-sphere are smoothly homotopic i and only i they have the same degree. we show that S" admits a f f smooth field of nonzero tangent vectors i and only i n i s odd. As an application. .+~(X). Thus the antipodal map of S" is homotopic to the identity. Thus we can think of v as a smooth function from x" to itself.X = 0 for all 2E S". . for each x E M . 7r) = -2.

C=+l c =+2 Figure 12. .)/ = I lv(4 I I I maps a small sphere centered a t z into the unit sphere. To explain what this means. let us consider the more general situation of a map f : M --+ N . Some examples. and the function zk defines a vector field with a zero of index -k. VECTOR FIELDS AND THE EULER NUMBER t=I A s a further application of the concept of degree. we study vector fields on other manifolds. . . Suppose that the vector jield v on U corresponds to V' = df o of-' on U' under a diffeomorphisiii f : U + U'.* The degree of this mapping is called the index L of v a t the zero z.The index 33 86./dt = v. T h e n the index of v at a n isolated ' zero z i s equal to the index of U at f(z). We must prove that this concept of index is invariant under diffeomorphism of U. * Each sphere is to be oriented as the boundary of the corresponding disk. Examples of plane vector fields DEFINITION. + If f is a diffeomorphism. = df 0 2. then clearly U' is uniquely determined by The notation 2)' U. 0. with a vector field on each manifold. are illustrated in Figure 12.(xl. Lemma 1. 2. It is these curves which are actually sketched in Figure 12. Consider first an open set U C R" and a smooth vector field v :U--+R" with an isolated zero a t the point z E t = O U . carries V ( X ) into v ' ( f ( x ) ) for each x E M . 0 f-1 will be used.) A zero with arbitrary index can be obtained as follows: In the plane of complex numbers the polynomial zk defines a smooth vector field with a zero of index k a t the origin. xJ. The function fib> U(. 1. with indices . (Intimately associated with v are the curves "tangent" to v which are obtained by solving the differential equations dx. vector fields v on A I and U' on N correspond under f The if the derivative df.1. .

* * In particular this index s u m i s a topological invariant of M : it does not depend on the particular choice of vector jield. This will be our first and last reference to homology theory.) We will prove part of this theorem. * + xmgm(tx) for all values of 1. If M has a boundary. and f. We may assume that z = f ( x ) = 0 and that U is convex. then. 4041. 5' = p 0 0 P ROOF. which is clearly isotoyi(* to the identity. + <t 1. This proves Lemma 2. f l = f . we can define the concept of index for a vector field w on an arbitrary manifold M as follows: If g : U + M is a parametrization of a neighborhood of z in M . and sketch a proof of the rest.+ sm-' assigns to each x E aX the outward unit normal vector a t x. 51. Hence the index of v = vo a t 0 must be equal to the index of U' = U . even as t f(x) = where g .. proceeding exactly as above. See [20. ( M ) . p.34 $6.+ xmgm(x). Then v' = p o v 0 p-l. It clearly will follow from Lemma 1 that L is well defined. . OF PROOF ax . 1 . If f preserves orientation. * See for example [22. t) F ( x . t-0 it is natural to define an isotopy F : R" X [O. which corresponds to v on U . then the index L of w a t x is defined to be equal to the index of the corresponding vector field dg-' 0 w o g on U a t the zero g-'(z). x1g1(x) 0. (A 2-dimensional version of this theorem was proved by Poincare in 1885. 0) = = f(tx)/t for 0 dfo(x>. These vector fields are all defined and nonzero on a sufficiently small sphere centered a t 0. p.' on f . for many values of m there exists an orientation prcserving diffeomorphism of the sphere S'" which is not smoothly isotopic to the identity. and note that F(xi t ) = + . Let X C R" be a compact m-manifold with boundary. First consider the special case of a compact domain in R"'. The s u m C L the indices at the zeros of such a vector jield i s equal to the Euler number* d f o ( x ) = lim f ( t z ) / t . . so the associated function e'(x) = v'(x)/l lv'(z)lj on the +sphere satisfies 0 p-1. R" 1+ by the formula F(x. * Here H i ( M ) denotes the i-th homology group of M . The full theorem was proved by Hopf [14] in 1926 after earlier partial results by Brouwer and Hadamard. denote the vector field d f t o v of. . Since the derivative a t 0 Evidently the degree of 8' equals the degree of fi. A n y orientation preserving diffeomorphism f of R" i s smoothly isotopic to the identity. Let U . We will study the following classical result: Let M be A compact manifold and w a smooth vector field on M with isolated zeros. The Gauss mapping 9 : Xlgl(tx) + * . Vector jields The index sum we construct a one-parameter family of embeddings f t : U+R" 36 Assuming Lemma 1. g m are suitable smooth functions. To prove that F is smooth. We may assume that f(0) can be defined by = 0. . The proof of Lemma 1 will be based on the proof of a quite different result: Lemma 2. ( U ) . (In contrast.(O) = 0 for all t.) with f o = identity. LEMMA 1. then w i s required to point outward at all boundary points.=O (-1)' rank H . which completes the proof of Lemma 1. To consider diffeomorphisms which reverse orientation it is sufficient to consider the special case of a reflection p . m x(M) = . Thus f is isotopic to the linear mapping dfn. we write f in the form* . of Poincare-Hopf Theorem. a t 0. This proves Lemma 1 for orientation preserving diff eomorphisms.

