1991 Mathematics Subject Classification. Primary 14J30, 14B15, 14C20; Secondary 32Q28.

- Abstract. We analyse the structure of the algebraic manifolds $Y$ of dimension 3 with ${H}^{i}(Y,{\Omega}_{Y}^{j})=0$ for all $j\ge 0$ , $i>0$ and ${h}^{0}(Y,{\mathcal{O}}_{Y})>1$ , by showing the deformation invariant of some open surfaces. Secondly, we show when a smooth threefold with nonconstant regular functions satisfies the vanishing Hodge cohomology. As an application of these theorems, we give the first example of non-affine, non-product threefolds $Y$ with this property by constructing a family of a certain type of open surfaces parametrized by the affine curve $\mathbb{C}-\left\{0\right\}$ such that the corresponding smooth completion $X$ has Kodaira dimension $-\infty $ and $D$ -dimension 1, where $D$ is the effective boundary divisor.

1 Introduction

The main problem in algebraic geometry is the classification problem([H1], page 55). Among algebraic varieties, affine varieties are basic, natural and important.

Even though we want to study non-affine varieties, we still need affine varieties since we can cover any variety by affine varieties and exam the local data then glue them together to get global information by cohomology. A fundamental theorem due to Serre says that a variety
$Y$
is affine if and only if
${H}^{i}(Y,\mathcal{\mathcal{F}})=0$
for any coherent sheaf
$\mathcal{\mathcal{F}}$
and any positive integer
$i$
. In analytic category, the corresponding important varieties are Stein varieties. And it is well-known that in both categories, regular or holomorphic forms are key tools to characterize the geometry and topology of the varieties. An old and natural question raised by J.-P. Serre [Se] is the following:

If
$Y$
is a complex manifold with
${H}^{i}(Y,{\Omega}_{Y}^{j})=0$
for all
$j\ge 0$
and
$i>0$
, then what is
$Y$
? Is Y Stein? Here
${\Omega}_{Y}^{j}$
is the sheaf of holomorphic
$j$
-forms and the cohomology is
$\stackrel{\u02c7}{\text{C}}$
ech cohomology. This is a very hard question. By Serre duality, we immediately see that
$Y$
is not compact. By a theorem of Yum-Tong Siu [Siu1], we know that for any analytic coherent sheaf
$\mathcal{\mathcal{F}}$
on
$Y$
,
${H}^{3}(Y,\mathcal{\mathcal{F}})=0$
. Then we know that
$Y$
contains no compact surfaces by a result of Francois Norguet and Yum-Tong Siu [NS]. In 1991, T.Peternell [P] proved that if
$Y$
is also holomorphically convex, i.e., for any discrete sequence
$\left\{{x}_{n}\right\}$
on
$Y$
, there is a holomorphic function
$h$
on
$Y$
such that
$h$
is unbounded on
$\left\{{x}_{n}\right\}$
:
${sup}_{n}\left|h\right({x}_{n}\left)\right|=\infty $
, then
$Y$
is Stein. In the same paper, Peternell asked that if
$Y$
is a smooth algebraic variety (scheme), is
$Y$
affine?

If dim
$Y=1$
, then
$Y$
is affine [H2]. In 1993, N.Mohan Kumar answered Peternell's question and classified all such algebraic surfaces completely.

Theorem(Mohan Kumar) Let
$Y$
be a smooth algebraic surface over
$\mathbb{C}$
with
${H}^{i}(Y,{\Omega}_{Y}^{j})=0$
for all
$j\ge 0$
and
$i>0$
, then
$Y$
is one of the following (1)
$Y$
is affine.

(2) Let
$C$
be an elliptic curve and
$E$
the unique nonsplit extension of
$\mathcal{O}$
${}_{C}$
by itself. Let
$X={\mathbb{P}}_{C}\left(E\right)$
and
$D$
be the canonical section, then
$Y=X-D$
.

(3) Let
$X$
be a projective rational surface with an effective divisor
$D=-K$
with
${D}^{2}=0$
,
$\mathcal{O}$
$\left(D\right){|}_{D}$
be nontorsion and the dual graph of
$D$
be
${\stackrel{~}{D}}_{8}$
or
${\stackrel{~}{E}}_{8}$
, then
$Y=X-D$
.

We know type (2) surface
$Y$
is Stein [H2]. It is the first Stein algebraic surface discovered by Serre which is not affine. Type (3) surface is a mystery. It is hard to check because the boundary divisor is not irreducible. We know some method when the divisor is irreducible [H2], [N], [Ued]. Surprisingly, to check the algebraic surface with vanishing Hodge cohomology to be Stein or not is very hard. T. Peternell proved that if
$Y$
is a nonsingular complex surface with a non-algebraic compactification, then
$Y$
is Stein [P] .

Let us focus on algebraic manifolds with dimension 3, i.e., irreducible nonsingular algebraic varieties with dimension 3 defined over
$\mathbb{C}$
. We want to know what
$Y$
is.

In our previous paper [Zh], we studied
$Y$
by looking at its smooth completion
$X$
and the fibre space from
$X$
to a smooth projective curve
$\overline{C}$
. If we can understand
$X$
and the boundary divisor
$D$
, then we can understand
$Y$
. Theorem(Zhang) If
${H}^{i}(Y,{\Omega}_{Y}^{j})=0$
for all
$j\ge 0$
and
$i>0$
and
${H}^{0}(Y,\mathcal{O}$
${}_{Y})\ne \mathbb{C}$
, then we have (1) There is a smooth projective curve
$\overline{C}$
, and a smooth affine curve
$C$
such that the following diagram commutes

where f is proper and surjective, every fibre of f over
$\overline{C}$
is connected, general fibre is smooth. Also general fibre of
$f{|}_{Y}$
is connected and smooth.

$$\begin{array}{ccc}Y& \hookrightarrow & X\end{array}$$ |

$$\begin{array}{ccc}\downarrow f{|}_{Y}& & \downarrow f\end{array}$$ |

$$\begin{array}{ccc}C& \hookrightarrow & \overline{C}\end{array}$$ |

(2) Every fibre
$S$
of
$f{|}_{Y}$
over
$C$
satisfies the same vanishing condition, i.e.,
${H}^{i}(S,{\Omega}_{S}^{j})=0$
.

We review two standard notions in complex algebraic geometry. Let
$X$
be an algebraic manifold and
$D$
be a divisor on
$X$
. Then associated to
$D$
we have a line bundle
${\mathcal{O}}_{X}\left(D\right)$
. If for all integers
$m>0$
we have
${H}^{0}(X,{\mathcal{O}}_{X}(mD\left)\right)=0$
, then we define the
$D$
-dimension of
$X$
, denoted by
$\kappa (D,X)$
, to be
$-\infty $
. If
${h}^{0}(X,{\mathcal{O}}_{X}(mD\left)\right)\ge 1$
for some
$m$
, choose a basis
$\{{f}_{0},{f}_{1},\cdot \cdot \cdot ,{f}_{n}\}$
of the linear space
${H}^{0}(X,{\mathcal{O}}_{X}(mD\left)\right)$
, it defines a rational map
${\Phi}_{mD}$
from
$X$
to the projective space
${\mathbb{P}}^{n}$
by sending a point
$x$
on
$X$
to
$\left({f}_{0}\right(x),{f}_{1}(x),\cdot \cdot \cdot ,{f}_{n}(x\left)\right)$
in
${\mathbb{P}}^{n}$
. Then we define
$\kappa (D,X)$
to be the maximal dimension of the images of the rational map
${\Phi}_{mD}$
, i.e.,
$$\kappa (D,X)={max}_{m}\{dim({\Phi}_{mD}\left(X\right)\left)\right\}.$$
Let
${K}_{X}$
be the canonical divisor of
$X$
, then the Kodaira dimension of
$X$
is the
${K}_{X}$
-dimension of
$X$
, denoted by
$\kappa \left(X\right)$
, i.e.,
$$\kappa \left(X\right)=\kappa ({K}_{X},X).$$

Theorem 1.1.
If
$Y$
is an algebraic manifold of dimension 3 with
${H}^{i}(Y,{\Omega}_{Y}^{j})=0$
for all
$j\ge 0$
and
$i>0$
and
${h}^{0}(Y,{\mathcal{O}}_{Y})>1$
, then in the diagram of the above theorem, we can not have mixed types of smooth fibres, i.e., all smooth fibres must be exactly one of the three kinds of surfaces in Mohan Kumar's above classification. Moreover, if one fibre is not affine, then
$X$
has Kodaira dimension
$-\infty $
and
$D$
-dimension 1.

Theorem 1.2.
Under the condition of the above theorem, if one smooth fibre
${S}_{0}$
of
$f{|}_{Y}$
over
${t}_{0}\in C$
is affine, then by removing finitely many fibres
${S}_{1}$
,
${S}_{2}$
,
$\cdot ,\cdot ,\cdot $
,
${S}_{m}$
from Y, the new threefold
${Y}^{\prime}=Y-\cup {S}_{i}$
is affine.

Now let
$Y$
be a smooth algebraic threefold embedded in a smooth projective threefold
$X$
. Let
$D$
be the boundary divisor with simple normal crossings. Let
$C$
be a smooth affine curve imbedded in smooth projective curve
$\overline{C}$
. Let
${F}_{n}={\Omega}_{X}^{j}\otimes {\mathcal{O}}_{X}\left(nD\right)$
.
Theorem 1.3.
If we have the commutative diagram

such that f is proper and surjective and each fibre
${X}_{t}$
over
$t\in C$
is of type (2) projective surface, then
${R}^{i}{f}_{*}{F}_{n}{|}_{C}$
is locally free for all
$n\gg 0$
and
${H}^{i}(Y,{\Omega}_{Y}^{j})=0$
for all
$j\ge 0$
and
$i>0$
.
If we also assume that every horizontal divisor
${D}_{i}$
(i.e.,
$f\left({D}_{i}\right)=\overline{C}$
) cuts each smooth fibre
${X}_{t}={f}^{-1}\left(t\right)$
over
$t\in C$
with a unique prime divisor (irreducible curve) on
${X}_{t}$
, then for type (3) fibres, the theorem still holds.

$$\begin{array}{ccc}Y& \hookrightarrow & X\end{array}$$ |

$$\begin{array}{ccc}\downarrow f{|}_{Y}& & \downarrow f\end{array}$$ |

$$\begin{array}{ccc}C& \hookrightarrow & \overline{C}\end{array}$$ |

Theorem 1.4.
In the above commutative diagram, if each fibre
${X}_{t}$
over
$t\in C$
is of type (3) projective surface, then
${R}^{i}{f}_{*}{F}_{n}{|}_{C}$
is locally free for all
$n\gg 0$
and
${H}^{i}(Y,{\Omega}_{Y}^{j})=0$
for all
$j\ge 0$
and
$i>0$
.
As a consequence of above theorems, we have the following existence theorem.

Theorem 1.5.
There exist nonaffine and nonproduct threefolds
$Y$
with
${H}^{i}(Y,{\Omega}_{Y}^{j})=0$
for all
$j\ge 0$
and
$i>0$
.
We will prove these theorems in the following sections. The proof of Theorem 1.3 and Theorem 1.4 is similar. We will just prove Theorem 1.3. The hardest part is Section 2, the proof of Theorem 1.1. The deformation theory of smooth projective surfaces is well-known but in our case, the fibre is open. We will carefully analyse how the sheaves
${F}_{n}$
change when restricted to each fibre to obtain the deformation invariant of the open surfaces.

Question Are the threefolds
$Y$
Stein in Theorem 1.3 and Theorem 1.4? In 1953, Serre [Se] also posed the following problem: Is the total space of a holomorphic fibre bundle with Stein base
$Z$
and Stein fibre
$F$
a Stein manifold? In some cases, it is true (Siu, Sibony, Hirschowitz, Mok, Diederich and Fornass). On the other hand, Skoda gave the first counter example and Demailly, G. Coeu
$\text{e\u0155}$
, J.J. Loeb also gave counter examples [FrG]. In the example in Section 5, every fibre is Stein and the base curve is Stein but we do not know whether the threefold is Stein. Convention Unless otherwise explicitly mentioned, we always use Zariski topology, i.e., an open set means a Zariski open set.

