Lectures On Levi Convexity

Of Complex Manifolds And

Cohomology Vanishing Theorems
By
E. Vesentini
Tata Institute Of Fundamental Research, Bombay
1967
Lectures On Levi Convexity
Of Complex Manifolds And
Cohomology Vanishing Theorems
By
E. Vesentini
Notes by
M.S. Raghunathan
No part of this book may be reproduced in any
form by print, microlm or any other means with-
out written permission from the Tata Institute of
Fundamental Research, Colaba, Bombay 5
Tata Institute of Fundamental Research,
Bombay
1967
Preface
These are notes of lectures which I gave at the Tata Institute of Funda-
mental Research in the Winter 1965.
Most of the material of these notes may be found in the papers [1],
[2] and [16] listed in the Bibliography at the end of this book.
My thanks are due to M.S. Raghunathan for the preparation of these
notes and to R. Narasimhan and M.S. Narasimhan for many helpful sug-
gestions.
E. Vesentini
iii
Contents
Preface iii
0 Prerequisites 1
1 Vanishing theorems for hermitian manifolds 7
1 The space [
p,q
. . . . . . . . . . . . . . . . . . . . . . 7
2 The space W
p,q
. . . . . . . . . . . . . . . . . . . . . . 15
3 W-ellipticity and a weak vanishing theorem . . . . . . . 25
4 Carleman inequalities . . . . . . . . . . . . . . . . . . . 30
2 W-ellipticity on Riemannian manifolds 39
5 W-ellipticity on Riemannian manifolds . . . . . . . . . . 39
6 A maximum principle . . . . . . . . . . . . . . . . . . . 44
7 Finite dimensionality of spaces of harmonic forms . . . . 50
8 Orthogonal decomposition in [
q
. . . . . . . . . . . . . 57
3 Local expressions for and the main inequality 61
9 Metrics and connections . . . . . . . . . . . . . . . . . 61
10 Local expressions for , and . . . . . . . . . . . . . 70
11 The main inequality . . . . . . . . . . . . . . . . . . . . 75
4 Vanishing Theorems 79
12 q-complete manifolds . . . . . . . . . . . . . . . . . . . 79
13 Holomorphic bundles over q-complete manifolds . . . . 85
14 Examples of q-complete manifolds . . . . . . . . . . . . 92
15 A theorem on the supports of analytic functionals . . . . 95
v
Chapter 0
Prerequisites
In this chapter we collect together some well known results which will 1
be used in the course of these lectures.
1. Let X be a connected, paracompact complex manifold of (com-
plex) dimension we denote by The sheaf of germs of homo-
morphic p-foxes end by
x,q
the sheaf of germs of 0

forms of
type (P, /). Further overset will denote the exterior dieren-
tiation with respect to

Then we have one.
Proposition. The sequence
0
P
A
p,0

A
p,1
. . .

A
p,n
0
is exact. Similarly if K
p,q
denotes the sheaf of germs of (p, q)
currents, the sequence
0
p
K
p,c

K
p,1
. . .

K
p,n
0
is exact.
We point out that the exactness of the rst step of the second se-
quence:
0
p
K
po
1
2 0. Prerequisites
is equivalent to the fact that every

closed distribution is a holo-
morphic function
1
(For a proof see for instance [5], Expos e XVIII, [22]) 2
The sheaves A
p,q
being ne, the cohomology group H
q
(X,
p
) is
canonically isomorphic to the qth cohomology group of either one
of the complexes
0
//
H

(X, A
p,o
)
//

//
H

(X, A
p,1
)

// . . .

//
H

(X, A
p,n
)
//
0
0
//
H

(X, K
p,o
)
//

//
H

(X, K
p,1
)

// . . .

//
H

(X, K
p,n
)
//
0.
2. We need the following theorem due to Leray.
Let G = U
i

iI
be a locally nite open covering of a paracompact
space X. Let / be a sheaf of abelian groups on X such that for
every q > 0, H
q
(U
i
o
. . . U
i
p
, /) = 0 for every set (i
0
, . . . i
p
) of
p-elements in I. Then the canonical map
H
q
(G, /) H
q
(x, /)
is an isomorphic for all q o
For a proof see for instance [13] 209-210
3. The following result enables us to apply the above theorem to
locally free sheaves over homomorphic function on a complex
manifold.
Let X be a (para compact) complex manifold of complex dimen-
sion n. Given a vector bundle E an X and any covering G =
U
i

iI
there is a renement V
j

jJ
such that for j
1
, . . . , j
k
with
V
j1
. . . V
jk
, for 0 p n, the sequence
1
The idea of the proof is the following. Let be any compactly supported function
of class C
k
with k > 0. Since

(T ) = 0, T is a homomorphic function hence it is
C

. Then T itself is a C

function [30, Vol. II, Theorem XXI, 50]. But

T = 0. Then
T is holomorphic.
3
0 H
0
(V
j1,... jk
,
p

O
E

H
0
(V
j1... jk
A

p1

O
E

)

1
. . .
. . .

O1
H
0
(V
j1... jk
, A
pn

O
E

) 0
where O is the structure sheaf, is exact, and V
j1... jk
is the intersec- 3
tion V
j1
. . . V
jk
and E is the locally free O sheaf associated to
E. (Since the sheaves, A
pi

O
E are ne sheaves, this implies that
H
p
(V
ji
. . .
jk

p

O
E) = 0
4. Let : E X be a holomorphic vector bundle on the complex
manifold X. Let G = (U
i
)
iI
be a covering of X and e
i j
: U
i
U
j

G = GL(n, C) be transition functions with respect to G

such that
G, e
i j
dene E. This means that we are given isomorphisms

i
: E/U
i
U
i
C
n
,
for i I such that e
i j
are the maps dened by
(
j

1
i
)(x, v) = (x, e
i j
(x)(v))
for x U
i
U
j
. In particular, we have for i, j, k I
e
ik
(x) = e
i j
(x)e
jk
(x)
provided that x U
i
U
j
U
k
.
For a vector bundle : E X, we denote, as before by E or (E)
the sheaf of germs of holomorphic section of E. (E) is a locally
free sheaf over the sheaf of germs of holomorphic functions on X
(we denote this latter sheaf by O). Then, in the above notation,
(E)

U
i
O
n
We x moreover the following notation:

p
(E) = E
O

p
, in particular

(E) = (E);
A
pq
(E) = E
O
A
pq
;
K
pq
(E) = E
O
K
pq
.
4 0. Prerequisites

p
(E) (resp. A
pq
(E), resp. K
pq
(E)) is the sheaf of germs of holo- 4
morphic E-valued p-forms (resp. dierentiable (p, q) forms with
values in E, resp. (p, q) currents with values in E ;
(X, A
pq
(E)) = Global(p, q)C

forms with values in E;

(X, K
pq
(E)) = Global(P, q) currents with values in E.
There is a one-one correspondence between global sections of
A
pq
(E) and collections (
i
)
iI
where each
i
is a vector valued
form,

i
=
_

i
.
.
.

m
i
_

_
each
k
i
being a scalar C

(p, q) forms on U
i
such that for x U
i

U
j
we have
i
(x) = e
i j
(x)
j
(x). One can set up a similar one-one
correspondence between sections of K
pq
(E) and families (
i
)
iI
of currents, each
k
i

i
being dened on U
i
, satisfying e
i j

j
=
i
on U
i
U
j
.
We denote by

the operators
1

:E
O
A
pq
E A
p,q+1
and 1

:E
O
K
pq
E K
p,q+1
.
Then the sequences
0
p
(E) A
p,o
(E)

. . .

A
p,n
(E) 0
and 0
p
(E) K
p,o
(E)

. . .

K
p,n
(E) 0
are exact. Moreover the sheaves A
pq
(E) and K
pq
(E) are ne.
Hence if denotes either the family of all closed subsets of X 5
or the family of all compact subsets of X, (more generally, any
paracompactifying family) and we set,

(X, A
pq
(E)) =
_
a section of A
pq
(E) over X, Supp
_
,
then H
q

(X,
p
(E))
_

(X, A
p,q
(E));

= 0
_

(X, A
p,q1
(E))
_
5
and similarly with the obvious notation,
H
q

(X,
p
(E))
_

(X, K
p,q
(E));

= 0
_

(X, K
p,q1
(E))
_
Chapter 1
Vanishing theorems for
hermitian manifolds
1 The space [
p,q
All manifolds considered are assumed to be connected and paracompact. 6
Let : E X be a holomorphic vector bundle on a paracompact
complex manifold X of complex dimension n.
Let G =
_
U
i
_
iI
be a locally nite open coordinate covering of X and
_
e
i j
: U
i
U
j
GL(mC)
_
i jII
transition functions dening E.
A hermitian metric along the bres of E is a collection
_
h
i
: U
i

GL(m, C)
_
iI
of C

-maps such that for xU

i
, h
i
(x) is a positive denite
hermitian matrix and for xU
i
U
j
, h
j
(x) =
t
e
i j
(x)h
i
(x)e
i j
(x).
Lemma 1.1. Every holomorphic vector bundle on a complex manifold
admits a hermitian metric.
Proof. Let
_
k
_
kI
be a C

-partition of unity subordinate to

_
U
k
_
kI
(=
G). Let
_
h
o
i
: U
i
GL(m, C)
_
iI
be any family of C

-functions such
that for i I, x U
i
, h
o
i
(x) is a positive denite hermitian matrix. Then
the family
_
h
o
i
_
iI
dened by
h
i
(x) =

k
(x)
t
e
ki
(x)h
o
k
(x)e
ki
(x)
7
8 1. Vanishing theorems for hermitian manifolds
is a metric along the bres of E.
In particular, the holomorphic tangent bundle H X admits a 7
hermitian metric. Let G =
_
U
i
_
iI
be a covering of X by means of
coordinate open sets. Let Z
1
i
, . . . , Z
n
i
be any system of coordinates in U
i
.
Then H X is dened with respect to G by means of the transition
functions J
i j
: U
i
U
j
GL(n, C) dened by
J
i j
(x) =
_

_
Z

j
Z

l

(x)
_

for x U
i
U
j
.
A hermitian metric on H is then a family of C

-hermitian -positive
denite matrix-valued functions g
i
: U
i
GL(n, C) such that
g =
t
J
i j
g
i
J
i j
Identifying hermitian matrices with hermitian bilinear forms on the tan-
gent spaces through the basis

Z
1
. . .

Z
n
of the holomorphic tangent
space, we may write g
i
as

g
i

dZ

i
d

Z

.
A hermitian metric on X denes on X (regarded as a dierentiable
manifold), a Riemannian metric: in fact, if z

i
= x

i
+ iy

i
, then we have
ds
2
=

g
i
dZ

i
dI

i
=

Re g
i
(dx

i
dx

i
+ dy

i
dy

i
)
in the coordinate open set U
i
, in terms of the local real coordinates x
i
, y
i
.
Since X carries a complex structure, it has a canonical orientation de-
ned by this structure. Hence X has a canonical structure of an oriented
Riemannian manifold. We denote by the volume form on X with re-
spect to this oriented Riemannian structure.
Let C
r
(X), 0 r 2n (n= complex dimension of X) denote the 8
space of complex valued exterior dierential forms of degree r. It is
a module over the algebra, F, (over the complex numbers) of complex
valued C

function on X
1. The space [
p,q
9
Denition 1.1. The star operator : C
r
(X) C
mr
(X) on an ori-
ented Riemannian manifold X of (real) dimension m is dened by the
formula
()(t
1
, . . . , t
mr
). = (t
1
) . . . (t
mr)
where C
r
(X), t
1
, . . . , t
mr
are tangent vectors to X at a point P and
is the canonical isomorphism of the F-module of tangent vector elds on
X onto the F-module of 1-forms. Clearly * is a real operator, i.e. =
(the bar denoting conjugation of complex numbers). Furthermore we
have
Lemma 1.2. Let (X, g) be an oriented Riemannian manifold of dimen-
sion m. Then
= (1)
r(m1)
for C
r
(X). (1 I)
Proof. ([18]) Let x X and t
1
, . . . , t
m
be an orthonormal basis of the
tangent space at x with respect to the Riemannian metric. Then it is
enough to check that
((t
i
1
) . . . (t
ir
)) = (1)
r(m1)
(t
i
1
) . . . (t
i
r
)
for every 0 < i
1
< i
2
< . . . < i
r
m, or again that for every sequence
0 < j
1
< . . . < j
r
m,
()(t
i
1
) . . . (t
i
r
)(t
j
1
, . . . t
j
r
)
=
_
(t
i
1
) . . . (t
i
r
)
_
(t
j
1
, . . . t
j
r
) (1)
r(m1)
.

Suppose now that ( j

1
, ..., j
r
) (i
1
, ..., i
r
). then 9
()((t
i
1
) ..., (t
i
r
)(t
j
1
, ..., t
j
r
).l
=
_
(t
i
1
) ... (t
i
r
)
_
(t
j
1
) (t
j
r
)
and the right hand side is zero because,
_
(t
i
1
) (t
i
r
)
_
(t

1
, . . . , t

mr
)l
10 1. Vanishing theorems for hermitian manifolds
= (t
i
1
) ... (t
i
r
) (t

1
) ... (t

mr
)
and here the right side is zero unless
_
i
1
, .., i
r
,
1
, ..,
mr
_
=
_
1, 2, ..., m
_
.
t
1
, ..., t
m
are so chosen that (t
1
) ... (t
m
) give the natural orienta-
tion,
((t
i
1
) ... (t
i
r
))(t

1
, ..., t

mr
) = 0
if (i
1
, ..., i
r
,
1
, ...,
mr
) (1, 2, ..m) and
((t
i
1
) ... (t
i
r
))(t

1
, ...t
mr
)l =
= (t
i
1
) (t
i
r
) (t

1
)...(t

mr
).
Now, the right side is clearly l where is the signature of the
permutation
(1, 2, ..., n) (t
i
1
, ..., t
i
r
, t

1
, ...t

mr
).
It follows that
((t
i
1
) ... (t
i
r
)) = (t

1
) ... (t

mr
)
where 0 <
1
< . . . <
mr
m are so chosen that 10
_
t
i
1
, ..., t
i
r
, t

1
, .., t

mr
_
= (1, 2, .., m).
Hence
()((t
i
1
)), (t
i
1
, ..., t
i
r
)
= (t

1
) .. (t

r
) (t
i
1
) ... (t
i
r
)
=

where

is the signature of the permutation

(1, 2, ..., m) (t

1
, .., t

r
, t
i
1
, .., t
i
r
).
Hence

= (1)
r(mr)
= (1)
r(m1)
. It follows that
()((t
i
1
)... (t
i
)(t
i
1
, .., t
i
r
) = (1)
r(m1)
_
(t
i
1
) ... (t
i
r
)
_
(t
i
1
, ..., t
i
r
).
1. The space [
p,q
11
For ( j
1
, j
2
, .., j
r
) (i
1
, .., i
r
) we have already seen
()(t
i
1
) .. (t
i
))(t
j
1
, ..., t
j
r
) = 0.
Evidently,
_
(t
i
1
) ... (t
i
)
_
(t
j
1
, .., t
j
r
) = 0.
Hence the lemma.
The isomorphisms between the F-module of C

tangent vector
elds on X onto the E-module C
1
extends to an isomorphism between
the r-th exterior power of these modules.
Let U be a coordinate open set on the dierentiable manifold X.
Let x
1
, ..., x
m
be a coordinate system on U. Then for the xed basis
dx
1
, ..., dx
m
for the space of dierentials at every point of U, any r-form 11
C
r
can be represented in U by
=
I
dx
I
,
where
I
is a C

function on U, I = (i
1
, ..., i
r
) is an r-tuple of indices
1 , i
1
< . . . < i
r
m = dim
R
X, and dx
I
= dx
1
i
. . . dx
1
r
.
Let g
i j
be the metric tensor in U, and let g
i j
(i, j = 1, ..., m) be the
elements of the inverse matrix g
1
= (g
i j
)
1
. Then
1
is dened in U
by its components (with respect to the basis

x
1
, ...,

x
m
)

i
1
...i
r
= g
i
1
k
1
. . . g
i
r
k
r

k
1
...k
r
.
If
()
J
dx
J
is the local representation in U of the (mr)-form , J being the multi-
index J = ( j
1
, . . . , j
mr
)(1 j
1
< . . . < j
mr
m), then
()
J
=
IJ

I
.
where =
1...
m
dx

... dx
m
is the representation of the volume ele-
ment in U(
1
...
m
=
_
det(g
i j
)) and
IJ
=

1...m
being the sign of the
summation (1, ...m) (I J).
12 1. Vanishing theorems for hermitian manifolds
The proof is straightforward. It follows that for , C
r
,
=
1
r!
_

I
_

_
(1, II)
the summation being over all multi-indices I = (i
1
, ..., i
r
) or
=

I
if we extend the summation only to the multi-indices I = (i
1
, ..., i
r
),
l i
1
< . . . < i
r
m clearly,
clearly, =
and =
_

_
. . . (1, III)
We now revert to the situation when the manifold X is a complex 12
manifold and the Riemannian structure is the one canonically associated
to a hermitian structure on X . Let C
pq
(X) be the complex vector space
of C

forms of type (p, q) on X .

Lemma 1.3. The -operator maps C
pq
(X) into C
nq,np
(X).
Proof. The isomorphism of the (real) tangent space to X at xX onto
space of dierentials extends to a complex linear isomorphism of the
complexied tangent space onto the space of complex valued dier-
entials. Denoting the extension again by and noting the fact that every
complex valued dierential form may be regarded as a multilinear form
on the complexied tangent space, we see that the formula
()(t
1
, . . . , t
mr
) = (t
1
) (t
mr
)
holds for complex tangent vectors (t
1
, . . . , t
mr
) as well. Moreover from
the denitions of the associated Riemannian structure, we see easily that
for every ,
_

z

i
_
is a linear combination of the dz

i
while
_

z

_
is
a linear combination of the dz

i
. Our assertion now follows from the
denitions of forms of type (p, q).
1. The space [
p,q
13
Now, let E be a holomorphic vector bundle on X dened by (G =
U
i

iI
; e
i j
: U
i
U
j
GL(n, C)).
Let h = (h
i
)
iI
be a hermitian metric along the bres. Let C
p,q
(X, E)
denote the space of C

- forms of type (p, q) on X with values in E

. The operator * of the associated Riemannian structure denes also a
canonical isomorphism
: C
p,q
(X, E) C
np,np
(X, E)
as follows. If a form in C
p,q
(X, E) is represented in U
i
by a column 13
_

_
.
.
.

1
i

m
i
_

_
where each
k
i
is a scalar (p, q) form, then is represented by
_

_
.
.
.

1
i

m
i
_

_
in U
i
. Next, the hermitian metric h enables us to dene an operator
# : C
p,q
(X, E) C
q, p
(X, E

),
where E

is the dual-bundle to E.(E

is dened by the transition func-

tions x t
1
e
i
(x)
with respect to the covering G = U
i

iI
). In fact if
we represent a (p, q)-form in U
i
by a column
_

_
.
.
.