Vector .). we obtain a new manifold with boundary. if a vector field on the disk D" points outward along the boundary. For tn odd it is equal to half the Euler number of a X . In particular. More generally consider a zero z of a vector field w on a manifold A l C Rk. : R"' + R"' is defined.(e') = ah/au. note that First - . t" form a basis for the tangent space TllIhc. SO that w(h(u)) = dh. +1 or . into the subspace T M . Consider first a vector field U on an open set U C R"' and think of v as a mapping U -+ R". some more preliminaries are needed. 1) v. + PROOF.)t' + c:v. c does not depend on the choice of U .) .36 $6. we have seen that vlUo can be deformed smoothly into the identity without introducing any new zeros. I U preserves orienf tation. and the degrees on the other boundary components add up to -E L . Removing a n €-ball around each zero.(v) = Therefore 2) Vati- aw(h(u))/au.e' be the vector field on U which corresponds to the vector Let = field w on M . is nonsingular. We must compute the image of t' = t'(u) under the linear transformation dw./au. dwh(-)(t') = d(w 0 h). so that dv. c DEFINITION. A n example with index sum +1 PROOF. If v : X + R" i s a smooth vector Jield with isolated zeros. But 6 I a X is homotopic to 9. Think of w as a map from llf to Rk so that the dcrivative dw.1 according as D i s positive or negative. The index of v at a nondegenerate zero z i s either according as the determinant of dv. Let e' denote the i-th basis vector of R" and let tt = dh.1 PROOF. T h e derivative dw. -+ Rk is defined. hence L = -1. It follows that z is a n isolated zero. then the index s u m L i s equal to the degree of the Gauss inapping from dX to S"-'. * .) in terms of the derivatives of v a t z.(.) Hence the index is certainly equal to +l. Hence the sum of the degrees of 8 restricted to the various boundary components is zero. = E: (av. (Compare Figure 13.(at'/azc. (See Lemmas 1. Let h U 4 A I be a parametrization of some neighborhood of z . . If v reverses orientation. The function 8 ( x ) = v(z)/ilv(x)ll maps this manifold into S"-'. of z into R". : T M .' ~ w 0 h. For example. actually carries T M . then L = + l . since it can be expressed as a constant times the integral over a X of the Gaussian curvature.fields The index sum 37 Lemma 3 (Hopf). Figure IS. We may assume that z = 0. to itself. Think of v as a diffeomorphism from some convex neighborhood U .) Therefore deg(g) as required. (The minus sign occurs since each small sphere gets the wrong orientation. By definition U = d h . L Lemma 5.. then similarly v can be deformed into a reflection. hence can be cmsidered as a linear transformation from and T M . Lemma 4.. i s positive or negative.. It is natural to try to compute the index of a vector field v at a zero z so that the vectors i'. The vector field v is nondegenerate a t z if the linear transformation dv. C Rk.(e') = aw(h(u))/au. Before extending this result to other manifolds.. The degree of g is also known as the "curvatura integra" of d X . 2. = 0 REMARK. and if v points out of X along the boundary. This integer is of course equal to the Euler number of X . If this linear transformation has determinant D # 0 then z i s a n isolated zero of w with index equal to 1 or .

See also Chern [S].U . p An easy computation shows that the gradient of grad p = is given by Thus dw. Thus the invariant L is equal to 0 or 2 according as m is odd or even. Hence the index of U’ a t the zero z is equal to the index L of v a t x .( p . is equal to the determinant of dv.)t’.). Thus the determinant of d w .) .r(z))/e. and the determinant L) of this linear transformation T M . let ~ ( x E)M denote the closest point of AI. At the north pole the vectors converge inward. * A different interpretation of this degree has been given by Allendoerfer and Fenchel: the degree of g can be expressed as the integral over M of a suitable curvature scalar. then each index is and multiplied by (. the index s u m L +u(44). This gives a new proof that every vector field on an even sphere has a zero. Now consider a compact. the set of all x E Rk with 15 . Problem 12. For E sufficiently small one can show that N . denote the closed eneighborhood of M(i. thus yielding an m-dimensional version of the classical Gauss-Bonnet theorem. hence the index is 1.(h) dw.(h) for all h E TA1. T h e index s u m 39 Combining 1) and 2). (See Figure 11. EXAMPLES. implies that L = 0 * For example. If E is sufficiently small.. . into itself.) Note that the vector x . boundaryless manifold M C Rk. Together with Lemma 4 this completes the proof./au.r(z)) Theorem 1. (Compurc $8. we see that dw. i s equal to the degree of the Gauss mapping* g : d N .l)m. + TAI. Computing the derivative of w :it a zero z E AI. [9].r(x)) arid u ( r ( x ) ) are mutually orthogonal. Problem 11. is a smooth manifold with boundary. I n particular this sum does not depend on the choice of vector j e l d .* At the south pole the vectors radiate outward. otherwise ~ ( x would not be the closest l for ) point of M . boundaryless manifold the invariant L is zero. maps T M . Kow according to Lemma 3 the index sum L is equal to the degree of g./au. For any odd-dimensional. (References [l]. dv. for each point 2 of the level surface d N . h for h E TM:. -+ Sk-’..(h) = = PROOF. and then evaluating a t the zero h-’(z) of obtain the formula 3) we We will also consider the squared distance function p(x) = 12 1 dwz(t*) = xi(dv. by setting w(z) = (x .r ( x ) is perpendicular to the tangent space of il a t ~(x). v can be defined by the formula u(z) = p . This proves Theorem 1. then the function r(z) is smooth a11d well defined. unit normal vector is given by the outward g(z) = grad p/l lgrad p [I Extend U = (z .) for 172 odd. Hence.Let N . since the inner product w ( x ) . For x E N .38 $6.r(2)1j2.e.) 2(x . For a n y vector field v on M with only nondegenerate zeros. For if the vector field v is replaced by . where p is the north pole.yII 5 1 for some y E M ) . to a vector field w on the neighborhood N .r(z)). The eneighborhood of M c L = (-1)” c 1. = p-’(cz).g ( x ) is equal to e > 0. (See $8.l)m. Kote that w can vanish only a t the zeros of v in M : this is clear since the two summands (x . z ) z . On the sphere S“ there exists a vector field v which points “north” a t every point. Vector jields U. Then w points outward along the boundary. hence the index is (. the equality + Figure 14. is equal to the determinant of the matrix (av.

