Let V be a 2-dimensional compact complex manifold. V is called a K-3 surface if : a) the irregularity q = dim H1(V, θ) of V vanishes and b) the first Chern class c1 of V vanishes. The canonical sheaf (of holo-morphic 2-forms) K of such a surface is trivial, since q = 0 implies that the Chern class map cx : Pic (V) → H2(V, Z) is injective : thus V has a nowhere zero holomorphic 2-form.

It is well known that many properties possessed by functions holomorphic and bounded in a region are also possessed by functions meromorphic and omitting three values. Noshiro [14] in 1938 and Lehto and Virtanen [12] in 1957 independently defined the notion of “normal functions” ; they and many others subsequently discovered that most properties concerning boundary behavior and value distribution acquired by meromorphic functions omitting three values in the unit disk (or more general, in a simply-connected region) are also valid properties of “normal functions” defined there. In their research on the problems of value distribution of normal functions, Lange [9], Gavrilov [5] and Gauthier [4] have discovered that functions normal in the disk are exactly those which omit three values “locally,” i.e., they do not possess any “p-sequence” (see above references). However, the definition of a function being normal in a region depends on the simply-connectedness of the region or its universal covering surface. It is thus difficult to judge if a function defined in an arbitrary region is normal.

In the study of moduli of polarized abelian varieties and of curves as well as in the theory of complex multiplications, the notion of fields of moduli for structures plays an essential role. This notion was first introduced by Matsusaka [7] for polarized varieties with some pleasing properties and later was given a more comprehensible treatment by Shimura [10] in the case of polarized abelian varieties or polarized abelian varieties with some further structures. Both authors discussed fields of moduli not only in algebraic geometry of characteristic zero but also in that of positive characteristic, but in the latter case the definition of fields of moduli seems somewhat artificial and there have been no essential applications of them so far.

By a harmonic space we mean a pair (X, H) where X is a locally compact, non-compact, connected, locally connected Hausdorff space ; and H is a sheaf of harmonic functions defined as follows: Suppose to each open set Ω ⊂ X there corresponds a linear space H(Ω) of finitely-continuous real-valued functions defined on Ω. Then H = {H(Ω)}Ω must satisfy the three axioms of Brelot (1) and in addition Axiom 4 of Loeb (4): 1 is H-superharmonic in X.

An important problem in the study of the Hilbert space PD(R) of Dirichlet finite solutions of Δu = Pu on a Riemann surface R is to know the behavior of the reproducing kernel in PD(R). The main result of this paper is that the reproducing kernel is strictly positive.

Let k be an algebraically closed field of characteristic p > 0, and let X be a curve defined over k. The aim of this paper is to study the behavior of the Frobenius map F*: H1(X, E) → H1(X, F*E) for a vector bundle E.

As an application of the Garleman-type estimation Hörmander [4], p. 221, has proved the following:

A solution (distribution) of the Tricomi equation

in an open set Ω in belongs to C∞(Ω) if it is in C∞(Ω-) where .

Throughout this paper, k is a perfect field of characteristic p > 0, R is a complete discrete valuation ring with residue field k and quotient field of characteristic zero, and Z is a connected smooth prescheme of finite type over k.

Let Xt(ω)) be a stochastic process with stationary independent increments on the N-dimensional Euclidean space RN, right continuous in t ≧ 0 and starting at the origin. Let C0(RN) be the Banach space of real-valued continuous functions on RN vanishing at infinity with norm . The process induces a transition semigroup of operators Tt on C0(RN) :

Ttf(x) = Ef(x + Xt).

The following theorem on infinite class field towers is well-known.

0. Introduction. Let k be an uncountable cardinal, and let λ be a regular cardinal less than k. Let I be a λ-saturated non-trivial ideal on k. Prikry, in his thesis, showed that, in certain Boolean extensions, k has a λ-saturated non-trivial ideal on k.

Let M be a compact C∞ Riemannian manifold, X a Killing vector field on M, and φt its 1-parameter group of isometries of M. In this, paper, we obtain some basic properties of the set of periodic points of φt. We show that the set of least periods is always finite, and the set P(X, t) of points of M having least period t for the vector field X is a totally geodesic submanifold, with possibly non-empty boundary. Moreover, we show there are at least m geometrically distinct closed geodesic orbits of φt, where m is the number of least periods which are not integral multiples of any other least period.

In this paper we discuss the boundary properties of a function f which is analytic in the open unit disc Δ and has Hadamard gaps—i.e.

(1)

where

. (2)

A holomorphic vector bundle E over a complex analytic manifold is said to be simple, if its global endomorphism ring Endc (E) is isomorphic to C. Projectifying the fibers of E, we get the associated projective bundle P(E) of E, If we can choose a system of constant transition functions of P(Exs), the projective bundle P(E) is said to be locally flat.

One of the main problems in potential theory is to determine the class of kernels satisfying the domination principle or the complete maximum principle. As to positive symmetric kernels this is settled to a certain extent, but as to non-symmetric kernels we have not yet obtain satisfactorily large explicit classes. In this note we shall give a class of positive non-symmetric convolution kernels on the real line satisfying the complete maximum principle.

Let Mn be a differentiable manifold (of class C∞). By a Morse function on Mn we mean a differentiable function whose critical points are all non-degenerate. If f is an immersion of Mn into a Euclidean space Rm, we may obtain Morse functions on Mn in the following way.

Tahara [1] has concretely determined the conjugate classes of finite subgroups of GL(3, Z). The group W5 in his list of groups of order 24 is in fact of order 12 and consists of the same matrices as the group W6 in his list of groups order 12. Hence, there are only 10 conjugate classes of subgroups of order 24 in GL(3, Z) and the total number of conjugate classes of finite subgroups is reduced to 73.