2 Proof of Theorem 1.1

Theorem(Iitaka) Let
$X$
be a normal projective variety and let
$D$
be an effective divisor on
$X$
. Then there exist two positive numbers
$\alpha $
and
$\beta $
such that for all sufficiently large
$n$
we have
$$\alpha {n}^{\kappa (D,X)}\le {h}^{0}(X,{\mathcal{O}}_{X}(nD\left)\right)\le \beta {n}^{\kappa (D,X)}.$$
For the proof of Iitaka's theorem, see Lecture 3 [I1] or Theorem 8.1 [Uen] .

The following two lemmas are known [Ku] .

Lemma 2.1.
Let
$S$
be a smooth open surface with
${H}^{i}(S,{\Omega}_{S}^{j})=0$
for all
$j\ge 0$
and
$i>0$
. Let
$\overline{S}$
be a smooth projective surface containing
$S$
and
$G$
be the divisor in Mohan Kumar's Theorem in the introduction.

Then there are three cases.

(1) If
$S$
is affine, then
$\kappa (G,\overline{S})=2$
and
$${h}^{0}(\overline{S},{\mathcal{O}}_{\overline{S}}(nG\left)\right)\ge c{n}^{2}$$
for some positive constant
$c$
and
$n\gg 0$
.

(2) If
$S$
is of type (2), then
$${h}^{0}(\overline{S},{\mathcal{O}}_{\overline{S}}(nG\left)\right)={h}^{1}(\overline{S},{\mathcal{O}}_{\overline{S}}(nG\left)\right)=1,{h}^{2}(\overline{S},{\mathcal{O}}_{\overline{S}}(nG\left)\right)=0$$
for all
$n\gg 0$
.

(3) If
$S$
is of type (3), then
$${h}^{0}(\overline{S},{\mathcal{O}}_{\overline{S}}(nG\left)\right)=1,{h}^{1}(\overline{S},{\mathcal{O}}_{\overline{S}}(nG\left)\right)={h}^{2}(\overline{S},{\mathcal{O}}_{\overline{S}}(nG\left)\right)=0$$
for all
$n\gg 0$
.
$Proof.$
(1) Since
$S$
is affine,
$\overline{S}-S$
is support of an ample divisor
$A$
[H2] . So
$\kappa (A,\overline{S})=\kappa (G,\overline{S})=2$
[I6] . The estimate is obvious by Iitaka's Theorem. (2) By Lemma 1.8 [Ku] , we have
${h}^{0}(\overline{S},{\mathcal{O}}_{\overline{S}}(nG\left)\right)=1$
. By Serre duality and Lemma 2.2, we have
${h}^{2}(\overline{S},{\mathcal{O}}_{\overline{S}}(nG\left)\right)=0$
for all
$n>2$
. By Lemma 2.7 [Ku] , we know
${p}_{g}={h}^{2}(\overline{S},{\mathcal{O}}_{\overline{S}})=0$
and
$q={h}^{1}(\overline{S},{\mathcal{O}}_{\overline{S}})=1$
. By Lemma 2.6 [Ku] ,
$G\cdot {K}_{\overline{S}}=0$
. By Corollary 1.7 [Ku] ,
${G}^{2}=0$
. By Riemann-Roch formula, it is easy to get
${h}^{1}(\overline{S},{\mathcal{O}}_{\overline{S}}(nG\left)\right)=1$
.

(3)Again by Lemma 1.8 [Ku] , we have
${h}^{0}(\overline{S},{\mathcal{O}}_{\overline{S}}(nG\left)\right)=1$
. By the proof of Lemma 3.1 [Ku] , we have
${h}^{1}(\overline{S},{\mathcal{O}}_{\overline{S}}(nG\left)\right)=0$
. By Serre duality and Lemma 2.2, we have
${h}^{2}(\overline{S},{\mathcal{O}}_{\overline{S}}(nG\left)\right)=0$
for
$n\gg 0$
.

Q.E.D.

Lemma 2.2.
Let
$S$
,
$\overline{S}$
and
$G$
be as above then we have (1) If
$\overline{S}$
is of type (2), then
${G}^{2}=0$
,
${K}_{\overline{S}}=-2G$
,
${p}_{g}=0$
and
$q=1$
.

(2) If
$\overline{S}$
is of type (3), then
${G}^{2}=0$
,
${K}_{\overline{S}}=-G$
,
${p}_{g}=q=0$
.

$Proof.$
(1) This is standard result for the ruled surface over an elliptic curve.
The proof can be found in [H1] , Chapter V, Section 2 or [Ku] .

(2) By Lemma 1.6 [Ku] , we know
${G}^{2}=0$
. By Lemma 3.1 [Ku] ,
${K}_{\overline{S}}=-G$
. Since
$\overline{S}$
is rational,
$q=0$
. By Lemma 2.1, Serre duality and Riemann-Roch formula,
${p}_{g}=0$
.

Q.E.D.

Let
$f$
:
$X\to Z$
be a morphism between varieties (schemes) with
$Z$
connected.
Let
${z}_{0}\in Z$
,
$k\left({z}_{0}\right)=K$
, and
${X}_{{z}_{0}}\sim ={X}_{0}$
. Then the other fibres
${X}_{z}$
of
$f$
are called deformations of
${X}_{0}$
, [H1], page 89. In the proof of Theorem 1.1, the deformation of a nonsingular complex surface
${X}_{0}$
means the following by the same notation:

Both
$X$
and
$Z$
are smooth and
$f$
is surjective, proper and flat morphism (i.e.,
${\mathcal{O}}_{X,x}$
is a flat
${\mathcal{O}}_{Z,f\left(x\right)}$
-module for all
$x\in X$
) such that the fibre over
${z}_{0}\in Z$
,
${X}_{{z}_{0}}\sim ={X}_{0}$
, [BaPV]. By [I4], we know that the deformation of a rational surface is again rational. By theorem (8.1), Chapter VI, [BaPV], the deformation of a ruled surface over a smooth curve of genus
$g\ge 1$
is also of the same type, i.e., has the same minimal model.

We need Kodaira's stability of
$(-1)$
-curves. It is Theorem 5 in [Kod1].

Theorem(Kodaira) Let
$f$
:
$X\to Z$
be a surjective, proper holomorphic map which is flat. If for some point
$0\in Z$
the fibre
${X}_{0}$
contains a
$(-1)$
-curve
${E}_{0}$
, then there is an open neighborhood (in complex topology)
$U$
of 0 in
$Z$
and a closed and connected submanifold
$E$
of
${f}^{-1}\left(U\right)$
such that
$E\cap {X}_{0}={E}_{0}$
and such that
$E\cap {X}_{t}={E}_{t}$
is a
$(-1)$
-curve for every
$t\in U$
.

Further, in Kodaira's Theorem, there is a
$g$
:
${X}^{\prime}\to U$
, a surjective, flat, proper holomorphic map such that the following diagram commutes:

where
$h{|}_{{X}_{t}}:{X}_{t}\to {X}_{t}^{\prime}$
is the blowing down of
${E}_{t}$
. Let us state the contraction part precisely. The proof is due to Suwa [I5], Appendix 1.

$$\begin{array}{ccc}X& \stackrel{h}{\u27f6}& {X}^{\prime}\end{array}$$ |

$$\begin{array}{ccc}\downarrow f& & \downarrow g\end{array}$$ |

$$\begin{array}{ccc}U& \stackrel{\approx}{\u27f6}& U\end{array}$$ |

Theorem(Suwa) Let
$X$
and
$Z$
be complex manifolds, and let
$f$
be a proper, surjective and flat holomorphic map from
$X$
to
$Z$
, such that every fibre
${X}_{z}$
is a smooth surface. If there exists a complex submanifold
$E$
of
$X$
such that its restriction to
${X}_{z}$
:
${E}_{z}=E\cap {X}_{z}$
is an irreducible exceptional curve of the first kind on
${X}_{z}$
at any
$z\in Z$
, then we can construct a complex manifold
${X}^{\prime}$
, which is proper over
$Z$
, and a holomorphic map
$h$
:
$X\to {X}^{\prime}$
over
$Z$
, such that
$h{|}_{{X}_{z}}$
:

${X}_{z}\to {X}_{z}^{\prime}$
shrinks
${E}_{z}$
to a point in
${X}_{z}^{\prime}$
for every point
$z\in Z$
, and such that
$h{|}_{X-E}$
:

$X-E\to {X}^{\prime}-h\left(E\right)$
is biholomorphic.

Upper Semicontinuity Theorem (Grauert, Grothendieck) Let
$f$
:
$X\to Z$
be a proper morphism of noetherian schemes,
$\mathcal{\mathcal{F}}$
a coherent sheaf on
$X$
, flat over
$Z$
, then (1) The
$i$
-th direct image
${R}^{i}{f}_{*}\mathcal{\mathcal{F}}$
is a coherent sheaf on
$Z$
for any nonnegative integer
$i$
.

(2) Let
${\mathcal{\mathcal{F}}}_{z}=\mathcal{\mathcal{F}}{|}_{{X}_{z}}$
, i.e., the sheaf
$\mathcal{\mathcal{F}}$
restricted to the fibre
${X}_{z}={f}^{-1}\left(z\right)$
, then the function
$${d}_{i}\left(z\right)={h}^{i}({X}_{z},{\mathcal{\mathcal{F}}}_{z})={dim}_{k\left(z\right)}{H}^{i}({X}_{z},{\mathcal{\mathcal{F}}}_{z})$$
is upper semi-continuous on
$z$
. That is, for any
$n\in \mathbb{Z}$
, the set
$\{z\in Z:{d}_{i}(z)\ge n\}$
is a closed set, where
$k\left(z\right)={\mathcal{O}}_{z}/{\mathcal{\mathcal{M}}}_{z}$
, the residue field at the point
$z$
.

(3) The Euler characteristic of the restriction sheaf
${\mathcal{\mathcal{F}}}_{z}$
$$\chi \left({\mathcal{\mathcal{F}}}_{z}\right)=\sum (-1{)}^{i}{dim}_{k\left(z\right)}{H}^{i}({X}_{z},{\mathcal{\mathcal{F}}}_{z})$$
is locally constant on
$Z$
.

(4) The following statements are equivalent:

(
$\text{i}$
)
${h}^{i}({X}_{z},{\mathcal{\mathcal{F}}}_{z})$
is a constant function on
$Z$
, (
$\text{ii}$
)
${R}^{i}{f}_{*}\mathcal{\mathcal{F}}$
is locally free sheaf on
$Z$
, and for all
$z\in Z$
, the natural map
$${R}^{i}{f}_{*}\mathcal{\mathcal{F}}{\otimes}_{{\mathcal{O}}_{z}}k\left(z\right)\u27f6{H}^{i}({X}_{z},{\mathcal{\mathcal{F}}}_{z})$$
is an isomorphism.

In addition, if these conditions are satisfied, then
$${R}^{i-1}{f}_{*}\mathcal{\mathcal{F}}{\otimes}_{{\mathcal{O}}_{z}}k\left(z\right)\u27f6{H}^{i-1}({X}_{z},{\mathcal{\mathcal{F}}}_{z})$$
is an isomorphism for all
$z\in Z$
.

For a proof, see [Mu] , page 46-53. For the analytic statement and proof, see [GR] .

In this section, from now on, we assume that the condition of Theorem 1.1 holds.

Theorem 1.1 is a direct consequence of the following Lemmas.