For details of this argument the reader is referred to Milnor [22. 361.e. More generally consider vector fields on a compact manifold M . then a theorem REMARK. L with the Euler number x(M). Vector fields The index s u m 41 L = 0 on a connected manifold M . If x : U--+ [ O . if defined as before by w(z) = v(r(z)) z . But if y is sufficiently small. differentiable of class Cm) manifold.N I . I n order to obtain the full strength of the PoincarBHopf theorem.field with isolated zeros can be replaced by a nondegenerate vector jield without altering the 1. then the vector field d ( z ) = U)(. three further steps are needed. The most pleasant way of doing field on M with but only a C'-manifold. STEP 2. The argument can nonetheless be carried out either by showing that our strong differentiability assumptions are not really necessary or by other methods. However. Furthermore. .40 56. . pp. Applying this argument locally we see that any vector . Thus N . * Clearly d is nondegenerate within N I . Morse showed that the sum of indices associated with such a gradient field is equal to the Euler number of M . will only be a continuous vector field near aM. Manifolds with boundary. then v' will have no zeros a t all within N .X(x)y is nondegenerate* within N . The extension w. then any vector field v which points outward along aM can again be extended over the neighborhood N . integer C STEP 3. and if y is a sufficiently small regular value of U . and hence does not change during this alteration. + this is the following: According to Marston Morse. there is some difficulty with smoothness around the boundary of M . Proving the theorem /or a vector Jield with degenerate zeros. The sum of the indices a t the zeros within N can be evaluated as the degree of the map 0 : aN + X"-'. so as to point outward along a N . STEP 1. 11 takes the value 1 on a small neighborhood N . Consider first a vector field v on an open set U with an isolated zero at z. . 29. is not a smooth (i. I f of Hopf asserts that there exists a vector field on M with no zeros a t all. it is always possible to find a real valued function on M whose ('gradient" is a nondegenerate vector field. If M C Rk has a boundary. of z and the value 0 outside a slightly larger neighborhood N .r ( z ) . Identification of the invariant It is sufficient to construct just one example of a nondegenerate vector i equal to x ( M ) .

~ " ( x of ff'(y). = 0) for (2. 1 1 so that that maps into U' under df. . due to Pontryagin..t ) (w'(x).n is called The the codimension of the submanifolds. f . f'y. This framed manifold (f-'(y). 1 between N and N' and a framing U of X . F Theorem A. . . (See Figure 15. 1 can be extended to a compact manifold 1 X C M X 10. but they all belong to a single framed cobordism class: F DEFINITION. If y' is another regular value off and b' is a positively then the framed manifold ( . First some definitions. E) E. For each x E f-'(y) recall from page 12 that dfb : TAP. f*b) will be called the Pontryagin manifold associated with f . for (z. uE(z. . A framing of the submanifold N C M is a smooth function t. Hence there is a unique vector DEFINITION.) DEFINITION. U") for the tangent space T ( S Y ) v . . Theorem B.) The pair ( N . f*d) is f'y) oriented basis for T(SI))III. . of normal vectors to N in M at x. Two mappings from M to s" are smoothly homotopic i and only if the associated Pontryagin inanifolds are framed cobordant. It will be convenient to use the notation tm = f*b for the resulting framing w'(x). . Two framed submanifolds ( N . N is cobordant to N' within M if the subset N X [ 0 . . THE degree of a mapping M -+ M' is defined only when the manifolds M and M' are oriented and have the same dimension. t) = (U'(. of M X [0. Again this is an equivalence relation. I] w'(x) E Tj-'(y): c TM. difference of dimensions m . We will study a generalization.The Pontryagin cmstruction @ 87. . which assigns to each x E N a basis b(x) = ( U ' . .). E) U N' X (1 - E. (See . Let N and N' be compact n-dimensional submanifolds of M with aN = aN' = aM = 0. and so that X does not intersect M X 0 U M X 1 except at the points of ax. t ) E N' X (1 0) I]. b) are framed cobordant if there 1 exists a cobordism X C A1 x [0. which is defined for a smooth map f:M--+S" from a n arbitrary compact. t ) E N X [0. ) . so that u'(x. Of course f has many Pontryagin manifolds. The map f induces a framing of the manifold f-'(y) as follows: Choose a positively oriented basis b = (U'.:Pastiny together two eobordisms within M Figure 16. vrn-'(x)) for the space TN: C T M . ) a = N x OUN' x x 1. -+ z'(sp)y maps the subspace Tf-'(y)=to zero and maps its orthogonal complement Tf-'(y)t isoinorphically onto T(Sv)II. corresponding to different choices of y and b. b) is called a framed submanifold of M . Now consider a smooth map f : AP + S" and a regular value y E S". Clearly cobordism is an equivalence relation.( ' . . FRAMED COBORDISM MxO MX I MX2 T H E PONTRYAGIN CONSTRUCTION Figure 15. boundaryless manifold to a sphere. framed cobordant to (-() f*b). b) and (N'.