1
i

m
i
_

_
, where each
k
i
is a scalar form of type (p, q), then # in U
i
is given by the column of
(p, q)-forms
(#)
i
=
_

_
h
i

1
i
h
i

2
i
h
i

m
i
_

_
That (#)
i

iI
dene forms with values in E

follows from the fact

h
i

i
= (t
e
ji
h
j
e
ji
e
i j

j
) = (t
e
ji
h
j

j
)

= e
t1
i
j
h
j

j
Remark . C
p,q
(X, E) has a structure of an F-module where F is the 14
ring of global complex-valued C

-functions. For this structure of F-

modules on the C
p,q
(X, E)
: C
p,q
(X, E) C
nq,np
(X, E)
14 1. Vanishing theorems for hermitian manifolds
is an F-linear isomorphism, while the operator
# : C
p,q
(X, E) C
q, p
(X, E

)
satises the condition
#( f .) = f (#) where f F, C
p,q
(X, E).
It follows that the operators * and # dene actually homomorphisms
: A
p,q
(E) A
nq,np
(E) (resp. A
p,q
(E)

A
q, p
(E

)
where A
rs
(E)(resp. A
rs
(E

))
is the vector-bundle of exterior dierential forms of type (r, s) with val-
ues in E (resp. E

); here * is C-linear while # is anti linear.

In the special case when E is trivial bundle, the map # : C

(X, E)
C

(X, E

) is simply the map

f f .
Let C
p,q
(X, E) and C
rs
(E

). Then ( ) is dened as a
scalar form of type (p + r, q + s). For , C
pq
(X, E) we set
A
E
(, )l = (# ). (We note that # = #.) If (supp supp )
is compact, then
_

_
X
# =
_

_
X
A
E
(, ) <
Let U be an open coordinate set in X and let , be represented in
U by
=
_
1
p!q!

a
A

B
dz
A
dz
B
_
, =
_
1
p!q!

a
AB
dz
A
dz
B
_
,
where: a = 1, . . . , m = rankE, A = (
1
, . . . ,
p
), B = (
1
, . . . ,
q
), 15
1
i
,
i
n, dz
A
= dz

1
. . . dz

p
, dz

= dz

1
. . . dz

q
.
Then it follows from (1 II) that
A(, ) =
1
p! q!
h
ba

a
AB

bAB
,
2. The space W
p,q
15
(h
ba
)
a,b=1,...,m
being the local representation in U of the metric along the
bres of E. We shall call A(, )
1
2
the length of the form .
Sometimes it will be denoted also by .
Let
p,q
(X, E), be the space of C

(p q) forms with compact

support. Then for , D
pq
(X, E) we can dene a scalar product:
(, ) =
_
X
(#) =
_
X
A
E
(, ).
Also, we denote (, ) by jj
2
; j j is in fact a norm: the hermitian
bilinear form (, ) denes on D
p,q
(X, E) the structure of a prehilbert
space over C. The completion of 1
p,q
(X, E) is denoted [
p,q
(X, E).
This latter space is referred to in the sequel also as the space of square
summable forms. Because of the Riesze-Fisher theorem [
pq
(X, E) can
be identied with the space of square forms of type (p, q)-i.e. the space
of such that
_
X
A(, ) < .
2 The space W
p,q
The operator : A
p,q
(E) A
p,q+1
(E) denes homomorphisms (again
denoted by the same symbol)
: C
p,q
(X, E) C
p,q+1
(X, E)
: D
p,q
(X, E) D
p,q+1
(X, E).
We can now dene the adjoint of with respect to hermitian metric 16
on X and a hermitian metric along the bres of the vector bundle :
E X, i.e. an operator
: C
p,q
(X, E) C
p,q1
(X, E)
satisfying
(, ) = (, ) (1.1)
for D
p,q
(X, E), D
p,q+1
(X, E). Such an operator, if it exists is
immediately seen to be unique. For the existence we have the
16 1. Vanishing theorems for hermitian manifolds
Lemma 1.4. If D
p,q
(X, E), D
p,q+1
(X, E) are Lipschitz continuous
forms with compact support, then (, ) = (, #
1
#).
Proof. We note rst that the scalar products on both sides are dened.
Firstly, the and # being C-linear (resp. C-anti linear) bundle homomor-
phisms, these can also be dened on arbitrary forms - that is arbitrary
sections of the vector bundle A
p,q
(E). Secondly, since and are Lips-
chitz continuous and

#
1
# are dened almost every where and
are bounded measurable forms with compact support. We have more-
over,
d( #) = ( #)
=
_
#
_
+ (1)
p+q
#
=
_
#
_
#
since = (1)
(p+q)(2npq)
= (1)
p+q
(here n is the complex dimension of X). Applying Stokes formula, we
obtain (since and are compact support),
0 =
_
X
( #)
_
X
#
i.e. (J, ) = (, ), which proves the lemma.
The operator depends on the hermitian metric on X and on the 17
hermitian metric along the bres of E. To emphasise this fact, we may
write sometimes
E
for .
We have already introduced the norm j j on D
p,q
(X, E) and denoted
the corresponding completion by [
p,q
(X, E). We will now introduce
another norm on D
p,q
(X, E).
Denition 1.2. For , D
p,q
(X, E).
a(, ) = (, ) + (, ) + (, ).
and N()
2
= a(, ).
As before we see that the hermitian scalar product a(, ) denes a
complex prehilbert-space structure on D
p,q
(X, E). We denote by W
p,q
2. The space W
p,q
17
(X, E) the corresponding completion. Clearly, we have for , D
p,q
(X, E) the inequality jj N(), so that the identity
D
p,q
(X, E) D
p,q
(X, E)
extends uniquely to a continuous linear map,
i : W
p,q
(X, E) [
p,q
(X, E).
Proposition 1.1. i is injective.
Proof. Let

D
p,q
(X, E) be a Cauchy sequence in N and assume that
j

j 0. Since

is a Cauchy-sequence in N,

and

are Cauchy
sequences in the norm j j. Hence

and

tend to a limit in
[
p,q+1
(X, E) and [
p,q1
(X, E) respectively.
We denote these limits by u and v. By our identication of [
p,q
(X, E) as the space of square summable E-valued forms, u and v may be
regarded as forms on X with values on E.
Now for any D
p,q+1
(X, E) 18
(u, ) = lim

, ) = lim

, ) = 0
since j

j 0. Since is arbitrary in D
p,q+1
(X, E), U = 0. Similarly
v = 0. That is,

0 and

0 in [
p,q+1
(X, E) and [
p,q1
(X, E)
respectively. Thus N(

) 0 as , hence the proposition.

By continuity we obtain from (1.1) that, if W
p,q
(X, E),
W
p,q+1
(X, E), then
(, ) = (, )
Clearly, we may now regard W
p,q
(X, E) also as a space of measur-
able E-valued forms on X. From the denitions, it is moreover clear that
if f W
p,q
(X, E), f and f in the distribution sense are currents repre-
sentable by square summable (p, q+1) and (p, q1) forms respectively.
In general W
p,q
(X, E) is not the space of all square summable forms of
type (p, q) whose and (in the sense of distributions) are again square
summable. We have however, the
18 1. Vanishing theorems for hermitian manifolds
Theorem 1.1. If the Riemannian metric (associated to the hermitian
metric) on X is complete, then
W
p,q
(X, E) =
_
[
p,

q
(X, E); [
p,q+1
(X, E); [
p,q1
(X, E)
_
Proof. We will rst establish three lemmas which are needed for the
proof.
Lemma A. There exists C

> 0 depending only on the dimension of X

such that for any scalar form u and any v C
p,q
(X, E),
A
E
(u v, u v)(x) C

u
2
A
E
(v, v)(x)
where u
2
denotes the length of the scalar form a). (We recall that 19
A
E
(, ) = #). This is simply a lemma on nite dimensional
vector space with scalar products. We omit the proof.
Lemma B. Let p

X be any point. The function (x) = d(p

, x) =
distance from 0 of x, is locally Lipschitz continuous and wherever has
partial derivatives, d
2
2n.
Proof. We have by the triangle inequality
(x) (y) d(x, y).
Now if U is a coordinate open set with coordinates (X
1
, . . . , X
m
) (X
considered as a dierentiable manifold of dimension m = 2n) and the
Riemannian metric is given on U by
_
g
i j
dx
i
x
j
then for any V 0 U,
there exists and such that

dx
i
2

g
i j
dx
i
dx
j

dx
i
2
so that if V is a ball about the origin in U, there exist constants C
1
C
2
> 0
such that
C
1
x y d(x, y) C
2
x y for x, y V.
Thus (x) (y) < C
2
x y for x, y V. Hence the rst
assertion.
2. The space W
p,q
19
For the second assertion consider about a point p a neighbourhood U
in which we can introduce geodesic polar-coordinates. Let (x
1
, . . . , x
m
)
be a coordinate system such that x
i
(p) = 0 for 1 i m.
For > 0 we can choose U so small that g
i j
<
i j
+ , i.e.
d(x, p)
2
(1 + )
_
x
2
i
. In such a coordinate system, we have setting
e
i
= (0, . . . , 1 . . . , 0) (1 at the i
th
place)

x
i
() = Lt
h0
(he
i
) (0)
h
On the other hand we have 20
(he
i
) (0) d(0, h, e
i
) (1 + )h
Since we are working with a coordinate system such that d(x, 0)
(1+
_
_
x
2
i
_ 1
2
. Hence is Lipschitz continuous and, if the derivatives ex-
ist,

x
i
(p) 1. Now, by denition d
2
=
_
g
i j

X
i

X
j
and in the coor-
dinate system introduced above, g
i j
(0) =
i j
so that
2
=
m
_
1=1
_
_

x
i
_
_
2

m. When X is a complex manifold of complex dimension n, we have,

d
2
2n.
Lemma C. Let

D
p,q

(X, E) =
_
[
p,q
(X, E), [
p,q+1
(X, E), Supp X
_

D
p,q

(X, E) =
_
[
p,q
(X, E),
[
p,q
(X, E), [
p,q1
(X, E), supp X
_
(where and are in the sense of distributions) and nally

D
p,q
(X, E) =

D
p,q

(X, E)

D
p,q

(X, E).
Then
20 1. Vanishing theorems for hermitian manifolds
(i) D
p,q
(X, E) is dense in

D
p,q

(X, E) w.r.t. the norm

P() : P()
2
= jj
2
+ jj
2
and
(ii) D
p,q
(X, E) is dense in

D
p,q

(X, E) with the norm

Q() : Q()
2
= jj
2
+ jj
2
(iii) D
p,q
(X, E) is dense in

D
pq
(X, E) with the norm
N : N()
2
= jj
2
+ jj
2
+ j
2
.
Proof. We will prove (i) ; the proofs for the other two cases are similar. 21
Let G = U
i

iI
be a locally nite covering of X such that E
U
i
is trivial.
Let m be the rank of E. Assume further that each U
i
is a coordinate open
set. Let V
i
V
i
U
i

iI
be a shrinking of U
i
. Let (
i
)
iI
be a partition
unity subordinate to V
i

iI
. Let

mathscrD
p,q(X,E)

. E
U
i
being
trivial, we may take
i

i
to be a C
m
-valued from on C
n
with compact
support in V
i
. For any > 0, we can nd a C
m
valued C

from
i
with compact support in V
i
C
n
, such that
i

i

i
< and
j
i

i
j < . This can be secured by the usual regularisation methods.
Since the support of is compact, we may take
i
= 0 except for a
nite number of i. Since E
U
i
is trivial and
i

i
were regarded as scalar
forms through suitable trivialisations, we may now revert the process
and consider
i
as E-valued C

-forms with support in V

i
. It follows
that =
_

i
is a C

-form with compact support and

j j = j

i
j

i
j
i

i

i
j < M
and similarly j j < M where M is the number of indices of the
nite set
i i I, U
i
support .
Since M is xed for a given and is at our choice, the lemma is
proved.
2. The space W
p,q
21
Lemma C enables us to prove the following more general statement.
Theorem 1.1. If the Riemannian metric (associated to the hermitian 22
metric) of X complete, then
i) D
p,q
(X, E) is dense in the space
_
[
p,q
(X, E), [
p,q+1
(X, E)
_
,
with respect to the norm P();
ii) D
p,q
(X, E) is dense in the space
_
[
p,q
(X, E), [
p,q1
(X, E)
_
,
with respect to the norm Q();
iii) D
p,q
(X, E) is dense in the space
_
[
p,q
(X, E), [
p,q+1
(X, E), [
p,q1
(X, E)
_
,
with respect to the norm N().
Proof. We will prove i). The proofs of ii) and iii) are similar.
In view of Lemma C it is sucient to prove that every distributive
form in [
p,q
(X, E) such that [
p,q+1
(X, E) can be approximated
as closely as we want by forms such that , are square summable
and supp , X.
Let : R
1
[0, 1] be a C

function such that

(i) (t) = 1 if t < 1 and (ii) (t) = 0 if t > 2.
Let M = Supp
t

d
dt
. Let d(x, y) be the distance function dened by
the complete Riemannian metric of X. Then we x a point p

X and 23
set (x) = d(x, p

) for x X, the function

(x) =
_
(x)

_
> 0
is locally Lipschitz, and where the derivatives

x
i
exist, then
d

t
_
(x)

_
d
2

2nM
2

2
22 1. Vanishing theorems for hermitian manifolds
in view of Lemma B. Since the metric is complete the ball of centre p
and radius c,
B
c
= p d(p, p

) < c
is relatively compact in X for all c > 0.
Consider now the form

. =

has compact support. It is

easily seen that

. =

is in

D
p,q
(X, E). We will now prove that

in the norm N as . In fact, we have rst of all

j

j = j(1

)j
XB

jj
XB

0 as .
Secondly
j

j = j

j
jj
XB

+ j

j
Now, j

j = A(

2
A(, ) by Lemma A.
On the other hand, one checks easily that almost everywhere

2
d

2nM
2

2
,
so that, we obtain
j

j jj
XB

+
C

jj.
Hence

0 as . This completes the proof of Theorem 24

1.1 (i).
Denition 1.3. : C
p,q
(X, E) C
p,q
(X, E) is the operator + .
The following result is easily checked.
Lemma 1.5. The operator dened above is strongly elliptic.
The operator depends on the metric on X and on the metric along
the bres of E. To emphasize this fact we may write
E
for .
For the operator we have the following result.
2. The space W
p,q
23
Theorem 1.2 (Stampacchia Inequality). We assume the Riemannian
metric on X (associated to the hermitian structure) to be complete. Let
p
o
X and for > 0, let B

= x d(x, p
o
) < . Then there exists a
constant A > 0 such that for every > 0, for every choice r < R of a
pair of positive reals r, R and every C
p,q
(X, E),

2
B
r
+
2
B
r
j
2
B
r
+
_
1

+
A
(R r)
2
_
jj
2
B
R
Proof. Let us choose as before a C

function : R
1
[0, 1] such that
(t) = 1 for t < 1 and (t) = 0 for t > 2.
Let M = sup
d
dt
. consider the function dened by
W(x) =
_
(x) + R 2r
R r
_
where (x) = d(p
o
, x). Evidently then, has support in B
R
and w 1
on B
r
(B
R
X).
Moreover, d
2
=
1
Rr

t
_
(x)+R2r
Rr
_
d

2
. It follows that d
2
25
2nM
2
(Rr)
2
.
Suppose now that C
pq
(X, E) and is any Lipschitz continuous
form with compact support in B
R
, then
(, )
B
R
+ (, )
B
R
= (, )
B
R
.
Set =
2
. We have then
=
2
+ 2
=
2
(2 ).
This leads to
(, )
B
R
+ (, )
B
R
= (,
2
) (, 2) + (, (2 )).
Now, by Schuarz inequality,
(,
2
)
B
R

1
2
jj
2
B
R
+
1
2
j
2
j
B
R

2
jj
2
B
R
+
1
2
jj
2
B
R
24 1. Vanishing theorems for hermitian manifolds
(, 2)
B
R

1
2
jj
2
B
R
+
4c
o
nM
2
(R r)
2
jj
2
B
R
and (, (2 ))
1
2
jj
2
B
R
+
4c
o
nM
2
(R r)
2
jj
2
B
R
(where c
o
is the positive constant which has been introduced in lemma
A)
If follows that 26
jj
2
B
R
+ jj
2
B
R
jj
2
B
R
+ jj
2
B
R
+
8c
o
nM
2
(R r)
2
jj
2
B
R
.
The inequality follows now from
jj
2
B
R
jj
2
B
and jj
2
B
R
jj
2
B
R
This completes the proof of theorem 1.2.
Corollary 1. For
2
C
p,q
(X, E) and for any > 0,
jj
2
+ jj
2
jj
2
+
1

jj
2
.
Proof. Set R = 2r in stampachia inequality and let r .
Corollary 2. If C
p,q
(X, E), jj < and jj < , then jj < ,
jj < . If square = 0, then = = 0.
Proof. The rst assertion follows from corollary 1. The second again
from corollary 1 since is arbitrary.
Remark . On any (paracompact) complex manifold X there exists a
complete hermitian metric. More exactly we shall prove that, given any
hermitian metric ds
2
on X, there exists a C

function F : X R such
that F ds
2
is a complete metric.
Proof. Let B

N
be a sequence of compact sets such that
B

o
B
+1
, B

= X.
3. W-ellipticity and a weak vanishing theorem 25
Let f

: X R be a C

function on X, satisfying the following 27

conditions:
0 f

1,
f

= 1 on B
+1
B

,
Supp f

B
+2
B
1
Let d(x, y) (x, y X) be the distance function determined by the
hermitian ds
2
, and let

= inf
xB

yB
+1
d(x, y).
Then

> 0. Consider the positive C

function
F(x) =
+

=0
f

(x)

and the hermitian metric

ds
2
= F(x)ds
2
. Let

d(x, y) be the distance
function determined by ds
2
. We have

d(B

, B
+1
)
x
inf
B
+1
B

F(x)d(B

, B
+1
)

= 1.
This implies that every Cauchy sequence for the distance

d(x, y) con-
verges. Hence

ds
2
is a complete hermitian metric.
3 W-ellipticity and a weak vanishing theorem
28
We now introduce a seminorm on W
p,q
(X, E) as follows: for , W
p,q
(X, E), let b(, ) = (, )+(, ); then b is a positive semi denite
form and b() denes a semi-norm on W
p,q
(X, E). the map
j : W
p,q
(X, E) W
p,q
b
(X, E)
26 1. Vanishing theorems for hermitian manifolds
where the latter space is W
p,q
(X, E) provided with the (not necessarily
Hausdor) topology dened by the seminorm b(, ) and j is the iden-
tity is clearly continuous: in fact, b(, ) N().
Denition 1.4. We say that E is W
p,q
elliptic with respect to the hermi-
tian metric on X and the and the hermitian metric along the bres of E,
if j admits a continuous inverse.
In particular, the W
p,q
-ellipticity of E implies that b(, )
1
2
is actu-
ally a norm. In fact, there is a K > 0 such that N()
2
Kb(, ). Since
by denition, N() = b(, ) + jj
2
, W
pq
-ellipticity is equivalent to fol-
lowing: there is a constant C > 0(= K 1 when K is as above) such
that
jj
2
Cb(, ) = C(jj
2
+ jj
2
). (1.2)
We shall call C a W
p,q
-ellipticity constant.
Proposition 1.2. Assume given a hermitian metric on X and a hermitian
metric along the bres of : E X. Suppose further that E is W
pq
-elliptic with reference to these hermitian metrics. Then there is a linear
map G : [
p,q
(X, E) W
p,q
(X, E) such that, for f [
p,q
(X, E),
W
p,q
(X, E), we have
( f , ) = ( Gf , ) + (Gf , ).
The linear map G is continuous more exactly: 29
b(Gf , Gf ) Cj f j
2
(1.3)
C being the constant which appears in (1.2). Moreover Gf is uniquely
determined by the above formula.
Proof. (, f ) denes a linear form on W
p,q
(X, E) which is evi-
dently continuous. By the Riesz representation theorem (since E is W
p,q
elliptic, b(, ) denes a Hilbert space structure on W
p,q
(X, E) equiva-
lent to that dened by N), there is a unique element Gf W
p,q
(X, E)
such that
( f , ) = b(Gf , ) = (Gf , ) + (Gf , ).
3. W-ellipticity and a weak vanishing theorem 27
We have, now
b(Gf , Gf ) = (Gf , Gf ) + (Gf , Gf ).
= ( f , Gf ) by the above equation.
Hence b(Gf , Gf )
2
j f j
2
.jGf j
2
Cj f j
2
b(Gf , Gf ) in view of W
pq
-
ellipticity. That is,
b(Gf , Gf ) Cj f j
2
.
We obtain
N(Gf )
2
(C + 1)j f j
2
.
This completes the proof of Proposition 1.2.
Corollary. Gf = f (in the sense of distributions).
Proof. For u D
p,q
(X, E), we have ( f , u) = (Gf , u) = (Gf , u).
This proves the lemma.
Proposition 1.3. Let E be W
p,q
-elliptic with respect to a given metric 30
along the bers and to a complete hermitian metric on X. In the notation
of Proposition 1.2 we have the following.
(i) If f [
p,q
(X, E) and f [
p,q+1
(X, E), then
Gf W
p,q+1
(X, E); Gf = f ;
jGf j
2

j f j
2
+ jGf j
2
for > 0.
(ii) If f [
p,q1
(X, E) and f [
p,q+1
(X, E), then
Gf W
p,q1
(X, E); Gf = f ;

