LECTURE 2: SMOOTH MANIFOLDS

1. Review of analysis
Let U be an open set in Rn , and f : U → R a continuous function. Recall that f
is said to be a C k -function, if all its partial derivatives of order at most k,
∂ |α| f ∂ |α| f
∂ α f := := , |α| = α1 + · · · + αn ≤ k
∂xα (∂x1 )α1 · · · (∂xn )αn
exist and are continuous on U . We say that f is a C ∞ function, or a smooth function,
if it is of class C k for all positive integers k. A function f is an analytic function (or
C ω ) if it is smooth and agrees with its Taylor series in a neighborhood of every point.
Note that not all smooth functions are analytic. A standard example is

0, x≤0
f (x) = − x1
e , x>0
is a C ∞ function defined on R but is not analytic at x = 0. (Check this!)
Now let U be an open set in Rn and V be an open set in Rm . Let
f = (f1 , · · · , fm ) : U → V
be a continuous map. We say f is C ∞ (or C k , or C ω ) if each component fi , 1 ≤ i ≤ m,
is a C ∞ (or C k , or C ω ) function. 1
Let f be a smooth map. The differential of f , df , assigns to each point x ∈ U a
linear map dfx : Rn → Rm whose matrix (with respect to the canonical basis) is the
Jacobian matrix of f at x,
 ∂f ∂f1

∂x
1
1 (x) · · · ∂x n (x)
dfx =  ... ... ..  .

. 
∂fm ∂fm
∂x1
(x) · · · ∂xn (x)
Note that for any v ∈ Rn , the directional derivative of f at x in the direction v is
f (x + hv) − f (x)
dfx (v) = lim .
h→0 h
The well-known chain rule asserts that if f : U → V and g : V → W are smooth
maps, so is the composition map g ◦ f : U → W , and
d(g ◦ f )x = dgf (x) ◦ dfx .
Definition 1.1. A smooth map f : U → V is a diffeomorphism if f is one-to-one and
onto, and f −1 : V → U is also smooth.
1In this course we will mainly consider C ∞ functions/maps. However, most definitions/theorems
can be easily extended to the C k setting.

1
2 LECTURE 2: SMOOTH MANIFOLDS

Obviously
• If f : U → V is a diffeomorphism, so is f −1 .
• If f : U → V and g : V → W are diffeomorphisms, so is g ◦ f .
As a consequence, we can prove the following smooth “invariance of domain” theorem:
Theorem 1.2. If f : U → V is a diffeomorphism, then for each x ∈ U , the differential
dfx is a linear isomorphism. In particular, dim U = dim V .
Proof. Applying the chain rule to f −1 ◦ f = idU , and notice that the differential of the
identity map idU : U → U is the identity transformation Id : Rn → Rn , we get
dff−1
(x) ◦ dfx = IdR .
n

The same argument applies to f ◦ f −1 , which yields

dfx ◦ dff−1
(x) = IdR .
m

By basic linear algebra, we conclude that m = n and that dfx is an isomorphism. 

The inverse of the previous theorem is not true. For example, we consider the map
f : R2 \ {0} → R2 \ {0}, (x1 , x2 ) 7→ ((x1 )2 − (x2 )2 , 2x1 x2 ).
Then at each point x ∈ R2 \ {0}, dfx is an isomorphism. However, f is not invertible
since f (x) = f (−x). (What is the map f if we identify R2 with C?)
The inverse function theorem is a partial inverse of the previous theorem, which
claims that an isomorphism in the linear category implies a local diffeomorphism in
the smooth category.
Theorem 1.3 (The inverse function theorem). If f : U → V is a smooth map, and
dfx is an isomorphism, then f is a local diffeomorphism near x, i.e. there exists a
neighborhood Ux containing x and a neighborhood Vf (x) containing f (x) such that
f |Ux : Ux → Vf (x)
is a diffeomorphism.
The theorem is a special case of the following
Theorem 1.4 (The implicit function theorem). Let W be an open set in Rnx × Rm y ,
m
and F = (F1 , · · · , Fm ) : W → R a smooth map. Let (x0 , y0 ) be a point in W so that
the m × m matrix  ∂F1 ∂F1

∂y 1 (x 0 , y0 ) · · · ∂y m (x 0 , y0 )
.. .. ..
.
 
 . . .
∂Fm ∂Fm
∂y m
(x0 , y0 ) ··· ∂y m
(x0 , y0 )
is nonsingular, then there exists a neighborhood U0 × V0 of (x0 , y0 ) in W and a smooth
map f : U0 → V0 so that
• f (x0 ) = y0 ,
• If we denote c = F (x0 , y0 ), then F −1 (c) ∩ (U0 × V0 ) is the graph of f , i.e.
F (x, f (x)) = c for all x ∈ U0 .
 LECTURE 2: SMOOTH MANIFOLDS 3