By abuse of notation we will often delete reference to f*b and speak simply of "the framed manifold f-'(y). Theorem C.f(z). R ) of matrices with positive determinant. f*b'). an isotopy) r r : S" S" so that rl(y) = z.' ( z ) = f-'rF'(z) proves Lemma 2. PROOF. and z i s suficiently close to y.yI( < e. then the Pontryagin manifold (f-'(y). It will be based on three lemmas. Herice F-'(z) C M X 10. f ( z ) 0 5 t < e. Choose a smooth path from b to b in the space of all positively ' oriented bases for T(S"). This is possible since this space of bases is framed cobordant to f-'(y) and so that g . Since the set f(C) of critical values is conipact. 1 --f Sv 1 by F(z. If f and g are smoothly homotopic and y i s a regular value for both. I]. Then F-'(z) is a framed manifold and provides a framed cobordism between f-'(z) and g-'(z). h ~ ) j codimension p o in M occurs as Pontryagin manifold for some smooth mapping f : M S". and hence is a regular value of f.e. choose a smooth one-parameter family of rotations (i. --f Thus the homotopy classes of maps are in one-one correspondence with the framed cobordism classes of submanifolds.44 $7. If y i s U regular value off. This proves Lemma 3. A n y compact /rained suhmanifold ( N . F(z. The proof of Theorem A will be very similar to the arguments in $04 and 5 . Framed cobordisin The Pontryagin construction 45 can be identified with the space GL'(p. f*b) i s framed cobordant to (f-'(y). . Framed submanifolds and a framed cobordism is a regular value for the mapping F . 1 ) = r. then f-'(y) i s framed cobordant to g-l(y).' ( z ) is framed cobordant to g-'(y).. Given z with I[z . then f-'(z) is framed cobordant to f-'(y). Such a path gives rise to the required framing of the cobordism f-'(y) X [0. t) = PROOF. 1 . PROOF. e --f Define the homotopy F : M X [0. and 3) each r. 11 is a framed manifold and provides a framed cobordism between the = f-'(y).e' < 1 5 1. and hence is connected. we can choose > 0 so that the e-neighborhood of y contains only regular values. and so that 1) r t is the identity for 0 5 t < E'. If b and b' are two dij'erent positively oriented bases at y. Choose a homotopy F with Lemma 1.'(z) lies on the great circle from y t o z . For each t note that z is a regular value of the composition rt 0 \ : ill -+ S". 2) rt equals r 1 for 1 . Lemma 3. t ) = g(z) Choose a regular value z for F which is close enough to y so that f-'(z) F ( z ." Lemma 2.e < t I 1. It follows a fortiori that Figure 16. This framed manifolds f-'(z) and (rl o f ) .

) The reader who is familiar with geodesics will have no difficulty in checking that g:NXU. y ) ) = y . Consider the mapping g : N X R” -+ M .llui12/e2). .. say to xo. t. . t. . Constructing a m a p with given Pontryagin manifold U> = dx’.) Clearly 0 is a regular value.. First suppose that M is the euclidean space R”’”. does what is expected of it.+ u / ( l .. is diffeomorphic to the full euclidean space R” under the correspondence u . with x’ converging to x. and we have contradicted the statement ’ 0). . For otherwise there would exist pairs (z. o ) is nonsingular. 0) = x . OF we can choose rotations r t :SP-+S” so that ro is the identity and r l ( y ) hence f-’(z) is framed cobordant to = z. + t”U”(X). Then clearly xo = x. 1 f f Product Neighborhood Theorem. boundaryless. of (2. hence g maps some neighborhood .-+M is well defined and smooth. dlffeomorphically onto an open set. + t. Given any two regular values y and z for f. This completes the proof of Theorem A. More precisely let g(z.. Thus f is homotopic to rl 0 f . this proves the Product Neighborhood Theorem for the special case M = E’”. x + t1U1(X) + * * . The proof of Theorem C will be based on the following: Let N C M be a framed submanifold of codimension p with framing b. * . The remainder of the proof proceeds exactly as before.. We will prove that g is one-one on the entire neighborhood N X U . and n-’(O)is precisely N with its given framing. for e sufficiently small. Assume that N is compact and that aN = aM = @.u”(x)I I in M which starts a t x with the initial velocity vector tlul(x) + . and with U -+ 0 and U --+ 0. Product neighborhoods do not exist for arbitrary subnianifolds. An unJramable submanifold T : V-+R” PROOF. providing that e > 0 is sufficiently small. ) Since g(x. .. t”) = by s ( g ( x . (See Figure 18. Choose a product representation g:NXIZ”-+VCM for a neighborhood V of N . U’) in N X R” with llull and I/u’I( U arbitrarily small and with !l & N Figure 18.) be the endpoirit of the geodesic segment of length I ltld ( x ) * . we could choose a sequence of such pairs with x converging.46 $7. Framed cobordism The Pontryagin construction 47 PROOF THEOREM A. = Since N is compact. Thus g maps N X U . of N X 0.. U’). t. REMARK.’(z) f-’(y). 0 ) E N x R” and so that each normal frame ~ ( x corresponds to the standard basis for R”. But U . where U . (Compare Figure 17. . .. PROOF OF THEOREM C. Some neighborhood of N in M i s diffeomorphic to the product N x R”. Furthermore the diffeomorphism can be chosen so that each x E N corresponds to (2.U”(X)/~ It#’(X) + + + * * + t g V ( x )I 1. and since dg. 0) E N X R” diffeomorphically onto an open set. Clearly d g ( z . For the general case it is necessary to replace straight lines in R”’” by geodesics in M . denotes the e-neighborhood of 0 in R”. as above. o . t. . . framed submanifold. and define the projection Figure 17.. that g is one-one in a neighborhood of (xo. ) # (x’. Let N C AI be a compact. defined by g ( x . (rl 0 f)-’(z) = f-’r.