Gf
2
f
2
+ Gf
2
f or > 0.
Proof. Since the metric is complete, by Theorem 1.1 1
p,q
(X, E) is
dense in the space
_
[
p,q
(X, E), [
p,q+1
(X, E)
_
provided with
the norm P()
2
= jj
2
+ jj
2
. Let f [
p,q
(X, E) and f [
p,q+1
(X, E). Then there is a sequence f
n
1
p,q
(X, E) such that j f
n
f j
28 1. Vanishing theorems for hermitian manifolds
0, j f
n
f j 0 as n . We have moreover, from Proposition 1.2,
that
N(G)
2
Kjj
2
for L
p,q
(X, E). Hence
jGf
n
Gf
m
j Kj f
n
f
m
j
and jGf
n
Gf
n
j + jGf
n
Gf
n
j Kj f
n
f
m
j.
Applying now Stampachia inequality (Corollary 2, Theorem 1.2) to
Gf
n
, and taking into account the fact that and commute, we obtain,
for > 0,
j(Gf
n
Gf
m
)j
2

( f

)j
2
+ j(G( f

)j
2
Since the right hand side tends to 0 as n, m , it follows that
_
f
n
_
is a Cauchy-sequence in [
p,q
(X, E). It is immediate then that 31
Gf = lim
n
Gf
n
[
p,q
(X, E). On the other hand (Gf ) = 0.
Hence Gf W
p,q+1
(X, E). The equation Gf = f follows from
the fact that and commute and the fact that Gf = f . The last
inequality is the Stampachia inequality applied to Gf . The proof of
Part (ii) of the proposition is entirely analogous.
Theorem 1.3. Let : E X be a holomorphic vector-bundle which
is W
p,q
-elliptic with respect to a complete hermitian metric on X and
a hermitian metric along the bres of E. Then if q > 0, given f
[
pq
(X, E) with f = 0, there is a unique x W
pq
(X, E) such that f =
x and x = 0. Moreover, we have jxj
2
Cj f j
2
, C being a W
p,q
-
ellipticity constant.
Proof. We set x = Gf . We have then Gf = f . (Corollary to Propo-
sition 1.2). Clearly f = 0 [
p,q+1
(X, E). Hence by (i) of proposition
1.3, we have Gf W
p,q+1
(X, E) and further jGf j
2
jGf j
2
+
1

jGf j
2
. On the other hand since and commute, it follows again
from the Corollary to Proposition 1.2, that Gf = 0. Since is ar-
bitrary, x = 0. It follows that (x, x) = (x, x) = 0 (note that
x W
p,q+1
(X, E): Proposition 1.3). Hence x = 0. Now = +
so that Gf = f leads to ( + )Gf = f that is, Gf = f .
3. W-ellipticity and a weak vanishing theorem 29
Finally (1.3) yields 32
jxj
2
Cj f j
2
.
The uniqueness of x satisfying x = f and x = 0 is easily
checked.
This completes the proof of the theorem.
It follows from the regularity theorem for elliptic systems that , if f
C
p,q
(X, E) [
p,q
then Gf (can be modied on a null set so that it)
will be of class C

on X.
In particular we have the
Corollary. If f 1
p,q
(X, E) and f = 0 then there exists C
p,q1
(X, E) such that = f
Remark. The corollary above clearly implies the following.
If E is W
pq
-elliptic with respect to a complete hermitian metric, then
the natural map
H
q
k
(X,
p
(E)) H
q
(X,
p
(E))
where the left-side stands for the q
th
cohomology with compact supports
of X with values in
p
(E), is the trivial map 0 for every
H
q
k
(X,
p
(E)).
Remark. Let G =
_
U
i
_
be a covering of X such that : E X is
dened with respect to G by holomorphic transitions e
i
j
: U
i
U
j

GL(m, C)(m = rank of E).
Let
_
h
i
_
iI
be a hermitian metric along the bers of E; then h
i
is a
C

function on U
i
whose values are positive denite hermitian matrices.
We have on U
i
U
j
33
h
i
=
t
e
ji
h
j
e
ji
,
and therefore t
h
1
i
=
t
(
te
1
ji
)
t
h
1
j
t
e
1
ji
30 1. Vanishing theorems for hermitian manifolds
That means that
_
t
h
1
i
_
denes a matric along the bres of the dual
bundle E

.
We note here for future reference the following identities, holding
with respect tot he metrics h
i
on E and t
h
1
i
on E

. They follow
immediately from the denitions of the operators involved.
A
E
(#, #) = A
E
(, ) (1.4)
# = (1)
p+q
#
E
, (, C
pq
(X, E)) (1.5)

E
# = (1)
p+q+1
#, (1.6)

E
# = #
E
. (1.7)
As a corollary we have that # denes an isometry of [
pq
(X, E) onto
[
np nq
(X, E) which maps W
pq
isometrically onto W
np nq
(X, E

).
Furthermore, if E is W
pq
-elliptic with respect to the metric h
i
on
E, then E

is W
np nq
- elliptic with respect to the metric t
h
1
i
on E

the W-ellipticity constants being the same.

4 Carleman inequalities
We will now formulate certain further conditions on the vector bundles 34
and show how these more stringent conditions lead to stronger vanishing
theorems than the one above.
We assume always that the hermitian metric denoted ds
2
on X is
complete. Let the given metric on E be denoted by h.
C
1
. There is given a C

function : X R
+
.
C
2
. For every non decreasing convex C

-function : R R.
E is W
pq
-elliptic with respect to (ds
2
, e

h
).
C
3
. The W
pq
-ellipticity constant is independent of.
That is, there is C > 0 independent of such that
j f j
2

C
_
j f j
2

+ j

f j
2

_
for f D
pq
(X, E) where j j

stands
for the norm with respect to (ds
2
, e
()
. h), and

denotes the -operator

with respect to these metrics.
4. Carleman inequalities 31
Conditions C
1
, C
2
, C
3
imply C

1
, C

2
, C

3
below. This weaker set of
conditions are sucient for the vanishing theorems we will now prove.
C

1
. There is given a C

function : X R
+
(this is the same as C
1
).
C

2
. Given C
o
> 0, there is a non-decreasing C

-function : R R
that (t) = 0 for t C
o
and (t) > 0 for t > C
o
, such that, E is
W
pq
elliptic with respect to (ds
2
, e
()
h) for every positive integer
.
C

3
. The W
pq
ellipticity constant is independent of that is, there is a 35
C > 0 such that for f D
pq
(X, E)
j f j
2

C
_
j f j
2

+ j

f j
2

_
,
where j j

stands for j j

and

for

.
Lemma 1.6. Assume given a constant C
o
> 0 and that the condition
C

1
, C

2
, C

3
above are satised (C
o
in condition C

2
is taken as the above
constant). For as in condition C

2
, let [
p,q

(X, E) denote the space

of E-valued forms on X which are square-summable with respect to
(ds
2
, e
()
.h). Then for f

[
p,q

(X, E)(q > 0) such that f = 0, there

exist

for every integer 0 such that [

p,q1

(X, E),

= f
and j

C j f j

(We assume that ds

2
is complete).
Proof. This follows from Theorem 1.3.
Theorem 1.4. Assume that conditions C

1
, C

2
, C

3
are satised. Then
for every square-summable form f of type (p, q)(q > 0) having compact
support such that f = 0, there exists [
p,q1
(X, E) such that, f =
, j j
2
C j f j
2
and (support of )
xj(x) < (sup (y)) y support f ).
Proof. Let C
o
= sup (x), x support f . Then we have
j f j
2

=
_
X
e
()
A( f , f )dx =
_
Supp . f
e
()
A( f , f )dX
32 1. Vanishing theorems for hermitian manifolds
=
_
Supp f
A( f , f )dx =
_
X
A( f , f )dX =j f j
2
Since ((x)) = 0 on the support of f . It follows that f

36
[
p,q

(X, E) for every integer > 0. Hence by Lemma 1.6, we can nd

[
p,q1

2
(X, E) for every Z
+
such that

= f and
2

C f
2
.
On the other hand, we have

so that,

C f
2
= C
1
say. It follows that we can assume (by passing to a subse-
quence if necessary) that

converges weakly in [
p,q1
(X, E) to a limit
. We have < C f . On the other hand, for every > 0,
_
(x)>c
0
+
e
()
A(

)dX C
1
and since is non-decreasing, we have,
e
(C
0
+)
_
C
0
+
A(

)dX C
1
.
It follows that
_
C
0
+
A(

)dX tends to zero and hence

0
almost everywhere in C
0
+ . (The integrand is positive for all
.) Hence = 0 on x (x) C
0
+ for every . Hence support
x

(x) C
0
. Finally for
u D
np,nq
(X, E

), (1)
p+q
(,

u) = (1)
p+q
_

u
= Lt(1)
p+q
_

u
=
_
f , u
_
.
Hence

= f in the sense of distributions. This completes the proof
of the theorem.
Remark. The proof contains the following lemma (which has no con- 37
nection with W
p,q
-ellipticity.
4. Carleman inequalities 33
Lemma 1.7. Suppose given a C

non-decreasing function : R R
such that (t) = 0 for t C
0
, (t) > 0 for t > C
0
, and a C

function
: X R. Suppose further that there is given an f [
pq
(X, E) such
that sup (x) on the support of f is C
0
. If for every integer > 0 there
exists

such that

= f and

C f
2
for a constant C
independent of , then there exists [
p,q1
(X, E) such that

= f ,

2
C f
2
and support x (x) C
0
. (Here by)

we
mean
_
e
()
A(

)dX).
Suppose now that the function in C

1
satises the following addi-
tional condition.
C

4
. For every C > 0, the set x (x) < C X.
Remark. If C

4
is satised in addition to C

1
, C

2
and C

3
then in Theorem
1.4, we can assert that the support of is compact.
Here we can state, as a corollary to Theorem 1.4 the following The-
orem 1.4

. If the hypothesis of Theorem 1.4 are fullled and if C

4
holds,
then
H
q
k
(X,
p
(E)) = 0.
Lemma 1.8. Let V [
p,q
(X, E) be the set V= V
p,q
(X, E) =
[
pq
(X, E)

there exists [
p,q1
(X, E) such that

= and N = 38
N
p,q
(X, E) = [
p,q
(X, E) = 0. Then N is the orthogonal
complement of V provided that the metric is complete.
Proof. Let [
p,q
(X, E). Then orthogonal complement of
V if and only if (,

) = 0 for every [
p,q1
(X, E) with


[
p,q1
(X, E). Since the metric is complete, this is equivalent to (,

) =
0 for every D
p,q1
(X, E). Hence the lemma.
Theorem 1.5. Assume conditions C

1
, C

2
, C

3
to hold for : E X. Let
f [
p,q+1
(X, E) be such that sup (x) on the support of f is C
0
. Suppose
further that ( f , g) = 0 for all g which belong to some [
p,q+1

, and are
such that

g = 0( = 1, 2, ....)(
1
). Then there exists [
p,q
(X, E) such
that

= f and support X (X) C
0
.
1
In view of the choice of , we have ( f , g) = ( f , g)

.
34 1. Vanishing theorems for hermitian manifolds
Let V
p q+1

be the space V
p q+1
constructed with respect to the met-
rics
(e
()
h, ds
2
). Since = 0 on supp f , then f [
pq

(X, E) for all > 0.

By the previous lemma
f V
p,q+1

for = 1, 2, ...
First assume that
f V
p q+1

for a particular .
In this case, there exists

[
p,q

(X, E) such that f =

Now by
Proposition 1.3, there is an X

= G

W
p,q

(X, E).
Such that

. We now set

. By Proposition 1.3

39
[
pq

(X, E). Further we have

=

(

)x

= f .
Since

= 0 , then by Theorem 1.1

W
pq

(X, E). On the other

hand, in view of W
p,q

-ellipticity,
j

j
2

C(j

j
2

+ j

j
2

) = Cj f j
2

= C
1
say C
1
is independent of since ((x)) = 0 for x support f .
Suppose now f V
pq+1

, then there exists a sequence f

i

V
pq+1

such that jf
i

- f j

0. Then choosing
i

W
pq

(X, E) is above for

each f
i

( V
pq+1

) we have

= f
i

j
i

Cj f
i

j
2

and j
i

j
2

Cj f
i

f
l

j
2

.
Hence
i

converge to a limit

in [
p,q

(X, E). Clearly

= F.
Further, from j
i

j
2

Cj f
i

j
2

, we deduce that j

j
2

Cj f j
2
. Now the
proof follows from Lemma 1.7.
The conditions C

1
, C

2
, C

3
below are dual to C

1
, C

2
, C

3
: that is
if C

1
, C

2
, C

3
hold for E-valued (p, q)-forms, then C

1
, C

2
, C

3
hold for
E

valued (n p, n q) forms.
C

1
. (= C

1
= C
1
) There is given a C

function : X R
+
.
4. Carleman inequalities 35
C

2
. For every C
o
0, there exists a non-decreasing C

-function, :
R R such that (t) = 0 if t C
0
and (t) > 0 for t > C
0
and E
is W
pq
-elliptic with respect to (ds
2
, e
()
h)
C

3
. The W
p,q
ellipticity constant is independent of = 1, 2, . . .
Then we have analogous to Theorem 1.5, the following 40
Theorem 1.6. Assume C

1
, C

2
, C

3
. Then if f [
p,q1
(X, E), is such
that sup (X) on support of f = C
0
, and if further ( f , u) = 0 for every u
[
p,q1

(X, E), such that

u = 0, then there exists [
p,q
(X, E) such
that f = and support X(X) C.
Theorem 1.7. Let T = T
p,q1
be a distribution valued form type (p, q
1) (that is a current of type (p, q 1)). Suppose further that

T
C
p,q
(X, E) and let K = support of T. Then for any neighbourhood U
of K, there is a form C
p,q1
(X, E) such that

T =

and support
V. In particular if K is compact, we can nd an with the above
property.
Proof. We recall (Prerequisites, 4) that we have ne resolutions
0
p
A
p,o

A
p,1
. . .

A
p,n
0
and 0
p
K
p,o

K
p,1
. . .

K
p,n
0.
Further there is canonical injection A
p,q
K
p,q
which is compatible
with the operator

. Hence by a standard result on cohomology of
sheaves the induced map of complexes

q0

(X, A
p,q
)

(X, K
p,q
)
is a homotopy equivalence of complexes for any para compactifying
family of closed sets on X. We consider in particular the family of
all closed sets of X which are contained in U. This is evidently a para-
compactifying family and theorem follows from the general result stated
above (for a more detailed proof, see [2], 97-99).
Theorems 1.5 and 1.7 together enable us to prove 41
36 1. Vanishing theorems for hermitian manifolds
Theorem 1.8. Let : E K be a holomorphic vector bundle. Assume
that for a suitable complete hermitian metric ds
2
on X and a suitable
hermitian metric h along the bres of E, conditions C

1
, C

2
, C

3
and
C

4
hold. Then the image

D
pq
(X, E) of

in D
p,q+1
(X, E) is a closed
subspace for the usual topology on D
pq
(X, E). Hence H
q+1
k
(X,
p
(E))
has a structure of a separated topological vector space.
Proof. The usual topology on D
pq
(H, E) may be described as fol-
lows. Let K

be an increasing sequence of compact sets such that

o
K

+ 1 and UK

= K. Further, let G = U
i

iI
be a locally nite
covering of X by open sets such that each U
i
is a relatively compact open
subset of a coordinate open set V
i
on X. Let G

= U
i
U
i
K

. We
topologize each D
pq
(K

, E) = D
pq
(X, E), supp

as follows:
for U
i
G

, U
i
may be regarded as a C

vector-valued function
i
on
U
i
; a fundamental system of neighbourhoods of zero in D
pq
(K

, E) is
given by,
_
D
pq
(K

, E)

<

, for every U
i
G

_
where

is an arbitrary family of positive reals.

D
pq
(K

, E) is a Frechet space. There is a natural injection

D
pq
(K

, E) D
pq
(K
+1
, E).