Lemma 2.3.
If one smooth fibre
${S}_{{t}_{0}}={S}_{0}$
is of type (2) or (3) open surface in Mohan Kumar's classification, then there is an affine open set
$U$
such that
${S}_{t}={f}^{-1}\left(t\right)-D$
over every
$t\in U$
is of the same type.
$Proof.$
Note that
${S}_{0}$
is not affine. Let
${X}_{0}={f}^{-1}\left({t}_{0}\right)$
. By Lemma 2.1 and Lemma 5.3 in [U], even though the divisor
${D}_{0}=D{|}_{{X}_{0}}$
contains exceptional curves of the first kind, we still have
${H}^{0}({X}_{0},\mathcal{O}(n{D}_{0}\left)\right)=\mathbb{C}$
for all nonnegative integer
$n$
. Let
${D}_{t}=D{|}_{{X}_{t}}$
, then
${D}_{t}$
is a connected curve on
${X}_{t}={f}^{-1}\left(t\right)$
since
${X}_{t}$
is smooth and
${H}^{i}({S}_{t},{\Omega}_{{S}_{t}}^{j})=0$
, Lemma 1.4, [Ku] . By upper semi-continuity, there is an affine open set
$U$
in
$C$
such that
${H}^{0}({X}_{t},\mathcal{O}(n{D}_{t}\left)\right)=\mathbb{C}$
since every
${D}_{t}$
is effective. Therefore every fibre
${S}_{t}$
over
$t\in U$
is not affine.

Secondly, if
${S}_{0}$
is of type (2) open surface in Mohan Kumar's classification, then
${p}_{g}\left({X}_{0}\right)={h}^{2}\left({\mathcal{O}}_{{X}_{0}}\right)=0$
and
$q\left({X}_{0}\right)={h}^{1}\left({\mathcal{O}}_{{X}_{0}}\right)=1$
(Lemma 2.2). Here the boundary divisor
${D}_{0}=D{|}_{{X}_{0}}$
may contain exceptional curves so
${X}_{0}$
may not be minimal. But
${p}_{g}$
is birational invariant and
$q$
is bimeromorphic invariant [BaPV] , page 107. Since
${R}^{i}{f}_{*}\left({\mathcal{O}}_{X}\right)$
and
${R}^{i}{f}_{*}\left({\mathcal{O}}_{X}\right({K}_{X}\left)\right)$
are locally free for all
$i\ge 0$
[Kol1, Kol2] , again by upper semi-continuity,
${p}_{g}\left({X}_{t}\right)=0$
and
$q\left({X}_{t}\right)=1$
for every
$t\in C$
. Now
${X}_{0}$
has the minimal model of a ruled surface over an elliptic curve, by the classification theorem (1.1), page 243 and deformation theorem (8.1), page 263, Chapter VI [BaPV], there is an affine open set
$U$
such that for every
$t\in U$
,
${X}_{t}$
has the same minimal model as
${X}_{0}$
in Mohan Kumar's theorem, i.e., type (2) projective surface.

Similarly, if
${S}_{0}$
is of the third type, then there is an affine open set
$U$
such that
${S}_{t}$
over every
$t\in U$
is of the same type by [I4] and [BaPV] since the deformation of a rational surface is still rational.

Q.E.D.

Remark 2.4.
If
$S$
is of type (2) open surface in Mohan Kumar's Theorem, then any point on
$S$
can not be contained in any exceptional curve of
$\overline{S}$
, where
$\overline{S}$
is a smooth completion of
$S$
. If
$\overline{S}$
is not minimal, then all exceptional curves are contained in the boundary
$\overline{S}-S$
.

Lemma 2.5.
If there is an affine open set
$U$
in
$C$
such that for every
$t\in U$
,
$t\ne {t}_{0}$
,
${S}_{t}={f}^{-1}\left(t\right)-D$
is of type (2) open surface of Mohan Kumar's classification, where
${t}_{0}$
is a fixed point of
$U$
, then
${S}_{0}$
must be of the same type surface.
$Proof.$
First,
${S}_{0}$
can not be of type (3) since
${X}_{t}={f}^{-1}\left(t\right)$
is not rational and the deformation of rational surfaces is still rational [I4] . By Mohan Kumar and Zhang's theorems in the introduction, there are three possible smooth fibres. So we only need to prove that
${S}_{0}$
is not affine. It suffices to prove that
${h}^{0}({X}_{0},{\mathcal{O}}_{{X}_{0}}(n{D}_{0}\left)\right)$
is bounded for all
$n$
. In fact, in our case, it is 1. Here
${X}_{0}={f}^{-1}\left({t}_{0}\right)$
,
${D}_{0}=D{|}_{{X}_{0}}$
and
${S}_{0}={X}_{0}-{D}_{0}$
.

By Kodaira's Stability Theorem of
$(-1)$
-curves and Suwa's Theorem, we may assume that
${D}_{0}$
has no exceptional curve of the first kind. So there is a small open set
$V$
in
$C$
(complex topology), for all points
$t\in V$
,
${D}_{t}=D{|}_{{X}_{t}}$
has no exceptional curves of the first kind. In fact, if there is
${t}_{1}\in V$
,
${D}_{{t}_{1}}$
has a component
${E}_{1}$
, such that
${E}_{1}$
is an exceptional curve of the first kind, then
${E}_{1}^{2}={E}_{1}\cdot {K}_{{X}_{1}}=-1$
, where
${X}_{1}={f}^{-1}\left({t}_{1}\right)$
. There is a prime component
$G$
of
$D$
in
$X$
such that
${E}_{1}\subset G$
. Let
${E}_{t}=G{|}_{{X}_{t}}$
for
$t\in V$
, then by upper semi-continuity, the Euler characteristic of
${\mathcal{O}}_{{X}_{t}}\left(n{E}_{t}\right)$
is constant for every
$t\in V$
and every
$n\ge 0$
. So for any
$n\ge 0$
, we have
$$\chi \left({\mathcal{O}}_{{X}_{t}}\right(n{E}_{t}\left)\right)=\chi \left({\mathcal{O}}_{{X}_{0}}\right(n{E}_{0}\left)\right).$$
By Riemann-Roch formula, for all
$n\ge 0$
, we have
$$\frac{1}{2}{n}^{2}{E}_{t}^{2}-\frac{1}{2}n{E}_{t}\cdot {K}_{{X}_{t}}=\frac{1}{2}{n}^{2}{E}_{1}^{2}-\frac{1}{2}n{E}_{1}\cdot {K}_{{X}_{1}}.$$
So
${E}_{t}^{2}={E}_{t}\cdot {K}_{{X}_{t}}=-1$
for all
$t\in V$
. In particular,
${E}_{0}^{2}={E}_{0}\cdot {K}_{0}=-1$
. This is impossible since
${D}_{0}$
has no
$(-1)$
-curves by our assumption. Thus for all
$t\in V$
and
$t\ne {t}_{0}$
,
${X}_{t}$
is type (2) surface, i.e., a minimal ruled surface over an elliptic curve.

But
${D}_{t}$
may have multiple. Let
${D}_{t}^{\prime}$
be the elliptic curve (a section) as in Mohan Kumar's classification, then there is a positive integer
$n\left(t\right)$
, depending on
$t$
such that
${D}_{t}=n\left(t\right){D}_{t}^{\prime}$
. Since the function
$n\left(t\right)$
is discrete, there is a dense subset
$B$
in
$V$
such that
$n\left(t\right)$
is a constant
$c$
for all
$t\in B$
. Let
${t}_{1}\in V-B$
. Considering divisor
$D+c{K}_{X}$
restricted to the corresponding fibre
${X}_{1}={f}^{-1}\left({t}_{1}\right)$
, by upper semi-continuity, we have
$${h}^{0}({X}_{{t}_{1}},{\mathcal{O}}_{{X}_{{t}_{1}}}({D}_{{t}_{1}}+2c{K}_{1}\left)\right)\ge {h}^{0}({X}_{t},{\mathcal{O}}_{{X}_{t}}({D}_{t}+2c{K}_{t}\left)\right)=1,$$
and
$${h}^{0}({X}_{{t}_{1}},{\mathcal{O}}_{{X}_{{t}_{1}}}(-{D}_{{t}_{1}}-2c{K}_{1}\left)\right)\ge {h}^{0}({X}_{t},{\mathcal{O}}_{{X}_{t}}(-{D}_{t}-2c{K}_{t}\left)\right)=1,$$
where
$t\in B$
,
${D}_{t}+2c{K}_{t}=c{D}_{t}^{\prime}+2c{K}_{t}=c({D}_{t}^{\prime}+2{K}_{t})=0$
(Lemma 2.2). So
${\mathcal{O}}_{{X}_{{t}_{1}}}({D}_{{t}_{1}}+2c{K}_{1})$
must be trivial, i.e.,
$${D}_{{t}_{1}}+2c{K}_{1}=n\left({t}_{1}\right){D}_{1}^{\prime}+2c{K}_{1}=-2n\left({t}_{1}\right){K}_{1}+2c{K}_{1}=0.$$
Therefore
$n\left({t}_{1}\right)=c$
for every
${t}_{1}\in V-B$
. Hence
${D}_{t}=c{D}_{t}^{\prime}$
for every
$t\in V$
. By changing coefficient locally, we may assume
$D{|}_{{X}_{t}}={D}_{t}^{\prime}$
, where
$2{D}_{t}^{\prime}+{K}_{t}=0$
.

Since
$2{D}_{t}+{K}_{{X}_{t}}=0$
for every
$t\ne {t}_{0}$
, considering the divisor
$2D+{K}_{X}$
restricted to every fibre
${X}_{t}$
, we have
$${h}^{0}({X}_{0},{\mathcal{O}}_{{X}_{0}}(2{D}_{0}+{K}_{0}\left)\right)\ge {h}^{0}({X}_{t},{\mathcal{O}}_{{X}_{t}}(2{D}_{t}+{K}_{{X}_{t}}\left)\right)=1.$$
On the other hand
$${h}^{0}({X}_{0},{\mathcal{O}}_{{X}_{0}}(-2{D}_{0}-{K}_{0}\left)\right)\ge {h}^{0}({X}_{t},{\mathcal{O}}_{{X}_{t}}(-2{D}_{t}-{K}_{{X}_{t}}\left)\right)=1.$$
So we have
$${h}^{0}({X}_{0},{\mathcal{O}}_{{X}_{0}}(2{D}_{0}+{K}_{0}\left)\right)={h}^{0}({X}_{0},{\mathcal{O}}_{{X}_{0}}(-2{D}_{0}-{K}_{0}\left)\right)=1.$$
Again this says that the sheaf
${\mathcal{O}}_{{X}_{0}}(2{D}_{0}+{K}_{0}))$
is trivial. Hence
$2{D}_{0}+{K}_{0}=0$
.

Since
${S}_{t}$
has vanishing Hodge cohomology, and
${X}_{0}$
is isomorphic to
${X}_{t}$
, we have
$${h}^{0}({X}_{0},{\mathcal{O}}_{{X}_{0}}(2n{D}_{0}\left)\right)={h}^{0}({X}_{0},{\mathcal{O}}_{{X}_{0}}(-n{K}_{0}\left)\right)$$
$$={h}^{0}({X}_{t},{\mathcal{O}}_{{X}_{t}}(-n{K}_{t}\left)\right)={h}^{0}({X}_{t},{\mathcal{O}}_{{X}_{t}}(2n{D}_{t}^{\prime}\left)\right)=1.$$
Therefore
${S}_{0}$
is not affine.

Q.E.D.

Remark 2.6.
Let
$U$
be covered by a set of small open discs
${U}_{i}$
. By the above argument, for each
$i$
, there is a constant
${c}_{i}$
such that for
$t\in {U}_{i}$
,
${D}_{t}={c}_{i}{D}_{t}^{\prime}$
, where
${D}_{i}^{\prime}$
is the irreducible boundary elliptic curve on
${X}_{t}$
. Since
$U$
is connected, all these
${c}_{i}^{\prime}s$
are equal. That is, there is constant
$c$
, such that for all
$t\in U$
,
${D}_{t}=c{D}_{t}^{\prime}$
.