PROOF. U) Lemma 4.u).h(g(x))] for x E M . . 0 F ( x .u) -~ l I l l ~ l l ~ for l c1 .with a similar inequality for G. deforms F = Fo into a mapping F .. U ) belong to the same open half-space in R”. &(x)) for x E V for x # V . Framed cobordisrn The Pontryagin construction 49 Now choose a smooth map cp : R” 3 x” which maps every x with 11x1 2 1into a base point so. So the honiotopy (1 . --f for z E N and 0 < llul[ < c. p(z) = h-l(z/A(\\z1\2)). and maps the open unit ball in R pdiffeomorphically* onto 8” .t)F(z.y -+ R” be stereographic projection. f*b) i s equal to (g-’(y). Choose a product representation for 0 < \lull < Min(c..u).. G : N X R” with --f R”. Making a corresponding deformation of the original mapping f . We will first find a constant c so that n-’(O> F(z. Let h : S” . U ) = [l . Define I g ( V ) do not contain the antipode fj of y.u > 0. Let f. l). Thus is suffices to deform f so that it coincides with g in some small neighborhood of N . = f(x>= s o dG(.u 2 . Z N X R” + V C M for a neighborhood V of N . g : M be smooth maps with a common regular value y.where h is the stereographic projection from SO and where X is a smooth monotone decreasing function with X ( t ) > 0 for t < 1 and X ( t ) = 0 for 1 2 1. for all 2 E N . 1 Hence I(F(. That is. 4 . f:M-+S” by f(z) = F-‘(O) = G-’(0) = N X 0. G(z.uI I I/U113 llU/j2 and w. a t least for I Iu[ < c. Since the corresponding Pontryagin manifold f-l(P(0)) = for all 5 E N . U ) and G ( x .so.@ $?. this clearly completes the proof of Lemma 4.U) + tG(x. In order to prove Theorem B we must first show that the Pontryagin S” manifold of a map determines its homotopy class. = (projection to R”) Clearly f is smooth.N + proves that f is smoothly homotopic to g.u). the points F(z. The hypothesis that f*b = g*b means that df.t). 0 for Now the honiotopy F t ( z .. c1 llull I.o.. g*b).5 with R”. (2) coincides with F for ( / U \ ( 2 c. then f i s smoothly homotopic to g.X ( U ) ~ ] F ( U ). U) * For example. that (1) coincides with G in the region \ ( U ( ( < c/2. It will be convenient to set N = f-l(y). and with dF. Identifying V with N X R” and identifying S” . this completes the proof of Theorem C. I By Taylor’s theorem IIF(z.. To avoid moving distant points we select a smooth map X : R p -+ R with X(U) X(U) = = 1 for lul 5 c/2 ll ljull 2 c. being careful not to map any new points into y during the deformation.’.u >0 is precisely equal to the framed manifold N .u). I f the framed manifold (f-’(y).c1 [lull3> 0 = f(4 for x e V = h-’[t*h(f(x)) (1 . we obtain corresponding mappings F . where V is small enough so that f ( V )and + X(U)~G(X. = dg. First suppose that f actually coincides with g throughout an entire neighborhood V of N .weeiiF and G will not map any new points into 0. and the point cp(0) is a regular value of f . and (3) has no new zeros. t) F(z. Then the homotopy bet.

For example if m = p 1 2 4. given a framed cobordism ( X . then two maps M are homotopic i and only i they have the same mod 2 degree. suppose that M is not orientable. r ~ ) between f-'(y) and 9-l (y). Conversely. t ) . The theory of framed cobordisni was iiitroduced by Pontryagiii in order to study homotopy classes of mappings S" + x" am --$ (S". (Compare $8. An easy argument then proves the following: whose Pontryagin manifold (F-'(y). It has since turned out to be easier to eriunierate homotopy classes of mappings by quite different. more algebraic methods. f by Lemma 4. The essential idea is to consider only mappings which carry the boundary into a base point so. - - - Theorem. (References [36]. M ) corresponds d d to the union operation for disjoint framed submanifolds of Interior (M). Hence F . F 4 ) is precisely equal to ( X .-T. however. On the other hand. Hu. then this set of homotopy classes can be given the structure of an abelian group.so) are in one-one correspondence with the cobordisni classes of franicd submanifolds N C Interior(M) of codimension p . Framed cobordism T h e Hopf theorem 61 PROOF F THEOREM B. f f -+ S'" 4 .and similarly F1 g. m).50 $7. Ren6 Thorn's work on cobordism is an important example of this. Let sgn(z) equal +1 or . With considerably more difficulty he was able to show that there are just two homotopy classes also in the case nz = p 2 2 4. Theory. However. and boundaryless. and C can easily be generalized so as to apply t o a manifold M with boundary. using framed 2-nianifolds. a double-edged tool. ( z ) = F(z. an argument completely analogous to the proof of Theorem C constructs a homotopy F : M X [0. Theorems A.p > 2 this approach to the problem runs into manifold difficulties. The composition operation in ?rp(M. REMARKS. Pontryagin proved this result by classifying framed 1-manifolds in S"'. Problem 17. Then given a basis for T M . let M be a connected and oriented manifold of dimension m = p. then O Lemma 3 asserts that the Pontryagin manifolds f f ' ( y ) and g-l(y) are framed cobordant.1]-+x" Theorem of Hopf. If M i s connected. which completes the proof of Theorem B. it conversely enables us to translate any information about homotopy into manifold theory. called the p-th cohomotopy group ?rP(Ml M ) . for m . Therefore f g . there are precisely two hoiiiotopy classes of mappings S"' + S".) + + THE HOPF THEOREM * See for example S. It not only allows us to translate information about manifolds into homotopy theory. and f have exactly the same Pontryagin manifold. But it is not difficult to see that the framed cobordisni class of the 0-manifold is uniquely determined by this integer sgn(z). Thus we have proved the following. If f and g are smoothly homotopic. If M i s cmnected but nonorientable. . Homotopy As an example. A framed submanifold of codimension p is just a finite set of points with a preferred basis at each. we can slide x around M in a closed loop so as to transform the given basis into one of opposite orientation. Then sgn(x) is clearly equal to the degree of the associated map M -+ S"'.) + with m > p. If p 2 am 1. Some of the deepest work in modern topology has come from the interplay of these two theories. Setting F .* Pontryagin's construction is. [all. B. note that the maps F . The homotopy classes of such mappings (AI.1 according as the preferred basis determines the right or wrong orientation. then two maps M -+ S"' are smoothly homotopic i and only i they have f f the same degree. oriented.