The image of D
pq
(K

, E) is a closed subset of D
pq
(K
+1
, E). The in- 42
duced topology on the image coincides with the topology of D
pq
(K

, E).
This shows that D
pq
(X, E) is a strict inductive limit of the Frechet spaces
D
pq
(K

, E) [8] 66-67; [9] (225-227). The usual topology of D

pq
(X, E)
is the inductive limit topology.
A subset of D
pq+1
(X, E) is closed if, and only if, it is sequentially
closed ([19], 228). Therefore it will be sucient to prove that

D
pq
(X, E) is sequentially closed in D
pq+1
(X, E).
Let
i

iN
be a sequence of (p, q)- forms of D
pq
(X, E). Such that
the sequence
i

iN
D
pq
(X, E) converges to an element of D
pq+1
(X, E). The sequence
i
is a bounded set in D
pq+1
(X, E).
4. Carleman inequalities 37
Hence there exists a compact K
s
such that
i
D
pq+1
(K
s
, E) for
i = 1, 2, . . ., and D
pq+1
(K
s
, E) ([8], 70; [19], 226).
Let c
o
= sup (x) on K
s
. Let : R R be a non decreasing C

function such that

(t) = 0 for t c
o
,
(t) > 0 for t > c
o
,
and let us consider the spaces [
pq+1

(X, E) = [
pq+1
()
(X, E) for =
1, 2, . . ..
Let g [
pq+1

(X, E) be such that

g = 0.
Then we have
(, g) = (, g)

= lim(
i
, g)

= lim(
i
,

g) = 0.
Thus by theorem 1.5 there exists a [
pq
(X, E) with compact
support (since we have assumed c

4
), such that 43
= .
Now, since in addition is C

, may be assumed to be a C

form,
i.e. D
pq
(X, E). Hence D
pq
(X, E). This completes the proof of
theorem 1.8.
Chapter 2
W-ellipticity on Riemannian
manifolds
5 W-ellipticity on Riemannian manifolds
We carry over the results proved for on a complex manifold to the 44
operator d-exterior dierentiation of ordinary forms. In the sequel until
further notice we consider only real valued forms. Through all this chap-
ter X will be a connected orientable Riemannian manifold. We choose
an orientation for X, and we denote by C
q
= C
q
(X) (resp. D
q
(X) = D
q
)
the space of C

q-forms (resp. C

forms with compact support). We

dened in Chapter 1 the operator
: C
q
C
nq
and the scalar product,
(, ) =
_
A(, )dX =
_
X

of two forms , in D
q
(X).
We dene the operator
: C
q
C
q1
39
40 2. W-ellipticity on Riemannian manifolds
as the linear dierential operator
(1)
q

1
d()
( C
q
(X)). If C
q
(X), C
q+1
(X), and if supp supp is
compact, then (d, ) = (, ). We also introduce the Laplace operator
= d + d;
is a strongly elliptic operator.
We have then 45
(, ) = (d, d) + (, )
for , D
q
(X).
On the space D
q
(X), we dene another scalar product: for ,
D
q
(X),
a(, ) = (, ) + (d, d) + (, ).
Then N() = a(, )
1
2
denes a norm on D
q
(X). We denote the comple-
tion of D
q
(X) under N by W
q
. If [
q
denotes the completion of D
q
(X)
with respect to the norm, j j
2
= (, ), then the identity extends canon-
ically to a continuous map
i : W
q
[
q
.
We have then the following result.
Proposition 2.1. i is an injection.
The proof is analogous to that of proposition 1.1. As in theorem
1.1 (Chapter 1), we have the following characterisation of W
q
(rather
i(W
q
)) when the Riemannian metric on X is complete.
Theorem 2.1. If the Riemannian metric on X is complete, then
W
q
(X) = [
q
(X), d [
q+1
(X), [
q1
(X).
Moreover D
q
(X) is dense in the space [
q
, d [
q1
,
with respect to the norm (j j
2
+ j d j
2
)
1
2
, and in the space
[
q
, [
q+1
with respect to the norm (j j
2
+ j j
2
)
1
2
.
Once again, we omit the proof which is analogous to that of Theorem
1.1 (Chapter 1). We state also, without proof, the analogue of Corollary
1 to Theorem 1.2 of Chapter 1.
5. W-ellipticity on Riemannian manifolds 41
Proposition 2.2. For C
q
and for any > 0 46
j d j
2
+ j j
2
j j
2
+
1

j j
2
.
Denition 2.1. The Riemannian manifold X is W
q
-elliptic if there is a
constant C > 0 such that
j j
2
C
_
j d j
2
+ j j
2
_
.
We shall call such a constant C a W
q
-ellipticity constant.
In further analogy to Chapter 1, we have nally the following
Theorem 2.2. If X is W
q
elliptic, then for f [
q
, there is a unique
x W
q
such tht f = x (in the sense of distributions). Further,
j x j c j f j
j dx j
2
+ j x j
2
C j f j
2
,
C being a W
q
ellipticity constant. If f is C

, then x C
q
W
q
. If the
metric on X is complete and j d f j< , then, for > 0,
j dx j
2

j d f j
2
+ j dx j
2
.
In particular, if df=0 , f=dx and dx=0.
Except for obvious modications, the proof is contained in Proposi-
tions 1.2 and 1.3 and Theorem 1.3 (Chapter 1).
Before we give criteria for W
pq
-ellipticity, on a complex manifold,
we give sucient conditions for W
q
-ellipticity, on an orientable Rie-
mannian manifold X. In order to do this we rst write down explicitly in 47
a coordinate open set the eect of the Laplacian on a C

pform.
Let (x, . . . x
n
) be a coordinate system on an open set U in X. Let the
Riemannian metric be

g
i j
dx
i
dx
j
in this open set. We denote the covariant derivation with respect to
the Riemannian metric. For a tensor ,

denotes the tensor

()(

x

).
42 2. W-ellipticity on Riemannian manifolds
( is regarded as a 1-form with values in the tensor bundle). In the
local-coordinate system, a form C
q
(X) may be written in the form.
=

1
1
q!

I
dx
I
where I runs over all q-tuples (i
1
, . . . , i
q
) of integers with 0 < i
j
n and
if I = (i
1
, . . . , i
q
) is a permutation (( j
i
), . . . , ( j
q
)) of I

= ( j
1
, . . . , j
q
),
then

I
=

I
.
(In particular, if I has repeated indices,
I
= 0). We have
A(, ) =
1
q!

I
(, C
q
)
For C
q
and a q-tuple I = (i
1
, . . . , i
q
),
d
iI
=

I
x
i
+

r
(1)
r

i
x
i
r
i
1
, . . . , i
r
, . . . i
q
=
i

I
+

r
(1)
r

ir

ii
1
, . . . i
r
. . . , i
q
(2.1)
as is seen by a direct computation. Similarly we have 48

J
=
i

i
J
(2.2)
where J is a (q 1)-tuple and
i
is the (q 1)-form dened by

i
J
=

g
i j

iJ
for any (q 1)-tuple J = ( j
1
, ... j
ql
).
From (2.1) and (2.2), it follows that
d
I
=
n

i=1

i
(d)
i
I
=
n

i=1

I

q

i=1
(1)
r

i
(
ir

i
)i
i
...

i
r
...i
q
(2.3)
5. W-ellipticity on Riemannian manifolds 43
(where
i
=
_
j
g
i j

j
) and
d
I
=
q

r=I
n

i=I
(1)
r

i
r

i
i
1
...

i
r
...i
q
(2.4)
Hence ()
I
=

I
+ ()
I
(2.5)
In the equation (2.5)
()
I
=

r
(1)
r
(
i
r

i

i

i
r
)
i
i
1
...

i
r
...i
q
=
q

r=1

j
(1)
r1
R
j
i
r

ji
1
...

i
r
...i
q
+
q

r,s=1
(1)
r+s
R
i
j
i
r
i
s

i. j.i
.
.

i
r
.

i
s
In the special case when is a function,
=
i

i
(2.6)
Lemma 2.1. For every D
q
we have 49
j j
2
+(K, ) =j d j
2
+ j j
2
(2.7)
Proof. From (2.6) above, we have setting
2
= A(, ) =
1

I
,
(
2
) =

2
=

1
q!

I
=
2
q!

i
(
i

I
.
I
)
=
2
q!

I
.
I
2
2
=

I
_
2
q!

2
q!
(K)
I

I
_
2
2
That is

2
= 2A(, ) 2A(K, ) 2A(, )...
If has compact support,
_

2
dx =
_
d
2
dX = 0
44 2. W-ellipticity on Riemannian manifolds
by Stokes formula.
On the other hand,
_
A(, )dX =j d j
2
+ j j
2
.
Hence
(K, )+ j j
2
=j d j
2
+ j j
2
: ( D
q
(X)).
If there exists a positive constant C such that
A(K, ) C
2
for every C
q
and at each point of X, then

2
C

(K, ) C

_
d
2
+
2
_
where C

=
1
C
. Hence the following.
Lemma 2.2. If there is a positive constant C such that 50
A(K , ) C A(, )
for every C
q
and at each point of X, then the Riemannian manifold
X is W
q
-elliptic.
6 A maximum principle
In the sequel we consider weaker condition on the expression A( , ).
We shall deal with the case where the quadratic from A( , ) is positive
semi- denite outside of a compact.
From now on we shall always assume X to be oriented and con-
nected, and the Riemannian metric of X to be complete.
Lemma 2.3. Assume A( , ) 0 (i.e. A(, ) positive semi-denite),
outside a compact set K in X. Then for W
q
, j j
2
< and (2.7)
holds.
6. A maximum principle 45
Proof. By Lemma 2.1 we have for 1
q
(X),
j j
2
+(, ) = (d , d) + (, ).
Since 7 is compact, there is C 0 such that (7, ) > CC
2
for every C
q
and each point of 7. Hence,
j j
2
+
_
X
A(k, ) C j j
2
k
+ j d j
2
+ j j
2
.
Since A(, ) 0 on X 7 we have for 1
q
,
j j
2
C
_
j j
2
+ j d j
2
+ j j
2
_
The lemma follows now from the fact that 1
q
is dense in W
q
. (Theorem
2.1.)
Remark. We have proved more: there is a C > 0 such that 51
jj
2
C
_
jj
2
K
+ jdj
2
+ jj
2
_
Let X be a manifold as in the above lemma.
Let C
q
. Identity (2.8) shows that
2
0 at each point of the
set Y where = 0 and A(K, ) 0
Applying a classical lemma of E. Hopf ([36], 26-30) we see that

2
cannot have a relative maximum at any point of Y. Thus, if X is
compact,
2
takes its, maximum in the set Supp f K.
If X is not compact we cannot draw the same conclusion. However
the following proposition will provide an estimate of
2
on X in terms
of Sup
2
on Supp f 7.
Proposition 2.3. Let X be a connected oriented and complete Rieman-
nian manifold. Assume give a compact set 7 on X such that for X
7, A(7, )(x) 0 for every C
q
. Suppose that C
q
[
q
is such
that = f [
q
. Then W
q
and sup
xX
(x)
2
C
0
where
C
0
=
Sup
xsupp f 7
(x)
2
.
46 2. W-ellipticity on Riemannian manifolds
For the proof of this proposition we utilise the following result
Lemma 2.4 (Ganey [12]). Let 7 be an oriented complete Riemannian
manifold and a C

1-form. Then if
_
X
dX < and
_
X
dX < ,
we have
_
X
dX = 0.
Proof. Let be a real value C

function on such that (t) = 1 for 52

t 0 and (t) = 0 for t 1. Let p
0
X be a xed point and = (x)
denote the distance function (from p
0
). Then (x) is locally Lipchitz as
was remarked earlier (Lemma B of Chapter 1). Let 0 < r < F and let us
consider the form

_
(x) r
F r
_

This form is locally Lipchitz and has compact support in the ball
x(x) R.
Hence by Stokes formula (which is applicable to Lipchitz continu-
ous form),
_
X
d
_
(x) r
R r
_
= 0;
We have,
_
X
1
R r

_
(x) r
r
_
d +
_
X

_
(x) r
R r
_
d = 0.
Now set R = 2r. Then
_
X
1
r

_
(x) r
r
_
d +
_
X

_
(x) r
r
_
d = 0
Now d
2
(lemma B of Chapter 1).
Hence, since is integrable, we have 53
_
X
d dX < .

6. A maximum principle 47
Hence, using the fact that

is bounded, we see that

1
r
_
X

_
(x) r
r
_
d 0 as .
It follows that
Lt
r
_
X

_
(x) r
r
_
d = 0
On the other hand, since
_
d dX < and
_
(x)r
r
_
is bounded
and tends to 1 as r ,
r
Lt

_

_
(X) r
r
_
d =
_
d .
This proves the lemma.
Proof (of Proposition 2.3) Let C
0
and C
1
be two positive constants,
0 < C
0
< C
1
, and let : R R be a C

function such that

(t) 0,

(t) 0,

(t) 0,
(t) = 0, for t C
0

(t) > 0, for t > C

0

(t) = 0, for t < C

0
and for t > C
1
,
and such that

(t) be bounded. Since

(t) has compact support, it is

bounded. We have then for as in the proposition,
(
2
) = d((
2
))
=
_

(
2
).d
2
_
= d
_

(
2
)d
2
_
= d
_

(
2
) d
2
_
=
_
d

(
2
) d
2
+

(
2
)d d
2
_
=
_

(
2
)d
2
d
2
+

2
d d
2
_
=

2
d
2

2
+

2
.(
2
).
48 2. W-ellipticity on Riemannian manifolds
In the view of (2.8), 54
(
2
) =

(
2
)d
2

2
+ 2

2
A(, ) A(, ) A(, )
On the other hand by proposition 2.2 we have for C
q
,

2
+
2

2
+
2
In particular, if [
q
and C
q
[
q
then d
2
< as also

2
< . In the view of theorem 2.1 this proves that W
q
.
Now
_
x
(
2
)dX =
_
X
d d(
2
)
We shall show that these integrals vanish.
Using Schwartz inequality, one checks that there exists a positive
constant C
2
such that
d
2
C
2
at each point of X. (2.10)
Hence, since

(t) is bounded it follows that

_
X
d(
2
)dX < .
We will prove now that
_
X
(
2
)dX < .
Letting C
3
= Sup

(t), we have, by (2.10),

_
X

(
2
)d
2

2
dX =
_
:(
2
2
)d
2

2
C

<
2
<C
1
dX C
3
_
d
2

2
C

<
2
<C
1
d
2

2
dX
C
2
Cc
3
_

2

2
C

<
2
<C
1
dX C
1
C
2
C
3
_
X

2
dX = C
1
C
2
C
3

2
Since W
q
, then by Lemma 2.3
2
< , and therefore 55
_
X

(
2
)d
2

2
dX <
6. A maximum principle 49
Since K is compact, there is a constant C
4
> 0 such that
A(7u, u) C
4
A(u, u)
for any u C
q
and at each point of 7. Then we have
_
X
A(K, )dX =
_
X7
A(7, )dX +
_
7
A(7, )dX

_
X
A(7, )dX + 2
_
7
A(7, )dX
(7, ) + 2C
2
2C
2
+ d
2
+
2
< .
Finally we have
_
X
A(, )dX
1
2
_
X
A(, )dX +
1
2
_
X
A(, )dX < .
Then, it follows from (2.9) and from the fact that

and

are
bounded, that
_
X
(
2
)dX < .
Hence, by Lemma 2.4,
_
X
(
2
)dX = 0, i.e. by (2.9)
_
X

(
2
)
2
(d
2
)
2
dX + 2
_
X

(
2
).A(7, )dX + 2
_
X

(
2
)
2
dX =
_
X
2

(
2
)A(, )dX. (2.11)
Let us consider now sup on Supp f k. If it is not nite there is 56
nothing to prove. If it is nite, we set C
0
= Sup
2
on K
_
Support f .
In view of our choice of the function all the terms on the left hand side
of (2.11) are non negative. We obtain
0 2
_
X

(
2
)
2
dX
_
X
2

(
2
)A(, )
50 2. W-ellipticity on Riemannian manifolds
= 2
_
X

(
2
).A( f , ).
Once again since
2
C
0
on support f , the integral on the right is
zero. Hence
_
X

(
2
)
2
dX = 0
By consequences we have also d
2
= 0, i.e. is a constant, on the
open set
2
> C
o
. Hence
2
C
o
on X. This concludes the proof.
Corollary. If [
q
C
q
and D
q
(in particular, if = 0), then
is bounded on X.
7 Finite dimensionality of spaces of harmonic forms
Let be a real number and let M
q

be the vector space

M
q

= [
q
, = .
we set M
q
= H
q
Lemma 2.5. M

W
q
and 0.
Proof. If M
q

, then [
q
. Thus, by the rst part of Proposition
2.3, W
q
. Since is a solution of the elliptic equation , we may 57
assume C
q
. We have moreover (since the metric is complete)
jdj
2
jdj
2
+ jdj
2
(
2
+ 1)jdj
2
,
jdj
2
jj
2
+ jj
2
(
2
+ 1)jj
2
.
Hence d W
q+1
, W
q1
and therefore
jj
2
= (, ) = jdj
2
+ jj
2
(2.12)
This proves that 0
It follows from (2.12) that, if = 0, i.e. if H
q
, then d = 0,
= 0.
7. Finite dimensionality of spaces of harmonic forms 51
Theorem 2.3. Let X be a connected, oriented and complete Riemannian
manifold and assume that A(7, ) C
2
for all C
q
for some
C > 0, outside a compact set K X. Then for < C, M
q

is nite
dimensional. In particular dim H
q
< .
Proof. Let M
q

. Since W
q
, then by Lemma 2.3
jj
2
+ (7, ) = jdj
2
+ jj
2
and by (2.12)
jj
2
+ (7, ) = jdj
2
+ jj
2
= jj
2
Now 7 being compact, there is a constant C
2
> 0 such that (7u, u)
C
2
u
2
for every u C
q
and at each point of 7. Since (7, ) C
2

2
on the complement X 7 we have
jj
2
+ Cjj
2
X7
C
2
jj
2
7
+ jj
2
or again,
jj
2
+ Cjj
2
(C + C
2
)jj
2
7
+ jj
2
Hence, 58
(C )jj
2
(C + C
2
)jj
2
K
(2.13)
Suppose now that M
q

is innite dimensional for < C, then the evalu-

ation map
(x) (x K)
(which associates to a form , the element of the q
th
exterior power of
the tangent space at x which is dened by ) is a linear map into a nite
dimensional space. Hence the kernel of this map is of nite codimen-
sion. Since the nite intersection of subspaces of nite codimension
is non-zero (over an innite eld !) it follows that given any sequence
x
1
, . . . , x

, . . . of points of K, we can nd forms

M
q

such that

(x
i
) = 0 for i . We may further assume that j

j = 1. It follows
then that we can choose a subsequence
k
=

k
of

such that
k
converges weakly to a limit in [
q
. Now this implies that M
q

,
since for any D
q
,
(, ) = Lt(
k
, ) = Lt(
k
, ) = (, ).
52 2. W-ellipticity on Riemannian manifolds
Hence we can assume C
q
. Now the
k
are solutions of the elliptic
equation = 0. Hence the weak convergence of
k
implies that
k
converge uniformly together with all partial derivatives on every com-
pact set. Clearly we have Lt
k
(x

) = 0 for every . If we choose x

to
be dense in K then on K since C
q
. Hence it follows that

k
< on K.
for any given > 0 and all large k. We conclude that 59
j
k
j
2
K
0
as k . On the other hand since
k
M
q

, we have by (2.13),
(C )j
k
j
2
(C + C
2
)j
K
j
2
K
.
Since C > 0 and j
k
j
2
= 1, we arrive at contradiction. Hence
M
q

is nite dimensional for every < C. This proves Theorem 2.1

completely.
Theorem 2.4. Under the hypothesis of Theorem 2.3 we have the fol-
lowing. Let S
q
denote the orthogonal complement of H
q
in the Hilbert
space W
q
. Then there is a C > 0 such that
jj
2
Cjdj
2
+ jj
2
for W
q
For the proof of the theorem we need the following result due to
Rellich.
Lemma 2.6. Let X be a Riemannian manifold and X Let

[
q
() be a sequence such that
j

j
2
< and j

j
2

<
(We denote
_

2
dX by j

j
2

). Suppose that

M for a xed constant M > 0. Then if is smooth, we can nd a
subsequence
k
=

k
of

such that
k

k

0 as k, k

.
If

[
q
2
(X) is a sequence such that j

j
2
+ j

j
2
M for a 60
xed constant M > 0, then we can nd a subsequence
k
=

k
such
that for each compact K

X,
j
k

k
j
K
0 as k, k

.
7. Finite dimensionality of spaces of harmonic forms 53
Proof. The set being relatively compact, it is sucient to prove the
lemma in the case q = 0, i.e. in the case of L
2
functions. For this proof
see e.g. [7], 339.
Proof of Theorem 2.4. Suppose that the theorem is false. Then we can
nd a sequence

W
q
such that
j d

j
2
+ j

j
2
+j

j
2
= 1
while jd

j + j

j
2

j
2
. In view of Remark under Lemma 2.3
this implies that
j

j
2
C
1
for some C
1
> 0. By Rellichs Lemma (Lemma 2.6) we can nd a [
q
(by passing of a subsequence if necessary) such that
j

0
for every relatively compact X.
In particular, we have
j

j
2
K
0.
We assert that is harmonic, that is = 0. In fact for u 1
q
, we 61
have
(, u) = Lt(