Lemma 2.7.
If there is an affine open set
$U$
in
$C$
such that for every
$t\in U$
,
$t\ne {t}_{0}$
,
${S}_{t}$
is of type (3) open surface, where
${t}_{0}$
is a fixed point of
$U$
, then
${S}_{0}$
must be of the same type.
$Proof$
. First,
${S}_{0}$
is not of type (2) open surface since
${X}_{0}$
is rational by Iitaka's theorem [I4]. We only need to prove that
${S}_{0}$
is not affine as above lemma. It suffices to prove that
${h}^{0}({X}_{0},{\mathcal{O}}_{{X}_{0}}(n{D}_{0}\left)\right)<c{n}^{2}$
for all positive number
$c$
(Lemma 2.1).

As in Lemma 2.5, we may assume that
${D}_{t}$
contains no exceptional curves of the first kind for every
$t$
in
$U$
although here the situation is more complicated. In fact, if there is an exceptional curve
${E}_{1}$
of the first kind in
${D}_{{t}_{1}}$
for some point
${t}_{1}$
in
$U$
, then locally analytically,
${E}_{1}$
sits in an irreducible nonsingular divisor
${D}_{1}$
of
$X$
, i.e.,
${D}_{1}$
is a prime component of
$D$
. (We may assume that
$D$
is an effective divisor on
$X$
with simple normal crossings [Zh] .) Now
$f$
is proper on
${D}_{1}$
and
${D}_{1}$
is a manifold. So we can apply Kodaira's extension theorem locally on
${D}_{1}$
near
${D}_{{t}_{1}}$
. More precisely, in our case, we can compute it directly. Since
${D}_{1}$
is smooth, for a small number
$\epsilon >0$
, in a neighborhood
$V=\{t\in C,|t-{t}_{1}|<\epsilon \}$
of
${t}_{1}$
,
${D}_{1}$
intersects every fibre
${X}_{t}$
with a prime divisor on
${X}_{t}$
. Since
${h}^{0}\left({\mathcal{O}}_{{X}_{t}}\right)=1$
and
${h}^{1}\left({\mathcal{O}}_{{X}_{t}}\right)={h}^{2}\left({\mathcal{O}}_{{X}_{t}}\right)=0$
(Lemma 2.1, Lemma 2.2), by the Riemann-Roch formula and upper semi-continuity, we have
$$\chi \left({\mathcal{O}}_{{X}_{t}}\right(n{E}_{t}\left)\right)=1+\frac{1}{2}{n}^{2}{E}_{t}^{2}-\frac{1}{2}n{E}_{t}\cdot {K}_{{X}_{t}}=1+\frac{1}{2}{n}^{2}{E}_{1}^{2}-\frac{1}{2}n{E}_{1}\cdot {K}_{{X}_{1}},$$
where
${E}_{t}={D}_{1}{|}_{{X}_{t}}$
. So
${E}_{t}$
is again an
$(-1)-$
curve on
${D}_{1}$
. This implies that all extended
$(-1)$
exceptional curves near
${D}_{{t}_{1}}$
sit in
${D}_{1}$
and do not meet
$Y$
. So after contraction,
$Y$
remains the same, that is, when contracting
$(-1)-$
curves, we only change the boundary
${D}_{t}$
but all open surfaces
${S}_{t}$
over
$t\in U$
remain unchanged.

If
${D}_{t}$
is the special divisor
${D}_{t}^{\prime}$
as in Mohan Kumar's Theorem, i.e., its dual graph is either
${\stackrel{~}{D}}_{8}$
or
${\stackrel{~}{E}}_{8}$
, then we have
${D}_{t}+{K}_{{X}_{t}}=0$
for every
$t\in U$
and
$t\ne {t}_{0}$
by Lemma 2.2. By the similar inequalities as in the proof of Lemma 2.5, we have
$${h}^{0}({X}_{0},{\mathcal{O}}_{{X}_{0}}({D}_{0}+{K}_{0}\left)\right)={h}^{0}({X}_{0},{\mathcal{O}}_{{X}_{0}}(-{D}_{0}-{K}_{0}\left)\right)=1.$$
Hence
${D}_{0}+{K}_{0}=0$
. Since
${X}_{0}$
is of type (3) projective surface and
${H}^{i}({S}_{0},{\Omega}_{{S}_{0}}^{j})=0$
, we know that
${S}_{0}$
is not affine and must be of type (3) open surface. But we can not guarantee that the dual graph of
${D}_{t}$
is either
${\stackrel{~}{D}}_{8}$
or
${\stackrel{~}{E}}_{8}$
. We only know
${D}_{t}$
has nine components and every prime component is isomorphic to
${\mathbb{P}}^{1}$
with self-intersection
$-2$
([Ku]). In Lemma 2.5, we always can assume that the special divisor
${D}_{t}$
in [Ku] on
${X}_{t}$
is the restriction of a global divisor
$D$
on
${X}_{t}$
since
${D}_{t}$
has only one component. Now the situation is more delicate.

Let
${D}_{t}^{\prime}$
be the special divisor of type (3) projective surface as above, i.e., its dual graph is either
${\stackrel{~}{D}}_{8}$
or
${\stackrel{~}{E}}_{8}$
,
${D}_{t}^{\prime}\cdot {D}_{t}^{\prime}={D}_{t}^{\prime}\cdot {K}_{t}=0$
and
${\mathcal{O}}_{{D}_{t}^{\prime}}\left({D}_{t}^{\prime}\right)$
is nontorsion [Ku] . For any nonnegative integer
$n$
, there is
$m$
such that
$m{D}_{t}^{\prime}-n{D}_{t}$
is effective.

For example, we may choose
$m=an$
where
$a$
is the maximum coefficient of
${D}_{t}$
's components. So
$$0<{h}^{0}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)\le {h}^{0}({X}_{t},{\mathcal{O}}_{{X}_{t}}(m{D}_{t}^{\prime}\left)\right)=1.$$
Therefore
${h}^{0}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)=1.$
By Serre duality,
${H}^{2}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)=0$
for all
$n\gg 0$
since
${K}_{{X}_{t}}$
and
${D}_{t}$
have the same support by Lemma 2.2. Consider
${h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)$
, there are three cases [Za].

Case 1.
${h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)$
is bounded, i.e., there is a positive integer
$k$
such that for all
$n\ge 0$
, we have
$${h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)\le k<\infty .$$
By Zariski's theorem, page 611, [Za] ,
${D}_{t}$
is arithmetically effective. By Riemann-Roch formula and Lemma 2.1, Lemma 2.2, we have
$${h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)=-\frac{1}{2}{n}^{2}{D}_{t}^{2}+\frac{1}{2}n{D}_{t}\cdot {K}_{{X}_{t}}.$$
This equality gives us
${D}_{t}^{2}={D}_{t}.{K}_{t}=0$
since
${h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)$
is bounded for all
$n$
. Then for every prime component
$E$
in
${D}_{t}$
, we have
$E.{D}_{t}=0$
since
${D}_{t}$
is arithmetically effective. By Lemma 1.7 [Ku], we know
${D}_{t}=n\left(t\right){D}_{t}^{\prime}$
, where the positive integer
$n\left(t\right)$
depends on the point
$t$
in
$U$
. So for every
$n\ge 0$
, by Lemma 2.1, we have
$${h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}(m{D}_{t}\left)\right)={h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}(mn\left(t\right){D}_{t}^{\prime}\left)\right)=0.$$
Now the Euler characteristic of
${\mathcal{O}}_{{X}_{0}}\left(n{D}_{0}\right)$
is
$$\chi \left({\mathcal{O}}_{{X}_{0}}\right(n{D}_{0}\left)\right)=1-\frac{1}{2}{n}^{2}{D}_{0}^{2}+\frac{1}{2}n{D}_{0}.{K}_{{X}_{0}}=1.$$
Thus again
${D}_{0}^{2}={D}_{0}\cdot {K}_{{X}_{0}}=0$
. By the same argument as in the above lemma and remark, there is a positive integer
$c$
such that for every
$t\in U$
and
$t\ne {t}_{0}$
,
${D}_{t}=c{D}_{t}^{\prime}$
. Consider the divisor
$D+c{K}_{X}$
on
$X$
, when restricted to
${X}_{0}$
, we have
$${h}^{0}({X}_{0},{\mathcal{O}}_{{X}_{0}}({D}_{0}+c{K}_{{X}_{0}}\left)\right)\ge 1,{h}^{0}({X}_{0},{\mathcal{O}}_{{X}_{0}}(-{D}_{0}-c{K}_{{X}_{0}}\left)\right)\ge 1.$$
So
${D}_{0}=-c{K}_{{X}_{0}}$
. Let
${D}_{0}=P+N$
be the Zariski decomposition of
${D}_{0}$
, then
$P$
is nef,
$N$
is definite negative (both are effective) and every component of
$N$
does not intersect
$P$
. Let
$E$
be a prime component of
$P$
. Locally analytically,
$E$
is contained in a prime divisor
$G$
of
$X$
. Let
$G{|}_{{X}_{t}}={E}_{t}$
. Apply upper semi-continuity and Riemann-Roch formula to
${\mathcal{O}}_{{X}_{0}}\left(nE\right)$
and
${\mathcal{O}}_{{X}_{t}}\left(n{E}_{t}\right)$
, we have
$E\cdot {K}_{0}={E}_{t}\cdot {K}_{t}=0$
, Lemma 3.1, [Ku] . Thus
$E\cdot {D}_{0}=E\cdot (-c{K}_{0})=0$
. If
$E\cdot P>0$
, then
$E\cdot P=E\cdot ({D}_{0}-N)=-E\cdot N>0$
. Therefore
$E\cdot N<0$
. This means that
$E$
must be a component of
$N$
which is a controdiction since no component of
$N$
intersects
$P$
.

So
${P}^{2}=0$
. By Corrollary 14.18, [Ba] ,
$\kappa ({D}_{0},{X}_{0})\le 1$
. By Lemma 2.1,
${S}_{0}$
is not affine.

Case 2. If
${h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)$
is as large as
$cn$
for some positive number
$c$
, then by Zariski's theorem [Za],
${D}_{t}$
is arithmetically effective and the intersection form of
${D}_{t}$
is negative definite. This contradicts Lemma 1.6 [Ku]. So this case can not happen.

Case 3. If
${h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)$
is as large as
$k{n}^{2}$
for some positive number
$k$
, then by Riemann-Roch formula, we know
${D}_{t}^{2}<0$
. Let
${D}_{t}=A+B$
be the Zariski decomposition such that
$A$
is arithmetically effective,
$B\ge 0$
is negative definite and every prime component of
$B$
does not meet
$A$
. There is a positive integer
${n}_{0}$
such that
${n}_{0}A$
and
${n}_{0}B$
are integral. Without loss of generality, we may assume
$A$
and
$B$
are integral. Since there is a positive integer
$l$
such that
$l{D}_{t}^{\prime}-{D}_{t}$
is effective, we have exact sequence
$$0\u27f6\mathcal{O}\left(n{D}_{t}\right)\u27f6\mathcal{O}\left(nl{D}_{t}^{\prime}\right)\u27f6Q\u27f60,$$
where
$Q$
is the cokernel. Hence we still have
${h}^{0}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)=1$
even though
${D}_{t}$
is different from
${D}_{t}^{\prime}$
. Therefore
$\kappa ({D}_{t},{X}_{t})=0$
by Iitaka's Theorem. This implies
${A}^{2}=0$
[Ba], [Za]. Since
$A$
is arithmetically effective and supp
${D}_{t}=$
supp
$A\cup $
supp
$B$
, for every prime component
$E$
of
${D}_{t}$
,
$E.A=0$
. By Corollary 1.7 [Ku], there is positive integer
${m}_{0}$
such that
$A={m}_{0}{D}_{t}^{\prime}$
. So
${D}_{t}^{2}={B}^{2}$
and
${D}_{t}-{D}_{t}^{\prime}\ge 0$
. Let
${D}_{0,i}$
be a prime component of
${D}_{0}=D{|}_{{X}_{0}}$
. Choose a small neighborhood
$V$
of
${t}_{0}$
such that locally analytically in
$V$
,
${D}_{0,i}$
lies in a unique prime divisor
${D}_{i}$
of
${f}^{-1}\left(V\right)$
.