PROBLEM 2.) is a smooth manifold. show that every map fiI" --$ S" is homotopic S" with degree PROBLEM 12. show that every continuous map X -+ S" can be uniformly approximated by a smooth map. Given M TAI = C Rk. PROBLEM 8. d(identity) = identity. Show that I is a smooth ' If M bounds a n oriented manifold X disjoint from N . the homotopy being smooth if f and g are smooth.Exercises manifold and that the tangent space 53 T r ( z .U ) E AI X Rk 1 v E Tillz) -+ is also a smooth manifold. If M and N are compact. vc T K x T N . X TN. .. -+ M by r ( x v ) = 2.~~)rnIXRk~~~TTnI. show that they are smoothly homotopic.If X is compact. then oriented. Define r : N . 1 PROBLEM 9. . Show that any smooth map f : M gives rise to a smooth map df : TM -+ N TN where IN CONCLUSION here are some problems for the reader. 88. EXERCISES PROBLEM 10. PROBLEM 3. show that the correspondence (x.y)/llx . + + + PROBLEM 13. the linking map PROBLEM 6. with total dimension m the degree of X is called the linking number l ( M . X : M X N . Show that the degree of a composition g 0 f is equal to the product (degree g) (degree f). Define the linking number for disjoint manifolds in the sphere X"+"". Show that any map s" -+ S" of odd degree must carry some pair of antipodal points into a pair of antipodal points. Prove that + l(N. y) E M X N with f(z) = y.+ Sk PROBLEM 7. y) = (x . prove that Z(M. Show that any map S" different from (. prove that f is homotopic to g. . n = k.yII. . Given smooth manifolds M C Rk and N C R'. (Brouwer). Show that r(z v) is closer to x U than any other point of M . and boundaryless. PROBLEM 5. PROBLEM 11. show that the tangent space T ( M x N ) c . The graph r of a smooth map f : M -+ N is defined to be the set of all (5. Show that every complex polynomial of degree n gives rise to a smooth map from the Gauss sphere S2 to itself of degree n. Using this retraction r. N ) . If m to a constant. . N ) = 0. N -+ c R'".M ) = (-l)(m+l)(n+l) w NI. If M is compact and boundaryless.l)"+' must have a fixed point.is equal to T M .. is defined by X(z. is equal to the graph of the linear map df. hood Theorem in $7. d(g 0 f ) = (dg) o ( d f ) . Similarly show that the normal bundle space E = ((~. of M in Rk.) + < p .g(x)11 < 2 for all x. PROBLEM 1. If two maps f and g from X to x"satisfy I If(x) . show that the tangent bundle space {(x. PROBLEM 4. If two smooth maps X -+ x" are continuously homotopic. prove the analogue of Problem 4 in which the sphere S" is replaced by a manifold M . Given disjoint manifolds M .4 H z from E to Rkmaps the E-neighborhood of M X 0 in E diffeomorphically (Compare the Product Neighboronto the eneighborhood N .

cl + ix2)/(x3 + i d ) i where h denotes stereographic projection to the complex plane. for each s E I . 11. This integer H(f) = l(f-’(y). the subspaces T N . prove that H ( g 0 f) is equal to H ( f ) multiplied by the square of the degree of g.41 = f-’W G’(Y). DEFINITION.) + nP(M) into * For example. 2.. t Thus I can have boundary points only if M has boundary points. A brief discussion of the classification problem for higher dimensional manifolds will also be given. and if the “velocity vector” d f . which has been assumed in the text. ( l ) E TM. f . prove that H(f) a composition S2P-l = 0. and T N J together generate TM. (Compare p. connected 1-dimensional manifold i s difeomorphic either to the circle S‘ or to some interval of real numbers. PROBLEM 16. and does not depend on the choice of y and z. (If n n’ < in this means that N f l N’ = @. prove that it can be deformed slightly f so as to intersect a given N‘ transversally. It may be finite or infinite. or half-open. U ’ --f If p 2 inz 1. use a diffeomorphism of the form f(t) = a tanh ( t ) + b. Lemma. The Hopf Jibration a : S3--+ S2is defined by a ( z l . If i t has two components.’ ( z ) ) depends only on the honlotopy class of f .2 2 . Then f(I) A g ( J ) has at most two components.) = h-’((. (Reference [15].” Let I be an interval. If the dimension p is odd. Ifit has only one component..) I N is a framed submanifold. closed.. 11. it follows that there are only four distinct connected 1-manifolds. Prove that H ( a ) = 1. 50. has unit length. Any given local parametrization I’ -+ M can be transformed into a parametrization by arc-length by a straightforward change of variables. Two submanifolds N and N‘ of A f are said to intersect transversally if. or (0.3.) Since any interval is diffeomorphic* either to [0.’ ( z ) can be oriented as in $5. use the disjoint union operation to make an abelian group. then the manifolds f-’(y). prove that K f 1 ( Y ) . If y # z are regular values for a 14. then f can be extended to a parametrization by arc-length of the union f(I) g ( J ) . . 1).) PROBLEM 15. f-’(z)) is defined. A n y smooth. open. + PROBLEM 17. I map f : S”-’ 4 S”. Let f : I M and g : J --$ M be parametrizations by arc-length. then M must be diffeomorphic to X .54 $8. 1 8”4 8” . a) Prove that this linking number is locally constant as a function of y. f . hence the linking number l(f-’(y)..9(”)1/ < for all x. Prove that the resulting intersection is a smooth manifold. f . (An interval is a connected subset of R which is not a point. for each x E N f l N’. where f APPENDIX CLASSIFYING 1-MAN1F O LD S Ilf(4 . b) I y and z are regular values of g also. Exercises PROBLEM THE HOPF NVARIANT. Use the transverse intersection operation to define a correspondence rIP(M)x rI*(M)3 rI”+“(M). A map f : I 4 M is a parametrization by arc-length if f maps I difleomorphically onto an open subset t of M . Let IIp(M) denote the set of all framed cobordism classes of codimension p in M. For W E WILL prove the following result. (0. g-‘(4). c) Prove that l(f-’(y). Theorem. f-’(zN = Vg-l(Y).’ ( z ) ) is called the Hopf invariant of f. The proof will make use of the concept of “arc-length. IIY . 2.