, u)
= Lt(d

, du) + (

, u)
Lt(jd

j jduj + j

j.juj)
lim
1

2
j

j(jduj + +juj) = 0
since j

j
2
is bounded. Hence W
q
H
q
.
Now we have for f W
q
,
jf j
2
+ (7 f , f ) = jdfj
2
+ j f j
2
.
54 2. W-ellipticity on Riemannian manifolds
We have (7 f , f ) C f
2
on X K . On the other hand there is a
C
2
0 such that (7 f , f ) > C
2
f
2
on K. Since jf j
2
o, we obtain,
Cj f j
2
(C + C
2
)j f j
2
k
+ jd f j
2
+ j f j
2
.
Setting f =

, we obtain,
Cj

j
2
(C + C
2
)j

j
2
k
+ jd(

)j
2
+ j(

)j
2
.
Now H
q
C
q
so that
(d, d) + (, ) = ( , ) = 0.
Hence
Cj

j
2
(C + C
2
)j

2
k
+ d

2
+

2
(C + C
2
)

2
k
+
1

j
2
.
It follows that

in [
q
on the whole of X. On the other hand 62
since
jd

j
2
+

2
d

and

converge to zero. Hence

converges in W
q
to a limit
satisfying darphi = 0, = 0, i.e. H
q
. But

S
q
and S
q
is a closed
subspace. Hence S
q
and therefore = 0. But then

0 in W
q
, and
this is absurd, since

2
+ d

2
+

2
= 1. It follows that there
exists C

> 0 such that for S

q
C

d
2
+
2

2
Remark. Since for H
q
, d = 0, = 0, S
q
is also the orthogonal
complement of H
q
with respect to the scalar product on W
q
induced by
[
q
.
The hypothesis of Theorem 2.3 can be weakened, further the space
H
q
.
7. Finite dimensionality of spaces of harmonic forms 55
Theorem 2.5. Let X be an oriented, connected and complete Rieman-
nian manifold and let A(7, ) 0 outside a compact set 7 for every
C
q
. Then dimH
q
< .
Proof. If dimH
q
= , then given any dense sequence x

of points
x

X, we can nd a sequence

of forms

H
q
such that

(x
i
) = 0 for i ,

2
= 1.
We assert now that

> 0 for any relatively compact open subset

K. In fact, we have

2
+ (7

) = d

2
+

2
= o since

H
q
.
Hence from the Remark under Lemma 2.3 it follows that there is a 63
constant C > 0 such that
j

j
2
< C j

j
2
K
(2.14)
Hence if j

= 0, then j

j
2
K
= 0 so that

= 0. But in the
case,

2
is a constant. Hence j

> 0, a contradiction.
We may therefore assume that for a xed K j

= 1. From
(2.14),j

j
2

+ j

j
2

< C

for some C

> 0. Hence by Rellichs

Lemma (Lemma 2.6), we can, by passing to a subsequence, if necessary,
assume that j

j
2

0 as , provided that is smooth.

Hence

converges to a limit in [
q
().
But since

is a sequence of solutions of an elliptic operator,

converges to uniformly on every compact subset of . Since

(x
i
) = 0 for i , we see that 0 on . Hence j

0 as
, a contradiction. It follows that H
q
is nite dimensional. Hence
the theorem. An estimate for the dimension of H
q
is provided by the
following.
Proposition 2.3. Assume that, under the same hypotheses for X, 7 0
on X. Then for H
q
, = 0. Hence, dimH
q

_
n
q
_
. Moreover, if
dimH
q
> 0, then vol X < .
56 2. W-ellipticity on Riemannian manifolds
Proof. Since now 64
j j
2
= jdj
2
+ jj
2
= 0 for H
q
we have = 0 for H
q
. A form invariant under parallel translation
is determined by its value at one point. Hence dim H
q

_
n
q
_
. If H
q
is non-zero, then jj
2
< ; on the other hand since = 0,
2
is a
constant and the last assertion follows.
Lemma 2.7. Under the hypotheses of Theorems 2.3 and 2.4, for every
sS
q
there exists a unique xS
q
such that
s = x
in the sense of distributions. Moreover
j x j C

j s j,
jdx
2
+ xj
2
C

j s j
2
C

being the constant which appears in Theorem 2.4. If s C

q
S
q
, then
(x can be modied on a null set in sch a way that) xC
q
S
q
and the
equation x = s holds in the ordinary sense.
Proof. S
q
as a (closed) subspace of the Hilbert space W
q
is a Hilbert
space. By theorem 2.4 the norm N is equivalent on S
q
to the norm
(d
2
+
2
)
1
2
. f (s, f ) is a continuous linear form on S
q
. Thus there
is a unique x S
q
such that
(s, f ) = (dx, d f ) + (x, f )for all f S
q
.
Let now be any element of W
q
and let h and be its orthogonal
projections into H
q
and S
q
. They are uniquely dened, and furthermore
= h +
Since h is orthogonal to s and dh = 0, h = 0, then 65
(s, ) = (s, ) = (dx, d) + (x, ) = (dx, d) + (x, ).
This proves the lemma.
8. Orthogonal decomposition in [
q
57
Since
W
q
= H
q
S
q
we can state the following.
Proposition 2.4. Under the hypotheses of Theorem 2.3, for every W
q
there exists a unique hH
q
and a unique xS
q
such that
= h + x
(in the sense of distributions). Moreover
x C

dx
2
+ x
2
C

2
8 Orthogonal decomposition in [
q
Let X be an oriented, complete and connected Riemannian manifold.
Assume that there exists a compact set K X and a positive constant C
such that
A(7, ) CA(, )
outside K, for each C
q
.
The space [
q
can be decomposed as direct sum of orthogonal sub-
spaces
[
q
= H
q
dD
q1
D
q+1
,
where dD
q1
and D
q+1
are the closure of d D
q1
and [1
q+1
with 66
respect to the norm j j.
Let [
q
; - if d = 0, then the orthogonal projection of into
1
q+1
is zero; if = 0, then the orthogonal projection into d1
q1
is
zero (see [17] 602-605; [29], 165).
Lemma 2.8. For every d1
q1
there is a unique form x S
q
such
that
= dx, dx = 0.
58 2. W-ellipticity on Riemannian manifolds
Moreover
j x j C

j j,
j x j
2
C

j j
2
(C

being the constant introduced in Theorem 2.4)

Proof.
s (, s)
is a continuous form on S
q
. Then there exists a unique x S
q
such that
(, s) = (dx, ds) + (x, s) for all s S
q
.
Moreover, by Theorem 2.4,
j x j
2
C

(j dx j
2
+ j x j
2
) = C

(, x) C

j j . j x j,
j dx j
2
+ j x j
2
= (, x) j j . j x j C

j j
2
.
Let now u 1
q
u can be written
u = h + s, h H
q
, s S
q
.
Since h d1
q1
in [
q
and dh = 0, h = 0, then (, u) = (, s) = 67
(dx, ds) + (x, s) = (dx, du) + (x, u) = (x, u) i.e.
= x
in the sense of distributions.
Let C
q
d1
q1
. By the regularity theorem, we can modify x on
a null set in such a way that x C
q
S
q
, and that the latter equation
holds in the ordinary sense. Thus, by proposition 2.2
j dx j
2

j d j
2
+ j dx j
2
= j dx j
2
for all > 0.
Hence
dx = 0, = dx.
8. Orthogonal decomposition in [
q
59
But then, by Theorem 2.1, x W
q1
, and therefore
j dx j
2
= (x, dx) = 0 i.e. dx = 0.
If is not C

, there exists a sequence u

of forms u

1
q1
such
that j du

j 0. Clearly du

H
q
; hence du

S
q
. Setting

= du

and applying to

the above argument, we can nd, for each

, a unique x

S
q
C
q
such that

= dx

, dx

= 0
j x

j C

j,
j x

j
2
C

j
2
Hence x

is a Cauchy sequence in S
q
. Let 68
x = lim x

.
Then
dx = 0,
j x j C

j j,
j x j
2
C

j j
2
,
and nally
= dx,
in the sense of distributions. This proves the lemma.
An analogous argument yields
Lemma 2.9. For any 1
q+1
there is a unique form y S
q
such that
= dy y = 0
Moreover
j y j C

j j
j dy j
2
C

j j
2
60 2. W-ellipticity on Riemannian manifolds
Theorem 2.6. Every current f [
q
can be decomposed as the sum of
three currents
f = h + d
1
+
1
with h H
q
,
1
W
q1
,
1
W
q+1
. Moreover
N(h)
2
=j h j
2
j f j
2
, N(
1
)
2
(C

+1) j j
2
, N(
1
)
2
(C

+1) j j
2
Proof. First of all any element f [
q
can be expressed in a unique way
as the sum
f = h + +
of three forms h H
q
, dD
q1
, D
q+1
69
f
2
= h
2
+
2
+
2
.
Next we apply to and Lemma 2.8 and 2.9 setting then
=
1
, = d
1
;

1
W
q1

1
W
q+1
.
Moreover
N(
1
)
2
=
1

2
+
2
(C

+ 1)
2
(C

+ 1) f
2
N(
1
)
2
=
1

2
+
2
(C

+ 1)
2
(C

+ 1) f
2

Remark. If the hypotheses of Theorem 2.3 are satised, not only for the
forms of degree q, but also for those degree q 1 and q + 1 then spaces
H
q1
, H
q+1
, S
q1
and S
q+1
can be introduced and one checks that

1
H
q1
, i.e.
1
S
q1

1
H
q+1
, i.e.
1
S
q+1
.
Chapter 3
Local expressions for and
the main inequality
9 Metrics and connections
We now go back to the case of holomorphic vector bundles on a complex 70
manifold.
Let x be a complex manifold and : E X a holomorphic vector
bundle. Let 1 = U
i

iI
be a covering of X such that : E X is
dened with respect to 1 by transition functions e
i j
: U
i
U
j
GL
(m, C) (E is of rank m).
Let (h
i
)
iI
be a hermitian metric along the bres of E: then h
i
are C

functions U
i
whose values are positive denite hermitian matrix such
that.
h
i
=
t
e
ji
h
j
e
ji
=
t
e
1
i j
h
j
e
1
i j
on U
i
U
j
.
Consider the 1-form l
i
= h
1
i
h
i
where is the exterior dierentia-
tion with respect holomorphic coordinates. We have for this family of
forms the following.
Lemma 3.1. If l
i
= h
1
i
h
i
, then l
i
are (1, 0) forms with values in M
m
(C)
and on U
i
U
j
we, have
l
i
= e
1
ji
l
j
e
ji
+ e
1
ji
e
ji
.
61
62 3. Local expressions for and the main inequality
In other words l
i
are the local representations of a connection of type 71
(1, 0) on the principal bundle associated to E.
Proof. We have
e
1
ji
l
j
e
ji
+ e
1
ji
e
ji
= e
1
ji
h
j
e
ji
+ e
1
ji
e
ji
Now h
i
=
t
e
ji
h
j
e
ji
, so that
h
1
i
h
i
= e
1
ji
h
1
j
t
e
1
ji
_

t
e
ji
h
j
e
ji
+
t
e
ji
h
j
e
ji
+
t
e
ji
h
j
e
ji
_
.
Since the e
i j
are holomorphic functions,

t
e
ji
= 0.
Hence h
1
i
h
i
= e
1
ji
h
1
j
h
j
e
ji
+ e
1
ji
e
i j
.
The principal bundles

: P X associated to E is dened, with
respect to the covering 1 = U
i
, by the transition function e
i j
acting
on the bre GL(m, C) as follows:
Z
i
= e
i j
Z
j
(Z
i
, Z
j
GL(m, C)).
The computations above shows that
Z
1
i
(l
i
Z
i
+ dZ
i
) = Z
1
j
(l
j
Z
j
+ dZ
j
) on

1
(U
i

(U
j
)
Hence
i
= Z
1
i
(l
i
Z
i
+dZ
i
) is the local representation on

1
(U
i
) of
a global 1-form with values in the Lie algebra of GL(m, C).
Let P

be the tangent space to P at a point P. Let Q

be 72
the subspace annihilated by . The family of these subspaces denes a
connection in P [26]. This proves the lemma.
The covariant derivation associated to the connection dened above
is a map
: (X, 7(E)) (X, 7(E

)
where for a vector-bundle F, 7(F) denotes the sheaf of germs of dier-
entiable sections of F, and

is the bundle of 1-forms with value in C.

9. Metrics and connections 63
(In the above we regard E as a dierentiable vector bundle). is dened
in an open set U
i
of the covering 1 as follows.
Let z
1
, . . . , z
n
be a system of local coordinates on U
i
. Since we are
given a local trivialisation over U
i
of E, a section over U
i
of E may
be regarded as function on X with values in C
m
(rank E = m). We then
have

() = ()
_

z

_
=
_

z

_
+ l
i
_

z

_
()
and

() = ()
_

z

_
=
_

z

_
.
It is easy to check that these local representation dened a global
map as above.
We consider next the curvature form of the connection.
Let be the curvature form of the connection form in the princi- 73
pal bundle P associated to E. The values of on a pair, u
1
, u
2
, of tangent
vector elds to P is given by the structure formula ([26], 34-35):
(u
1
, u
2
) = d(u
1
, u
2
) +
1
2
[(u
1
), (u
2
)].
Hence the component of type (1,1) of is given on
1
(U
i
) by

i
= Z
1
i
s
i
Z
i
,
where s
i
= l
i
.
We shall call s the curvature form of the (connection l
i
dened by the)
metric h
i
along the bers of E. On U
i
U
j
s
i
= e
i j
s
j
e
ji
.
Hence we have
Lemma 3.2. The curvature form of the connection dened above is
given in U
i
by the matrix valued 2-form
(s
a
t
)
i
= s
i
= l
i
Hence it is a form of type (1,1) with values in the adjoint bundle End
(E).
64 3. Local expressions for and the main inequality
The bundle

decomposes canonically as a direct sum

where

(resp.

) is the holomorphic tangent bundle (resp. anti holo- 74

morphic tangent bundle) of X. (The two bundles in the above decompo-
sition are regarded as dierentiable bundles. As a dierentiable bundle

(resp. (

), is simply the bundle of C

forms on X of type (1,0)

(resp. type (0,1)). The decomposition above gives a direct-sum repre-
sentations
7(

) 7(

) 7(

),
and for any holomorphic vector bundle E on X,
(X, 7(E

)) (X, 7(E

)) (X, 7(E

)).
Now if we are given a connection on E, it denes as we have re-
marked a covariant derivation
: (X, 7(E)) (X, 7(E

));
composing with the natural projections dened by the direct sum de-
composition above we obtain maps

: (X, 7(E)) (X, 7(E

))
and

: (X, 7(E)) (X, 7(E

)).
Remarks. (1) As it has been remarked at the end of 3, the metric h
i

on E induces a metric on the dual E

: this is simply given by the

family of positive denite matrix-valued functions
_
t
h
1
i
_
iI
. The
corresponding connection is given locally on U
i
by the form
t
l
i
. 75
It is also easy to see that the curvature form of this connection is
given by
t
s
i

iI
(2) Let

E denote the anti holomorphic bundle associated to E.

E
is the bundle with transition functions e
i j
. Then
t
h
i
=

h
i
dene a
metric on

E as well. In this case, the local forms

l
i
=

h
1
i

h
i
9. Metrics and connections 65
dene a connection of type (0,1). When we speak of a connection
on

E without further comment, it will always be of this connection
(of course, it is necessary to assume given a metric or at least a
connection on E.)
(3) Suppose that E and F are holomorphic vector bundles on X and
that we given hermitian metrics h
1
i

iI
and h
2
i

iI
on E and F re-
spectively. (We assume here, as we may, that E and F are dened
by means of transition function e
i j
f
i j
with respect to the same
covering 1 = U
i

iI
). Then the family of matrix valued functions
h
1
i
h
2
i
denes a metric h
i
along the bres of E F. We have
then
(h
i
) = (h
1
i
h
2
i
) = h
1
i
h
2
i
+ h
1
i
h
2
i
so that l
i
= h
1
i
h
i
= h
1
i
h
1
i
I
r
2
+ I
r
1
h
2
1
i
h
2
i
where r
1
(resp. r
2
) is the rank of E (resp. rank F) and for an 76
integer m, I
m
denotes the (m m) identity matrix. One sees im-
mediately then that the curvature form s
i
is given by the formula
s
i
= l
i
= s
1
i
I
r
2
+ I
r
1
s
2
i
.
It is clear that the above considerations can be carried over to the
case when one or both of E and F are anti holomorphic.
In the sequel, we call a connection on a (dierential) vector bundle
(with complex bres) a connection (resp. - connection) if the associ-
ated connection form is of type (1,0) (resp. type (0,1)). The connection
l
i
on a holomorphic bundle E dened with respect to a metric h
i
is then a
connection while that on the conjugate bundle E is a connection.
Now if h
i

iI
is a hermitian metric along the bres of E, then it
can be regarded as a section h of the bundle E

. We have also a
canonical connection on E

. We denote the covariant derivative on

section of E

again by :
: (X, 7(E

)) (X, 7(E

)).
66 3. Local expressions for and the main inequality
Proposition 3.1. h = 0
Proof. Let z
1
, .., z
n
be a coordinate system on U
1
. In this proof we write
l for l
i
and h for h
i
.
We have 77

h = (h)
_

z

_
=

h
ab

c
l
c
b
h
ac
=
x
h
ab

(h
1
)
cd

h
db
h
ac
=

h
ab

d
a

h
db
=

h
ab

h
ab
= 0.
Similarly one shows that

h = h(

z

) = 0.
Hence the proposition.
(Here h is regarded as a function from U
i
into
_
C
m
C
m
_

using the
given trivialisation of E U
i
. A vector t C
m
is an m-tuple denoted
(t
a
)
1am
and tC
n
is again an m-tuple but is denoted (t
a
)
1am
. Further
conventions are as follows: a vector in the dual of C
m
is an m-tuple but
with subscripts instead of super scripts. In other words we write
t =

t
a
e
a
for t C
m
t =

t
a
e
a
for t

C
m
t =

t
a
e

a
for t C
m

t =

t
a
e

a
for t C
m

where e
a
(resp. e
a
, e

a
, e

a
is the canonical (resp. conjugate of the
canonical, dual of the canonical, conjugate dual of the canonical) basis
of C
m
).
Proposition 3.2. Let z
1
, . . . , z
n
be a coordinate system on U
i
. Then in 78
U
i
, the curvature form can be written as
a
i
= (s
a
b

)
ab
d z

dz
,
9. Metrics and connections 67
where
(s
a
b

)
ab
=

(3.1)
Proof. Let t = (t
a
) be a section of the bundle over U
i
.
Then

t =

t = (

t
a
)
Hence
(

t)
a
=

t
a

b
l
a
b

t
b
where l =
_
l
a
b
dz

is the connection form.

t =

t + (l
a
b
)(t) =

t
a
+

b
l
a
b
t
b
so that

t =

t
a
+

l
a
b
t
b
+

b
l
a
b

t
b
=

t
a
+ s
a
b

t
b
+

b
l
a
b

t
b
.
Hence (

t)
a
=
_
b
s
a
b

t
b
.
This proves the proposition.
The equation (3.1) is known as the Bianchi Identity.
(We remark that s is a form of type (1,1).).
The following lemma is easy to prove. 79
Lemma 3.3. # = #.
Next we specialize the preceding considerations to the case E =
o
the homomorphic tangent bundle. Let z
1
, . . . , z
n
be a local coordinate
system in an open set U X. Let
g =

g

dz

dz

be the expression for the hermitian metric in this coordinate system. The
associated -connection is then given in U by the (1,0)- form

dz

68 3. Local expressions for and the main inequality

where C

g

g

z

.
(here, as is usual, g

is dened by (g

)(g

) = Identity).
That means that
_

_

z

_
=
_
C

= 0
(3.2)
where
: (X, 7()) (X, 7(
o

))
is the covariant derivation.
The - connection dened above is not in general symmetric: C