${D}_{i}$
cuts every fibre
${X}_{t}$
,
$t\in V$
with an irreducible
$(-2)$
-curve. So over
$V$
, there is one to one correspondence between the prime divisor of
${D}_{t}$
and the prime divisor of
${f}^{-1}\left(V\right)$
. We may rearrange the coefficients of
${D}_{i}$
locally as in the proof of the above lemma such that
${D}_{t}=c{D}_{t}^{\prime}$
,
${t}_{0}\ne t\in V$
. So
${h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)=0$
for all
$t\in V$
,
$t\ne {t}_{0}$
. Then we reduce Case 3 to Case 1. This proves that
${S}_{0}$
is not affine.

Q.E.D.

Remark 2.8.
If
${D}_{t}$
is not the special divisor as in Mohan Kumar's Theorem, i.e.,
${D}_{t}$
has different coefficients from
${D}_{t}^{\prime}$
but they have the same support, then we still have
${D}_{t}\cdot {K}_{t}=0$
by Lemma 3.1 [Ku] . Since
${h}^{0}({X}_{t},{\mathcal{O}}_{{X}_{t}}(nB\left)\right)=1$
, by Riemann-Roch formula,
${h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)=-\frac{1}{2}{n}^{2}{D}_{t}^{2}=-\frac{1}{2}{n}^{2}{B}^{2}={h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}(nB\left)\right)\sim c{n}^{2}.$
Thus
${B}^{2}<0$
. Since
$B$
is definite negative, by Lemma 1.6 [Ku], we know that the support of
$B$
is strictly smaller than the support of
${D}_{t}$
.

Remark 2.9.
Let
${X}_{t}$
be of type (3) projective surface and
${D}_{t}^{\prime}$
the special divisor as above. Let
$E$
be any prime component of
${D}_{t}^{\prime}$
. Then
${E}^{2}=-2$
[Ku]. Since the canonical divisor
${K}_{t}=-{D}_{t}^{\prime}$
, by Riemann-Roch,
${h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}^{\prime}+E\left)\right)={n}^{2}$
.

Combining with the above argument, for any divisor
${D}_{t}$
with the same support as
${D}_{t}^{\prime}$
, we know either
${h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)=0$
or
${h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)\sim c{n}^{2}$
for some positive integer
$c$
.

Lemma 2.10.
If
${S}_{0}$
is affine then there is an affine open set
$U$
in
$C$
such that for every
$t\in U$
,
${S}_{t}$
is affine.
$Proof.$
This is direct consequence of the above lemmas since
${S}_{0}$
only can be one of the three surfaces.

Q.E.D.

Lemma 2.11.
If there is an affine open set
$U$
in
$C$
such that for every
$t\ne {t}_{0}$
,
${S}_{t}$
is affine, then
${S}_{0}$
is affine.
$Proof.$
This is an immediate conclusion of Mohan Kumar's classification and upper semi-continuity theorem.

Q.E.D.

The first half of Theorem 1.1 follows from the above lemmas. The second half is a direct consequence of Theorem 5.11 and Theorem 6.12 in [Uen] . In fact, if one smooth fibre
${X}_{0}$
is not affine, then all smooth fibres are not affine by the above lemmas. Since
${X}_{0}$
is a ruled surface,
$\kappa \left({X}_{0}\right)=-\infty .$
So
$$\kappa \left(X\right)\le \kappa \left({X}_{0}\right)+1=-\infty .$$
By Lemma 2.1,
$$0<\kappa (D,X)\le \kappa ({D}_{t},{X}_{t})+1=1.$$
This completes the proof of Theorem 1.1.
3 Proof of Theorem 1.2

Lemma 3.1 (Goodman, Hartshorne).
Let
$V$
be a scheme and
$D$
be an effective Cartier divisor on
$V$
. Let
$U=V-SuppD$
and
$F$
be any coherent sheaf on
$V$
, then for every
$i\ge 0,$

$${lim}_{\overrightarrow{n}}{H}^{i}(V,F\otimes \mathcal{O}(nD\left)\right)\sim ={H}^{i}(U,F{|}_{U}).$$
This lemma enables us to transfer the cohomology information from
$Y$
to its completion
$X$
.
$Proof$
$of$
$Theorem$
$1.2$
. The idea is to prove for any coherent sheaf
$F$
on
${Y}^{\prime}$
,
${H}^{i}({Y}^{\prime},{F}_{{Y}^{\prime}})=0$
for all
$i>0$
. Since the dimension of
${Y}^{\prime}$
is 3, we only need to consider
$i=1,2,3.$
We use the technique in [Zh] with some modification. We present all details for completeness.

Notice that
${Y}^{\prime}\subset Y$
. Let
${F}_{{Y}^{\prime}}$
be any coherent sheaf on
${Y}^{\prime}$
, then it can be extended to a coherent sheaf
${F}_{X}$
on
$X$
and
${F}_{Y}{|}_{{S}_{t}}$
,
${F}_{X}{|}_{{X}_{t}}$
are coherent, page 115, page 126, [H1] . We will not distinguish them and just write
$F$
. Since general fibre
${X}_{t}$
over
$t\in C$
is smooth and irreducible [Zh] and for any
$F$
, there is an open set
$U$
in
$C$
such that
${R}^{i}{f}_{*}F$
is locally free on
$U$
, we may assume that
${R}^{i}{f}_{*}F$
is locally free on
$C$
and every fibre over
$C$
is smooth and irreducible.

(1) Proof of
${H}^{3}(Y,F)=0$
.

Since
${S}_{t}$
is affine for every
$t$
in
$C$
, we have
${H}^{i}({S}_{t},F{|}_{{S}_{t}})=0$
for every
$i>0$
. Let
${F}_{n}=F\otimes {\mathcal{O}}_{X}\left(nD\right)$
and
${F}_{n,t}=F\otimes {\mathcal{O}}_{X}\left(nD\right){|}_{{X}_{t}}$
. By Goodman and Hartshorne's Lemma,
$${lim}_{\overrightarrow{n}}{H}^{i}({X}_{t},{F}_{n,t})=0$$
for all
$i>0$
and
$t\in C$
. Since each fibre has dimension 2, we have
${H}^{3}({X}_{t},{F}_{n,t})=0$
for all
$n\ge 0$
and
$t\in C$
. By upper semi-continuity,
${R}^{3}{f}_{*}{F}_{n}=0$
for all
$n$
. Again by Goodman and Hartshorne's Lemma,
$${H}^{3}(Y,F)={lim}_{\overrightarrow{n}}{H}^{3}\left({f}^{-1}\right(C),{F}_{n})={lim}_{\overrightarrow{n}}{R}^{3}{f}_{*}{F}_{n}\left(C\right)=0.$$
(2) Proof of
${H}^{2}(Y,F)=0$
.

It suffices to prove the claim for locally free sheaves. In fact, suppose
${H}^{2}(Y,L)=0$
for any locally free sheaf
$L$
on
$X$
. For any coherent sheaf
$F$
on
$X$
, there is a locally free sheaf
$L$
on
$X$
such that we have the surjective map
$L\u27f6F$
. Let
$K$
be the kernel, then we have short exact sequence on
$Y$
$$0\u27f6K\u27f6L\u27f6F\u27f60.$$
By step 1, we know
${H}^{3}(Y,K)=0$
since
$K$
is also coherent [H1] . So
${H}^{2}(Y,L)=0$
implies
${H}^{2}(Y,F)=0$
.

So we may assume that
$F$
is a locally free sheaf on
$X$
.

Let
$t\in C$
. From the exact sequence
$$0\u27f6{\mathcal{O}}_{X}\left(nD\right)\u27f6{\mathcal{O}}_{X}\left(\right(n+1\left)D\right)\u27f6{\mathcal{O}}_{D}\left(\right(n+1\left)D\right)\u27f60,$$
tensoring with
$F$
then with
${\mathcal{O}}_{{X}_{t}}$
, we have
$$0\u27f6{F}_{n,t}\u27f6{F}_{n+1,t}\u27f6{F}_{n+1,t}{|}_{{D}_{t}}\u27f60.$$
Since
${D}_{t}$
is a curve,
${H}^{2}({X}_{t},{F}_{n+1,t}{|}_{{D}_{t}})=0$
for all
$n\ge 0$
and
$t\in C$
. So the map
${H}^{2}({X}_{t},{F}_{n,t})\to {H}^{2}({X}_{t},{F}_{n+1,t})$
is surjective. Since
${S}_{t}={X}_{t}-{D}_{t}$
is affine, by Goodman and Hartshorne's Lemma,
$${lim}_{\overrightarrow{n}}{H}^{2}({X}_{t},{F}_{n,t})={H}^{2}({S}_{t},F)=0.$$
So for any
$t\in C$
, there is a positive integer
$n\left(t\right)$
, depending on
$t$
such that for every
$n\ge n\left(t\right)$
,
${H}^{2}({X}_{t},{F}_{n,t})=0$
.

Given any
$n$
, there is an affine open set
${U}_{n}$
of
$C$
such that
${R}^{2}{f}_{*}{F}_{n}$
is locally free on
${U}_{n}$
. By the same argument as in the next paragraph, the intersection of these infinitely many open sets is not empty. Now fix some
${t}_{0}$
in
$\cap {U}_{n}$
such that
${H}^{2}({X}_{{t}_{0}},{F}_{n,{t}_{0}})=0$
for every
$n\ge n\left({t}_{0}\right)$
and there is an open neighborhood
${U}_{0}$
of
${t}_{0}$
in
$\overline{C}$
such that
${R}^{2}{f}_{*}{F}_{n\left({t}_{0}\right)}$
is locally free on
${U}_{0}$
. Then
${H}^{2}({X}_{t},{F}_{n\left({t}_{0}\right),t})=0$
for every
$t$
in
${U}_{0}$
. So
${H}^{2}({X}_{t},{F}_{n,t})=0$
for every
$t$
in
${U}_{0}$
and every
$n\ge n\left({t}_{0}\right)$
. Let
$C-{U}_{0}=\{{t}_{1},{t}_{2},...,{t}_{k}\}$
, choose
${n}_{0}=max\left(n\right({t}_{0}),n({t}_{1}),...,n({t}_{k}\left)\right)$
, then
${H}^{2}({X}_{t},{F}_{n,t})=0$
for every
$t\in C$
and every
$n\ge {n}_{0}$
. By upper semi-continuity theorem,
$({R}^{2}{f}_{*}{F}_{n}{)}_{t}/\mathcal{P}({R}^{2}{f}_{*}{F}_{n}{)}_{t}=0$
for all points
$t$
in
$C$
. By Nakayama's lemma,
${R}^{2}{f}_{*}{F}_{n}{|}_{C}=0$
. By Goodman and Hartshorne's Lemma
$${H}^{2}(Y,F)={lim}_{\overrightarrow{n}}{H}^{2}\left({f}^{-1}\right(C),{F}_{n})={lim}_{\overrightarrow{n}}{R}^{2}{f}_{*}{F}_{n}\left(C\right)=0.$$
(3) Proof of
${H}^{1}({Y}^{\prime},F)=0$
, where
${Y}^{\prime}$
is an open subset of
$Y$
obtained by removing finitely many fibres from
$Y$
.

