The Hodge group (aka special Mumford-Tate group) of a complex abelian variety $X$ is a certain linear reductive algebraic group over the rationals that is closely related to the endomorphism ring of $X$. (For example, the Hodge group is commutative if and only if $X$ is an abelian variety of CM-type.) In this talk I discuss" lower bounds" for the center of Hodge groups of superelliptic jacobians. (This is a joint work with Jiangwei Xue.)

The values of the partial zeta functions for an abelian extension of number fields at non-positive integers are rational numbers with known bounds on their denominators. David Hayes conjectured that when the associated fields satisfy certain algebraic conditions, the bound at s=0 can be sharpened. I will present a counterexample to Hayes's conjecture. I will then propose a new conjecture sharpening the bounds at arbitrary non-positive integers that implies a weaker version of Hayes conjecture at s=0. I will conclude by proving that the new conjecture is a consequence of the Coates-Sinnott conjecture.

Many constructions in algebraic geometry require one to choose a point
outside a countable union of subvarieties. Over $\C$ this is always
possible. Over a countable field, a countable union of subvarieties
can cover all the closed points. Let $k$ be a finitely generated
field of characteristic zero and let $\kbar$ be an algebraic closure.
Let $A$ be a semiabelian variety defined over $k$, and let $\End(A)$
be the ring of endomorphisms of $A$ over $\kbar$. Let $X\subset A$ be
a subvariety of smaller dimension. We show that $\Union_{f\in
\End(A)} f(X(\kbar))$ does not equal $A(\kbar)$. Bogomolov and
Tschinkel show that the above is false for $k$ equal to an algebraic
closure of a finite field, and use the result to show that on any
Kummer surface over such $k$, the union of all rational curves covers
all of the closed points. We give further examples of such problems.