80
C

; in fact this connection is symmetric if and only if the hermitian

metric g is Kahler. We set then
S =

1
2
(C

)dz

dz

.
Clearly S is an alternating (2,0) form with values in the tangent bun-
dle
0
. It is called the torsion form of the - connection. Its vanishing
characterises the Kahler metrics.
A hermitian metric on the holomorphic tangent bundle denes a Rie-
mannian metric as well on X. We have corresponding to this metric a
Riemannian connection on X. The associated covariant derivative is a
map
D : (X7()) (X, 7(

)).
In particular, we have
D : (X7(
0
)) (X, 7(
0

)).
Denoting, as is usual, the Riemann-Christofel symbols by

,..
etc, one can prove easily that these are related to the C

as follows:

=
1
2
(C

+ C

)
9. Metrics and connections 69

r
S

Since

as a dierentiable vector bundle, we have a direct

sum decomposition
(X, 7(

)) (X, 7(

)) (X, 7(

))
and hence a natural projection 81
(X, 7(

)) (X, 7(

));
composing this projection with D and , we obtain two linear maps

: (X, 7(
o
)) (X, 7(
o

o
))
and D

: (X, 7(
o
)) (X, 7(
o

o
)).
If the metric is Kahler, these two maps coincide. In a similar way,
composing the natural projection
(X, 7(

)) (X, 7(

o
))
with and D we dene two linear maps,

: (X, 7
o
)) (X, 7(
o

o

))
D

: (X, 7
o
)) (X, 7(
o

o
))
which coincide if the hermitian metric of X is a Kahler metric.
Suppose now that we are given a vector bundle E on X and hermitian
metrics on X and along the bres of E. We then have canonical -
connections on the bundle E and
o
and a - connection on
o
. These
connections extend canonically to connections on each of the bundles
p

o

q

o
E
We denote by the covariant derivation in any of these bundles: 82
: (X, 7(
p

o

q

o
E)) (X, 7(
p

o

q

))
70 3. Local expressions for and the main inequality
Once again, composing with the natural projections, we can dene

: (X, 7(
p

o

q

E)) (X, 7(
p

o

q

o
E

o
)),

: (X, 7(
p

o
q

o
E)) (X, 7(
p

o
q

o
E E

o
),
Let s =
_
,
s
a
b
d z

dz

be the curvature form of the -connection

on E in a co-ordinate open set Y over which the bundle E is trivial, the
complex analytic coordinates in U being (z
1
, . . . , z
n
) : if (rank E)= m,
a, b run through 1 to m and for xed , , (s
a
b
) is an (m m)-matrix.
In the same coordinate neighborhood, let
L =

,
L

d z

dz

be the curvature form of the -connection on the holomorphic tangent

bundle
o
. For xed , , (L

) is an (n n)-matrix.
10 Local expressions for , and
We will now obtain a local expression for the operator
= +
in terms of

and the forms L and s above.

We adopt the following notation : for a p- tuple A = (
1
, . . .
p
), we 83
set dz
A
= dz

1
. . . dz

p
(resp. dz
A
= d z

1
. . . d z

p
). Then if for
C
pq(X,E)
, we set

_

z
1
, . . . ,

z
p
,

z
1
, . . . , . . .

z
q
_
=
a
A

B
we have =

A,B
1
p!q!

a
A

B
dz
A
d z
B
where A = (
1
, . . . ,
p
) and B = (
1
, . . . ,
q
) for a p-tuple A and a (q+1)-
tuple (
1
, . . . ,
q+1
) we have
(

)
a
A

1
. . .

q+1
= (1)
p
q+1

r=1
(1)
r1

a
A

1
...
r
...
q+1
.
10. Local expressions for , and 71
On the other hand

1
...

r...
q+1
=

a
A

1
...

r
...

q+1

ir
C

a
A

1
...

i1

i+1
...

r
...

q+1
=
r

a
A

1
...

r
...

q+1

ir
B

r
+ S

a
A

1
...

i1

i+1
...

r
...
q+1
where C

has been dened before and 84

B

=
1
2
(C

+ C

)
while S

=
1
2
(C

)
so that S =

dz

dz

is the torsion of the connection on

.
We have therefore
(

a
)
A

1
...

q+1
= (1)
p

(1)
r1

a
A

1
...

r
...

q+1
+ (1)
p

ir
(1)
r1

S

a
A

1
...()
1
...

r
...

q+1
.
Let 85
S : C
pq
(X, E) C
p,q+1
(X, E)
be the operator dened by
(S )
a
A

1
...

q+1
= (1)
p

ir
(1)
r1

S

a
A
1
...
i1

i+1

q1
(3.4)
and set

=

+ S, (3.5)
so that
(

)
a
A

1
...

q+1
= (1)
p

(1)
r1

1
...

r
...
q1
(3.6)
Let

= #
1

# : C
pq
(X, E) C
p,q1
(X, E) (3.7)
72 3. Local expressions for and the main inequality
and
T = #
1
S # : C
pq
(X, E) C
p,q1
(X, E) (3.8)
so that
=

+ T. (3.9)
Finally let
=

If the hermitian metric on X is a Kahler metric, then 86

, =

, =

.
For C
pq
(X, E)
_

_
a
AB

= (1)
p1

a
A

, (3.10)
so that, exactly as in the case of the Laplacian in Chapter 2, we have
_

_
a
AB
=

a
AB
+
q

r=1
(1)
r1
(

)
a
A

r
(3.11)
where

= g

,
and A = (
1
, ........
p
), B = (
1
, ......
q
), B

r
= (
1
, .......,

r
, ....,
q
).
In view of the Ricci identity, the summand of (3.11) can be ex-
pressed by
q

r=1
(1)
r1
(

)
a
A

r
= (

7)
a
AB (3.12)
where

7 is a mapping

7 : C
pq
(X, E) C
pq
(X, E),
which is linear over C

functions, whose local expression involves lin-

early (with integral coecients) only the coecients of the curvature 87
10. Local expressions for , and 73
forms, s and L, of E and

.
By Remark (3) after Lemma 3.2 we have
_

7
_
a
AB
=
q

r=1
(1)
r
s
a
b
r

b
A

r
+
_

7

_
a
AB
, (3.13)
where

involves only the curvature tensor of

, and is completely
independent of E.
Formula (3.11) can be also written as
(

)
a
AB
=

a
AB
+ (

7)
a
AB
. (3.14)
Now
= (

) = (

+ S )(

+ T) + (

+ T)(

+ S )
=

T + T

S + S

+ S T + TS.
It follows that
Lemma 3.4. For any C
pq
(X, E)
()
a
AB
= (

)
a
AB
+ (F
1
)
a
AB
+ (F
2

)
a
AB
+ (F
3

)
a
AB
where
F
1
: C
pq
(X, E) C
pq
(X, E),
F
2
: C
pq
(X, E

o
) C
pq
(X, E),
F
3
: C
pq
(X, E

o
) C
pq
(X, E),
are linear over C

functions. Their local expression involves the tensor 88

tensor and its rst derivatives.
If the metric on X is a Kahler metric, then F
1
0, F
2
0, F
3
0.
Applying proposition 3.1 to the hermitian metric on X we have that
= .
74 3. Local expressions for and the main inequality
Hence, by (3.14),

7
1

7 . (3.15)
If the metric on X is Kahler, this identity becomes

1
=

7
1

7 .
A direct computation shows that, in this case,

7
1

7 does not
depend on the curvature form of the Kahler metric. We have in fact (see
[2], 103)
((

7
1

7))
a
AB
=
q

i=1
(1)
i
s
a
b
i

b
AB

i
+
p

j=1
(1)
j
s
a
b
j

b
A

j

B
+ s
a
b

b
AB
where A

j
= (
1
, . . . ,
j
, ....,
p
).
Starting from this formula, it is easy to check that

7
1

7 = e(s) e(s),
where 89
e(s) : C
pq
(X, E) C
p+1,q+1
(X, E)
is the linear mapping locally dened by
(e(s))
i
=

1s
a
ib

b
,
and is the classical operator of the Kahler geometry (see e.g. [35],
42). Thus we have, in the Kahler case,

1
= e(s) e(s)
which, by (1.7), can also be written

E
#
1

E
# = e(s) e(s).
This formula, which was rst obtained in [4], 483, yields W-ellipti-
city conditions on K ahler manifolds (see [2], ).
11. The main inequality 75
If the metric on X is not Kahler, then it follows from Lemma 3.4 and
from (3.15) that

1
=

K
1

K +G
1
+ G
2

+ G
3

where
G
1
: C
pq
(X, E) C
pq
(X, E),
G
2
: C
pq
(X, E

) C
pq
(X, E),
G
3
: C
pq
(X, E

) C
pq
(X, E),
are linear over C

functions. Their local expressions involve the torsion

tensor and its rst covariant derivatives. If the hermitian metric is a 90
K ahler metric, then G
1
0, G
2
0, G
3
0.
11 The main inequality
We shall now establish an integral inequality which, under convenient
hypotheses, yields a sucient condition for the W
pq
- ellipticity of E.
Let C
pq
(X, E)(q > 0). Let and be two tangent vector elds
to X, dened by
= (

= h
ba

a
A

B

b

AB

= 0)
= (

= 0,

= h
ba

a
A

B
.
b

AB

)
We have ( being the complex dimension of X)
div =
2n

i=1
D
i

i
=
n

=1

,
S

,
where

= h
ba

a
A

B
.
b aB

+ h
ba

a
A

B
.
B

b

AB

= h
ba

a
A

B
.
b aB

+ h
ba

a
A

B
.
B

b

AB

+ h
ba
(

)
a
A

B
.
b

AB

.
76 3. Local expressions for and the main inequality
The last summand can be evaluated using the Ricci identity. We 91
have, by (3.12).
h
ba
(

)
a
A

B
.
b

AB
=
1
q
h
ba
(

K)
a
A

B

b

AB
= p!(q1)!A(

K, ).
Furthermore a direct computation, starting from (3.6), shows that
A(

,

) = A(

)
1
p!(q 1)!
h
ba

a
A

B
.

b

AB

.
Hence we have

= h
ba

a
A

B
.
b

AB

+ p!(q 1)!
_
A(

K, ) + A(

) A(

,

)
_
.
Now
div =
2n

i=1
D
i

i
=

,
S

= h
ba

a
A

B
.
b

AB

+ h
ba

a
A

B
.

b

AB

= h
ba

a
A

B
.
b

AB

+ p!(q 1)!A(

,

), by (3.10).
Thus 92
div div = p!(q 1)!
_
A(

K, ) + A(

)
A(

) A(

)
_
2
_
S

_
.
Let D
pq
(X, E). Then by Stokes theorem we have
j

j
2
+(

K, ) = j

j
2
+j

j
2
+
2
p!(q 1)!
_
X
_
S

_
dX.
(3.16)
Let S be the length of the torsion form. Applying lemma A of
Chapter 1 and the Schwartz inequality we get from (3.4), (3.5) and from
(3.8), (3.9) the following estimates
j

j
2
2(j
2
+ jS j
2
) 2jj
2
+ c
_
X
S
2
A(, )dX
11. The main inequality 77
= 2jj
2
+ +c( S
2
, ),
j

j
2
2(jj
2
+ jTj
2
) 2jj
2
+ c
_
X
S
2
A(, )dX
= 2jj
2
+ c( S
2
, ),
c being a universal positive constant (depending only on the dimension
of X).
Furthermore, again by the Schwartz inequality, we have 93
1
p!(q 1)!

1
p!(q 1)!
S j q S A(

)
1
2
A(, )
1
2

q
2
S
2
A(, ) +
q
2
A(

),
1
p!(q 1)!

1
p!(q 1)!
S j q S A(

)
1
2
A(, )
1
2
q

2
S
2
A(, ) +
q
2
A(

),
for any > 0.
Substituting these estimates in (3.16) we obtain for any > 0
_
1
q

_
j

j
2
+
_

K
_
2c + 2q +
qc

_
S
2
,
_
2jj
2
+ +2
_
1 +
q

_
jj
2
.
Setting, for instance = 2q , and
7 =

7
_
5
2
c + 4n
2
_
S
2
.Id : C
pq
(X, E) C
pq
(X, E) (3.17)
we obtain the following
Proposition 3.3. For any D
pq
(X, E) (q > 0) we have
1
2
j

j
2
+ (7, ) 3
_
jj
2
+ jj
2
_
.
78 3. Local expressions for and the main inequality
If the hermitian metric on X is a K ahler metric, the above proposition
can be considerably sharpened. In fact, in this case, S 0, = 0, 94

= . It follows from (3.16) that for every D

pq
(X, E) (q > 0) satises
the identity
j

j
2
+ (7, ) = jj
2
+ jj
2
(K ahler) (3.18)
where we have set 7 =

7. Also in the hermitian case the choice of the

numerical constants can be improved.
As a corollary to proposition 3.1 we have the following
Theorem 3.1. If there exists a positive constant C
2
such that
A(7, ) C
2
A(, )
for every C
pq
(X, E) (q > 0) and at each point of X, then E is W
p,q
-
elliptic.
Remark. Eect on 7 of a change of the hermitian metric on E.
Let : XR be any C

- function. We wish to examine how the

operator 7 above changes when we replace the hermitian metric h on E
by h

= e

h. (The hermitian metric on the base X remains undisturbed).

First, the curvature form s

of h

is given by
s

= s +

dz

dz

.I
m
,
where s, as above, is the curvature form of h; (z

)
1n
is a local coordi-
nate system in an open set U X on which a local trivialization of E is
assumed given; the symbol I
m
stands for the (m m) identity matrix. It
is immediate from (3.13) that
(7

)
a
AB
= (7
a
)
AB
+

i
(1)
i

2

a
AB

i
.
where 7

is the analogue of 7, dened now with respect to the new 95

hermitian metric h

.
Chapter 4
Vanishing Theorems
12 q-complete manifolds
Let X be a complex manifold and : X R a C

-function. The Levi 96

formL() of is the hermitian quadratic dierential form on X dened
as follows: let z
1
, ..., z
n
be a coordinate system on an open set U X;
then
L()
_

z

,

z

_
=

.
(L() is thus a hermitian form on the holomorphic tangent space).
Denition 4.1. A C

function : X R is strongly q-pseudo-convex

if the Levi form L() has at least (n q) positive eigen-values at every
point of X.
Denition 4.2. A complex manifold X is q-complete if there exists a C

function : X R such that

(i) is strongly q-pseudo-convex
(ii) for c R, the set
_
x x X, (x) < c
_
is relatively compact in X.
Lemma 4.1 (E. Calabi). Let H be a hermitian quadratic dierential
form on X and G a hermitian metric on X. Assume that H has at least p
79
80 4. Vanishing Theorems
positive eigen values. Let
1
(x), ...,
n
(x) be the eigen values of H (w.r.t.
G) at x in decreasing order:
r
(x)
r+1
(x)
Then given c
1
, c
2
> 0, G can be so chosen that 97
l
H
(x) = c
1

p
(x) + c
2
Inf(0,
n
(x)) > 0 for all x X.
Proof. Let G be any complete hermitian metric whatever. Let
1
(x), . . . ,

n
(x) be the eigen-values of H with reference to G arranged in de-
creasing order. We construct now a metric G on X whose eigen val-
ues are functions of
_

i
(x)
_
1in
as follows: let : X R be a C

function (we will impose conditions on letter); let U be a coordi-

nate open set in X with holomorphic coordinates (z
1
, ..., z
n
); in U, we
have G = G
U
dz

dz

so that (G
U
)

is a function whose values

are positive denite hermitian matrices; then the matrix valued function

G
U
= (

G
U
)

where

G =
_

G
U
dz

dz

in U is dened by

G
1
U
= G
1
U

=
r=0
(x)
r
(r + 1)!
(H
U
G
1
U
)
r
where H
U
is the matrix valued function (H
U
) dened by
H =

H
U
dz

d z

in U.
We now assert that

G
U
dene a global hermitian deerential form on
X and under a suitable choice of , it is positive denite. To see that

G
U
denes a global hermitian dierential form on X, we need only prove
the following. Let V be another coordinate open set with coordinates
complex (w
1
, ..., w
n
). Let J =
(z
1
, ..., z
n
)
(w
1
, ..., w
n
)
be the Jacobian matrix. As 98
before let G
V
= (G
V
) be dened by G =
_
G
V
dw

d w

in V. Then
if

G
V
is dened starting from G
V
as

G
U
from G
U
, we have
J

G
t
U
J =

G
V
.
12. q-complete manifolds 81
We have in fact, writing J

for
t

J
1
,
J

G
1
U
J
1
= G
1
V
J
_

r=0
(x)
1
(r + 1)!
(H
U
G
1
U
)
r
_

_
J
1
, since JG
U
t

J = G
V
.
It follows from the above that
J

G
1
U
J
1
= G
1
V

r=0
(x)
r
(r + 1)!
(JH
U
G
1
U
J
1
)
r
= G
1
V

r=0
(x)
r
(r = 1)!
(H
V
J

G
1
U
J
1
)
r
since JH
t
U
JH
V
. Hence we obtain
J

G
1
U
J
1
= G
1
V

r=0
((x))
r
(r + 1)!
(H
V
G
1
V
)
r
=

G
1
V
This proves that

G
U
denes on X a global hermitian dierential. We
next show that

G is positive denite. For this we look for the eigen-
values of

G with reference to G. TO compute these, we may assume, in
the above formula for

G
U
, that G
U
is the identity matrix
Then we have 99

G
U
=
+

r=0
(x)
r
(r + 1)!
H
r
U
It follows that the eigen values of

G
U
are
_

_
+

r=0
(x)
r
(r + 1)!

q
(x)
r
_

_
1qn
.
It is easily seen that these are all strictly greater than zero: this as-
sertion simply means this: f (t) =
e
t
1
t
=

_
r=0
t
r
(r+1)!
for t 0, f (0) = 1
(which is continuous in t) is everywhere greater than 0.
82 4. Vanishing Theorems
We will now look for conditions on such that

G satises our re-
quirements. From the formula for

G, we have
H
U

G
1
U
=
+

r=0
(x)
r
(r + 1)
(H
U
G
1
U
)
r+1
.
Now the eigen values of H with respect to

G (resp. G) are simply
those of the matrix H
U

G
1
U
(resp. H
U
G
1
U
), Hence these eigen-values

q
(X) of H with reference to

G are
f ((x),
q
(x))
where f (s, t) is the function on R
2
dened by
f (s, t) =
t

r=0
S
r
(r + 1)
t
r+1
Since
f (s,r)
t
= e
st
> 0 for any, f (s, t) is monotone increasing in t.
Hence we have 100

r
(x)
r+1
(x) for 1 r n 1.
Moreover f (s, t) t for s 0. Thus , if we choose (x) 0 for
every xX, then
q
(x)
q
(x) > 0.
The choice of (x) is now made as follows . Let , for every integer
> 0. B