Let
${F}_{n}$
be as above. For any fixed
$n$
, there is an open set
${U}_{n}$
in
$\overline{C}$
, such that
${R}^{1}{f}_{*}{F}_{n}$
is locally free on
${U}_{n}$
. Let
${U}_{n}=\overline{C}\backslash {A}_{n}$
, where
${A}_{n}$
is closed in
$\overline{C}$
, i.e., it consists only finitely many points of
$\overline{C}$
. Since any complete metric space is a Baire space, Chapter 9, [Bo2] ,
$B=\overline{C}\backslash \cup {A}_{n}=\cap {U}_{n}$
is a dense subset of
$\overline{C}$
in complex topology. Hence for every point
$t$
in
$B$
, all stalks
$({R}^{1}{f}_{*}{F}_{n}{)}_{t}$
are locally free. Write
$B$
as a union of connected subsets
${B}_{m}$
,
$B=\cup {B}_{m}$
, then there is one
${B}_{m}$
, such that
${B}_{m}$
is dense in
$\overline{C}$
and connected in complex topology. So we may assume that
$B$
is connected. Again by upper-semicontinuity theorem, for every point
$t$
in
$C$
and every
$n\ge {n}_{0}$
, since
${R}^{2}{f}_{*}{F}_{n}{|}_{C}=0$
, we have
$$({R}^{1}{f}_{*}{F}_{n}{)}_{t}\otimes \mathbb{C}\sim ={H}^{1}({X}_{t},{F}_{n,t}).$$
For any
$m$
,
${h}^{1}({X}_{t},{F}_{m,t})$
is constant on
$B$
since
${R}^{1}{f}_{*}{F}_{m}$
is locally free at every point
$t$
in
$B$
and
$B$
is connected. So for the above
$n$
and for all points
$t$
in
$B$
, there is
$l$
such that the map
$${H}^{1}({X}_{t},{F}_{n,t})\u27f6{H}^{1}({X}_{t},{F}_{n+l,t})$$
is zero. Moreover, for every point
$t$
in
$C$
and sufficiently large
$n$
, we have the following commutative diagram

The map
$\beta $
is zero for every
$t\in B$
, so the map
$$\alpha :({R}^{1}{f}_{*}{F}_{n}{)}_{t}/\mathcal{P}({R}^{1}{f}_{*}{F}_{n}{)}_{t}\u27f6({R}^{1}{f}_{*}{F}_{n+l}{)}_{t}/\mathcal{P}({R}^{1}{f}_{*}{F}_{n+1}{)}_{t}$$
is zero for all points
$t$
in
$B$
. By the local freeness, this says for every point
$t$
in
$B$
, the stalks satisfy
$${lim}_{\overrightarrow{n}}({R}^{1}{f}_{*}{F}_{n}{)}_{t}=0.$$
To see this, fix a point
${t}_{0}$
in
$B$
, for any sufficiently large
$n$
and for the above
$l$
, choose an affine open set
$V$
containing
${t}_{0}$
such that both
${R}^{1}{f}_{*}{F}_{n}$
and
${R}^{1}{f}_{*}{F}_{n+l}$
are locally free on
$V$
. So there are two positive integers
${m}_{1}$
and
${m}_{2}$
such that
${R}^{1}{f}_{*}{F}_{n}\left(V\right)=\mathcal{O}(V{)}^{{m}_{1}}$
and
${R}^{1}{f}_{*}{F}_{n+l}\left(V\right)=\mathcal{O}(V{)}^{{m}_{2}}$
. Now for infinitely many maximal ideals
$\mathcal{P}$
, we have commutative diagram

Since
$\psi (\mathcal{O}(V{)}^{{m}_{1}})\subset \cap \mathcal{P}\mathcal{O}(V{)}^{{m}_{2}}=0$
, where
$\mathcal{P}$
runs over infinitely many maximal ideals of
$\mathcal{O}\left(V\right)$
, we have
$\psi (\mathcal{O}(V{)}^{{m}_{1}})=0$
. This proves
$${lim}_{\overrightarrow{n}}({R}^{1}{f}_{*}{F}_{n}{)}_{t}=0.$$
Since the direct limit of
${R}^{1}{f}_{*}{F}_{n}$
is quasi-coherent, its support is locally closed.

$$\begin{array}{ccc}{R}^{1}{f}_{*}{F}_{n}\otimes \mathbb{C}\left(t\right)& \stackrel{\approx}{\u27f6}& {H}^{1}({X}_{t},{F}_{n,t})\end{array}$$ |

$$\begin{array}{ccc}\downarrow \alpha & & \downarrow \beta \end{array}$$ |

$$\begin{array}{ccc}{R}^{1}{f}_{*}{F}_{n+l}\otimes \mathbb{C}\left(t\right)& \stackrel{\approx}{\u27f6}& {H}^{1}({X}_{t},{F}_{n+l,t}).\end{array}$$ |

$$\begin{array}{ccc}\mathcal{O}(V{)}^{{m}_{1}}& \stackrel{\psi}{\u27f6}& \mathcal{O}(V{)}^{{m}_{2}}\end{array}$$ |

$$\begin{array}{ccc}\downarrow {\pi}_{1}& & \downarrow {\pi}_{2}\end{array}$$ |

$$\begin{array}{ccc}\mathcal{O}(V{)}^{{m}_{1}}/\mathcal{P}\mathcal{O}(V{)}^{{m}_{1}}& \stackrel{\phi}{\u27f6}& \mathcal{O}(V{)}^{{m}_{2}}/\mathcal{P}\mathcal{O}(V{)}^{{m}_{2}}.\end{array}$$ |

Now
$B$
is dense and connected in complex topology, there exists an affine open set
$U$
in
$\overline{C}$
such that on
$U$
, the direct limit
$${lim}_{\overrightarrow{n}}{R}^{1}{f}_{*}{F}_{n}{|}_{U}=0.$$
Let
${Y}^{\prime}={f}^{-1}\left(U\right)-D$
, by Goodman and Hartshorne's Lemma, we have
$${H}^{1}({Y}^{\prime},F)={lim}_{\overrightarrow{n}}{H}^{1}\left({f}^{-1}\right(U),{F}_{n})={lim}_{\overrightarrow{n}}{R}^{1}{f}_{*}{F}_{n}\left(U\right)=0.$$
This finishes the proof of Theorem 1.2.

Remark 3.2.
In our above proof of step 3, we encounter the following two questions if we do not know the local freeness of of
${R}^{1}{f}_{*}{F}_{n}$
.

(1) Let
$U$
be a smooth affine curve, then
$\mathcal{O}\left(U\right)=A$
is a Dedekind domain. Let
$N$
be a finitely generated module over
$A$
, then under what condition,
$\cap (\mathcal{P}N)=0$
? Where
$\mathcal{P}$
runs over all maximal ideals of
$A$
. A sufficient condition is that
$N$
is projective module. But this condition is too strong. Our
$N$
is defined by cohomology. It is hard to see it is projective or not. Definitely finitely generated module is not sufficient. For example, let
$U={\mathbb{A}}^{1}$
,
$A=\mathcal{O}\left(U\right)=\mathbb{C}\left[x\right]$
,
$N=\mathbb{C}\left[x\right]/\left({x}^{2}\right)$
, then
$\cap (\mathcal{P}N)\ne 0.$
(2) Let
$A$
be a Dedekind domain and
$\mathcal{P}$
as above, let
$({M}_{n},{f}_{n})$
be a direct system of finitely generated
$A$
-modules. If
$${lim}_{\overrightarrow{n}}({M}_{n}/\mathcal{P}{M}_{n})=0,$$
under what conditions, can we say that
$${lim}_{\overrightarrow{n}}{M}_{n}=0?$$
Again all
${M}_{n}$
being finitely generated is not sufficient. For example, let
$A=\mathbb{C}\left[\right[t\left]\right]$
, the ring of formal power series, let
${M}_{n}={t}^{-n}A$
, then
$${lim}_{\overrightarrow{n}}{M}_{n}=\mathbb{K}\ne 0,$$
where
$\mathbb{K}=\mathbb{C}\left(\right(t\left)\right)$
but
$${lim}_{\overrightarrow{n}}{M}_{n}/\mathcal{P}{M}_{n}=0.$$

4 Proof of Theorem 1.3

Lemma 4.1.
${R}^{i}{f}_{*}{\mathcal{O}}_{X}\left(nD\right)$
is locally free for all
$i\ge 0$
and
$n\gg 0$
.
$Proof$
. Since each fibre has dimension 2, by upper semi-continuity theorem [GH, Mu] ,
${R}^{i}{f}_{*}{\mathcal{O}}_{X}\left(nD\right)=0$
for all
$i>2$
and
$n\ge 0$
. By Lemma 2.1, since each fibre
${X}_{t}$
is of type (2), we have
$${h}^{0}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)={h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)=1\text{and}{h}^{2}({X}_{t},{\mathcal{O}}_{{X}_{t}}(n{D}_{t}\left)\right)=0$$
for all
$t\in C$
and
$n\gg 0$
.

Q.E.D.

Lemma 4.2.
${R}^{i}{f}_{*}{\Omega}_{X}^{3}\left(nD\right)={R}^{i}{f}_{*}{\mathcal{O}}_{X}({K}_{X}+nD)$
is locally free for all
$i\ge 0$
and
$n\gg 0$
.
$Proof$
. Since
${X}_{t}$
is smooth, we have
${K}_{X}+D{|}_{{X}_{t}}={K}_{{X}_{t}}={K}_{t}$
. So
${\Omega}_{X}^{3}\left(nD\right){|}_{{X}_{t}}={\mathcal{O}}_{X}({K}_{X}+nD){|}_{{X}_{t}}={\mathcal{O}}_{{X}_{t}}({K}_{t}+(n-1\left){D}_{t}\right),$
where
${D}_{t}=D{|}_{{X}_{t}}.$
By Lemma 2.1,
$${h}^{0}({X}_{t},{\Omega}_{X}^{3}(nD\left){|}_{{X}_{t}}\right)={h}^{0}({X}_{t},{\mathcal{O}}_{{X}_{t}}({K}_{t}+(n-1){D}_{t})=1,$$
$${h}^{1}({X}_{t},{\Omega}_{X}^{3}(nD\left){|}_{{X}_{t}}\right)={h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}({K}_{t}+(n-1){D}_{t})=1,$$
and
$${h}^{2}({X}_{t},{\Omega}_{X}^{3}(nD\left){|}_{{X}_{t}}\right)={h}^{2}({X}_{t},{\mathcal{O}}_{{X}_{t}}({K}_{t}+(n-1){D}_{t})=0$$
for all
$n\gg 0$
. This proves the local freeness.

Q.E.D.

Lemma 4.3.
${R}^{i}{f}_{*}{\Omega}_{X}^{1}\left(nD\right)$
is locally free for all
$i\ge 0$
and
$n\gg 0$
.
$Proof$
. From the exact sequences ([H1], II, Theorem 8.17 and [GrH], page 157)
$$0\u27f6{\mathcal{O}}_{{X}_{t}}\u27f6{\Omega}_{X}^{1}{|}_{{X}_{t}}\u27f6{\Omega}_{{X}_{t}}^{1}\u27f60,$$
tensoring with
${\mathcal{O}}_{X}\left(nD\right)$
, we have
$$0\u27f6{\mathcal{O}}_{{X}_{t}}\left(n{D}_{t}\right)\u27f6{\Omega}_{X}^{1}\left(nD\right){|}_{{X}_{t}}\u27f6{\Omega}_{{X}_{t}}^{1}\left(n{D}_{t}\right)\u27f60.$$
We will prove that for any two points
$t,{t}^{\prime}\in C$
and all
$n\gg 0$
, we have
$${h}^{i}({X}_{t},{\Omega}_{X}^{1}(nD\left){|}_{{X}_{t}}\right)={h}^{i}({X}_{{t}^{\prime}},{\Omega}_{X}^{1}(nD\left){|}_{{X}_{{t}^{\prime}}}\right).$$
Then by upper semi-contituity theorem [GR, Mu] , we are done. By the above short exact sequences, for both fibres
${X}_{t}$
and
${X}_{{t}^{\prime}}$
, we have the commutative diagram

where the natural map
$\phi $
is defined as follows. If
$\xi \in {H}^{0}\left({\Omega}_{X}^{1}\right(nD\left){|}_{{X}_{t}}\right)$
is contained in the image of
${H}^{0}\left({\mathcal{O}}_{{X}_{t}}\right(n{D}_{t}\left)\right)=\mathbb{C}$
, then there is a number
$a\in \mathbb{C}$
such that
$\xi ={\alpha}_{1}\left(a\right)$
. Let
$\phi \left(a\right)={\beta}_{1}\left(a\right).$
If
$\xi $
is not contained in the image of
${\alpha}_{1}$
, then
${\alpha}_{2}\left(\xi \right)\in {H}^{0}\left({\Omega}_{{X}_{t}}^{1}\right(n{D}_{t}\left)\right)$
and
${\alpha}_{3}\circ {\alpha}_{2}\left(\xi \right)=0$
. So there is
$\eta \in {H}^{0}\left({\Omega}_{X}^{1}\right(nD\left){|}_{{X}_{{t}^{\prime}}}\right)$
such that
${\beta}_{3}\circ {\beta}_{2}\left(\eta \right)=0.$
Let
$\phi \left(\xi \right)=\eta $
, by 5-Lemma [La] , we have
$${H}^{0}\left({\Omega}_{X}^{1}\right(nD\left){|}_{{X}_{t}}\right)={H}^{0}\left({\Omega}_{X}^{1}\right(nD\left){|}_{{X}_{{t}^{\prime}}}\right).$$
Similarly, we have
$${H}^{i}\left({\Omega}_{X}^{1}\right(nD\left){|}_{{X}_{t}}\right)={H}^{i}\left({\Omega}_{X}^{1}\right(nD\left){|}_{{X}_{{t}^{\prime}}}\right)$$
for
$i>0$
.