then g-l o f extends to a linear map L : R f Now f and g o L piece together to yield the required extension F : I U L-'(J) + f ( I ) U g ( J ) . Since g-' o f is one-one and single valued. Furthermore the derivative of g-' o f is equal to i l everywhere. being compact and open in AI.) Furthermore. PROOF O F CLASSIFICATION THEOREM. -+ R. If M is not diffeoniorphic to A". sin 0) = = f(t) g(t) for a for c < t < d. For if the open set f ( I ) were not all of M . with slope say 1. < t < 8. there would be a limit point z of f ( I ) in M . If has two components. For 2-dimensional manifolds. there can be a t most one of these segments ending on each of the four edges of the rectangle I x J . t ) with f(s) = g(t).f ( I ) . . the two must have the same slope. we may assume that y = c and 6 = d. Translating the interval J = (y. Any parametrization by arclength can be extended to one f :I+M which is maximal in the sense that f cannot be extended over any larger interval as a parametrization by arc-length: it is only necessary to extend f as far as possible to the left and then as far as possible to the right. Parametrizing a neighborhood of x by arc-length and applying the lemma.* But for high dimensional simply connected manifolds there has been much progress in recent years.a ) . Clearly 9-l 0 f maps some relatively open subset of I diffeomorphically onto a relatively open subset of J . we will prove that f is onto. and hence is a diffeoniorphism. a thorough exposition has been given by KerBkjArtb [17]. Since r is closed and 9-l 0 f is locally a diffeoniorphism. consisting of all (s. 8 -I A U Figure 19. '1 The image h ( S ' ) . Consider the graph r C I x J . REMAHKS. these line segments cannot end in the interior of I x J . (See l'apakyrialtopoulos [as]. must be the entire manifold M . * See Markov [19]. Then J? is a closed subset of I X J made up of line segments of slope f1. the required difieomorphism h:S'-+M is defined by the formula h(cos 0. Hence has a t most two components. so that a + <b l c < d < a <p.) lcor coinpact nianifolds of dimension 2 4 the classification problem is actually unsolvable. and also a forthcoming paper by Boone.66 Appendix Classifying I-manifolds 57 PROOF. Now setting 8 = 2?rt/(a . (See Figure 19. Three of the possibilities for r i I r is connected. This proves the lemma. we would see that f can be extended over a larger interval. This contradicts the assumption that f is maximal and hence completes the proof. Haken. they must be arranged as in the left-hand rectangle of Figure 19. as exemplified by the work of Smale [31]and Wall [37]. p ) if necessary. The study of %dimensional manifolds is very much a topic of current research. For manifolds of higher dimension the classification problem becomes quite formidable. but must extend to the boundary. and Poknaru in Fundamenta Mathematicae. if there are two components.

Press..“On total curvatures of Riemannian manifolds. Mathematical Analysis. Dieudonnb. J. U.Wiylie. Jour. 32 (74). let me recommend Hilton and Wylie [ 1I]. For the reader who wishes to pursue the study of differential topology. Foundations of Modern Analysis..” Mat.S. Calculus of Several Variables. Brouwer. W. MacKenzic. New York: Harper & Row. Soc. C. Brown. Hilton. Soc. M.I nier. 443-464. “Tht.. : . 38 (1935). and R. 1057. L. Hcwling. . New York: Academic Press. 379-394.. 1960. Hoinology Theory. 15 (1940). and Sternberg [35]. let me recommend Milnor [22].) Chern. London Math. 1. Math.. Cambridge Univ.. Annalen 71 (1012). S. Ya. 45 (1944). 97-115. C. “Uber Abbildung von ?tlannigfaltigkeitcn”. ilrlath. 1965.5-22. and S. Introduction to DiJerentiable Jlanilolds. 62 (1!140). 1960.) Fenchel. 1963. I h l w numlw of a Ricmann m:inifold. T.BIBLIOGRAPHY THE following is a nliscellaneous list consisting of original sources and of reconimended textbooks. de Rham [29]. J.“Functional dependence. P. 243-248... (1953). illcndocrfcr.” . (In Russian. Lang [18].. A. and Pontryagin [as]. Jlath. Dubovickii.. L. 747-752.” Trans. “On differentiable mappings of an n-dimensional cube into a k-dimensional cube.4postol. Steenrod [34].” Jour. Sbornik N. Alunkres [25]. Mass. (See Theorem 3-111. 13. “A simple intrinsic proof of the Gauss-Ronnet formula for closed Riemannian manifolds.” Annals of ddath. New York: McGraw-Hill.itltlison-Wcslcy. Auslander. .. Hu [16]. A. E. Amer. S. Goffman. The survey articles [23] and [32] should also prove useful. For background knowledge in closely related fields.