=
_
x d(x, x
0
)
_
for some x
0
X, the distance being i the
metric G. The B

are then compact. Let b

= Inf
xB

(
p
(x)).
Then b
1
b
2
. . . b
+1
. . ..
Let b(x) be a C

function on X such that b(x) > 0 for x X and

b(x) < b

in B

B
1
. Then clearly b(x)
p
(x).
Finally let (x) be a C

function on X such that (x) d(x, x

0
), and
k >
_
C
2
C
1
b
1
be a real constant. Set (x) =
2ke
(x)
b
2
(x)
. We have then

q
(x) = f ((x),
p
(x)) =
p
(x) +
(x)
2!

p
(x)
2
+
12. q-complete manifolds 83
so that
p
(x)
ke
(x)
b
2
(x)

p
(x)
2
ke
(x)
k.
On the other hand,

n
(x) = f ((x),
n
(x)) =
1
(x)
_
e
(x)
n
(x)
1
_

1
(x)
=
b
2
(x)
2ke
(x)

b
2
1
k
,
C
1

p
(x) + C
2
Inf(0,
n
(x)) C
1
k C
2
b
2
1
k

1
k
(C
1
k
2
C
2
b
2
1
) > 0.
This proves the lemma.
Combining this with the Remark at the end of 2, we can assert 101
moreover that G can be chosen to be complete and satisfy the condition
of Lemma 4.1.
Let : X R be a strongly q-pseudoconvex function on X. As-
suming as a form H in the above lemma the Levi form L(), and taking
p = n q, c
1
= 1, c
2
= n, we obtain
Lemma 4.2. Let : X R be a strongly q-pseudoconvex function on
X. Then there exists a complete hermitian metric on X such that at each
point of X.
l
L()
(x) =
nq
(x) + n Inf (0,
n
(x)) > 0.
Let u C
rs
(X, E) and let
L()u, u =
1
p!q!
h
ba

u
a
A

u
b

AB

Lemma 4.3. If : X R is strongly q-pseudoconvex then for any

u C
rs
(X, E), with s q +1 the following inequality holds at any point
of X
L()u, u l
A(u,u)
L()
.
Proof. At any point x X, we have
L()u, u = h
ba

(x)u
AB

AB

84 4. Vanishing Theorems
Since 102
s q + 1, i.e s + n q n + 1,
then every s-tuple of indices contains at least one the indices 1, 2 . . . , n
q of the positive eigenvalues
1
(x), . . . ,
nq
(x). Hence
1
p!q!
h
ba
nq

=1
u
AB

AB

A(u, u),
and therefore
L()u, u
nq
(x)A(u, u) +
1
p!q!
_

nq+1
(x)u
Anq+1B

u
Anq+1B
1
+
n
(x)u
AnB
u

AnB

_

nq
(x)A(u, u) + Inf(o,
n
(x))A(u, u).
This proves our lemma.
Lemma 4.4. be a strongly q-pseudo convex function on X and a
positive real valued function on R such that

(t) 0,

(t) > 0.
Then
L(())(u, u) =

()L()(u, u) + (

)()

2
Hence
L(())(u, u)

()L()(u, u),
and l
L(())

()l
L().
The Lemma follows from a direct computation of the Levi form
L(()) and from the fact since

()

2
is positive semi denite,
the eigenvalues of L(()) are not than the corresponding eigenvalues 103
of

()L().
13. Holomorphic bundles over q-complete manifolds 85
Lemma 4.5. Let X be a hermitian manifold and a strongly q-pseudo
convex proper function such that at each x X
l
L()
(x) =
nq
(X) + n Inf(0,
n
(x)) > 0,
where
p
are the eigen values of L() with respect to the metric on
X arranged in decreasing order. Then given any continuous function
g : X R there exists a sequence a

1<
of real constants such that
for any C

function : R R satisfying (i)

(t) > 0 (ii)

(t) 0 and
(iii)

(t) a

for t < + 1, we have

l
L(())
(x) > g(x).
Proof. Since the sets K

= x (x) + 1 are compact and

l
L()
(x) > 0 for all x X, we can nd a

such that
a

l
L()
(x) > g(x) for x K

the lemma now follows from Lemma 4.4

13 Holomorphic bundles over q-complete
manifolds
Lemma 4.6. Let E be a holomorphic vector bundle on a q-complete
complex manifold. Let : X R be a strongly q-pseudo-convex C

function on X such that x(x) < c X for every c R. Let ds

2
be a
hermitian metric on X such that l
L()
(x) > 0 with respect to this metric.
Let h be any hermitian metric on X. Then there is a sequence (a

)
1<
of real constants such that the following property holds.
Let be any C

functions on R such that

(t) a

for t +1 104
and

(t) > 0,

(t) 0 for all t. Let 7 and (resp. 7

)
be the opera-
tors (linear over C

functions) from C
rs
(X, E) C
rs
(X, E) dened in
chapter 3. (See (3.12) and (3.17)) with respect to the hermitian metrics
ds
2
and h (resp. h = e
()h
. Then for C
r,s
(X, E) with s > q, we
have
A

(7

, ) e
()
A(, ).
86 4. Vanishing Theorems
Proof. From the Remark at the end of Chapter 3,
7

= 7

(1)
i

2
()

z
z

a
A

i
.
It follows that
A

(7

, ) = e
()
L(())(, ) + e
()
A(7, )
e
()
l
L(()
+ f A(, ),
Where f : X is a continuous function depending on 7. By
Lemma 4.4. (a

)
1<
can be so chosen that for any convex satisfying
(i)

(t) > 0, (ii)

(t) 0 (iii)

(t) > a

for t < +1,

we have
c
1
l
L(()
+ f 1.
Hence
A
()
(7

, ) e
()
A(, ).

Theorem 4.1. Let X be a q-complete manifold so that there exists : 105

X satisfying : (i) is strongly q-pseudoconvex and (ii), for c
, x (x) c is compact X. Let E be a holomorphic vector bundle
on X. Then there exists a complete hermitian metric ds
2
on X and a
hermitian metric h along the bres of E and real constants a

1<
and c > 0 such that for every C

function : satisfying
(i)

(t) > 0, (ii)

(t) 0 and (iii)

(t) a

for v t < +1
we have
3(jj

+ j

j
2

) jj
2

for every D
r,s
(X, E), s > q.
Hence E is W
rs
-elliptic for s > q, with respect to the metric e
()
h.
Proof. From proposition 3.1., we have
(7

, ) 3
_
jj
2

+ j

j
2
_
.
(The subscript means that the operators scalar product etc. are
dened with reference to the metric ds
2
on the base and e
()
h along
the bres of E). The theorems then follow from Lemma 4.2, Lemma 4.6
and the above inequality.
13. Holomorphic bundles over q-complete manifolds 87
Remark . If we choose a satisfying the conditions of Theorem 4.1,
then for any function : such that

(t) 0,

(t) 0 for
t , + again satises the conditions of the theorems.
Consequently if we denote the metric on the base by ds
2
and the
hermitian metric e
()
h by h

, with respect to (ds

2
, h

) we have the fol- 106

lowing properties for s > q:
1) There is a C

function : X R such that (x) 0

2) For every C

non decreasing convex function : R RE is W

rs

elliptic with respect to (ds

2
, e
()
h

)
3) The W
rs
ellipticity constant is independent of .
4) The set B
c
= x x X, (x) c is compact in X for every c R.
These conditions imply conditions C

1
, C

2
, C

3
, C

4
of 4.
From Theorem 4.1 and from Theorem 1.3 we have the following
corollary.
Theorem 4.2. Let X be a q-complete complex manifold and E a holo-
morphic vector bundle on X. Then for
s q + 1, r 0, H
s
(X,
r
(E)) = 0.
Proof. In fact, for C
rs
(X, E)(s > q), there is a suitable as in
Theorem 4.1 above (i.e : is a C

function such that

(t) >
0,

(t) 0 and

(t) > a

for t < + 1, a

1<
being chosen so
that Theorem 4.1 holds for this sequence of real constants) such that
jj
2

=
_
X
e
()
A(, )dX < .
On the other hand, Theorem 4.1. ensures us that with respect to the
metric e
()
h (h as in Theorem 4.1.) and the complete hermitian metric
ds
2
(as in Theorem 4.1.) on X, we have W
rs
ellipticity for s > q. We
deduce Theorem 4.2. from Theorem 1.3 of Chapter 1.
88 4. Vanishing Theorems
Theorem 4.3. The cohomology groups H
s
k
(X,
r
(E)) (cohomology with 107
compact supports) vanish for s n q 1 for any vector bundle E on
the q-complete complex manifold X. Moreover, H
pq
k
(X,
r
(E)) has a
structure of a separated topological vector space.
Proof. The theorem can be proved by applying Serres duality to Theo-
rem 4.2. It can also be given the following direct proof, which is based
on the results of 4.
Let us choose a metric h

along the bres of the dual bundle E

in
such a way that conditions 1), 2), 3), 4) of the Remark after Theorem
4.1. be satised (by E

). Corresponding to the metric e

()
h

on E

,
we choose the dual metric
t
(e
()
h

)
1
= e
()
t
h
1
on E. Let
C
rs
(X, E), with s n q 1. Then # C
nrns
(X, E

). Since
n s q + 1, # satises the inequality
A
E

,
(7
E

,
#, #) A
E

,
(#, #),
i.e. A
E

,
(7
E

,
#, #) A
E,
(, ).

Hence, for any D

rs
(X, E), with s n q 1 we have
jj
2
E,
3(j #j
2
E

,
+ j
E

,
#j
2
E

,
)
i.e. by (1.4), (1.5), (1.6),
jj
2
E,
3(jj
2
E,
+ j
E,
j
2
)
This inequality holds for any D
rs
(X, E), with s n q 1, and 108
for any C

, nondecreasing, convex function : .

Our theorem follows from Theorem 1.4. of 4 and from Theorem
1.8.
Let now X be a q-complete manifold and : X R
+
a strongly
q-pseudo convex function on X such that for
c , B
c
= xx X, (x) < c X. We set B
c
= Y for some c .
Lemma 4.7. B
c
is q-complete.
13. Holomorphic bundles over q-complete manifolds 89
Proof. Let be a C

, nondecreasing convex function on (, c) such

that t t , (t) < t

is relatively compact in (, c) for every

t

. Then the two conditions of Denitions 4.2 are satised by the

function (). This proves the lemma
We adopt the following notation
F
rs
(X, E) =
_
L
rs
loc
(X, E); = 0
_
.
Q
rs
(X, E) =
_
L
rs

(X, E) for positive integer , = 0

_
.
Finally : F
rs
(X, E) F
rs
(Y, E) is the restriction map. Clearly
Q
rs
(X, E) CF
rs
(X, E) and we have
Theorem 4.4. If s q, (Q
rs
(X, E)) is dense in F
rs
(Y, E). (The metric
on X and that along the bres are chosen as ds
2
and h respectively in
the Remark after Theorem 4.1.)
Proof. We denote Q
rs
(X, E) simply by Q. Let be a continuous linear
functional on F
rs
(Y, E) which vanishes on r(Q). It is sucient to prove
that vanishes on all of F
rs
(Y, E).
First, we extend to a continuous linear form on L
rs
loc
(Y, E) (Hahn- 109
Banach extension theorem). By the representation theorem it follows
that there is a L
rs
(Y, E) with compact support in Y such that (u) =
(, u) for every u L
rs
loc
(Y, E).
Let c

= sup(x) x . Support . Since Support is compact in

Y = B
c
, then
c

< c.
Now, let : be a real C

nondecreasing, convex function

such that
(t) = 0 for t < c

(t) = 0 for t < c

,
0

(t) 1.
Then, for t > c

,
(t) t c

t.
90 4. Vanishing Theorems
Let
u L
rs
()
(X, E) for some > 0, u = 0.
Then u Q, because
juj
2

=
_
X
e

A(u, u)dX
_
e
()
A(u, u)dX = juj
2
()
Thus we have
(, u) = ((u)) = 0
Now under these conditions, it follows from Theorem 1.6, taking 110
into account the remark after Theorem 4.1 that there exists X L
r,s+1
such that
(i) = X and (ii) support X
_
x(x) c

_
.
Let c

< c
1
< c and introduce a complete hermitian metric on Y
which coincides with the given one B
1
. With respect to the new metric,
L
rs
(Y, E), X L
r,s+1
(Y, E)
Now let u F
rs
(Y, E) be an arbitrary form such that u = 0. Then
there is a C

, nondecreasing function : (, c) such that (t) =

0 for t c
1
, and
u L
rs
()
(Y, E).
We have then
(, u) = (X, u) = (
()
X, u)
()
= (X, u)
()
= 0.
Hence the linear form vanishes on F
rs
(Y, E). This proves the the-
orem.
Corollary 1. Q
rs
(X, E) is dense in F
rs
(X, E) (in the topology of L
2

convergence on compact sets).

Proof. This is simply the case X = Y (note that in the proof of Theorem
4.4, the case c = is covered).
Corollary 2. (F
rs
(X, E)) is dense in (F
rs
(Y, E). 111
13. Holomorphic bundles over q-complete manifolds 91
Proof. In fact F
rs
(X, E).
Suppose now that X is a stein manifold. Then X is is 0-complete.
Hence we can nd a C

function : X R such that L() is positive

denite and such that for c R,
_
xx X, X < c
_
X. If then E
is a holomorphic vector bundle on X, we obtain from Corollary 1, the
following result.
Corollary 3. Let Q(X, E) denote the space of holomorphic sections of
E over the Stein manifold X. There exists a complete hermitian metric
on X and a hermitian metric along the bres of E such that the set
f f Q(X, E);
_
X
e

f
2
< for some integer is dense in Q(X, E)
in the topology of uniform convergence on compact sets.
In view of Corollary 1, to prove Corollary 3, we need only prove the
following
Lemma 4.8. Let D be a domain in C
n
and f
m
a sequence of holomor-
phic functions on D such that for every compact K D, f
m
converges
in L
2
(K) (= space of square summable functions on K). Then f
m
con-
verges uniformly on every compact set K.
Proof. Let K be a compact set contained in D. Then there is a constant
c = c(K) > 0 such that for every point z

K. The poly-disc P
z

= z

z
i
z
i
o
c of radius c is contained in D and
_
z

K
P
z
o
D. We
have then from the Cauchy-integral formula,
f (z

) =
1
(c
2
)
n
_
P
zo
f (z)dv. (dv = Lebesque measure in C
n
)
It follows on applying this to f
2
, that 112
f (z

)
2

1
volP
z

_
P
z
o
f
2
dv
1
volP
z

_
P
z
o
j f
2
K

where K

is the (compact) closure of

_
z

K
P
z
o
. The lemma is immediate
from the above inequality.
In the special case q = 0, Corollary 2 together with the lemma above
yields the following result.
92 4. Vanishing Theorems
Theorem 4.5. Let X be a Stein manifold and : X R a strongly
0-pseudo convex C

function on X such that for

c R,
_
X (X) < c
_
X. Then the pair (X, B
c
), c R is
a Runge pair. In other words, the set of holomorphic functions on B
c
which are restrictions of global holomorphic functions on X is dense in
the space of all holomorphic functions on B
c
in the topology of uniform
convergence on compact subsets (of B
c
).
14 Examples of q-complete manifolds
Let X be an m-complete manifold and f
1
, . . . , f
q+1
be any q+1 holomor-
phic functions on V. Let Z =
_
x f
i
(x) = 0 for 1 i q + 1
_
.
Assertion Y = V Z is (m + q)-complete.
Proof. Let : V R be a strongly m-pseudo-convex function V such
that for every c R,
B
c
=
_
x x V, (x) < c
_
X.
Because of the second condition, we can choose a C

function 113
: R R such that

(t) > 0,

(t) 0,
(t) > 0 for t R, (t) as t and further
() >
q+1

i=1
f
i
f
i
on Y.
Set = () log
q+1
_
i=1
f
i

f
i
. Since L(log
q+1
_
i=1
f
i

f
i
) is positive semi-
denite with at most q positive eigen-values, strongly (m+ q)-pseudo
convex. It remains to prove that
B

c
= X x Y, (x) < c y
14. Examples of q-complete manifolds 93
for every c R. Now (x) c implies that
()
q+1

i=1
log f
i

f
i
c
or again that
e
()
q+1
_
i=1
f
i
f
i
e
c
,
i,e.,
q+1

i=1
f
i
f
i
e
()c
> > 0.

Hence B
c
X U, where U is a neighborhood of Z. On the other
hand; since () >
q+1
_
i=1
f
i
f
i
, e
()
/() < e
c
, hence () < e
c
= C, on B

c
. 114
Hence B

c
B
c
. On the other hand since B

c
is closed in X Z in
X U, it is closed in X. Hence B

c
is compact. This concludes the proof
of our assertion.
In particular, if X is a Stein manifold, then Y is q-complete.
If Z is a complete intersection, then q+1 is the complex codimension
of the submanifold Z of X. Let now Z be any analytic subset of the Stein
manifold X, of complex codimension q + 1 at each point. To Y = X Z
q-complete?
If q = 0, and if Z is a divisor, then the answer to this question is
positive i.e. Y = X Z is a Stein manifold as it was shown by Serre
([6] p. 50) R.R. Simha [31] has proved that, if q = 0, if X is a complex
codimension 2 and if Z is an analytic set in X of complex codimension
1 at each point, then Y = X Z is a Stein space. Examples ([23] Satz
12, 17; [14]) show that this result is false when dim
C
X > Z.
An example, due to G. Sorani and V. Villani [33] , shows that if
q > 0 then Y is not necessarily q-complete. Before discerning that
example, we shall establish the following
94 4. Vanishing Theorems
Theorem 4.6 ([32]). Let X be q-complete of complex dimension n. Let

n
denote the sheaf of germs of holomorphic n-forms on X. The natural
map
H
q
(X,
n
) H
n+q
(X, C)
is surjective.
Proof. For any complex manifold, we have spectral sequence (E
st
r
)
o
115
r such that E

is the associated graded group of H

(X, C), and

E
st
1
= H
t
(X,
s
) [11]. In the case when X is q-complete, H
t
(X,
s
) = 0
for t > q. In particular for t +s = n+q, E
st
t
= 0 for all (s, t) dierent from
(n, q). Hence E
s,t
r
= 0 for (s, t) (n, q), s + t = n + q. Moreover, all of
E
n,q
1
are cycles for d
1
since
n+1
= 0. Hence E
n,q
1
E
n,q
2
is surjective.
For r 2, E
nr,q+r1
r
= 0 since q + r 1 > q and E
n+r,q+r+1
= 0. Hence
E
n,q
2
E
n,q

. Hence H
n+q
(X, C) E
n,q

= E
n,q
2
. (We note that E
s,t

= 0 for
s + t = n + q, (s, t) (n, q)). This proves the theorem.
We consider now the subsets of C
2n
Z
1
=
_
z C
2n

z
i
= 0 for i > n
_
, X
1
= C
2n
Z
1
,
Z
2
=
_
z C
2n

z
i
= 0 for i n
_
, X
2
= C
2n
Z
2
,
Z = Z
1
Z
2
, Y = C
2n
Z = X
1
X
2
.
The analytic set Z C
2n
has complex codimension n at each point.
We shall prove that Y is not (n 1)-complete, when n 2.
Since the point 0 is a complete intersection of codimension 2n in
C
2n
, then U = C
2n
0 is (2n 1)-complete. Let
2n
be the sheaf of
group of holomorphic 2n-forms. Then by Theorem 4.6.
H
2n1
(U,
2n
) H
2n1
(U, C)
is surjective. On the other hand, since U is contractible on the unit 116
sphere of C
2n
, then H
2n1
(U, C) C. Hence
H
2n1
(U,
2n
) 0.
The exact cohomology sequence of Mayer-Victoris yields
15. A theorem on the supports of analytic functionals 95
H
r
(U,
2n
) H
r
(X
1
,
2n
) H
r
(X
z
,
2n
)
H
r
(X
1
X
2
,
2n
) H
r+1
(U,
2n
) (4.1)
Since Z
i
(i = 1, 2) is a complete intersection of codimension n, X
i
is
(n 1)-complete.
Hence
H
r
(X
i
,
2n
) = 0 f or r n(i = 1, 2).
It follows from (4.1) that
H
r
(X
1
X
2
,
2n
) H
r+1
(U,
2n
) for r n.
Then
H
2n2
(X
1
X
2
,
2n
) 0.
If n 2, then 2n 2 n. This shows that X
1
X
2
is not (n 1)-
complete.
15 A theorem on the supports of analytic function-
als
We shall apply the methods developed above to obtain-following the
ideas of [16], 2.5 with minor technical changes - a generalization, due
to Martinean [23] of a theorem originating with P olya [27]. We rst
prove two propositions which are of independent interest.
Proposition 4.1. Let be a C

plurisubharmonic function on C
n
. Let 117
jj

denote the norm of a C

form of type (p, q) with respect to

the Euclidean metric on the base and the metric e

on the trivial line

bundle. Let = + 2 log(1 + z
2
). Then there is a constant C > 0 such
that for any C

form of type (p, q), q 1, with

= 0, jj

< ,
there is a C

form

of type (p, q1) with

= , j

Cjj

.
Proof. Consider the operator 7

with respect to the metric e

. We
have, since the Euclidean metric is at,
A

(7

u, u) = e

L()u, u
96 4. Vanishing Theorems
2e

(1 +

z
2
)
2
A(u, u) = 2(1 + z
2
)
2
AA

(u, u),
since L()u, u 2L(log(1 +z
2
))u, u 2(q +z
2
)
2
A(u, u). Hence
7

is positive denite with eigenvalues 2(1 + z

2
)
2
at the point z;
7
1

is also positive denite, and its eigenvalues are

1
2
(1 + z
2
)
2
. By
consequence we have
A

(u, v) A

(7
1

u, u)A

(7

v, v)

1
2
A

((1 + z
2
)
2
u, u)A

(7

v, v).
Hence, for forms u, v of type (p, q), we have
(u, v)

1
2
j(1 + z
2
)uj
2

(7

v, v)

, (4.2)
whenever the forms u, v are such that the norms on the right are nite.