$$\begin{array}{cccc}0\u27f6{H}^{0}\left({\mathcal{O}}_{{X}_{t}}\right(n{D}_{t}\left)\right)& \stackrel{{\alpha}_{1}}{\u27f6}{H}^{0}\left({\Omega}_{X}^{1}\right(nD\left){|}_{{X}_{t}}\right)& \stackrel{{\alpha}_{2}}{\u27f6}{H}^{0}\left({\Omega}_{{X}_{t}}^{1}\right(n{D}_{t}\left)\right)& \stackrel{{\alpha}_{3}}{\u27f6}{H}^{1}\left({\mathcal{O}}_{{X}_{t}}\right(n{D}_{t}\left)\right)\end{array}$$ |

$$\begin{array}{ccccc}\parallel & \downarrow \phi & \parallel & \parallel & \end{array}$$ |

$$\begin{array}{cccc}0\u27f6{H}^{0}\left({\mathcal{O}}_{{X}_{{t}^{\prime}}}\right(n{D}_{{t}^{\prime}}\left)\right)& \stackrel{{\beta}_{1}}{\u27f6}{H}^{0}\left({\Omega}_{X}^{1}\right(nD\left){|}_{{X}_{{t}^{\prime}}}\right)& \stackrel{{\beta}_{2}}{\u27f6}{H}^{0}\left({\Omega}_{{X}_{{t}^{\prime}}}^{1}\right(n{D}_{{t}^{\prime}}\left)\right)& \stackrel{{\beta}_{3}}{\u27f6}{H}^{1}\left({\mathcal{O}}_{{X}_{{t}^{\prime}}}\right(n{D}_{{t}^{\prime}}\left)\right),\end{array}$$ |

Q.E.D.

Lemma 4.4.
${R}^{i}{f}_{*}{\Omega}_{X}^{2}\left(nD\right)$
is locally free for all
$i\ge 0$
and
$n\gg 0$
.
$Proof.$
Notice that we have the short exact sequence ([H1] , II, Theorem 8.17 and [GrH] , page 157)
$$0\u27f6{\Omega}_{{X}_{t}}^{1}\u27f6{\Omega}_{X}^{2}{|}_{{X}_{t}}\u27f6{\Omega}_{{X}_{t}}^{2}\u27f60.$$
Tensoring with
${\mathcal{O}}_{X}\left(nD\right)$
, we have
$$0\u27f6{\Omega}_{{X}_{t}}^{1}\left(n{D}_{t}\right)\u27f6{\Omega}_{X}^{2}\left(nD\right){|}_{{X}_{t}}\u27f6{\Omega}_{{X}_{t}}^{2}\left(n{D}_{t}\right)\u27f60.$$
By Lemma 2.1, for every
$t\in C$
, we know
$${h}^{0}({X}_{t},{\Omega}_{{X}_{t}}^{2}(n{D}_{t}\left)\right)={h}^{1}({X}_{t},{\mathcal{O}}_{{X}_{t}}({K}_{t}+n{D}_{t}\left)\right)=1\text{and}{h}^{2}({X}_{t},{\mathcal{O}}_{{X}_{t}}({K}_{t}+n{D}_{t}\left)\right)=0.$$
By the same argument as the above lemma, for any two points
$t,{t}^{\prime}\in C$
and all
$n\gg 0$
, writing the long exact sequences for
$t$
and
${t}^{\prime}$
, we have
$${h}^{i}({X}_{t},{\Omega}_{X}^{2}(nD\left){|}_{{X}_{t}}\right)={h}^{i}({X}_{{t}^{\prime}},{\Omega}_{X}^{2}(nD\left){|}_{{X}_{{t}^{\prime}}}\right).$$

Q.E.D.

Lemma 4.5.
For every
$t\in C$
,
${H}^{i}({S}_{t},{\Omega}_{Y}^{j}{|}_{{S}_{t}})=0$
for all
$j\ge 0$
and
$i>0$
, where
${S}_{t}={X}_{t}-{D}_{t}$
.
$Proof.$
Since
${S}_{t}$
is a surface, we only need to consider
$i=1,2.$
The claim is obvious for
${\mathcal{O}}_{Y}$
. Since
${S}_{t}$
is smooth, we have exact sequence
$$0\u27f6{\phi}_{t}/{\phi}_{t}^{2}={\mathcal{O}}_{{S}_{t}}\u27f6{\Omega}_{Y}^{1}{|}_{{S}_{t}}\u27f6{\Omega}_{{S}_{t}}^{1}\u27f60,$$
where
${\phi}_{t}$
is the defining sheaf of
${S}_{t}$
. Therefore the claim holds for
${\Omega}_{Y}^{1}{|}_{{S}_{t}}$
. Since the normal sheaf
$${\mathcal{N}}_{{S}_{t}/Y}=\mathcal{\mathscr{H}}om({\phi}_{t}/{\phi}_{t}^{2},{\mathcal{O}}_{{S}_{t}})={\mathcal{O}}_{{S}_{t}},$$
we have
$${\omega}_{{S}_{t}}\sim ={\omega}_{Y}\otimes {\mathcal{N}}_{{S}_{t}/Y}\sim ={\omega}_{Y}\otimes {\mathcal{O}}_{{S}_{t}}={\omega}_{Y}{|}_{{S}_{t}}.$$
Hence the claim holds for
${\Omega}_{Y}^{3}{|}_{{S}_{t}}$
. From the exact sequence
$$0\u27f6{\Omega}_{{S}_{t}}^{1}\u27f6{\Omega}_{Y}^{2}{|}_{{S}_{t}}\u27f6{\Omega}_{{S}_{t}}^{2}\u27f60,$$
we get the claim for
${\Omega}_{Y}^{2}{|}_{{S}_{t}}$
.

Q.E.D.

Lemma 4.6.
For all
$j\ge 0$
,
${H}^{2}(Y,{\Omega}_{Y}^{j})=0$
.
$Proof.$
The sheaves
${\Omega}_{X}^{j}\otimes {\mathcal{O}}_{X}\left(nD\right)$
are locally free by the above lemmas for all
$n\gg 0$
. By upper semi-continuity theorem, there is an integer
${n}_{0}$
such that for all
$n\ge {n}_{0}$
,
${R}^{2}{f}_{*}{\Omega}_{X}^{j}\otimes {\mathcal{O}}_{X}\left(nD\right){|}_{C}=0$
. By Goodman and Hartshorne's Lemma [GH] , we have
$${H}^{2}(Y,{\Omega}_{Y}^{j})={lim}_{\overrightarrow{n}}{H}^{2}\left({f}^{-1}\right(C),{\Omega}_{X}^{j}\otimes {\mathcal{O}}_{X}(nD\left)\right)={lim}_{\overrightarrow{n}}{R}^{2}{f}_{*}{\Omega}_{X}^{j}\otimes {\mathcal{O}}_{X}\left(nD\right)\left(C\right)=0.$$

Q.E.D.

Lemma 4.7.
For all
$j\ge 0$
,
${H}^{1}(Y,{\Omega}_{Y}^{j})=0$
.
$Proof.$
By the local freeness lemmas and Goodman and Hartshorne's Lemma, we have
$${H}^{1}(Y,{\Omega}_{Y}^{j})={lim}_{\overrightarrow{n}}{H}^{1}\left({f}^{-1}\right(C),{\Omega}_{X}^{j}\otimes {\mathcal{O}}_{X}(nD\left)\right)={lim}_{\overrightarrow{n}}{R}^{1}{f}_{*}{\Omega}_{X}^{j}\otimes {\mathcal{O}}_{X}\left(nD\right)\left(C\right)=0.$$

Q.E.D.

5 Proof of Theorem 1.5

We will prove Theorem 1.5 by construct an example. Let
${C}_{t}$
be a smooth projective elliptic curve defined by
${y}^{2}=x(x-1)(x-t)$
,
$t\ne 0,1$
. Let
$Z$
be the elliptic surface defined by the same equation, then we have surjective morphism from
$Z$
to
$C=\mathbb{C}-\{0,1\}$
such that for every
$t\in C$
, the fibre
${f}^{-1}\left(t\right)={C}_{t}$
.

Lemma 5.1.
There is a rank 2 vector bundle
$E$
on
$Z$
such that when restricted to
${C}_{t}$
,
$E{|}_{{C}_{t}}={E}_{t}$
is the unique nonsplit extension of
${\mathcal{O}}_{{C}_{t}}$
by
${\mathcal{O}}_{{C}_{t}}$
, where
$f$
is the morphism from
$Z$
to
$C$
.
$Proof$
. Since
$f:Z\u27f6C$
is an elliptic fibration, for every
$t$
, we have
$${h}^{1}\left({f}^{-1}\right(t),{\mathcal{O}}_{{f}^{-1}\left(t\right)})={h}^{1}\left({\mathcal{O}}_{{C}_{t}}\right)=1.$$
So
$${R}^{1}{f}_{*}{\mathcal{O}}_{Z}\otimes \mathbb{C}\left(t\right)\sim ={H}^{1}({C}_{t},{\mathcal{O}}_{{C}_{t}})\sim =\mathbb{C}.$$
It gives us
$$({R}^{1}{f}_{*}{\mathcal{O}}_{Z}{)}_{t}/{\mathcal{P}}_{t}({R}^{1}{f}_{*}{\mathcal{O}}_{Z}{)}_{t}\sim =\mathbb{C}.$$
By Nakayama's lemma,
${R}^{1}{f}_{*}{\mathcal{O}}_{Z}$
is a line bundle on
$C$
. Since
$\mathbb{C}[x,1/x,1/(x-1\left)\right]$
is principle ideal domain [La] , the Picard group of
$C$
is trivial, i.e., any line bundle on
$C$
is trivial. Therefore
${R}^{1}{f}_{*}{\mathcal{O}}_{Z}\sim ={\mathcal{O}}_{C}$
and
$${H}^{1}(Z,{\mathcal{O}}_{Z})={R}^{1}{f}_{*}{\mathcal{O}}_{Z}\left(C\right)={\mathcal{O}}_{C}\left(C\right)=\mathbb{C}[x,\frac{1}{x},\frac{1}{x-1}].$$
Given any exact sequence of vector bundles
$$0\u27f6{\mathcal{O}}_{Z}\u27f6E\u27f6{\mathcal{O}}_{Z}\u27f60,$$
let
$\xi $
be the image of unit of
${H}^{0}(Z,{\mathcal{O}}_{Z})$
in
${H}^{1}(Z,{\mathcal{O}}_{Z})$
, we get an element of
${H}^{1}(Z,{\mathcal{O}}_{Z})$
. Conversely, given any element
$\xi $
in
${H}^{1}(Z,{\mathcal{O}}_{Z})$
, we can get an exact sequence as above. The procedure is the following. Take any (degree) large ample line bundle
$L$
on the elliptic surface
$Z$
, for any positive integer
$n$
, we have an exact sequence
$$0\u27f6{\mathcal{O}}_{Z}\stackrel{\alpha}{\u27f6}{L}^{\oplus n}\stackrel{\beta}{\u27f6}G\u27f60,$$
where
$G$
is the quotient which is a vector bundle. We may assume
${H}^{1}(Z,L)=0$
by raising the degree of
$L$
since
$L$
is ample. So we have surjective map
${H}^{0}(Z,G)\twoheadrightarrow {H}^{1}(Z,{\mathcal{O}}_{Z})$
. Hence
$\xi $
can be lifted to an element
$\eta $
in
${H}^{0}(Z,G)$
. This element
$\eta $
defines a map from
${\mathcal{O}}_{Z}$
to
$G$
,
$\eta :{\mathcal{O}}_{Z}\to G$
, sending 1 to
$\eta $
. Let
$E={\beta}^{-1}\left(\eta \right({\mathcal{O}}_{Z}\left)\right)$
, then we have exact sequence
$$0\u27f6{\mathcal{O}}_{Z}\stackrel{\alpha}{\u27f6}E\stackrel{\beta}{\u27f6}\eta \left({\mathcal{O}}_{Z}\right)={\mathcal{O}}_{Z}\u27f60.$$
So there is one-to-one correspondence between the elements of
${H}^{1}(Z,{\mathcal{O}}_{Z})$
and the above exact sequences. Further, we have commutative diagram