2. Smooth Manifolds
We would like to define smooth structures on topological manifolds so that one
can do calculus on it. In particular, we should be able to talk about smoothness of
continuous functions on a given smooth manifold M . Since near each point in M , one
has a chart {ϕ, U, V } which identify the open set U in M with the open set V in Rn ,
it is natural identify any function f on U with the function f ◦ ϕ−1 on V , and use
the smoothness of f ◦ ϕ−1 to define the smoothness of f itself. This idea is of course
correct. The only issue is that a point on M could sit in many different open charts,
and the smoothness of a function at this point should be independent of the choice of
chart. In other words, if both ϕ and ψ are chart maps near a point, we want the maps
f ◦ ϕ−1 and f ◦ ψ −1 to be simultaneously smooth or non-smooth. This amounts to
require ϕ ◦ ψ −1 to be smooth. With this as our intuition, we define
Definition 2.1. Let M be a topological manifold. We say two charts {ϕα , Uα , Vα } and
{ϕβ , Uβ , Vβ } of M are compatible if the transition map
ϕαβ = ϕβ ◦ ϕ−1
α : ϕα (Uα ∩ Uβ ) → ϕβ (Uα ∩ Uβ )
is a diffeomorphism.

Note that both ϕα (Uα ∩ Uβ ) and ϕβ (Uα ∩ Uβ ) are open in Rn , so the smoothness
of ϕαβ is well understood.
Example. (Spheres - continued). Recall that the n-sphere S n admits two atlas
{ϕ+ , U+ , Rn } and {ϕ− , U− , Rn }, where
U± = S n \ {(0, · · · , 0, ∓1)}
and
1
ϕ± (x1 , · · · , xn , xn+1 ) = n+1
(x1 , · · · , xn ).
1±x
It follows that on ϕ− (U+ ∩ U− ) = Rn \ {0},
ϕ−+ (y 1 , · · · , y n ) = ϕ+ ◦ ϕ−1 1 n
− (y , · · · , y )
 
1 1 n 1 2 n 2


= ϕ+ 2y , · · · , 2y , −1 + (y ) + · · · + (y )
1 + (y 1 )2 + · · · + (y n )2
1
= 1 2 (y 1 , · · · , y n ),
(y ) + · · · + (y n )2
which is a diffeomorphism from Rn \ {0} to itself. So these two charts are compatible.
4 LECTURE 2: SMOOTH MANIFOLDS

Example. (Projective Spaces - continued). Recall that the n dimensional real

projective space RPn has n + 1 charts {ϕi , Ui , Rn }, 1 ≤ i ≤ n + 1, where
Ui = {[x1 : · · · : xn+1 ] | xi 6= 0}
and
x1 xi−1 xi+1 xn+1
 
1 n+1
ϕi ([x : · · · : x ]) = ,··· , i , i ,··· , i .
xi x x x
Without loss of generality, let’s verify that ϕ1,n+1 is a diffeomorphism between
ϕ1 (U1 ∩ Un+1 ) = {(y 1 , · · · , y n ) | y n 6= 0} =: Vn
and
ϕn+1 (U1 ∩ Un+1 ) = {(y 1 , · · · , y n ) | y 1 6= 0} =: V1 .
In fact, by definition
ϕ1,n+1 (y 1 , · · · , y n ) = ϕn+1 ◦ ϕ−1 1 n
1 (y , · · · , y )

= ϕn+1 ([1 : y 1 : · · · : y n ])
1 y1 y n−1
 
= , ,··· , n
yn yn y
which is obviously a diffeomorphism from Vn to V1 . Similarly one can show that all
transition maps ϕij are diffeomorphisms.
Definition 2.2. (1) An atlas A on M is a collection of charts
S {ϕα , Uα , Vα } such that
all charts in A are compatible to each other, and satisfies α Uα = M .
(2) Two atlas on M are said to be equivalent if their union is still an atlas on M .
Example. We can define two atlas on R by A = {ϕ1 , R, R} and B = {ϕ2 , R, R}, where
ϕ1 (x) = x and ϕ2 (x) = x3 . Then they are non-equivalent atlas since
ϕ21 (x) = ϕ1 ◦ ϕ−1
2 (x) = x
1/3

is not smooth.
Definition 2.3. An n-dimensional smooth manifold is an n-dimensional topological
manifold M equipped with an equivalence class of atlas. This equivalence class is called
its smooth structure.

So a smooth manifold is a pair (M, A). In the future we will always omit A if there
is no confusion of the smooth structure.
Remark. Similarly one can define C k manifolds, real analytic (=C ω ) manifolds and
complex manifolds. For example, a complex manifold is a Hausdorff and second count-
able topological space that locally looks like Cn , so that the transition maps are all
bi-holomorphic.
Example. If a topological manifold M can be covered by a single chart, then such a
chart automatically determines a smooth structure on M . So in particular, the graph
of any continuous function on an open domain of Rn is a smooth manifold. (However,
it is possible that such a graph is not a smooth submanifold of Rn+1 .)
 LECTURE 2: SMOOTH MANIFOLDS 5

Example. All the examples we studied in lecture 1, including graphs, S n , RPn etc, with
the charts we described, are smooth manifolds. In fact, they are also C ω manifolds.
Remark. Some deep results from differential topology:
• There exists topological manifolds that do not admits smooth structure. The
first example was a compact 10-dimensional manifold found by M. Kervaire.
• If a topological manifold admits a C 1 structure, it also admits a C ∞ structure.
• Any manifold M admits a finite atlas consisting of dim M + 1 charts (not
necessarily connected).