On D
p,q
(C
n
), we introduce the norm M by 118
M(u)
2
= (7

u, u)

+ j

uj
2

+ j

uj
2

,
and let V
p,q
be the completing of D
p,q
with respect to this norm. Obvi-
ously V
pq
L
pq
loc
. For u D
p,q
, we have, by (3.18),
(7

u, u) j

uj
2

+ j

uj
2

.
Thus the norm M on D
p,q
is equivalent to that dened by
M

(u)
2
= j

uj
2

+ j

uj
2

.
Let L
p,q
loc
, and let jj

< . Then, for u D

p,q
, we have, by
(4.2) (dropping the factor
1
2
as we may)
(u, )

2
j(1 + z
2
)j
2

(u)
2
= jj
2

(u)
2
(4.3)
15. A theorem on the supports of analytic functionals 97
Thus, the linear form u (u, )

is continuous on V
p,q
so that
there exists a unique x V
p,q
with
(u, )

= (

u,

x)

+ (

u,

x)

= (

u, x)

, u D
p,q
.
Thus we have
=

x
and, moreover, by (4.3) 119
M

(x)
2
= (x, )

j j

(x),
so that j x j
2

+ j

x j
2

j j
2

.
Let us assume now that is C

and that = o. Then by Corol-

lary 1 to Theorem 1.2, we have, for any > 0,
j

x j

j j
2

+ j x j
2

= j x j

,
and letting 0, we obtain

x = 0,
so that

x =

x;
since, as proved above, j

x j

j j

, we obtain the proposition

on setting

x.
We point out that the above proposition depends on a weaker condi-
tion than W
p,q
-ellipticity, in fact, proposition 4.1 is a particular case of
the following general statement, which can be established by a similar
argument.
Proposition 4.2. Let X be a complex manifold endowed with a complete 120
hermitian metric. Let : E X be a holomorphic vector bundle on X
and let h be a hermitian metric along the bres of E. Assume that
A(7, ) > 0
98 4. Vanishing Theorems
at each point of X and for every C
pq
(X, E). Let (x) be the least
eigenvalue of the (positive denite) hermitian from A(7, )
x
(x X)
acting on the space of the (p, q)-form with values in E. Then (x) > 0.
If C
pq
(X, E), = 0 is such that
_
X
1
(x)
A(, )dX < ,
there exists a from C
p,q1
(X, E) such that
= ,
_
X
A(, )dX 3
_
X
1
(x)
A(, )dX.
Proposition 4.3. Let be a plurisubharmonic function on C
n
such that
there exists a constant C > 0 for which (z) (z

) C whenever
z z

1. Let V be a (complex) subspace of C

n
of codimension k,
and f a holomorphic function on V such that
_
f
2
e

d
V
v < (d
V
v =
Lebesgue measure on V). Then, there exits a holomorphic function F on
C
n
with FV = f and
_
C
n
F
2
e

(1 + z
2
)
3k
dv M
_
V
f
2
e

d
V
v,
where M is a constant independent of f ; here dv = d
C
n v is the Lebesgue 121
measure in C
n
Proof. We may clearly suppose that V has codimension 1 in C
n
. We
assume that V is the subspace z
n
= 0.
Then f is a holomorphic function in C
n
, depending only on z
1
, . . . ,
z
n1
. Let : R R be a C

function on R, such that

0 (t) 1,
(t) = 1 for t
1
4
(t) = 0 for t 1.
Let C
1
= Sup

d
dt

.
15. A theorem on the supports of analytic functionals 99
We will construct a function in such a way that
F(z
1
, . . . , z
n
) = (z
n

2
) f (z
1
, . . . , z
n1
) z
n
(z
1
, . . . , z
n
)
satises all the requirements of the lemma.
Let
= (z
n

2
) f ,
is a C

, -closed form, such that z

n

1
2
. Thus the form =
1
z
n

is a (0, 1)- from which is C

and -closed on C
n
, i.e. C
01
(C
n
, C).
Moreover
Supp
_
z = (z
1
, . . . , z
n
) z
n

1
2
1
_
.
Since 122
(z, ..., z
n1
, z
n
) (z, ..., z
n1
, 0) C for z
n
1,
then
_
z
n
<1
e

f
2
dv e
C
_
V
e

f
2
dv
V
On the other hand
jj
2

=
_
1
2
<z
n
<1
e

A(, )dv (2)

2
_
1
2
<z
n
<1
e

f
2
d C
2
_
V
e

f
2
dv
V
(C
2
= 4C
2
1
e
C
). In particular, jj
2

<
Assume now that is C

. By Proposition 4.1, there exists a C

function : C
n
C such that
=
_
C
n
e

(1+ z
2
)
n

2
dv jj
2

c
2
_
V
e

f
2
dv.
If is not C

, we introduce a regularising function, i.e. a C

func-
tion (z) = ( z ) 0, on C
n
, with support z 1, and such that
_
C
n
dv = 1. Given 0 < < 1, the function

(z) =
_
C
n
(z z

)(z

)dv(z

)
100 4. Vanishing Theorems
is a C

plurisubharmonic function on C
n
, such that

(z) (z) as 0.
Moreover 123

(z) (z)
_
C
n
(z

) (z z

) (z) d(z

) C,
and therefore

(z)

(z

) 3C for z z

.
Hence there exists a C

function

: C
n
C, such that

=
_

n
e

(1+ z
2
)
2

2
dv e
c
_
C
n
e

(1+ z
2
)
2

2
dv C
3
_
V
e

f
2
dv
V
.(C
3
= C
2
1
e
5c
)
Letting 0, we can nd a distribution such that
=
1
_
C
n
e

(1+ z
2
)
2

2
dv C
3
_
V
e

f
2
dv
V
.
Consider now the distribution
F = ( z
n

2
) f z
n
.
We have
F = f z
n
= 0.
Hence F is a holomorphic function C
n
.
Furthermore 124
_
C
n
e

(1+ z
2
)
3
F
2
dv
15. A theorem on the supports of analytic functionals 101
2
_
z
n
<
e

(1+ z
2
)
3
f
2
dv + 2
_
C
z
n

2
e

(1+ z
2
)
3

2
dv
2e
C
_
V
e

f
2
dv
V
+ 2
_
C
n
e

(1+ z
2
)
2

2
dv
C
4
_
V
e

f
2
dv
V
(C
4
= 2e
C
+ 2C
3
).
This proves our proposition.
Lemma 4.9. Let be as in Proposition 4.2. Let F : C
n
C be a
holomorphic function such that
_
C
e
2
(1+ z
2
)
N
F
2
dv = M
2
< ,
for some N 0. Then there exists a positive constant C
5
depending only
on C and on N such that
F(z) C
5
Me
(z)
(1+ z )
N
e
(z)
.
Proof. Let z
0
C
n
and let P(z
0
) be the poly-disc with centre z
0
and
radius 1. Then by Cauchys formula we have
F(z
0
)
2
=
1

n
_
P(z
0
)
F(z)
2
dv,
whence 125
F(z
o
)
2

n
_
P(z
o
)
F
2
d.
Furthermore
e
(z
o
)
e

(z) for z P
1
(z
o
)
and
1
(1+ z
o
)
2

_
1 + 2n
1+ z
_
2

(1 + 2n)
2
1+ z
2
f or z P
1
(z
o
)
102 4. Vanishing Theorems
In conclusion
e
2(z
o
)
(1+ z
o
)
2N
F(z
o
)
2

e
2
(1 + 2n)
2N

n
_
P(z
o
)
e
2
(1+ z
2
)
F
2
dv

e
2
(1 + 2n)
2N

n
M
2
.
This concludes the proof of our lemma.
Before going on to state the theorem of Martinean-p olya, we need
some denitions.
Let H = H (C
n
) denote the vector space of holomorphic function
on C
n
with the topology of compact convergence. An analytic func-
tional is a continuous linear functional on H is said to be supported
by a compact set K if for any open set U K there is a constant M
U
such that sort any f H we have
( f ) M
U
sup
zU
f (z) .
For z = (z
1
, . . . , z
n
), = (
1
, . . . ,
n
) C
n
, let (z, ) =
_
z
i

i
. For 126
C
n
, let e

H be dened by e

(z) = e
(z,)
. The Laplace transform
of is the holomorphic function

on C
n
dened by

() = (e

).
Let K be compact and H the function dened by H
K
() = sup
zK
Re
(z, ). Clearly H
K
is continuous, and hence, being the supremum of
plurisubharmonic functions, is itself plurisubharmonic. If K

denotes
the set z C
n
d(z, K) , we have H
K

() = H
K
() +
Theorem 4.7 (Martineao-Polya). In order that the analytic functional
be supported by the convex compact set K, it is necessary and sucient
that for every > 0, there exist a constant C

> 0 such that

() C

e
H
K
()+
.
15. A theorem on the supports of analytic functionals 103
Proof. Suppose that is supported by K. Then, by denition
() = (e

) C

sup
zK

(z) = C

e
sup
K
(z,)
C

e
H
K
()+
To prove the converse, we proceed as follows. H is a closed sub-
space of the space C of continuous function on C
n
; hence (Hahn - Ba-
nach) extends to a linear functional : C C, hence denes a
measure. It is clearly sucient to prove that any > 0, there exists
distribution with support in K

such that
( f ) = for all f H .
Let

be the space of distributions with compact support in C

n
. For 127

, let denote the Fourier transform of considered as a function

of 2n real variables u
1
, . . . , u
2n
:
(u
1
, . . . , u
2n
) = (e

u
),
where e

u
(z) = exp (i(u
1
Re z
1
+ u
2
Imz
1
+ . . . + u
2n
Imz
n
)).

Clearly has an extension to a holomorphic function on C

2n
; further
since linear combination of the function e

are dense in H , we have

(g) = (g) for all g H if and only if (e

) = (e

) for all , i.e. if

and only if

() =(i
1
,
i
, ..., i
n
,
n
). (4.4)
Therefore, because of the Paley-Wiener theorem, it is sucient to
construct an entire function on C
2n
satisfying (4.4), for which, we
have further,
(u) C

(1+ u )
N(u)
. (4.5)
for some N > 0; here (u) = sup
xK

(x
1
Imu
1
+ . . . + x
2n
Imu
2n
) and z
j
=
x
2j1
+ ix
2j
.
Consider the subspace V of C
2n
consisting of points (i
i
,
1
, ..., i

n
,
n
), where = (
1
, ...,
n
) C
n
. If u = (i
1
, ...,
n
), we have
(u) = H
K

() = H
K
() + .
104 4. Vanishing Theorems
Further, (u) is plurisubharmonic in C
2n
and there exists a constant 128
C > 0 such that (u) (u

) C for u u

1. Clearly

denes
a holomorphic function f on V if we set f (i
1
, ...,
n
) =

(
1
, ...,
n
).
Since for any > 0,

() C

e
H
K
()+
by hypothesis and (u) =
H
K
() + on V, we have
f (u) C

e
(u)+()u
on V.
If < , we clearly have therefore
_
V
f (u)
2
e
2(u)
d
V
v < .
By Proposition 4.2 there exists F, holomorphic in C
2n
such that
FV = f , and
_
C
2n
F(u)
2
e
2(u)
(1 + u
2
)
3n
dv < .
By Lemma 4.9 this implies that there exists a constant C > 0 such
that
F(u) M(1 + u)
3n
e
(u)
,
so that if we set (u) = F(u), the conditions (4.4) and (4.5) are satised.
This completes the proof of the theorem.
Bibliography
[1] A. Andreotti and H. Grauert: Th eor` ems de nitude pour la coho- 129
mogipdes espaces complexes, Bull. Soc. Math. Frances 90(1962),
193-259.
[2] A. Andreotti and E. Vesentini: Carleman estimates for the Laplace
Beltrami equation on comples manifoldes, instit. Heautes

Etudes
Sci., Publ. Math. 25.81-130: Erratum ibd, to appearss
[3] A. Andreotti and E. Vesentini Les th eor` emes fondaientaux de la
th eorie des espaces holomorhiquement complets, S eminaire de
G eometrie Di erentielle Vol IV (1962-63) 1-31; Paris, Secr etariat
Math emateque.
[4] S. Bochner: Tensor elds with nite bases, Ann. of Math.,
53(1951), 400-411.
[5] E. Calabi and E. Vesentini: On compact, locally symmetric Kahler
manifolds, Ann. of Math., 71(1960), 472-507.
[6] H. Cartan: S eminaire E.N.S., 1953-54; Paris, Secr etariat
Math ematique.
[7] H. Cartan: Vari et es analytiques complexes et cohomologie, Col-
loque sur les fonctiones de plusieurs Variables, Bruxelles, 1953;
41-55.
[8] J. Deny and J.L. Lions: Les espaces de B. Levi, Ann. Inst. Fourier.
5(1953-54), 305-370.
105
106 BIBLIOGRAPHY
[9] J. Dieudonn e and L. schwartz: La dualit e dans les espaces (F) et
(LF), Annales de I Institut Fourier, 1(1949), 61-101.
[10] K.O. Friedrichs: On the dierentiability of the solutions of linear
elliptic dierential equations, Comm. Pure Appl. Math., 6(1953),
299-325.
[11] K.O. Friedrichs: Dierential forms on Riemannian manifolds,
Comm. Pure. Appl. Math., 8(1955) 551-590.
[12] A. Frolicher: Relations between the cohomology groups of Dol-
beault and topological invariants, Proc. Nat. Acad. Sci. U.S.A.,
41(1955), 641-644.
[13] M.P. Ganey: A special Stokess theorem for complete riemannian 130
manifolds, Ann. of Math., 60(1954), 140-145.
[14] R. Godement: Topologie alg ebrioue et th eorie et theorie des fais-
ceaux, Hermann, Paris, 1958.
[15] H. Grauert and R. Remmert: Singularitaten komplexer Man-
nigtigkeiten and Riemannscher Gebiete, Math. Zeitschr.,
67(1957), 103-128.
[16] L. Hormander: L
2
estimates and existence theorems for the -
operator, Dienrential Analysis, OXford University Press, 1964,
65-79.
[17] L. Hormander: Existence theorems for the -operator by L
2
meth-
ods, Acta mathematica, 113 (1965), 89-152.
[18] K. Kodaira: Harmonic elds in Riemannian manifolds (General-
ized potential theory), Ann. of math., 50 (1959), 587-665.
[19] J.L. Koszul: Vari et es Kahl eriennes, S ao Paulo, 1957;
mimeographed notes.
[20] G. K othe: Topologische lineare R aume, Springer-Verlag, 1960.
BIBLIOGRAPHY 107
[21] A. Lichn erowicz: Th eorie globale des connexion et des groupes
dholonmie, Cremonese, Roma, 1955.
[22] A. Lichn erowicz: G eom etrie des groupes de transformationes,
Dunod, Paris, 1958.
[23] B. Malgrange: Lectures on the theory of functions of several com-
plex variables, Tata Institute of Fundamental Research, Bombay,
1958.
[24] A. Martineau: Sur les fonctionnelles anaytiques et la trans-
formation de Fourier-Borel, Journal dAnalyse Math ematique,
XI(1963), 1-164.
[25] T. Meis: Die Minimale Blmtterzhl der Konkretisierungen einer
kompakten Riemannschen Flmche Schriftenreihe des Mathema-
tischen Instituts der Universitmt munster, Heft 16, 1960.
[26] M.S. Narasimhan: A remark on curvature and the Dirichlet prob- 131
lem, Bull. Amer. Math. Soc., 65(1959), 363-364.
[27] K. Nomizu: Lie groups and dierential geometry, Publications of
the Mathematical Society of Japan, 1956.
[28] G. Polya: Untersuchungen uber L ucken and Singularit aten von
Potenzreihen, Math. Zeitschr., 29 (1929) 549-640.
[29] F. Rellich: Ein Satz uber mittlere Konvergenz, Nachr. Ges.
G ottingen (math.-phys. Kl.), 1930, 30-35.
[30] G. de Rham: Vari et es di erentiables, Hermann, Paris, 1955.
[31] L. Schwartz: Th eories des distributioes, Hermann, Paris; Tome I,
1957; Tome II, 1959.
[32] R.R. Simha: On the complement of a curve on a Stein space of
dimension two, Math. Zeitschr., 82(1963), 63-66.
[33] G. Sorani: Homologie des q-paires de Runge, Annali Scuola Nor-
male Superiore, Pisa, (3), 17 (1963), 319-322.
108 BIBLIOGRAPHY
[34] G. Sorani-and V. Vesentini: Spazi q-completi e coomologia,
Mimeographed notes, 1964.
[35] E. Vesentini: Sulla coomologia delle variet a dierenziaoili, Rend.
Seminario Matematico e sico di Milano, 34 (1964), 19-30.
[36] A. Weil: Introduction a l etude des vari et es k ahl eri ennes, Her-
mann, Paris, 1958.
[37] K. Yano and S. Bochner: Curvature and Betti numbers, Ann. of
Math. Studies, 32, Princeton, 1953.