Since
$\mathbb{C}\subset {H}^{1}(Z,{\mathcal{O}}_{Z})=\mathbb{C}[x,\frac{1}{x},\frac{1}{x-1}]$
,
$1\in {H}^{1}(Z,{\mathcal{O}}_{Z})$
. This nonzero element
$1$
corresponds to a rank 2 vector bundle
$E$
such that when restricted to every fibre
${C}_{t}$
, it is the nonsplit extension of
${\mathcal{O}}_{{C}_{t}}$
by
${\mathcal{O}}_{{C}_{t}}$
. In fact, in the natural restriction map
$${H}^{1}(Z,{\mathcal{O}}_{Z})\u27f6{H}^{1}({C}_{t},{\mathcal{O}}_{{C}_{t}}),$$
$1$
goes to
$1$
. A nonzero element of
${H}^{1}({C}_{t},{\mathcal{O}}_{{C}_{t}})$
determines a nonsplit extension of
${\mathcal{O}}_{{C}_{t}}$
by
${\mathcal{O}}_{{C}_{t}}$
.

$$\begin{array}{cccc}0\u27f6{\mathcal{O}}_{Z}& \u27f6{L}^{\oplus n}& \u27f6G& \u27f60\end{array}$$ |

$$\begin{array}{cccc}\parallel & \uparrow & \uparrow & \end{array}$$ |

$$\begin{array}{cccc}0\u27f6{\mathcal{O}}_{Z}& \u27f6E& \u27f6{\mathcal{O}}_{Z}& \u27f60.\end{array}$$ |

Q.E.D.

Lemma 5.2.
There is a divisor
$D$
on
$X={\mathbb{P}}_{Z}\left(E\right)$
such that when restricted to
${X}_{t}={\mathbb{P}}_{{C}_{t}}\left({E}_{t}\right)$
,
$D{|}_{{X}_{t}}={D}_{t}$
is the canonical section of
${X}_{t}$
.
$Proof.$
By the above lemma, we have surjective map from
$E$
to
${\mathcal{O}}_{Z}$
. It corresponds to a section
$\sigma :Z\u27f6X.$
When restricted to
${C}_{t}$
,
$\sigma {|}_{{C}_{t}}={\sigma}_{t}:{C}_{t}\u27f6{X}_{t}$
is the unique nonsplit extension of
${\mathcal{O}}_{{C}_{t}}$
by
${\mathcal{O}}_{{C}_{t}}$
.

Q.E.D.

Let
$Y=X-D$
, by the proof Theorem 1.3, we have
${H}^{i}(Y,{\Omega}_{Y}^{j})=0$
for all
$i>0$
and
$j\ge 0$
. By now we have constructed a non-affine, non-product example of threefold
$Y$
with vanishing Hodge cohomology. This proves Theorem 1.5.
Acknowledgments I would like to express my thanks to the following professors for helpful discussions: Michael Artin, Steven Dale Cutkosky, Dan Edidin, Robert Gunning, J
$\text{a\u0301}$
nos Kol
$\text{a\u013a}$
r, N.Mohan Kumar, Kenji Matsuki, Thomas Peternell, Bangere P.Purnaprajna, Zhenbo Qin, A. Prabhakar Rao, Yum-Tong Siu, David Wright and Qi Zhang. References

- Arapura, D., Complex Algebraic Varieties and their Cohomology, Lecture Notes, 2003.
- Artin, M., Some numerical criteria for contractibility of curves on algebraic surfaces. Amer. J. Math. 84(1962), 485-496.
- B $\stackrel{\u02d8}{\text{a}}$ descu, Lucian, Algebraic Surfaces, Universitext, Springer-Verlag, 2001.
- Barth, W.; Peters, C.; Van de Ven, A., Compact Complex Surfaces, Springer-Verlag, second edition, 2004.
- Bourbaki, N., Commutative Algebra, Springer-Verlag, 1989.
- Bourbaki, N., General Topology, Chapters 5-10, Springer-Verlag, 1989.
- Cutkosky, S. D., Zariski decomposition of divisors on algebraic varieties, Duke Math. J. 53(1986), no. 1, 149-156.
- Cutkosky, S. D.; Srinivas, V. On a problem of Zarisky on dimensions of linear systems. Ann. of Math. (2) 137 (1993), no. 3, 531-559.
- Friedman, Robert; Qin, Zhenbo On Complex surfaces diffeomorphic to rational surfaces. Invent. Math. 120(1995), no.1, 81-117.
- Fritzsche, K.; Grauert, H., From Holomorphic Functions to Complex Manifolds, GTM 213, Springer-Verlag, 2002.
- Goodman, J., Hartshorne, R., Schemes with finite-dimensional cohomology groups, American Journal of Mathematics, v. 91, Issue 1, 258-266, 1969.
- Grauert, H., Remmert, R., Coherent Analytic Sheaves, Springer-Verlag, 1984.
- Griffiths, P. and Harris, J., Principals of Algebraic Geometry, John Wiley $\&$ Sons, Inc., 1994.
- Hartshorne, R., Algebraic Geometry, Springer-Verlag, 1997.
- Hartshorne, R., Ample Subvarieties of Algebraic Varieties, Lecture Notes in Mathematics, 156, Springer-Verlag, 1970.
- Hartshorne, R., Local Cohomology, Lecture Notes in Mathe., 41, Springer-Verlag, 1967.
- Iitaka, S., Birational Geometry for Open Varieties, Les Presses de l'Universite de Montreal, 1981.
- Iitaka, S., Birational Geometry of Algebraic Varieties, ICM, 1983.
- Iitaka, S., Birational geometry and logarithmic forms, Recent Progress of Algebraic Geometry in Japan, North-Holland Mathematics Studies 73, 1-27.
- Iitaka, S., Deformation of compact complex surfaces I, Global Analysis, papers in honor of K. Kodaira, Princeton Univ. Press, 1969, 267-272.
- Iitaka, S., Deformation of compact complex surfaces II, J. Math. Soc. Japan 22, 1970, 247-261.
- Iitaka, S., Some applications of logarithmic Kodaira dimension, Proc. Int. Symp. algebraic Geometry, Kyoto, 1978.
- Kawamata, Y., On the extension problem of pluricanonical forms, Algebraic geometry: Hirzebruch 70 (Warsaw, 1998), 193-207, Contemp. Math., 241. Amer. Math. Soc., Province, RI, 1999.
- Kawamata, Y., Deformations of canonical singularities. J. Amer. Math. Soc. 12(1999), no. 1, 85-92.
- Kodaira, K., On stability of compact submanifold of complex manifolds, Amer. J. Math. 85(1963), 79-94.
- Kol $\text{a\u013a}$ r, J., Higher Direct Images of Dualizing Sheaves I, Ann. of Math., v.123, 1(1986), 11-42.
- Kol $\text{a\u013a}$ r, J., Higher Direct Images of Dualizing Sheaves, II, Ann. of Math., v.124, 1(1986), 171-202.
- Kol $\text{a\u013a}$ r, J., Mori, S., Birational Geometry of Algebraic Varieties, Cambridge University Press, 1998.
- Kosarew, Siegmund; Peternell, Thomas Formal cohomology, analytic cohomology and nonalgebraic manifolds. Compositio Math. 74 (1990), no. 3, 299–325.
- Kumar, N. Mohan, Affine-Like Surfaces, Journal of Algebraic Geometry, 2(1993), 689-703.
- Lang, S., Algebra, 1993.
- Matsuki, Kenji, Introduction to the Mori program. Universitext. Springer-Verlag, New York, 2002.
- Luo, Tie, Global 2-forms on regular 3-folds of general type. Duke Math. J. 71 (1993), no.3, 859-869.
- Luo, Tie, Global holomorphic 2-forms and pluricanonical systems on threefolds. Math. Ann. 318 (2000), no. 4, 707-730.
- Luo, Tie; Zhang, Qi, Holomorphic forms on threefolds, preprint, 2003.
- Mumford, D., Abelian Varieties, Oxford University Press, 1970.
- Neeman, A., Ueda theory: Theorems and problems, Mem. Amer. Math. Soc., No.415 (1989), 1-123.
- Norguet, F.; Siu, Y.T. Holomorphic convexity of spaces of analytic cycles. Bull. Soc. Math. France 105, 191-223(1977).
- Peternell, T., Hodge-kohomolodie und Steinsche Mannigfaltigkeiten, Complex Analysis, Aspects of Mathematics, Vieweg-Verlag, 1990, 235-246.
- Qin, Zhenbo, Stable rank- $2$ bundles on simply connected elliptic surfaces. Duke Math. J. 67(1992), no. 3, 557-569.
- Serre, J. P., Quelques probl $\text{e\u0300}$ mes globaux relatifs aus vari $\text{e\u0301}$ t $\text{e\u0301}$ s deStein, Collected Papers, Vol.1, Springer-Verlag(1985), 259-270.
- Simpson, Carlos, The construction problem in K $\ddot{\text{a}}$ hler geometry, to appear in the volume ”Different Faces of Geometry”, International Mathematics Series, vol.3.
- Siu, Y.-T., Analytic sheaf cohomology of dimension n of n-dimensional complex spaces. Trans. Amer. Math. Soc. 143, 77-94(1969).
- Siu, Y.-T., Invariance of plurigenera, Invent. math. 134, 661-673(1998).
- Siu, Y.-T., Invariance of plurigenera and torsion-freeness of direct image sheaves of pluricanonical bundles. Finite or infinite dimensional complex analysis and applications, 45-83, Adv. Complex Anal. Appl., 2, Kluwer Acad. Publ., Dordrecht, 2004.
- Ueda, T., On the neighborhood of a compact complex curve with topologically trivial normal bundle, J. Math. Kyoto Univ. 22-4, 583-607, 1983.
- Ueno, K., Classification Theory of Algebraic Varieties and Compact Complex Spaces, Lecture Notes in Mathematics, v.439, 1975, Springer-Verlag.
- Zariski, O., The theorem of Riemann-Roch for high multiples of an effective divisor on an algebraic surface, Ann. of Math. (2), 76(1962), 560-616.
- Zhang, Jing, Threefolds with vanishing Hodge cohomology, Trans. Amer. Math. Soc. 357, Number 5, 1977-1994.
- Zhang, Qi, Global holomorphic one-forms on projective manifolds with ample canonical bundles. J. Algebraic Geometry 6 (1997), 777-787.

^{1 }
Department of Mathematics, University of Missouri, Columbia, MO 65211, USA email: zhangj@math.missouri.edu