[25] Munkres.S. Ser. [28] . (Translated from Trudy Inst. 1964. Amer. “Quelques propriBtBs globales des vanet& difErentiables. 1960. 1955. Jour. New York: Academic Press. “Abbildungsklassen n-dimensionaler Mannigfaltigkeiten. 1962.) Princeton Univ. . Bull. Press. [ 131 Hopf.” Amer. 1 T.” L’Enseignement math.. 69 (1963). [24] Morse. Press.. 147-149. Annalen 96 (1926).” Annals of Math. A..-T. C. pp. pp. C.” Fundamenta Mathematicae 25 (1935). Lectures on Diflerential Geometry. 1171 KerBkjdrt6. 163-189. L. 300-306. Introduction to Differentiable Manifolds.” Proc. [IS] Lang.“A survey of cobordism theory. “Differential topology.. The Topology of Fibre Bundles. [21] . 1 (1935).. Nauk.“A survey of some recent developments in differential topology.“Smooth manifolds and their applications in homotopy theory. 1951. 1963. A. 1955. [19] Markov. Press. Princeton Univ.. Soc. Math. 40 (1939). “On manifolds homeomorphic to the 7-sphere. J. “Generalized PoincarB’s conjecture in dimensions greater than four. Annalen 96 (1926). 427-440.. L.” . J. 433-440. . “The measure of the critical points of differentiable maps. “A proof of the nonretractibility of a cell onto its boundary. [34] Steenrod. Berlin: Springer. Z€eZvet.“Uber die Abbildungen von Spharen auf Spharen niedrigerer Dimension. (Annals Studies 51. Press. Math. “A classification of continuous transformations of a complex into a sphere. S. .” Annals of Math. H. Translations. S. New York: Benjamin. 1231 ----. Calculus on Manifolds. 1321 . 74 (1961). “Vektorfelder in n-dimensionalen Mannigfaltigkeiten.. 1-114. “Classification of (n . 883-890.” Duke Math. 14 (1963).) [2O] Milnor. 1958. [33] Spivak. C. “A function not constant on a connected set of critical points. Homotopy Theory. 1960.” Annals of Math. (In Russian. Varidte‘s diffkrentiables. Princeton Univ. 28 (1954).” Doklady Akad. 1 . T. Congress of Math. . 75 (1962). [37] Wall. SOC... pp. Cambridge Univ..” Proceedings Intern. A. Paris: Hermann. 64 (1956). New York: Wiley. Saaty. I1 (1959).R.S. M. v.) [29] Rham. 62-70. 48 (1942). 364-365. ‘‘Insolubility of the problem of homeomorphy. 209-224..60 Bibliography Bibliography 61 [12] Hirsch.” Commentarii Math. 1161 Hu.Morse Theory. S. [38] Whitney. G. 1923. (Comptes Rendues) 19 (1938). U. (Annals Studies 54). A. [26] Papakyriakopoulos. N. Amer. R. hlath. S.. [36] Thom.” dnnals of Math. “The theory of three-dimensional manifolds since 1950. Congress of Math.. Steklov 45 (1955). 1221 . Math. 1311 Smale.” Proceedings Intern. ~ 5 1. 165-183.. 17-86. M. H..1)-connected 2n-manifolds.” Math. Amer. [30] Sard. “The behavior of a function on its critical set.. Elementary Differential Topology. [27] Pontryagin. de. ~ 4 1__ . S. 399-405. Soc. D. [35] Sternberg. 1959. etl. New York: Interscience. ... 2. New York: I’renticeHall. 16-23. R. S. 1964. Soc. 1963. 225-250. 391-406. Cambridge Univ. 514-517. 131-145.” illath. 8 (1962).” Lectures on Modern Mathematics. Vorlesungen uber Topologie.” Bull. I’. 1965. Press.

31. Milnor in December. 2-7 diffeomorphism. antipodal niap. 16 smooth manifolds. 10. 21. 1. 11. 11 degree of a map. A. 14 Brown.INDEX PAGE-BARBOUR LECTURE SERIES The Page-Barbour Lccture Foundation was foundcd in 1007 by a gift from Mrs. 55 Poincark. 1 corresponding vector fields. 8.. J. 38 Gauss mapping. 17 fundamental theorem of algebra. 35 Pontryagin.. 36. 12 Hirsch. 7 regular value. 53 Morse. 54 index (of a zcro of a vector field).13. 2s outward vector. 35.40. L. 20 derivative of a map. E. I>. 20. as the forty-seventh series sponsored by the Foundation... 51 (dimension. 5. 42 orientation. 42.27. M. 1. 46 Fubini theorem. 13. M. H. 22. 10. 44. 59 dimension of a manifold.. 11.. 40. 42 Pontryagin manifold. 30. 53 normal vectors.20. 52 boundary. 34 linking. A.. 51. 36. 20. 14 Euler number. 12 . 42. 35. 1963. 8 inward vector. 3. 7 disk. 1 differential topology.14. 1.. 16. 43. 3234 index sum. A. 35. 8. 26 isotopy. 11 critical value. 11 chain rule for derivatives. 1 with boundary. 2 by arc-length. 38 half-space. 26 parametrization.43 Sard. 35-41 inverse function theorem. 12. Thomas Nelson Page (n6e Barbour) and the Honorable Thomas Nelson Page for the purpose of bringing to the University of Virginia each session a series of lectures by an eminent person in some field of scholarly endeavor. 32 critical point. 8 Gauss-Bonnet theorem. 40 framed cobordism. 43 product neighborhood theorem. 7 cobordism. 50 coordinate system. L. 26 of a boundary. 52 Hopf. The materials in this volume were presented by Professor John W. 40 nondegencrate zcro (of a vector field) 37. 4. 28 mod two. 10 Morse. 35. B. 40 normal bundle. 12 Brouwer. 14 homotopy. 43 framed submanifold. H. 46 regular point. 27 of F-'(y).. 28 fixed point theorem. 42 cohomotopy. 8. 52 degree. 21.

1 sphere. 12. 2 vector fields. 27 df.. !*a. 53 tangent space. 2 lnrrcz stereographic projection.) the classification problem.. 2-5 a t a boundary point. 32-41 Weierstrass approximation theorem. Inc. 1 ax. The types are Modern Number 8 and Century Schoolbook.. Y). 26 smooth maps (= smooth mappings). 9. Virginia and bound by Russell Rutter Co. Inc. 26 tangent vector. 30. 57 of dimension zero.. 2 TM. Richmond.64 smooth manifolds (cont. #P(Y)l 14 12 43 8 12 1 *-1. 2-7 Dm. 13 -~ INDEX O F SYMBOLS deg (f. Hm. Y. New York. 14 TOPOLOGY FROM T H E DIFFERENTIABLE VIEWPOINT was composed and printed by The William Byrd Press. 14. 2 of dimension one. N. 48 tangent bundle. 2-5 IIXII. Design is by Edward Foss. s! o.. . and the paper is Oxford's Bookbuildcrs Plate. Rk. 55 oriented.